diff options
Diffstat (limited to 'drivers/gpu/drm')
870 files changed, 28042 insertions, 12779 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 8d16f280b695..384b798a9bad 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -62,7 +62,7 @@ subdir-ccflags-$(CONFIG_DRM_AMDGPU_WERROR) += -Werror amdgpu-y := amdgpu_drv.o # add KMS driver -amdgpu-y += amdgpu_device.o amdgpu_kms.o \ +amdgpu-y += amdgpu_device.o amdgpu_doorbell_mgr.o amdgpu_kms.o \ amdgpu_atombios.o atombios_crtc.o amdgpu_connectors.o \ atom.o amdgpu_fence.o amdgpu_ttm.o amdgpu_object.o amdgpu_gart.o \ amdgpu_encoders.o amdgpu_display.o amdgpu_i2c.o \ @@ -98,7 +98,7 @@ amdgpu-y += \ vega20_reg_init.o nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \ nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o \ sienna_cichlid.o smu_v13_0_10.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o \ - nbio_v7_9.o aqua_vanjaram_reg_init.o + nbio_v7_9.o aqua_vanjaram.o # add DF block amdgpu-y += \ @@ -129,7 +129,8 @@ amdgpu-y += \ vega10_ih.o \ vega20_ih.o \ navi10_ih.o \ - ih_v6_0.o + ih_v6_0.o \ + ih_v6_1.o # add PSP block amdgpu-y += \ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 8f2255b3a38a..a79d53bdbe13 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -192,7 +192,6 @@ extern int amdgpu_emu_mode; extern uint amdgpu_smu_memory_pool_size; extern int amdgpu_smu_pptable_id; extern uint amdgpu_dc_feature_mask; -extern uint amdgpu_freesync_vid_mode; extern uint amdgpu_dc_debug_mask; extern uint amdgpu_dc_visual_confirm; extern uint amdgpu_dm_abm_level; @@ -1033,7 +1032,6 @@ struct amdgpu_device { bool has_pr3; bool ucode_sysfs_en; - bool psp_sysfs_en; /* Chip product information */ char product_number[20]; @@ -1128,7 +1126,7 @@ void amdgpu_device_wreg(struct amdgpu_device *adev, void amdgpu_device_indirect_wreg_ext(struct amdgpu_device *adev, u64 reg_addr, u32 reg_data); void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, - uint32_t reg, uint32_t v); + uint32_t reg, uint32_t v, uint32_t xcc_id); void amdgpu_mm_wreg8(struct amdgpu_device *adev, uint32_t offset, uint8_t value); uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset); @@ -1506,4 +1504,8 @@ static inline bool amdgpu_is_tmz(struct amdgpu_device *adev) int amdgpu_in_reset(struct amdgpu_device *adev); +extern const struct attribute_group amdgpu_vram_mgr_attr_group; +extern const struct attribute_group amdgpu_gtt_mgr_attr_group; +extern const struct attribute_group amdgpu_flash_attr_group; + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 385c6acb5728..2bca37044ad0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -706,7 +706,7 @@ int amdgpu_acpi_pcie_performance_request(struct amdgpu_device *adev, atcs_input.size = sizeof(struct atcs_pref_req_input); /* client id (bit 2-0: func num, 7-3: dev num, 15-8: bus num) */ - atcs_input.client_id = adev->pdev->devfn | (adev->pdev->bus->number << 8); + atcs_input.client_id = pci_dev_id(adev->pdev); atcs_input.valid_flags_mask = ATCS_VALID_FLAGS_MASK; atcs_input.flags = ATCS_WAIT_FOR_COMPLETION; if (advertise) @@ -776,7 +776,7 @@ int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, atcs_input.size = sizeof(struct atcs_pwr_shift_input); /* dGPU id (bit 2-0: func num, 7-3: dev num, 15-8: bus num) */ - atcs_input.dgpu_id = adev->pdev->devfn | (adev->pdev->bus->number << 8); + atcs_input.dgpu_id = pci_dev_id(adev->pdev); atcs_input.dev_acpi_state = dev_state; atcs_input.drv_state = drv_state; @@ -868,7 +868,7 @@ static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm) if (!numa_info) { struct sysinfo info; - numa_info = kzalloc(sizeof *numa_info, GFP_KERNEL); + numa_info = kzalloc(sizeof(*numa_info), GFP_KERNEL); if (!numa_info) return NULL; @@ -1141,7 +1141,7 @@ int amdgpu_acpi_get_tmr_info(struct amdgpu_device *adev, u64 *tmr_offset, if (!tmr_offset || !tmr_size) return -EINVAL; - bdf = (adev->pdev->bus->number << 8) | adev->pdev->devfn; + bdf = pci_dev_id(adev->pdev); dev_info = amdgpu_acpi_get_dev(bdf); if (!dev_info) return -ENOENT; @@ -1162,7 +1162,7 @@ int amdgpu_acpi_get_mem_info(struct amdgpu_device *adev, int xcc_id, if (!numa_info) return -EINVAL; - bdf = (adev->pdev->bus->number << 8) | adev->pdev->devfn; + bdf = pci_dev_id(adev->pdev); dev_info = amdgpu_acpi_get_dev(bdf); if (!dev_info) return -ENOENT; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index b4fcad0e62f7..25d5fda5b243 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -226,16 +226,6 @@ void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm) kgd2kfd_suspend(adev->kfd.dev, run_pm); } -int amdgpu_amdkfd_resume_iommu(struct amdgpu_device *adev) -{ - int r = 0; - - if (adev->kfd.dev) - r = kgd2kfd_resume_iommu(adev->kfd.dev); - - return r; -} - int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm) { int r = 0; @@ -452,9 +442,7 @@ void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev, mem_info->local_mem_size_public, mem_info->local_mem_size_private); - if (amdgpu_sriov_vf(adev)) - mem_info->mem_clk_max = adev->clock.default_mclk / 100; - else if (adev->pm.dpm_enabled) { + if (adev->pm.dpm_enabled) { if (amdgpu_emu_mode == 1) mem_info->mem_clk_max = 0; else @@ -473,9 +461,7 @@ uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev) uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev) { /* the sclk is in quantas of 10kHz */ - if (amdgpu_sriov_vf(adev)) - return adev->clock.default_sclk / 100; - else if (adev->pm.dpm_enabled) + if (adev->pm.dpm_enabled) return amdgpu_dpm_get_sclk(adev, false) / 100; else return 100; @@ -492,7 +478,7 @@ void amdgpu_amdkfd_get_cu_info(struct amdgpu_device *adev, struct kfd_cu_info *c cu_info->cu_active_number = acu_info.number; cu_info->cu_ao_mask = acu_info.ao_cu_mask; memcpy(&cu_info->cu_bitmap[0], &acu_info.bitmap[0], - sizeof(acu_info.bitmap)); + sizeof(cu_info->cu_bitmap)); cu_info->num_shader_engines = adev->gfx.config.max_shader_engines; cu_info->num_shader_arrays_per_engine = adev->gfx.config.max_sh_per_se; cu_info->num_cu_per_sh = adev->gfx.config.max_cu_per_sh; @@ -830,3 +816,53 @@ u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id) return adev->gmc.real_vram_size; } } + +int amdgpu_amdkfd_unmap_hiq(struct amdgpu_device *adev, u32 doorbell_off, + u32 inst) +{ + struct amdgpu_kiq *kiq = &adev->gfx.kiq[inst]; + struct amdgpu_ring *kiq_ring = &kiq->ring; + struct amdgpu_ring_funcs *ring_funcs; + struct amdgpu_ring *ring; + int r = 0; + + if (!kiq->pmf || !kiq->pmf->kiq_unmap_queues) + return -EINVAL; + + ring_funcs = kzalloc(sizeof(*ring_funcs), GFP_KERNEL); + if (!ring_funcs) + return -ENOMEM; + + ring = kzalloc(sizeof(*ring), GFP_KERNEL); + if (!ring) { + r = -ENOMEM; + goto free_ring_funcs; + } + + ring_funcs->type = AMDGPU_RING_TYPE_COMPUTE; + ring->doorbell_index = doorbell_off; + ring->funcs = ring_funcs; + + spin_lock(&kiq->ring_lock); + + if (amdgpu_ring_alloc(kiq_ring, kiq->pmf->unmap_queues_size)) { + spin_unlock(&kiq->ring_lock); + r = -ENOMEM; + goto free_ring; + } + + kiq->pmf->kiq_unmap_queues(kiq_ring, ring, RESET_QUEUES, 0, 0); + + if (kiq_ring->sched.ready && !adev->job_hang) + r = amdgpu_ring_test_helper(kiq_ring); + + spin_unlock(&kiq->ring_lock); + +free_ring: + kfree(ring); + +free_ring_funcs: + kfree(ring_funcs); + + return r; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 1e4cc1fe88fc..2fe9860725bd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -148,7 +148,6 @@ int amdgpu_amdkfd_init(void); void amdgpu_amdkfd_fini(void); void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm); -int amdgpu_amdkfd_resume_iommu(struct amdgpu_device *adev); int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm); void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, const void *ih_ring_entry); @@ -251,6 +250,8 @@ int amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(struct amdgpu_device *dst, int amdgpu_amdkfd_get_pcie_bandwidth_mbytes(struct amdgpu_device *adev, bool is_min); int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev, uint32_t *payload); +int amdgpu_amdkfd_unmap_hiq(struct amdgpu_device *adev, u32 doorbell_off, + u32 inst); /* Read user wptr from a specified user address space with page fault * disabled. The memory must be pinned and mapped to the hardware when @@ -397,7 +398,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, const struct kgd2kfd_shared_resources *gpu_resources); void kgd2kfd_device_exit(struct kfd_dev *kfd); void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm); -int kgd2kfd_resume_iommu(struct kfd_dev *kfd); int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm); int kgd2kfd_pre_reset(struct kfd_dev *kfd); int kgd2kfd_post_reset(struct kfd_dev *kfd); @@ -437,11 +437,6 @@ static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) { } -static int __maybe_unused kgd2kfd_resume_iommu(struct kfd_dev *kfd) -{ - return 0; -} - static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) { return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c index 60f9e027fb66..aff08321e976 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.c @@ -23,6 +23,7 @@ #include "amdgpu_amdkfd.h" #include "amdgpu_amdkfd_arcturus.h" #include "amdgpu_amdkfd_gfx_v9.h" +#include "amdgpu_amdkfd_aldebaran.h" #include "gc/gc_9_4_2_offset.h" #include "gc/gc_9_4_2_sh_mask.h" #include <uapi/linux/kfd_ioctl.h> @@ -36,7 +37,7 @@ * initialize the debug mode registers after it has disabled GFX off during the * debug session. */ -static uint32_t kgd_aldebaran_enable_debug_trap(struct amdgpu_device *adev, +uint32_t kgd_aldebaran_enable_debug_trap(struct amdgpu_device *adev, bool restore_dbg_registers, uint32_t vmid) { @@ -107,7 +108,7 @@ static uint32_t kgd_aldebaran_set_wave_launch_trap_override(struct amdgpu_device return data; } -static uint32_t kgd_aldebaran_set_wave_launch_mode(struct amdgpu_device *adev, +uint32_t kgd_aldebaran_set_wave_launch_mode(struct amdgpu_device *adev, uint8_t wave_launch_mode, uint32_t vmid) { @@ -125,7 +126,8 @@ static uint32_t kgd_gfx_aldebaran_set_address_watch( uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid) + uint32_t debug_vmid, + uint32_t inst) { uint32_t watch_address_high; uint32_t watch_address_low; @@ -161,12 +163,6 @@ static uint32_t kgd_gfx_aldebaran_set_address_watch( return watch_address_cntl; } -static uint32_t kgd_gfx_aldebaran_clear_address_watch(struct amdgpu_device *adev, - uint32_t watch_id) -{ - return 0; -} - const struct kfd2kgd_calls aldebaran_kfd2kgd = { .program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_gfx_v9_set_pasid_vmid_mapping, @@ -191,7 +187,7 @@ const struct kfd2kgd_calls aldebaran_kfd2kgd = { .set_wave_launch_trap_override = kgd_aldebaran_set_wave_launch_trap_override, .set_wave_launch_mode = kgd_aldebaran_set_wave_launch_mode, .set_address_watch = kgd_gfx_aldebaran_set_address_watch, - .clear_address_watch = kgd_gfx_aldebaran_clear_address_watch, + .clear_address_watch = kgd_gfx_v9_clear_address_watch, .get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times, .build_grace_period_packet_info = kgd_gfx_v9_build_grace_period_packet_info, .program_trap_handler_settings = kgd_gfx_v9_program_trap_handler_settings, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.h new file mode 100644 index 000000000000..a7bdaf8d82dd --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_aldebaran.h @@ -0,0 +1,27 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +uint32_t kgd_aldebaran_enable_debug_trap(struct amdgpu_device *adev, + bool restore_dbg_registers, + uint32_t vmid); +uint32_t kgd_aldebaran_set_wave_launch_mode(struct amdgpu_device *adev, + uint8_t wave_launch_mode, + uint32_t vmid); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c index 5b4b7f8b92a5..490c8f5ddb60 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c @@ -22,6 +22,7 @@ #include "amdgpu.h" #include "amdgpu_amdkfd.h" #include "amdgpu_amdkfd_gfx_v9.h" +#include "amdgpu_amdkfd_aldebaran.h" #include "gc/gc_9_4_3_offset.h" #include "gc/gc_9_4_3_sh_mask.h" #include "athub/athub_1_8_0_offset.h" @@ -32,6 +33,7 @@ #include "soc15.h" #include "sdma/sdma_4_4_2_offset.h" #include "sdma/sdma_4_4_2_sh_mask.h" +#include <uapi/linux/kfd_ioctl.h> static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd) { @@ -361,6 +363,156 @@ static int kgd_gfx_v9_4_3_hqd_load(struct amdgpu_device *adev, void *mqd, return 0; } +/* returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_gfx_v9_4_3_disable_debug_trap(struct amdgpu_device *adev, + bool keep_trap_enabled, + uint32_t vmid) +{ + uint32_t data = 0; + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); + + return data; +} + +static int kgd_gfx_v9_4_3_validate_trap_override_request( + struct amdgpu_device *adev, + uint32_t trap_override, + uint32_t *trap_mask_supported) +{ + *trap_mask_supported &= KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION | + KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START | + KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END; + + if (trap_override != KFD_DBG_TRAP_OVERRIDE_OR && + trap_override != KFD_DBG_TRAP_OVERRIDE_REPLACE) + return -EPERM; + + return 0; +} + +static uint32_t trap_mask_map_sw_to_hw(uint32_t mask) +{ + uint32_t trap_on_start = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START) ? 1 : 0; + uint32_t trap_on_end = (mask & KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END) ? 1 : 0; + uint32_t excp_en = mask & (KFD_DBG_TRAP_MASK_FP_INVALID | + KFD_DBG_TRAP_MASK_FP_INPUT_DENORMAL | + KFD_DBG_TRAP_MASK_FP_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_FP_OVERFLOW | + KFD_DBG_TRAP_MASK_FP_UNDERFLOW | + KFD_DBG_TRAP_MASK_FP_INEXACT | + KFD_DBG_TRAP_MASK_INT_DIVIDE_BY_ZERO | + KFD_DBG_TRAP_MASK_DBG_ADDRESS_WATCH | + KFD_DBG_TRAP_MASK_DBG_MEMORY_VIOLATION); + uint32_t ret; + + ret = REG_SET_FIELD(0, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, excp_en); + ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START, trap_on_start); + ret = REG_SET_FIELD(ret, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END, trap_on_end); + + return ret; +} + +static uint32_t trap_mask_map_hw_to_sw(uint32_t mask) +{ + uint32_t ret = REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, EXCP_EN); + + if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_START)) + ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_START; + + if (REG_GET_FIELD(mask, SPI_GDBG_PER_VMID_CNTL, TRAP_ON_END)) + ret |= KFD_DBG_TRAP_MASK_TRAP_ON_WAVE_END; + + return ret; +} + +/* returns TRAP_EN, EXCP_EN and EXCP_REPLACE. */ +static uint32_t kgd_gfx_v9_4_3_set_wave_launch_trap_override( + struct amdgpu_device *adev, + uint32_t vmid, + uint32_t trap_override, + uint32_t trap_mask_bits, + uint32_t trap_mask_request, + uint32_t *trap_mask_prev, + uint32_t kfd_dbg_trap_cntl_prev) + +{ + uint32_t data = 0; + + *trap_mask_prev = trap_mask_map_hw_to_sw(kfd_dbg_trap_cntl_prev); + + data = (trap_mask_bits & trap_mask_request) | + (*trap_mask_prev & ~trap_mask_request); + data = trap_mask_map_sw_to_hw(data); + + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, trap_override); + + return data; +} + +#define TCP_WATCH_STRIDE (regTCP_WATCH1_ADDR_H - regTCP_WATCH0_ADDR_H) +static uint32_t kgd_gfx_v9_4_3_set_address_watch( + struct amdgpu_device *adev, + uint64_t watch_address, + uint32_t watch_address_mask, + uint32_t watch_id, + uint32_t watch_mode, + uint32_t debug_vmid, + uint32_t inst) +{ + uint32_t watch_address_high; + uint32_t watch_address_low; + uint32_t watch_address_cntl; + + watch_address_cntl = 0; + watch_address_low = lower_32_bits(watch_address); + watch_address_high = upper_32_bits(watch_address) & 0xffff; + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MODE, + watch_mode); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + MASK, + watch_address_mask >> 7); + + watch_address_cntl = REG_SET_FIELD(watch_address_cntl, + TCP_WATCH0_CNTL, + VALID, + 1); + + WREG32_RLC((SOC15_REG_OFFSET(GC, GET_INST(GC, inst), + regTCP_WATCH0_ADDR_H) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_high); + + WREG32_RLC((SOC15_REG_OFFSET(GC, GET_INST(GC, inst), + regTCP_WATCH0_ADDR_L) + + (watch_id * TCP_WATCH_STRIDE)), + watch_address_low); + + return watch_address_cntl; +} + +static uint32_t kgd_gfx_v9_4_3_clear_address_watch(struct amdgpu_device *adev, + uint32_t watch_id) +{ + return 0; +} + const struct kfd2kgd_calls gc_9_4_3_kfd2kgd = { .program_sh_mem_settings = kgd_gfx_v9_program_sh_mem_settings, .set_pasid_vmid_mapping = kgd_gfx_v9_4_3_set_pasid_vmid_mapping, @@ -379,6 +531,19 @@ const struct kfd2kgd_calls gc_9_4_3_kfd2kgd = { kgd_gfx_v9_get_atc_vmid_pasid_mapping_info, .set_vm_context_page_table_base = kgd_gfx_v9_set_vm_context_page_table_base, + .get_cu_occupancy = kgd_gfx_v9_get_cu_occupancy, .program_trap_handler_settings = - kgd_gfx_v9_program_trap_handler_settings + kgd_gfx_v9_program_trap_handler_settings, + .build_grace_period_packet_info = + kgd_gfx_v9_build_grace_period_packet_info, + .get_iq_wait_times = kgd_gfx_v9_get_iq_wait_times, + .enable_debug_trap = kgd_aldebaran_enable_debug_trap, + .disable_debug_trap = kgd_gfx_v9_4_3_disable_debug_trap, + .validate_trap_override_request = + kgd_gfx_v9_4_3_validate_trap_override_request, + .set_wave_launch_trap_override = + kgd_gfx_v9_4_3_set_wave_launch_trap_override, + .set_wave_launch_mode = kgd_aldebaran_set_wave_launch_mode, + .set_address_watch = kgd_gfx_v9_4_3_set_address_watch, + .clear_address_watch = kgd_gfx_v9_4_3_clear_address_watch }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c index 8ad7a7779e14..69810b3f1c63 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c @@ -886,7 +886,8 @@ uint32_t kgd_gfx_v10_set_address_watch(struct amdgpu_device *adev, uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid) + uint32_t debug_vmid, + uint32_t inst) { uint32_t watch_address_high; uint32_t watch_address_low; @@ -968,7 +969,8 @@ uint32_t kgd_gfx_v10_clear_address_watch(struct amdgpu_device *adev, * deq_retry_wait_time -- Wait Count for Global Wave Syncs. */ void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, - uint32_t *wait_times) + uint32_t *wait_times, + uint32_t inst) { *wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2)); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h index e6b70196071a..67bcaa3d4226 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.h @@ -44,10 +44,13 @@ uint32_t kgd_gfx_v10_set_address_watch(struct amdgpu_device *adev, uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid); + uint32_t debug_vmid, + uint32_t inst); uint32_t kgd_gfx_v10_clear_address_watch(struct amdgpu_device *adev, uint32_t watch_id); -void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times); +void kgd_gfx_v10_get_iq_wait_times(struct amdgpu_device *adev, + uint32_t *wait_times, + uint32_t inst); void kgd_gfx_v10_build_grace_period_packet_info(struct amdgpu_device *adev, uint32_t wait_times, uint32_t grace_period, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c index 91c3574ebed3..d67d003bada2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c @@ -637,7 +637,7 @@ static uint32_t kgd_gfx_v11_disable_debug_trap(struct amdgpu_device *adev, { uint32_t data = 0; - data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, keep_trap_enabled); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_EN, 0); data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, EXCP_REPLACE, 0); @@ -743,7 +743,8 @@ static uint32_t kgd_gfx_v11_set_address_watch(struct amdgpu_device *adev, uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid) + uint32_t debug_vmid, + uint32_t inst) { uint32_t watch_address_high; uint32_t watch_address_low; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 51d93fb13ea3..3c45a188b701 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -822,7 +822,8 @@ uint32_t kgd_gfx_v9_set_address_watch(struct amdgpu_device *adev, uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid) + uint32_t debug_vmid, + uint32_t inst) { uint32_t watch_address_high; uint32_t watch_address_low; @@ -903,10 +904,12 @@ uint32_t kgd_gfx_v9_clear_address_watch(struct amdgpu_device *adev, * deq_retry_wait_time -- Wait Count for Global Wave Syncs. */ void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, - uint32_t *wait_times) + uint32_t *wait_times, + uint32_t inst) { - *wait_times = RREG32(SOC15_REG_OFFSET(GC, 0, mmCP_IQ_WAIT_TIME2)); + *wait_times = RREG32(SOC15_REG_OFFSET(GC, GET_INST(GC, inst), + mmCP_IQ_WAIT_TIME2)); } void kgd_gfx_v9_set_vm_context_page_table_base(struct amdgpu_device *adev, @@ -1105,7 +1108,7 @@ void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev, *reg_data = wait_times; /* - * The CP cannont handle a 0 grace period input and will result in + * The CP cannot handle a 0 grace period input and will result in * an infinite grace period being set so set to 1 to prevent this. */ if (grace_period == 0) @@ -1128,9 +1131,9 @@ void kgd_gfx_v9_program_trap_handler_settings(struct amdgpu_device *adev, * Program TBA registers */ WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TBA_LO, - lower_32_bits(tba_addr >> 8)); + lower_32_bits(tba_addr >> 8)); WREG32_SOC15(GC, GET_INST(GC, inst), mmSQ_SHADER_TBA_HI, - upper_32_bits(tba_addr >> 8)); + upper_32_bits(tba_addr >> 8)); /* * Program TMA registers diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h index 5f54bff0db49..ce424615f59b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.h @@ -89,10 +89,13 @@ uint32_t kgd_gfx_v9_set_address_watch(struct amdgpu_device *adev, uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid); + uint32_t debug_vmid, + uint32_t inst); uint32_t kgd_gfx_v9_clear_address_watch(struct amdgpu_device *adev, uint32_t watch_id); -void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, uint32_t *wait_times); +void kgd_gfx_v9_get_iq_wait_times(struct amdgpu_device *adev, + uint32_t *wait_times, + uint32_t inst); void kgd_gfx_v9_build_grace_period_packet_info(struct amdgpu_device *adev, uint32_t wait_times, uint32_t grace_period, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index edb35a88cb78..7d6daf8d2bfa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -39,7 +39,6 @@ #include "amdgpu_xgmi.h" #include "kfd_priv.h" #include "kfd_smi_events.h" -#include <drm/ttm/ttm_tt.h> /* Userptr restore delay, just long enough to allow consecutive VM * changes to accumulate @@ -1662,7 +1661,8 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( alloc_flags |= (flags & KFD_IOC_ALLOC_MEM_FLAGS_PUBLIC) ? AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED : 0; } - xcp_id = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id; + xcp_id = fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION ? + 0 : fpriv->xcp_id; } else if (flags & KFD_IOC_ALLOC_MEM_FLAGS_GTT) { domain = alloc_domain = AMDGPU_GEM_DOMAIN_GTT; alloc_flags = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c index f4e3c133a16c..73ee14f7a9a4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c @@ -1776,7 +1776,7 @@ static ssize_t amdgpu_atombios_get_vbios_version(struct device *dev, struct amdgpu_device *adev = drm_to_adev(ddev); struct atom_context *ctx = adev->mode_info.atom_context; - return sysfs_emit(buf, "%s\n", ctx->vbios_version); + return sysfs_emit(buf, "%s\n", ctx->vbios_ver_str); } static DEVICE_ATTR(vbios_version, 0444, amdgpu_atombios_get_vbios_version, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h index b639a80ee3fc..0811474e8fd3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h @@ -89,8 +89,7 @@ struct atom_memory_info { #define MAX_AC_TIMING_ENTRIES 16 -struct atom_memory_clock_range_table -{ +struct atom_memory_clock_range_table { u8 num_entries; u8 rsv[3]; u32 mclk[MAX_AC_TIMING_ENTRIES]; @@ -118,14 +117,12 @@ struct atom_mc_reg_table { #define MAX_VOLTAGE_ENTRIES 32 -struct atom_voltage_table_entry -{ +struct atom_voltage_table_entry { u16 value; u32 smio_low; }; -struct atom_voltage_table -{ +struct atom_voltage_table { u32 count; u32 mask_low; u32 phase_delay; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index 0b7f4c4d58e5..fb2681dd6b33 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -58,7 +58,7 @@ uint32_t amdgpu_atomfirmware_query_firmware_capability(struct amdgpu_device *ade if (amdgpu_atom_parse_data_header(adev->mode_info.atom_context, index, &size, &frev, &crev, &data_offset)) { /* support firmware_info 3.1 + */ - if ((frev == 3 && crev >=1) || (frev > 3)) { + if ((frev == 3 && crev >= 1) || (frev > 3)) { firmware_info = (union firmware_info *) (mode_info->atom_context->bios + data_offset); fw_cap = le32_to_cpu(firmware_info->v31.firmware_capability); @@ -217,6 +217,7 @@ union umc_info { struct atom_umc_info_v3_1 v31; struct atom_umc_info_v3_2 v32; struct atom_umc_info_v3_3 v33; + struct atom_umc_info_v4_0 v40; }; union vram_info { @@ -508,9 +509,8 @@ bool amdgpu_atomfirmware_mem_ecc_supported(struct amdgpu_device *adev) if (amdgpu_atom_parse_data_header(mode_info->atom_context, index, &size, &frev, &crev, &data_offset)) { + umc_info = (union umc_info *)(mode_info->atom_context->bios + data_offset); if (frev == 3) { - umc_info = (union umc_info *) - (mode_info->atom_context->bios + data_offset); switch (crev) { case 1: umc_config = le32_to_cpu(umc_info->v31.umc_config); @@ -533,6 +533,20 @@ bool amdgpu_atomfirmware_mem_ecc_supported(struct amdgpu_device *adev) /* unsupported crev */ return false; } + } else if (frev == 4) { + switch (crev) { + case 0: + umc_config1 = le32_to_cpu(umc_info->v40.umc_config1); + ecc_default_enabled = + (umc_config1 & UMC_CONFIG1__ENABLE_ECC_CAPABLE) ? true : false; + break; + default: + /* unsupported crev */ + return false; + } + } else { + /* unsupported frev */ + return false; } } @@ -597,7 +611,7 @@ bool amdgpu_atomfirmware_ras_rom_addr(struct amdgpu_device *adev, index, &size, &frev, &crev, &data_offset)) { /* support firmware_info 3.4 + */ - if ((frev == 3 && crev >=4) || (frev > 3)) { + if ((frev == 3 && crev >= 4) || (frev > 3)) { firmware_info = (union firmware_info *) (mode_info->atom_context->bios + data_offset); /* The ras_rom_i2c_slave_addr should ideally @@ -850,7 +864,7 @@ int amdgpu_atomfirmware_get_fw_reserved_fb_size(struct amdgpu_device *adev) firmware_info = (union firmware_info *)(ctx->bios + data_offset); - if (frev !=3) + if (frev != 3) return -EINVAL; switch (crev) { @@ -909,7 +923,7 @@ int amdgpu_atomfirmware_asic_init(struct amdgpu_device *adev, bool fb_reset) } index = get_index_into_master_table(atom_master_list_of_command_functions_v2_1, - asic_init); + asic_init); if (amdgpu_atom_parse_cmd_header(mode_info->atom_context, index, &frev, &crev)) { if (frev == 2 && crev >= 1) { memset(&asic_init_ps_v2_1, 0, sizeof(asic_init_ps_v2_1)); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index d6d986be906a..375f02002579 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -74,24 +74,29 @@ struct atpx_mux { u16 mux; } __packed; -bool amdgpu_has_atpx(void) { +bool amdgpu_has_atpx(void) +{ return amdgpu_atpx_priv.atpx_detected; } -bool amdgpu_has_atpx_dgpu_power_cntl(void) { +bool amdgpu_has_atpx_dgpu_power_cntl(void) +{ return amdgpu_atpx_priv.atpx.functions.power_cntl; } -bool amdgpu_is_atpx_hybrid(void) { +bool amdgpu_is_atpx_hybrid(void) +{ return amdgpu_atpx_priv.atpx.is_hybrid; } -bool amdgpu_atpx_dgpu_req_power_for_displays(void) { +bool amdgpu_atpx_dgpu_req_power_for_displays(void) +{ return amdgpu_atpx_priv.atpx.dgpu_req_power_for_displays; } #if defined(CONFIG_ACPI) -void *amdgpu_atpx_get_dhandle(void) { +void *amdgpu_atpx_get_dhandle(void) +{ return amdgpu_atpx_priv.dhandle; } #endif @@ -134,7 +139,7 @@ static union acpi_object *amdgpu_atpx_call(acpi_handle handle, int function, /* Fail only if calling the method fails and ATPX is supported */ if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) { - printk("failed to evaluate ATPX got %s\n", + pr_err("failed to evaluate ATPX got %s\n", acpi_format_exception(status)); kfree(buffer.pointer); return NULL; @@ -190,7 +195,7 @@ static int amdgpu_atpx_validate(struct amdgpu_atpx *atpx) size = *(u16 *) info->buffer.pointer; if (size < 10) { - printk("ATPX buffer is too small: %zu\n", size); + pr_err("ATPX buffer is too small: %zu\n", size); kfree(info); return -EINVAL; } @@ -223,11 +228,11 @@ static int amdgpu_atpx_validate(struct amdgpu_atpx *atpx) atpx->is_hybrid = false; if (valid_bits & ATPX_MS_HYBRID_GFX_SUPPORTED) { if (amdgpu_atpx_priv.quirks & AMDGPU_PX_QUIRK_FORCE_ATPX) { - printk("ATPX Hybrid Graphics, forcing to ATPX\n"); + pr_warn("ATPX Hybrid Graphics, forcing to ATPX\n"); atpx->functions.power_cntl = true; atpx->is_hybrid = false; } else { - printk("ATPX Hybrid Graphics\n"); + pr_notice("ATPX Hybrid Graphics\n"); /* * Disable legacy PM methods only when pcie port PM is usable, * otherwise the device might fail to power off or power on. @@ -269,7 +274,7 @@ static int amdgpu_atpx_verify_interface(struct amdgpu_atpx *atpx) size = *(u16 *) info->buffer.pointer; if (size < 8) { - printk("ATPX buffer is too small: %zu\n", size); + pr_err("ATPX buffer is too small: %zu\n", size); err = -EINVAL; goto out; } @@ -278,8 +283,8 @@ static int amdgpu_atpx_verify_interface(struct amdgpu_atpx *atpx) memcpy(&output, info->buffer.pointer, size); /* TODO: check version? */ - printk("ATPX version %u, functions 0x%08x\n", - output.version, output.function_bits); + pr_notice("ATPX version %u, functions 0x%08x\n", + output.version, output.function_bits); amdgpu_atpx_parse_functions(&atpx->functions, output.function_bits); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c index b582b83c4984..38ccec913f00 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bios.c @@ -460,7 +460,7 @@ bool amdgpu_get_bios(struct amdgpu_device *adev) return false; success: - adev->is_atom_fw = (adev->asic_type >= CHIP_VEGA10) ? true : false; + adev->is_atom_fw = adev->asic_type >= CHIP_VEGA10; return true; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 456e385333b6..b8280be6225d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -41,13 +41,13 @@ struct amdgpu_cgs_device { ((struct amdgpu_cgs_device *)cgs_device)->adev -static uint32_t amdgpu_cgs_read_register(struct cgs_device *cgs_device, unsigned offset) +static uint32_t amdgpu_cgs_read_register(struct cgs_device *cgs_device, unsigned int offset) { CGS_FUNC_ADEV; return RREG32(offset); } -static void amdgpu_cgs_write_register(struct cgs_device *cgs_device, unsigned offset, +static void amdgpu_cgs_write_register(struct cgs_device *cgs_device, unsigned int offset, uint32_t value) { CGS_FUNC_ADEV; @@ -56,7 +56,7 @@ static void amdgpu_cgs_write_register(struct cgs_device *cgs_device, unsigned of static uint32_t amdgpu_cgs_read_ind_register(struct cgs_device *cgs_device, enum cgs_ind_reg space, - unsigned index) + unsigned int index) { CGS_FUNC_ADEV; switch (space) { @@ -84,7 +84,7 @@ static uint32_t amdgpu_cgs_read_ind_register(struct cgs_device *cgs_device, static void amdgpu_cgs_write_ind_register(struct cgs_device *cgs_device, enum cgs_ind_reg space, - unsigned index, uint32_t value) + unsigned int index, uint32_t value) { CGS_FUNC_ADEV; switch (space) { @@ -163,38 +163,38 @@ static uint16_t amdgpu_get_firmware_version(struct cgs_device *cgs_device, uint16_t fw_version = 0; switch (type) { - case CGS_UCODE_ID_SDMA0: - fw_version = adev->sdma.instance[0].fw_version; - break; - case CGS_UCODE_ID_SDMA1: - fw_version = adev->sdma.instance[1].fw_version; - break; - case CGS_UCODE_ID_CP_CE: - fw_version = adev->gfx.ce_fw_version; - break; - case CGS_UCODE_ID_CP_PFP: - fw_version = adev->gfx.pfp_fw_version; - break; - case CGS_UCODE_ID_CP_ME: - fw_version = adev->gfx.me_fw_version; - break; - case CGS_UCODE_ID_CP_MEC: - fw_version = adev->gfx.mec_fw_version; - break; - case CGS_UCODE_ID_CP_MEC_JT1: - fw_version = adev->gfx.mec_fw_version; - break; - case CGS_UCODE_ID_CP_MEC_JT2: - fw_version = adev->gfx.mec_fw_version; - break; - case CGS_UCODE_ID_RLC_G: - fw_version = adev->gfx.rlc_fw_version; - break; - case CGS_UCODE_ID_STORAGE: - break; - default: - DRM_ERROR("firmware type %d do not have version\n", type); - break; + case CGS_UCODE_ID_SDMA0: + fw_version = adev->sdma.instance[0].fw_version; + break; + case CGS_UCODE_ID_SDMA1: + fw_version = adev->sdma.instance[1].fw_version; + break; + case CGS_UCODE_ID_CP_CE: + fw_version = adev->gfx.ce_fw_version; + break; + case CGS_UCODE_ID_CP_PFP: + fw_version = adev->gfx.pfp_fw_version; + break; + case CGS_UCODE_ID_CP_ME: + fw_version = adev->gfx.me_fw_version; + break; + case CGS_UCODE_ID_CP_MEC: + fw_version = adev->gfx.mec_fw_version; + break; + case CGS_UCODE_ID_CP_MEC_JT1: + fw_version = adev->gfx.mec_fw_version; + break; + case CGS_UCODE_ID_CP_MEC_JT2: + fw_version = adev->gfx.mec_fw_version; + break; + case CGS_UCODE_ID_RLC_G: + fw_version = adev->gfx.rlc_fw_version; + break; + case CGS_UCODE_ID_STORAGE: + break; + default: + DRM_ERROR("firmware type %d do not have version\n", type); + break; } return fw_version; } @@ -205,7 +205,7 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, { CGS_FUNC_ADEV; - if ((CGS_UCODE_ID_SMU != type) && (CGS_UCODE_ID_SMU_SK != type)) { + if (type != CGS_UCODE_ID_SMU && type != CGS_UCODE_ID_SMU_SK) { uint64_t gpu_addr; uint32_t data_size; const struct gfx_firmware_header_v1_0 *header; @@ -232,7 +232,7 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, info->mc_addr = gpu_addr; info->version = (uint16_t)le32_to_cpu(header->header.ucode_version); - if (CGS_UCODE_ID_CP_MEC == type) + if (type == CGS_UCODE_ID_CP_MEC) info->image_size = le32_to_cpu(header->jt_offset) << 2; info->fw_version = amdgpu_get_firmware_version(cgs_device, type); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 977e1804718d..efdb1c48f431 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -127,7 +127,6 @@ static int amdgpu_cs_p1_user_fence(struct amdgpu_cs_parser *p, { struct drm_gem_object *gobj; unsigned long size; - int r; gobj = drm_gem_object_lookup(p->filp, data->handle); if (gobj == NULL) @@ -137,23 +136,14 @@ static int amdgpu_cs_p1_user_fence(struct amdgpu_cs_parser *p, drm_gem_object_put(gobj); size = amdgpu_bo_size(p->uf_bo); - if (size != PAGE_SIZE || (data->offset + 8) > size) { - r = -EINVAL; - goto error_unref; - } + if (size != PAGE_SIZE || data->offset > (size - 8)) + return -EINVAL; - if (amdgpu_ttm_tt_get_usermm(p->uf_bo->tbo.ttm)) { - r = -EINVAL; - goto error_unref; - } + if (amdgpu_ttm_tt_get_usermm(p->uf_bo->tbo.ttm)) + return -EINVAL; *offset = data->offset; - return 0; - -error_unref: - amdgpu_bo_unref(&p->uf_bo); - return r; } static int amdgpu_cs_p1_bo_handles(struct amdgpu_cs_parser *p, @@ -293,7 +283,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p, if (!p->gang_size) { ret = -EINVAL; - goto free_partial_kdata; + goto free_all_kdata; } for (i = 0; i < p->gang_size; ++i) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 56e89e76ff17..a4faea4fa0b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -154,7 +154,7 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, } else { r = get_user(value, (uint32_t *)buf); if (!r) - amdgpu_mm_wreg_mmio_rlc(adev, *pos >> 2, value); + amdgpu_mm_wreg_mmio_rlc(adev, *pos >> 2, value, 0); } if (r) { result = r; @@ -283,7 +283,7 @@ static ssize_t amdgpu_debugfs_regs2_op(struct file *f, char __user *buf, u32 off } else { r = get_user(value, (uint32_t *)buf); if (!r) - amdgpu_mm_wreg_mmio_rlc(adev, offset >> 2, value); + amdgpu_mm_wreg_mmio_rlc(adev, offset >> 2, value, rd->id.xcc_id); } if (r) { result = r; @@ -375,7 +375,7 @@ static int amdgpu_debugfs_gprwave_open(struct inode *inode, struct file *file) { struct amdgpu_debugfs_gprwave_data *rd; - rd = kzalloc(sizeof *rd, GFP_KERNEL); + rd = kzalloc(sizeof(*rd), GFP_KERNEL); if (!rd) return -ENOMEM; rd->adev = file_inode(file)->i_private; @@ -388,6 +388,7 @@ static int amdgpu_debugfs_gprwave_open(struct inode *inode, struct file *file) static int amdgpu_debugfs_gprwave_release(struct inode *inode, struct file *file) { struct amdgpu_debugfs_gprwave_data *rd = file->private_data; + mutex_destroy(&rd->lock); kfree(file->private_data); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index a2cdde0ca0a7..30c4f5cca02c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -159,76 +159,11 @@ static ssize_t amdgpu_device_get_pcie_replay_count(struct device *dev, return sysfs_emit(buf, "%llu\n", cnt); } -static DEVICE_ATTR(pcie_replay_count, S_IRUGO, +static DEVICE_ATTR(pcie_replay_count, 0444, amdgpu_device_get_pcie_replay_count, NULL); static void amdgpu_device_get_pcie_info(struct amdgpu_device *adev); -/** - * DOC: product_name - * - * The amdgpu driver provides a sysfs API for reporting the product name - * for the device - * The file product_name is used for this and returns the product name - * as returned from the FRU. - * NOTE: This is only available for certain server cards - */ - -static ssize_t amdgpu_device_get_product_name(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct drm_device *ddev = dev_get_drvdata(dev); - struct amdgpu_device *adev = drm_to_adev(ddev); - - return sysfs_emit(buf, "%s\n", adev->product_name); -} - -static DEVICE_ATTR(product_name, S_IRUGO, - amdgpu_device_get_product_name, NULL); - -/** - * DOC: product_number - * - * The amdgpu driver provides a sysfs API for reporting the part number - * for the device - * The file product_number is used for this and returns the part number - * as returned from the FRU. - * NOTE: This is only available for certain server cards - */ - -static ssize_t amdgpu_device_get_product_number(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct drm_device *ddev = dev_get_drvdata(dev); - struct amdgpu_device *adev = drm_to_adev(ddev); - - return sysfs_emit(buf, "%s\n", adev->product_number); -} - -static DEVICE_ATTR(product_number, S_IRUGO, - amdgpu_device_get_product_number, NULL); - -/** - * DOC: serial_number - * - * The amdgpu driver provides a sysfs API for reporting the serial number - * for the device - * The file serial_number is used for this and returns the serial number - * as returned from the FRU. - * NOTE: This is only available for certain server cards - */ - -static ssize_t amdgpu_device_get_serial_number(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct drm_device *ddev = dev_get_drvdata(dev); - struct amdgpu_device *adev = drm_to_adev(ddev); - - return sysfs_emit(buf, "%s\n", adev->serial); -} - -static DEVICE_ATTR(serial_number, S_IRUGO, - amdgpu_device_get_serial_number, NULL); /** * amdgpu_device_supports_px - Is the device a dGPU with ATPX power control @@ -370,10 +305,16 @@ size_t amdgpu_device_aper_access(struct amdgpu_device *adev, loff_t pos, if (write) { memcpy_toio(addr, buf, count); + /* Make sure HDP write cache flush happens without any reordering + * after the system memory contents are sent over PCIe device + */ mb(); amdgpu_device_flush_hdp(adev, NULL); } else { amdgpu_device_invalidate_hdp(adev, NULL); + /* Make sure HDP read cache is invalidated before issuing a read + * to the PCIe device + */ mb(); memcpy_fromio(buf, addr, count); } @@ -481,8 +422,7 @@ uint32_t amdgpu_device_rreg(struct amdgpu_device *adev, /* * MMIO register read with bytes helper functions * @offset:bytes offset from MMIO start - * -*/ + */ /** * amdgpu_mm_rreg8 - read a memory mapped IO register @@ -506,8 +446,8 @@ uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset) * MMIO register write with bytes helper functions * @offset:bytes offset from MMIO start * @value: the value want to be written to the register - * -*/ + */ + /** * amdgpu_mm_wreg8 - read a memory mapped IO register * @@ -571,7 +511,8 @@ void amdgpu_device_wreg(struct amdgpu_device *adev, * this function is invoked only for the debugfs register access */ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, - uint32_t reg, uint32_t v) + uint32_t reg, uint32_t v, + uint32_t xcc_id) { if (amdgpu_device_skip_hw_access(adev)) return; @@ -580,7 +521,7 @@ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, adev->gfx.rlc.funcs && adev->gfx.rlc.funcs->is_rlcg_access_range) { if (adev->gfx.rlc.funcs->is_rlcg_access_range(adev, reg)) - return amdgpu_sriov_wreg(adev, reg, v, 0, 0); + return amdgpu_sriov_wreg(adev, reg, v, 0, 0, xcc_id); } else if ((reg * 4) >= adev->rmmio_size) { adev->pcie_wreg(adev, reg * 4, v); } else { @@ -589,94 +530,6 @@ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, } /** - * amdgpu_mm_rdoorbell - read a doorbell dword - * - * @adev: amdgpu_device pointer - * @index: doorbell index - * - * Returns the value in the doorbell aperture at the - * requested doorbell index (CIK). - */ -u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index) -{ - if (amdgpu_device_skip_hw_access(adev)) - return 0; - - if (index < adev->doorbell.num_kernel_doorbells) { - return readl(adev->doorbell.ptr + index); - } else { - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); - return 0; - } -} - -/** - * amdgpu_mm_wdoorbell - write a doorbell dword - * - * @adev: amdgpu_device pointer - * @index: doorbell index - * @v: value to write - * - * Writes @v to the doorbell aperture at the - * requested doorbell index (CIK). - */ -void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v) -{ - if (amdgpu_device_skip_hw_access(adev)) - return; - - if (index < adev->doorbell.num_kernel_doorbells) { - writel(v, adev->doorbell.ptr + index); - } else { - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); - } -} - -/** - * amdgpu_mm_rdoorbell64 - read a doorbell Qword - * - * @adev: amdgpu_device pointer - * @index: doorbell index - * - * Returns the value in the doorbell aperture at the - * requested doorbell index (VEGA10+). - */ -u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index) -{ - if (amdgpu_device_skip_hw_access(adev)) - return 0; - - if (index < adev->doorbell.num_kernel_doorbells) { - return atomic64_read((atomic64_t *)(adev->doorbell.ptr + index)); - } else { - DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); - return 0; - } -} - -/** - * amdgpu_mm_wdoorbell64 - write a doorbell Qword - * - * @adev: amdgpu_device pointer - * @index: doorbell index - * @v: value to write - * - * Writes @v to the doorbell aperture at the - * requested doorbell index (VEGA10+). - */ -void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v) -{ - if (amdgpu_device_skip_hw_access(adev)) - return; - - if (index < adev->doorbell.num_kernel_doorbells) { - atomic64_set((atomic64_t *)(adev->doorbell.ptr + index), v); - } else { - DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); - } -} - -/** * amdgpu_device_indirect_rreg - read an indirect register * * @adev: amdgpu_device pointer @@ -1032,13 +885,20 @@ static void amdgpu_block_invalid_wreg(struct amdgpu_device *adev, */ static int amdgpu_device_asic_init(struct amdgpu_device *adev) { + int ret; + amdgpu_asic_pre_asic_init(adev); if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3) || - adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0)) - return amdgpu_atomfirmware_asic_init(adev, true); - else + adev->ip_versions[GC_HWIP][0] >= IP_VERSION(11, 0, 0)) { + amdgpu_psp_wait_for_bootloader(adev); + ret = amdgpu_atomfirmware_asic_init(adev, true); + return ret; + } else { return amdgpu_atom_asic_init(adev->mode_info.atom_context); + } + + return 0; } /** @@ -1078,7 +938,7 @@ static void amdgpu_device_mem_scratch_fini(struct amdgpu_device *adev) * @registers: pointer to the register array * @array_size: size of the register array * - * Programs an array or registers with and and or masks. + * Programs an array or registers with and or masks. * This is a helper for setting golden registers. */ void amdgpu_device_program_register_sequence(struct amdgpu_device *adev, @@ -1136,83 +996,6 @@ int amdgpu_device_pci_reset(struct amdgpu_device *adev) } /* - * GPU doorbell aperture helpers function. - */ -/** - * amdgpu_device_doorbell_init - Init doorbell driver information. - * - * @adev: amdgpu_device pointer - * - * Init doorbell driver information (CIK) - * Returns 0 on success, error on failure. - */ -static int amdgpu_device_doorbell_init(struct amdgpu_device *adev) -{ - - /* No doorbell on SI hardware generation */ - if (adev->asic_type < CHIP_BONAIRE) { - adev->doorbell.base = 0; - adev->doorbell.size = 0; - adev->doorbell.num_kernel_doorbells = 0; - adev->doorbell.ptr = NULL; - return 0; - } - - if (pci_resource_flags(adev->pdev, 2) & IORESOURCE_UNSET) - return -EINVAL; - - amdgpu_asic_init_doorbell_index(adev); - - /* doorbell bar mapping */ - adev->doorbell.base = pci_resource_start(adev->pdev, 2); - adev->doorbell.size = pci_resource_len(adev->pdev, 2); - - if (adev->enable_mes) { - adev->doorbell.num_kernel_doorbells = - adev->doorbell.size / sizeof(u32); - } else { - adev->doorbell.num_kernel_doorbells = - min_t(u32, adev->doorbell.size / sizeof(u32), - adev->doorbell_index.max_assignment+1); - if (adev->doorbell.num_kernel_doorbells == 0) - return -EINVAL; - - /* For Vega, reserve and map two pages on doorbell BAR since SDMA - * paging queue doorbell use the second page. The - * AMDGPU_DOORBELL64_MAX_ASSIGNMENT definition assumes all the - * doorbells are in the first page. So with paging queue enabled, - * the max num_kernel_doorbells should + 1 page (0x400 in dword) - */ - if (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(4, 0, 0) && - adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(4, 2, 0)) - adev->doorbell.num_kernel_doorbells += 0x400; - } - - adev->doorbell.ptr = ioremap(adev->doorbell.base, - adev->doorbell.num_kernel_doorbells * - sizeof(u32)); - if (adev->doorbell.ptr == NULL) - return -ENOMEM; - - return 0; -} - -/** - * amdgpu_device_doorbell_fini - Tear down doorbell driver information. - * - * @adev: amdgpu_device pointer - * - * Tear down doorbell driver information (CIK) - */ -static void amdgpu_device_doorbell_fini(struct amdgpu_device *adev) -{ - iounmap(adev->doorbell.ptr); - adev->doorbell.ptr = NULL; -} - - - -/* * amdgpu_device_wb_*() * Writeback is the method by which the GPU updates special pages in memory * with the status of certain GPU events (fences, ring pointers,etc.). @@ -1321,10 +1104,13 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) int rbar_size = pci_rebar_bytes_to_size(adev->gmc.real_vram_size); struct pci_bus *root; struct resource *res; - unsigned i; + unsigned int i; u16 cmd; int r; + if (!IS_ENABLED(CONFIG_PHYS_ADDR_T_64BIT)) + return 0; + /* Bypass for VF */ if (amdgpu_sriov_vf(adev)) return 0; @@ -1359,7 +1145,7 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) cmd & ~PCI_COMMAND_MEMORY); /* Free the VRAM and doorbell BAR, we most likely need to move both. */ - amdgpu_device_doorbell_fini(adev); + amdgpu_doorbell_fini(adev); if (adev->asic_type >= CHIP_BONAIRE) pci_release_resource(adev->pdev, 2); @@ -1376,7 +1162,7 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) /* When the doorbell or fb BAR isn't available we have no chance of * using the device. */ - r = amdgpu_device_doorbell_init(adev); + r = amdgpu_doorbell_init(adev); if (r || (pci_resource_flags(adev->pdev, 0) & IORESOURCE_UNSET)) return -ENODEV; @@ -1387,9 +1173,8 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) static bool amdgpu_device_read_bios(struct amdgpu_device *adev) { - if (hweight32(adev->aid_mask) && (adev->flags & AMD_IS_APU)) { + if (hweight32(adev->aid_mask) && (adev->flags & AMD_IS_APU)) return false; - } return true; } @@ -1425,6 +1210,7 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev) if (adev->asic_type == CHIP_FIJI) { int err; uint32_t fw_ver; + err = request_firmware(&adev->pm.fw, "amdgpu/fiji_smc.bin", adev->dev); /* force vPost if error occured */ if (err) @@ -1527,6 +1313,7 @@ static unsigned int amdgpu_device_vga_set_decode(struct pci_dev *pdev, bool state) { struct amdgpu_device *adev = drm_to_adev(pci_get_drvdata(pdev)); + amdgpu_asic_set_vga_state(adev, state); if (state) return VGA_RSRC_LEGACY_IO | VGA_RSRC_LEGACY_MEM | @@ -1549,7 +1336,8 @@ static void amdgpu_device_check_block_size(struct amdgpu_device *adev) { /* defines number of bits in page table versus page directory, * a page is 4KB so we have 12 bits offset, minimum 9 bits in the - * page table and the remaining bits are in the page directory */ + * page table and the remaining bits are in the page directory + */ if (amdgpu_vm_block_size == -1) return; @@ -1781,7 +1569,7 @@ static bool amdgpu_switcheroo_can_switch(struct pci_dev *pdev) { struct drm_device *dev = pci_get_drvdata(pdev); - /* + /* * FIXME: open_count is protected by drm_global_mutex but that would lead to * locking inversion with the driver load path. And the access here is * completely racy anyway. So don't bother with locking for now. @@ -3426,7 +3214,7 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev) * * Main resume function for hardware IPs. The hardware IPs * are split into two resume functions because they are - * are also used in in recovering from a GPU reset and some additional + * also used in recovering from a GPU reset and some additional * steps need to be take between them. In this case (S3/S4) they are * run sequentially. * Returns 0 on success, negative error code on failure. @@ -3435,12 +3223,6 @@ static int amdgpu_device_ip_resume(struct amdgpu_device *adev) { int r; - if (!adev->in_s0ix) { - r = amdgpu_amdkfd_resume_iommu(adev); - if (r) - return r; - } - r = amdgpu_device_ip_resume_phase1(adev); if (r) return r; @@ -3528,8 +3310,7 @@ bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type) #else default: if (amdgpu_dc > 0) - DRM_INFO_ONCE("Display Core has been requested via kernel parameter " - "but isn't supported by ASIC, ignoring\n"); + DRM_INFO_ONCE("Display Core has been requested via kernel parameter but isn't supported by ASIC, ignoring\n"); return false; #endif } @@ -3685,9 +3466,6 @@ static void amdgpu_device_check_iommu_direct_map(struct amdgpu_device *adev) } static const struct attribute *amdgpu_dev_attributes[] = { - &dev_attr_product_name.attr, - &dev_attr_product_number.attr, - &dev_attr_serial_number.attr, &dev_attr_pcie_replay_count.attr, NULL }; @@ -3696,10 +3474,11 @@ static void amdgpu_device_set_mcbp(struct amdgpu_device *adev) { if (amdgpu_mcbp == 1) adev->gfx.mcbp = true; - - if ((adev->ip_versions[GC_HWIP][0] >= IP_VERSION(9, 0, 0)) && - (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 0, 0)) && - adev->gfx.num_gfx_rings) + else if (amdgpu_mcbp == 0) + adev->gfx.mcbp = false; + else if ((adev->ip_versions[GC_HWIP][0] >= IP_VERSION(9, 0, 0)) && + (adev->ip_versions[GC_HWIP][0] < IP_VERSION(10, 0, 0)) && + adev->gfx.num_gfx_rings) adev->gfx.mcbp = true; if (amdgpu_sriov_vf(adev)) @@ -3777,7 +3556,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, pdev->subsystem_vendor, pdev->subsystem_device, pdev->revision); /* mutex initialization are all done here so we - * can recall function without having locking issues */ + * can recall function without having locking issues + */ mutex_init(&adev->firmware.mutex); mutex_init(&adev->pm.mutex); mutex_init(&adev->gfx.gpu_clock_mutex); @@ -3854,11 +3634,11 @@ int amdgpu_device_init(struct amdgpu_device *adev, atomic_set(&adev->pm.pwr_state[i], POWER_STATE_UNKNOWN); adev->rmmio = ioremap(adev->rmmio_base, adev->rmmio_size); - if (adev->rmmio == NULL) { + if (!adev->rmmio) return -ENOMEM; - } + DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base); - DRM_INFO("register mmio size: %u\n", (unsigned)adev->rmmio_size); + DRM_INFO("register mmio size: %u\n", (unsigned int)adev->rmmio_size); /* * Reset domain needs to be present early, before XGMI hive discovered @@ -3926,7 +3706,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, dev_info(adev->dev, "PCIE atomic ops is not supported\n"); /* doorbell bar mapping and doorbell index init*/ - amdgpu_device_doorbell_init(adev); + amdgpu_doorbell_init(adev); if (amdgpu_emu_mode == 1) { /* post the asic on emulation mode */ @@ -4069,14 +3849,6 @@ fence_driver_init: } else adev->ucode_sysfs_en = true; - r = amdgpu_psp_sysfs_init(adev); - if (r) { - adev->psp_sysfs_en = false; - if (!amdgpu_sriov_vf(adev)) - DRM_ERROR("Creating psp sysfs failed\n"); - } else - adev->psp_sysfs_en = true; - /* * Register gpu instance before amdgpu_device_enable_mgpu_fan_boost. * Otherwise the mgpu fan boost feature will be skipped due to the @@ -4109,6 +3881,8 @@ fence_driver_init: if (r) dev_err(adev->dev, "Could not create amdgpu device attr\n"); + amdgpu_fru_sysfs_init(adev); + if (IS_ENABLED(CONFIG_PERF_EVENTS)) r = amdgpu_pmu_init(adev); if (r) @@ -4120,7 +3894,8 @@ fence_driver_init: /* if we have > 1 VGA cards, then disable the amdgpu VGA resources */ /* this will fail for cards that aren't VGA class devices, just - * ignore it */ + * ignore it + */ if ((adev->pdev->class >> 8) == PCI_CLASS_DISPLAY_VGA) vga_client_register(adev->pdev, amdgpu_device_vga_set_decode); @@ -4172,7 +3947,7 @@ static void amdgpu_device_unmap_mmio(struct amdgpu_device *adev) unmap_mapping_range(adev->ddev.anon_inode->i_mapping, 0, 0, 1); /* Unmap all mapped bars - Doorbell, registers and VRAM */ - amdgpu_device_doorbell_fini(adev); + amdgpu_doorbell_fini(adev); iounmap(adev->rmmio); adev->rmmio = NULL; @@ -4203,7 +3978,7 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) /* make sure IB test finished before entering exclusive mode * to avoid preemption on IB test - * */ + */ if (amdgpu_sriov_vf(adev)) { amdgpu_virt_request_full_gpu(adev, false); amdgpu_virt_fini_data_exchange(adev); @@ -4226,9 +4001,8 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) amdgpu_pm_sysfs_fini(adev); if (adev->ucode_sysfs_en) amdgpu_ucode_sysfs_fini(adev); - if (adev->psp_sysfs_en) - amdgpu_psp_sysfs_fini(adev); sysfs_remove_files(&adev->dev->kobj, amdgpu_dev_attributes); + amdgpu_fru_sysfs_fini(adev); /* disable ras feature must before hw fini */ amdgpu_ras_pre_fini(adev); @@ -4286,7 +4060,7 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev) iounmap(adev->rmmio); adev->rmmio = NULL; - amdgpu_device_doorbell_fini(adev); + amdgpu_doorbell_fini(adev); drm_dev_exit(idx); } @@ -4367,6 +4141,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, true); cancel_delayed_work_sync(&adev->delayed_init_work); + flush_delayed_work(&adev->gfx.gfx_off_delay_work); amdgpu_ras_suspend(adev); @@ -4745,6 +4520,10 @@ retry: r = amdgpu_virt_reset_gpu(adev); if (r) return r; + amdgpu_irq_gpu_reset_resume_helper(adev); + + /* some sw clean up VF needs to do before recover */ + amdgpu_virt_post_reset(adev); /* Resume IP prior to SMC */ r = amdgpu_device_ip_reinit_early_sriov(adev); @@ -4771,7 +4550,6 @@ retry: amdgpu_put_xgmi_hive(hive); if (!r) { - amdgpu_irq_gpu_reset_resume_helper(adev); r = amdgpu_ib_ring_tests(adev); amdgpu_amdkfd_post_reset(adev); @@ -4897,9 +4675,12 @@ int amdgpu_device_mode1_reset(struct amdgpu_device *adev) } if (ret) - dev_err(adev->dev, "GPU mode1 reset failed\n"); + goto mode1_reset_failed; amdgpu_device_load_pci_state(adev->pdev); + ret = amdgpu_psp_wait_for_bootloader(adev); + if (ret) + goto mode1_reset_failed; /* wait for asic to come out of reset */ for (i = 0; i < adev->usec_timeout; i++) { @@ -4910,7 +4691,17 @@ int amdgpu_device_mode1_reset(struct amdgpu_device *adev) udelay(1); } + if (i >= adev->usec_timeout) { + ret = -ETIMEDOUT; + goto mode1_reset_failed; + } + amdgpu_atombios_scratch_regs_engine_hung(adev, false); + + return 0; + +mode1_reset_failed: + dev_err(adev->dev, "GPU mode1 reset failed\n"); return ret; } @@ -4939,8 +4730,9 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, if (!ring || !ring->sched.thread) continue; - /*clear job fence from fence drv to avoid force_completion - *leave NULL and vm flush fence in fence drv */ + /* Clear job fence from fence drv to avoid force_completion + * leave NULL and vm flush fence in fence drv + */ amdgpu_fence_driver_clear_job_fences(ring); /* after all hw jobs are reset, hw fence is meaningless, so force_completion */ @@ -4954,7 +4746,7 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, r = amdgpu_reset_prepare_hwcontext(adev, reset_context); /* If reset handler not implemented, continue; otherwise return */ - if (r == -ENOSYS) + if (r == -EOPNOTSUPP) r = 0; else return r; @@ -5051,7 +4843,7 @@ static void amdgpu_reset_capture_coredumpm(struct amdgpu_device *adev) struct drm_device *dev = adev_to_drm(adev); ktime_get_ts64(&adev->reset_time); - dev_coredumpm(dev->dev, THIS_MODULE, adev, 0, GFP_KERNEL, + dev_coredumpm(dev->dev, THIS_MODULE, adev, 0, GFP_NOWAIT, amdgpu_devcoredump_read, amdgpu_devcoredump_free); } #endif @@ -5072,7 +4864,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, reset_context->reset_device_list = device_list_handle; r = amdgpu_reset_perform_reset(tmp_adev, reset_context); /* If reset handler not implemented, continue; otherwise return */ - if (r == -ENOSYS) + if (r == -EOPNOTSUPP) r = 0; else return r; @@ -5150,9 +4942,6 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, dev_warn(tmp_adev->dev, "asic atom init failed!"); } else { dev_info(tmp_adev->dev, "GPU reset succeeded, trying to resume\n"); - r = amdgpu_amdkfd_resume_iommu(tmp_adev); - if (r) - goto out; r = amdgpu_device_ip_resume_phase1(tmp_adev); if (r) @@ -5561,9 +5350,8 @@ skip_hw_reset: if (adev->enable_mes && adev->ip_versions[GC_HWIP][0] != IP_VERSION(11, 0, 3)) amdgpu_mes_self_test(tmp_adev); - if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled) { + if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled) drm_helper_resume_force_mode(adev_to_drm(tmp_adev)); - } if (tmp_adev->asic_reset_res) r = tmp_adev->asic_reset_res; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 8e1cfc87122d..7d5e7ad28ba8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -65,6 +65,7 @@ #include "soc21.h" #include "navi10_ih.h" #include "ih_v6_0.h" +#include "ih_v6_1.h" #include "gfx_v10_0.h" #include "gfx_v11_0.h" #include "sdma_v5_0.h" @@ -1389,6 +1390,7 @@ union gc_info { struct gc_info_v1_1 v1_1; struct gc_info_v1_2 v1_2; struct gc_info_v2_0 v2; + struct gc_info_v2_1 v2_1; }; static int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev) @@ -1464,6 +1466,15 @@ static int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev) adev->gfx.config.num_sc_per_sh = le32_to_cpu(gc_info->v2.gc_num_sc_per_se) / le32_to_cpu(gc_info->v2.gc_num_sh_per_se); adev->gfx.config.num_packer_per_sc = le32_to_cpu(gc_info->v2.gc_num_packer_per_sc); + if (gc_info->v2.header.version_minor == 1) { + adev->gfx.config.gc_num_tcp_per_sa = le32_to_cpu(gc_info->v2_1.gc_num_tcp_per_sh); + adev->gfx.config.gc_tcp_size_per_cu = le32_to_cpu(gc_info->v2_1.gc_tcp_size_per_cu); + adev->gfx.config.gc_num_sdp_interface = le32_to_cpu(gc_info->v2_1.gc_num_sdp_interface); /* per XCD */ + adev->gfx.config.gc_num_cu_per_sqc = le32_to_cpu(gc_info->v2_1.gc_num_cu_per_sqc); + adev->gfx.config.gc_l1_instruction_cache_size_per_sqc = le32_to_cpu(gc_info->v2_1.gc_instruction_cache_size_per_sqc); + adev->gfx.config.gc_l1_data_cache_size_per_sqc = le32_to_cpu(gc_info->v2_1.gc_scalar_data_cache_size_per_sqc); + adev->gfx.config.gc_tcc_size = le32_to_cpu(gc_info->v2_1.gc_tcc_size); /* per XCD */ + } break; default: dev_err(adev->dev, @@ -1477,6 +1488,7 @@ static int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev) union mall_info { struct mall_info_v1_0 v1; + struct mall_info_v2_0 v2; }; static int amdgpu_discovery_get_mall_info(struct amdgpu_device *adev) @@ -1517,6 +1529,10 @@ static int amdgpu_discovery_get_mall_info(struct amdgpu_device *adev) adev->gmc.mall_size = mall_size; adev->gmc.m_half_use = half_use; break; + case 2: + mall_size_per_umc = le32_to_cpu(mall_info->v2.mall_size_per_umc); + adev->gmc.mall_size = mall_size_per_umc * adev->gmc.num_umc; + break; default: dev_err(adev->dev, "Unhandled MALL info table %d.%d\n", @@ -1702,6 +1718,9 @@ static int amdgpu_discovery_set_ih_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(6, 0, 2): amdgpu_device_ip_block_add(adev, &ih_v6_0_ip_block); break; + case IP_VERSION(6, 1, 0): + amdgpu_device_ip_block_add(adev, &ih_v6_1_ip_block); + break; default: dev_err(adev->dev, "Failed to add ih ip block(OSSSYS_HWIP:0x%x)\n", @@ -1750,6 +1769,7 @@ static int amdgpu_discovery_set_psp_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(13, 0, 8): case IP_VERSION(13, 0, 10): case IP_VERSION(13, 0, 11): + case IP_VERSION(14, 0, 0): amdgpu_device_ip_block_add(adev, &psp_v13_0_ip_block); break; case IP_VERSION(13, 0, 4): @@ -1968,6 +1988,7 @@ static int amdgpu_discovery_set_sdma_ip_blocks(struct amdgpu_device *adev) case IP_VERSION(6, 0, 1): case IP_VERSION(6, 0, 2): case IP_VERSION(6, 0, 3): + case IP_VERSION(6, 1, 0): amdgpu_device_ip_block_add(adev, &sdma_v6_0_ip_block); break; default: @@ -2447,6 +2468,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) break; case IP_VERSION(6, 0, 0): case IP_VERSION(6, 0, 1): + case IP_VERSION(6, 1, 0): adev->hdp.funcs = &hdp_v6_0_funcs; break; default: @@ -2509,6 +2531,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) break; case IP_VERSION(13, 0, 6): case IP_VERSION(13, 0, 8): + case IP_VERSION(14, 0, 0): adev->smuio.funcs = &smuio_v13_0_6_funcs; break; default: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index b702f499f5fb..363e6a2cad8c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -38,6 +38,8 @@ #include <linux/pci.h> #include <linux/pm_runtime.h> #include <drm/drm_crtc_helper.h> +#include <drm/drm_damage_helper.h> +#include <drm/drm_drv.h> #include <drm/drm_edid.h> #include <drm/drm_fb_helper.h> #include <drm/drm_gem_framebuffer_helper.h> @@ -124,7 +126,7 @@ static void amdgpu_display_flip_work_func(struct work_struct *__work) struct drm_crtc *crtc = &amdgpu_crtc->base; unsigned long flags; - unsigned i; + unsigned int i; int vpos, hpos; for (i = 0; i < work->shared_count; ++i) @@ -201,7 +203,7 @@ int amdgpu_display_crtc_page_flip_target(struct drm_crtc *crtc, u64 tiling_flags; int i, r; - work = kzalloc(sizeof *work, GFP_KERNEL); + work = kzalloc(sizeof(*work), GFP_KERNEL); if (work == NULL) return -ENOMEM; @@ -332,13 +334,15 @@ int amdgpu_display_crtc_set_config(struct drm_mode_set *set, adev = drm_to_adev(dev); /* if we have active crtcs and we don't have a power ref, - take the current one */ + * take the current one + */ if (active && !adev->have_disp_power_ref) { adev->have_disp_power_ref = true; return ret; } /* if we have no active crtcs, then drop the power ref - we got before */ + * we got before + */ if (!active && adev->have_disp_power_ref) { pm_runtime_put_autosuspend(dev->dev); adev->have_disp_power_ref = false; @@ -507,11 +511,10 @@ bool amdgpu_display_ddc_probe(struct amdgpu_connector *amdgpu_connector, if (amdgpu_connector->router.ddc_valid) amdgpu_i2c_router_select_ddc_port(amdgpu_connector); - if (use_aux) { + if (use_aux) ret = i2c_transfer(&amdgpu_connector->ddc_bus->aux.ddc, msgs, 2); - } else { + else ret = i2c_transfer(&amdgpu_connector->ddc_bus->adapter, msgs, 2); - } if (ret != 2) /* Couldn't find an accessible DDC on this connector */ @@ -520,20 +523,40 @@ bool amdgpu_display_ddc_probe(struct amdgpu_connector *amdgpu_connector, * EDID header starts with: * 0x00,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0x00. * Only the first 6 bytes must be valid as - * drm_edid_block_valid() can fix the last 2 bytes */ + * drm_edid_block_valid() can fix the last 2 bytes + */ if (drm_edid_header_is_valid(buf) < 6) { /* Couldn't find an accessible EDID on this - * connector */ + * connector + */ return false; } return true; } +static int amdgpu_dirtyfb(struct drm_framebuffer *fb, struct drm_file *file, + unsigned int flags, unsigned int color, + struct drm_clip_rect *clips, unsigned int num_clips) +{ + + if (file) + return -ENOSYS; + + return drm_atomic_helper_dirtyfb(fb, file, flags, color, clips, + num_clips); +} + static const struct drm_framebuffer_funcs amdgpu_fb_funcs = { .destroy = drm_gem_fb_destroy, .create_handle = drm_gem_fb_create_handle, }; +static const struct drm_framebuffer_funcs amdgpu_fb_funcs_atomic = { + .destroy = drm_gem_fb_destroy, + .create_handle = drm_gem_fb_create_handle, + .dirty = amdgpu_dirtyfb +}; + uint32_t amdgpu_display_supported_domains(struct amdgpu_device *adev, uint64_t bo_flags) { @@ -1136,7 +1159,11 @@ static int amdgpu_display_gem_fb_verify_and_init(struct drm_device *dev, if (ret) goto err; - ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs); + if (drm_drv_uses_atomic_modeset(dev)) + ret = drm_framebuffer_init(dev, &rfb->base, + &amdgpu_fb_funcs_atomic); + else + ret = drm_framebuffer_init(dev, &rfb->base, &amdgpu_fb_funcs); if (ret) goto err; @@ -1216,8 +1243,10 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev, obj = drm_gem_object_lookup(file_priv, mode_cmd->handles[0]); if (obj == NULL) { - drm_dbg_kms(dev, "No GEM object associated to handle 0x%08X, " - "can't create framebuffer\n", mode_cmd->handles[0]); + drm_dbg_kms(dev, + "No GEM object associated to handle 0x%08X, can't create framebuffer\n", + mode_cmd->handles[0]); + return ERR_PTR(-ENOENT); } @@ -1410,6 +1439,7 @@ bool amdgpu_display_crtc_scaling_mode_fixup(struct drm_crtc *crtc, } if (amdgpu_crtc->rmx_type != RMX_OFF) { fixed20_12 a, b; + a.full = dfixed_const(src_v); b.full = dfixed_const(dst_v); amdgpu_crtc->vsc.full = dfixed_div(a, b); @@ -1429,7 +1459,7 @@ bool amdgpu_display_crtc_scaling_mode_fixup(struct drm_crtc *crtc, * * \param dev Device to query. * \param pipe Crtc to query. - * \param flags Flags from caller (DRM_CALLED_FROM_VBLIRQ or 0). + * \param flags from caller (DRM_CALLED_FROM_VBLIRQ or 0). * For driver internal use only also supports these flags: * * USE_REAL_VBLANKSTART to use the real start of vblank instead @@ -1504,8 +1534,8 @@ int amdgpu_display_get_crtc_scanoutpos(struct drm_device *dev, /* Called from driver internal vblank counter query code? */ if (flags & GET_DISTANCE_TO_VBLANKSTART) { - /* Caller wants distance from real vbl_start in *hpos */ - *hpos = *vpos - vbl_start; + /* Caller wants distance from real vbl_start in *hpos */ + *hpos = *vpos - vbl_start; } /* Fudge vblank to start a few scanlines earlier to handle the @@ -1527,7 +1557,7 @@ int amdgpu_display_get_crtc_scanoutpos(struct drm_device *dev, /* In vblank? */ if (in_vbl) - ret |= DRM_SCANOUTPOS_IN_VBLANK; + ret |= DRM_SCANOUTPOS_IN_VBLANK; /* Called from driver internal vblank counter query code? */ if (flags & GET_DISTANCE_TO_VBLANKSTART) { @@ -1635,6 +1665,7 @@ int amdgpu_display_suspend_helper(struct amdgpu_device *adev) if (amdgpu_crtc->cursor_bo && !adev->enable_virtual_display) { struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo); + r = amdgpu_bo_reserve(aobj, true); if (r == 0) { amdgpu_bo_unpin(aobj); @@ -1642,9 +1673,9 @@ int amdgpu_display_suspend_helper(struct amdgpu_device *adev) } } - if (fb == NULL || fb->obj[0] == NULL) { + if (!fb || !fb->obj[0]) continue; - } + robj = gem_to_amdgpu_bo(fb->obj[0]); if (!amdgpu_display_robj_is_fb(adev, robj)) { r = amdgpu_bo_reserve(robj, true); @@ -1671,6 +1702,7 @@ int amdgpu_display_resume_helper(struct amdgpu_device *adev) if (amdgpu_crtc->cursor_bo && !adev->enable_virtual_display) { struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo); + r = amdgpu_bo_reserve(aobj, true); if (r == 0) { r = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h index f637574644c0..09f6727e7c73 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h @@ -31,10 +31,15 @@ struct amdgpu_doorbell { /* doorbell mmio */ resource_size_t base; resource_size_t size; - u32 __iomem *ptr; /* Number of doorbells reserved for amdgpu kernel driver */ u32 num_kernel_doorbells; + + /* Kernel doorbells */ + struct amdgpu_bo *kernel_doorbells; + + /* For CPU access of doorbells */ + uint32_t *cpu_addr; }; /* Reserved doorbells for amdgpu (including multimedia). @@ -90,8 +95,7 @@ struct amdgpu_doorbell_index { uint32_t xcc_doorbell_range; }; -typedef enum _AMDGPU_DOORBELL_ASSIGNMENT -{ +enum AMDGPU_DOORBELL_ASSIGNMENT { AMDGPU_DOORBELL_KIQ = 0x000, AMDGPU_DOORBELL_HIQ = 0x001, AMDGPU_DOORBELL_DIQ = 0x002, @@ -109,10 +113,10 @@ typedef enum _AMDGPU_DOORBELL_ASSIGNMENT AMDGPU_DOORBELL_IH = 0x1E8, AMDGPU_DOORBELL_MAX_ASSIGNMENT = 0x3FF, AMDGPU_DOORBELL_INVALID = 0xFFFF -} AMDGPU_DOORBELL_ASSIGNMENT; +}; + +enum AMDGPU_VEGA20_DOORBELL_ASSIGNMENT { -typedef enum _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT -{ /* Compute + GFX: 0~255 */ AMDGPU_VEGA20_DOORBELL_KIQ = 0x000, AMDGPU_VEGA20_DOORBELL_HIQ = 0x001, @@ -176,10 +180,10 @@ typedef enum _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT = 0x1F7, AMDGPU_VEGA20_DOORBELL_INVALID = 0xFFFF -} AMDGPU_VEGA20_DOORBELL_ASSIGNMENT; +}; + +enum AMDGPU_NAVI10_DOORBELL_ASSIGNMENT { -typedef enum _AMDGPU_NAVI10_DOORBELL_ASSIGNMENT -{ /* Compute + GFX: 0~255 */ AMDGPU_NAVI10_DOORBELL_KIQ = 0x000, AMDGPU_NAVI10_DOORBELL_HIQ = 0x001, @@ -227,13 +231,12 @@ typedef enum _AMDGPU_NAVI10_DOORBELL_ASSIGNMENT AMDGPU_NAVI10_DOORBELL_MAX_ASSIGNMENT = 0x18F, AMDGPU_NAVI10_DOORBELL_INVALID = 0xFFFF -} AMDGPU_NAVI10_DOORBELL_ASSIGNMENT; +}; /* * 64bit doorbell, offset are in QWORD, occupy 2KB doorbell space */ -typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT -{ +enum AMDGPU_DOORBELL64_ASSIGNMENT { /* * All compute related doorbells: kiq, hiq, diq, traditional compute queue, user queue, should locate in * a continues range so that programming CP_MEC_DOORBELL_RANGE_LOWER/UPPER can cover this range. @@ -309,9 +312,10 @@ typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT AMDGPU_DOORBELL64_MAX_ASSIGNMENT = 0xFF, AMDGPU_DOORBELL64_INVALID = 0xFFFF -} AMDGPU_DOORBELL64_ASSIGNMENT; +}; + +enum AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1 { -typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1 { /* XCC0: 0x00 ~20, XCC1: 20 ~ 2F ... */ /* KIQ/HIQ/DIQ */ @@ -330,22 +334,32 @@ typedef enum _AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1 { AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_END = 0x19F, /* IH: 0x1A0 ~ 0x1AF */ AMDGPU_DOORBELL_LAYOUT1_IH = 0x1A0, - /* VCN: 0x1B0 ~ 0x1D4 */ + /* VCN: 0x1B0 ~ 0x1E8 */ AMDGPU_DOORBELL_LAYOUT1_VCN_START = 0x1B0, - AMDGPU_DOORBELL_LAYOUT1_VCN_END = 0x1D4, + AMDGPU_DOORBELL_LAYOUT1_VCN_END = 0x1E8, AMDGPU_DOORBELL_LAYOUT1_FIRST_NON_CP = AMDGPU_DOORBELL_LAYOUT1_sDMA_ENGINE_START, AMDGPU_DOORBELL_LAYOUT1_LAST_NON_CP = AMDGPU_DOORBELL_LAYOUT1_VCN_END, - AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT = 0x1D4, + AMDGPU_DOORBELL_LAYOUT1_MAX_ASSIGNMENT = 0x1E8, AMDGPU_DOORBELL_LAYOUT1_INVALID = 0xFFFF -} AMDGPU_DOORBELL_ASSIGNMENT_LAYOUT1; +}; u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index); void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v); u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index); void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v); +/* + * GPU doorbell aperture helpers function. + */ +int amdgpu_doorbell_init(struct amdgpu_device *adev); +void amdgpu_doorbell_fini(struct amdgpu_device *adev); +int amdgpu_doorbell_create_kernel_doorbells(struct amdgpu_device *adev); +uint32_t amdgpu_doorbell_index_on_bar(struct amdgpu_device *adev, + struct amdgpu_bo *db_bo, + uint32_t doorbell_index); + #define RDOORBELL32(index) amdgpu_mm_rdoorbell(adev, (index)) #define WDOORBELL32(index, v) amdgpu_mm_wdoorbell(adev, (index), (v)) #define RDOORBELL64(index) amdgpu_mm_rdoorbell64(adev, (index)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c new file mode 100644 index 000000000000..da4be0bbb446 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell_mgr.c @@ -0,0 +1,230 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright 2022 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include "amdgpu.h" + +/** + * amdgpu_mm_rdoorbell - read a doorbell dword + * + * @adev: amdgpu_device pointer + * @index: doorbell index + * + * Returns the value in the doorbell aperture at the + * requested doorbell index (CIK). + */ +u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index) +{ + if (amdgpu_device_skip_hw_access(adev)) + return 0; + + if (index < adev->doorbell.num_kernel_doorbells) + return readl(adev->doorbell.cpu_addr + index); + + DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); + return 0; +} + +/** + * amdgpu_mm_wdoorbell - write a doorbell dword + * + * @adev: amdgpu_device pointer + * @index: doorbell index + * @v: value to write + * + * Writes @v to the doorbell aperture at the + * requested doorbell index (CIK). + */ +void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v) +{ + if (amdgpu_device_skip_hw_access(adev)) + return; + + if (index < adev->doorbell.num_kernel_doorbells) + writel(v, adev->doorbell.cpu_addr + index); + else + DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); +} + +/** + * amdgpu_mm_rdoorbell64 - read a doorbell Qword + * + * @adev: amdgpu_device pointer + * @index: doorbell index + * + * Returns the value in the doorbell aperture at the + * requested doorbell index (VEGA10+). + */ +u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index) +{ + if (amdgpu_device_skip_hw_access(adev)) + return 0; + + if (index < adev->doorbell.num_kernel_doorbells) + return atomic64_read((atomic64_t *)(adev->doorbell.cpu_addr + index)); + + DRM_ERROR("reading beyond doorbell aperture: 0x%08x!\n", index); + return 0; +} + +/** + * amdgpu_mm_wdoorbell64 - write a doorbell Qword + * + * @adev: amdgpu_device pointer + * @index: doorbell index + * @v: value to write + * + * Writes @v to the doorbell aperture at the + * requested doorbell index (VEGA10+). + */ +void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v) +{ + if (amdgpu_device_skip_hw_access(adev)) + return; + + if (index < adev->doorbell.num_kernel_doorbells) + atomic64_set((atomic64_t *)(adev->doorbell.cpu_addr + index), v); + else + DRM_ERROR("writing beyond doorbell aperture: 0x%08x!\n", index); +} + +/** + * amdgpu_doorbell_index_on_bar - Find doorbell's absolute offset in BAR + * + * @adev: amdgpu_device pointer + * @db_bo: doorbell object's bo + * @db_index: doorbell relative index in this doorbell object + * + * returns doorbell's absolute index in BAR + */ +uint32_t amdgpu_doorbell_index_on_bar(struct amdgpu_device *adev, + struct amdgpu_bo *db_bo, + uint32_t doorbell_index) +{ + int db_bo_offset; + + db_bo_offset = amdgpu_bo_gpu_offset_no_check(db_bo); + + /* doorbell index is 32 bit but doorbell's size is 64-bit, so *2 */ + return db_bo_offset / sizeof(u32) + doorbell_index * 2; +} + +/** + * amdgpu_doorbell_create_kernel_doorbells - Create kernel doorbells for graphics + * + * @adev: amdgpu_device pointer + * + * Creates doorbells for graphics driver usages. + * returns 0 on success, error otherwise. + */ +int amdgpu_doorbell_create_kernel_doorbells(struct amdgpu_device *adev) +{ + int r; + int size; + + /* Reserve first num_kernel_doorbells (page-aligned) for kernel ops */ + size = ALIGN(adev->doorbell.num_kernel_doorbells * sizeof(u32), PAGE_SIZE); + + /* Allocate an extra page for MES kernel usages (ring test) */ + adev->mes.db_start_dw_offset = size / sizeof(u32); + size += PAGE_SIZE; + + r = amdgpu_bo_create_kernel(adev, + size, + PAGE_SIZE, + AMDGPU_GEM_DOMAIN_DOORBELL, + &adev->doorbell.kernel_doorbells, + NULL, + (void **)&adev->doorbell.cpu_addr); + if (r) { + DRM_ERROR("Failed to allocate kernel doorbells, err=%d\n", r); + return r; + } + + adev->doorbell.num_kernel_doorbells = size / sizeof(u32); + return 0; +} + +/* + * GPU doorbell aperture helpers function. + */ +/** + * amdgpu_doorbell_init - Init doorbell driver information. + * + * @adev: amdgpu_device pointer + * + * Init doorbell driver information (CIK) + * Returns 0 on success, error on failure. + */ +int amdgpu_doorbell_init(struct amdgpu_device *adev) +{ + + /* No doorbell on SI hardware generation */ + if (adev->asic_type < CHIP_BONAIRE) { + adev->doorbell.base = 0; + adev->doorbell.size = 0; + adev->doorbell.num_kernel_doorbells = 0; + return 0; + } + + if (pci_resource_flags(adev->pdev, 2) & IORESOURCE_UNSET) + return -EINVAL; + + amdgpu_asic_init_doorbell_index(adev); + + /* doorbell bar mapping */ + adev->doorbell.base = pci_resource_start(adev->pdev, 2); + adev->doorbell.size = pci_resource_len(adev->pdev, 2); + + adev->doorbell.num_kernel_doorbells = + min_t(u32, adev->doorbell.size / sizeof(u32), + adev->doorbell_index.max_assignment + 1); + if (adev->doorbell.num_kernel_doorbells == 0) + return -EINVAL; + + /* + * For Vega, reserve and map two pages on doorbell BAR since SDMA + * paging queue doorbell use the second page. The + * AMDGPU_DOORBELL64_MAX_ASSIGNMENT definition assumes all the + * doorbells are in the first page. So with paging queue enabled, + * the max num_kernel_doorbells should + 1 page (0x400 in dword) + */ + if (adev->asic_type >= CHIP_VEGA10) + adev->doorbell.num_kernel_doorbells += 0x400; + + return 0; +} + +/** + * amdgpu_doorbell_fini - Tear down doorbell driver information. + * + * @adev: amdgpu_device pointer + * + * Tear down doorbell driver information (CIK) + */ +void amdgpu_doorbell_fini(struct amdgpu_device *adev) +{ + amdgpu_bo_free_kernel(&adev->doorbell.kernel_doorbells, + NULL, + (void **)&adev->doorbell.cpu_addr); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index e90f730eb715..81edf66dbea8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -26,30 +26,30 @@ #include <drm/drm_drv.h> #include <drm/drm_fbdev_generic.h> #include <drm/drm_gem.h> -#include <drm/drm_vblank.h> #include <drm/drm_managed.h> -#include "amdgpu_drv.h" - #include <drm/drm_pciids.h> -#include <linux/module.h> -#include <linux/pm_runtime.h> -#include <linux/vga_switcheroo.h> #include <drm/drm_probe_helper.h> -#include <linux/mmu_notifier.h> -#include <linux/suspend.h> +#include <drm/drm_vblank.h> + #include <linux/cc_platform.h> #include <linux/dynamic_debug.h> +#include <linux/module.h> +#include <linux/mmu_notifier.h> +#include <linux/pm_runtime.h> +#include <linux/suspend.h> +#include <linux/vga_switcheroo.h> #include "amdgpu.h" -#include "amdgpu_irq.h" +#include "amdgpu_amdkfd.h" #include "amdgpu_dma_buf.h" -#include "amdgpu_sched.h" +#include "amdgpu_drv.h" #include "amdgpu_fdinfo.h" -#include "amdgpu_amdkfd.h" - +#include "amdgpu_irq.h" +#include "amdgpu_psp.h" #include "amdgpu_ras.h" -#include "amdgpu_xgmi.h" #include "amdgpu_reset.h" +#include "amdgpu_sched.h" +#include "amdgpu_xgmi.h" #include "../amdxcp/amdgpu_xcp_drv.h" /* @@ -187,7 +187,6 @@ int amdgpu_mes_kiq; int amdgpu_noretry = -1; int amdgpu_force_asic_type = -1; int amdgpu_tmz = -1; /* auto */ -uint amdgpu_freesync_vid_mode; int amdgpu_reset_method = -1; /* auto */ int amdgpu_num_kcq = -1; int amdgpu_smartshift_bias; @@ -313,9 +312,7 @@ module_param_named(msi, amdgpu_msi, int, 0444); * jobs is 10000. The timeout for compute is 60000. */ MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: for bare metal 10000 for non-compute jobs and 60000 for compute jobs; " - "for passthrough or sriov, 10000 for all jobs." - " 0: keep default value. negative: infinity timeout), " - "format: for bare metal [Non-Compute] or [GFX,Compute,SDMA,Video]; " + "for passthrough or sriov, 10000 for all jobs. 0: keep default value. negative: infinity timeout), format: for bare metal [Non-Compute] or [GFX,Compute,SDMA,Video]; " "for passthrough or sriov [all jobs] or [GFX,Compute,SDMA,Video]."); module_param_string(lockup_timeout, amdgpu_lockup_timeout, sizeof(amdgpu_lockup_timeout), 0444); @@ -350,8 +347,9 @@ module_param_named(aspm, amdgpu_aspm, int, 0444); * Override for runtime power management control for dGPUs. The amdgpu driver can dynamically power down * the dGPUs when they are idle if supported. The default is -1 (auto enable). * Setting the value to 0 disables this functionality. + * Setting the value to -2 is auto enabled with power down when displays are attached. */ -MODULE_PARM_DESC(runpm, "PX runtime pm (2 = force enable with BAMACO, 1 = force enable with BACO, 0 = disable, -1 = auto)"); +MODULE_PARM_DESC(runpm, "PX runtime pm (2 = force enable with BAMACO, 1 = force enable with BACO, 0 = disable, -1 = auto, -2 = autowith displays)"); module_param_named(runpm, amdgpu_runtime_pm, int, 0444); /** @@ -584,7 +582,7 @@ module_param_named(timeout_period, amdgpu_watchdog_timer.period, uint, 0644); */ #ifdef CONFIG_DRM_AMDGPU_SI -#if defined(CONFIG_DRM_RADEON) || defined(CONFIG_DRM_RADEON_MODULE) +#if IS_ENABLED(CONFIG_DRM_RADEON) || IS_ENABLED(CONFIG_DRM_RADEON_MODULE) int amdgpu_si_support = 0; MODULE_PARM_DESC(si_support, "SI support (1 = enabled, 0 = disabled (default))"); #else @@ -603,7 +601,7 @@ module_param_named(si_support, amdgpu_si_support, int, 0444); */ #ifdef CONFIG_DRM_AMDGPU_CIK -#if defined(CONFIG_DRM_RADEON) || defined(CONFIG_DRM_RADEON_MODULE) +#if IS_ENABLED(CONFIG_DRM_RADEON) || IS_ENABLED(CONFIG_DRM_RADEON_MODULE) int amdgpu_cik_support = 0; MODULE_PARM_DESC(cik_support, "CIK support (1 = enabled, 0 = disabled (default))"); #else @@ -620,8 +618,7 @@ module_param_named(cik_support, amdgpu_cik_support, int, 0444); * E.g. 0x1 = 256Mbyte, 0x2 = 512Mbyte, 0x4 = 1 Gbyte, 0x8 = 2GByte. The default is 0 (disabled). */ MODULE_PARM_DESC(smu_memory_pool_size, - "reserve gtt for smu debug usage, 0 = disable," - "0x1 = 256Mbyte, 0x2 = 512Mbyte, 0x4 = 1 Gbyte, 0x8 = 2GByte"); + "reserve gtt for smu debug usage, 0 = disable,0x1 = 256Mbyte, 0x2 = 512Mbyte, 0x4 = 1 Gbyte, 0x8 = 2GByte"); module_param_named(smu_memory_pool_size, amdgpu_smu_memory_pool_size, uint, 0444); /** @@ -759,20 +756,6 @@ MODULE_PARM_DESC(debug_largebar, "Debug large-bar flag used to simulate large-bar capability on non-large bar machine (0 = disable, 1 = enable)"); /** - * DOC: ignore_crat (int) - * Ignore CRAT table during KFD initialization. By default, KFD uses the ACPI CRAT - * table to get information about AMD APUs. This option can serve as a workaround on - * systems with a broken CRAT table. - * - * Default is auto (according to asic type, iommu_v2, and crat table, to decide - * whether use CRAT) - */ -int ignore_crat; -module_param(ignore_crat, int, 0444); -MODULE_PARM_DESC(ignore_crat, - "Ignore CRAT table during KFD initialization (0 = auto (default), 1 = ignore CRAT)"); - -/** * DOC: halt_if_hws_hang (int) * Halt if HWS hang is detected. Default value, 0, disables the halt on hang. * Setting 1 enables halt on hang. @@ -791,9 +774,9 @@ module_param(hws_gws_support, bool, 0444); MODULE_PARM_DESC(hws_gws_support, "Assume MEC2 FW supports GWS barriers (false = rely on FW version check (Default), true = force supported)"); /** - * DOC: queue_preemption_timeout_ms (int) - * queue preemption timeout in ms (1 = Minimum, 9000 = default) - */ + * DOC: queue_preemption_timeout_ms (int) + * queue preemption timeout in ms (1 = Minimum, 9000 = default) + */ int queue_preemption_timeout_ms = 9000; module_param(queue_preemption_timeout_ms, int, 0644); MODULE_PARM_DESC(queue_preemption_timeout_ms, "queue preemption timeout in ms (1 = Minimum, 9000 = default)"); @@ -889,32 +872,6 @@ MODULE_PARM_DESC(tmz, "Enable TMZ feature (-1 = auto (default), 0 = off, 1 = on) module_param_named(tmz, amdgpu_tmz, int, 0444); /** - * DOC: freesync_video (uint) - * Enable the optimization to adjust front porch timing to achieve seamless - * mode change experience when setting a freesync supported mode for which full - * modeset is not needed. - * - * The Display Core will add a set of modes derived from the base FreeSync - * video mode into the corresponding connector's mode list based on commonly - * used refresh rates and VRR range of the connected display, when users enable - * this feature. From the userspace perspective, they can see a seamless mode - * change experience when the change between different refresh rates under the - * same resolution. Additionally, userspace applications such as Video playback - * can read this modeset list and change the refresh rate based on the video - * frame rate. Finally, the userspace can also derive an appropriate mode for a - * particular refresh rate based on the FreeSync Mode and add it to the - * connector's mode list. - * - * Note: This is an experimental feature. - * - * The default value: 0 (off). - */ -MODULE_PARM_DESC( - freesync_video, - "Enable freesync modesetting optimization feature (0 = off (default), 1 = on)"); -module_param_named(freesync_video, amdgpu_freesync_vid_mode, uint, 0444); - -/** * DOC: reset_method (int) * GPU reset method (-1 = auto (default), 0 = legacy, 1 = mode0, 2 = mode1, 3 = mode2, 4 = baco) */ @@ -2417,7 +2374,6 @@ static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work) amdgpu_amdkfd_device_init(adev); amdgpu_ttm_set_buffer_funcs_status(adev, true); } - return; } static int amdgpu_pmops_prepare(struct device *dev) @@ -2541,24 +2497,26 @@ static int amdgpu_runtime_idle_check_display(struct device *dev) struct drm_connector_list_iter iter; int ret = 0; - /* XXX: Return busy if any displays are connected to avoid - * possible display wakeups after runtime resume due to - * hotplug events in case any displays were connected while - * the GPU was in suspend. Remove this once that is fixed. - */ - mutex_lock(&drm_dev->mode_config.mutex); - drm_connector_list_iter_begin(drm_dev, &iter); - drm_for_each_connector_iter(list_connector, &iter) { - if (list_connector->status == connector_status_connected) { - ret = -EBUSY; - break; + if (amdgpu_runtime_pm != -2) { + /* XXX: Return busy if any displays are connected to avoid + * possible display wakeups after runtime resume due to + * hotplug events in case any displays were connected while + * the GPU was in suspend. Remove this once that is fixed. + */ + mutex_lock(&drm_dev->mode_config.mutex); + drm_connector_list_iter_begin(drm_dev, &iter); + drm_for_each_connector_iter(list_connector, &iter) { + if (list_connector->status == connector_status_connected) { + ret = -EBUSY; + break; + } } - } - drm_connector_list_iter_end(&iter); - mutex_unlock(&drm_dev->mode_config.mutex); + drm_connector_list_iter_end(&iter); + mutex_unlock(&drm_dev->mode_config.mutex); - if (ret) - return ret; + if (ret) + return ret; + } if (adev->dc_enabled) { struct drm_crtc *crtc; @@ -2614,6 +2572,7 @@ static int amdgpu_pmops_runtime_suspend(struct device *dev) /* wait for all rings to drain before suspending */ for (i = 0; i < AMDGPU_MAX_RINGS; i++) { struct amdgpu_ring *ring = adev->rings[i]; + if (ring && ring->sched.ready) { ret = amdgpu_fence_wait_empty(ring); if (ret) @@ -2738,6 +2697,7 @@ long amdgpu_drm_ioctl(struct file *filp, struct drm_file *file_priv = filp->private_data; struct drm_device *dev; long ret; + dev = file_priv->minor->dev; ret = pm_runtime_get_sync(dev->dev); if (ret < 0) @@ -2802,9 +2762,8 @@ int amdgpu_file_to_fpriv(struct file *filp, struct amdgpu_fpriv **fpriv) if (!filp) return -EINVAL; - if (filp->f_op != &amdgpu_driver_kms_fops) { + if (filp->f_op != &amdgpu_driver_kms_fops) return -EINVAL; - } file = filp->private_data; *fpriv = file->driver_priv; @@ -2891,16 +2850,13 @@ static struct pci_error_handlers amdgpu_pci_err_handler = { .resume = amdgpu_pci_resume, }; -extern const struct attribute_group amdgpu_vram_mgr_attr_group; -extern const struct attribute_group amdgpu_gtt_mgr_attr_group; - static const struct attribute_group *amdgpu_sysfs_groups[] = { &amdgpu_vram_mgr_attr_group, &amdgpu_gtt_mgr_attr_group, + &amdgpu_flash_attr_group, NULL, }; - static struct pci_driver amdgpu_kms_pci_driver = { .name = DRIVER_NAME, .id_table = pciidlist, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_eeprom.c index 7d2a908438e9..e71768661ca8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_eeprom.c @@ -183,6 +183,8 @@ static int amdgpu_eeprom_xfer(struct i2c_adapter *i2c_adap, u32 eeprom_addr, { const struct i2c_adapter_quirks *quirks = i2c_adap->quirks; u16 limit; + u16 ps; /* Partial size */ + int res = 0, r; if (!quirks) limit = 0; @@ -200,28 +202,25 @@ static int amdgpu_eeprom_xfer(struct i2c_adapter *i2c_adap, u32 eeprom_addr, eeprom_addr, buf_size, read ? "read" : "write", EEPROM_OFFSET_SIZE); return -EINVAL; - } else { - u16 ps; /* Partial size */ - int res = 0, r; - - /* The "limit" includes all data bytes sent/received, - * which would include the EEPROM_OFFSET_SIZE bytes. - * Account for them here. - */ - limit -= EEPROM_OFFSET_SIZE; - for ( ; buf_size > 0; - buf_size -= ps, eeprom_addr += ps, eeprom_buf += ps) { - ps = min(limit, buf_size); - - r = __amdgpu_eeprom_xfer(i2c_adap, eeprom_addr, - eeprom_buf, ps, read); - if (r < 0) - return r; - res += r; - } + } - return res; + /* The "limit" includes all data bytes sent/received, + * which would include the EEPROM_OFFSET_SIZE bytes. + * Account for them here. + */ + limit -= EEPROM_OFFSET_SIZE; + for ( ; buf_size > 0; + buf_size -= ps, eeprom_addr += ps, eeprom_buf += ps) { + ps = min(limit, buf_size); + + r = __amdgpu_eeprom_xfer(i2c_adap, eeprom_addr, + eeprom_buf, ps, read); + if (r < 0) + return r; + res += r; } + + return res; } int amdgpu_eeprom_read(struct i2c_adapter *i2c_adap, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c index 13d7413d4ca3..6038b5021b27 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c @@ -89,7 +89,7 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file) drm_printf(p, "pasid:\t%u\n", fpriv->vm.pasid); drm_printf(p, "drm-driver:\t%s\n", file->minor->dev->driver->name); drm_printf(p, "drm-pdev:\t%04x:%02x:%02x.%d\n", domain, bus, dev, fn); - drm_printf(p, "drm-client-id:\t%Lu\n", vm->immediate.fence_context); + drm_printf(p, "drm-client-id:\t%llu\n", vm->immediate.fence_context); drm_printf(p, "drm-memory-vram:\t%llu KiB\n", stats.vram/1024UL); drm_printf(p, "drm-memory-gtt: \t%llu KiB\n", stats.gtt/1024UL); drm_printf(p, "drm-memory-cpu: \t%llu KiB\n", stats.cpu/1024UL); @@ -109,7 +109,7 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file) if (!usage[hw_ip]) continue; - drm_printf(p, "drm-engine-%s:\t%Ld ns\n", amdgpu_ip_name[hw_ip], + drm_printf(p, "drm-engine-%s:\t%lld ns\n", amdgpu_ip_name[hw_ip], ktime_to_ns(usage[hw_ip])); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c index c694b41f6461..7537f5aa76f0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c @@ -552,6 +552,41 @@ int amdgpu_fence_driver_sw_init(struct amdgpu_device *adev) } /** + * amdgpu_fence_need_ring_interrupt_restore - helper function to check whether + * fence driver interrupts need to be restored. + * + * @ring: ring that to be checked + * + * Interrupts for rings that belong to GFX IP don't need to be restored + * when the target power state is s0ix. + * + * Return true if need to restore interrupts, false otherwise. + */ +static bool amdgpu_fence_need_ring_interrupt_restore(struct amdgpu_ring *ring) +{ + struct amdgpu_device *adev = ring->adev; + bool is_gfx_power_domain = false; + + switch (ring->funcs->type) { + case AMDGPU_RING_TYPE_SDMA: + /* SDMA 5.x+ is part of GFX power domain so it's covered by GFXOFF */ + if (adev->ip_versions[SDMA0_HWIP][0] >= IP_VERSION(5, 0, 0)) + is_gfx_power_domain = true; + break; + case AMDGPU_RING_TYPE_GFX: + case AMDGPU_RING_TYPE_COMPUTE: + case AMDGPU_RING_TYPE_KIQ: + case AMDGPU_RING_TYPE_MES: + is_gfx_power_domain = true; + break; + default: + break; + } + + return !(adev->in_s0ix && is_gfx_power_domain); +} + +/** * amdgpu_fence_driver_hw_fini - tear down the fence driver * for all possible rings. * @@ -579,7 +614,8 @@ void amdgpu_fence_driver_hw_fini(struct amdgpu_device *adev) amdgpu_fence_driver_force_completion(ring); if (!drm_dev_is_unplugged(adev_to_drm(adev)) && - ring->fence_drv.irq_src) + ring->fence_drv.irq_src && + amdgpu_fence_need_ring_interrupt_restore(ring)) amdgpu_irq_put(adev, ring->fence_drv.irq_src, ring->fence_drv.irq_type); @@ -655,7 +691,8 @@ void amdgpu_fence_driver_hw_init(struct amdgpu_device *adev) continue; /* enable the interrupt */ - if (ring->fence_drv.irq_src) + if (ring->fence_drv.irq_src && + amdgpu_fence_need_ring_interrupt_restore(ring)) amdgpu_irq_get(adev, ring->fence_drv.irq_src, ring->fence_drv.irq_type); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c index 4620c4712ce3..9c66d98af6d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.c @@ -60,10 +60,10 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev, u32 *fru_addr) switch (adev->asic_type) { case CHIP_VEGA20: /* D161 and D163 are the VG20 server SKUs */ - if (strnstr(atom_ctx->vbios_version, "D161", - sizeof(atom_ctx->vbios_version)) || - strnstr(atom_ctx->vbios_version, "D163", - sizeof(atom_ctx->vbios_version))) { + if (strnstr(atom_ctx->vbios_pn, "D161", + sizeof(atom_ctx->vbios_pn)) || + strnstr(atom_ctx->vbios_pn, "D163", + sizeof(atom_ctx->vbios_pn))) { if (fru_addr) *fru_addr = FRU_EEPROM_MADDR_6; return true; @@ -72,22 +72,23 @@ static bool is_fru_eeprom_supported(struct amdgpu_device *adev, u32 *fru_addr) } case CHIP_ALDEBARAN: /* All Aldebaran SKUs have an FRU */ - if (!strnstr(atom_ctx->vbios_version, "D673", - sizeof(atom_ctx->vbios_version))) + if (!strnstr(atom_ctx->vbios_pn, "D673", + sizeof(atom_ctx->vbios_pn))) if (fru_addr) *fru_addr = FRU_EEPROM_MADDR_6; return true; case CHIP_SIENNA_CICHLID: - if (strnstr(atom_ctx->vbios_version, "D603", - sizeof(atom_ctx->vbios_version))) { - if (strnstr(atom_ctx->vbios_version, "D603GLXE", - sizeof(atom_ctx->vbios_version))) { + if (strnstr(atom_ctx->vbios_pn, "D603", + sizeof(atom_ctx->vbios_pn))) { + if (strnstr(atom_ctx->vbios_pn, "D603GLXE", + sizeof(atom_ctx->vbios_pn))) { return false; - } else { - if (fru_addr) - *fru_addr = FRU_EEPROM_MADDR_6; - return true; } + + if (fru_addr) + *fru_addr = FRU_EEPROM_MADDR_6; + return true; + } else { return false; } @@ -211,3 +212,92 @@ Out: kfree(pia); return 0; } + +/** + * DOC: product_name + * + * The amdgpu driver provides a sysfs API for reporting the product name + * for the device + * The file product_name is used for this and returns the product name + * as returned from the FRU. + * NOTE: This is only available for certain server cards + */ + +static ssize_t amdgpu_fru_product_name_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + + return sysfs_emit(buf, "%s\n", adev->product_name); +} + +static DEVICE_ATTR(product_name, 0444, amdgpu_fru_product_name_show, NULL); + +/** + * DOC: product_number + * + * The amdgpu driver provides a sysfs API for reporting the part number + * for the device + * The file product_number is used for this and returns the part number + * as returned from the FRU. + * NOTE: This is only available for certain server cards + */ + +static ssize_t amdgpu_fru_product_number_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + + return sysfs_emit(buf, "%s\n", adev->product_number); +} + +static DEVICE_ATTR(product_number, 0444, amdgpu_fru_product_number_show, NULL); + +/** + * DOC: serial_number + * + * The amdgpu driver provides a sysfs API for reporting the serial number + * for the device + * The file serial_number is used for this and returns the serial number + * as returned from the FRU. + * NOTE: This is only available for certain server cards + */ + +static ssize_t amdgpu_fru_serial_number_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + + return sysfs_emit(buf, "%s\n", adev->serial); +} + +static DEVICE_ATTR(serial_number, 0444, amdgpu_fru_serial_number_show, NULL); + +static const struct attribute *amdgpu_fru_attributes[] = { + &dev_attr_product_name.attr, + &dev_attr_product_number.attr, + &dev_attr_serial_number.attr, + NULL +}; + +int amdgpu_fru_sysfs_init(struct amdgpu_device *adev) +{ + if (!is_fru_eeprom_supported(adev, NULL)) + return 0; + + return sysfs_create_files(&adev->dev->kobj, amdgpu_fru_attributes); +} + +void amdgpu_fru_sysfs_fini(struct amdgpu_device *adev) +{ + if (!is_fru_eeprom_supported(adev, NULL)) + return; + + sysfs_remove_files(&adev->dev->kobj, amdgpu_fru_attributes); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h index 1308d976d60e..c817db17cfa7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fru_eeprom.h @@ -25,5 +25,7 @@ #define __AMDGPU_FRU_EEPROM_H__ int amdgpu_fru_get_product_info(struct amdgpu_device *adev); +int amdgpu_fru_sysfs_init(struct amdgpu_device *adev); +void amdgpu_fru_sysfs_fini(struct amdgpu_device *adev); #endif // __AMDGPU_FRU_EEPROM_H__ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c index 2ca3c329de6d..2d4b67175b55 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fw_attestation.c @@ -32,17 +32,15 @@ #include "soc15_common.h" #define FW_ATTESTATION_DB_COOKIE 0x143b6a37 -#define FW_ATTESTATION_RECORD_VALID 1 +#define FW_ATTESTATION_RECORD_VALID 1 #define FW_ATTESTATION_MAX_SIZE 4096 -typedef struct FW_ATT_DB_HEADER -{ +struct FW_ATT_DB_HEADER { uint32_t AttDbVersion; /* version of the fwar feature */ uint32_t AttDbCookie; /* cookie as an extra check for corrupt data */ -} FW_ATT_DB_HEADER; +}; -typedef struct FW_ATT_RECORD -{ +struct FW_ATT_RECORD { uint16_t AttFwIdV1; /* Legacy FW Type field */ uint16_t AttFwIdV2; /* V2 FW ID field */ uint32_t AttFWVersion; /* FW Version */ @@ -50,7 +48,7 @@ typedef struct FW_ATT_RECORD uint8_t AttSource; /* FW source indicator */ uint8_t RecordValid; /* Indicates whether the record is a valid entry */ uint32_t AttFwTaId; /* Ta ID (only in TA Attestation Table) */ -} FW_ATT_RECORD; +}; static ssize_t amdgpu_fw_attestation_debugfs_read(struct file *f, char __user *buf, @@ -60,15 +58,15 @@ static ssize_t amdgpu_fw_attestation_debugfs_read(struct file *f, struct amdgpu_device *adev = (struct amdgpu_device *)file_inode(f)->i_private; uint64_t records_addr = 0; uint64_t vram_pos = 0; - FW_ATT_DB_HEADER fw_att_hdr = {0}; - FW_ATT_RECORD fw_att_record = {0}; + struct FW_ATT_DB_HEADER fw_att_hdr = {0}; + struct FW_ATT_RECORD fw_att_record = {0}; - if (size < sizeof(FW_ATT_RECORD)) { + if (size < sizeof(struct FW_ATT_RECORD)) { DRM_WARN("FW attestation input buffer not enough memory"); return -EINVAL; } - if ((*pos + sizeof(FW_ATT_DB_HEADER)) >= FW_ATTESTATION_MAX_SIZE) { + if ((*pos + sizeof(struct FW_ATT_DB_HEADER)) >= FW_ATTESTATION_MAX_SIZE) { DRM_WARN("FW attestation out of bounds"); return 0; } @@ -83,8 +81,8 @@ static ssize_t amdgpu_fw_attestation_debugfs_read(struct file *f, if (*pos == 0) { amdgpu_device_vram_access(adev, vram_pos, - (uint32_t*)&fw_att_hdr, - sizeof(FW_ATT_DB_HEADER), + (uint32_t *)&fw_att_hdr, + sizeof(struct FW_ATT_DB_HEADER), false); if (fw_att_hdr.AttDbCookie != FW_ATTESTATION_DB_COOKIE) { @@ -96,20 +94,20 @@ static ssize_t amdgpu_fw_attestation_debugfs_read(struct file *f, } amdgpu_device_vram_access(adev, - vram_pos + sizeof(FW_ATT_DB_HEADER) + *pos, - (uint32_t*)&fw_att_record, - sizeof(FW_ATT_RECORD), + vram_pos + sizeof(struct FW_ATT_DB_HEADER) + *pos, + (uint32_t *)&fw_att_record, + sizeof(struct FW_ATT_RECORD), false); if (fw_att_record.RecordValid != FW_ATTESTATION_RECORD_VALID) return 0; - if (copy_to_user(buf, (void*)&fw_att_record, sizeof(FW_ATT_RECORD))) + if (copy_to_user(buf, (void *)&fw_att_record, sizeof(struct FW_ATT_RECORD))) return -EINVAL; - *pos += sizeof(FW_ATT_RECORD); + *pos += sizeof(struct FW_ATT_RECORD); - return sizeof(FW_ATT_RECORD); + return sizeof(struct FW_ATT_RECORD); } static const struct file_operations amdgpu_fw_attestation_debugfs_ops = { @@ -136,7 +134,7 @@ void amdgpu_fw_attestation_debugfs_init(struct amdgpu_device *adev) return; debugfs_create_file("amdgpu_fw_attestation", - S_IRUSR, + 0400, adev_to_drm(adev)->primary->debugfs_root, adev, &amdgpu_fw_attestation_debugfs_ops); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 7da871972a8e..ca4d2d430e28 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -182,11 +182,10 @@ static int amdgpu_gem_object_open(struct drm_gem_object *obj, return r; bo_va = amdgpu_vm_bo_find(vm, abo); - if (!bo_va) { + if (!bo_va) bo_va = amdgpu_vm_bo_add(adev, vm, abo); - } else { + else ++bo_va->ref_count; - } amdgpu_bo_unreserve(abo); return 0; } @@ -290,6 +289,10 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data, uint32_t handle, initial_domain; int r; + /* reject DOORBELLs until userspace code to use it is available */ + if (args->in.domains & AMDGPU_GEM_DOMAIN_DOORBELL) + return -EINVAL; + /* reject invalid gem flags */ if (flags & ~(AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED | AMDGPU_GEM_CREATE_NO_CPU_ACCESS | @@ -461,9 +464,9 @@ int amdgpu_mode_dumb_mmap(struct drm_file *filp, struct amdgpu_bo *robj; gobj = drm_gem_object_lookup(filp, handle); - if (gobj == NULL) { + if (!gobj) return -ENOENT; - } + robj = gem_to_amdgpu_bo(gobj); if (amdgpu_ttm_tt_get_usermm(robj->tbo.ttm) || (robj->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)) { @@ -480,6 +483,7 @@ int amdgpu_gem_mmap_ioctl(struct drm_device *dev, void *data, { union drm_amdgpu_gem_mmap *args = data; uint32_t handle = args->in.handle; + memset(args, 0, sizeof(*args)); return amdgpu_mode_dumb_mmap(filp, dev, handle, &args->out.addr_ptr); } @@ -506,7 +510,7 @@ unsigned long amdgpu_gem_timeout(uint64_t timeout_ns) timeout_jiffies = nsecs_to_jiffies(ktime_to_ns(timeout)); /* clamp timeout to avoid unsigned-> signed overflow */ - if (timeout_jiffies > MAX_SCHEDULE_TIMEOUT ) + if (timeout_jiffies > MAX_SCHEDULE_TIMEOUT) return MAX_SCHEDULE_TIMEOUT - 1; return timeout_jiffies; @@ -524,9 +528,9 @@ int amdgpu_gem_wait_idle_ioctl(struct drm_device *dev, void *data, long ret; gobj = drm_gem_object_lookup(filp, handle); - if (gobj == NULL) { + if (!gobj) return -ENOENT; - } + robj = gem_to_amdgpu_bo(gobj); ret = dma_resv_wait_timeout(robj->tbo.base.resv, DMA_RESV_USAGE_READ, true, timeout); @@ -553,7 +557,7 @@ int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data, struct amdgpu_bo *robj; int r = -1; - DRM_DEBUG("%d \n", args->handle); + DRM_DEBUG("%d\n", args->handle); gobj = drm_gem_object_lookup(filp, args->handle); if (gobj == NULL) return -ENOENT; @@ -680,7 +684,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, if (args->va_address < AMDGPU_VA_RESERVED_SIZE) { dev_dbg(dev->dev, - "va_address 0x%LX is in reserved area 0x%LX\n", + "va_address 0x%llx is in reserved area 0x%llx\n", args->va_address, AMDGPU_VA_RESERVED_SIZE); return -EINVAL; } @@ -688,7 +692,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, if (args->va_address >= AMDGPU_GMC_HOLE_START && args->va_address < AMDGPU_GMC_HOLE_END) { dev_dbg(dev->dev, - "va_address 0x%LX is in VA hole 0x%LX-0x%LX\n", + "va_address 0x%llx is in VA hole 0x%llx-0x%llx\n", args->va_address, AMDGPU_GMC_HOLE_START, AMDGPU_GMC_HOLE_END); return -EINVAL; @@ -808,9 +812,9 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data, int r; gobj = drm_gem_object_lookup(filp, args->handle); - if (gobj == NULL) { + if (!gobj) return -ENOENT; - } + robj = gem_to_amdgpu_bo(gobj); r = amdgpu_bo_reserve(robj, false); @@ -936,9 +940,9 @@ int amdgpu_mode_dumb_create(struct drm_file *file_priv, r = drm_gem_handle_create(file_priv, gobj, &handle); /* drop reference from allocate - handle holds it now */ drm_gem_object_put(gobj); - if (r) { + if (r) return r; - } + args->handle = handle; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index a33d4bc34cee..2382921710ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -110,9 +110,9 @@ bool amdgpu_gfx_is_me_queue_enabled(struct amdgpu_device *adev, * The bitmask of CUs to be disabled in the shader array determined by se and * sh is stored in mask[se * max_sh + sh]. */ -void amdgpu_gfx_parse_disable_cu(unsigned *mask, unsigned max_se, unsigned max_sh) +void amdgpu_gfx_parse_disable_cu(unsigned int *mask, unsigned int max_se, unsigned int max_sh) { - unsigned se, sh, cu; + unsigned int se, sh, cu; const char *p; memset(mask, 0, sizeof(*mask) * max_se * max_sh); @@ -124,6 +124,7 @@ void amdgpu_gfx_parse_disable_cu(unsigned *mask, unsigned max_se, unsigned max_s for (;;) { char *next; int ret = sscanf(p, "%u.%u.%u", &se, &sh, &cu); + if (ret < 3) { DRM_ERROR("amdgpu: could not parse disable_cu\n"); return; @@ -349,7 +350,7 @@ void amdgpu_gfx_kiq_fini(struct amdgpu_device *adev, int xcc_id) } int amdgpu_gfx_kiq_init(struct amdgpu_device *adev, - unsigned hpd_size, int xcc_id) + unsigned int hpd_size, int xcc_id) { int r; u32 *hpd; @@ -376,7 +377,7 @@ int amdgpu_gfx_kiq_init(struct amdgpu_device *adev, /* create MQD for each compute/gfx queue */ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, - unsigned mqd_size, int xcc_id) + unsigned int mqd_size, int xcc_id) { int r, i, j; struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_id]; @@ -407,8 +408,11 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, /* prepare MQD backup */ kiq->mqd_backup = kmalloc(mqd_size, GFP_KERNEL); - if (!kiq->mqd_backup) - dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); + if (!kiq->mqd_backup) { + dev_warn(adev->dev, + "no memory to create MQD backup for ring %s\n", ring->name); + return -ENOMEM; + } } if (adev->asic_type >= CHIP_NAVI10 && amdgpu_async_gfx_ring) { @@ -427,8 +431,10 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, ring->mqd_size = mqd_size; /* prepare MQD backup */ adev->gfx.me.mqd_backup[i] = kmalloc(mqd_size, GFP_KERNEL); - if (!adev->gfx.me.mqd_backup[i]) + if (!adev->gfx.me.mqd_backup[i]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); + return -ENOMEM; + } } } } @@ -449,8 +455,10 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, ring->mqd_size = mqd_size; /* prepare MQD backup */ adev->gfx.mec.mqd_backup[j] = kmalloc(mqd_size, GFP_KERNEL); - if (!adev->gfx.mec.mqd_backup[j]) + if (!adev->gfx.mec.mqd_backup[j]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); + return -ENOMEM; + } } } @@ -692,15 +700,8 @@ void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable) if (adev->gfx.gfx_off_req_count == 0 && !adev->gfx.gfx_off_state) { - /* If going to s2idle, no need to wait */ - if (adev->in_s0ix) { - if (!amdgpu_dpm_set_powergating_by_smu(adev, - AMD_IP_BLOCK_TYPE_GFX, true)) - adev->gfx.gfx_off_state = true; - } else { - schedule_delayed_work(&adev->gfx.gfx_off_delay_work, + schedule_delayed_work(&adev->gfx.gfx_off_delay_work, delay); - } } } else { if (adev->gfx.gfx_off_req_count == 0) { @@ -1281,11 +1282,11 @@ static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev, return sysfs_emit(buf, "%s\n", supported_partition); } -static DEVICE_ATTR(current_compute_partition, S_IRUGO | S_IWUSR, +static DEVICE_ATTR(current_compute_partition, 0644, amdgpu_gfx_get_current_compute_partition, amdgpu_gfx_set_compute_partition); -static DEVICE_ATTR(available_compute_partition, S_IRUGO, +static DEVICE_ATTR(available_compute_partition, 0444, amdgpu_gfx_get_available_compute_partition, NULL); int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index a4ff515ce896..0ca95c4d4bfb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -43,6 +43,7 @@ #define AMDGPU_GFX_LBPW_DISABLED_MODE 0x00000008L #define AMDGPU_MAX_GC_INSTANCES 8 +#define KGD_MAX_QUEUES 128 #define AMDGPU_MAX_GFX_QUEUES KGD_MAX_QUEUES #define AMDGPU_MAX_COMPUTE_QUEUES KGD_MAX_QUEUES @@ -241,6 +242,9 @@ struct amdgpu_gfx_config { uint32_t gc_gl1c_per_sa; uint32_t gc_gl1c_size_per_instance; uint32_t gc_gl2c_per_gpu; + uint32_t gc_tcp_size_per_cu; + uint32_t gc_num_cu_per_sqc; + uint32_t gc_tcc_size; }; struct amdgpu_cu_info { @@ -254,7 +258,7 @@ struct amdgpu_cu_info { uint32_t number; uint32_t ao_cu_mask; uint32_t ao_cu_bitmap[4][4]; - uint32_t bitmap[4][4]; + uint32_t bitmap[AMDGPU_MAX_GC_INSTANCES][4][4]; }; struct amdgpu_gfx_ras { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 56d73fade568..fdc25cd559b6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -331,6 +331,8 @@ struct amdgpu_gmc { u64 VM_CONTEXT_PAGE_TABLE_END_ADDR_LO32[16]; u64 VM_CONTEXT_PAGE_TABLE_END_ADDR_HI32[16]; u64 MC_VM_MX_L1_TLB_CNTL; + + u64 noretry_flags; }; #define amdgpu_gmc_flush_gpu_tlb(adev, vmid, vmhub, type) ((adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (vmhub), (type))) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c index ebeddc9a37e9..6aa3b1d845ab 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c @@ -62,7 +62,7 @@ * Returns 0 on success, error on failure. */ int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm, - unsigned size, enum amdgpu_ib_pool_type pool_type, + unsigned int size, enum amdgpu_ib_pool_type pool_type, struct amdgpu_ib *ib) { int r; @@ -123,7 +123,7 @@ void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, * a CONST_IB), it will be put on the ring prior to the DE IB. Prior * to SI there was just a DE IB. */ -int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, +int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs, struct amdgpu_ib *ibs, struct amdgpu_job *job, struct dma_fence **f) { @@ -131,16 +131,16 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, struct amdgpu_ib *ib = &ibs[0]; struct dma_fence *tmp = NULL; bool need_ctx_switch; - unsigned patch_offset = ~0; + unsigned int patch_offset = ~0; struct amdgpu_vm *vm; uint64_t fence_ctx; uint32_t status = 0, alloc_size; - unsigned fence_flags = 0; + unsigned int fence_flags = 0; bool secure, init_shadow; u64 shadow_va, csa_va, gds_va; int vmid = AMDGPU_JOB_GET_VMID(job); - unsigned i; + unsigned int i; int r = 0; bool need_pipe_sync = false; @@ -282,7 +282,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs, amdgpu_ring_emit_gfx_shadow(ring, 0, 0, 0, false, 0); if (ring->funcs->init_cond_exec) { - unsigned ce_offset = ~0; + unsigned int ce_offset = ~0; ce_offset = amdgpu_ring_init_cond_exec(ring); if (ce_offset != ~0 && ring->funcs->patch_cond_exec) @@ -385,7 +385,7 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev) { long tmo_gfx, tmo_mm; int r, ret = 0; - unsigned i; + unsigned int i; tmo_mm = tmo_gfx = AMDGPU_IB_TEST_TIMEOUT; if (amdgpu_sriov_vf(adev)) { @@ -402,7 +402,7 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev) /* for CP & SDMA engines since they are scheduled together so * need to make the timeout width enough to cover the time * cost waiting for it coming back under RUNTIME only - */ + */ tmo_gfx = 8 * AMDGPU_IB_TEST_TIMEOUT; } else if (adev->gmc.xgmi.hive_id) { tmo_gfx = AMDGPU_IB_TEST_GFX_XGMI_TIMEOUT; @@ -465,13 +465,13 @@ static int amdgpu_debugfs_sa_info_show(struct seq_file *m, void *unused) { struct amdgpu_device *adev = m->private; - seq_printf(m, "--------------------- DELAYED --------------------- \n"); + seq_puts(m, "--------------------- DELAYED ---------------------\n"); amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DELAYED], m); - seq_printf(m, "-------------------- IMMEDIATE -------------------- \n"); + seq_puts(m, "-------------------- IMMEDIATE --------------------\n"); amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_IMMEDIATE], m); - seq_printf(m, "--------------------- DIRECT ---------------------- \n"); + seq_puts(m, "--------------------- DIRECT ----------------------\n"); amdgpu_sa_bo_dump_debug_info(&adev->ib_pools[AMDGPU_IB_POOL_DIRECT], m); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index fceb3b384955..f3b0aaf3ebc6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -138,6 +138,7 @@ void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih) /** * amdgpu_ih_ring_write - write IV to the ring buffer * + * @adev: amdgpu_device pointer * @ih: ih ring to write to * @iv: the iv to write * @num_dw: size of the iv in dw @@ -145,8 +146,8 @@ void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih) * Writes an IV to the ring buffer using the CPU and increment the wptr. * Used for testing and delegating IVs to a software ring. */ -void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv, - unsigned int num_dw) +void amdgpu_ih_ring_write(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, + const uint32_t *iv, unsigned int num_dw) { uint32_t wptr = le32_to_cpu(*ih->wptr_cpu) >> 2; unsigned int i; @@ -161,6 +162,9 @@ void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv, if (wptr != READ_ONCE(ih->rptr)) { wmb(); WRITE_ONCE(*ih->wptr_cpu, cpu_to_le32(wptr)); + } else if (adev->irq.retry_cam_enabled) { + dev_warn_once(adev->dev, "IH soft ring buffer overflow 0x%X, 0x%X\n", + wptr, ih->rptr); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h index dd1c2eded6b9..6c6184f0dbc1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h @@ -27,6 +27,9 @@ /* Maximum number of IVs processed at once */ #define AMDGPU_IH_MAX_NUM_IVS 32 +#define IH_RING_SIZE (256 * 1024) +#define IH_SW_RING_SIZE (8 * 1024) /* enough for 256 CAM entries */ + struct amdgpu_device; struct amdgpu_iv_entry; @@ -97,8 +100,8 @@ struct amdgpu_ih_funcs { int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, unsigned ring_size, bool use_bus_addr); void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); -void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv, - unsigned int num_dw); +void amdgpu_ih_ring_write(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, + const uint32_t *iv, unsigned int num_dw); int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 5273decc5753..fa6d0adcec20 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -493,7 +493,7 @@ void amdgpu_irq_delegate(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry, unsigned int num_dw) { - amdgpu_ih_ring_write(&adev->irq.ih_soft, entry->iv_entry, num_dw); + amdgpu_ih_ring_write(adev, &adev->irq.ih_soft, entry->iv_entry, num_dw); schedule_work(&adev->irq.ih_soft_work); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index cca5a495611f..d30dc0b718c7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -557,6 +557,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) crtc = (struct drm_crtc *)minfo->crtcs[i]; if (crtc && crtc->base.id == info->mode_crtc.id) { struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); + ui32 = amdgpu_crtc->crtc_id; found = 1; break; @@ -575,7 +576,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) if (ret) return ret; - ret = copy_to_user(out, &ip, min((size_t)size, sizeof(ip))); + ret = copy_to_user(out, &ip, min_t(size_t, size, sizeof(ip))); return ret ? -EFAULT : 0; } case AMDGPU_INFO_HW_IP_COUNT: { @@ -721,17 +722,18 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) ? -EFAULT : 0; } case AMDGPU_INFO_READ_MMR_REG: { - unsigned n, alloc_size; + unsigned int n, alloc_size; uint32_t *regs; - unsigned se_num = (info->read_mmr_reg.instance >> + unsigned int se_num = (info->read_mmr_reg.instance >> AMDGPU_INFO_MMR_SE_INDEX_SHIFT) & AMDGPU_INFO_MMR_SE_INDEX_MASK; - unsigned sh_num = (info->read_mmr_reg.instance >> + unsigned int sh_num = (info->read_mmr_reg.instance >> AMDGPU_INFO_MMR_SH_INDEX_SHIFT) & AMDGPU_INFO_MMR_SH_INDEX_MASK; /* set full masks if the userspace set all bits - * in the bitfields */ + * in the bitfields + */ if (se_num == AMDGPU_INFO_MMR_SE_INDEX_MASK) se_num = 0xffffffff; else if (se_num >= AMDGPU_GFX_MAX_SE) @@ -837,7 +839,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) memcpy(&dev_info->cu_ao_bitmap[0], &adev->gfx.cu_info.ao_cu_bitmap[0], sizeof(adev->gfx.cu_info.ao_cu_bitmap)); memcpy(&dev_info->cu_bitmap[0], &adev->gfx.cu_info.bitmap[0], - sizeof(adev->gfx.cu_info.bitmap)); + sizeof(dev_info->cu_bitmap)); dev_info->vram_type = adev->gmc.vram_type; dev_info->vram_bit_width = adev->gmc.vram_width; dev_info->vce_harvest_config = adev->vce.harvest_config; @@ -896,7 +898,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return ret; } case AMDGPU_INFO_VCE_CLOCK_TABLE: { - unsigned i; + unsigned int i; struct drm_amdgpu_info_vce_clock_table vce_clk_table = {}; struct amd_vce_state *vce_state; @@ -938,12 +940,17 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) struct atom_context *atom_context; atom_context = adev->mode_info.atom_context; - memcpy(vbios_info.name, atom_context->name, sizeof(atom_context->name)); - memcpy(vbios_info.vbios_pn, atom_context->vbios_pn, sizeof(atom_context->vbios_pn)); - vbios_info.version = atom_context->version; - memcpy(vbios_info.vbios_ver_str, atom_context->vbios_ver_str, - sizeof(atom_context->vbios_ver_str)); - memcpy(vbios_info.date, atom_context->date, sizeof(atom_context->date)); + if (atom_context) { + memcpy(vbios_info.name, atom_context->name, + sizeof(atom_context->name)); + memcpy(vbios_info.vbios_pn, atom_context->vbios_pn, + sizeof(atom_context->vbios_pn)); + vbios_info.version = atom_context->version; + memcpy(vbios_info.vbios_ver_str, atom_context->vbios_ver_str, + sizeof(atom_context->vbios_ver_str)); + memcpy(vbios_info.date, atom_context->date, + sizeof(atom_context->date)); + } return copy_to_user(out, &vbios_info, min((size_t)size, sizeof(vbios_info))) ? -EFAULT : 0; @@ -1017,7 +1024,7 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) case AMDGPU_INFO_SENSOR_GPU_AVG_POWER: /* get average GPU power */ if (amdgpu_dpm_read_sensor(adev, - AMDGPU_PP_SENSOR_GPU_POWER, + AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&ui32, &ui32_size)) { return -EINVAL; } @@ -1102,6 +1109,9 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) struct drm_amdgpu_info_video_caps *caps; int r; + if (!adev->asic_funcs->query_video_codecs) + return -EINVAL; + switch (info->video_cap.type) { case AMDGPU_INFO_VIDEO_CAPS_DECODE: r = amdgpu_asic_query_video_codecs(adev, false, &codecs); @@ -1229,13 +1239,13 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) pasid = 0; } - r = amdgpu_vm_init(adev, &fpriv->vm); + r = amdgpu_xcp_open_device(adev, fpriv, file_priv); if (r) goto error_pasid; - r = amdgpu_xcp_open_device(adev, fpriv, file_priv); + r = amdgpu_vm_init(adev, &fpriv->vm, fpriv->xcp_id); if (r) - goto error_vm; + goto error_pasid; r = amdgpu_vm_set_pasid(adev, &fpriv->vm, pasid); if (r) @@ -1719,7 +1729,7 @@ static int amdgpu_debugfs_firmware_info_show(struct seq_file *m, void *unused) seq_printf(m, "MES feature version: %u, firmware version: 0x%08x\n", fw_info.feature, fw_info.ver); - seq_printf(m, "VBIOS version: %s\n", ctx->vbios_version); + seq_printf(m, "VBIOS version: %s\n", ctx->vbios_pn); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index 3e3dadd6d0f3..b6015157763a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -39,120 +39,70 @@ int amdgpu_mes_doorbell_process_slice(struct amdgpu_device *adev) PAGE_SIZE); } -int amdgpu_mes_alloc_process_doorbells(struct amdgpu_device *adev, - unsigned int *doorbell_index) -{ - int r = ida_simple_get(&adev->mes.doorbell_ida, 2, - adev->mes.max_doorbell_slices, - GFP_KERNEL); - if (r > 0) - *doorbell_index = r; - - return r; -} - -void amdgpu_mes_free_process_doorbells(struct amdgpu_device *adev, - unsigned int doorbell_index) -{ - if (doorbell_index) - ida_simple_remove(&adev->mes.doorbell_ida, doorbell_index); -} - -unsigned int amdgpu_mes_get_doorbell_dw_offset_in_bar( - struct amdgpu_device *adev, - uint32_t doorbell_index, - unsigned int doorbell_id) -{ - return ((doorbell_index * - amdgpu_mes_doorbell_process_slice(adev)) / sizeof(u32) + - doorbell_id * 2); -} - -static int amdgpu_mes_queue_doorbell_get(struct amdgpu_device *adev, +static int amdgpu_mes_kernel_doorbell_get(struct amdgpu_device *adev, struct amdgpu_mes_process *process, int ip_type, uint64_t *doorbell_index) { unsigned int offset, found; + struct amdgpu_mes *mes = &adev->mes; - if (ip_type == AMDGPU_RING_TYPE_SDMA) { + if (ip_type == AMDGPU_RING_TYPE_SDMA) offset = adev->doorbell_index.sdma_engine[0]; - found = find_next_zero_bit(process->doorbell_bitmap, - AMDGPU_MES_MAX_NUM_OF_QUEUES_PER_PROCESS, - offset); - } else { - found = find_first_zero_bit(process->doorbell_bitmap, - AMDGPU_MES_MAX_NUM_OF_QUEUES_PER_PROCESS); - } + else + offset = 0; - if (found >= AMDGPU_MES_MAX_NUM_OF_QUEUES_PER_PROCESS) { + found = find_next_zero_bit(mes->doorbell_bitmap, mes->num_mes_dbs, offset); + if (found >= mes->num_mes_dbs) { DRM_WARN("No doorbell available\n"); return -ENOSPC; } - set_bit(found, process->doorbell_bitmap); - - *doorbell_index = amdgpu_mes_get_doorbell_dw_offset_in_bar(adev, - process->doorbell_index, found); + set_bit(found, mes->doorbell_bitmap); + /* Get the absolute doorbell index on BAR */ + *doorbell_index = mes->db_start_dw_offset + found * 2; return 0; } -static void amdgpu_mes_queue_doorbell_free(struct amdgpu_device *adev, +static void amdgpu_mes_kernel_doorbell_free(struct amdgpu_device *adev, struct amdgpu_mes_process *process, uint32_t doorbell_index) { - unsigned int old, doorbell_id; + unsigned int old, rel_index; + struct amdgpu_mes *mes = &adev->mes; - doorbell_id = doorbell_index - - (process->doorbell_index * - amdgpu_mes_doorbell_process_slice(adev)) / sizeof(u32); - doorbell_id /= 2; - - old = test_and_clear_bit(doorbell_id, process->doorbell_bitmap); + /* Find the relative index of the doorbell in this object */ + rel_index = (doorbell_index - mes->db_start_dw_offset) / 2; + old = test_and_clear_bit(rel_index, mes->doorbell_bitmap); WARN_ON(!old); } static int amdgpu_mes_doorbell_init(struct amdgpu_device *adev) { - size_t doorbell_start_offset; - size_t doorbell_aperture_size; - size_t doorbell_process_limit; - size_t aggregated_doorbell_start; int i; + struct amdgpu_mes *mes = &adev->mes; - aggregated_doorbell_start = (adev->doorbell_index.max_assignment + 1) * sizeof(u32); - aggregated_doorbell_start = - roundup(aggregated_doorbell_start, PAGE_SIZE); - - doorbell_start_offset = aggregated_doorbell_start + PAGE_SIZE; - doorbell_start_offset = - roundup(doorbell_start_offset, - amdgpu_mes_doorbell_process_slice(adev)); - - doorbell_aperture_size = adev->doorbell.size; - doorbell_aperture_size = - rounddown(doorbell_aperture_size, - amdgpu_mes_doorbell_process_slice(adev)); - - if (doorbell_aperture_size > doorbell_start_offset) - doorbell_process_limit = - (doorbell_aperture_size - doorbell_start_offset) / - amdgpu_mes_doorbell_process_slice(adev); - else - return -ENOSPC; - - adev->mes.doorbell_id_offset = doorbell_start_offset / sizeof(u32); - adev->mes.max_doorbell_slices = doorbell_process_limit; + /* Bitmap for dynamic allocation of kernel doorbells */ + mes->doorbell_bitmap = bitmap_zalloc(PAGE_SIZE / sizeof(u32), GFP_KERNEL); + if (!mes->doorbell_bitmap) { + DRM_ERROR("Failed to allocate MES doorbell bitmap\n"); + return -ENOMEM; + } - /* allocate Qword range for aggregated doorbell */ - for (i = 0; i < AMDGPU_MES_PRIORITY_NUM_LEVELS; i++) - adev->mes.aggregated_doorbells[i] = - aggregated_doorbell_start / sizeof(u32) + i * 2; + mes->num_mes_dbs = PAGE_SIZE / AMDGPU_ONE_DOORBELL_SIZE; + for (i = 0; i < AMDGPU_MES_PRIORITY_NUM_LEVELS; i++) { + adev->mes.aggregated_doorbells[i] = mes->db_start_dw_offset + i * 2; + set_bit(i, mes->doorbell_bitmap); + } - DRM_INFO("max_doorbell_slices=%zu\n", doorbell_process_limit); return 0; } +static void amdgpu_mes_doorbell_free(struct amdgpu_device *adev) +{ + bitmap_free(adev->mes.doorbell_bitmap); +} + int amdgpu_mes_init(struct amdgpu_device *adev) { int i, r; @@ -251,6 +201,7 @@ void amdgpu_mes_fini(struct amdgpu_device *adev) amdgpu_device_wb_free(adev, adev->mes.sch_ctx_offs); amdgpu_device_wb_free(adev, adev->mes.query_status_fence_offs); amdgpu_device_wb_free(adev, adev->mes.read_val_offs); + amdgpu_mes_doorbell_free(adev); idr_destroy(&adev->mes.pasid_idr); idr_destroy(&adev->mes.gang_id_idr); @@ -279,15 +230,6 @@ int amdgpu_mes_create_process(struct amdgpu_device *adev, int pasid, return -ENOMEM; } - process->doorbell_bitmap = - kzalloc(DIV_ROUND_UP(AMDGPU_MES_MAX_NUM_OF_QUEUES_PER_PROCESS, - BITS_PER_BYTE), GFP_KERNEL); - if (!process->doorbell_bitmap) { - DRM_ERROR("failed to allocate doorbell bitmap\n"); - kfree(process); - return -ENOMEM; - } - /* allocate the process context bo and map it */ r = amdgpu_bo_create_kernel(adev, AMDGPU_MES_PROC_CTX_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, @@ -314,15 +256,6 @@ int amdgpu_mes_create_process(struct amdgpu_device *adev, int pasid, goto clean_up_ctx; } - /* allocate the starting doorbell index of the process */ - r = amdgpu_mes_alloc_process_doorbells(adev, &process->doorbell_index); - if (r < 0) { - DRM_ERROR("failed to allocate doorbell for process\n"); - goto clean_up_pasid; - } - - DRM_DEBUG("process doorbell index = %d\n", process->doorbell_index); - INIT_LIST_HEAD(&process->gang_list); process->vm = vm; process->pasid = pasid; @@ -332,15 +265,12 @@ int amdgpu_mes_create_process(struct amdgpu_device *adev, int pasid, amdgpu_mes_unlock(&adev->mes); return 0; -clean_up_pasid: - idr_remove(&adev->mes.pasid_idr, pasid); - amdgpu_mes_unlock(&adev->mes); clean_up_ctx: + amdgpu_mes_unlock(&adev->mes); amdgpu_bo_free_kernel(&process->proc_ctx_bo, &process->proc_ctx_gpu_addr, &process->proc_ctx_cpu_ptr); clean_up_memory: - kfree(process->doorbell_bitmap); kfree(process); return r; } @@ -386,7 +316,6 @@ void amdgpu_mes_destroy_process(struct amdgpu_device *adev, int pasid) idr_remove(&adev->mes.gang_id_idr, gang->gang_id); } - amdgpu_mes_free_process_doorbells(adev, process->doorbell_index); idr_remove(&adev->mes.pasid_idr, pasid); amdgpu_mes_unlock(&adev->mes); @@ -408,7 +337,6 @@ void amdgpu_mes_destroy_process(struct amdgpu_device *adev, int pasid) amdgpu_bo_free_kernel(&process->proc_ctx_bo, &process->proc_ctx_gpu_addr, &process->proc_ctx_cpu_ptr); - kfree(process->doorbell_bitmap); kfree(process); } @@ -643,6 +571,8 @@ int amdgpu_mes_add_hw_queue(struct amdgpu_device *adev, int gang_id, unsigned long flags; int r; + memset(&queue_input, 0, sizeof(struct mes_add_queue_input)); + /* allocate the mes queue buffer */ queue = kzalloc(sizeof(struct amdgpu_mes_queue), GFP_KERNEL); if (!queue) { @@ -680,7 +610,7 @@ int amdgpu_mes_add_hw_queue(struct amdgpu_device *adev, int gang_id, *queue_id = queue->queue_id = r; /* allocate a doorbell index for the queue */ - r = amdgpu_mes_queue_doorbell_get(adev, gang->process, + r = amdgpu_mes_kernel_doorbell_get(adev, gang->process, qprops->queue_type, &qprops->doorbell_off); if (r) @@ -738,7 +668,7 @@ int amdgpu_mes_add_hw_queue(struct amdgpu_device *adev, int gang_id, return 0; clean_up_doorbell: - amdgpu_mes_queue_doorbell_free(adev, gang->process, + amdgpu_mes_kernel_doorbell_free(adev, gang->process, qprops->doorbell_off); clean_up_queue_id: spin_lock_irqsave(&adev->mes.queue_id_lock, flags); @@ -793,7 +723,7 @@ int amdgpu_mes_remove_hw_queue(struct amdgpu_device *adev, int queue_id) queue_id); list_del(&queue->list); - amdgpu_mes_queue_doorbell_free(adev, gang->process, + amdgpu_mes_kernel_doorbell_free(adev, gang->process, queue->doorbell_off); amdgpu_mes_unlock(&adev->mes); @@ -1378,7 +1308,7 @@ int amdgpu_mes_self_test(struct amdgpu_device *adev) goto error_pasid; } - r = amdgpu_vm_init(adev, vm); + r = amdgpu_vm_init(adev, vm, -1); if (r) { DRM_ERROR("failed to initialize vm\n"); goto error_pasid; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h index 2d6ac30b7135..a27b424ffe00 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h @@ -27,6 +27,7 @@ #include "amdgpu_irq.h" #include "kgd_kfd_interface.h" #include "amdgpu_gfx.h" +#include "amdgpu_doorbell.h" #include <linux/sched/mm.h> #define AMDGPU_MES_MAX_COMPUTE_PIPES 8 @@ -76,7 +77,6 @@ struct amdgpu_mes { uint32_t kiq_version; uint32_t total_max_queue; - uint32_t doorbell_id_offset; uint32_t max_doorbell_slices; uint64_t default_process_quantum; @@ -128,6 +128,11 @@ struct amdgpu_mes { int (*kiq_hw_init)(struct amdgpu_device *adev); int (*kiq_hw_fini)(struct amdgpu_device *adev); + /* MES doorbells */ + uint32_t db_start_dw_offset; + uint32_t num_mes_dbs; + unsigned long *doorbell_bitmap; + /* ip specific functions */ const struct amdgpu_mes_funcs *funcs; }; @@ -142,7 +147,6 @@ struct amdgpu_mes_process { uint64_t process_quantum; struct list_head gang_list; uint32_t doorbell_index; - unsigned long *doorbell_bitmap; struct mutex doorbell_lock; }; @@ -224,6 +228,7 @@ struct mes_add_queue_input { uint32_t is_kfd_process; uint32_t is_aql_queue; uint32_t queue_size; + uint32_t exclusively_scheduled; }; struct mes_remove_queue_input { @@ -386,14 +391,6 @@ int amdgpu_mes_ctx_unmap_meta_data(struct amdgpu_device *adev, int amdgpu_mes_self_test(struct amdgpu_device *adev); -int amdgpu_mes_alloc_process_doorbells(struct amdgpu_device *adev, - unsigned int *doorbell_index); -void amdgpu_mes_free_process_doorbells(struct amdgpu_device *adev, - unsigned int doorbell_index); -unsigned int amdgpu_mes_get_doorbell_dw_offset_in_bar( - struct amdgpu_device *adev, - uint32_t doorbell_index, - unsigned int doorbell_id); int amdgpu_mes_doorbell_process_slice(struct amdgpu_device *adev); /* diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index a3bc00577a7c..51ca544a7094 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -45,6 +45,22 @@ int amdgpu_nbio_ras_sw_init(struct amdgpu_device *adev) return 0; } +u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev) +{ + if (adev->nbio.funcs && adev->nbio.funcs->get_pcie_replay_count) + return adev->nbio.funcs->get_pcie_replay_count(adev); + + return 0; +} + +void amdgpu_nbio_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0, + uint64_t *count1) +{ + if (adev->nbio.funcs->get_pcie_usage) + adev->nbio.funcs->get_pcie_usage(adev, count0, count1); + +} + int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) { int r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index 8ab8ae01f87c..6cf7a8829a52 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -99,6 +99,9 @@ struct amdgpu_nbio_funcs { int (*get_compute_partition_mode)(struct amdgpu_device *adev); u32 (*get_memory_partition_mode)(struct amdgpu_device *adev, u32 *supp_modes); + u64 (*get_pcie_replay_count)(struct amdgpu_device *adev); + void (*get_pcie_usage)(struct amdgpu_device *adev, uint64_t *count0, + uint64_t *count1); }; struct amdgpu_nbio { @@ -111,5 +114,8 @@ struct amdgpu_nbio { }; int amdgpu_nbio_ras_sw_init(struct amdgpu_device *adev); +void amdgpu_nbio_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0, uint64_t *count1); int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); +u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index f7905bce0de1..ace837cfa0a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -158,6 +158,14 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain) c++; } + if (domain & AMDGPU_GEM_DOMAIN_DOORBELL) { + places[c].fpfn = 0; + places[c].lpfn = 0; + places[c].mem_type = AMDGPU_PL_DOORBELL; + places[c].flags = 0; + c++; + } + if (domain & AMDGPU_GEM_DOMAIN_GTT) { places[c].fpfn = 0; places[c].lpfn = 0; @@ -477,7 +485,7 @@ static bool amdgpu_bo_validate_size(struct amdgpu_device *adev, goto fail; } - /* TODO add more domains checks, such as AMDGPU_GEM_DOMAIN_CPU */ + /* TODO add more domains checks, such as AMDGPU_GEM_DOMAIN_CPU, _DOMAIN_DOORBELL */ return true; fail: @@ -1029,6 +1037,7 @@ void amdgpu_bo_unpin(struct amdgpu_bo *bo) } else if (bo->tbo.resource->mem_type == TTM_PL_TT) { atomic64_sub(amdgpu_bo_size(bo), &adev->gart_pin_size); } + } static const char * const amdgpu_vram_names[] = { @@ -1575,23 +1584,31 @@ u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m) { struct dma_buf_attachment *attachment; struct dma_buf *dma_buf; - unsigned int domain; const char *placement; unsigned int pin_count; u64 size; - domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type); - switch (domain) { - case AMDGPU_GEM_DOMAIN_VRAM: - placement = "VRAM"; - break; - case AMDGPU_GEM_DOMAIN_GTT: - placement = " GTT"; - break; - case AMDGPU_GEM_DOMAIN_CPU: - default: - placement = " CPU"; - break; + if (dma_resv_trylock(bo->tbo.base.resv)) { + unsigned int domain; + domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type); + switch (domain) { + case AMDGPU_GEM_DOMAIN_VRAM: + if (amdgpu_bo_in_cpu_visible_vram(bo)) + placement = "VRAM VISIBLE"; + else + placement = "VRAM"; + break; + case AMDGPU_GEM_DOMAIN_GTT: + placement = "GTT"; + break; + case AMDGPU_GEM_DOMAIN_CPU: + default: + placement = "CPU"; + break; + } + dma_resv_unlock(bo->tbo.base.resv); + } else { + placement = "UNKNOWN"; } size = amdgpu_bo_size(bo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 05496b97ef93..f3ee83cdf97e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -182,6 +182,8 @@ static inline unsigned amdgpu_mem_type_to_domain(u32 mem_type) return AMDGPU_GEM_DOMAIN_GWS; case AMDGPU_PL_OA: return AMDGPU_GEM_DOMAIN_OA; + case AMDGPU_PL_DOORBELL: + return AMDGPU_GEM_DOMAIN_DOORBELL; default: break; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c index 71ee361d0972..6e91ea1de5aa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c @@ -276,9 +276,8 @@ static void amdgpu_perf_read(struct perf_event *event) (!pe->adev->df.funcs->pmc_get_count)) return; + prev = local64_read(&hwc->prev_count); do { - prev = local64_read(&hwc->prev_count); - switch (hwc->config_base) { case AMDGPU_PMU_EVENT_CONFIG_TYPE_DF: case AMDGPU_PMU_EVENT_CONFIG_TYPE_XGMI: @@ -289,7 +288,7 @@ static void amdgpu_perf_read(struct perf_event *event) count = 0; break; } - } while (local64_cmpxchg(&hwc->prev_count, prev, count) != prev); + } while (!local64_try_cmpxchg(&hwc->prev_count, &prev, count)); local64_add(count - prev, &event->count); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 6d676bdd1505..429ef212c1f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -45,9 +45,6 @@ #define AMD_VBIOS_FILE_MAX_SIZE_B (1024*1024*3) -static int psp_sysfs_init(struct amdgpu_device *adev); -static void psp_sysfs_fini(struct amdgpu_device *adev); - static int psp_load_smu_fw(struct psp_context *psp); static int psp_rap_terminate(struct psp_context *psp); static int psp_securedisplay_terminate(struct psp_context *psp); @@ -148,6 +145,7 @@ static int psp_init_sriov_microcode(struct psp_context *psp) break; case IP_VERSION(13, 0, 6): ret = psp_init_cap_microcode(psp, ucode_prefix); + ret &= psp_init_ta_microcode(psp, ucode_prefix); break; case IP_VERSION(13, 0, 10): adev->virt.autoload_ucode_id = AMDGPU_UCODE_ID_CP_MES1_DATA; @@ -180,9 +178,11 @@ static int psp_early_init(void *handle) psp->autoload_supported = false; break; case IP_VERSION(11, 0, 0): + case IP_VERSION(11, 0, 7): + adev->psp.sup_pd_fw_up = !amdgpu_sriov_vf(adev); + fallthrough; case IP_VERSION(11, 0, 5): case IP_VERSION(11, 0, 9): - case IP_VERSION(11, 0, 7): case IP_VERSION(11, 0, 11): case IP_VERSION(11, 5, 0): case IP_VERSION(11, 0, 12): @@ -202,8 +202,8 @@ static int psp_early_init(void *handle) case IP_VERSION(13, 0, 3): case IP_VERSION(13, 0, 5): case IP_VERSION(13, 0, 8): - case IP_VERSION(13, 0, 10): case IP_VERSION(13, 0, 11): + case IP_VERSION(14, 0, 0): psp_v13_0_set_psp_funcs(psp); psp->autoload_supported = true; break; @@ -215,8 +215,10 @@ static int psp_early_init(void *handle) break; case IP_VERSION(13, 0, 0): case IP_VERSION(13, 0, 7): + case IP_VERSION(13, 0, 10): psp_v13_0_set_psp_funcs(psp); psp->autoload_supported = true; + adev->psp.sup_ifwi_up = !amdgpu_sriov_vf(adev); break; case IP_VERSION(13, 0, 4): psp_v13_0_4_set_psp_funcs(psp); @@ -437,14 +439,15 @@ static int psp_sw_init(void *handle) /* If psp runtime database exists, then * only enable two stage memory training * when TWO_STAGE_DRAM_TRAINING bit is set - * in runtime database */ + * in runtime database + */ mem_training_ctx->enable_mem_training = true; } } else { - /* If psp runtime database doesn't exist or - * is invalid, force enable two stage memory - * training */ + /* If psp runtime database doesn't exist or is + * invalid, force enable two stage memory training + */ mem_training_ctx->enable_mem_training = true; } @@ -462,13 +465,6 @@ static int psp_sw_init(void *handle) } } - if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) || - adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7)) { - ret = psp_sysfs_init(adev); - if (ret) - return ret; - } - ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG, amdgpu_sriov_vf(adev) ? AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT, @@ -498,11 +494,11 @@ static int psp_sw_init(void *handle) return 0; failed2: - amdgpu_bo_free_kernel(&psp->fw_pri_bo, - &psp->fw_pri_mc_addr, &psp->fw_pri_buf); -failed1: amdgpu_bo_free_kernel(&psp->fence_buf_bo, &psp->fence_buf_mc_addr, &psp->fence_buf); +failed1: + amdgpu_bo_free_kernel(&psp->fw_pri_bo, + &psp->fw_pri_mc_addr, &psp->fw_pri_buf); return ret; } @@ -520,10 +516,6 @@ static int psp_sw_fini(void *handle) amdgpu_ucode_release(&psp->cap_fw); amdgpu_ucode_release(&psp->toc_fw); - if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 0) || - adev->ip_versions[MP0_HWIP][0] == IP_VERSION(11, 0, 7)) - psp_sysfs_fini(adev); - kfree(cmd); cmd = NULL; @@ -807,7 +799,8 @@ static int psp_tmr_init(struct psp_context *psp) tmr_size = PSP_TMR_SIZE(psp->adev); /* For ASICs support RLC autoload, psp will parse the toc - * and calculate the total size of TMR needed */ + * and calculate the total size of TMR needed + */ if (!amdgpu_sriov_vf(psp->adev) && psp->toc.start_addr && psp->toc.size_bytes && @@ -1147,9 +1140,9 @@ int psp_ta_init_shared_buf(struct psp_context *psp, struct ta_mem_context *mem_ctx) { /* - * Allocate 16k memory aligned to 4k from Frame Buffer (local - * physical) for ta to host memory - */ + * Allocate 16k memory aligned to 4k from Frame Buffer (local + * physical) for ta to host memory + */ return amdgpu_bo_create_kernel(psp->adev, mem_ctx->shared_mem_size, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT, @@ -1738,7 +1731,8 @@ int psp_ras_trigger_error(struct psp_context *psp, return -EINVAL; /* If err_event_athub occurs error inject was successful, however - return status from TA is no long reliable */ + * return status from TA is no long reliable + */ if (amdgpu_ras_intr_triggered()) return 0; @@ -2084,6 +2078,17 @@ int psp_securedisplay_invoke(struct psp_context *psp, uint32_t ta_cmd_id) } /* SECUREDISPLAY end */ +int amdgpu_psp_wait_for_bootloader(struct amdgpu_device *adev) +{ + struct psp_context *psp = &adev->psp; + int ret = 0; + + if (!amdgpu_sriov_vf(adev) && psp->funcs && psp->funcs->wait_for_bootloader != NULL) + ret = psp->funcs->wait_for_bootloader(psp); + + return ret; +} + static int psp_hw_start(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; @@ -2459,8 +2464,8 @@ static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode, return ret; } -static int psp_execute_non_psp_fw_load(struct psp_context *psp, - struct amdgpu_firmware_info *ucode) +int psp_execute_ip_fw_load(struct psp_context *psp, + struct amdgpu_firmware_info *ucode) { int ret = 0; struct psp_gfx_cmd_resp *cmd = acquire_psp_cmd_buf(psp); @@ -2503,7 +2508,7 @@ static int psp_load_smu_fw(struct psp_context *psp) DRM_WARN("Failed to set MP1 state prepare for reload\n"); } - ret = psp_execute_non_psp_fw_load(psp, ucode); + ret = psp_execute_ip_fw_load(psp, ucode); if (ret) DRM_ERROR("PSP load smu failed!\n"); @@ -2545,7 +2550,7 @@ int psp_load_fw_list(struct psp_context *psp, for (i = 0; i < ucode_count; ++i) { ucode = ucode_list[i]; psp_print_fw_hdr(psp, ucode); - ret = psp_execute_non_psp_fw_load(psp, ucode); + ret = psp_execute_ip_fw_load(psp, ucode); if (ret) return ret; } @@ -2587,12 +2592,13 @@ static int psp_load_non_psp_fw(struct psp_context *psp) ucode->ucode_id == AMDGPU_UCODE_ID_SDMA2 || ucode->ucode_id == AMDGPU_UCODE_ID_SDMA3)) /* PSP only receive one SDMA fw for sienna_cichlid, - * as all four sdma fw are same */ + * as all four sdma fw are same + */ continue; psp_print_fw_hdr(psp, ucode); - ret = psp_execute_non_psp_fw_load(psp, ucode); + ret = psp_execute_ip_fw_load(psp, ucode); if (ret) return ret; @@ -2652,8 +2658,8 @@ static int psp_load_fw(struct amdgpu_device *adev) if (adev->gmc.xgmi.num_physical_nodes > 1) { ret = psp_xgmi_initialize(psp, false, true); /* Warning the XGMI seesion initialize failure - * Instead of stop driver initialization - */ + * Instead of stop driver initialization + */ if (ret) dev_err(psp->adev->dev, "XGMI: Failed to initialize XGMI session\n"); @@ -2931,19 +2937,6 @@ int psp_rlc_autoload_start(struct psp_context *psp) return ret; } -int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx, - uint64_t cmd_gpu_addr, int cmd_size) -{ - struct amdgpu_firmware_info ucode = {0}; - - ucode.ucode_id = inst_idx ? AMDGPU_UCODE_ID_VCN1_RAM : - AMDGPU_UCODE_ID_VCN0_RAM; - ucode.mc_addr = cmd_gpu_addr; - ucode.ucode_size = cmd_size; - - return psp_execute_non_psp_fw_load(&adev->psp, &ucode); -} - int psp_ring_cmd_submit(struct psp_context *psp, uint64_t cmd_buf_mc_addr, uint64_t fence_mc_addr, @@ -3584,6 +3577,11 @@ void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size drm_dev_exit(idx); } +/** + * DOC: usbc_pd_fw + * Reading from this file will retrieve the USB-C PD firmware version. Writing to + * this file will trigger the update process. + */ static DEVICE_ATTR(usbc_pd_fw, 0644, psp_usbc_pd_fw_sysfs_read, psp_usbc_pd_fw_sysfs_write); @@ -3624,7 +3622,7 @@ static ssize_t amdgpu_psp_vbflash_write(struct file *filp, struct kobject *kobj, adev->psp.vbflash_image_size += count; mutex_unlock(&adev->psp.mutex); - dev_info(adev->dev, "VBIOS flash write PSP done"); + dev_dbg(adev->dev, "IFWI staged for update"); return count; } @@ -3644,7 +3642,7 @@ static ssize_t amdgpu_psp_vbflash_read(struct file *filp, struct kobject *kobj, if (adev->psp.vbflash_image_size == 0) return -EINVAL; - dev_info(adev->dev, "VBIOS flash to PSP started"); + dev_dbg(adev->dev, "PSP IFWI flash process initiated"); ret = amdgpu_bo_create_kernel(adev, adev->psp.vbflash_image_size, AMDGPU_GPU_PAGE_SIZE, @@ -3669,14 +3667,32 @@ rel_buf: adev->psp.vbflash_image_size = 0; if (ret) { - dev_err(adev->dev, "Failed to load VBIOS FW, err = %d", ret); + dev_err(adev->dev, "Failed to load IFWI, err = %d", ret); return ret; } - dev_info(adev->dev, "VBIOS flash to PSP done"); + dev_dbg(adev->dev, "PSP IFWI flash process done"); return 0; } +/** + * DOC: psp_vbflash + * Writing to this file will stage an IFWI for update. Reading from this file + * will trigger the update process. + */ +static struct bin_attribute psp_vbflash_bin_attr = { + .attr = {.name = "psp_vbflash", .mode = 0660}, + .size = 0, + .write = amdgpu_psp_vbflash_write, + .read = amdgpu_psp_vbflash_read, +}; + +/** + * DOC: psp_vbflash_status + * The status of the flash process. + * 0: IFWI flash not complete. + * 1: IFWI flash complete. + */ static ssize_t amdgpu_psp_vbflash_status(struct device *dev, struct device_attribute *attr, char *buf) @@ -3693,39 +3709,49 @@ static ssize_t amdgpu_psp_vbflash_status(struct device *dev, return sysfs_emit(buf, "0x%x\n", vbflash_status); } +static DEVICE_ATTR(psp_vbflash_status, 0440, amdgpu_psp_vbflash_status, NULL); -static const struct bin_attribute psp_vbflash_bin_attr = { - .attr = {.name = "psp_vbflash", .mode = 0660}, - .size = 0, - .write = amdgpu_psp_vbflash_write, - .read = amdgpu_psp_vbflash_read, +static struct bin_attribute *bin_flash_attrs[] = { + &psp_vbflash_bin_attr, + NULL }; -static DEVICE_ATTR(psp_vbflash_status, 0440, amdgpu_psp_vbflash_status, NULL); +static struct attribute *flash_attrs[] = { + &dev_attr_psp_vbflash_status.attr, + &dev_attr_usbc_pd_fw.attr, + NULL +}; -int amdgpu_psp_sysfs_init(struct amdgpu_device *adev) +static umode_t amdgpu_flash_attr_is_visible(struct kobject *kobj, struct attribute *attr, int idx) { - int ret = 0; + struct device *dev = kobj_to_dev(kobj); + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); - if (amdgpu_sriov_vf(adev)) - return -EINVAL; + if (attr == &dev_attr_usbc_pd_fw.attr) + return adev->psp.sup_pd_fw_up ? 0660 : 0; - switch (adev->ip_versions[MP0_HWIP][0]) { - case IP_VERSION(13, 0, 0): - case IP_VERSION(13, 0, 7): - case IP_VERSION(13, 0, 10): - ret = sysfs_create_bin_file(&adev->dev->kobj, &psp_vbflash_bin_attr); - if (ret) - dev_err(adev->dev, "Failed to create device file psp_vbflash"); - ret = device_create_file(adev->dev, &dev_attr_psp_vbflash_status); - if (ret) - dev_err(adev->dev, "Failed to create device file psp_vbflash_status"); - return ret; - default: - return 0; - } + return adev->psp.sup_ifwi_up ? 0440 : 0; } +static umode_t amdgpu_bin_flash_attr_is_visible(struct kobject *kobj, + struct bin_attribute *attr, + int idx) +{ + struct device *dev = kobj_to_dev(kobj); + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + + return adev->psp.sup_ifwi_up ? 0660 : 0; +} + +const struct attribute_group amdgpu_flash_attr_group = { + .attrs = flash_attrs, + .bin_attrs = bin_flash_attrs, + .is_bin_visible = amdgpu_bin_flash_attr_is_visible, + .is_visible = amdgpu_flash_attr_is_visible, +}; + const struct amd_ip_funcs psp_ip_funcs = { .name = "psp", .early_init = psp_early_init, @@ -3744,27 +3770,6 @@ const struct amd_ip_funcs psp_ip_funcs = { .set_powergating_state = psp_set_powergating_state, }; -static int psp_sysfs_init(struct amdgpu_device *adev) -{ - int ret = device_create_file(adev->dev, &dev_attr_usbc_pd_fw); - - if (ret) - DRM_ERROR("Failed to create USBC PD FW control file!"); - - return ret; -} - -void amdgpu_psp_sysfs_fini(struct amdgpu_device *adev) -{ - sysfs_remove_bin_file(&adev->dev->kobj, &psp_vbflash_bin_attr); - device_remove_file(adev->dev, &dev_attr_psp_vbflash_status); -} - -static void psp_sysfs_fini(struct amdgpu_device *adev) -{ - device_remove_file(adev->dev, &dev_attr_usbc_pd_fw); -} - const struct amdgpu_ip_block_version psp_v3_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_PSP, .major = 3, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 2cae0b1a0b8a..3e67ed63e638 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -39,6 +39,8 @@ #define PSP_TMR_ALIGNMENT 0x100000 #define PSP_FW_NAME_LEN 0x24 +extern const struct attribute_group amdgpu_flash_attr_group; + enum psp_shared_mem_size { PSP_ASD_SHARED_MEM_SIZE = 0x0, PSP_XGMI_SHARED_MEM_SIZE = 0x4000, @@ -78,8 +80,7 @@ enum psp_bootloader_cmd { PSP_BL__LOAD_TOS_SPL_TABLE = 0x10000000, }; -enum psp_ring_type -{ +enum psp_ring_type { PSP_RING_TYPE__INVALID = 0, /* * These values map to the way the PSP kernel identifies the @@ -89,8 +90,7 @@ enum psp_ring_type PSP_RING_TYPE__KM = 2 /* Kernel mode ring (formerly called GPCOM) */ }; -struct psp_ring -{ +struct psp_ring { enum psp_ring_type ring_type; struct psp_gfx_rb_frame *ring_mem; uint64_t ring_mem_mc_addr; @@ -107,9 +107,9 @@ enum psp_reg_prog_id { PSP_REG_LAST }; -struct psp_funcs -{ +struct psp_funcs { int (*init_microcode)(struct psp_context *psp); + int (*wait_for_bootloader)(struct psp_context *psp); int (*bootloader_load_kdb)(struct psp_context *psp); int (*bootloader_load_spl)(struct psp_context *psp); int (*bootloader_load_sysdrv)(struct psp_context *psp); @@ -133,6 +133,7 @@ struct psp_funcs int (*read_usbc_pd_fw)(struct psp_context *psp, uint32_t *fw_ver); int (*update_spirom)(struct psp_context *psp, uint64_t fw_pri_mc_addr); int (*vbflash_stat)(struct psp_context *psp); + int (*fatal_error_recovery_quirk)(struct psp_context *psp); }; struct ta_funcs { @@ -307,10 +308,9 @@ struct psp_runtime_scpm_entry { enum psp_runtime_scpm_authentication scpm_status; }; -struct psp_context -{ - struct amdgpu_device *adev; - struct psp_ring km_ring; +struct psp_context { + struct amdgpu_device *adev; + struct psp_ring km_ring; struct psp_gfx_cmd_resp *cmd; const struct psp_funcs *funcs; @@ -339,7 +339,7 @@ struct psp_context uint64_t tmr_mc_addr; /* asd firmware */ - const struct firmware *asd_fw; + const struct firmware *asd_fw; /* toc firmware */ const struct firmware *toc_fw; @@ -384,9 +384,13 @@ struct psp_context uint32_t boot_cfg_bitmask; - char *vbflash_tmp_buf; - size_t vbflash_image_size; - bool vbflash_done; + /* firmware upgrades supported */ + bool sup_pd_fw_up; + bool sup_ifwi_up; + + char *vbflash_tmp_buf; + size_t vbflash_image_size; + bool vbflash_done; }; struct amdgpu_psp_funcs { @@ -443,6 +447,10 @@ struct amdgpu_psp_funcs { ((psp)->funcs->vbflash_stat ? \ (psp)->funcs->vbflash_stat((psp)) : -EINVAL) +#define psp_fatal_error_recovery_quirk(psp) \ + ((psp)->funcs->fatal_error_recovery_quirk ? \ + (psp)->funcs->fatal_error_recovery_quirk((psp)) : 0) + extern const struct amd_ip_funcs psp_ip_funcs; extern const struct amdgpu_ip_block_version psp_v3_1_ip_block; @@ -458,9 +466,10 @@ extern int psp_wait_for(struct psp_context *psp, uint32_t reg_index, extern int psp_wait_for_spirom_update(struct psp_context *psp, uint32_t reg_index, uint32_t field_val, uint32_t mask, uint32_t msec_timeout); +int psp_execute_ip_fw_load(struct psp_context *psp, + struct amdgpu_firmware_info *ucode); + int psp_gpu_reset(struct amdgpu_device *adev); -int psp_update_vcn_sram(struct amdgpu_device *adev, int inst_idx, - uint64_t cmd_gpu_addr, int cmd_size); int psp_ta_init_shared_buf(struct psp_context *psp, struct ta_mem_context *mem_ctx); @@ -525,6 +534,6 @@ int psp_spatial_partition(struct psp_context *psp, int mode); int is_psp_fw_valid(struct psp_bin_desc bin); -int amdgpu_psp_sysfs_init(struct amdgpu_device *adev); -void amdgpu_psp_sysfs_fini(struct amdgpu_device *adev); +int amdgpu_psp_wait_for_bootloader(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 8aaa427f8c0f..937c54fc7174 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -35,6 +35,7 @@ #include "amdgpu_xgmi.h" #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h" #include "nbio_v4_3.h" +#include "nbio_v7_9.h" #include "atom.h" #include "amdgpu_reset.h" @@ -757,28 +758,27 @@ static int __amdgpu_ras_feature_enable(struct amdgpu_device *adev, return 0; } -static int amdgpu_ras_check_feature_allowed(struct amdgpu_device *adev, - struct ras_common_if *head) -{ - if (amdgpu_ras_is_feature_allowed(adev, head) || - amdgpu_ras_is_poison_mode_supported(adev)) - return 1; - else - return 0; -} - /* wrapper of psp_ras_enable_features */ int amdgpu_ras_feature_enable(struct amdgpu_device *adev, struct ras_common_if *head, bool enable) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); union ta_ras_cmd_input *info; - int ret = 0; + int ret; if (!con) return -EINVAL; - if (head->block == AMDGPU_RAS_BLOCK__GFX) { + /* Do not enable ras feature if it is not allowed */ + if (enable && + head->block != AMDGPU_RAS_BLOCK__GFX && + !amdgpu_ras_is_feature_allowed(adev, head)) + return 0; + + /* Only enable gfx ras feature from host side */ + if (head->block == AMDGPU_RAS_BLOCK__GFX && + !amdgpu_sriov_vf(adev) && + !amdgpu_ras_intr_triggered()) { info = kzalloc(sizeof(union ta_ras_cmd_input), GFP_KERNEL); if (!info) return -ENOMEM; @@ -794,32 +794,23 @@ int amdgpu_ras_feature_enable(struct amdgpu_device *adev, .error_type = amdgpu_ras_error_to_ta(head->type), }; } - } - - /* Do not enable if it is not allowed. */ - if (enable && !amdgpu_ras_check_feature_allowed(adev, head)) - goto out; - /* Only enable ras feature operation handle on host side */ - if (head->block == AMDGPU_RAS_BLOCK__GFX && - !amdgpu_sriov_vf(adev) && - !amdgpu_ras_intr_triggered()) { ret = psp_ras_enable_features(&adev->psp, info, enable); if (ret) { dev_err(adev->dev, "ras %s %s failed poison:%d ret:%d\n", enable ? "enable":"disable", get_ras_block_str(head), amdgpu_ras_is_poison_mode_supported(adev), ret); - goto out; + return ret; } + + kfree(info); } /* setup the obj */ __amdgpu_ras_feature_enable(adev, head, enable); -out: - if (head->block == AMDGPU_RAS_BLOCK__GFX) - kfree(info); - return ret; + + return 0; } /* Only used in device probe stage and called only once. */ @@ -1061,7 +1052,8 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev, info->ce_count = obj->err_data.ce_count; if (err_data.ce_count) { - if (adev->smuio.funcs && + if (!adev->aid_mask && + adev->smuio.funcs && adev->smuio.funcs->get_socket_id && adev->smuio.funcs->get_die_id) { dev_info(adev->dev, "socket: %d, die: %d " @@ -1081,7 +1073,8 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev, } } if (err_data.ue_count) { - if (adev->smuio.funcs && + if (!adev->aid_mask && + adev->smuio.funcs && adev->smuio.funcs->get_socket_id && adev->smuio.funcs->get_die_id) { dev_info(adev->dev, "socket: %d, die: %d " @@ -1159,7 +1152,8 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, } /* Calculate XGMI relative offset */ - if (adev->gmc.xgmi.num_physical_nodes > 1) { + if (adev->gmc.xgmi.num_physical_nodes > 1 && + info->head.block != AMDGPU_RAS_BLOCK__GFX) { block_info.address = amdgpu_xgmi_get_relative_phy_addr(adev, block_info.address); @@ -2072,6 +2066,8 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) if (ras->gpu_reset_flags & AMDGPU_RAS_GPU_RESET_MODE1_RESET) { ras->gpu_reset_flags &= ~AMDGPU_RAS_GPU_RESET_MODE1_RESET; set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); + + psp_fatal_error_recovery_quirk(&adev->psp); } } @@ -2405,6 +2401,7 @@ static bool amdgpu_ras_asic_supported(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) { switch (adev->ip_versions[MP0_HWIP][0]) { case IP_VERSION(13, 0, 2): + case IP_VERSION(13, 0, 6): return true; default: return false; @@ -2414,6 +2411,7 @@ static bool amdgpu_ras_asic_supported(struct amdgpu_device *adev) if (adev->asic_type == CHIP_IP_DISCOVERY) { switch (adev->ip_versions[MP0_HWIP][0]) { case IP_VERSION(13, 0, 0): + case IP_VERSION(13, 0, 6): case IP_VERSION(13, 0, 10): return true; default: @@ -2440,10 +2438,10 @@ static void amdgpu_ras_get_quirks(struct amdgpu_device *adev) if (!ctx) return; - if (strnstr(ctx->vbios_version, "D16406", - sizeof(ctx->vbios_version)) || - strnstr(ctx->vbios_version, "D36002", - sizeof(ctx->vbios_version))) + if (strnstr(ctx->vbios_pn, "D16406", + sizeof(ctx->vbios_pn)) || + strnstr(ctx->vbios_pn, "D36002", + sizeof(ctx->vbios_pn))) adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__GFX); } @@ -2515,8 +2513,18 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev) /* hw_supported needs to be aligned with RAS block mask. */ adev->ras_hw_enabled &= AMDGPU_RAS_BLOCK_MASK; - adev->ras_enabled = amdgpu_ras_enable == 0 ? 0 : - adev->ras_hw_enabled & amdgpu_ras_mask; + + /* + * Disable ras feature for aqua vanjaram + * by default on apu platform. + */ + if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 6) && + adev->gmc.is_app_apu) + adev->ras_enabled = amdgpu_ras_enable != 1 ? 0 : + adev->ras_hw_enabled & amdgpu_ras_mask; + else + adev->ras_enabled = amdgpu_ras_enable == 0 ? 0 : + adev->ras_hw_enabled & amdgpu_ras_mask; } static void amdgpu_ras_counte_dw(struct work_struct *work) @@ -2642,6 +2650,10 @@ int amdgpu_ras_init(struct amdgpu_device *adev) * check DF RAS */ adev->nbio.ras = &nbio_v4_3_ras; break; + case IP_VERSION(7, 9, 0): + if (!adev->gmc.is_app_apu) + adev->nbio.ras = &nbio_v7_9_ras; + break; default: /* nbio ras is not available */ break; @@ -2765,23 +2777,28 @@ int amdgpu_ras_block_late_init(struct amdgpu_device *adev, goto cleanup; } - r = amdgpu_ras_sysfs_create(adev, ras_block); - if (r) - goto interrupt; + if (ras_obj->hw_ops && + (ras_obj->hw_ops->query_ras_error_count || + ras_obj->hw_ops->query_ras_error_status)) { + r = amdgpu_ras_sysfs_create(adev, ras_block); + if (r) + goto interrupt; - /* Those are the cached values at init. - */ - query_info = kzalloc(sizeof(struct ras_query_if), GFP_KERNEL); - if (!query_info) - return -ENOMEM; - memcpy(&query_info->head, ras_block, sizeof(struct ras_common_if)); + /* Those are the cached values at init. + */ + query_info = kzalloc(sizeof(*query_info), GFP_KERNEL); + if (!query_info) + return -ENOMEM; + memcpy(&query_info->head, ras_block, sizeof(struct ras_common_if)); - if (amdgpu_ras_query_error_count(adev, &ce_count, &ue_count, query_info) == 0) { - atomic_set(&con->ras_ce_count, ce_count); - atomic_set(&con->ras_ue_count, ue_count); + if (amdgpu_ras_query_error_count(adev, &ce_count, &ue_count, query_info) == 0) { + atomic_set(&con->ras_ce_count, ce_count); + atomic_set(&con->ras_ue_count, ue_count); + } + + kfree(query_info); } - kfree(query_info); return 0; interrupt: @@ -2958,10 +2975,6 @@ int amdgpu_ras_fini(struct amdgpu_device *adev) void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev) { - amdgpu_ras_check_supported(adev); - if (!adev->ras_hw_enabled) - return; - if (atomic_cmpxchg(&amdgpu_ras_in_intr, 0, 1) == 0) { struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); @@ -3136,6 +3149,10 @@ int amdgpu_ras_is_supported(struct amdgpu_device *adev, * that the ras block supports ras function. */ if (!ret && + (block == AMDGPU_RAS_BLOCK__GFX || + block == AMDGPU_RAS_BLOCK__SDMA || + block == AMDGPU_RAS_BLOCK__VCN || + block == AMDGPU_RAS_BLOCK__JPEG) && amdgpu_ras_is_poison_mode_supported(adev) && amdgpu_ras_get_ras_block(adev, block, 0)) ret = 1; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c index 0648dfe559af..595d5e535aca 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c @@ -160,6 +160,8 @@ static bool __is_ras_eeprom_supported(struct amdgpu_device *adev) case IP_VERSION(13, 0, 2): /* Aldebaran */ case IP_VERSION(13, 0, 10): return true; + case IP_VERSION(13, 0, 6): + return (adev->gmc.is_app_apu) ? false : true; default: return false; } @@ -194,9 +196,9 @@ static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev, /* VEGA20 and ARCTURUS */ if (adev->asic_type == CHIP_VEGA20) control->i2c_address = EEPROM_I2C_MADDR_0; - else if (strnstr(atom_ctx->vbios_version, + else if (strnstr(atom_ctx->vbios_pn, "D342", - sizeof(atom_ctx->vbios_version))) + sizeof(atom_ctx->vbios_pn))) control->i2c_address = EEPROM_I2C_MADDR_0; else control->i2c_address = EEPROM_I2C_MADDR_4; @@ -205,13 +207,14 @@ static bool __get_eeprom_i2c_addr(struct amdgpu_device *adev, control->i2c_address = EEPROM_I2C_MADDR_0; return true; case IP_VERSION(13, 0, 2): - if (strnstr(atom_ctx->vbios_version, "D673", - sizeof(atom_ctx->vbios_version))) + if (strnstr(atom_ctx->vbios_pn, "D673", + sizeof(atom_ctx->vbios_pn))) control->i2c_address = EEPROM_I2C_MADDR_4; else control->i2c_address = EEPROM_I2C_MADDR_0; return true; case IP_VERSION(13, 0, 0): + case IP_VERSION(13, 0, 6): case IP_VERSION(13, 0, 10): control->i2c_address = EEPROM_I2C_MADDR_4; return true; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h index 5c4f93ee0c57..3c988cc406e4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h @@ -90,6 +90,7 @@ static inline void amdgpu_res_first(struct ttm_resource *res, cur->node = block; break; case TTM_PL_TT: + case AMDGPU_PL_DOORBELL: node = to_ttm_range_mgr_node(res)->mm_nodes; while (start >= node->size << PAGE_SHIFT) start -= node++->size << PAGE_SHIFT; @@ -152,6 +153,7 @@ static inline void amdgpu_res_next(struct amdgpu_res_cursor *cur, uint64_t size) cur->size = min(amdgpu_vram_mgr_block_size(block), cur->remaining); break; case TTM_PL_TT: + case AMDGPU_PL_DOORBELL: node = cur->node; cur->node = ++node; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c index eec41ad30406..5fed06ffcc6b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c @@ -87,7 +87,7 @@ int amdgpu_reset_prepare_hwcontext(struct amdgpu_device *adev, reset_handler = adev->reset_cntl->get_reset_handler( adev->reset_cntl, reset_context); if (!reset_handler) - return -ENOSYS; + return -EOPNOTSUPP; return reset_handler->prepare_hwcontext(adev->reset_cntl, reset_context); @@ -103,7 +103,7 @@ int amdgpu_reset_perform_reset(struct amdgpu_device *adev, reset_handler = adev->reset_cntl->get_reset_handler( adev->reset_cntl, reset_context); if (!reset_handler) - return -ENOSYS; + return -EOPNOTSUPP; ret = reset_handler->perform_reset(adev->reset_cntl, reset_context); if (ret) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 028ff075db51..e2ab303ad270 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -389,7 +389,7 @@ static inline void amdgpu_ring_write_multiple(struct amdgpu_ring *ring, occupied = ring->wptr & ring->buf_mask; dst = (void *)&ring->ring[occupied]; chunk1 = ring->buf_mask + 1 - occupied; - chunk1 = (chunk1 >= count_dw) ? count_dw: chunk1; + chunk1 = (chunk1 >= count_dw) ? count_dw : chunk1; chunk2 = count_dw - chunk1; chunk1 <<= 2; chunk2 <<= 2; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c index b779ee4bbaa7..e1ee1c7117fb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.c @@ -397,7 +397,7 @@ void amdgpu_sw_ring_ib_begin(struct amdgpu_ring *ring) struct amdgpu_ring_mux *mux = &adev->gfx.muxer; WARN_ON(!ring->is_sw_ring); - if (ring->hw_prio > AMDGPU_RING_PRIO_DEFAULT) { + if (adev->gfx.mcbp && ring->hw_prio > AMDGPU_RING_PRIO_DEFAULT) { if (amdgpu_mcbp_scan(mux) > 0) amdgpu_mcbp_trigger_preempt(mux); return; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h index b22d4fb2a847..d3186b570b82 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring_mux.h @@ -56,6 +56,15 @@ enum amdgpu_ring_mux_offset_type { AMDGPU_MUX_OFFSET_TYPE_CE, }; +enum ib_complete_status { + /* IB not started/reset value, default value. */ + IB_COMPLETION_STATUS_DEFAULT = 0, + /* IB preempted, started but not completed. */ + IB_COMPLETION_STATUS_PREEMPTED = 1, + /* IB completed. */ + IB_COMPLETION_STATUS_COMPLETED = 2, +}; + struct amdgpu_ring_mux { struct amdgpu_ring *real_ring; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h index 80b263646966..b591d33af264 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h @@ -26,6 +26,8 @@ #include "clearstate_defs.h" +#define AMDGPU_MAX_RLC_INSTANCES 8 + /* firmware ID used in rlc toc */ typedef enum _FIRMWARE_ID_ { FIRMWARE_ID_INVALID = 0, @@ -201,7 +203,7 @@ struct amdgpu_rlc { u32 cp_table_size; /* safe mode for updating CG/PG state */ - bool in_safe_mode[8]; + bool in_safe_mode[AMDGPU_MAX_RLC_INSTANCES]; const struct amdgpu_rlc_funcs *funcs; /* for firmware data */ @@ -257,7 +259,7 @@ struct amdgpu_rlc { bool rlcg_reg_access_supported; /* registers for rlcg indirect reg access */ - struct amdgpu_rlcg_reg_access_ctrl reg_access_ctrl; + struct amdgpu_rlcg_reg_access_ctrl reg_access_ctrl[AMDGPU_MAX_RLC_INSTANCES]; }; void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev, int xcc_id); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c index c6b4337eb20c..10df731998b2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c @@ -81,7 +81,7 @@ int amdgpu_sa_bo_new(struct amdgpu_sa_manager *sa_manager, unsigned int size) { struct drm_suballoc *sa = drm_suballoc_new(&sa_manager->base, size, - GFP_KERNEL, true, 0); + GFP_KERNEL, false, 0); if (IS_ERR(sa)) { *sa_bo = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index dacf281d2b21..e2b9392d7f0d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -239,9 +239,6 @@ int amdgpu_sdma_init_microcode(struct amdgpu_device *adev, sizeof(struct amdgpu_sdma_instance)); } - if (amdgpu_sriov_vf(adev)) - return 0; - DRM_DEBUG("psp_load == '%s'\n", adev->firmware.load_type == AMDGPU_FW_LOAD_PSP ? "true" : "false"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h index 525dffbe046a..2fd1bfb35916 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h @@ -432,7 +432,7 @@ TRACE_EVENT(amdgpu_vm_flush, ), TP_printk("ring=%s, id=%u, hub=%u, pd_addr=%010Lx", __get_str(ring), __entry->vmid, - __entry->vm_hub,__entry->pd_addr) + __entry->vm_hub, __entry->pd_addr) ); DECLARE_EVENT_CLASS(amdgpu_pasid, @@ -494,7 +494,7 @@ TRACE_EVENT(amdgpu_cs_bo_status, ); TRACE_EVENT(amdgpu_bo_move, - TP_PROTO(struct amdgpu_bo* bo, uint32_t new_placement, uint32_t old_placement), + TP_PROTO(struct amdgpu_bo *bo, uint32_t new_placement, uint32_t old_placement), TP_ARGS(bo, new_placement, old_placement), TP_STRUCT__entry( __field(struct amdgpu_bo *, bo) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 0534ab716809..4e51dce3aab5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -49,7 +49,6 @@ #include <drm/ttm/ttm_tt.h> #include <drm/amdgpu_drm.h> -#include <drm/drm_drv.h> #include "amdgpu.h" #include "amdgpu_object.h" @@ -127,6 +126,7 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo, case AMDGPU_PL_GDS: case AMDGPU_PL_GWS: case AMDGPU_PL_OA: + case AMDGPU_PL_DOORBELL: placement->num_placement = 0; placement->num_busy_placement = 0; return; @@ -496,9 +496,11 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict, if (old_mem->mem_type == AMDGPU_PL_GDS || old_mem->mem_type == AMDGPU_PL_GWS || old_mem->mem_type == AMDGPU_PL_OA || + old_mem->mem_type == AMDGPU_PL_DOORBELL || new_mem->mem_type == AMDGPU_PL_GDS || new_mem->mem_type == AMDGPU_PL_GWS || - new_mem->mem_type == AMDGPU_PL_OA) { + new_mem->mem_type == AMDGPU_PL_OA || + new_mem->mem_type == AMDGPU_PL_DOORBELL) { /* Nothing to save here */ ttm_bo_move_null(bo, new_mem); goto out; @@ -582,6 +584,12 @@ static int amdgpu_ttm_io_mem_reserve(struct ttm_device *bdev, mem->bus.offset += adev->gmc.aper_base; mem->bus.is_iomem = true; break; + case AMDGPU_PL_DOORBELL: + mem->bus.offset = mem->start << PAGE_SHIFT; + mem->bus.offset += adev->doorbell.base; + mem->bus.is_iomem = true; + mem->bus.caching = ttm_uncached; + break; default: return -EINVAL; } @@ -596,6 +604,10 @@ static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo, amdgpu_res_first(bo->resource, (u64)page_offset << PAGE_SHIFT, 0, &cursor); + + if (bo->resource->mem_type == AMDGPU_PL_DOORBELL) + return ((uint64_t)(adev->doorbell.base + cursor.start)) >> PAGE_SHIFT; + return (adev->gmc.aper_base + cursor.start) >> PAGE_SHIFT; } @@ -1305,6 +1317,7 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem) flags |= AMDGPU_PTE_VALID; if (mem && (mem->mem_type == TTM_PL_TT || + mem->mem_type == AMDGPU_PL_DOORBELL || mem->mem_type == AMDGPU_PL_PREEMPT)) { flags |= AMDGPU_PTE_SYSTEM; @@ -1924,6 +1937,20 @@ int amdgpu_ttm_init(struct amdgpu_device *adev) DRM_INFO("amdgpu: %uM of GTT memory ready.\n", (unsigned int)(gtt_size / (1024 * 1024))); + /* Initiailize doorbell pool on PCI BAR */ + r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_DOORBELL, adev->doorbell.size / PAGE_SIZE); + if (r) { + DRM_ERROR("Failed initializing doorbell heap.\n"); + return r; + } + + /* Create a boorbell page for kernel usages */ + r = amdgpu_doorbell_create_kernel_doorbells(adev); + if (r) { + DRM_ERROR("Failed to initialize kernel doorbells.\n"); + return r; + } + /* Initialize preemptible memory pool */ r = amdgpu_preempt_mgr_init(adev); if (r) { @@ -2392,7 +2419,7 @@ static ssize_t amdgpu_iomem_read(struct file *f, char __user *buf, struct page *p; void *ptr; - bytes = bytes < size ? bytes : size; + bytes = min(bytes, size); /* Translate the bus address to a physical address. If * the domain is NULL it means there is no IOMMU active @@ -2447,7 +2474,7 @@ static ssize_t amdgpu_iomem_write(struct file *f, const char __user *buf, struct page *p; void *ptr; - bytes = bytes < size ? bytes : size; + bytes = min(bytes, size); addr = dom ? iommu_iova_to_phys(dom, addr) : addr; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 6d0d66e40db9..65ec82141a8e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -33,12 +33,16 @@ #define AMDGPU_PL_GWS (TTM_PL_PRIV + 1) #define AMDGPU_PL_OA (TTM_PL_PRIV + 2) #define AMDGPU_PL_PREEMPT (TTM_PL_PRIV + 3) +#define AMDGPU_PL_DOORBELL (TTM_PL_PRIV + 4) #define AMDGPU_GTT_MAX_TRANSFER_SIZE 512 #define AMDGPU_GTT_NUM_TRANSFER_WINDOWS 2 #define AMDGPU_POISON 0xd0bed0be +extern const struct attribute_group amdgpu_vram_mgr_attr_group; +extern const struct attribute_group amdgpu_gtt_mgr_attr_group; + struct hmm_range; struct amdgpu_gtt_mgr { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index 16807ff96dc9..8beefc045e14 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -703,6 +703,8 @@ FW_VERSION_ATTR(sdma_fw_version, 0444, sdma.instance[0].fw_version); FW_VERSION_ATTR(sdma2_fw_version, 0444, sdma.instance[1].fw_version); FW_VERSION_ATTR(vcn_fw_version, 0444, vcn.fw_version); FW_VERSION_ATTR(dmcu_fw_version, 0444, dm.dmcu_fw_version); +FW_VERSION_ATTR(mes_fw_version, 0444, mes.sched_version & AMDGPU_MES_VERSION_MASK); +FW_VERSION_ATTR(mes_kiq_fw_version, 0444, mes.kiq_version & AMDGPU_MES_VERSION_MASK); static struct attribute *fw_attrs[] = { &dev_attr_vce_fw_version.attr, &dev_attr_uvd_fw_version.attr, @@ -716,6 +718,7 @@ static struct attribute *fw_attrs[] = { &dev_attr_smc_fw_version.attr, &dev_attr_sdma_fw_version.attr, &dev_attr_sdma2_fw_version.attr, &dev_attr_vcn_fw_version.attr, &dev_attr_dmcu_fw_version.attr, &dev_attr_imu_fw_version.attr, + &dev_attr_mes_fw_version.attr, &dev_attr_mes_kiq_fw_version.attr, NULL }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index ae455aab5d29..36b55d2bd51a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -1239,3 +1239,18 @@ int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev) return 0; } + +int amdgpu_vcn_psp_update_sram(struct amdgpu_device *adev, int inst_idx, + enum AMDGPU_UCODE_ID ucode_id) +{ + struct amdgpu_firmware_info ucode = { + .ucode_id = (ucode_id ? ucode_id : + (inst_idx ? AMDGPU_UCODE_ID_VCN1_RAM : + AMDGPU_UCODE_ID_VCN0_RAM)), + .mc_addr = adev->vcn.inst[inst_idx].dpg_sram_gpu_addr, + .ucode_size = ((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr - + (uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr), + }; + + return psp_execute_ip_fw_load(&adev->psp, &ucode); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index 92d5534df5f4..a3eed90b6af0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -161,6 +161,7 @@ } while (0) #define AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE (1 << 2) +#define AMDGPU_FW_SHARED_FLAG_0_DRM_KEY_INJECT (1 << 4) #define AMDGPU_VCN_FW_SHARED_FLAG_0_RB (1 << 6) #define AMDGPU_VCN_MULTI_QUEUE_FLAG (1 << 8) #define AMDGPU_VCN_SW_RING_FLAG (1 << 9) @@ -180,6 +181,8 @@ #define AMDGPU_VCN_SMU_DPM_INTERFACE_DGPU (0) #define AMDGPU_VCN_SMU_DPM_INTERFACE_APU (1) +#define AMDGPU_DRM_KEY_INJECT_WORKAROUND_VCNFW_ASD_HANDSHAKING 2 + enum fw_queue_mode { FW_QUEUE_RING_RESET = 1, FW_QUEUE_DPG_HOLD_OFF = 2, @@ -343,6 +346,11 @@ struct amdgpu_fw_shared_rb_setup { uint32_t reserved[6]; }; +struct amdgpu_fw_shared_drm_key_wa { + uint8_t method; + uint8_t reserved[3]; +}; + struct amdgpu_vcn4_fw_shared { uint32_t present_flag_0; uint8_t pad[12]; @@ -352,6 +360,7 @@ struct amdgpu_vcn4_fw_shared { uint8_t pad2[20]; struct amdgpu_fw_shared_rb_setup rb_setup; struct amdgpu_fw_shared_smu_interface_info smu_dpm_interface; + struct amdgpu_fw_shared_drm_key_wa drm_key_wa; }; struct amdgpu_vcn_fwlog { @@ -414,4 +423,7 @@ int amdgpu_vcn_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev); +int amdgpu_vcn_psp_update_sram(struct amdgpu_device *adev, int inst_idx, + enum AMDGPU_UCODE_ID ucode_id); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 41aa853a07d2..96857ae7fb5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -520,7 +520,7 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev) tmp = ((struct amd_sriov_msg_pf2vf_info *)pf2vf_info)->mm_bw_management[i].encode_max_frame_pixels; adev->virt.encode_max_frame_pixels = max(tmp, adev->virt.encode_max_frame_pixels); } - if((adev->virt.decode_max_dimension_pixels > 0) || (adev->virt.encode_max_dimension_pixels > 0)) + if ((adev->virt.decode_max_dimension_pixels > 0) || (adev->virt.encode_max_dimension_pixels > 0)) adev->virt.is_mm_bw_enabled = true; adev->unique_id = @@ -835,6 +835,16 @@ enum amdgpu_sriov_vf_mode amdgpu_virt_get_sriov_vf_mode(struct amdgpu_device *ad return mode; } +void amdgpu_virt_post_reset(struct amdgpu_device *adev) +{ + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(11, 0, 3)) { + /* force set to GFXOFF state after reset, + * to avoid some invalid operation before GC enable + */ + adev->gfx.is_poweron = false; + } +} + bool amdgpu_virt_fw_load_skip_check(struct amdgpu_device *adev, uint32_t ucode_id) { switch (adev->ip_versions[MP0_HWIP][0]) { @@ -845,6 +855,17 @@ bool amdgpu_virt_fw_load_skip_check(struct amdgpu_device *adev, uint32_t ucode_i return false; else return true; + case IP_VERSION(11, 0, 9): + case IP_VERSION(11, 0, 7): + /* black list for CHIP_NAVI12 and CHIP_SIENNA_CICHLID */ + if (ucode_id == AMDGPU_UCODE_ID_RLC_G + || ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_CNTL + || ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_GPM_MEM + || ucode_id == AMDGPU_UCODE_ID_RLC_RESTORE_LIST_SRM_MEM + || ucode_id == AMDGPU_UCODE_ID_SMC) + return true; + else + return false; case IP_VERSION(13, 0, 10): /* white list */ if (ucode_id == AMDGPU_UCODE_ID_CAP @@ -954,7 +975,7 @@ static bool amdgpu_virt_get_rlcg_reg_access_flag(struct amdgpu_device *adev, return ret; } -static u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag) +static u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v, u32 flag, u32 xcc_id) { struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; uint32_t timeout = 50000; @@ -972,7 +993,12 @@ static u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v return 0; } - reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + if (adev->gfx.xcc_mask && (((1 << xcc_id) & adev->gfx.xcc_mask) == 0)) { + dev_err(adev->dev, "invalid xcc\n"); + return 0; + } + + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[xcc_id]; scratch_reg0 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg0; scratch_reg1 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg1; scratch_reg2 = (void __iomem *)adev->rmmio + 4 * reg_access_ctrl->scratch_reg2; @@ -1037,13 +1063,13 @@ static u32 amdgpu_virt_rlcg_reg_rw(struct amdgpu_device *adev, u32 offset, u32 v void amdgpu_sriov_wreg(struct amdgpu_device *adev, u32 offset, u32 value, - u32 acc_flags, u32 hwip) + u32 acc_flags, u32 hwip, u32 xcc_id) { u32 rlcg_flag; if (!amdgpu_sriov_runtime(adev) && amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, true, &rlcg_flag)) { - amdgpu_virt_rlcg_reg_rw(adev, offset, value, rlcg_flag); + amdgpu_virt_rlcg_reg_rw(adev, offset, value, rlcg_flag, xcc_id); return; } @@ -1054,13 +1080,13 @@ void amdgpu_sriov_wreg(struct amdgpu_device *adev, } u32 amdgpu_sriov_rreg(struct amdgpu_device *adev, - u32 offset, u32 acc_flags, u32 hwip) + u32 offset, u32 acc_flags, u32 hwip, u32 xcc_id) { u32 rlcg_flag; if (!amdgpu_sriov_runtime(adev) && amdgpu_virt_get_rlcg_reg_access_flag(adev, acc_flags, hwip, false, &rlcg_flag)) - return amdgpu_virt_rlcg_reg_rw(adev, offset, 0, rlcg_flag); + return amdgpu_virt_rlcg_reg_rw(adev, offset, 0, rlcg_flag, xcc_id); if (acc_flags & AMDGPU_REGS_NO_KIQ) return RREG32_NO_KIQ(offset); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 4f7bab52282a..fabb83e9d9ae 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -355,9 +355,10 @@ void amdgpu_virt_update_sriov_video_codec(struct amdgpu_device *adev, struct amdgpu_video_codec_info *decode, uint32_t decode_array_size); void amdgpu_sriov_wreg(struct amdgpu_device *adev, u32 offset, u32 value, - u32 acc_flags, u32 hwip); + u32 acc_flags, u32 hwip, u32 xcc_id); u32 amdgpu_sriov_rreg(struct amdgpu_device *adev, - u32 offset, u32 acc_flags, u32 hwip); + u32 offset, u32 acc_flags, u32 hwip, u32 xcc_id); bool amdgpu_virt_fw_load_skip_check(struct amdgpu_device *adev, uint32_t ucode_id); +void amdgpu_virt_post_reset(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c index 53ff91fc6cf6..7148a216ae2f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c @@ -55,8 +55,9 @@ static enum hrtimer_restart amdgpu_vkms_vblank_simulate(struct hrtimer *timer) DRM_WARN("%s: vblank timer overrun\n", __func__); ret = drm_crtc_handle_vblank(crtc); + /* Don't queue timer again when vblank is disabled. */ if (!ret) - DRM_ERROR("amdgpu_vkms failure on handling vblank"); + return HRTIMER_NORESTART; return HRTIMER_RESTART; } @@ -81,7 +82,7 @@ static void amdgpu_vkms_disable_vblank(struct drm_crtc *crtc) { struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); - hrtimer_cancel(&amdgpu_crtc->vblank_timer); + hrtimer_try_to_cancel(&amdgpu_crtc->vblank_timer); } static bool amdgpu_vkms_get_vblank_timestamp(struct drm_crtc *crtc, @@ -500,8 +501,6 @@ static int amdgpu_vkms_sw_init(void *handle) adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true; - adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true; - r = amdgpu_display_modeset_create_props(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index d0e6009a985b..f5daadcec865 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -112,9 +112,9 @@ struct amdgpu_prt_cb { }; /** - * struct amdgpu_vm_tlb_seq_cb - Helper to increment the TLB flush sequence + * struct amdgpu_vm_tlb_seq_struct - Helper to increment the TLB flush sequence */ -struct amdgpu_vm_tlb_seq_cb { +struct amdgpu_vm_tlb_seq_struct { /** * @vm: pointer to the amdgpu_vm structure to set the fence sequence on */ @@ -829,7 +829,7 @@ error: static void amdgpu_vm_tlb_seq_cb(struct dma_fence *fence, struct dma_fence_cb *cb) { - struct amdgpu_vm_tlb_seq_cb *tlb_cb; + struct amdgpu_vm_tlb_seq_struct *tlb_cb; tlb_cb = container_of(cb, typeof(*tlb_cb), cb); atomic64_inc(&tlb_cb->vm->tlb_seq); @@ -867,7 +867,7 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct dma_fence **fence) { struct amdgpu_vm_update_params params; - struct amdgpu_vm_tlb_seq_cb *tlb_cb; + struct amdgpu_vm_tlb_seq_struct *tlb_cb; struct amdgpu_res_cursor cursor; enum amdgpu_sync_mode sync_mode; int r, idx; @@ -2117,13 +2117,14 @@ long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout) * * @adev: amdgpu_device pointer * @vm: requested vm + * @xcp_id: GPU partition selection id * * Init @vm fields. * * Returns: * 0 for success, error for failure. */ -int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) +int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, int32_t xcp_id) { struct amdgpu_bo *root_bo; struct amdgpu_bo_vm *root; @@ -2173,7 +2174,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm) vm->evicting = false; r = amdgpu_vm_pt_create(adev, vm, adev->vm_manager.root_level, - false, &root); + false, &root, xcp_id); if (r) goto error_free_delayed; root_bo = &root->bo; @@ -2275,16 +2276,13 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm) goto unreserve_bo; vm->update_funcs = &amdgpu_vm_cpu_funcs; + r = amdgpu_vm_pt_map_tables(adev, vm); + if (r) + goto unreserve_bo; + } else { vm->update_funcs = &amdgpu_vm_sdma_funcs; } - /* - * Make sure root PD gets mapped. As vm_update_mode could be changed - * when turning a GFX VM into a compute VM. - */ - r = vm->update_funcs->map_table(to_amdgpu_bo_vm(vm->root.bo)); - if (r) - goto unreserve_bo; dma_fence_put(vm->last_update); vm->last_update = dma_fence_get_stub(); @@ -2600,7 +2598,7 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid, /* Intentionally setting invalid PTE flag * combination to force a no-retry-fault */ - flags = AMDGPU_PTE_SNOOPED | AMDGPU_PTE_PRT; + flags = AMDGPU_VM_NORETRY_FLAGS; value = 0; } else if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_NEVER) { /* Redirect the access to the dummy page */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 45bcf46c6f59..204ab13184ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -86,7 +86,13 @@ struct amdgpu_mem_stats; /* PDE Block Fragment Size for VEGA10 */ #define AMDGPU_PDE_BFS(a) ((uint64_t)a << 59) +/* Flag combination to set no-retry with TF disabled */ +#define AMDGPU_VM_NORETRY_FLAGS (AMDGPU_PTE_EXECUTABLE | AMDGPU_PDE_PTE | \ + AMDGPU_PTE_TF) +/* Flag combination to set no-retry with TF enabled */ +#define AMDGPU_VM_NORETRY_FLAGS_TF (AMDGPU_PTE_VALID | AMDGPU_PTE_SYSTEM | \ + AMDGPU_PTE_PRT) /* For GFX9 */ #define AMDGPU_PTE_MTYPE_VG10(a) ((uint64_t)(a) << 57) #define AMDGPU_PTE_MTYPE_VG10_MASK AMDGPU_PTE_MTYPE_VG10(3ULL) @@ -394,7 +400,7 @@ int amdgpu_vm_set_pasid(struct amdgpu_device *adev, struct amdgpu_vm *vm, u32 pasid); long amdgpu_vm_wait_idle(struct amdgpu_vm *vm, long timeout); -int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm); +int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, int32_t xcp_id); int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm); void amdgpu_vm_release_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm); void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm); @@ -476,7 +482,8 @@ void amdgpu_vm_get_memory(struct amdgpu_vm *vm, int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo_vm *vmbo, bool immediate); int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, - int level, bool immediate, struct amdgpu_bo_vm **vmbo); + int level, bool immediate, struct amdgpu_bo_vm **vmbo, + int32_t xcp_id); void amdgpu_vm_pt_free_root(struct amdgpu_device *adev, struct amdgpu_vm *vm); bool amdgpu_vm_pt_is_root_clean(struct amdgpu_device *adev, struct amdgpu_vm *vm); @@ -492,6 +499,8 @@ void amdgpu_vm_pt_free_work(struct work_struct *work); void amdgpu_debugfs_vm_bo_info(struct amdgpu_vm *vm, struct seq_file *m); #endif +int amdgpu_vm_pt_map_tables(struct amdgpu_device *adev, struct amdgpu_vm *vm); + /** * amdgpu_vm_tlb_seq - return tlb flush sequence number * @vm: the amdgpu_vm structure to query diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c index 31913ae86de6..6e31621452de 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c @@ -31,6 +31,7 @@ */ static int amdgpu_vm_cpu_map_table(struct amdgpu_bo_vm *table) { + table->bo.flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED; return amdgpu_bo_kmap(&table->bo, NULL); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c index dea1a64be44d..96d601e209b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c @@ -498,11 +498,12 @@ exit: * @level: the page table level * @immediate: use a immediate update * @vmbo: pointer to the buffer object pointer + * @xcp_id: GPU partition id */ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, - int level, bool immediate, struct amdgpu_bo_vm **vmbo) + int level, bool immediate, struct amdgpu_bo_vm **vmbo, + int32_t xcp_id) { - struct amdgpu_fpriv *fpriv = container_of(vm, struct amdgpu_fpriv, vm); struct amdgpu_bo_param bp; struct amdgpu_bo *bo; struct dma_resv *resv; @@ -535,7 +536,7 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, bp.type = ttm_bo_type_kernel; bp.no_wait_gpu = immediate; - bp.xcp_id_plus1 = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id + 1; + bp.xcp_id_plus1 = xcp_id + 1; if (vm->root.bo) bp.resv = vm->root.bo->tbo.base.resv; @@ -561,7 +562,7 @@ int amdgpu_vm_pt_create(struct amdgpu_device *adev, struct amdgpu_vm *vm, bp.type = ttm_bo_type_kernel; bp.resv = bo->tbo.base.resv; bp.bo_ptr_size = sizeof(struct amdgpu_bo); - bp.xcp_id_plus1 = fpriv->xcp_id == ~0 ? 0 : fpriv->xcp_id + 1; + bp.xcp_id_plus1 = xcp_id + 1; r = amdgpu_bo_create(adev, &bp, &(*vmbo)->shadow); @@ -606,7 +607,8 @@ static int amdgpu_vm_pt_alloc(struct amdgpu_device *adev, return 0; amdgpu_vm_eviction_unlock(vm); - r = amdgpu_vm_pt_create(adev, vm, cursor->level, immediate, &pt); + r = amdgpu_vm_pt_create(adev, vm, cursor->level, immediate, &pt, + vm->root.bo->xcp_id); amdgpu_vm_eviction_lock(vm); if (r) return r; @@ -778,6 +780,27 @@ int amdgpu_vm_pde_update(struct amdgpu_vm_update_params *params, 1, 0, flags); } +/** + * amdgpu_vm_pte_update_noretry_flags - Update PTE no-retry flags + * + * @adev: amdgpu_device pointer + * @flags: pointer to PTE flags + * + * Update PTE no-retry flags when TF is enabled. + */ +static void amdgpu_vm_pte_update_noretry_flags(struct amdgpu_device *adev, + uint64_t *flags) +{ + /* + * Update no-retry flags with the corresponding TF + * no-retry combination. + */ + if ((*flags & AMDGPU_VM_NORETRY_FLAGS) == AMDGPU_VM_NORETRY_FLAGS) { + *flags &= ~AMDGPU_VM_NORETRY_FLAGS; + *flags |= adev->gmc.noretry_flags; + } +} + /* * amdgpu_vm_pte_update_flags - figure out flags for PTE updates * @@ -804,6 +827,16 @@ static void amdgpu_vm_pte_update_flags(struct amdgpu_vm_update_params *params, flags |= AMDGPU_PTE_EXECUTABLE; } + /* + * Update no-retry flags to use the no-retry flag combination + * with TF enabled. The AMDGPU_VM_NORETRY_FLAGS flag combination + * does not work when TF is enabled. So, replace them with + * AMDGPU_VM_NORETRY_FLAGS_TF flag combination which works for + * all cases. + */ + if (level == AMDGPU_VM_PTB) + amdgpu_vm_pte_update_noretry_flags(adev, &flags); + /* APUs mapping system memory may need different MTYPEs on different * NUMA nodes. Only do this for contiguous ranges that can be assumed * to be on the same NUMA node. @@ -1044,3 +1077,31 @@ int amdgpu_vm_ptes_update(struct amdgpu_vm_update_params *params, return 0; } + +/** + * amdgpu_vm_pt_map_tables - have bo of root PD cpu accessible + * @adev: amdgpu device structure + * @vm: amdgpu vm structure + * + * make root page directory and everything below it cpu accessible. + */ +int amdgpu_vm_pt_map_tables(struct amdgpu_device *adev, struct amdgpu_vm *vm) +{ + struct amdgpu_vm_pt_cursor cursor; + struct amdgpu_vm_bo_base *entry; + + for_each_amdgpu_vm_pt_dfs_safe(adev, vm, NULL, cursor, entry) { + + struct amdgpu_bo_vm *bo; + int r; + + if (entry->bo) { + bo = to_amdgpu_bo_vm(entry->bo); + r = vm->update_funcs->map_table(bo); + if (r) + return r; + } + } + + return 0; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c index d175e862f222..565a1fa436d4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c @@ -239,8 +239,13 @@ static int amdgpu_xcp_dev_alloc(struct amdgpu_device *adev) for (i = 1; i < MAX_XCP; i++) { ret = amdgpu_xcp_drm_dev_alloc(&p_ddev); - if (ret) + if (ret == -ENOSPC) { + dev_warn(adev->dev, + "Skip xcp node #%d when out of drm node resource.", i); + return 0; + } else if (ret) { return ret; + } /* Redirect all IOCTLs to the primary device */ adev->xcp_mgr->xcp[i].rdev = p_ddev->render->dev; @@ -328,6 +333,9 @@ int amdgpu_xcp_dev_register(struct amdgpu_device *adev, return 0; for (i = 1; i < MAX_XCP; i++) { + if (!adev->xcp_mgr->xcp[i].ddev) + break; + ret = drm_dev_register(adev->xcp_mgr->xcp[i].ddev, ent->driver_data); if (ret) return ret; @@ -345,6 +353,9 @@ void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev) return; for (i = 1; i < MAX_XCP; i++) { + if (!adev->xcp_mgr->xcp[i].ddev) + break; + p_ddev = adev->xcp_mgr->xcp[i].ddev; drm_dev_unplug(p_ddev); p_ddev->render->dev = adev->xcp_mgr->xcp[i].rdev; @@ -363,7 +374,7 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev, if (!adev->xcp_mgr) return 0; - fpriv->xcp_id = ~0; + fpriv->xcp_id = AMDGPU_XCP_NO_PARTITION; for (i = 0; i < MAX_XCP; ++i) { if (!adev->xcp_mgr->xcp[i].ddev) break; @@ -381,7 +392,7 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev, } } - fpriv->vm.mem_id = fpriv->xcp_id == ~0 ? -1 : + fpriv->vm.mem_id = fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION ? -1 : adev->xcp_mgr->xcp[fpriv->xcp_id].mem_id; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h index 0f8026d64ea5..9a1036aeec2a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h @@ -37,6 +37,8 @@ #define AMDGPU_XCP_FL_NONE 0 #define AMDGPU_XCP_FL_LOCKED (1 << 0) +#define AMDGPU_XCP_NO_PARTITION (~0) + struct amdgpu_fpriv; enum AMDGPU_XCP_IP_BLOCK { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 03dc59cbe8aa..7e91b24784e5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -500,6 +500,7 @@ struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) hive = kzalloc(sizeof(*hive), GFP_KERNEL); if (!hive) { dev_err(adev->dev, "XGMI: allocation failed\n"); + ret = -ENOMEM; hive = NULL; goto pro_end; } diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c index 16471b81a1f5..d0fc62784e82 100644 --- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c +++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c @@ -68,7 +68,7 @@ static void aqua_vanjaram_set_xcp_id(struct amdgpu_device *adev, enum AMDGPU_XCP_IP_BLOCK ip_blk; uint32_t inst_mask; - ring->xcp_id = ~0; + ring->xcp_id = AMDGPU_XCP_NO_PARTITION; if (adev->xcp_mgr->mode == AMDGPU_XCP_MODE_NONE) return; @@ -134,7 +134,7 @@ static int aqua_vanjaram_xcp_sched_list_update( for (i = 0; i < AMDGPU_MAX_RINGS; i++) { ring = adev->rings[i]; - if (!ring || !ring->sched.ready) + if (!ring || !ring->sched.ready || ring->no_scheduler) continue; aqua_vanjaram_xcp_gpu_sched_update(adev, ring, ring->xcp_id); @@ -177,7 +177,7 @@ static int aqua_vanjaram_select_scheds( u32 sel_xcp_id; int i; - if (fpriv->xcp_id == ~0) { + if (fpriv->xcp_id == AMDGPU_XCP_NO_PARTITION) { u32 least_ref_cnt = ~0; fpriv->xcp_id = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index 5f610e9a5f0f..9f63ddb89b75 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -1438,6 +1438,8 @@ static void atom_get_vbios_pn(struct atom_context *ctx) ctx->vbios_pn[count] = 0; } + + pr_info("ATOM BIOS: %s\n", ctx->vbios_pn); } static void atom_get_vbios_version(struct atom_context *ctx) @@ -1460,11 +1462,9 @@ struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios) int base; struct atom_context *ctx = kzalloc(sizeof(struct atom_context), GFP_KERNEL); - char *str; struct _ATOM_ROM_HEADER *atom_rom_header; struct _ATOM_MASTER_DATA_TABLE *master_table; struct _ATOM_FIRMWARE_INFO *atom_fw_info; - u16 idx; if (!ctx) return NULL; @@ -1502,16 +1502,6 @@ struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios) return NULL; } - idx = CU16(ATOM_ROM_PART_NUMBER_PTR); - if (idx == 0) - idx = 0x80; - - str = CSTR(idx); - if (*str != '\0') { - pr_info("ATOM BIOS: %s\n", str); - strscpy(ctx->vbios_version, str, sizeof(ctx->vbios_version)); - } - atom_rom_header = (struct _ATOM_ROM_HEADER *)CSTR(base); if (atom_rom_header->usMasterDataTableOffset != 0) { master_table = (struct _ATOM_MASTER_DATA_TABLE *) diff --git a/drivers/gpu/drm/amd/amdgpu/atom.h b/drivers/gpu/drm/amd/amdgpu/atom.h index 0c1839824520..c11cf18a0f18 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.h +++ b/drivers/gpu/drm/amd/amdgpu/atom.h @@ -33,7 +33,6 @@ struct drm_device; #define ATOM_ATI_MAGIC_PTR 0x30 #define ATOM_ATI_MAGIC " 761295520" #define ATOM_ROM_TABLE_PTR 0x48 -#define ATOM_ROM_PART_NUMBER_PTR 0x6E #define ATOM_ROM_MAGIC "ATOM" #define ATOM_ROM_MAGIC_PTR 4 @@ -118,12 +117,15 @@ struct drm_device; struct card_info { struct drm_device *dev; - void (* reg_write)(struct card_info *, uint32_t, uint32_t); /* filled by driver */ - uint32_t (* reg_read)(struct card_info *, uint32_t); /* filled by driver */ - void (* mc_write)(struct card_info *, uint32_t, uint32_t); /* filled by driver */ - uint32_t (* mc_read)(struct card_info *, uint32_t); /* filled by driver */ - void (* pll_write)(struct card_info *, uint32_t, uint32_t); /* filled by driver */ - uint32_t (* pll_read)(struct card_info *, uint32_t); /* filled by driver */ + void (*reg_write)(struct card_info *info, + u32 reg, uint32_t val); /* filled by driver */ + uint32_t (*reg_read)(struct card_info *info, uint32_t reg); /* filled by driver */ + void (*mc_write)(struct card_info *info, + u32 reg, uint32_t val); /* filled by driver */ + uint32_t (*mc_read)(struct card_info *info, uint32_t reg); /* filled by driver */ + void (*pll_write)(struct card_info *info, + u32 reg, uint32_t val); /* filled by driver */ + uint32_t (*pll_read)(struct card_info *info, uint32_t reg); /* filled by driver */ }; struct atom_context { @@ -143,7 +145,6 @@ struct atom_context { int io_mode; uint32_t *scratch; int scratch_size_bytes; - char vbios_version[20]; uint8_t name[STRLEN_LONG]; uint8_t vbios_pn[STRLEN_LONG]; @@ -154,10 +155,10 @@ struct atom_context { extern int amdgpu_atom_debug; -struct atom_context *amdgpu_atom_parse(struct card_info *, void *); -int amdgpu_atom_execute_table(struct atom_context *, int, uint32_t *); -int amdgpu_atom_asic_init(struct atom_context *); -void amdgpu_atom_destroy(struct atom_context *); +struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios); +int amdgpu_atom_execute_table(struct atom_context *ctx, int index, uint32_t *params); +int amdgpu_atom_asic_init(struct atom_context *ctx); +void amdgpu_atom_destroy(struct atom_context *ctx); bool amdgpu_atom_parse_data_header(struct atom_context *ctx, int index, uint16_t *size, uint8_t *frev, uint8_t *crev, uint16_t *data_start); bool amdgpu_atom_parse_cmd_header(struct atom_context *ctx, int index, diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index 5641cf05d856..e63abdf52b6c 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1574,17 +1574,8 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &bridge_cfg); - pcie_capability_read_word(adev->pdev, PCI_EXP_LNKCTL, - &gpu_cfg); - - tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); - - tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(adev->pdev, PCI_EXP_LNKCTL, - tmp16); + pcie_capability_set_word(root, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); + pcie_capability_set_word(adev->pdev, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); tmp = RREG32_PCIE(ixPCIE_LC_STATUS1); max_lw = (tmp & PCIE_LC_STATUS1__LC_DETECTED_LINK_WIDTH_MASK) >> @@ -1637,21 +1628,14 @@ static void cik_pcie_gen3_enable(struct amdgpu_device *adev) msleep(100); /* linkctl */ - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(root, PCI_EXP_LNKCTL, - tmp16); - - pcie_capability_read_word(adev->pdev, - PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(adev->pdev, - PCI_EXP_LNKCTL, - tmp16); + pcie_capability_clear_and_set_word(root, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + bridge_cfg & + PCI_EXP_LNKCTL_HAWD); + pcie_capability_clear_and_set_word(adev->pdev, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + gpu_cfg & + PCI_EXP_LNKCTL_HAWD); /* linkctl2 */ pcie_capability_read_word(root, PCI_EXP_LNKCTL2, diff --git a/drivers/gpu/drm/amd/amdgpu/cik_ih.c b/drivers/gpu/drm/amd/amdgpu/cik_ih.c index df385ffc9768..6f7c031dd197 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_ih.c @@ -442,8 +442,7 @@ static void cik_ih_set_interrupt_funcs(struct amdgpu_device *adev) adev->irq.ih_funcs = &cik_ih_funcs; } -const struct amdgpu_ip_block_version cik_ih_ip_block = -{ +const struct amdgpu_ip_block_version cik_ih_ip_block = { .type = AMD_IP_BLOCK_TYPE_IH, .major = 2, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 9a24ed463abd..584cd5277f92 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -52,8 +52,7 @@ static void dce_v10_0_set_display_funcs(struct amdgpu_device *adev); static void dce_v10_0_set_irq_funcs(struct amdgpu_device *adev); -static const u32 crtc_offsets[] = -{ +static const u32 crtc_offsets[] = { CRTC0_REGISTER_OFFSET, CRTC1_REGISTER_OFFSET, CRTC2_REGISTER_OFFSET, @@ -63,8 +62,7 @@ static const u32 crtc_offsets[] = CRTC6_REGISTER_OFFSET }; -static const u32 hpd_offsets[] = -{ +static const u32 hpd_offsets[] = { HPD0_REGISTER_OFFSET, HPD1_REGISTER_OFFSET, HPD2_REGISTER_OFFSET, @@ -121,30 +119,26 @@ static const struct { .hpd = DISP_INTERRUPT_STATUS_CONTINUE5__DC_HPD6_INTERRUPT_MASK } }; -static const u32 golden_settings_tonga_a11[] = -{ +static const u32 golden_settings_tonga_a11[] = { mmDCI_CLK_CNTL, 0x00000080, 0x00000000, mmFBC_DEBUG_COMP, 0x000000f0, 0x00000070, mmFBC_MISC, 0x1f311fff, 0x12300000, mmHDMI_CONTROL, 0x31000111, 0x00000011, }; -static const u32 tonga_mgcg_cgcg_init[] = -{ +static const u32 tonga_mgcg_cgcg_init[] = { mmXDMA_CLOCK_GATING_CNTL, 0xffffffff, 0x00000100, mmXDMA_MEM_POWER_CNTL, 0x00000101, 0x00000000, }; -static const u32 golden_settings_fiji_a10[] = -{ +static const u32 golden_settings_fiji_a10[] = { mmDCI_CLK_CNTL, 0x00000080, 0x00000000, mmFBC_DEBUG_COMP, 0x000000f0, 0x00000070, mmFBC_MISC, 0x1f311fff, 0x12300000, mmHDMI_CONTROL, 0x31000111, 0x00000011, }; -static const u32 fiji_mgcg_cgcg_init[] = -{ +static const u32 fiji_mgcg_cgcg_init[] = { mmXDMA_CLOCK_GATING_CNTL, 0xffffffff, 0x00000100, mmXDMA_MEM_POWER_CNTL, 0x00000101, 0x00000000, }; @@ -1425,8 +1419,7 @@ static void dce_v10_0_audio_enable(struct amdgpu_device *adev, enable ? AZALIA_F0_CODEC_PIN_CONTROL_HOT_PLUG_CONTROL__AUDIO_ENABLED_MASK : 0); } -static const u32 pin_offsets[] = -{ +static const u32 pin_offsets[] = { AUD0_REGISTER_OFFSET, AUD1_REGISTER_OFFSET, AUD2_REGISTER_OFFSET, @@ -1811,8 +1804,7 @@ static void dce_v10_0_afmt_fini(struct amdgpu_device *adev) } } -static const u32 vga_control_regs[6] = -{ +static const u32 vga_control_regs[6] = { mmD1VGA_CONTROL, mmD2VGA_CONTROL, mmD3VGA_CONTROL, @@ -3651,8 +3643,7 @@ static void dce_v10_0_set_irq_funcs(struct amdgpu_device *adev) adev->hpd_irq.funcs = &dce_v10_0_hpd_irq_funcs; } -const struct amdgpu_ip_block_version dce_v10_0_ip_block = -{ +const struct amdgpu_ip_block_version dce_v10_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 10, .minor = 0, @@ -3660,8 +3651,7 @@ const struct amdgpu_ip_block_version dce_v10_0_ip_block = .funcs = &dce_v10_0_ip_funcs, }; -const struct amdgpu_ip_block_version dce_v10_1_ip_block = -{ +const struct amdgpu_ip_block_version dce_v10_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 10, .minor = 1, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index d421a268c9ff..f2b3cb5ed6be 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -53,8 +53,7 @@ static void dce_v8_0_set_display_funcs(struct amdgpu_device *adev); static void dce_v8_0_set_irq_funcs(struct amdgpu_device *adev); -static const u32 crtc_offsets[6] = -{ +static const u32 crtc_offsets[6] = { CRTC0_REGISTER_OFFSET, CRTC1_REGISTER_OFFSET, CRTC2_REGISTER_OFFSET, @@ -63,8 +62,7 @@ static const u32 crtc_offsets[6] = CRTC5_REGISTER_OFFSET }; -static const u32 hpd_offsets[] = -{ +static const u32 hpd_offsets[] = { HPD0_REGISTER_OFFSET, HPD1_REGISTER_OFFSET, HPD2_REGISTER_OFFSET, @@ -1345,9 +1343,9 @@ static void dce_v8_0_audio_write_sad_regs(struct drm_encoder *encoder) if (sad->channels > max_channels) { value = (sad->channels << AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__MAX_CHANNELS__SHIFT) | - (sad->byte2 << + (sad->byte2 << AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__DESCRIPTOR_BYTE_2__SHIFT) | - (sad->freq << + (sad->freq << AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0__SUPPORTED_FREQUENCIES__SHIFT); max_channels = sad->channels; } @@ -1379,8 +1377,7 @@ static void dce_v8_0_audio_enable(struct amdgpu_device *adev, enable ? AZALIA_F0_CODEC_PIN_CONTROL_HOT_PLUG_CONTROL__AUDIO_ENABLED_MASK : 0); } -static const u32 pin_offsets[7] = -{ +static const u32 pin_offsets[7] = { (0x1780 - 0x1780), (0x1786 - 0x1780), (0x178c - 0x1780), @@ -1740,8 +1737,7 @@ static void dce_v8_0_afmt_fini(struct amdgpu_device *adev) } } -static const u32 vga_control_regs[6] = -{ +static const u32 vga_control_regs[6] = { mmD1VGA_CONTROL, mmD2VGA_CONTROL, mmD3VGA_CONTROL, @@ -1895,9 +1891,9 @@ static int dce_v8_0_crtc_do_set_base(struct drm_crtc *crtc, case DRM_FORMAT_XBGR8888: case DRM_FORMAT_ABGR8888: fb_format = ((GRPH_DEPTH_32BPP << GRPH_CONTROL__GRPH_DEPTH__SHIFT) | - (GRPH_FORMAT_ARGB8888 << GRPH_CONTROL__GRPH_FORMAT__SHIFT)); + (GRPH_FORMAT_ARGB8888 << GRPH_CONTROL__GRPH_FORMAT__SHIFT)); fb_swap = ((GRPH_RED_SEL_B << GRPH_SWAP_CNTL__GRPH_RED_CROSSBAR__SHIFT) | - (GRPH_BLUE_SEL_R << GRPH_SWAP_CNTL__GRPH_BLUE_CROSSBAR__SHIFT)); + (GRPH_BLUE_SEL_R << GRPH_SWAP_CNTL__GRPH_BLUE_CROSSBAR__SHIFT)); #ifdef __BIG_ENDIAN fb_swap |= (GRPH_ENDIAN_8IN32 << GRPH_SWAP_CNTL__GRPH_ENDIAN_SWAP__SHIFT); #endif @@ -3151,7 +3147,7 @@ static int dce_v8_0_pageflip_irq(struct amdgpu_device *adev, spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags); works = amdgpu_crtc->pflip_works; - if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){ + if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED) { DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != " "AMDGPU_FLIP_SUBMITTED(%d)\n", amdgpu_crtc->pflip_status, @@ -3544,8 +3540,7 @@ static void dce_v8_0_set_irq_funcs(struct amdgpu_device *adev) adev->hpd_irq.funcs = &dce_v8_0_hpd_irq_funcs; } -const struct amdgpu_ip_block_version dce_v8_0_ip_block = -{ +const struct amdgpu_ip_block_version dce_v8_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 8, .minor = 0, @@ -3553,8 +3548,7 @@ const struct amdgpu_ip_block_version dce_v8_0_ip_block = .funcs = &dce_v8_0_ip_funcs, }; -const struct amdgpu_ip_block_version dce_v8_1_ip_block = -{ +const struct amdgpu_ip_block_version dce_v8_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 8, .minor = 1, @@ -3562,8 +3556,7 @@ const struct amdgpu_ip_block_version dce_v8_1_ip_block = .funcs = &dce_v8_0_ip_funcs, }; -const struct amdgpu_ip_block_version dce_v8_2_ip_block = -{ +const struct amdgpu_ip_block_version dce_v8_2_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 8, .minor = 2, @@ -3571,8 +3564,7 @@ const struct amdgpu_ip_block_version dce_v8_2_ip_block = .funcs = &dce_v8_0_ip_funcs, }; -const struct amdgpu_ip_block_version dce_v8_3_ip_block = -{ +const struct amdgpu_ip_block_version dce_v8_3_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 8, .minor = 3, @@ -3580,8 +3572,7 @@ const struct amdgpu_ip_block_version dce_v8_3_ip_block = .funcs = &dce_v8_0_ip_funcs, }; -const struct amdgpu_ip_block_version dce_v8_5_ip_block = -{ +const struct amdgpu_ip_block_version dce_v8_5_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 8, .minor = 5, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 44af8022b89f..9032d7a24d7c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -271,8 +271,7 @@ MODULE_FIRMWARE("amdgpu/gc_10_3_7_mec.bin"); MODULE_FIRMWARE("amdgpu/gc_10_3_7_mec2.bin"); MODULE_FIRMWARE("amdgpu/gc_10_3_7_rlc.bin"); -static const struct soc15_reg_golden golden_settings_gc_10_1[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_1[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x00400014), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_CPF_CLK_CTRL, 0xfcff8fff, 0xf8000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CLK_CTRL, 0xcd000000, 0x0d000100), @@ -315,13 +314,11 @@ static const struct soc15_reg_golden golden_settings_gc_10_1[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00c00000, 0x00c00000) }; -static const struct soc15_reg_golden golden_settings_gc_10_0_nv10[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_0_nv10[] = { /* Pending on emulation bring up */ }; -static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_0_nv10[] = -{ +static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_0_nv10[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xe0000000, 0x0), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), @@ -1376,8 +1373,7 @@ static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_0_nv10[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xFFFFFFFF, 0xe0000000) }; -static const struct soc15_reg_golden golden_settings_gc_10_1_1[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_1_1[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0xffffffff, 0x003c0014), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_GS_NGG_CLK_CTRL, 0xffff8fff, 0xffff8100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_IA_CLK_CTRL, 0xffff0fff, 0xffff0100), @@ -1418,8 +1414,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_1[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00c00000, 0x00c00000), }; -static const struct soc15_reg_golden golden_settings_gc_10_1_2[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_1_2[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCB_HW_CONTROL_4, 0x003e001f, 0x003c0014), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_GS_NGG_CLK_CTRL, 0xffff8fff, 0xffff8100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_IA_CLK_CTRL, 0xffff0fff, 0xffff0100), @@ -1464,13 +1459,11 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00c00000) }; -static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_1_nv14[] = { /* Pending on emulation bring up */ }; -static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_nv14[] = -{ +static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_nv14[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xE0000000L, 0x0), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), @@ -2093,13 +2086,11 @@ static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_nv14[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xFFFFFFFF, 0xe0000000) }; -static const struct soc15_reg_golden golden_settings_gc_10_1_2_nv12[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_1_2_nv12[] = { /* Pending on emulation bring up */ }; -static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_2_nv12[] = -{ +static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_2_nv12[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xe0000000L, 0x0), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xffffff, 0x0), SOC15_REG_GOLDEN_VALUE(GC, 0, mmRLC_SPM_GLB_SAMPLEDELAY_IND_ADDR, 0xFFFFFFFF, 0x28), @@ -3154,8 +3145,7 @@ static const struct soc15_reg_golden golden_settings_gc_rlc_spm_10_1_2_nv12[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmGRBM_GFX_INDEX, 0xFFFFFFFF, 0xe0000000) }; -static const struct soc15_reg_golden golden_settings_gc_10_3[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0x78000000, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_PS_CLK_CTRL, 0xff7f0fff, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0xff7f0fff, 0x30000100), @@ -3164,7 +3154,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG3, 0xffffffff, 0x00000280), SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_DEBUG4, 0xffffffff, 0x00800000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmDB_EXCEPTION_CONTROL, 0x7fff0f1f, 0x00b80000), - SOC15_REG_GOLDEN_VALUE(GC, 0 ,mmGCEA_SDP_TAG_RESERVE0, 0xffffffff, 0x10100100), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCEA_SDP_TAG_RESERVE0, 0xffffffff, 0x10100100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCEA_SDP_TAG_RESERVE1, 0xffffffff, 0x17000088), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCR_GENERAL_CNTL_Sienna_Cichlid, 0x1ff1ffff, 0x00000500), SOC15_REG_GOLDEN_VALUE(GC, 0, mmGCUTCL2_CGTT_CLK_CTRL_Sienna_Cichlid, 0xff000000, 0xff008080), @@ -3201,13 +3191,11 @@ static const struct soc15_reg_golden golden_settings_gc_10_3[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffbfffff, 0x00a00000) }; -static const struct soc15_reg_golden golden_settings_gc_10_3_sienna_cichlid[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3_sienna_cichlid[] = { /* Pending on emulation bring up */ }; -static const struct soc15_reg_golden golden_settings_gc_10_3_2[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3_2[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_PS_CLK_CTRL, 0xff7f0fff, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0xff7f0fff, 0x30000100), @@ -3254,8 +3242,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_2[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG, 0x00000020, 0x00000020), }; -static const struct soc15_reg_golden golden_settings_gc_10_3_vangogh[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3_vangogh[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0xff7f0fff, 0x30000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA1_CLK_CTRL, 0xff7f0fff, 0x7e000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x000000e4), @@ -3285,8 +3272,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_vangogh[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmLDS_CONFIG, 0x00000020, 0x00000020), }; -static const struct soc15_reg_golden golden_settings_gc_10_3_3[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3_3[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x000000e4), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c200), @@ -3309,8 +3295,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_3[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffffffff, 0x00100000) }; -static const struct soc15_reg_golden golden_settings_gc_10_3_4[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3_4[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0x78000000, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA0_CLK_CTRL, 0x30000000, 0x30000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_RA1_CLK_CTRL, 0x7e000000, 0x7e000100), @@ -3380,7 +3365,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_3_5[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER7_SELECT, 0xf0f001ff, 0x00000000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER8_SELECT, 0xf0f001ff, 0x00000000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmSQ_PERFCOUNTER9_SELECT, 0xf0f001ff, 0x00000000), - SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX,0xfff7ffff, 0x01030000), + SOC15_REG_GOLDEN_VALUE(GC, 0, mmTA_CNTL_AUX, 0xfff7ffff, 0x01030000), SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0xffbfffff, 0x00a00000) }; @@ -3421,8 +3406,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_0_cyan_skillfish[] = SOC15_REG_GOLDEN_VALUE(GC, 0, mmUTCL1_CTRL, 0x00800000, 0x00800000) }; -static const struct soc15_reg_golden golden_settings_gc_10_3_6[] = -{ +static const struct soc15_reg_golden golden_settings_gc_10_3_6[] = { SOC15_REG_GOLDEN_VALUE(GC, 0, mmCGTT_SPI_CS_CLK_CTRL, 0xff7f0fff, 0x78000100), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCH_PIPE_STEER, 0x000000ff, 0x00000044), SOC15_REG_GOLDEN_VALUE(GC, 0, mmCPF_GCR_CNTL, 0x0007ffff, 0x0000c200), @@ -3506,6 +3490,8 @@ static void gfx_v10_3_set_power_brake_sequence(struct amdgpu_device *adev); static void gfx_v10_0_ring_invalidate_tlbs(struct amdgpu_ring *ring, uint16_t pasid, uint32_t flush_type, bool all_hub, uint8_t dst_sel); +static void gfx_v10_0_update_spm_vmid_internal(struct amdgpu_device *adev, + unsigned int vmid); static void gfx10_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) { @@ -3714,8 +3700,8 @@ static void gfx_v10_0_init_golden_registers(struct amdgpu_device *adev) break; case IP_VERSION(10, 3, 4): soc15_program_register_sequence(adev, - golden_settings_gc_10_3_4, - (const u32)ARRAY_SIZE(golden_settings_gc_10_3_4)); + golden_settings_gc_10_3_4, + (const u32)ARRAY_SIZE(golden_settings_gc_10_3_4)); break; case IP_VERSION(10, 3, 5): soc15_program_register_sequence(adev, @@ -3782,7 +3768,7 @@ static int gfx_v10_0_ring_test_ring(struct amdgpu_ring *ring) struct amdgpu_device *adev = ring->adev; uint32_t scratch = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG0); uint32_t tmp = 0; - unsigned i; + unsigned int i; int r; WREG32(scratch, 0xCAFEDEAD); @@ -3820,7 +3806,7 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout) struct amdgpu_device *adev = ring->adev; struct amdgpu_ib ib; struct dma_fence *f = NULL; - unsigned index; + unsigned int index; uint64_t gpu_addr; volatile uint32_t *cpu_ptr; long r; @@ -3951,7 +3937,7 @@ static bool gfx_v10_0_navi10_gfxoff_should_enable(struct amdgpu_device *adev) break; } - return ret ; + return ret; } static void gfx_v10_0_check_gfxoff_flag(struct amdgpu_device *adev) @@ -4151,7 +4137,7 @@ static void gfx_v10_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) { struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; - reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[0]; reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG0); reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG1); reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG2); @@ -4159,14 +4145,14 @@ static void gfx_v10_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) reg_access_ctrl->grbm_cntl = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_CNTL); reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_INDEX); switch (adev->ip_versions[GC_HWIP][0]) { - case IP_VERSION(10, 3, 0): - reg_access_ctrl->spare_int = - SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT_0_Sienna_Cichlid); - break; - default: - reg_access_ctrl->spare_int = - SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT); - break; + case IP_VERSION(10, 3, 0): + reg_access_ctrl->spare_int = + SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT_0_Sienna_Cichlid); + break; + default: + reg_access_ctrl->spare_int = + SOC15_REG_OFFSET(GC, 0, mmRLC_SPARE_INT); + break; } adev->gfx.rlc.rlcg_reg_access_supported = true; } @@ -4187,11 +4173,6 @@ static int gfx_v10_0_rlc_init(struct amdgpu_device *adev) return r; } - /* init spm vmid with 0xf */ - if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); - - return 0; } @@ -4213,7 +4194,7 @@ static int gfx_v10_0_mec_init(struct amdgpu_device *adev) int r; u32 *hpd; const __le32 *fw_data = NULL; - unsigned fw_size; + unsigned int fw_size; u32 *fw = NULL; size_t mec_hpd_size; @@ -4295,7 +4276,8 @@ static void gfx_v10_0_read_wave_data(struct amdgpu_device *adev, uint32_t xcc_id { /* in gfx10 the SIMD_ID is specified as part of the INSTANCE * field when performing a select_se_sh so it should be - * zero here */ + * zero here + */ WARN_ON(simd != 0); /* type 2 wave data */ @@ -4474,7 +4456,7 @@ static int gfx_v10_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id, static int gfx_v10_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, int mec, int pipe, int queue) { - unsigned irq_type; + unsigned int irq_type; struct amdgpu_ring *ring; unsigned int hw_prio; @@ -4795,7 +4777,8 @@ static u32 gfx_v10_0_init_pa_sc_tile_steering_override(struct amdgpu_device *ade uint32_t pa_sc_tile_steering_override; /* for ASICs that integrates GFX v10.3 - * pa_sc_tile_steering_override should be set to 0 */ + * pa_sc_tile_steering_override should be set to 0 + */ if (adev->ip_versions[GC_HWIP][0] >= IP_VERSION(10, 3, 0)) return 0; @@ -4871,8 +4854,10 @@ static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev) nv_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); - /* Initialize all compute VMIDs to have no GDS, GWS, or OA - access. These should be enabled by FW for target VMIDs. */ + /* + * Initialize all compute VMIDs to have no GDS, GWS, or OA + * access. These should be enabled by FW for target VMIDs. + */ for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { WREG32_SOC15_OFFSET(GC, 0, mmGDS_VMID0_BASE, 2 * i, 0); WREG32_SOC15_OFFSET(GC, 0, mmGDS_VMID0_SIZE, 2 * i, 0); @@ -5108,8 +5093,10 @@ static void gfx_v10_0_rlc_smu_handshake_cntl(struct amdgpu_device *adev, static void gfx_v10_0_rlc_start(struct amdgpu_device *adev) { - /* TODO: enable rlc & smu handshake until smu - * and gfxoff feature works as expected */ + /* + * TODO: enable rlc & smu handshake until smu + * and gfxoff feature works as expected + */ if (!(amdgpu_pp_feature_mask & PP_GFXOFF_MASK)) gfx_v10_0_rlc_smu_handshake_cntl(adev, false); @@ -5132,7 +5119,7 @@ static int gfx_v10_0_rlc_load_microcode(struct amdgpu_device *adev) { const struct rlc_firmware_header_v2_0 *hdr; const __le32 *fw_data; - unsigned i, fw_size; + unsigned int i, fw_size; if (!adev->gfx.rlc_fw) return -EINVAL; @@ -5169,6 +5156,8 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev) gfx_v10_0_init_csb(adev); + gfx_v10_0_update_spm_vmid_internal(adev, 0xf); + if (!amdgpu_sriov_vf(adev)) /* enable RLC SRM */ gfx_v10_0_rlc_enable_srm(adev); } else { @@ -5199,6 +5188,8 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev) gfx_v10_0_init_csb(adev); + gfx_v10_0_update_spm_vmid_internal(adev, 0xf); + adev->gfx.rlc.funcs->start(adev); if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) { @@ -5207,6 +5198,7 @@ static int gfx_v10_0_rlc_resume(struct amdgpu_device *adev) return r; } } + return 0; } @@ -5674,11 +5666,10 @@ static int gfx_v10_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, enable ? 0 : 1); - if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 2)) { + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(10, 1, 2)) WREG32_SOC15_RLC(GC, 0, mmCP_ME_CNTL, tmp); - } else { + else WREG32_SOC15(GC, 0, mmCP_ME_CNTL, tmp); - } if (adev->job_hang && !enable) return 0; @@ -5700,7 +5691,7 @@ static int gfx_v10_0_cp_gfx_load_pfp_microcode(struct amdgpu_device *adev) int r; const struct gfx_firmware_header_v1_0 *pfp_hdr; const __le32 *fw_data; - unsigned i, fw_size; + unsigned int i, fw_size; uint32_t tmp; uint32_t usec_timeout = 50000; /* wait for 50ms */ @@ -5778,7 +5769,7 @@ static int gfx_v10_0_cp_gfx_load_ce_microcode(struct amdgpu_device *adev) int r; const struct gfx_firmware_header_v1_0 *ce_hdr; const __le32 *fw_data; - unsigned i, fw_size; + unsigned int i, fw_size; uint32_t tmp; uint32_t usec_timeout = 50000; /* wait for 50ms */ @@ -5855,7 +5846,7 @@ static int gfx_v10_0_cp_gfx_load_me_microcode(struct amdgpu_device *adev) int r; const struct gfx_firmware_header_v1_0 *me_hdr; const __le32 *fw_data; - unsigned i, fw_size; + unsigned int i, fw_size; uint32_t tmp; uint32_t usec_timeout = 50000; /* wait for 50ms */ @@ -6243,7 +6234,7 @@ static int gfx_v10_0_cp_compute_load_microcode(struct amdgpu_device *adev) { const struct gfx_firmware_header_v1_0 *mec_hdr; const __le32 *fw_data; - unsigned i; + unsigned int i; u32 tmp; u32 usec_timeout = 50000; /* Wait for 50 ms */ @@ -6922,8 +6913,10 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev) { uint32_t data, pattern = 0xDEADBEEF; - /* check if mmVGT_ESGS_RING_SIZE_UMD - * has been remapped to mmVGT_ESGS_RING_SIZE */ + /* + * check if mmVGT_ESGS_RING_SIZE_UMD + * has been remapped to mmVGT_ESGS_RING_SIZE + */ switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(10, 3, 0): case IP_VERSION(10, 3, 2): @@ -6934,12 +6927,10 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_UMD, pattern); if (RREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_Sienna_Cichlid) == pattern) { - WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_UMD , data); + WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_UMD, data); return true; - } else { - WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_Sienna_Cichlid, data); - return false; } + WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_Sienna_Cichlid, data); break; case IP_VERSION(10, 3, 1): case IP_VERSION(10, 3, 3): @@ -6954,12 +6945,12 @@ static bool gfx_v10_0_check_grbm_cam_remapping(struct amdgpu_device *adev) if (RREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE) == pattern) { WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE_UMD, data); return true; - } else { - WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE, data); - return false; } + WREG32_SOC15(GC, 0, mmVGT_ESGS_RING_SIZE, data); break; } + + return false; } static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev) @@ -6969,8 +6960,10 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) return; - /* initialize cam_index to 0 - * index will auto-inc after each data writting */ + /* + * Initialize cam_index to 0 + * index will auto-inc after each data writing + */ WREG32_SOC15(GC, 0, mmGRBM_CAM_INDEX, 0); switch (adev->ip_versions[GC_HWIP][0]) { @@ -7100,6 +7093,7 @@ static void gfx_v10_0_setup_grbm_cam_remapping(struct amdgpu_device *adev) static void gfx_v10_0_disable_gpa_mode(struct amdgpu_device *adev) { uint32_t data; + data = RREG32_SOC15(GC, 0, mmCPC_PSP_DEBUG); data |= CPC_PSP_DEBUG__GPA_OVERRIDE_MASK; WREG32_SOC15(GC, 0, mmCPC_PSP_DEBUG, data); @@ -7216,7 +7210,7 @@ static bool gfx_v10_0_is_idle(void *handle) static int gfx_v10_0_wait_for_idle(void *handle) { - unsigned i; + unsigned int i; u32 tmp; struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -7471,7 +7465,7 @@ static bool gfx_v10_0_is_rlc_enabled(struct amdgpu_device *adev) static void gfx_v10_0_set_safe_mode(struct amdgpu_device *adev, int xcc_id) { uint32_t data; - unsigned i; + unsigned int i; data = RLC_SAFE_MODE__CMD_MASK; data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT); @@ -7900,12 +7894,11 @@ static int gfx_v10_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } -static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +static void gfx_v10_0_update_spm_vmid_internal(struct amdgpu_device *adev, + unsigned int vmid) { u32 reg, data; - amdgpu_gfx_off_ctrl(adev, false); - /* not for *_SOC15 */ reg = SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_MC_CNTL); if (amdgpu_sriov_is_pp_one_vf(adev)) @@ -7920,6 +7913,13 @@ static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data); else WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); +} + +static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned int vmid) +{ + amdgpu_gfx_off_ctrl(adev, false); + + gfx_v10_0_update_spm_vmid_internal(adev, vmid); amdgpu_gfx_off_ctrl(adev, true); } @@ -8297,7 +8297,7 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, struct amdgpu_ib *ib, uint32_t flags) { - unsigned vmid = AMDGPU_JOB_GET_VMID(job); + unsigned int vmid = AMDGPU_JOB_GET_VMID(job); u32 header, control = 0; if (ib->flags & AMDGPU_IB_FLAG_CE) @@ -8338,7 +8338,7 @@ static void gfx_v10_0_ring_emit_ib_compute(struct amdgpu_ring *ring, struct amdgpu_ib *ib, uint32_t flags) { - unsigned vmid = AMDGPU_JOB_GET_VMID(job); + unsigned int vmid = AMDGPU_JOB_GET_VMID(job); u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24); if (ring->is_mes_queue) @@ -8373,7 +8373,7 @@ static void gfx_v10_0_ring_emit_ib_compute(struct amdgpu_ring *ring, } static void gfx_v10_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, - u64 seq, unsigned flags) + u64 seq, unsigned int flags) { bool write64bit = flags & AMDGPU_FENCE_FLAG_64BIT; bool int_sel = flags & AMDGPU_FENCE_FLAG_INT; @@ -8429,7 +8429,7 @@ static void gfx_v10_0_ring_invalidate_tlbs(struct amdgpu_ring *ring, } static void gfx_v10_0_ring_emit_vm_flush(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { if (ring->is_mes_queue) gfx_v10_0_ring_invalidate_tlbs(ring, 0, 0, false, 0); @@ -8511,9 +8511,9 @@ static void gfx_v10_0_ring_emit_cntxcntl(struct amdgpu_ring *ring, amdgpu_ring_write(ring, 0); } -static unsigned gfx_v10_0_ring_emit_init_cond_exec(struct amdgpu_ring *ring) +static unsigned int gfx_v10_0_ring_emit_init_cond_exec(struct amdgpu_ring *ring) { - unsigned ret; + unsigned int ret; amdgpu_ring_write(ring, PACKET3(PACKET3_COND_EXEC, 3)); amdgpu_ring_write(ring, lower_32_bits(ring->cond_exe_gpu_addr)); @@ -8525,9 +8525,10 @@ static unsigned gfx_v10_0_ring_emit_init_cond_exec(struct amdgpu_ring *ring) return ret; } -static void gfx_v10_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigned offset) +static void gfx_v10_0_ring_emit_patch_cond_exec(struct amdgpu_ring *ring, unsigned int offset) { - unsigned cur; + unsigned int cur; + BUG_ON(offset > ring->buf_mask); BUG_ON(ring->ring[offset] != 0x55aa55aa); @@ -8750,7 +8751,7 @@ static void gfx_v10_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, } static void gfx_v10_0_ring_soft_recovery(struct amdgpu_ring *ring, - unsigned vmid) + unsigned int vmid) { struct amdgpu_device *adev = ring->adev; uint32_t value = 0; @@ -8859,7 +8860,7 @@ static void gfx_v10_0_set_compute_eop_interrupt_state(struct amdgpu_device *adev static int gfx_v10_0_set_eop_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { switch (type) { @@ -8956,7 +8957,7 @@ static int gfx_v10_0_eop_irq(struct amdgpu_device *adev, static int gfx_v10_0_set_priv_reg_fault_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { switch (state) { @@ -8975,7 +8976,7 @@ static int gfx_v10_0_set_priv_reg_fault_state(struct amdgpu_device *adev, static int gfx_v10_0_set_priv_inst_fault_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { switch (state) { @@ -9342,7 +9343,7 @@ static void gfx_v10_0_set_rlc_funcs(struct amdgpu_device *adev) static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev) { - unsigned total_cu = adev->gfx.config.max_cu_per_sh * + unsigned int total_cu = adev->gfx.config.max_cu_per_sh * adev->gfx.config.max_sh_per_se * adev->gfx.config.max_shader_engines; @@ -9423,7 +9424,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, { int i, j, k, counter, active_cu_number = 0; u32 mask, bitmap, ao_bitmap, ao_cu_mask = 0; - unsigned disable_masks[4 * 2]; + unsigned int disable_masks[4 * 2]; if (!adev || !cu_info) return -EINVAL; @@ -9448,7 +9449,7 @@ static int gfx_v10_0_get_cu_info(struct amdgpu_device *adev, gfx_v10_0_set_user_wgp_inactive_bitmap_per_sh( adev, disable_masks[i * 2 + j]); bitmap = gfx_v10_0_get_cu_active_bitmap_per_sh(adev); - cu_info->bitmap[i][j] = bitmap; + cu_info->bitmap[0][i][j] = bitmap; for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { if (bitmap & mask) { @@ -9540,8 +9541,7 @@ static void gfx_v10_3_set_power_brake_sequence(struct amdgpu_device *adev) (0x1 << DIDT_SQ_THROTTLE_CTRL__PWRBRK_STALL_EN__SHIFT)); } -const struct amdgpu_ip_block_version gfx_v10_0_ip_block = -{ +const struct amdgpu_ip_block_version gfx_v10_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_GFX, .major = 10, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 3a7af59e83ca..762d7a19f1be 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -471,8 +471,12 @@ static void gfx_v11_0_check_fw_cp_gfx_shadow(struct amdgpu_device *adev) case IP_VERSION(11, 0, 3): if ((adev->gfx.me_fw_version >= 1505) && (adev->gfx.pfp_fw_version >= 1600) && - (adev->gfx.mec_fw_version >= 512)) - adev->gfx.cp_gfx_shadow = true; + (adev->gfx.mec_fw_version >= 512)) { + if (amdgpu_sriov_vf(adev)) + adev->gfx.cp_gfx_shadow = true; + else + adev->gfx.cp_gfx_shadow = false; + } break; default: adev->gfx.cp_gfx_shadow = false; @@ -663,7 +667,7 @@ static void gfx_v11_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) { struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; - reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[0]; reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG0); reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG1); reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, 0, regSCRATCH_REG2); @@ -4650,26 +4654,6 @@ static int gfx_v11_0_early_init(void *handle) return gfx_v11_0_init_microcode(adev); } -static int gfx_v11_0_ras_late_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct ras_common_if *gfx_common_if; - int ret; - - gfx_common_if = kzalloc(sizeof(struct ras_common_if), GFP_KERNEL); - if (!gfx_common_if) - return -ENOMEM; - - gfx_common_if->block = AMDGPU_RAS_BLOCK__GFX; - - ret = amdgpu_ras_feature_enable(adev, gfx_common_if, true); - if (ret) - dev_warn(adev->dev, "Failed to enable gfx11 ras feature\n"); - - kfree(gfx_common_if); - return 0; -} - static int gfx_v11_0_late_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -4683,12 +4667,6 @@ static int gfx_v11_0_late_init(void *handle) if (r) return r; - if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(11, 0, 3)) { - r = gfx_v11_0_ras_late_init(handle); - if (r) - return r; - } - return 0; } @@ -6390,7 +6368,7 @@ static int gfx_v11_0_get_cu_info(struct amdgpu_device *adev, * SE6: {SH0,SH1} --> {bitmap[2][2], bitmap[2][3]} * SE7: {SH0,SH1} --> {bitmap[3][2], bitmap[3][3]} */ - cu_info->bitmap[i % 4][j + (i / 4) * 2] = bitmap; + cu_info->bitmap[0][i % 4][j + (i / 4) * 2] = bitmap; for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { if (bitmap & mask) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index da6caff78c22..34f9211b2679 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -3577,7 +3577,7 @@ static void gfx_v6_0_get_cu_info(struct amdgpu_device *adev) gfx_v6_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * 2 + j]); bitmap = gfx_v6_0_get_cu_enabled(adev); - cu_info->bitmap[i][j] = bitmap; + cu_info->bitmap[0][i][j] = bitmap; for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { if (bitmap & mask) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 8c174c11eaee..c2faf6b4c2fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -90,8 +90,7 @@ MODULE_FIRMWARE("amdgpu/mullins_ce.bin"); MODULE_FIRMWARE("amdgpu/mullins_rlc.bin"); MODULE_FIRMWARE("amdgpu/mullins_mec.bin"); -static const struct amdgpu_gds_reg_offset amdgpu_gds_reg_offset[] = -{ +static const struct amdgpu_gds_reg_offset amdgpu_gds_reg_offset[] = { {mmGDS_VMID0_BASE, mmGDS_VMID0_SIZE, mmGDS_GWS_VMID0, mmGDS_OA_VMID0}, {mmGDS_VMID1_BASE, mmGDS_VMID1_SIZE, mmGDS_GWS_VMID1, mmGDS_OA_VMID1}, {mmGDS_VMID2_BASE, mmGDS_VMID2_SIZE, mmGDS_GWS_VMID2, mmGDS_OA_VMID2}, @@ -110,8 +109,7 @@ static const struct amdgpu_gds_reg_offset amdgpu_gds_reg_offset[] = {mmGDS_VMID15_BASE, mmGDS_VMID15_SIZE, mmGDS_GWS_VMID15, mmGDS_OA_VMID15} }; -static const u32 spectre_rlc_save_restore_register_list[] = -{ +static const u32 spectre_rlc_save_restore_register_list[] = { (0x0e00 << 16) | (0xc12c >> 2), 0x00000000, (0x0e00 << 16) | (0xc140 >> 2), @@ -557,8 +555,7 @@ static const u32 spectre_rlc_save_restore_register_list[] = (0x0e00 << 16) | (0x9600 >> 2), }; -static const u32 kalindi_rlc_save_restore_register_list[] = -{ +static const u32 kalindi_rlc_save_restore_register_list[] = { (0x0e00 << 16) | (0xc12c >> 2), 0x00000000, (0x0e00 << 16) | (0xc140 >> 2), @@ -933,7 +930,8 @@ static int gfx_v7_0_init_microcode(struct amdgpu_device *adev) case CHIP_MULLINS: chip_name = "mullins"; break; - default: BUG(); + default: + BUG(); } snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_pfp.bin", chip_name); @@ -2759,8 +2757,7 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev) return 0; } -struct hqd_registers -{ +struct hqd_registers { u32 cp_mqd_base_addr; u32 cp_mqd_base_addr_hi; u32 cp_hqd_active; @@ -5122,13 +5119,13 @@ static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev) gfx_v7_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * 2 + j]); bitmap = gfx_v7_0_get_cu_active_bitmap(adev); - cu_info->bitmap[i][j] = bitmap; + cu_info->bitmap[0][i][j] = bitmap; - for (k = 0; k < adev->gfx.config.max_cu_per_sh; k ++) { + for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { if (bitmap & mask) { if (counter < ao_cu_num) ao_bitmap |= mask; - counter ++; + counter++; } mask <<= 1; } @@ -5150,8 +5147,7 @@ static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev) cu_info->lds_size = 64; } -const struct amdgpu_ip_block_version gfx_v7_1_ip_block = -{ +const struct amdgpu_ip_block_version gfx_v7_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_GFX, .major = 7, .minor = 1, @@ -5159,8 +5155,7 @@ const struct amdgpu_ip_block_version gfx_v7_1_ip_block = .funcs = &gfx_v7_0_ip_funcs, }; -const struct amdgpu_ip_block_version gfx_v7_2_ip_block = -{ +const struct amdgpu_ip_block_version gfx_v7_2_ip_block = { .type = AMD_IP_BLOCK_TYPE_GFX, .major = 7, .minor = 2, @@ -5168,8 +5163,7 @@ const struct amdgpu_ip_block_version gfx_v7_2_ip_block = .funcs = &gfx_v7_0_ip_funcs, }; -const struct amdgpu_ip_block_version gfx_v7_3_ip_block = -{ +const struct amdgpu_ip_block_version gfx_v7_3_ip_block = { .type = AMD_IP_BLOCK_TYPE_GFX, .major = 7, .minor = 3, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 51c1745c8369..885ebd703260 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -7121,7 +7121,7 @@ static void gfx_v8_0_get_cu_info(struct amdgpu_device *adev) gfx_v8_0_set_user_cu_inactive_bitmap( adev, disable_masks[i * 2 + j]); bitmap = gfx_v8_0_get_cu_active_bitmap(adev); - cu_info->bitmap[i][j] = bitmap; + cu_info->bitmap[0][i][j] = bitmap; for (k = 0; k < adev->gfx.config.max_cu_per_sh; k ++) { if (bitmap & mask) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 65577eca58f1..fd61574a737c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -762,6 +762,8 @@ static void gfx_v9_0_query_ras_error_count(struct amdgpu_device *adev, static int gfx_v9_0_ras_error_inject(struct amdgpu_device *adev, void *inject_if, uint32_t instance_mask); static void gfx_v9_0_reset_ras_error_count(struct amdgpu_device *adev); +static void gfx_v9_0_update_spm_vmid_internal(struct amdgpu_device *adev, + unsigned int vmid); static void gfx_v9_0_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) @@ -1497,7 +1499,7 @@ static void gfx_v9_0_init_always_on_cu_mask(struct amdgpu_device *adev) amdgpu_gfx_select_se_sh(adev, i, j, 0xffffffff, 0); for (k = 0; k < adev->gfx.config.max_cu_per_sh; k ++) { - if (cu_info->bitmap[i][j] & mask) { + if (cu_info->bitmap[0][i][j] & mask) { if (counter == pg_always_on_cu_num) WREG32_SOC15(GC, 0, mmRLC_PG_ALWAYS_ON_CU_MASK, cu_bitmap); if (counter < always_on_cu_num) @@ -1632,7 +1634,7 @@ static void gfx_v9_0_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) { struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; - reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl; + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[0]; reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG0); reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG1); reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, 0, mmSCRATCH_REG2); @@ -1667,22 +1669,6 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev) return r; } - switch (adev->ip_versions[GC_HWIP][0]) { - case IP_VERSION(9, 2, 2): - case IP_VERSION(9, 1, 0): - gfx_v9_0_init_lbpw(adev); - break; - case IP_VERSION(9, 4, 0): - gfx_v9_4_init_lbpw(adev); - break; - default: - break; - } - - /* init spm vmid with 0xf */ - if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); - return 0; } @@ -2942,12 +2928,14 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev) switch (adev->ip_versions[GC_HWIP][0]) { case IP_VERSION(9, 2, 2): case IP_VERSION(9, 1, 0): + gfx_v9_0_init_lbpw(adev); if (amdgpu_lbpw == 0) gfx_v9_0_enable_lbpw(adev, false); else gfx_v9_0_enable_lbpw(adev, true); break; case IP_VERSION(9, 4, 0): + gfx_v9_4_init_lbpw(adev); if (amdgpu_lbpw > 0) gfx_v9_0_enable_lbpw(adev, true); else @@ -2957,6 +2945,8 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev) break; } + gfx_v9_0_update_spm_vmid_internal(adev, 0xf); + adev->gfx.rlc.funcs->start(adev); return 0; @@ -4881,12 +4871,11 @@ static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } -static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +static void gfx_v9_0_update_spm_vmid_internal(struct amdgpu_device *adev, + unsigned int vmid) { u32 reg, data; - amdgpu_gfx_off_ctrl(adev, false); - reg = SOC15_REG_OFFSET(GC, 0, mmRLC_SPM_MC_CNTL); if (amdgpu_sriov_is_pp_one_vf(adev)) data = RREG32_NO_KIQ(reg); @@ -4900,6 +4889,13 @@ static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data); else WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); +} + +static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned int vmid) +{ + amdgpu_gfx_off_ctrl(adev, false); + + gfx_v9_0_update_spm_vmid_internal(adev, vmid); amdgpu_gfx_off_ctrl(adev, true); } @@ -5230,6 +5226,9 @@ static void gfx_v9_0_ring_patch_de_meta(struct amdgpu_ring *ring, de_payload_cpu_addr = adev->virt.csa_cpu_addr + payload_offset; } + ((struct v9_de_ib_state *)de_payload_cpu_addr)->ib_completion_status = + IB_COMPLETION_STATUS_PREEMPTED; + if (offset + (payload_size >> 2) <= ring->buf_mask + 1) { memcpy((void *)&ring->ring[offset], de_payload_cpu_addr, payload_size); } else { @@ -7234,7 +7233,7 @@ static int gfx_v9_0_get_cu_info(struct amdgpu_device *adev, * SE6,SH0 --> bitmap[2][1] * SE7,SH0 --> bitmap[3][1] */ - cu_info->bitmap[i % 4][j + i / 4] = bitmap; + cu_info->bitmap[0][i % 4][j + i / 4] = bitmap; for (k = 0; k < adev->gfx.config.max_cu_per_sh; k ++) { if (bitmap & mask) { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index 9e3b835bdbb2..18ce5fe45f6f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -46,6 +46,7 @@ MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin"); #define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L #define GOLDEN_GB_ADDR_CONFIG 0x2a114042 +#define CP_HQD_PERSISTENT_STATE_DEFAULT 0xbe05301 struct amdgpu_gfx_ras gfx_v9_4_3_ras; @@ -195,16 +196,16 @@ static void gfx_v9_4_3_init_golden_registers(struct amdgpu_device *adev) num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < num_xcc; i++) { dev_inst = GET_INST(GC, i); - if (dev_inst >= 2) - WREG32_SOC15(GC, dev_inst, regGRBM_MCM_ADDR, 0x4); + WREG32_SOC15(GC, dev_inst, regGB_ADDR_CONFIG, + GOLDEN_GB_ADDR_CONFIG); /* Golden settings applied by driver for ASIC with rev_id 0 */ if (adev->rev_id == 0) { - WREG32_SOC15(GC, dev_inst, regGB_ADDR_CONFIG, - GOLDEN_GB_ADDR_CONFIG); - WREG32_FIELD15_PREREG(GC, dev_inst, TCP_UTCL1_CNTL1, REDUCE_FIFO_DEPTH_BY_2, 2); + } else { + WREG32_FIELD15_PREREG(GC, dev_inst, TCP_UTCL1_CNTL2, + SPARE, 0x1); } } } @@ -339,13 +340,11 @@ static uint64_t gfx_v9_4_3_get_gpu_clock_counter(struct amdgpu_device *adev) { uint64_t clock; - amdgpu_gfx_off_ctrl(adev, false); mutex_lock(&adev->gfx.gpu_clock_mutex); WREG32_SOC15(GC, GET_INST(GC, 0), regRLC_CAPTURE_GPU_CLOCK_COUNT, 1); clock = (uint64_t)RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_GPU_CLOCK_COUNT_LSB) | ((uint64_t)RREG32_SOC15(GC, GET_INST(GC, 0), regRLC_GPU_CLOCK_COUNT_MSB) << 32ULL); mutex_unlock(&adev->gfx.gpu_clock_mutex); - amdgpu_gfx_off_ctrl(adev, true); return clock; } @@ -624,7 +623,7 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev, int num_xccs_per_xcp) { int ret, i, num_xcc; - u32 tmp = 0; + u32 tmp = 0, regval; if (adev->psp.funcs) { ret = psp_spatial_partition(&adev->psp, @@ -632,23 +631,24 @@ static int gfx_v9_4_3_switch_compute_partition(struct amdgpu_device *adev, num_xccs_per_xcp); if (ret) return ret; - } else { - num_xcc = NUM_XCC(adev->gfx.xcc_mask); + } - for (i = 0; i < num_xcc; i++) { - tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, NUM_XCC_IN_XCP, - num_xccs_per_xcp); - tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID, - i % num_xccs_per_xcp); + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + + for (i = 0; i < num_xcc; i++) { + tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, NUM_XCC_IN_XCP, + num_xccs_per_xcp); + tmp = REG_SET_FIELD(tmp, CP_HYP_XCP_CTL, VIRTUAL_XCC_ID, + i % num_xccs_per_xcp); + regval = RREG32_SOC15(GC, GET_INST(GC, i), regCP_HYP_XCP_CTL); + if (regval != tmp) WREG32_SOC15(GC, GET_INST(GC, i), regCP_HYP_XCP_CTL, tmp); - } - ret = 0; } adev->gfx.num_xcc_per_xcp = num_xccs_per_xcp; - return ret; + return 0; } static int gfx_v9_4_3_ih_to_xcc_inst(struct amdgpu_device *adev, int ih_node) @@ -863,11 +863,15 @@ static int gfx_v9_4_3_sw_init(void *handle) if (r) return r; - r = amdgpu_gfx_sysfs_init(adev); + r = amdgpu_gfx_ras_sw_init(adev); if (r) return r; - return amdgpu_gfx_ras_sw_init(adev); + + if (!amdgpu_sriov_vf(adev)) + r = amdgpu_gfx_sysfs_init(adev); + + return r; } static int gfx_v9_4_3_sw_fini(void *handle) @@ -888,7 +892,8 @@ static int gfx_v9_4_3_sw_fini(void *handle) gfx_v9_4_3_mec_fini(adev); amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj); gfx_v9_4_3_free_microcode(adev); - amdgpu_gfx_sysfs_fini(adev); + if (!amdgpu_sriov_vf(adev)) + amdgpu_gfx_sysfs_fini(adev); return 0; } @@ -900,6 +905,7 @@ static void gfx_v9_4_3_xcc_init_compute_vmid(struct amdgpu_device *adev, int i; uint32_t sh_mem_config; uint32_t sh_mem_bases; + uint32_t data; /* * Configure apertures: @@ -919,6 +925,11 @@ static void gfx_v9_4_3_xcc_init_compute_vmid(struct amdgpu_device *adev, /* CP and shaders */ WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSH_MEM_CONFIG, sh_mem_config); WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSH_MEM_BASES, sh_mem_bases); + + /* Enable trap for each kfd vmid. */ + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regSPI_GDBG_PER_VMID_CNTL); + data = REG_SET_FIELD(data, SPI_GDBG_PER_VMID_CNTL, TRAP_EN, 1); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regSPI_GDBG_PER_VMID_CNTL, data); } soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); @@ -1037,32 +1048,6 @@ static void gfx_v9_4_3_xcc_disable_gpa_mode(struct amdgpu_device *adev, int xcc_ WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCPC_PSP_DEBUG, data); } -static void gfx_v9_4_3_xcc_program_xcc_id(struct amdgpu_device *adev, - int xcc_id) -{ - uint32_t tmp = 0; - int num_xcc; - - num_xcc = NUM_XCC(adev->gfx.xcc_mask); - switch (num_xcc) { - /* directly config VIRTUAL_XCC_ID to 0 for 1-XCC */ - case 1: - WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, 0x8); - break; - case 2: - case 4: - case 6: - case 8: - tmp = (xcc_id % adev->gfx.num_xcc_per_xcp) << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, VIRTUAL_XCC_ID); - tmp = tmp | (adev->gfx.num_xcc_per_xcp << REG_FIELD_SHIFT(CP_HYP_XCP_CTL, NUM_XCC_IN_XCP)); - WREG32_SOC15(GC, GET_INST(GC, xcc_id), regCP_HYP_XCP_CTL, tmp); - - break; - default: - break; - } -} - static bool gfx_v9_4_3_is_rlc_enabled(struct amdgpu_device *adev) { uint32_t rlc_setting; @@ -1101,6 +1086,24 @@ static void gfx_v9_4_3_xcc_unset_safe_mode(struct amdgpu_device *adev, WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_SAFE_MODE, data); } +static void gfx_v9_4_3_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev) +{ + int xcc_id, num_xcc; + struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl; + + num_xcc = NUM_XCC(adev->gfx.xcc_mask); + for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) { + reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[GET_INST(GC, xcc_id)]; + reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regSCRATCH_REG0); + reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regSCRATCH_REG1); + reg_access_ctrl->scratch_reg2 = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regSCRATCH_REG2); + reg_access_ctrl->scratch_reg3 = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regSCRATCH_REG3); + reg_access_ctrl->grbm_cntl = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regGRBM_GFX_CNTL); + reg_access_ctrl->grbm_idx = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regGRBM_GFX_INDEX); + reg_access_ctrl->spare_int = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regRLC_SPARE_INT); + } +} + static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev) { /* init spm vmid with 0xf */ @@ -1736,7 +1739,7 @@ static int gfx_v9_4_3_xcc_q_fini_register(struct amdgpu_ring *ring, WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_IQ_TIMER, 0); WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_IB_CONTROL, 0); - WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE, 0); + WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PERSISTENT_STATE, CP_HQD_PERSISTENT_STATE_DEFAULT); WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, 0x40000000); WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_DOORBELL_CONTROL, 0); WREG32_SOC15_RLC(GC, GET_INST(GC, xcc_id), regCP_HQD_PQ_RPTR, 0); @@ -1920,9 +1923,6 @@ static int gfx_v9_4_3_xcc_cp_resume(struct amdgpu_device *adev, int xcc_id) return r; } - /* set the virtual and physical id based on partition_mode */ - gfx_v9_4_3_xcc_program_xcc_id(adev, xcc_id); - r = gfx_v9_4_3_xcc_kiq_resume(adev, xcc_id); if (r) return r; @@ -2181,6 +2181,9 @@ static int gfx_v9_4_3_early_init(void *handle) gfx_v9_4_3_set_gds_init(adev); gfx_v9_4_3_set_rlc_funcs(adev); + /* init rlcg reg access ctrl */ + gfx_v9_4_3_init_rlcg_reg_access_ctrl(adev); + return gfx_v9_4_3_init_microcode(adev); } @@ -2197,6 +2200,10 @@ static int gfx_v9_4_3_late_init(void *handle) if (r) return r; + if (adev->gfx.ras && + adev->gfx.ras->enable_watchdog_timer) + adev->gfx.ras->enable_watchdog_timer(adev); + return 0; } @@ -2220,15 +2227,6 @@ static void gfx_v9_4_3_xcc_update_sram_fgcg(struct amdgpu_device *adev, WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CGTT_MGCG_OVERRIDE, data); - def = data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CLK_CNTL); - - if (enable) - data &= ~RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK; - else - data |= RLC_CLK_CNTL__RLC_SRAM_CLK_GATER_OVERRIDE_MASK; - - if (def != data) - WREG32_SOC15(GC, GET_INST(GC, xcc_id), regRLC_CLK_CNTL, data); } static void gfx_v9_4_3_xcc_update_repeater_fgcg(struct amdgpu_device *adev, @@ -4043,6 +4041,35 @@ static void gfx_v9_4_3_inst_reset_ras_err_status(struct amdgpu_device *adev, gfx_v9_4_3_inst_reset_sq_timeout_status(adev, xcc_id); } +static void gfx_v9_4_3_inst_enable_watchdog_timer(struct amdgpu_device *adev, + void *ras_error_status, int xcc_id) +{ + uint32_t i; + uint32_t data; + + data = RREG32_SOC15(GC, GET_INST(GC, 0), regSQ_TIMEOUT_CONFIG); + data = REG_SET_FIELD(data, SQ_TIMEOUT_CONFIG, TIMEOUT_FATAL_DISABLE, + amdgpu_watchdog_timer.timeout_fatal_disable ? 1 : 0); + + if (amdgpu_watchdog_timer.timeout_fatal_disable && + (amdgpu_watchdog_timer.period < 1 || + amdgpu_watchdog_timer.period > 0x23)) { + dev_warn(adev->dev, "Watchdog period range is 1 to 0x23\n"); + amdgpu_watchdog_timer.period = 0x23; + } + data = REG_SET_FIELD(data, SQ_TIMEOUT_CONFIG, PERIOD_SEL, + amdgpu_watchdog_timer.period); + + mutex_lock(&adev->grbm_idx_mutex); + for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { + gfx_v9_4_3_xcc_select_se_sh(adev, i, 0xffffffff, 0xffffffff, xcc_id); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regSQ_TIMEOUT_CONFIG, data); + } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); + mutex_unlock(&adev->grbm_idx_mutex); +} + static void gfx_v9_4_3_query_ras_error_count(struct amdgpu_device *adev, void *ras_error_status) { @@ -4065,6 +4092,11 @@ static void gfx_v9_4_3_reset_ras_error_status(struct amdgpu_device *adev) amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_reset_ras_err_status); } +static void gfx_v9_4_3_enable_watchdog_timer(struct amdgpu_device *adev) +{ + amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_enable_watchdog_timer); +} + static const struct amd_ip_funcs gfx_v9_4_3_ip_funcs = { .name = "gfx_v9_4_3", .early_init = gfx_v9_4_3_early_init, @@ -4227,7 +4259,7 @@ static void gfx_v9_4_3_set_gds_init(struct amdgpu_device *adev) } static void gfx_v9_4_3_set_user_cu_inactive_bitmap(struct amdgpu_device *adev, - u32 bitmap) + u32 bitmap, int xcc_id) { u32 data; @@ -4237,15 +4269,15 @@ static void gfx_v9_4_3_set_user_cu_inactive_bitmap(struct amdgpu_device *adev, data = bitmap << GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT; data &= GC_USER_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK; - WREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_SHADER_ARRAY_CONFIG, data); + WREG32_SOC15(GC, GET_INST(GC, xcc_id), regGC_USER_SHADER_ARRAY_CONFIG, data); } -static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev) +static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev, int xcc_id) { u32 data, mask; - data = RREG32_SOC15(GC, GET_INST(GC, 0), regCC_GC_SHADER_ARRAY_CONFIG); - data |= RREG32_SOC15(GC, GET_INST(GC, 0), regGC_USER_SHADER_ARRAY_CONFIG); + data = RREG32_SOC15(GC, GET_INST(GC, xcc_id), regCC_GC_SHADER_ARRAY_CONFIG); + data |= RREG32_SOC15(GC, GET_INST(GC, xcc_id), regGC_USER_SHADER_ARRAY_CONFIG); data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK; data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT; @@ -4258,7 +4290,7 @@ static u32 gfx_v9_4_3_get_cu_active_bitmap(struct amdgpu_device *adev) static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, struct amdgpu_cu_info *cu_info) { - int i, j, k, counter, active_cu_number = 0; + int i, j, k, counter, xcc_id, active_cu_number = 0; u32 mask, bitmap, ao_bitmap, ao_cu_mask = 0; unsigned disable_masks[4 * 4]; @@ -4277,46 +4309,38 @@ static int gfx_v9_4_3_get_cu_info(struct amdgpu_device *adev, adev->gfx.config.max_sh_per_se); mutex_lock(&adev->grbm_idx_mutex); - for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { - for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { - mask = 1; - ao_bitmap = 0; - counter = 0; - gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff, 0); - gfx_v9_4_3_set_user_cu_inactive_bitmap( - adev, disable_masks[i * adev->gfx.config.max_sh_per_se + j]); - bitmap = gfx_v9_4_3_get_cu_active_bitmap(adev); - - /* - * The bitmap(and ao_cu_bitmap) in cu_info structure is - * 4x4 size array, and it's usually suitable for Vega - * ASICs which has 4*2 SE/SH layout. - * But for Arcturus, SE/SH layout is changed to 8*1. - * To mostly reduce the impact, we make it compatible - * with current bitmap array as below: - * SE4,SH0 --> bitmap[0][1] - * SE5,SH0 --> bitmap[1][1] - * SE6,SH0 --> bitmap[2][1] - * SE7,SH0 --> bitmap[3][1] - */ - cu_info->bitmap[i % 4][j + i / 4] = bitmap; - - for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { - if (bitmap & mask) { - if (counter < adev->gfx.config.max_cu_per_sh) - ao_bitmap |= mask; - counter++; + for (xcc_id = 0; xcc_id < NUM_XCC(adev->gfx.xcc_mask); xcc_id++) { + for (i = 0; i < adev->gfx.config.max_shader_engines; i++) { + for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) { + mask = 1; + ao_bitmap = 0; + counter = 0; + gfx_v9_4_3_xcc_select_se_sh(adev, i, j, 0xffffffff, xcc_id); + gfx_v9_4_3_set_user_cu_inactive_bitmap( + adev, + disable_masks[i * adev->gfx.config.max_sh_per_se + j], + xcc_id); + bitmap = gfx_v9_4_3_get_cu_active_bitmap(adev, xcc_id); + + cu_info->bitmap[xcc_id][i][j] = bitmap; + + for (k = 0; k < adev->gfx.config.max_cu_per_sh; k++) { + if (bitmap & mask) { + if (counter < adev->gfx.config.max_cu_per_sh) + ao_bitmap |= mask; + counter++; + } + mask <<= 1; } - mask <<= 1; + active_cu_number += counter; + if (i < 2 && j < 2) + ao_cu_mask |= (ao_bitmap << (i * 16 + j * 8)); + cu_info->ao_cu_bitmap[i][j] = ao_bitmap; } - active_cu_number += counter; - if (i < 2 && j < 2) - ao_cu_mask |= (ao_bitmap << (i * 16 + j * 8)); - cu_info->ao_cu_bitmap[i % 4][j + i / 4] = ao_bitmap; } + gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, + xcc_id); } - gfx_v9_4_3_xcc_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff, - 0); mutex_unlock(&adev->grbm_idx_mutex); cu_info->number = active_cu_number; @@ -4393,4 +4417,5 @@ struct amdgpu_gfx_ras gfx_v9_4_3_ras = { .ras_block = { .hw_ops = &gfx_v9_4_3_ras_ops, }, + .enable_watchdog_timer = &gfx_v9_4_3_enable_watchdog_timer, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index d94cc1ec7242..cdc290a474a9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -103,7 +103,7 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev) min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); if (adev->apu_flags & AMD_APU_IS_RAVEN2) - /* + /* * Raven2 has a HW issue that it is unable to use the * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. * So here is the workaround that increase system @@ -248,7 +248,7 @@ static void gfxhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev) static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; - unsigned num_level, block_size; + unsigned int num_level, block_size; uint32_t tmp; int i; @@ -308,7 +308,7 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v1_0_program_invalidation(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; - unsigned i; + unsigned int i; for (i = 0 ; i < 18; ++i) { WREG32_SOC15_OFFSET(GC, 0, mmVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, @@ -375,6 +375,7 @@ static void gfxhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, bool value) { u32 tmp; + tmp = RREG32_SOC15(GC, 0, mmVM_L2_PROTECTION_FAULT_CNTL); tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c index 4dabf910334b..0834af771549 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c @@ -140,7 +140,7 @@ gfxhub_v1_2_xcc_init_system_aperture_regs(struct amdgpu_device *adev, min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18); if (adev->apu_flags & AMD_APU_IS_RAVEN2) - /* + /* * Raven2 has a HW issue that it is unable to use the * vram which is out of MC_VM_SYSTEM_APERTURE_HIGH_ADDR. * So here is the workaround that increase system @@ -315,7 +315,7 @@ static void gfxhub_v1_2_xcc_setup_vmid_config(struct amdgpu_device *adev, uint32_t xcc_mask) { struct amdgpu_vmhub *hub; - unsigned num_level, block_size; + unsigned int num_level, block_size; uint32_t tmp; int i, j; @@ -402,25 +402,6 @@ static void gfxhub_v1_2_xcc_program_invalidation(struct amdgpu_device *adev, static int gfxhub_v1_2_xcc_gart_enable(struct amdgpu_device *adev, uint32_t xcc_mask) { - uint32_t tmp_mask; - int i; - - tmp_mask = xcc_mask; - /* - * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, because they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - if (amdgpu_sriov_vf(adev)) { - for_each_inst(i, tmp_mask) { - i = ffs(tmp_mask) - 1; - WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15_RLC(GC, GET_INST(GC, i), regMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); - } - } - /* GART Enable. */ gfxhub_v1_2_xcc_init_gart_aperture_regs(adev, xcc_mask); gfxhub_v1_2_xcc_init_system_aperture_regs(adev, xcc_mask); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index f173a61c6c15..a041c6c970e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -31,7 +31,7 @@ #include "soc15_common.h" -static const char *gfxhub_client_ids[] = { +static const char * const gfxhub_client_ids[] = { "CB/DB", "Reserved", "GE1", @@ -332,7 +332,7 @@ static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v2_0_program_invalidation(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; - unsigned i; + unsigned int i; for (i = 0 ; i < 18; ++i) { WREG32_SOC15_OFFSET(GC, 0, mmGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, @@ -393,6 +393,7 @@ static void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev, bool value) { u32 tmp; + tmp = RREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_CNTL); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c index d8fc3e8088cd..7708d5ded7b8 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c @@ -34,7 +34,7 @@ #define mmGCUTCL2_HARVEST_BYPASS_GROUPS_YELLOW_CARP 0x16f8 #define mmGCUTCL2_HARVEST_BYPASS_GROUPS_YELLOW_CARP_BASE_IDX 0 -static const char *gfxhub_client_ids[] = { +static const char * const gfxhub_client_ids[] = { "CB/DB", "Reserved", "GE1", @@ -341,7 +341,7 @@ static void gfxhub_v2_1_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v2_1_program_invalidation(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; - unsigned i; + unsigned int i; for (i = 0 ; i < 18; ++i) { WREG32_SOC15_OFFSET(GC, 0, mmGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, @@ -582,6 +582,7 @@ static void gfxhub_v2_1_utcl2_harvest(struct amdgpu_device *adev) static void gfxhub_v2_1_save_regs(struct amdgpu_device *adev) { int i; + adev->gmc.VM_L2_CNTL = RREG32_SOC15(GC, 0, mmGCVM_L2_CNTL); adev->gmc.VM_L2_CNTL2 = RREG32_SOC15(GC, 0, mmGCVM_L2_CNTL2); adev->gmc.VM_DUMMY_PAGE_FAULT_CNTL = RREG32_SOC15(GC, 0, mmGCVM_DUMMY_PAGE_FAULT_CNTL); @@ -616,6 +617,7 @@ static void gfxhub_v2_1_save_regs(struct amdgpu_device *adev) static void gfxhub_v2_1_restore_regs(struct amdgpu_device *adev) { int i; + WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL, adev->gmc.VM_L2_CNTL); WREG32_SOC15(GC, 0, mmGCVM_L2_CNTL2, adev->gmc.VM_L2_CNTL2); WREG32_SOC15(GC, 0, mmGCVM_DUMMY_PAGE_FAULT_CNTL, adev->gmc.VM_DUMMY_PAGE_FAULT_CNTL); @@ -679,9 +681,8 @@ static void gfxhub_v2_1_halt(struct amdgpu_device *adev) tmp = RREG32_SOC15(GC, 0, mmGRBM_STATUS2); } - if (!time) { + if (!time) DRM_WARN("failed to wait for GRBM(EA) idle\n"); - } } const struct amdgpu_gfxhub_funcs gfxhub_v2_1_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c index c53147f9c9fc..e1c76c070ba9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c @@ -30,7 +30,7 @@ #include "navi10_enum.h" #include "soc15_common.h" -static const char *gfxhub_client_ids[] = { +static const char * const gfxhub_client_ids[] = { "CB/DB", "Reserved", "GE1", @@ -340,7 +340,7 @@ static void gfxhub_v3_0_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v3_0_program_invalidation(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; - unsigned i; + unsigned int i; for (i = 0 ; i < 18; ++i) { WREG32_SOC15_OFFSET(GC, 0, regGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c index ae777487d72e..07f369c7a1ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c @@ -33,7 +33,7 @@ #define regGCVM_L2_CNTL4_DEFAULT 0x000000c1 #define regGCVM_L2_CNTL5_DEFAULT 0x00003fe0 -static const char *gfxhub_client_ids[] = { +static const char * const gfxhub_client_ids[] = { "CB/DB", "Reserved", "GE1", @@ -345,7 +345,7 @@ static void gfxhub_v3_0_3_setup_vmid_config(struct amdgpu_device *adev) static void gfxhub_v3_0_3_program_invalidation(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_GFXHUB(0)]; - unsigned i; + unsigned int i; for (i = 0 ; i < 18; ++i) { WREG32_SOC15_OFFSET(GC, 0, regGCVM_INVALIDATE_ENG0_ADDR_RANGE_LO32, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 0c8a47989576..fa87a85e1017 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -53,16 +53,9 @@ #include "amdgpu_reset.h" -#if 0 -static const struct soc15_reg_golden golden_settings_navi10_hdp[] = -{ - /* TODO add golden setting for hdp */ -}; -#endif - static int gmc_v10_0_ecc_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { return 0; @@ -70,7 +63,7 @@ static int gmc_v10_0_ecc_interrupt_state(struct amdgpu_device *adev, static int gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev, - struct amdgpu_irq_src *src, unsigned type, + struct amdgpu_irq_src *src, unsigned int type, enum amdgpu_interrupt_state state) { switch (state) { @@ -109,9 +102,11 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { + uint32_t vmhub_index = entry->client_id == SOC15_IH_CLIENTID_VMC ? + AMDGPU_MMHUB0(0) : AMDGPU_GFXHUB(0); + struct amdgpu_vmhub *hub = &adev->vmhub[vmhub_index]; bool retry_fault = !!(entry->src_data[1] & 0x80); bool write_fault = !!(entry->src_data[1] & 0x20); - struct amdgpu_vmhub *hub = &adev->vmhub[entry->vmid_src]; struct amdgpu_task_info task_info; uint32_t status = 0; u64 addr; @@ -164,8 +159,7 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); dev_err(adev->dev, - "[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, " - "for process %s pid %d thread %s pid %d)\n", + "[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, for process %s pid %d thread %s pid %d)\n", entry->vmid_src ? "mmhub" : "gfxhub", entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, @@ -244,7 +238,7 @@ static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, u32 inv_req = hub->vmhub_funcs->get_invalidate_req(vmid, flush_type); u32 tmp; /* Use register 17 for GART */ - const unsigned eng = 17; + const unsigned int eng = 17; unsigned int i; unsigned char hub_ip = 0; @@ -346,7 +340,7 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) && down_read_trylock(&adev->reset_domain->sem)) { struct amdgpu_vmhub *hub = &adev->vmhub[vmhub]; - const unsigned eng = 17; + const unsigned int eng = 17; u32 inv_req = hub->vmhub_funcs->get_invalidate_req(vmid, flush_type); u32 req = hub->vm_inv_eng0_req + hub->eng_distance * eng; u32 ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng; @@ -477,12 +471,12 @@ static int gmc_v10_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, } static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { bool use_semaphore = gmc_v10_0_use_invalidate_semaphore(ring->adev, ring->vm_hub); struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->vm_hub]; uint32_t req = hub->vmhub_funcs->get_invalidate_req(vmid, 0); - unsigned eng = ring->vm_inv_eng; + unsigned int eng = ring->vm_inv_eng; /* * It may lose gpuvm invalidate acknowldege state across power-gating @@ -524,8 +518,8 @@ static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, return pd_addr; } -static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid, - unsigned pasid) +static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int vmid, + unsigned int pasid) { struct amdgpu_device *adev = ring->adev; uint32_t reg; @@ -645,10 +639,10 @@ static void gmc_v10_0_get_vm_pte(struct amdgpu_device *adev, AMDGPU_PTE_MTYPE_NV10(MTYPE_UC); } -static unsigned gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev) +static unsigned int gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL); - unsigned size; + unsigned int size; if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) { size = AMDGPU_VBIOS_VGA_ALLOCATION; @@ -751,6 +745,7 @@ static int gmc_v10_0_early_init(void *handle) adev->gmc.private_aperture_start = 0x1000000000000000ULL; adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + adev->gmc.noretry_flags = AMDGPU_VM_NORETRY_FLAGS_TF; return 0; } @@ -972,7 +967,7 @@ static int gmc_v10_0_sw_init(void *handle) r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { - printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); + dev_warn(adev->dev, "amdgpu: No suitable DMA available.\n"); return r; } @@ -1081,7 +1076,7 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) gmc_v10_0_flush_gpu_tlb(adev, 0, AMDGPU_GFXHUB(0), 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", - (unsigned)(adev->gmc.gart_size >> 20), + (unsigned int)(adev->gmc.gart_size >> 20), (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo)); return 0; @@ -1255,8 +1250,7 @@ const struct amd_ip_funcs gmc_v10_0_ip_funcs = { .get_clockgating_state = gmc_v10_0_get_clockgating_state, }; -const struct amdgpu_ip_block_version gmc_v10_0_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v10_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 10, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index c571f0d95994..e3b76fd28d15 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -50,7 +50,7 @@ static int gmc_v11_0_ecc_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { return 0; @@ -58,7 +58,7 @@ static int gmc_v11_0_ecc_interrupt_state(struct amdgpu_device *adev, static int gmc_v11_0_vm_fault_interrupt_state(struct amdgpu_device *adev, - struct amdgpu_irq_src *src, unsigned type, + struct amdgpu_irq_src *src, unsigned int type, enum amdgpu_interrupt_state state) { switch (state) { @@ -97,7 +97,9 @@ static int gmc_v11_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - struct amdgpu_vmhub *hub = &adev->vmhub[entry->vmid_src]; + uint32_t vmhub_index = entry->client_id == SOC21_IH_CLIENTID_VMC ? + AMDGPU_MMHUB0(0) : AMDGPU_GFXHUB(0); + struct amdgpu_vmhub *hub = &adev->vmhub[vmhub_index]; uint32_t status = 0; u64 addr; @@ -124,8 +126,7 @@ static int gmc_v11_0_process_interrupt(struct amdgpu_device *adev, amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); dev_err(adev->dev, - "[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, " - "for process %s pid %d thread %s pid %d)\n", + "[%s] page fault (src_id:%u ring:%u vmid:%u pasid:%u, for process %s pid %d thread %s pid %d)\n", entry->vmid_src ? "mmhub" : "gfxhub", entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, @@ -198,7 +199,7 @@ static void gmc_v11_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid, u32 inv_req = hub->vmhub_funcs->get_invalidate_req(vmid, flush_type); u32 tmp; /* Use register 17 for GART */ - const unsigned eng = 17; + const unsigned int eng = 17; unsigned int i; unsigned char hub_ip = 0; @@ -296,7 +297,7 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, if ((adev->gfx.kiq[0].ring.sched.ready || adev->mes.ring.sched.ready) && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) { struct amdgpu_vmhub *hub = &adev->vmhub[vmhub]; - const unsigned eng = 17; + const unsigned int eng = 17; u32 inv_req = hub->vmhub_funcs->get_invalidate_req(vmid, flush_type); u32 req = hub->vm_inv_eng0_req + hub->eng_distance * eng; u32 ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng; @@ -309,7 +310,6 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, mutex_lock(&adev->mman.gtt_window_lock); gmc_v11_0_flush_vm_hub(adev, vmid, vmhub, 0); mutex_unlock(&adev->mman.gtt_window_lock); - return; } /** @@ -379,12 +379,12 @@ static int gmc_v11_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, } static uint64_t gmc_v11_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { bool use_semaphore = gmc_v11_0_use_invalidate_semaphore(ring->adev, ring->vm_hub); struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->vm_hub]; uint32_t req = hub->vmhub_funcs->get_invalidate_req(vmid, 0); - unsigned eng = ring->vm_inv_eng; + unsigned int eng = ring->vm_inv_eng; /* * It may lose gpuvm invalidate acknowldege state across power-gating @@ -426,8 +426,8 @@ static uint64_t gmc_v11_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, return pd_addr; } -static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid, - unsigned pasid) +static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int vmid, + unsigned int pasid) { struct amdgpu_device *adev = ring->adev; uint32_t reg; @@ -547,10 +547,10 @@ static void gmc_v11_0_get_vm_pte(struct amdgpu_device *adev, AMDGPU_PTE_MTYPE_NV10(MTYPE_UC); } -static unsigned gmc_v11_0_get_vbios_fb_size(struct amdgpu_device *adev) +static unsigned int gmc_v11_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32_SOC15(DCE, 0, regD1VGA_CONTROL); - unsigned size; + unsigned int size; if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) { size = AMDGPU_VBIOS_VGA_ALLOCATION; @@ -651,6 +651,7 @@ static int gmc_v11_0_early_init(void *handle) adev->gmc.private_aperture_start = 0x1000000000000000ULL; adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + adev->gmc.noretry_flags = AMDGPU_VM_NORETRY_FLAGS_TF; return 0; } @@ -727,9 +728,9 @@ static int gmc_v11_0_mc_init(struct amdgpu_device *adev) adev->gmc.visible_vram_size = adev->gmc.real_vram_size; /* set the gart size */ - if (amdgpu_gart_size == -1) { + if (amdgpu_gart_size == -1) adev->gmc.gart_size = 512ULL << 20; - } else + else adev->gmc.gart_size = (u64)amdgpu_gart_size << 20; gmc_v11_0_vram_gtt_location(adev, &adev->gmc); @@ -823,7 +824,7 @@ static int gmc_v11_0_sw_init(void *handle) r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { - printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); + dev_warn(adev->dev, "amdgpu: No suitable DMA available.\n"); return r; } @@ -926,7 +927,7 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev) gmc_v11_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", - (unsigned)(adev->gmc.gart_size >> 20), + (unsigned int)(adev->gmc.gart_size >> 20), (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo)); return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index aa754c95a0b3..5b837a65fad2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -120,7 +120,8 @@ static int gmc_v6_0_init_microcode(struct amdgpu_device *adev) case CHIP_HAINAN: chip_name = "hainan"; break; - default: BUG(); + default: + BUG(); } /* this memory configuration requires special firmware */ @@ -178,9 +179,8 @@ static int gmc_v6_0_mc_load_microcode(struct amdgpu_device *adev) WREG32(mmMC_SEQ_IO_DEBUG_DATA, le32_to_cpup(new_io_mc_regs++)); } /* load the MC ucode */ - for (i = 0; i < ucode_size; i++) { + for (i = 0; i < ucode_size; i++) WREG32(mmMC_SEQ_SUP_PGM, le32_to_cpup(new_fw_data++)); - } /* put the engine back into the active state */ WREG32(mmMC_SEQ_SUP_CNTL, 0x00000008); @@ -208,6 +208,7 @@ static void gmc_v6_0_vram_gtt_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc) { u64 base = RREG32(mmMC_VM_FB_LOCATION) & 0xFFFF; + base <<= 24; amdgpu_gmc_vram_location(adev, mc, base); @@ -228,9 +229,8 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev) } WREG32(mmHDP_REG_COHERENCY_FLUSH_CNTL, 0); - if (gmc_v6_0_wait_for_idle((void *)adev)) { + if (gmc_v6_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); - } if (adev->mode_info.num_crtc) { u32 tmp; @@ -256,9 +256,8 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev) WREG32(mmMC_VM_AGP_TOP, 0x0FFFFFFF); WREG32(mmMC_VM_AGP_BOT, 0x0FFFFFFF); - if (gmc_v6_0_wait_for_idle((void *)adev)) { + if (gmc_v6_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); - } } static int gmc_v6_0_mc_init(struct amdgpu_device *adev) @@ -269,13 +268,13 @@ static int gmc_v6_0_mc_init(struct amdgpu_device *adev) int r; tmp = RREG32(mmMC_ARB_RAMCFG); - if (tmp & (1 << 11)) { + if (tmp & (1 << 11)) chansize = 16; - } else if (tmp & MC_ARB_RAMCFG__CHANSIZE_MASK) { + else if (tmp & MC_ARB_RAMCFG__CHANSIZE_MASK) chansize = 64; - } else { + else chansize = 32; - } + tmp = RREG32(mmMC_SHARED_CHMAP); switch ((tmp & MC_SHARED_CHMAP__NOOFCHAN_MASK) >> MC_SHARED_CHMAP__NOOFCHAN__SHIFT) { case 0: @@ -352,7 +351,7 @@ static void gmc_v6_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } static uint64_t gmc_v6_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { uint32_t reg; @@ -405,11 +404,11 @@ static void gmc_v6_0_set_fault_enable_default(struct amdgpu_device *adev, } /** - + * gmc_v8_0_set_prt - set PRT VM fault - + * - + * @adev: amdgpu_device pointer - + * @enable: enable/disable VM fault handling for PRT - +*/ + * gmc_v8_0_set_prt() - set PRT VM fault + * + * @adev: amdgpu_device pointer + * @enable: enable/disable VM fault handling for PRT + */ static void gmc_v6_0_set_prt(struct amdgpu_device *adev, bool enable) { u32 tmp; @@ -547,7 +546,7 @@ static int gmc_v6_0_gart_enable(struct amdgpu_device *adev) gmc_v6_0_flush_gpu_tlb(adev, 0, 0, 0); dev_info(adev->dev, "PCIE GART of %uM enabled (table at 0x%016llX).\n", - (unsigned)(adev->gmc.gart_size >> 20), + (unsigned int)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); return 0; } @@ -787,15 +786,16 @@ static int gmc_v6_0_late_init(void *handle) return 0; } -static unsigned gmc_v6_0_get_vbios_fb_size(struct amdgpu_device *adev) +static unsigned int gmc_v6_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32(mmD1VGA_CONTROL); - unsigned size; + unsigned int size; if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) { size = AMDGPU_VBIOS_VGA_ALLOCATION; } else { u32 viewport = RREG32(mmVIEWPORT_SIZE); + size = (REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_HEIGHT) * REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_WIDTH) * 4); @@ -814,6 +814,7 @@ static int gmc_v6_0_sw_init(void *handle) adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; } else { u32 tmp = RREG32(mmMC_SEQ_MISC0); + tmp &= MC_SEQ_MISC0__MT__MASK; adev->gmc.vram_type = gmc_v6_0_convert_vram_type(tmp); } @@ -964,7 +965,7 @@ static bool gmc_v6_0_is_idle(void *handle) static int gmc_v6_0_wait_for_idle(void *handle) { - unsigned i; + unsigned int i; struct amdgpu_device *adev = (struct amdgpu_device *)handle; for (i = 0; i < adev->usec_timeout; i++) { @@ -995,10 +996,8 @@ static int gmc_v6_0_soft_reset(void *handle) if (srbm_soft_reset) { gmc_v6_0_mc_stop(adev); - if (gmc_v6_0_wait_for_idle(adev)) { + if (gmc_v6_0_wait_for_idle(adev)) dev_warn(adev->dev, "Wait for GMC idle timed out !\n"); - } - tmp = RREG32(mmSRBM_SOFT_RESET); tmp |= srbm_soft_reset; @@ -1023,7 +1022,7 @@ static int gmc_v6_0_soft_reset(void *handle) static int gmc_v6_0_vm_fault_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { u32 tmp; @@ -1141,8 +1140,7 @@ static void gmc_v6_0_set_irq_funcs(struct amdgpu_device *adev) adev->gmc.vm_fault.funcs = &gmc_v6_0_irq_funcs; } -const struct amdgpu_ip_block_version gmc_v6_0_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v6_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 6, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index acd2b407860f..6a6929ac2748 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -58,16 +58,14 @@ MODULE_FIRMWARE("amdgpu/bonaire_mc.bin"); MODULE_FIRMWARE("amdgpu/hawaii_mc.bin"); MODULE_FIRMWARE("amdgpu/topaz_mc.bin"); -static const u32 golden_settings_iceland_a11[] = -{ +static const u32 golden_settings_iceland_a11[] = { mmVM_PRT_APERTURE0_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE1_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE2_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE3_LOW_ADDR, 0x0fffffff, 0x0fffffff }; -static const u32 iceland_mgcg_cgcg_init[] = -{ +static const u32 iceland_mgcg_cgcg_init[] = { mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104 }; @@ -151,7 +149,8 @@ static int gmc_v7_0_init_microcode(struct amdgpu_device *adev) case CHIP_KABINI: case CHIP_MULLINS: return 0; - default: BUG(); + default: + return -EINVAL; } snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mc.bin", chip_name); @@ -237,6 +236,7 @@ static void gmc_v7_0_vram_gtt_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc) { u64 base = RREG32(mmMC_VM_FB_LOCATION) & 0xFFFF; + base <<= 24; amdgpu_gmc_vram_location(adev, mc, base); @@ -266,9 +266,9 @@ static void gmc_v7_0_mc_program(struct amdgpu_device *adev) } WREG32(mmHDP_REG_COHERENCY_FLUSH_CNTL, 0); - if (gmc_v7_0_wait_for_idle((void *)adev)) { + if (gmc_v7_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); - } + if (adev->mode_info.num_crtc) { /* Lockout access through VGA aperture*/ tmp = RREG32(mmVGA_HDP_CONTROL); @@ -290,9 +290,8 @@ static void gmc_v7_0_mc_program(struct amdgpu_device *adev) WREG32(mmMC_VM_AGP_BASE, 0); WREG32(mmMC_VM_AGP_TOP, 0x0FFFFFFF); WREG32(mmMC_VM_AGP_BOT, 0x0FFFFFFF); - if (gmc_v7_0_wait_for_idle((void *)adev)) { + if (gmc_v7_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); - } WREG32(mmBIF_FB_EN, BIF_FB_EN__FB_READ_EN_MASK | BIF_FB_EN__FB_WRITE_EN_MASK); @@ -324,11 +323,11 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev) /* Get VRAM informations */ tmp = RREG32(mmMC_ARB_RAMCFG); - if (REG_GET_FIELD(tmp, MC_ARB_RAMCFG, CHANSIZE)) { + if (REG_GET_FIELD(tmp, MC_ARB_RAMCFG, CHANSIZE)) chansize = 64; - } else { + else chansize = 32; - } + tmp = RREG32(mmMC_SHARED_CHMAP); switch (REG_GET_FIELD(tmp, MC_SHARED_CHMAP, NOOFCHAN)) { case 0: @@ -472,7 +471,7 @@ static void gmc_v7_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } static uint64_t gmc_v7_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { uint32_t reg; @@ -488,8 +487,8 @@ static uint64_t gmc_v7_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, return pd_addr; } -static void gmc_v7_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid, - unsigned pasid) +static void gmc_v7_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int vmid, + unsigned int pasid) { amdgpu_ring_emit_wreg(ring, mmIH_VMID_0_LUT + vmid, pasid); } @@ -700,7 +699,7 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev) gmc_v7_0_flush_gpu_tlb(adev, 0, 0, 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", - (unsigned)(adev->gmc.gart_size >> 20), + (unsigned int)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); return 0; } @@ -761,7 +760,7 @@ static void gmc_v7_0_gart_disable(struct amdgpu_device *adev) * Print human readable fault information (CIK). */ static void gmc_v7_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, - u32 addr, u32 mc_client, unsigned pasid) + u32 addr, u32 mc_client, unsigned int pasid) { u32 vmid = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID); u32 protections = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, @@ -942,6 +941,7 @@ static int gmc_v7_0_early_init(void *handle) adev->gmc.shared_aperture_end + 1; adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + adev->gmc.noretry_flags = AMDGPU_VM_NORETRY_FLAGS_TF; return 0; } @@ -956,15 +956,16 @@ static int gmc_v7_0_late_init(void *handle) return 0; } -static unsigned gmc_v7_0_get_vbios_fb_size(struct amdgpu_device *adev) +static unsigned int gmc_v7_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32(mmD1VGA_CONTROL); - unsigned size; + unsigned int size; if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) { size = AMDGPU_VBIOS_VGA_ALLOCATION; } else { u32 viewport = RREG32(mmVIEWPORT_SIZE); + size = (REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_HEIGHT) * REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_WIDTH) * 4); @@ -984,6 +985,7 @@ static int gmc_v7_0_sw_init(void *handle) adev->gmc.vram_type = AMDGPU_VRAM_TYPE_UNKNOWN; } else { u32 tmp = RREG32(mmMC_SEQ_MISC0); + tmp &= MC_SEQ_MISC0__MT__MASK; adev->gmc.vram_type = gmc_v7_0_convert_vram_type(tmp); } @@ -1152,7 +1154,7 @@ static bool gmc_v7_0_is_idle(void *handle) static int gmc_v7_0_wait_for_idle(void *handle) { - unsigned i; + unsigned int i; u32 tmp; struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -1190,10 +1192,8 @@ static int gmc_v7_0_soft_reset(void *handle) if (srbm_soft_reset) { gmc_v7_0_mc_stop(adev); - if (gmc_v7_0_wait_for_idle((void *)adev)) { + if (gmc_v7_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for GMC idle timed out !\n"); - } - tmp = RREG32(mmSRBM_SOFT_RESET); tmp |= srbm_soft_reset; @@ -1219,7 +1219,7 @@ static int gmc_v7_0_soft_reset(void *handle) static int gmc_v7_0_vm_fault_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { u32 tmp; @@ -1383,8 +1383,7 @@ static void gmc_v7_0_set_irq_funcs(struct amdgpu_device *adev) adev->gmc.vm_fault.funcs = &gmc_v7_0_irq_funcs; } -const struct amdgpu_ip_block_version gmc_v7_0_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v7_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 7, .minor = 0, @@ -1392,8 +1391,7 @@ const struct amdgpu_ip_block_version gmc_v7_0_ip_block = .funcs = &gmc_v7_0_ip_funcs, }; -const struct amdgpu_ip_block_version gmc_v7_4_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v7_4_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 7, .minor = 4, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 85dead2a5702..5af235202513 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -64,8 +64,7 @@ MODULE_FIRMWARE("amdgpu/polaris11_k_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris10_k_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris12_k_mc.bin"); -static const u32 golden_settings_tonga_a11[] = -{ +static const u32 golden_settings_tonga_a11[] = { mmMC_ARB_WTM_GRPWT_RD, 0x00000003, 0x00000000, mmMC_HUB_RDREQ_DMIF_LIMIT, 0x0000007f, 0x00000028, mmMC_HUB_WDP_UMC, 0x00007fb6, 0x00000991, @@ -75,34 +74,29 @@ static const u32 golden_settings_tonga_a11[] = mmVM_PRT_APERTURE3_LOW_ADDR, 0x0fffffff, 0x0fffffff, }; -static const u32 tonga_mgcg_cgcg_init[] = -{ +static const u32 tonga_mgcg_cgcg_init[] = { mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104 }; -static const u32 golden_settings_fiji_a10[] = -{ +static const u32 golden_settings_fiji_a10[] = { mmVM_PRT_APERTURE0_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE1_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE2_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE3_LOW_ADDR, 0x0fffffff, 0x0fffffff, }; -static const u32 fiji_mgcg_cgcg_init[] = -{ +static const u32 fiji_mgcg_cgcg_init[] = { mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104 }; -static const u32 golden_settings_polaris11_a11[] = -{ +static const u32 golden_settings_polaris11_a11[] = { mmVM_PRT_APERTURE0_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE1_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE2_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE3_LOW_ADDR, 0x0fffffff, 0x0fffffff }; -static const u32 golden_settings_polaris10_a11[] = -{ +static const u32 golden_settings_polaris10_a11[] = { mmMC_ARB_WTM_GRPWT_RD, 0x00000003, 0x00000000, mmVM_PRT_APERTURE0_LOW_ADDR, 0x0fffffff, 0x0fffffff, mmVM_PRT_APERTURE1_LOW_ADDR, 0x0fffffff, 0x0fffffff, @@ -110,19 +104,16 @@ static const u32 golden_settings_polaris10_a11[] = mmVM_PRT_APERTURE3_LOW_ADDR, 0x0fffffff, 0x0fffffff }; -static const u32 cz_mgcg_cgcg_init[] = -{ +static const u32 cz_mgcg_cgcg_init[] = { mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104 }; -static const u32 stoney_mgcg_cgcg_init[] = -{ +static const u32 stoney_mgcg_cgcg_init[] = { mmATC_MISC_CG, 0xffffffff, 0x000c0200, mmMC_MEM_POWER_LS, 0xffffffff, 0x00000104 }; -static const u32 golden_settings_stoney_common[] = -{ +static const u32 golden_settings_stoney_common[] = { mmMC_HUB_RDREQ_UVD, MC_HUB_RDREQ_UVD__PRESCALE_MASK, 0x00000004, mmMC_RD_GRP_OTH, MC_RD_GRP_OTH__UVD_MASK, 0x00600000 }; @@ -260,7 +251,8 @@ static int gmc_v8_0_init_microcode(struct amdgpu_device *adev) case CHIP_STONEY: case CHIP_VEGAM: return 0; - default: BUG(); + default: + return -EINVAL; } snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_mc.bin", chip_name); @@ -448,9 +440,9 @@ static void gmc_v8_0_mc_program(struct amdgpu_device *adev) } WREG32(mmHDP_REG_COHERENCY_FLUSH_CNTL, 0); - if (gmc_v8_0_wait_for_idle((void *)adev)) { + if (gmc_v8_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); - } + if (adev->mode_info.num_crtc) { /* Lockout access through VGA aperture*/ tmp = RREG32(mmVGA_HDP_CONTROL); @@ -483,9 +475,8 @@ static void gmc_v8_0_mc_program(struct amdgpu_device *adev) WREG32(mmMC_VM_AGP_BASE, 0); WREG32(mmMC_VM_AGP_TOP, 0x0FFFFFFF); WREG32(mmMC_VM_AGP_BOT, 0x0FFFFFFF); - if (gmc_v8_0_wait_for_idle((void *)adev)) { + if (gmc_v8_0_wait_for_idle((void *)adev)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); - } WREG32(mmBIF_FB_EN, BIF_FB_EN__FB_READ_EN_MASK | BIF_FB_EN__FB_WRITE_EN_MASK); @@ -517,11 +508,11 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev) /* Get VRAM informations */ tmp = RREG32(mmMC_ARB_RAMCFG); - if (REG_GET_FIELD(tmp, MC_ARB_RAMCFG, CHANSIZE)) { + if (REG_GET_FIELD(tmp, MC_ARB_RAMCFG, CHANSIZE)) chansize = 64; - } else { + else chansize = 32; - } + tmp = RREG32(mmMC_SHARED_CHMAP); switch (REG_GET_FIELD(tmp, MC_SHARED_CHMAP, NOOFCHAN)) { case 0: @@ -671,7 +662,7 @@ static void gmc_v8_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, } static uint64_t gmc_v8_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { uint32_t reg; @@ -687,8 +678,8 @@ static uint64_t gmc_v8_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, return pd_addr; } -static void gmc_v8_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid, - unsigned pasid) +static void gmc_v8_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int vmid, + unsigned int pasid) { amdgpu_ring_emit_wreg(ring, mmIH_VMID_0_LUT + vmid, pasid); } @@ -759,11 +750,11 @@ static void gmc_v8_0_set_fault_enable_default(struct amdgpu_device *adev, } /** - * gmc_v8_0_set_prt - set PRT VM fault + * gmc_v8_0_set_prt() - set PRT VM fault * * @adev: amdgpu_device pointer * @enable: enable/disable VM fault handling for PRT -*/ + */ static void gmc_v8_0_set_prt(struct amdgpu_device *adev, bool enable) { u32 tmp; @@ -940,7 +931,7 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev) gmc_v8_0_flush_gpu_tlb(adev, 0, 0, 0); DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n", - (unsigned)(adev->gmc.gart_size >> 20), + (unsigned int)(adev->gmc.gart_size >> 20), (unsigned long long)table_addr); return 0; } @@ -1001,7 +992,7 @@ static void gmc_v8_0_gart_disable(struct amdgpu_device *adev) * Print human readable fault information (VI). */ static void gmc_v8_0_vm_decode_fault(struct amdgpu_device *adev, u32 status, - u32 addr, u32 mc_client, unsigned pasid) + u32 addr, u32 mc_client, unsigned int pasid) { u32 vmid = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, VMID); u32 protections = REG_GET_FIELD(status, VM_CONTEXT1_PROTECTION_FAULT_STATUS, @@ -1056,6 +1047,7 @@ static int gmc_v8_0_early_init(void *handle) adev->gmc.shared_aperture_end + 1; adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + adev->gmc.noretry_flags = AMDGPU_VM_NORETRY_FLAGS_TF; return 0; } @@ -1070,15 +1062,16 @@ static int gmc_v8_0_late_init(void *handle) return 0; } -static unsigned gmc_v8_0_get_vbios_fb_size(struct amdgpu_device *adev) +static unsigned int gmc_v8_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32(mmD1VGA_CONTROL); - unsigned size; + unsigned int size; if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) { size = AMDGPU_VBIOS_VGA_ALLOCATION; } else { u32 viewport = RREG32(mmVIEWPORT_SIZE); + size = (REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_HEIGHT) * REG_GET_FIELD(viewport, VIEWPORT_SIZE, VIEWPORT_WIDTH) * 4); @@ -1282,7 +1275,7 @@ static bool gmc_v8_0_is_idle(void *handle) static int gmc_v8_0_wait_for_idle(void *handle) { - unsigned i; + unsigned int i; u32 tmp; struct amdgpu_device *adev = (struct amdgpu_device *)handle; @@ -1318,13 +1311,15 @@ static bool gmc_v8_0_check_soft_reset(void *handle) srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_MC, 1); } + if (srbm_soft_reset) { adev->gmc.srbm_soft_reset = srbm_soft_reset; return true; - } else { - adev->gmc.srbm_soft_reset = 0; - return false; } + + adev->gmc.srbm_soft_reset = 0; + + return false; } static int gmc_v8_0_pre_soft_reset(void *handle) @@ -1335,9 +1330,8 @@ static int gmc_v8_0_pre_soft_reset(void *handle) return 0; gmc_v8_0_mc_stop(adev); - if (gmc_v8_0_wait_for_idle(adev)) { + if (gmc_v8_0_wait_for_idle(adev)) dev_warn(adev->dev, "Wait for GMC idle timed out !\n"); - } return 0; } @@ -1386,7 +1380,7 @@ static int gmc_v8_0_post_soft_reset(void *handle) static int gmc_v8_0_vm_fault_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { u32 tmp; @@ -1747,8 +1741,7 @@ static void gmc_v8_0_set_irq_funcs(struct amdgpu_device *adev) adev->gmc.vm_fault.funcs = &gmc_v8_0_irq_funcs; } -const struct amdgpu_ip_block_version gmc_v8_0_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v8_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 8, .minor = 0, @@ -1756,8 +1749,7 @@ const struct amdgpu_ip_block_version gmc_v8_0_ip_block = .funcs = &gmc_v8_0_ip_funcs, }; -const struct amdgpu_ip_block_version gmc_v8_1_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v8_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 8, .minor = 1, @@ -1765,8 +1757,7 @@ const struct amdgpu_ip_block_version gmc_v8_1_ip_block = .funcs = &gmc_v8_0_ip_funcs, }; -const struct amdgpu_ip_block_version gmc_v8_5_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v8_5_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 8, .minor = 5, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 67e669e0141c..f9a5a2c0573e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -81,7 +81,7 @@ #define MAX_MEM_RANGES 8 -static const char *gfxhub_client_ids[] = { +static const char * const gfxhub_client_ids[] = { "CB", "DB", "IA", @@ -332,14 +332,12 @@ static const char *mmhub_client_ids_aldebaran[][2] = { [384+0][1] = "OSS", }; -static const struct soc15_reg_golden golden_settings_mmhub_1_0_0[] = -{ +static const struct soc15_reg_golden golden_settings_mmhub_1_0_0[] = { SOC15_REG_GOLDEN_VALUE(MMHUB, 0, mmDAGB1_WRCLI2, 0x00000007, 0xfe5fe0fa), SOC15_REG_GOLDEN_VALUE(MMHUB, 0, mmMMEA1_DRAM_WR_CLI2GRP_MAP0, 0x00000030, 0x55555565) }; -static const struct soc15_reg_golden golden_settings_athub_1_0_0[] = -{ +static const struct soc15_reg_golden golden_settings_athub_1_0_0[] = { SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL, 0x0000ff00, 0x00000800), SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL2, 0x00ff00ff, 0x00080008) }; @@ -416,13 +414,14 @@ static const uint32_t ecc_umc_mcumc_ctrl_mask_addrs[] = { static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { u32 bits, i, tmp, reg; /* Devices newer then VEGA10/12 shall have these programming - sequences performed by PSP BL */ + * sequences performed by PSP BL + */ if (adev->asic_type >= CHIP_VEGA20) return 0; @@ -466,7 +465,7 @@ static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev, static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { struct amdgpu_vmhub *hub; @@ -631,8 +630,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, amdgpu_vm_get_task_info(adev, entry->pasid, &task_info); dev_err(adev->dev, - "[%s] %s page fault (src_id:%u ring:%u vmid:%u " - "pasid:%u, for process %s pid %d thread %s pid %d)\n", + "[%s] %s page fault (src_id:%u ring:%u vmid:%u pasid:%u, for process %s pid %d thread %s pid %d)\n", hub_name, retry_fault ? "retry" : "no-retry", entry->src_id, entry->ring_id, entry->vmid, entry->pasid, task_info.process_name, task_info.tgid, @@ -816,7 +814,7 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, uint32_t vmhub, uint32_t flush_type) { bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(adev, vmhub); - const unsigned eng = 17; + const unsigned int eng = 17; u32 j, inv_req, inv_req2, tmp; struct amdgpu_vmhub *hub; @@ -1033,13 +1031,13 @@ static int gmc_v9_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev, } static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, - unsigned vmid, uint64_t pd_addr) + unsigned int vmid, uint64_t pd_addr) { bool use_semaphore = gmc_v9_0_use_invalidate_semaphore(ring->adev, ring->vm_hub); struct amdgpu_device *adev = ring->adev; struct amdgpu_vmhub *hub = &adev->vmhub[ring->vm_hub]; uint32_t req = gmc_v9_0_get_invalidate_req(vmid, 0); - unsigned eng = ring->vm_inv_eng; + unsigned int eng = ring->vm_inv_eng; /* * It may lose gpuvm invalidate acknowldege state across power-gating @@ -1081,8 +1079,8 @@ static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring, return pd_addr; } -static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid, - unsigned pasid) +static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int vmid, + unsigned int pasid) { struct amdgpu_device *adev = ring->adev; uint32_t reg; @@ -1373,10 +1371,10 @@ static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev, } } -static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) +static unsigned int gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev) { u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL); - unsigned size; + unsigned int size; /* TODO move to DC so GMC doesn't need to hard-code DCN registers */ @@ -1622,6 +1620,7 @@ static int gmc_v9_0_early_init(void *handle) adev->gmc.private_aperture_start = 0x1000000000000000ULL; adev->gmc.private_aperture_end = adev->gmc.private_aperture_start + (4ULL << 30) - 1; + adev->gmc.noretry_flags = AMDGPU_VM_NORETRY_FLAGS_TF; return 0; } @@ -1999,6 +1998,19 @@ static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev) return 0; } +static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev) +{ + static const u32 regBIF_BIOS_SCRATCH_4 = 0x50; + u32 vram_info; + + if (!amdgpu_sriov_vf(adev)) { + vram_info = RREG32(regBIF_BIOS_SCRATCH_4); + adev->gmc.vram_vendor = vram_info & 0xF; + } + adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; + adev->gmc.vram_width = 128 * 64; +} + static int gmc_v9_0_sw_init(void *handle) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits; @@ -2011,15 +2023,12 @@ static int gmc_v9_0_sw_init(void *handle) spin_lock_init(&adev->gmc.invalidate_lock); - if (!(adev->bios) || adev->gmc.is_app_apu) { + if (adev->ip_versions[GC_HWIP][0] == IP_VERSION(9, 4, 3)) { + gmc_v9_4_3_init_vram_info(adev); + } else if (!adev->bios) { if (adev->flags & AMD_IS_APU) { - if (adev->gmc.is_app_apu) { - adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; - adev->gmc.vram_width = 128 * 64; - } else { - adev->gmc.vram_type = AMDGPU_VRAM_TYPE_DDR4; - adev->gmc.vram_width = 64 * 64; - } + adev->gmc.vram_type = AMDGPU_VRAM_TYPE_DDR4; + adev->gmc.vram_width = 64 * 64; } else { adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM; adev->gmc.vram_width = 128 * 64; @@ -2150,7 +2159,7 @@ static int gmc_v9_0_sw_init(void *handle) dma_addr_bits = adev->ip_versions[GC_HWIP][0] >= IP_VERSION(9, 4, 2) ? 48:44; r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(dma_addr_bits)); if (r) { - printk(KERN_WARNING "amdgpu: No suitable DMA available.\n"); + dev_warn(adev->dev, "amdgpu: No suitable DMA available.\n"); return r; } adev->need_swiotlb = drm_need_swiotlb(dma_addr_bits); @@ -2304,7 +2313,7 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) return r; DRM_INFO("PCIE GART of %uM enabled.\n", - (unsigned)(adev->gmc.gart_size >> 20)); + (unsigned int)(adev->gmc.gart_size >> 20)); if (adev->gmc.pdb0_bo) DRM_INFO("PDB0 located at 0x%016llX\n", (unsigned long long)amdgpu_bo_gpu_offset(adev->gmc.pdb0_bo)); @@ -2490,8 +2499,7 @@ const struct amd_ip_funcs gmc_v9_0_ip_funcs = { .get_clockgating_state = gmc_v9_0_get_clockgating_state, }; -const struct amdgpu_ip_block_version gmc_v9_0_ip_block = -{ +const struct amdgpu_ip_block_version gmc_v9_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_GMC, .major = 9, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c index b02e1cef78a7..ec0c8f8b465a 100644 --- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c @@ -494,7 +494,8 @@ static int ih_v6_0_self_irq(struct amdgpu_device *adev, *adev->irq.ih1.wptr_cpu = wptr; schedule_work(&adev->irq.ih1_work); break; - default: break; + default: + break; } return 0; } @@ -535,7 +536,7 @@ static int ih_v6_0_sw_init(void *handle) * use bus address for ih ring by psp bl */ use_bus_addr = (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) ? false : true; - r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr); if (r) return r; @@ -548,7 +549,7 @@ static int ih_v6_0_sw_init(void *handle) /* initialize ih control register offset */ ih_v6_0_init_register_offset(adev); - r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, true); if (r) return r; @@ -759,8 +760,7 @@ static void ih_v6_0_set_interrupt_funcs(struct amdgpu_device *adev) adev->irq.ih_funcs = &ih_v6_0_funcs; } -const struct amdgpu_ip_block_version ih_v6_0_ip_block = -{ +const struct amdgpu_ip_block_version ih_v6_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_IH, .major = 6, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c new file mode 100644 index 000000000000..8fb05eae340a --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c @@ -0,0 +1,769 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include <linux/pci.h> + +#include "amdgpu.h" +#include "amdgpu_ih.h" + +#include "oss/osssys_6_1_0_offset.h" +#include "oss/osssys_6_1_0_sh_mask.h" + +#include "soc15_common.h" +#include "ih_v6_1.h" + +#define MAX_REARM_RETRY 10 + +static void ih_v6_1_set_interrupt_funcs(struct amdgpu_device *adev); + +/** + * ih_v6_1_init_register_offset - Initialize register offset for ih rings + * + * @adev: amdgpu_device pointer + * + * Initialize register offset ih rings (IH_V6_0). + */ +static void ih_v6_1_init_register_offset(struct amdgpu_device *adev) +{ + struct amdgpu_ih_regs *ih_regs; + + /* ih ring 2 is removed + * ih ring and ih ring 1 are available */ + if (adev->irq.ih.ring_size) { + ih_regs = &adev->irq.ih.ih_regs; + ih_regs->ih_rb_base = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_BASE); + ih_regs->ih_rb_base_hi = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_BASE_HI); + ih_regs->ih_rb_cntl = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_CNTL); + ih_regs->ih_rb_wptr = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_WPTR); + ih_regs->ih_rb_rptr = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_RPTR); + ih_regs->ih_doorbell_rptr = SOC15_REG_OFFSET(OSSSYS, 0, regIH_DOORBELL_RPTR); + ih_regs->ih_rb_wptr_addr_lo = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_WPTR_ADDR_LO); + ih_regs->ih_rb_wptr_addr_hi = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_WPTR_ADDR_HI); + ih_regs->psp_reg_id = PSP_REG_IH_RB_CNTL; + } + + if (adev->irq.ih1.ring_size) { + ih_regs = &adev->irq.ih1.ih_regs; + ih_regs->ih_rb_base = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_BASE_RING1); + ih_regs->ih_rb_base_hi = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_BASE_HI_RING1); + ih_regs->ih_rb_cntl = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_CNTL_RING1); + ih_regs->ih_rb_wptr = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_WPTR_RING1); + ih_regs->ih_rb_rptr = SOC15_REG_OFFSET(OSSSYS, 0, regIH_RB_RPTR_RING1); + ih_regs->ih_doorbell_rptr = SOC15_REG_OFFSET(OSSSYS, 0, regIH_DOORBELL_RPTR_RING1); + ih_regs->psp_reg_id = PSP_REG_IH_RB_CNTL_RING1; + } +} + +/** + * force_update_wptr_for_self_int - Force update the wptr for self interrupt + * + * @adev: amdgpu_device pointer + * @threshold: threshold to trigger the wptr reporting + * @timeout: timeout to trigger the wptr reporting + * @enabled: Enable/disable timeout flush mechanism + * + * threshold input range: 0 ~ 15, default 0, + * real_threshold = 2^threshold + * timeout input range: 0 ~ 20, default 8, + * real_timeout = (2^timeout) * 1024 / (socclk_freq) + * + * Force update wptr for self interrupt ( >= SIENNA_CICHLID). + */ +static void +force_update_wptr_for_self_int(struct amdgpu_device *adev, + u32 threshold, u32 timeout, bool enabled) +{ + u32 ih_cntl, ih_rb_cntl; + + ih_cntl = RREG32_SOC15(OSSSYS, 0, regIH_CNTL2); + ih_rb_cntl = RREG32_SOC15(OSSSYS, 0, regIH_RB_CNTL_RING1); + + ih_cntl = REG_SET_FIELD(ih_cntl, IH_CNTL2, + SELF_IV_FORCE_WPTR_UPDATE_TIMEOUT, timeout); + ih_cntl = REG_SET_FIELD(ih_cntl, IH_CNTL2, + SELF_IV_FORCE_WPTR_UPDATE_ENABLE, enabled); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL_RING1, + RB_USED_INT_THRESHOLD, threshold); + + if (amdgpu_sriov_vf(adev) && amdgpu_sriov_reg_indirect_ih(adev)) { + if (psp_reg_program(&adev->psp, PSP_REG_IH_RB_CNTL_RING1, ih_rb_cntl)) + return; + } else { + WREG32_SOC15(OSSSYS, 0, regIH_RB_CNTL_RING1, ih_rb_cntl); + } + + WREG32_SOC15(OSSSYS, 0, regIH_CNTL2, ih_cntl); +} + +/** + * ih_v6_1_toggle_ring_interrupts - toggle the interrupt ring buffer + * + * @adev: amdgpu_device pointer + * @ih: amdgpu_ih_ring pointer + * @enable: true - enable the interrupts, false - disable the interrupts + * + * Toggle the interrupt ring buffer (IH_V6_0) + */ +static int ih_v6_1_toggle_ring_interrupts(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih, + bool enable) +{ + struct amdgpu_ih_regs *ih_regs; + uint32_t tmp; + + ih_regs = &ih->ih_regs; + + tmp = RREG32(ih_regs->ih_rb_cntl); + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, RB_ENABLE, (enable ? 1 : 0)); + /* enable_intr field is only valid in ring0 */ + if (ih == &adev->irq.ih) + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, ENABLE_INTR, (enable ? 1 : 0)); + + if (amdgpu_sriov_vf(adev) && amdgpu_sriov_reg_indirect_ih(adev)) { + if (psp_reg_program(&adev->psp, ih_regs->psp_reg_id, tmp)) + return -ETIMEDOUT; + } else { + WREG32(ih_regs->ih_rb_cntl, tmp); + } + + if (enable) { + ih->enabled = true; + } else { + /* set rptr, wptr to 0 */ + WREG32(ih_regs->ih_rb_rptr, 0); + WREG32(ih_regs->ih_rb_wptr, 0); + ih->enabled = false; + ih->rptr = 0; + } + + return 0; +} + +/** + * ih_v6_1_toggle_interrupts - Toggle all the available interrupt ring buffers + * + * @adev: amdgpu_device pointer + * @enable: enable or disable interrupt ring buffers + * + * Toggle all the available interrupt ring buffers (IH_V6_0). + */ +static int ih_v6_1_toggle_interrupts(struct amdgpu_device *adev, bool enable) +{ + struct amdgpu_ih_ring *ih[] = {&adev->irq.ih, &adev->irq.ih1}; + int i; + int r; + + for (i = 0; i < ARRAY_SIZE(ih); i++) { + if (ih[i]->ring_size) { + r = ih_v6_1_toggle_ring_interrupts(adev, ih[i], enable); + if (r) + return r; + } + } + + return 0; +} + +static uint32_t ih_v6_1_rb_cntl(struct amdgpu_ih_ring *ih, uint32_t ih_rb_cntl) +{ + int rb_bufsz = order_base_2(ih->ring_size / 4); + + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, + MC_SPACE, ih->use_bus_addr ? 2 : 4); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, + WPTR_OVERFLOW_CLEAR, 1); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, + WPTR_OVERFLOW_ENABLE, 1); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, RB_SIZE, rb_bufsz); + /* Ring Buffer write pointer writeback. If enabled, IH_RB_WPTR register + * value is written to memory + */ + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, + WPTR_WRITEBACK_ENABLE, 1); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_SNOOP, 1); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_RO, 0); + ih_rb_cntl = REG_SET_FIELD(ih_rb_cntl, IH_RB_CNTL, MC_VMID, 0); + + return ih_rb_cntl; +} + +static uint32_t ih_v6_1_doorbell_rptr(struct amdgpu_ih_ring *ih) +{ + u32 ih_doorbell_rtpr = 0; + + if (ih->use_doorbell) { + ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, + IH_DOORBELL_RPTR, OFFSET, + ih->doorbell_index); + ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, + IH_DOORBELL_RPTR, + ENABLE, 1); + } else { + ih_doorbell_rtpr = REG_SET_FIELD(ih_doorbell_rtpr, + IH_DOORBELL_RPTR, + ENABLE, 0); + } + return ih_doorbell_rtpr; +} + +/** + * ih_v6_1_enable_ring - enable an ih ring buffer + * + * @adev: amdgpu_device pointer + * @ih: amdgpu_ih_ring pointer + * + * Enable an ih ring buffer (IH_V6_0) + */ +static int ih_v6_1_enable_ring(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih) +{ + struct amdgpu_ih_regs *ih_regs; + uint32_t tmp; + + ih_regs = &ih->ih_regs; + + /* Ring Buffer base. [39:8] of 40-bit address of the beginning of the ring buffer*/ + WREG32(ih_regs->ih_rb_base, ih->gpu_addr >> 8); + WREG32(ih_regs->ih_rb_base_hi, (ih->gpu_addr >> 40) & 0xff); + + tmp = RREG32(ih_regs->ih_rb_cntl); + tmp = ih_v6_1_rb_cntl(ih, tmp); + if (ih == &adev->irq.ih) + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, RPTR_REARM, !!adev->irq.msi_enabled); + if (ih == &adev->irq.ih1) { + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_ENABLE, 0); + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, RB_FULL_DRAIN_ENABLE, 1); + } + + if (amdgpu_sriov_vf(adev) && amdgpu_sriov_reg_indirect_ih(adev)) { + if (psp_reg_program(&adev->psp, ih_regs->psp_reg_id, tmp)) { + DRM_ERROR("PSP program IH_RB_CNTL failed!\n"); + return -ETIMEDOUT; + } + } else { + WREG32(ih_regs->ih_rb_cntl, tmp); + } + + if (ih == &adev->irq.ih) { + /* set the ih ring 0 writeback address whether it's enabled or not */ + WREG32(ih_regs->ih_rb_wptr_addr_lo, lower_32_bits(ih->wptr_addr)); + WREG32(ih_regs->ih_rb_wptr_addr_hi, upper_32_bits(ih->wptr_addr) & 0xFFFF); + } + + /* set rptr, wptr to 0 */ + WREG32(ih_regs->ih_rb_wptr, 0); + WREG32(ih_regs->ih_rb_rptr, 0); + + WREG32(ih_regs->ih_doorbell_rptr, ih_v6_1_doorbell_rptr(ih)); + + return 0; +} + +/** + * ih_v6_1_irq_init - init and enable the interrupt ring + * + * @adev: amdgpu_device pointer + * + * Allocate a ring buffer for the interrupt controller, + * enable the RLC, disable interrupts, enable the IH + * ring buffer and enable it. + * Called at device load and reume. + * Returns 0 for success, errors for failure. + */ +static int ih_v6_1_irq_init(struct amdgpu_device *adev) +{ + struct amdgpu_ih_ring *ih[] = {&adev->irq.ih, &adev->irq.ih1}; + u32 ih_chicken; + u32 tmp; + int ret; + int i; + + /* disable irqs */ + ret = ih_v6_1_toggle_interrupts(adev, false); + if (ret) + return ret; + + adev->nbio.funcs->ih_control(adev); + + if (unlikely((adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) || + (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO))) { + if (ih[0]->use_bus_addr) { + ih_chicken = RREG32_SOC15(OSSSYS, 0, regIH_CHICKEN); + ih_chicken = REG_SET_FIELD(ih_chicken, + IH_CHICKEN, MC_SPACE_GPA_ENABLE, 1); + WREG32_SOC15(OSSSYS, 0, regIH_CHICKEN, ih_chicken); + } + } + + for (i = 0; i < ARRAY_SIZE(ih); i++) { + if (ih[i]->ring_size) { + ret = ih_v6_1_enable_ring(adev, ih[i]); + if (ret) + return ret; + } + } + + /* update doorbell range for ih ring 0 */ + adev->nbio.funcs->ih_doorbell_range(adev, ih[0]->use_doorbell, + ih[0]->doorbell_index); + + tmp = RREG32_SOC15(OSSSYS, 0, regIH_STORM_CLIENT_LIST_CNTL); + tmp = REG_SET_FIELD(tmp, IH_STORM_CLIENT_LIST_CNTL, + CLIENT18_IS_STORM_CLIENT, 1); + WREG32_SOC15(OSSSYS, 0, regIH_STORM_CLIENT_LIST_CNTL, tmp); + + tmp = RREG32_SOC15(OSSSYS, 0, regIH_INT_FLOOD_CNTL); + tmp = REG_SET_FIELD(tmp, IH_INT_FLOOD_CNTL, FLOOD_CNTL_ENABLE, 1); + WREG32_SOC15(OSSSYS, 0, regIH_INT_FLOOD_CNTL, tmp); + + /* GC/MMHUB UTCL2 page fault interrupts are configured as + * MSI storm capable interrupts by deafult. The delay is + * used to avoid ISR being called too frequently + * when page fault happens on several continuous page + * and thus avoid MSI storm */ + tmp = RREG32_SOC15(OSSSYS, 0, regIH_MSI_STORM_CTRL); + tmp = REG_SET_FIELD(tmp, IH_MSI_STORM_CTRL, + DELAY, 3); + WREG32_SOC15(OSSSYS, 0, regIH_MSI_STORM_CTRL, tmp); + + pci_set_master(adev->pdev); + + /* enable interrupts */ + ret = ih_v6_1_toggle_interrupts(adev, true); + if (ret) + return ret; + /* enable wptr force update for self int */ + force_update_wptr_for_self_int(adev, 0, 8, true); + + if (adev->irq.ih_soft.ring_size) + adev->irq.ih_soft.enabled = true; + + return 0; +} + +/** + * ih_v6_1_irq_disable - disable interrupts + * + * @adev: amdgpu_device pointer + * + * Disable interrupts on the hw. + */ +static void ih_v6_1_irq_disable(struct amdgpu_device *adev) +{ + force_update_wptr_for_self_int(adev, 0, 8, false); + ih_v6_1_toggle_interrupts(adev, false); + + /* Wait and acknowledge irq */ + mdelay(1); +} + +/** + * ih_v6_1_get_wptr - get the IH ring buffer wptr + * + * @adev: amdgpu_device pointer + * @ih: amdgpu_ih_ring pointer + * + * Get the IH ring buffer wptr from either the register + * or the writeback memory buffer. Also check for + * ring buffer overflow and deal with it. + * Returns the value of the wptr. + */ +static u32 ih_v6_1_get_wptr(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih) +{ + u32 wptr, tmp; + struct amdgpu_ih_regs *ih_regs; + + wptr = le32_to_cpu(*ih->wptr_cpu); + ih_regs = &ih->ih_regs; + + if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) + goto out; + + wptr = RREG32_NO_KIQ(ih_regs->ih_rb_wptr); + if (!REG_GET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW)) + goto out; + wptr = REG_SET_FIELD(wptr, IH_RB_WPTR, RB_OVERFLOW, 0); + + /* When a ring buffer overflow happen start parsing interrupt + * from the last not overwritten vector (wptr + 32). Hopefully + * this should allow us to catch up. + */ + tmp = (wptr + 32) & ih->ptr_mask; + dev_warn(adev->dev, "IH ring buffer overflow " + "(0x%08X, 0x%08X, 0x%08X)\n", + wptr, ih->rptr, tmp); + ih->rptr = tmp; + + tmp = RREG32_NO_KIQ(ih_regs->ih_rb_cntl); + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1); + WREG32_NO_KIQ(ih_regs->ih_rb_cntl, tmp); +out: + return (wptr & ih->ptr_mask); +} + +/** + * ih_v6_1_irq_rearm - rearm IRQ if lost + * + * @adev: amdgpu_device pointer + * @ih: amdgpu_ih_ring pointer + * + */ +static void ih_v6_1_irq_rearm(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih) +{ + uint32_t v = 0; + uint32_t i = 0; + struct amdgpu_ih_regs *ih_regs; + + ih_regs = &ih->ih_regs; + + /* Rearm IRQ / re-write doorbell if doorbell write is lost */ + for (i = 0; i < MAX_REARM_RETRY; i++) { + v = RREG32_NO_KIQ(ih_regs->ih_rb_rptr); + if ((v < ih->ring_size) && (v != ih->rptr)) + WDOORBELL32(ih->doorbell_index, ih->rptr); + else + break; + } +} + +/** + * ih_v6_1_set_rptr - set the IH ring buffer rptr + * + * @adev: amdgpu_device pointer + * @ih: amdgpu_ih_ring pointer + * + * Set the IH ring buffer rptr. + */ +static void ih_v6_1_set_rptr(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih) +{ + struct amdgpu_ih_regs *ih_regs; + + if (ih->use_doorbell) { + /* XXX check if swapping is necessary on BE */ + *ih->rptr_cpu = ih->rptr; + WDOORBELL32(ih->doorbell_index, ih->rptr); + + if (amdgpu_sriov_vf(adev)) + ih_v6_1_irq_rearm(adev, ih); + } else { + ih_regs = &ih->ih_regs; + WREG32(ih_regs->ih_rb_rptr, ih->rptr); + } +} + +/** + * ih_v6_1_self_irq - dispatch work for ring 1 + * + * @adev: amdgpu_device pointer + * @source: irq source + * @entry: IV with WPTR update + * + * Update the WPTR from the IV and schedule work to handle the entries. + */ +static int ih_v6_1_self_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + uint32_t wptr = cpu_to_le32(entry->src_data[0]); + + switch (entry->ring_id) { + case 1: + *adev->irq.ih1.wptr_cpu = wptr; + schedule_work(&adev->irq.ih1_work); + break; + default: + break; + } + return 0; +} + +static const struct amdgpu_irq_src_funcs ih_v6_1_self_irq_funcs = { + .process = ih_v6_1_self_irq, +}; + +static void ih_v6_1_set_self_irq_funcs(struct amdgpu_device *adev) +{ + adev->irq.self_irq.num_types = 0; + adev->irq.self_irq.funcs = &ih_v6_1_self_irq_funcs; +} + +static int ih_v6_1_early_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + ih_v6_1_set_interrupt_funcs(adev); + ih_v6_1_set_self_irq_funcs(adev); + return 0; +} + +static int ih_v6_1_sw_init(void *handle) +{ + int r; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool use_bus_addr; + + r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_IH, 0, + &adev->irq.self_irq); + + if (r) + return r; + + /* use gpu virtual address for ih ring + * until ih_checken is programmed to allow + * use bus address for ih ring by psp bl */ + use_bus_addr = + (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) ? false : true; + r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr); + if (r) + return r; + + adev->irq.ih.use_doorbell = true; + adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1; + + adev->irq.ih1.ring_size = 0; + adev->irq.ih2.ring_size = 0; + + /* initialize ih control register offset */ + ih_v6_1_init_register_offset(adev); + + r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true); + if (r) + return r; + + r = amdgpu_irq_init(adev); + + return r; +} + +static int ih_v6_1_sw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + amdgpu_irq_fini_sw(adev); + + return 0; +} + +static int ih_v6_1_hw_init(void *handle) +{ + int r; + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + r = ih_v6_1_irq_init(adev); + if (r) + return r; + + return 0; +} + +static int ih_v6_1_hw_fini(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + ih_v6_1_irq_disable(adev); + + return 0; +} + +static int ih_v6_1_suspend(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + return ih_v6_1_hw_fini(adev); +} + +static int ih_v6_1_resume(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + return ih_v6_1_hw_init(adev); +} + +static bool ih_v6_1_is_idle(void *handle) +{ + /* todo */ + return true; +} + +static int ih_v6_1_wait_for_idle(void *handle) +{ + /* todo */ + return -ETIMEDOUT; +} + +static int ih_v6_1_soft_reset(void *handle) +{ + /* todo */ + return 0; +} + +static void ih_v6_1_update_clockgating_state(struct amdgpu_device *adev, + bool enable) +{ + uint32_t data, def, field_val; + + if (adev->cg_flags & AMD_CG_SUPPORT_IH_CG) { + def = data = RREG32_SOC15(OSSSYS, 0, regIH_CLK_CTRL); + field_val = enable ? 0 : 1; + data = REG_SET_FIELD(data, IH_CLK_CTRL, + DBUS_MUX_CLK_SOFT_OVERRIDE, field_val); + data = REG_SET_FIELD(data, IH_CLK_CTRL, + OSSSYS_SHARE_CLK_SOFT_OVERRIDE, field_val); + data = REG_SET_FIELD(data, IH_CLK_CTRL, + LIMIT_SMN_CLK_SOFT_OVERRIDE, field_val); + data = REG_SET_FIELD(data, IH_CLK_CTRL, + DYN_CLK_SOFT_OVERRIDE, field_val); + data = REG_SET_FIELD(data, IH_CLK_CTRL, + REG_CLK_SOFT_OVERRIDE, field_val); + if (def != data) + WREG32_SOC15(OSSSYS, 0, regIH_CLK_CTRL, data); + } + + return; +} + +static int ih_v6_1_set_clockgating_state(void *handle, + enum amd_clockgating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + ih_v6_1_update_clockgating_state(adev, + state == AMD_CG_STATE_GATE); + return 0; +} + +static void ih_v6_1_update_ih_mem_power_gating(struct amdgpu_device *adev, + bool enable) +{ + uint32_t ih_mem_pwr_cntl; + + /* Disable ih sram power cntl before switch powergating mode */ + ih_mem_pwr_cntl = RREG32_SOC15(OSSSYS, 0, regIH_MEM_POWER_CTRL); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_CTRL_EN, 0); + WREG32_SOC15(OSSSYS, 0, regIH_MEM_POWER_CTRL, ih_mem_pwr_cntl); + + /* It is recommended to set mem powergating mode to DS mode */ + if (enable) { + /* mem power mode */ + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_LS_EN, 0); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_DS_EN, 1); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_SD_EN, 0); + /* cam mem power mode */ + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_RETRY_INT_CAM_MEM_POWER_LS_EN, 0); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_RETRY_INT_CAM_MEM_POWER_DS_EN, 1); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_RETRY_INT_CAM_MEM_POWER_SD_EN, 0); + /* re-enable power cntl */ + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_CTRL_EN, 1); + } else { + /* mem power mode */ + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_LS_EN, 0); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_DS_EN, 0); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_SD_EN, 0); + /* cam mem power mode */ + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_RETRY_INT_CAM_MEM_POWER_LS_EN, 0); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_RETRY_INT_CAM_MEM_POWER_DS_EN, 0); + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_RETRY_INT_CAM_MEM_POWER_SD_EN, 0); + /* re-enable power cntl*/ + ih_mem_pwr_cntl = REG_SET_FIELD(ih_mem_pwr_cntl, IH_MEM_POWER_CTRL, + IH_BUFFER_MEM_POWER_CTRL_EN, 1); + } + + WREG32_SOC15(OSSSYS, 0, regIH_MEM_POWER_CTRL, ih_mem_pwr_cntl); +} + +static int ih_v6_1_set_powergating_state(void *handle, + enum amd_powergating_state state) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + bool enable = (state == AMD_PG_STATE_GATE); + + if (adev->pg_flags & AMD_PG_SUPPORT_IH_SRAM_PG) + ih_v6_1_update_ih_mem_power_gating(adev, enable); + + return 0; +} + +static void ih_v6_1_get_clockgating_state(void *handle, u64 *flags) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + if (!RREG32_SOC15(OSSSYS, 0, regIH_CLK_CTRL)) + *flags |= AMD_CG_SUPPORT_IH_CG; + + return; +} + +static const struct amd_ip_funcs ih_v6_1_ip_funcs = { + .name = "ih_v6_1", + .early_init = ih_v6_1_early_init, + .late_init = NULL, + .sw_init = ih_v6_1_sw_init, + .sw_fini = ih_v6_1_sw_fini, + .hw_init = ih_v6_1_hw_init, + .hw_fini = ih_v6_1_hw_fini, + .suspend = ih_v6_1_suspend, + .resume = ih_v6_1_resume, + .is_idle = ih_v6_1_is_idle, + .wait_for_idle = ih_v6_1_wait_for_idle, + .soft_reset = ih_v6_1_soft_reset, + .set_clockgating_state = ih_v6_1_set_clockgating_state, + .set_powergating_state = ih_v6_1_set_powergating_state, + .get_clockgating_state = ih_v6_1_get_clockgating_state, +}; + +static const struct amdgpu_ih_funcs ih_v6_1_funcs = { + .get_wptr = ih_v6_1_get_wptr, + .decode_iv = amdgpu_ih_decode_iv_helper, + .decode_iv_ts = amdgpu_ih_decode_iv_ts_helper, + .set_rptr = ih_v6_1_set_rptr +}; + +static void ih_v6_1_set_interrupt_funcs(struct amdgpu_device *adev) +{ + adev->irq.ih_funcs = &ih_v6_1_funcs; +} + +const struct amdgpu_ip_block_version ih_v6_1_ip_block = { + .type = AMD_IP_BLOCK_TYPE_IH, + .major = 6, + .minor = 0, + .rev = 0, + .funcs = &ih_v6_1_ip_funcs, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_1.h b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.h new file mode 100644 index 000000000000..2232bc5cbd09 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.h @@ -0,0 +1,28 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef __IH_V6_1_IH_H__ +#define __IH_V6_1_IH_H__ + +extern const struct amdgpu_ip_block_version ih_v6_1_ip_block; + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index c25d4a07350b..1c8116d75f63 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -807,8 +807,7 @@ static void jpeg_v2_0_set_irq_funcs(struct amdgpu_device *adev) adev->jpeg.inst->irq.funcs = &jpeg_v2_0_irq_funcs; } -const struct amdgpu_ip_block_version jpeg_v2_0_ip_block = -{ +const struct amdgpu_ip_block_version jpeg_v2_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_JPEG, .major = 2, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index 79791379fc2b..df4440c21bbf 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -479,7 +479,7 @@ static int jpeg_v3_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = state == AMD_CG_STATE_GATE; if (enable) { if (!jpeg_v3_0_is_idle(handle)) diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index a707d407fbd0..3eb3dcd56b57 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -626,7 +626,7 @@ static int jpeg_v4_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = state == AMD_CG_STATE_GATE; if (enable) { if (!jpeg_v4_0_is_idle(handle)) diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c index ce2b22f7e4e4..1de79d660285 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -26,6 +26,7 @@ #include "soc15.h" #include "soc15d.h" #include "jpeg_v4_0_3.h" +#include "mmsch_v4_0_3.h" #include "vcn/vcn_4_0_3_offset.h" #include "vcn/vcn_4_0_3_sh_mask.h" @@ -41,6 +42,7 @@ static void jpeg_v4_0_3_set_irq_funcs(struct amdgpu_device *adev); static int jpeg_v4_0_3_set_powergating_state(void *handle, enum amd_powergating_state state); static void jpeg_v4_0_3_set_ras_funcs(struct amdgpu_device *adev); +static void jpeg_v4_0_3_dec_ring_set_wptr(struct amdgpu_ring *ring); static int amdgpu_ih_srcid_jpeg[] = { VCN_4_0__SRCID__JPEG_DECODE, @@ -109,9 +111,20 @@ static int jpeg_v4_0_3_sw_init(void *handle) ring = &adev->jpeg.inst[i].ring_dec[j]; ring->use_doorbell = true; ring->vm_hub = AMDGPU_MMHUB0(adev->jpeg.inst[i].aid_id); - ring->doorbell_index = - (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + - 1 + j + 9 * jpeg_inst; + if (!amdgpu_sriov_vf(adev)) { + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 1 + j + 9 * jpeg_inst; + } else { + if (j < 4) + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 4 + j + 32 * jpeg_inst; + else + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 8 + j + 32 * jpeg_inst; + } sprintf(ring->name, "jpeg_dec_%d.%d", adev->jpeg.inst[i].aid_id, j); r = amdgpu_ring_init(adev, ring, 512, &adev->jpeg.inst->irq, 0, AMDGPU_RING_PRIO_DEFAULT, NULL); @@ -160,6 +173,119 @@ static int jpeg_v4_0_3_sw_fini(void *handle) return r; } +static int jpeg_v4_0_3_start_sriov(struct amdgpu_device *adev) +{ + struct amdgpu_ring *ring; + uint64_t ctx_addr; + uint32_t param, resp, expected; + uint32_t tmp, timeout; + + struct amdgpu_mm_table *table = &adev->virt.mm_table; + uint32_t *table_loc; + uint32_t table_size; + uint32_t size, size_dw, item_offset; + uint32_t init_status; + int i, j, jpeg_inst; + + struct mmsch_v4_0_cmd_direct_write + direct_wt = { {0} }; + struct mmsch_v4_0_cmd_end end = { {0} }; + struct mmsch_v4_0_3_init_header header; + + direct_wt.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_WRITE; + end.cmd_header.command_type = + MMSCH_COMMAND__END; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; i++) { + jpeg_inst = GET_INST(JPEG, i); + + memset(&header, 0, sizeof(struct mmsch_v4_0_3_init_header)); + header.version = MMSCH_VERSION; + header.total_size = sizeof(struct mmsch_v4_0_3_init_header) >> 2; + + table_loc = (uint32_t *)table->cpu_addr; + table_loc += header.total_size; + + item_offset = header.total_size; + + for (j = 0; j < adev->jpeg.num_jpeg_rings; j++) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + table_size = 0; + + tmp = SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_LOW); + MMSCH_V4_0_INSERT_DIRECT_WT(tmp, lower_32_bits(ring->gpu_addr)); + tmp = SOC15_REG_OFFSET(JPEG, 0, regUVD_JMI0_UVD_LMI_JRBC_RB_64BIT_BAR_HIGH); + MMSCH_V4_0_INSERT_DIRECT_WT(tmp, upper_32_bits(ring->gpu_addr)); + tmp = SOC15_REG_OFFSET(JPEG, 0, regUVD_JRBC0_UVD_JRBC_RB_SIZE); + MMSCH_V4_0_INSERT_DIRECT_WT(tmp, ring->ring_size / 4); + + if (j <= 3) { + header.mjpegdec0[j].table_offset = item_offset; + header.mjpegdec0[j].init_status = 0; + header.mjpegdec0[j].table_size = table_size; + } else { + header.mjpegdec1[j - 4].table_offset = item_offset; + header.mjpegdec1[j - 4].init_status = 0; + header.mjpegdec1[j - 4].table_size = table_size; + } + header.total_size += table_size; + item_offset += table_size; + } + + MMSCH_V4_0_INSERT_END(); + + /* send init table to MMSCH */ + size = sizeof(struct mmsch_v4_0_3_init_header); + table_loc = (uint32_t *)table->cpu_addr; + memcpy((void *)table_loc, &header, size); + + ctx_addr = table->gpu_addr; + WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_CTX_ADDR_LO, lower_32_bits(ctx_addr)); + WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_CTX_ADDR_HI, upper_32_bits(ctx_addr)); + + tmp = RREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_VMID); + tmp &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK; + tmp |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT); + WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_VMID, tmp); + + size = header.total_size; + WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_CTX_SIZE, size); + + WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_MAILBOX_RESP, 0); + + param = 0x00000001; + WREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_MAILBOX_HOST, param); + tmp = 0; + timeout = 1000; + resp = 0; + expected = MMSCH_VF_MAILBOX_RESP__OK; + init_status = + ((struct mmsch_v4_0_3_init_header *)(table_loc))->mjpegdec0[i].init_status; + while (resp != expected) { + resp = RREG32_SOC15(VCN, jpeg_inst, regMMSCH_VF_MAILBOX_RESP); + + if (resp != 0) + break; + udelay(10); + tmp = tmp + 10; + if (tmp >= timeout) { + DRM_ERROR("failed to init MMSCH. TIME-OUT after %d usec"\ + " waiting for regMMSCH_VF_MAILBOX_RESP "\ + "(expected=0x%08x, readback=0x%08x)\n", + tmp, expected, resp); + return -EBUSY; + } + } + if (resp != expected && resp != MMSCH_VF_MAILBOX_RESP__INCOMPLETE && + init_status != MMSCH_VF_ENGINE_STATUS__PASS) + DRM_ERROR("MMSCH init status is incorrect! readback=0x%08x, header init status for jpeg: %x\n", + resp, init_status); + + } + return 0; +} + /** * jpeg_v4_0_3_hw_init - start and test JPEG block * @@ -172,31 +298,47 @@ static int jpeg_v4_0_3_hw_init(void *handle) struct amdgpu_ring *ring; int i, j, r, jpeg_inst; - for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { - jpeg_inst = GET_INST(JPEG, i); + if (amdgpu_sriov_vf(adev)) { + r = jpeg_v4_0_3_start_sriov(adev); + if (r) + return r; - ring = adev->jpeg.inst[i].ring_dec; + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + ring->wptr = 0; + ring->wptr_old = 0; + jpeg_v4_0_3_dec_ring_set_wptr(ring); + ring->sched.ready = true; + } + } + } else { + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + jpeg_inst = GET_INST(JPEG, i); - if (ring->use_doorbell) - adev->nbio.funcs->vcn_doorbell_range( - adev, ring->use_doorbell, - (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + - 9 * jpeg_inst, - adev->jpeg.inst[i].aid_id); + ring = adev->jpeg.inst[i].ring_dec; - for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { - ring = &adev->jpeg.inst[i].ring_dec[j]; if (ring->use_doorbell) - WREG32_SOC15_OFFSET( - VCN, GET_INST(VCN, i), - regVCN_JPEG_DB_CTRL, - (ring->pipe ? (ring->pipe - 0x15) : 0), - ring->doorbell_index + adev->nbio.funcs->vcn_doorbell_range( + adev, ring->use_doorbell, + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 9 * jpeg_inst, + adev->jpeg.inst[i].aid_id); + + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + if (ring->use_doorbell) + WREG32_SOC15_OFFSET( + VCN, GET_INST(VCN, i), + regVCN_JPEG_DB_CTRL, + (ring->pipe ? (ring->pipe - 0x15) : 0), + ring->doorbell_index << VCN_JPEG_DB_CTRL__OFFSET__SHIFT | - VCN_JPEG_DB_CTRL__EN_MASK); - r = amdgpu_ring_test_helper(ring); - if (r) - return r; + VCN_JPEG_DB_CTRL__EN_MASK); + r = amdgpu_ring_test_helper(ring); + if (r) + return r; + } } } DRM_DEV_INFO(adev->dev, "JPEG decode initialized successfully.\n"); @@ -218,8 +360,10 @@ static int jpeg_v4_0_3_hw_fini(void *handle) cancel_delayed_work_sync(&adev->jpeg.idle_work); - if (adev->jpeg.cur_state != AMD_PG_STATE_GATE) - ret = jpeg_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE); + if (!amdgpu_sriov_vf(adev)) { + if (adev->jpeg.cur_state != AMD_PG_STATE_GATE) + ret = jpeg_v4_0_3_set_powergating_state(adev, AMD_PG_STATE_GATE); + } return ret; } @@ -785,7 +929,7 @@ static int jpeg_v4_0_3_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = state == AMD_CG_STATE_GATE; int i; for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c index 36a123e6c8ee..eb06d749876f 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.c @@ -909,10 +909,12 @@ static int mes_v10_1_mqd_sw_init(struct amdgpu_device *adev, /* prepare MQD backup */ adev->mes.mqd_backup[pipe] = kmalloc(mqd_size, GFP_KERNEL); - if (!adev->mes.mqd_backup[pipe]) + if (!adev->mes.mqd_backup[pipe]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); + return -ENOMEM; + } return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 1bdaa00c0b46..6827d547042e 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -210,9 +210,7 @@ static int mes_v11_0_add_hw_queue(struct amdgpu_mes *mes, mes_add_queue_pkt.is_aql_queue = input->is_aql_queue; mes_add_queue_pkt.gds_size = input->queue_size; - /* For KFD, gds_size is re-used for queue size (needed in MES for AQL queues) */ - mes_add_queue_pkt.is_aql_queue = input->is_aql_queue; - mes_add_queue_pkt.gds_size = input->queue_size; + mes_add_queue_pkt.exclusively_scheduled = input->exclusively_scheduled; return mes_v11_0_submit_pkt_and_poll_completion(mes, &mes_add_queue_pkt, sizeof(mes_add_queue_pkt), @@ -790,8 +788,7 @@ static int mes_v11_0_mqd_init(struct amdgpu_ring *ring) DOORBELL_SOURCE, 0); tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_HIT, 0); - } - else + } else tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_DOORBELL_CONTROL, DOORBELL_EN, 0); mqd->cp_hqd_pq_doorbell_control = tmp; @@ -1019,10 +1016,12 @@ static int mes_v11_0_mqd_sw_init(struct amdgpu_device *adev, /* prepare MQD backup */ adev->mes.mqd_backup[pipe] = kmalloc(mqd_size, GFP_KERNEL); - if (!adev->mes.mqd_backup[pipe]) + if (!adev->mes.mqd_backup[pipe]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); + return -ENOMEM; + } return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c index 5e8b493f8699..784c4e077470 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c @@ -413,18 +413,6 @@ static void mmhub_v1_8_program_invalidation(struct amdgpu_device *adev) static int mmhub_v1_8_gart_enable(struct amdgpu_device *adev) { - if (amdgpu_sriov_vf(adev)) { - /* - * MC_VM_FB_LOCATION_BASE/TOP is NULL for VF, becuase they are - * VF copy registers so vbios post doesn't program them, for - * SRIOV driver need to program them - */ - WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_BASE, - adev->gmc.vram_start >> 24); - WREG32_SOC15(MMHUB, 0, regMC_VM_FB_LOCATION_TOP, - adev->gmc.vram_end >> 24); - } - /* GART Enable. */ mmhub_v1_8_init_gart_aperture_regs(adev); mmhub_v1_8_init_system_aperture_regs(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c index 8bd0fc8d9d25..1dce053a4c4d 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c @@ -331,7 +331,7 @@ static void mmhub_v2_3_setup_vmid_config(struct amdgpu_device *adev) static void mmhub_v2_3_program_invalidation(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; - unsigned i; + unsigned int i; for (i = 0; i < 18; ++i) { WREG32_SOC15_OFFSET(MMHUB, 0, @@ -406,6 +406,7 @@ static void mmhub_v2_3_set_fault_enable_default(struct amdgpu_device *adev, bool value) { u32 tmp; + tmp = RREG32_SOC15(MMHUB, 0, mmMMVM_L2_PROTECTION_FAULT_CNTL); tmp = REG_SET_FIELD(tmp, MMVM_L2_PROTECTION_FAULT_CNTL, RANGE_PROTECTION_FAULT_ENABLE_DEFAULT, value); @@ -499,11 +500,11 @@ mmhub_v2_3_update_medium_grain_clock_gating(struct amdgpu_device *adev, if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG)) { data &= ~MM_ATC_L2_CGTT_CLK_CTRL__SOFT_OVERRIDE_MASK; data1 &= ~(DAGB0_CNTL_MISC2__DISABLE_WRREQ_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK); + DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK); } else { data |= MM_ATC_L2_CGTT_CLK_CTRL__SOFT_OVERRIDE_MASK; @@ -593,13 +594,13 @@ static void mmhub_v2_3_get_clockgating(struct amdgpu_device *adev, u64 *flags) /* AMD_CG_SUPPORT_MC_MGCG */ if (!(data & (DAGB0_CNTL_MISC2__DISABLE_WRREQ_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | - DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK)) + DAGB0_CNTL_MISC2__DISABLE_WRRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDREQ_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_RDRET_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBWR_CG_MASK | + DAGB0_CNTL_MISC2__DISABLE_TLBRD_CG_MASK)) && !(data1 & MM_ATC_L2_CGTT_CLK_CTRL__SOFT_OVERRIDE_MASK)) { - *flags |= AMD_CG_SUPPORT_MC_MGCG; + *flags |= AMD_CG_SUPPORT_MC_MGCG; } /* AMD_CG_SUPPORT_MC_LS */ diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index e790f890aec6..5718e4d40e66 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -108,7 +108,7 @@ static void mmhub_v9_4_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmi } static void mmhub_v9_4_init_system_aperture_regs(struct amdgpu_device *adev, - int hubid) + int hubid) { uint64_t value; uint32_t tmp; @@ -1568,7 +1568,7 @@ static int mmhub_v9_4_get_ras_error_count(struct amdgpu_device *adev, uint32_t sec_cnt, ded_cnt; for (i = 0; i < ARRAY_SIZE(mmhub_v9_4_ras_fields); i++) { - if(mmhub_v9_4_ras_fields[i].reg_offset != reg->reg_offset) + if (mmhub_v9_4_ras_fields[i].reg_offset != reg->reg_offset) continue; sec_cnt = (value & diff --git a/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0_3.h b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0_3.h new file mode 100644 index 000000000000..db7eb5260295 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/mmsch_v4_0_3.h @@ -0,0 +1,37 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef __MMSCH_V4_0_3_H__ +#define __MMSCH_V4_0_3_H__ + +#include "amdgpu_vcn.h" +#include "mmsch_v4_0.h" + +struct mmsch_v4_0_3_init_header { + uint32_t version; + uint32_t total_size; + struct mmsch_v4_0_table_info vcn0; + struct mmsch_v4_0_table_info mjpegdec0[4]; + struct mmsch_v4_0_table_info mjpegdec1[4]; +}; +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index cae1aaa4ddb6..6a68ee946f1c 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -183,12 +183,10 @@ send_request: if (req != IDH_REQ_GPU_INIT_DATA) { pr_err("Doesn't get msg:%d from pf, error=%d\n", event, r); return r; - } - else /* host doesn't support REQ_GPU_INIT_DATA handshake */ + } else /* host doesn't support REQ_GPU_INIT_DATA handshake */ adev->virt.req_init_data_ver = 0; } else { - if (req == IDH_REQ_GPU_INIT_DATA) - { + if (req == IDH_REQ_GPU_INIT_DATA) { adev->virt.req_init_data_ver = RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW1); diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c index 288c414babdf..59f53c743362 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c @@ -334,7 +334,7 @@ static void xgpu_vi_mailbox_send_ack(struct amdgpu_device *adev) break; } mdelay(1); - timeout -=1; + timeout -= 1; reg = RREG32_NO_KIQ(mmMAILBOX_CONTROL); } diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index eec13cb5bf75..b6a8478dabf4 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -565,7 +565,7 @@ static int navi10_ih_sw_init(void *handle) use_bus_addr = false; else use_bus_addr = true; - r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr); if (r) return r; @@ -578,7 +578,7 @@ static int navi10_ih_sw_init(void *handle) /* initialize ih control registers offset */ navi10_ih_init_register_offset(adev); - r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, true); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.c b/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.c index d5ed9e0e1a5f..e5b5b0f4940f 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v4_3.c @@ -345,6 +345,9 @@ static void nbio_v4_3_init_registers(struct amdgpu_device *adev) data &= ~RCC_DEV0_EPF2_STRAP2__STRAP_NO_SOFT_RESET_DEV0_F2_MASK; WREG32_SOC15(NBIO, 0, regRCC_DEV0_EPF2_STRAP2, data); } + if (amdgpu_sriov_vf(adev)) + adev->rmmio_remap.reg_offset = SOC15_REG_OFFSET(NBIO, 0, + regBIF_BX_DEV0_EPF0_VF0_HDP_MEM_COHERENCY_FLUSH_CNTL) << 2; } static u32 nbio_v4_3_get_rom_offset(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index cd1a02d30420..f85eec05d218 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -32,6 +32,18 @@ #define NPS_MODE_MASK 0x000000FFL +/* Core 0 Port 0 counter */ +#define smnPCIEP_NAK_COUNTER 0x1A340218 + +#define smnPCIE_PERF_CNTL_TXCLK3 0x1A38021c +#define smnPCIE_PERF_CNTL_TXCLK7 0x1A380888 +#define smnPCIE_PERF_COUNT_CNTL 0x1A380200 +#define smnPCIE_PERF_COUNT0_TXCLK3 0x1A380220 +#define smnPCIE_PERF_COUNT0_TXCLK7 0x1A38088C +#define smnPCIE_PERF_COUNT0_UPVAL_TXCLK3 0x1A3808F8 +#define smnPCIE_PERF_COUNT0_UPVAL_TXCLK7 0x1A380918 + + static void nbio_v7_9_remap_hdp_registers(struct amdgpu_device *adev) { WREG32_SOC15(NBIO, 0, regBIF_BX0_REMAP_HDP_MEM_FLUSH_CNTL, @@ -425,6 +437,93 @@ static void nbio_v7_9_init_registers(struct amdgpu_device *adev) XCC_DOORBELL_FENCE__SHUB_SLV_MODE_MASK); } + + if (!amdgpu_sriov_vf(adev)) { + u32 baco_cntl; + for_each_inst(i, adev->aid_mask) { + baco_cntl = RREG32_SOC15(NBIO, i, regBIF_BX0_BACO_CNTL); + if (baco_cntl & (BIF_BX0_BACO_CNTL__BACO_DUMMY_EN_MASK | + BIF_BX0_BACO_CNTL__BACO_EN_MASK)) { + baco_cntl &= ~( + BIF_BX0_BACO_CNTL__BACO_DUMMY_EN_MASK | + BIF_BX0_BACO_CNTL__BACO_EN_MASK); + dev_dbg(adev->dev, + "Unsetting baco dummy mode %x", + baco_cntl); + WREG32_SOC15(NBIO, i, regBIF_BX0_BACO_CNTL, + baco_cntl); + } + } + } +} + +static u64 nbio_v7_9_get_pcie_replay_count(struct amdgpu_device *adev) +{ + u32 val, nak_r, nak_g; + + if (adev->flags & AMD_IS_APU) + return 0; + + /* Get the number of NAKs received and generated */ + val = RREG32_PCIE(smnPCIEP_NAK_COUNTER); + nak_r = val & 0xFFFF; + nak_g = val >> 16; + + /* Add the total number of NAKs, i.e the number of replays */ + return (nak_r + nak_g); +} + +static void nbio_v7_9_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0, + uint64_t *count1) +{ + uint32_t perfctrrx = 0; + uint32_t perfctrtx = 0; + + /* This reports 0 on APUs, so return to avoid writing/reading registers + * that may or may not be different from their GPU counterparts + */ + if (adev->flags & AMD_IS_APU) + return; + + /* Use TXCLK3 counter group for rx event */ + /* Use TXCLK7 counter group for tx event */ + /* Set the 2 events that we wish to watch, defined above */ + /* 40 is event# for received msgs */ + /* 2 is event# of posted requests sent */ + perfctrrx = REG_SET_FIELD(perfctrrx, PCIE_PERF_CNTL_TXCLK3, EVENT0_SEL, 40); + perfctrtx = REG_SET_FIELD(perfctrtx, PCIE_PERF_CNTL_TXCLK7, EVENT0_SEL, 2); + + /* Write to enable desired perf counters */ + WREG32_PCIE(smnPCIE_PERF_CNTL_TXCLK3, perfctrrx); + WREG32_PCIE(smnPCIE_PERF_CNTL_TXCLK7, perfctrtx); + + /* Zero out and enable SHADOW_WR + * Write 0x6: + * Bit 1 = Global Shadow wr(1) + * Bit 2 = Global counter reset enable(1) + */ + WREG32_PCIE(smnPCIE_PERF_COUNT_CNTL, 0x00000006); + + /* Enable Gloabl Counter + * Write 0x1: + * Bit 0 = Global Counter Enable(1) + */ + WREG32_PCIE(smnPCIE_PERF_COUNT_CNTL, 0x00000001); + + msleep(1000); + + /* Disable Global Counter, Reset and enable SHADOW_WR + * Write 0x6: + * Bit 1 = Global Shadow wr(1) + * Bit 2 = Global counter reset enable(1) + */ + WREG32_PCIE(smnPCIE_PERF_COUNT_CNTL, 0x00000006); + + /* Get the upper and lower count */ + *count0 = RREG32_PCIE(smnPCIE_PERF_COUNT0_TXCLK3) | + ((uint64_t)RREG32_PCIE(smnPCIE_PERF_COUNT0_UPVAL_TXCLK3) << 32); + *count1 = RREG32_PCIE(smnPCIE_PERF_COUNT0_TXCLK7) | + ((uint64_t)RREG32_PCIE(smnPCIE_PERF_COUNT0_UPVAL_TXCLK7) << 32); } const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { @@ -450,4 +549,193 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { .get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode, .get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode, .init_registers = nbio_v7_9_init_registers, + .get_pcie_replay_count = nbio_v7_9_get_pcie_replay_count, + .get_pcie_usage = nbio_v7_9_get_pcie_usage, +}; + +static void nbio_v7_9_query_ras_error_count(struct amdgpu_device *adev, + void *ras_error_status) +{ + return; +} + +static void nbio_v7_9_handle_ras_controller_intr_no_bifring(struct amdgpu_device *adev) +{ + uint32_t bif_doorbell_intr_cntl; + struct ras_manager *obj = amdgpu_ras_find_obj(adev, adev->nbio.ras_if); + struct ras_err_data err_data = {0, 0, 0, NULL}; + struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); + + bif_doorbell_intr_cntl = RREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL); + + if (REG_GET_FIELD(bif_doorbell_intr_cntl, + BIF_BX0_BIF_DOORBELL_INT_CNTL, RAS_CNTLR_INTERRUPT_STATUS)) { + /* driver has to clear the interrupt status when bif ring is disabled */ + bif_doorbell_intr_cntl = REG_SET_FIELD(bif_doorbell_intr_cntl, + BIF_BX0_BIF_DOORBELL_INT_CNTL, + RAS_CNTLR_INTERRUPT_CLEAR, 1); + WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL, bif_doorbell_intr_cntl); + + if (!ras->disable_ras_err_cnt_harvest) { + /* + * clear error status after ras_controller_intr + * according to hw team and count ue number + * for query + */ + nbio_v7_9_query_ras_error_count(adev, &err_data); + + /* logging on error cnt and printing for awareness */ + obj->err_data.ue_count += err_data.ue_count; + obj->err_data.ce_count += err_data.ce_count; + + if (err_data.ce_count) + dev_info(adev->dev, "%ld correctable hardware " + "errors detected in %s block, " + "no user action is needed.\n", + obj->err_data.ce_count, + get_ras_block_str(adev->nbio.ras_if)); + + if (err_data.ue_count) + dev_info(adev->dev, "%ld uncorrectable hardware " + "errors detected in %s block\n", + obj->err_data.ue_count, + get_ras_block_str(adev->nbio.ras_if)); + } + + dev_info(adev->dev, "RAS controller interrupt triggered " + "by NBIF error\n"); + + /* ras_controller_int is dedicated for nbif ras error, + * not the global interrupt for sync flood + */ + amdgpu_ras_reset_gpu(adev); + } +} + +static void nbio_v7_9_handle_ras_err_event_athub_intr_no_bifring(struct amdgpu_device *adev) +{ + uint32_t bif_doorbell_intr_cntl; + + bif_doorbell_intr_cntl = RREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL); + + if (REG_GET_FIELD(bif_doorbell_intr_cntl, + BIF_BX0_BIF_DOORBELL_INT_CNTL, RAS_ATHUB_ERR_EVENT_INTERRUPT_STATUS)) { + /* driver has to clear the interrupt status when bif ring is disabled */ + bif_doorbell_intr_cntl = REG_SET_FIELD(bif_doorbell_intr_cntl, + BIF_BX0_BIF_DOORBELL_INT_CNTL, + RAS_ATHUB_ERR_EVENT_INTERRUPT_CLEAR, 1); + + WREG32_SOC15(NBIO, 0, regBIF_BX0_BIF_DOORBELL_INT_CNTL, bif_doorbell_intr_cntl); + + amdgpu_ras_global_ras_isr(adev); + } +} + +static int nbio_v7_9_set_ras_controller_irq_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *src, + unsigned type, + enum amdgpu_interrupt_state state) +{ + /* Dummy function, there is no initialization operation in driver */ + + return 0; +} + +static int nbio_v7_9_process_ras_controller_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + /* By design, the ih cookie for ras_controller_irq should be written + * to BIFring instead of general iv ring. However, due to known bif ring + * hw bug, it has to be disabled. There is no chance the process function + * will be involked. Just left it as a dummy one. + */ + return 0; +} + +static int nbio_v7_9_set_ras_err_event_athub_irq_state(struct amdgpu_device *adev, + struct amdgpu_irq_src *src, + unsigned type, + enum amdgpu_interrupt_state state) +{ + /* Dummy function, there is no initialization operation in driver */ + + return 0; +} + +static int nbio_v7_9_process_err_event_athub_irq(struct amdgpu_device *adev, + struct amdgpu_irq_src *source, + struct amdgpu_iv_entry *entry) +{ + /* By design, the ih cookie for err_event_athub_irq should be written + * to BIFring instead of general iv ring. However, due to known bif ring + * hw bug, it has to be disabled. There is no chance the process function + * will be involked. Just left it as a dummy one. + */ + return 0; +} + +static const struct amdgpu_irq_src_funcs nbio_v7_9_ras_controller_irq_funcs = { + .set = nbio_v7_9_set_ras_controller_irq_state, + .process = nbio_v7_9_process_ras_controller_irq, +}; + +static const struct amdgpu_irq_src_funcs nbio_v7_9_ras_err_event_athub_irq_funcs = { + .set = nbio_v7_9_set_ras_err_event_athub_irq_state, + .process = nbio_v7_9_process_err_event_athub_irq, +}; + +static int nbio_v7_9_init_ras_controller_interrupt (struct amdgpu_device *adev) +{ + int r; + + /* init the irq funcs */ + adev->nbio.ras_controller_irq.funcs = + &nbio_v7_9_ras_controller_irq_funcs; + adev->nbio.ras_controller_irq.num_types = 1; + + /* register ras controller interrupt */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_BIF, + NBIF_7_4__SRCID__RAS_CONTROLLER_INTERRUPT, + &adev->nbio.ras_controller_irq); + + return r; +} + +static int nbio_v7_9_init_ras_err_event_athub_interrupt (struct amdgpu_device *adev) +{ + + int r; + + /* init the irq funcs */ + adev->nbio.ras_err_event_athub_irq.funcs = + &nbio_v7_9_ras_err_event_athub_irq_funcs; + adev->nbio.ras_err_event_athub_irq.num_types = 1; + + /* register ras err event athub interrupt */ + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_BIF, + NBIF_7_4__SRCID__ERREVENT_ATHUB_INTERRUPT, + &adev->nbio.ras_err_event_athub_irq); + + return r; +} + +const struct amdgpu_ras_block_hw_ops nbio_v7_9_ras_hw_ops = { + .query_ras_error_count = nbio_v7_9_query_ras_error_count, +}; + +struct amdgpu_nbio_ras nbio_v7_9_ras = { + .ras_block = { + .ras_comm = { + .name = "pcie_bif", + .block = AMDGPU_RAS_BLOCK__PCIE_BIF, + .type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE, + }, + .hw_ops = &nbio_v7_9_ras_hw_ops, + .ras_late_init = amdgpu_nbio_ras_late_init, + }, + .handle_ras_controller_intr_no_bifring = nbio_v7_9_handle_ras_controller_intr_no_bifring, + .handle_ras_err_event_athub_intr_no_bifring = nbio_v7_9_handle_ras_err_event_athub_intr_no_bifring, + .init_ras_controller_interrupt = nbio_v7_9_init_ras_controller_interrupt, + .init_ras_err_event_athub_interrupt = nbio_v7_9_init_ras_err_event_athub_interrupt, }; diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.h b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.h index 8e04eb484328..73709771950d 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.h +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.h @@ -28,5 +28,6 @@ extern const struct nbio_hdp_flush_reg nbio_v7_9_hdp_flush_reg; extern const struct amdgpu_nbio_funcs nbio_v7_9_funcs; +extern struct amdgpu_nbio_ras nbio_v7_9_ras; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 51523b27a186..13aca808ecab 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -67,21 +67,18 @@ static const struct amd_ip_funcs nv_common_ip_funcs; /* Navi */ -static const struct amdgpu_video_codec_info nv_video_codecs_encode_array[] = -{ +static const struct amdgpu_video_codec_info nv_video_codecs_encode_array[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2304, 0)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 4096, 2304, 0)}, }; -static const struct amdgpu_video_codecs nv_video_codecs_encode = -{ +static const struct amdgpu_video_codecs nv_video_codecs_encode = { .codec_count = ARRAY_SIZE(nv_video_codecs_encode_array), .codec_array = nv_video_codecs_encode_array, }; /* Navi1x */ -static const struct amdgpu_video_codec_info nv_video_codecs_decode_array[] = -{ +static const struct amdgpu_video_codec_info nv_video_codecs_decode_array[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2, 4096, 4096, 3)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4, 4096, 4096, 5)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, @@ -91,8 +88,7 @@ static const struct amdgpu_video_codec_info nv_video_codecs_decode_array[] = {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9, 8192, 4352, 0)}, }; -static const struct amdgpu_video_codecs nv_video_codecs_decode = -{ +static const struct amdgpu_video_codecs nv_video_codecs_decode = { .codec_count = ARRAY_SIZE(nv_video_codecs_decode_array), .codec_array = nv_video_codecs_decode_array, }; @@ -108,8 +104,7 @@ static const struct amdgpu_video_codecs sc_video_codecs_encode = { .codec_array = sc_video_codecs_encode_array, }; -static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn0[] = -{ +static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn0[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2, 4096, 4096, 3)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4, 4096, 4096, 5)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, @@ -120,8 +115,7 @@ static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn0[] {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1, 8192, 4352, 0)}, }; -static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn1[] = -{ +static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn1[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2, 4096, 4096, 3)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4, 4096, 4096, 5)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, @@ -131,27 +125,23 @@ static const struct amdgpu_video_codec_info sc_video_codecs_decode_array_vcn1[] {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9, 8192, 4352, 0)}, }; -static const struct amdgpu_video_codecs sc_video_codecs_decode_vcn0 = -{ +static const struct amdgpu_video_codecs sc_video_codecs_decode_vcn0 = { .codec_count = ARRAY_SIZE(sc_video_codecs_decode_array_vcn0), .codec_array = sc_video_codecs_decode_array_vcn0, }; -static const struct amdgpu_video_codecs sc_video_codecs_decode_vcn1 = -{ +static const struct amdgpu_video_codecs sc_video_codecs_decode_vcn1 = { .codec_count = ARRAY_SIZE(sc_video_codecs_decode_array_vcn1), .codec_array = sc_video_codecs_decode_array_vcn1, }; /* SRIOV Sienna Cichlid, not const since data is controlled by host */ -static struct amdgpu_video_codec_info sriov_sc_video_codecs_encode_array[] = -{ +static struct amdgpu_video_codec_info sriov_sc_video_codecs_encode_array[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2160, 0)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 7680, 4352, 0)}, }; -static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn0[] = -{ +static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn0[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2, 4096, 4096, 3)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4, 4096, 4096, 5)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, @@ -162,8 +152,7 @@ static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn0[] {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1, 8192, 4352, 0)}, }; -static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn1[] = -{ +static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn1[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2, 4096, 4096, 3)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4, 4096, 4096, 5)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, @@ -173,20 +162,17 @@ static struct amdgpu_video_codec_info sriov_sc_video_codecs_decode_array_vcn1[] {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9, 8192, 4352, 0)}, }; -static struct amdgpu_video_codecs sriov_sc_video_codecs_encode = -{ +static struct amdgpu_video_codecs sriov_sc_video_codecs_encode = { .codec_count = ARRAY_SIZE(sriov_sc_video_codecs_encode_array), .codec_array = sriov_sc_video_codecs_encode_array, }; -static struct amdgpu_video_codecs sriov_sc_video_codecs_decode_vcn0 = -{ +static struct amdgpu_video_codecs sriov_sc_video_codecs_decode_vcn0 = { .codec_count = ARRAY_SIZE(sriov_sc_video_codecs_decode_array_vcn0), .codec_array = sriov_sc_video_codecs_decode_array_vcn0, }; -static struct amdgpu_video_codecs sriov_sc_video_codecs_decode_vcn1 = -{ +static struct amdgpu_video_codecs sriov_sc_video_codecs_decode_vcn1 = { .codec_count = ARRAY_SIZE(sriov_sc_video_codecs_decode_array_vcn1), .codec_array = sriov_sc_video_codecs_decode_array_vcn1, }; @@ -536,8 +522,7 @@ static void nv_program_aspm(struct amdgpu_device *adev) } -const struct amdgpu_ip_block_version nv_common_ip_block = -{ +const struct amdgpu_ip_block_version nv_common_ip_block = { .type = AMD_IP_BLOCK_TYPE_COMMON, .major = 1, .minor = 0, @@ -572,16 +557,6 @@ static bool nv_need_reset_on_init(struct amdgpu_device *adev) return false; } -static uint64_t nv_get_pcie_replay_count(struct amdgpu_device *adev) -{ - - /* TODO - * dummy implement for pcie_replay_count sysfs interface - * */ - - return 0; -} - static void nv_init_doorbell_index(struct amdgpu_device *adev) { adev->doorbell_index.kiq = AMDGPU_NAVI10_DOORBELL_KIQ; @@ -642,8 +617,7 @@ static int nv_update_umd_stable_pstate(struct amdgpu_device *adev, return 0; } -static const struct amdgpu_asic_funcs nv_asic_funcs = -{ +static const struct amdgpu_asic_funcs nv_asic_funcs = { .read_disabled_bios = &nv_read_disabled_bios, .read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom, .read_register = &nv_read_register, @@ -656,7 +630,7 @@ static const struct amdgpu_asic_funcs nv_asic_funcs = .init_doorbell_index = &nv_init_doorbell_index, .need_full_reset = &nv_need_full_reset, .need_reset_on_init = &nv_need_reset_on_init, - .get_pcie_replay_count = &nv_get_pcie_replay_count, + .get_pcie_replay_count = &amdgpu_nbio_get_pcie_replay_count, .supports_baco = &amdgpu_dpm_is_baco_supported, .pre_asic_init = &nv_pre_asic_init, .update_umd_stable_pstate = &nv_update_umd_stable_pstate, @@ -889,7 +863,8 @@ static int nv_common_early_init(void *handle) AMD_CG_SUPPORT_ATHUB_LS | AMD_CG_SUPPORT_IH_CG | AMD_CG_SUPPORT_VCN_MGCG | - AMD_CG_SUPPORT_JPEG_MGCG; + AMD_CG_SUPPORT_JPEG_MGCG | + AMD_CG_SUPPORT_SDMA_MGCG; adev->pg_flags = AMD_PG_SUPPORT_GFX_PG | AMD_PG_SUPPORT_VCN | AMD_PG_SUPPORT_VCN_DPG | @@ -950,7 +925,8 @@ static int nv_common_early_init(void *handle) AMD_CG_SUPPORT_ATHUB_LS | AMD_CG_SUPPORT_IH_CG | AMD_CG_SUPPORT_VCN_MGCG | - AMD_CG_SUPPORT_JPEG_MGCG; + AMD_CG_SUPPORT_JPEG_MGCG | + AMD_CG_SUPPORT_SDMA_MGCG; adev->pg_flags = AMD_PG_SUPPORT_VCN | AMD_PG_SUPPORT_VCN_DPG | AMD_PG_SUPPORT_JPEG | diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index f9cb0d2c89d1..469eed084976 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -49,6 +49,9 @@ MODULE_FIRMWARE("amdgpu/psp_13_0_10_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_11_toc.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_11_ta.bin"); MODULE_FIRMWARE("amdgpu/psp_13_0_6_sos.bin"); +MODULE_FIRMWARE("amdgpu/psp_13_0_6_ta.bin"); +MODULE_FIRMWARE("amdgpu/psp_14_0_0_toc.bin"); +MODULE_FIRMWARE("amdgpu/psp_14_0_0_ta.bin"); /* For large FW files the time to complete can be very long */ #define USBC_PD_POLLING_LIMIT_S 240 @@ -93,6 +96,7 @@ static int psp_v13_0_init_microcode(struct psp_context *psp) case IP_VERSION(13, 0, 5): case IP_VERSION(13, 0, 8): case IP_VERSION(13, 0, 11): + case IP_VERSION(14, 0, 0): err = psp_init_toc_microcode(psp, ucode_prefix); if (err) return err; @@ -129,21 +133,42 @@ static bool psp_v13_0_is_sos_alive(struct psp_context *psp) return sol_reg != 0x0; } -static int psp_v13_0_wait_for_bootloader(struct psp_context *psp) +static int psp_v13_0_wait_for_vmbx_ready(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; + int retry_loop, ret; - int ret; - int retry_loop; + for (retry_loop = 0; retry_loop < 70; retry_loop++) { + /* Wait for bootloader to signify that is + ready having bit 31 of C2PMSG_33 set to 1 */ + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_33), + 0x80000000, 0xffffffff, false); + + if (ret == 0) + break; + } + + if (ret) + dev_warn(adev->dev, "Bootloader wait timed out"); + + return ret; +} + +static int psp_v13_0_wait_for_bootloader(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + int retry_loop, ret; + /* Wait for bootloader to signify that it is ready having bit 31 of + * C2PMSG_35 set to 1. All other bits are expected to be cleared. + * If there is an error in processing command, bits[7:0] will be set. + * This is applicable for PSP v13.0.6 and newer. + */ for (retry_loop = 0; retry_loop < 10; retry_loop++) { - /* Wait for bootloader to signify that is - ready having bit 31 of C2PMSG_35 set to 1 */ - ret = psp_wait_for(psp, - SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), - 0x80000000, - 0x80000000, - false); + ret = psp_wait_for( + psp, SOC15_REG_OFFSET(MP0, 0, regMP0_SMN_C2PMSG_35), + 0x80000000, 0xffffffff, false); if (ret == 0) return 0; @@ -152,6 +177,19 @@ static int psp_v13_0_wait_for_bootloader(struct psp_context *psp) return ret; } +static int psp_v13_0_wait_for_bootloader_steady_state(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + + if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 6)) { + psp_v13_0_wait_for_vmbx_ready(psp); + + return psp_v13_0_wait_for_bootloader(psp); + } + + return 0; +} + static int psp_v13_0_bootloader_load_component(struct psp_context *psp, struct psp_bin_desc *bin_desc, enum psp_bootloader_cmd bl_cmd) @@ -686,8 +724,30 @@ static int psp_v13_0_vbflash_status(struct psp_context *psp) return RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_115); } +static int psp_v13_0_fatal_error_recovery_quirk(struct psp_context *psp) +{ + struct amdgpu_device *adev = psp->adev; + + if (adev->ip_versions[MP0_HWIP][0] == IP_VERSION(13, 0, 10)) { + uint32_t reg_data; + /* MP1 fatal error: trigger PSP dram read to unhalt PSP + * during MP1 triggered sync flood. + */ + reg_data = RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_67); + WREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_67, reg_data + 0x10); + + /* delay 1000ms for the mode1 reset for fatal error + * to be recovered back. + */ + msleep(1000); + } + + return 0; +} + static const struct psp_funcs psp_v13_0_funcs = { .init_microcode = psp_v13_0_init_microcode, + .wait_for_bootloader = psp_v13_0_wait_for_bootloader_steady_state, .bootloader_load_kdb = psp_v13_0_bootloader_load_kdb, .bootloader_load_spl = psp_v13_0_bootloader_load_spl, .bootloader_load_sysdrv = psp_v13_0_bootloader_load_sysdrv, @@ -705,7 +765,8 @@ static const struct psp_funcs psp_v13_0_funcs = { .load_usbc_pd_fw = psp_v13_0_load_usbc_pd_fw, .read_usbc_pd_fw = psp_v13_0_read_usbc_pd_fw, .update_spirom = psp_v13_0_update_spirom, - .vbflash_stat = psp_v13_0_vbflash_status + .vbflash_stat = psp_v13_0_vbflash_status, + .fatal_error_recovery_quirk = psp_v13_0_fatal_error_recovery_quirk, }; void psp_v13_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 5c4d4df9cf94..1cc34efb455b 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -237,17 +237,15 @@ static void sdma_v5_0_init_golden_registers(struct amdgpu_device *adev) // emulation only, won't work on real chip // navi10 real chip need to use PSP to load firmware static int sdma_v5_0_init_microcode(struct amdgpu_device *adev) -{ int ret, i; - - if (amdgpu_sriov_vf(adev) && (adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(5, 0, 5))) - return 0; +{ + int ret, i; for (i = 0; i < adev->sdma.num_instances; i++) { ret = amdgpu_sdma_init_microcode(adev, i, false); if (ret) return ret; } - + return ret; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index a7b230e5a26d..2b3ebebc4299 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -1507,6 +1507,30 @@ static int sdma_v5_2_process_illegal_inst_irq(struct amdgpu_device *adev, return 0; } +static bool sdma_v5_2_firmware_mgcg_support(struct amdgpu_device *adev, + int i) +{ + switch (adev->ip_versions[SDMA0_HWIP][0]) { + case IP_VERSION(5, 2, 1): + if (adev->sdma.instance[i].fw_version < 70) + return false; + break; + case IP_VERSION(5, 2, 3): + if (adev->sdma.instance[i].fw_version < 47) + return false; + break; + case IP_VERSION(5, 2, 7): + if (adev->sdma.instance[i].fw_version < 9) + return false; + break; + default: + return true; + } + + return true; + +} + static void sdma_v5_2_update_medium_grain_clock_gating(struct amdgpu_device *adev, bool enable) { @@ -1515,7 +1539,7 @@ static void sdma_v5_2_update_medium_grain_clock_gating(struct amdgpu_device *ade for (i = 0; i < adev->sdma.num_instances; i++) { - if (adev->sdma.instance[i].fw_version < 70 && adev->ip_versions[SDMA0_HWIP][0] == IP_VERSION(5, 2, 1)) + if (!sdma_v5_2_firmware_mgcg_support(adev, i)) adev->cg_flags &= ~AMD_CG_SUPPORT_SDMA_MGCG; if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) { @@ -1589,6 +1613,7 @@ static int sdma_v5_2_set_clockgating_state(void *handle, case IP_VERSION(5, 2, 5): case IP_VERSION(5, 2, 6): case IP_VERSION(5, 2, 3): + case IP_VERSION(5, 2, 7): sdma_v5_2_update_medium_grain_clock_gating(adev, state == AMD_CG_STATE_GATE); sdma_v5_2_update_medium_grain_light_sleep(adev, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 3b03dda854fd..45be0af2570b 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -48,6 +48,7 @@ MODULE_FIRMWARE("amdgpu/sdma_6_0_0.bin"); MODULE_FIRMWARE("amdgpu/sdma_6_0_1.bin"); MODULE_FIRMWARE("amdgpu/sdma_6_0_2.bin"); MODULE_FIRMWARE("amdgpu/sdma_6_0_3.bin"); +MODULE_FIRMWARE("amdgpu/sdma_6_1_0.bin"); #define SDMA1_REG_OFFSET 0x600 #define SDMA0_HYP_DEC_REG_START 0x5880 diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index f64b87b11b1b..4b81f29e5fd5 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -2276,17 +2276,8 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &bridge_cfg); - pcie_capability_read_word(adev->pdev, PCI_EXP_LNKCTL, - &gpu_cfg); - - tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); - - tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(adev->pdev, PCI_EXP_LNKCTL, - tmp16); + pcie_capability_set_word(root, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); + pcie_capability_set_word(adev->pdev, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); tmp = RREG32_PCIE(PCIE_LC_STATUS1); max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT; @@ -2331,21 +2322,14 @@ static void si_pcie_gen3_enable(struct amdgpu_device *adev) mdelay(100); - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(root, PCI_EXP_LNKCTL, - tmp16); - - pcie_capability_read_word(adev->pdev, - PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(adev->pdev, - PCI_EXP_LNKCTL, - tmp16); + pcie_capability_clear_and_set_word(root, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + bridge_cfg & + PCI_EXP_LNKCTL_HAWD); + pcie_capability_clear_and_set_word(adev->pdev, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + gpu_cfg & + PCI_EXP_LNKCTL_HAWD); pcie_capability_read_word(root, PCI_EXP_LNKCTL2, &tmp16); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index afcaeadda4c7..f5be40d7ba36 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -559,8 +559,10 @@ soc15_asic_reset_method(struct amdgpu_device *adev) */ if (amdgpu_gpu_recovery == 4 || amdgpu_gpu_recovery == 5) return AMD_RESET_METHOD_MODE2; + else if (!(adev->flags & AMD_IS_APU)) + return AMD_RESET_METHOD_MODE1; else - return AMD_RESET_METHOD_NONE; + return AMD_RESET_METHOD_MODE2; default: break; } @@ -893,9 +895,9 @@ static const struct amdgpu_asic_funcs aqua_vanjaram_asic_funcs = .get_config_memsize = &soc15_get_config_memsize, .need_full_reset = &soc15_need_full_reset, .init_doorbell_index = &aqua_vanjaram_doorbell_index_init, - .get_pcie_usage = &vega20_get_pcie_usage, + .get_pcie_usage = &amdgpu_nbio_get_pcie_usage, .need_reset_on_init = &soc15_need_reset_on_init, - .get_pcie_replay_count = &soc15_get_pcie_replay_count, + .get_pcie_replay_count = &amdgpu_nbio_get_pcie_replay_count, .supports_baco = &soc15_supports_baco, .pre_asic_init = &soc15_pre_asic_init, .query_video_codecs = &soc15_query_video_codecs, diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 96948a59f8dd..da683afa0222 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -37,65 +37,65 @@ #define SOC15_REG_OFFSET1(ip, inst, reg, offset) \ (adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)+(offset)) -#define __WREG32_SOC15_RLC__(reg, value, flag, hwip) \ +#define __WREG32_SOC15_RLC__(reg, value, flag, hwip, inst) \ ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.rlcg_reg_access_supported) ? \ - amdgpu_sriov_wreg(adev, reg, value, flag, hwip) : \ + amdgpu_sriov_wreg(adev, reg, value, flag, hwip, inst) : \ WREG32(reg, value)) -#define __RREG32_SOC15_RLC__(reg, flag, hwip) \ +#define __RREG32_SOC15_RLC__(reg, flag, hwip, inst) \ ((amdgpu_sriov_vf(adev) && adev->gfx.rlc.funcs && adev->gfx.rlc.rlcg_reg_access_supported) ? \ - amdgpu_sriov_rreg(adev, reg, flag, hwip) : \ + amdgpu_sriov_rreg(adev, reg, flag, hwip, inst) : \ RREG32(reg)) #define WREG32_FIELD15(ip, idx, reg, field, val) \ __WREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][idx][mm##reg##_BASE_IDX] + mm##reg, \ (__RREG32_SOC15_RLC__( \ adev->reg_offset[ip##_HWIP][idx][mm##reg##_BASE_IDX] + mm##reg, \ - 0, ip##_HWIP) & \ + 0, ip##_HWIP, idx) & \ ~REG_FIELD_MASK(reg, field)) | (val) << REG_FIELD_SHIFT(reg, field), \ - 0, ip##_HWIP) + 0, ip##_HWIP, idx) #define WREG32_FIELD15_PREREG(ip, idx, reg_name, field, val) \ __WREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][idx][reg##reg_name##_BASE_IDX] + reg##reg_name, \ (__RREG32_SOC15_RLC__( \ adev->reg_offset[ip##_HWIP][idx][reg##reg_name##_BASE_IDX] + reg##reg_name, \ - 0, ip##_HWIP) & \ + 0, ip##_HWIP, idx) & \ ~REG_FIELD_MASK(reg_name, field)) | (val) << REG_FIELD_SHIFT(reg_name, field), \ - 0, ip##_HWIP) + 0, ip##_HWIP, idx) #define RREG32_SOC15(ip, inst, reg) \ __RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \ - 0, ip##_HWIP) + 0, ip##_HWIP, inst) -#define RREG32_SOC15_IP(ip, reg) __RREG32_SOC15_RLC__(reg, 0, ip##_HWIP) +#define RREG32_SOC15_IP(ip, reg) __RREG32_SOC15_RLC__(reg, 0, ip##_HWIP, 0) -#define RREG32_SOC15_IP_NO_KIQ(ip, reg) __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ, ip##_HWIP) +#define RREG32_SOC15_IP_NO_KIQ(ip, reg) __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ, ip##_HWIP, 0) #define RREG32_SOC15_NO_KIQ(ip, inst, reg) \ __RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \ - AMDGPU_REGS_NO_KIQ, ip##_HWIP) + AMDGPU_REGS_NO_KIQ, ip##_HWIP, inst) #define RREG32_SOC15_OFFSET(ip, inst, reg, offset) \ __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + (reg)) + \ - (offset), 0, ip##_HWIP) + (offset), 0, ip##_HWIP, inst) #define WREG32_SOC15(ip, inst, reg, value) \ __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg), \ - value, 0, ip##_HWIP) + value, 0, ip##_HWIP, inst) #define WREG32_SOC15_IP(ip, reg, value) \ - __WREG32_SOC15_RLC__(reg, value, 0, ip##_HWIP) + __WREG32_SOC15_RLC__(reg, value, 0, ip##_HWIP, 0) #define WREG32_SOC15_IP_NO_KIQ(ip, reg, value) \ - __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ, ip##_HWIP) + __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ, ip##_HWIP, 0) #define WREG32_SOC15_NO_KIQ(ip, inst, reg, value) \ __WREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, \ - value, AMDGPU_REGS_NO_KIQ, ip##_HWIP) + value, AMDGPU_REGS_NO_KIQ, ip##_HWIP, inst) #define WREG32_SOC15_OFFSET(ip, inst, reg, offset, value) \ __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, \ - value, 0, ip##_HWIP) + value, 0, ip##_HWIP, inst) #define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask) \ amdgpu_device_wait_on_rreg(adev, inst, \ @@ -108,16 +108,16 @@ #reg, expected_value, mask) #define WREG32_RLC(reg, value) \ - __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_RLC, GC_HWIP) + __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_RLC, GC_HWIP, 0) -#define WREG32_RLC_EX(prefix, reg, value) \ +#define WREG32_RLC_EX(prefix, reg, value, inst) \ do { \ if (amdgpu_sriov_fullaccess(adev)) { \ uint32_t i = 0; \ uint32_t retries = 50000; \ - uint32_t r0 = adev->reg_offset[GC_HWIP][0][prefix##SCRATCH_REG0_BASE_IDX] + prefix##SCRATCH_REG0; \ - uint32_t r1 = adev->reg_offset[GC_HWIP][0][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG1; \ - uint32_t spare_int = adev->reg_offset[GC_HWIP][0][prefix##RLC_SPARE_INT_BASE_IDX] + prefix##RLC_SPARE_INT; \ + uint32_t r0 = adev->reg_offset[GC_HWIP][inst][prefix##SCRATCH_REG0_BASE_IDX] + prefix##SCRATCH_REG0; \ + uint32_t r1 = adev->reg_offset[GC_HWIP][inst][prefix##SCRATCH_REG1_BASE_IDX] + prefix##SCRATCH_REG1; \ + uint32_t spare_int = adev->reg_offset[GC_HWIP][inst][prefix##RLC_SPARE_INT_BASE_IDX] + prefix##RLC_SPARE_INT; \ WREG32(r0, value); \ WREG32(r1, (reg | 0x80000000)); \ WREG32(spare_int, 0x1); \ @@ -136,17 +136,17 @@ /* shadow the registers in the callback function */ #define WREG32_SOC15_RLC_SHADOW(ip, inst, reg, value) \ - __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg), value, AMDGPU_REGS_RLC, GC_HWIP) + __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg), value, AMDGPU_REGS_RLC, GC_HWIP, inst) /* for GC only */ #define RREG32_RLC(reg) \ __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_RLC, GC_HWIP) #define WREG32_RLC_NO_KIQ(reg, value, hwip) \ - __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ | AMDGPU_REGS_RLC, hwip) + __WREG32_SOC15_RLC__(reg, value, AMDGPU_REGS_NO_KIQ | AMDGPU_REGS_RLC, hwip, 0) #define RREG32_RLC_NO_KIQ(reg, hwip) \ - __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ | AMDGPU_REGS_RLC, hwip) + __RREG32_SOC15_RLC__(reg, AMDGPU_REGS_NO_KIQ | AMDGPU_REGS_RLC, hwip, 0) #define WREG32_SOC15_RLC_SHADOW_EX(prefix, ip, inst, reg, value) \ do { \ @@ -167,32 +167,32 @@ } while (0) #define RREG32_SOC15_RLC(ip, inst, reg) \ - __RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, AMDGPU_REGS_RLC, ip##_HWIP) + __RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg, AMDGPU_REGS_RLC, ip##_HWIP, inst) #define WREG32_SOC15_RLC(ip, inst, reg, value) \ do { \ uint32_t target_reg = adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg;\ - __WREG32_SOC15_RLC__(target_reg, value, AMDGPU_REGS_RLC, ip##_HWIP); \ + __WREG32_SOC15_RLC__(target_reg, value, AMDGPU_REGS_RLC, ip##_HWIP, inst); \ } while (0) #define WREG32_SOC15_RLC_EX(prefix, ip, inst, reg, value) \ do { \ uint32_t target_reg = adev->reg_offset[GC_HWIP][inst][reg##_BASE_IDX] + reg;\ - WREG32_RLC_EX(prefix, target_reg, value); \ + WREG32_RLC_EX(prefix, target_reg, value, inst); \ } while (0) #define WREG32_FIELD15_RLC(ip, idx, reg, field, val) \ __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][idx][mm##reg##_BASE_IDX] + mm##reg), \ (__RREG32_SOC15_RLC__(adev->reg_offset[ip##_HWIP][idx][mm##reg##_BASE_IDX] + mm##reg, \ - AMDGPU_REGS_RLC, ip##_HWIP) & \ + AMDGPU_REGS_RLC, ip##_HWIP, idx) & \ ~REG_FIELD_MASK(reg, field)) | (val) << REG_FIELD_SHIFT(reg, field), \ - AMDGPU_REGS_RLC, ip##_HWIP) + AMDGPU_REGS_RLC, ip##_HWIP, idx) #define WREG32_SOC15_OFFSET_RLC(ip, inst, reg, offset, value) \ - __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, value, AMDGPU_REGS_RLC, ip##_HWIP) + __WREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, value, AMDGPU_REGS_RLC, ip##_HWIP, inst) #define RREG32_SOC15_OFFSET_RLC(ip, inst, reg, offset) \ - __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, AMDGPU_REGS_RLC, ip##_HWIP) + __RREG32_SOC15_RLC__((adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg) + offset, AMDGPU_REGS_RLC, ip##_HWIP, inst) /* inst equals to ext for some IPs */ #define RREG32_SOC15_EXT(ip, inst, reg, ext) \ diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index e5e5d68a4d70..8b2ff2b281b0 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -48,33 +48,28 @@ static const struct amd_ip_funcs soc21_common_ip_funcs; /* SOC21 */ -static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_encode_array_vcn0[] = -{ +static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_encode_array_vcn0[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2304, 0)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 4096, 2304, 0)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1, 8192, 4352, 0)}, }; -static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_encode_array_vcn1[] = -{ +static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_encode_array_vcn1[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 2304, 0)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 4096, 2304, 0)}, }; -static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_encode_vcn0 = -{ +static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_encode_vcn0 = { .codec_count = ARRAY_SIZE(vcn_4_0_0_video_codecs_encode_array_vcn0), .codec_array = vcn_4_0_0_video_codecs_encode_array_vcn0, }; -static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_encode_vcn1 = -{ +static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_encode_vcn1 = { .codec_count = ARRAY_SIZE(vcn_4_0_0_video_codecs_encode_array_vcn1), .codec_array = vcn_4_0_0_video_codecs_encode_array_vcn1, }; -static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_decode_array_vcn0[] = -{ +static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_decode_array_vcn0[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 8192, 4352, 186)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG, 4096, 4096, 0)}, @@ -82,22 +77,19 @@ static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_decode_array_ {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1, 8192, 4352, 0)}, }; -static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_decode_array_vcn1[] = -{ +static const struct amdgpu_video_codec_info vcn_4_0_0_video_codecs_decode_array_vcn1[] = { {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC, 4096, 4096, 52)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC, 8192, 4352, 186)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG, 4096, 4096, 0)}, {codec_info_build(AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9, 8192, 4352, 0)}, }; -static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_decode_vcn0 = -{ +static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_decode_vcn0 = { .codec_count = ARRAY_SIZE(vcn_4_0_0_video_codecs_decode_array_vcn0), .codec_array = vcn_4_0_0_video_codecs_decode_array_vcn0, }; -static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_decode_vcn1 = -{ +static const struct amdgpu_video_codecs vcn_4_0_0_video_codecs_decode_vcn1 = { .codec_count = ARRAY_SIZE(vcn_4_0_0_video_codecs_decode_array_vcn1), .codec_array = vcn_4_0_0_video_codecs_decode_array_vcn1, }; @@ -445,8 +437,7 @@ static void soc21_program_aspm(struct amdgpu_device *adev) adev->nbio.funcs->program_aspm(adev); } -const struct amdgpu_ip_block_version soc21_common_ip_block = -{ +const struct amdgpu_ip_block_version soc21_common_ip_block = { .type = AMD_IP_BLOCK_TYPE_COMMON, .major = 1, .minor = 0, @@ -484,16 +475,6 @@ static bool soc21_need_reset_on_init(struct amdgpu_device *adev) return false; } -static uint64_t soc21_get_pcie_replay_count(struct amdgpu_device *adev) -{ - - /* TODO - * dummy implement for pcie_replay_count sysfs interface - * */ - - return 0; -} - static void soc21_init_doorbell_index(struct amdgpu_device *adev) { adev->doorbell_index.kiq = AMDGPU_NAVI10_DOORBELL_KIQ; @@ -547,8 +528,7 @@ static int soc21_update_umd_stable_pstate(struct amdgpu_device *adev, return 0; } -static const struct amdgpu_asic_funcs soc21_asic_funcs = -{ +static const struct amdgpu_asic_funcs soc21_asic_funcs = { .read_disabled_bios = &soc21_read_disabled_bios, .read_bios_from_rom = &amdgpu_soc15_read_bios_from_rom, .read_register = &soc21_read_register, @@ -561,7 +541,7 @@ static const struct amdgpu_asic_funcs soc21_asic_funcs = .init_doorbell_index = &soc21_init_doorbell_index, .need_full_reset = &soc21_need_full_reset, .need_reset_on_init = &soc21_need_reset_on_init, - .get_pcie_replay_count = &soc21_get_pcie_replay_count, + .get_pcie_replay_count = &amdgpu_nbio_get_pcie_replay_count, .supports_baco = &amdgpu_dpm_is_baco_supported, .pre_asic_init = &soc21_pre_asic_init, .query_video_codecs = &soc21_query_video_codecs, @@ -786,7 +766,7 @@ static int soc21_common_hw_init(void *handle) * for the purpose of expose those registers * to process space */ - if (adev->nbio.funcs->remap_hdp_registers) + if (adev->nbio.funcs->remap_hdp_registers && !amdgpu_sriov_vf(adev)) adev->nbio.funcs->remap_hdp_registers(adev); /* enable the doorbell aperture */ adev->nbio.funcs->enable_doorbell_aperture(adev, true); diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c index b08905d1c00f..917707bba7f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c @@ -493,8 +493,7 @@ static void tonga_ih_set_interrupt_funcs(struct amdgpu_device *adev) adev->irq.ih_funcs = &tonga_ih_funcs; } -const struct amdgpu_ip_block_version tonga_ih_ip_block = -{ +const struct amdgpu_ip_block_version tonga_ih_ip_block = { .type = AMD_IP_BLOCK_TYPE_IH, .major = 3, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c index 0fef925b6602..5534c769b655 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c @@ -815,8 +815,7 @@ static const struct amd_ip_funcs uvd_v3_1_ip_funcs = { .set_powergating_state = uvd_v3_1_set_powergating_state, }; -const struct amdgpu_ip_block_version uvd_v3_1_ip_block = -{ +const struct amdgpu_ip_block_version uvd_v3_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_UVD, .major = 3, .minor = 1, diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index abaa4463e906..86d1d46e1e5e 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -679,11 +679,11 @@ static void uvd_v7_0_mc_resume(struct amdgpu_device *adev) if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW, i == 0 ? - adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].tmr_mc_addr_lo: + adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].tmr_mc_addr_lo : adev->firmware.ucode[AMDGPU_UCODE_ID_UVD1].tmr_mc_addr_lo); WREG32_SOC15(UVD, i, mmUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH, i == 0 ? - adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].tmr_mc_addr_hi: + adev->firmware.ucode[AMDGPU_UCODE_ID_UVD].tmr_mc_addr_hi : adev->firmware.ucode[AMDGPU_UCODE_ID_UVD1].tmr_mc_addr_hi); WREG32_SOC15(UVD, i, mmUVD_VCPU_CACHE_OFFSET0, 0); offset = 0; @@ -1908,8 +1908,7 @@ static void uvd_v7_0_set_irq_funcs(struct amdgpu_device *adev) } } -const struct amdgpu_ip_block_version uvd_v7_0_ip_block = -{ +const struct amdgpu_ip_block_version uvd_v7_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_UVD, .major = 7, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index 8def62c83ffd..18f6e62af339 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -998,8 +998,7 @@ static void vce_v3_0_set_irq_funcs(struct amdgpu_device *adev) adev->vce.irq.funcs = &vce_v3_0_irq_funcs; }; -const struct amdgpu_ip_block_version vce_v3_0_ip_block = -{ +const struct amdgpu_ip_block_version vce_v3_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_VCE, .major = 3, .minor = 0, @@ -1007,8 +1006,7 @@ const struct amdgpu_ip_block_version vce_v3_0_ip_block = .funcs = &vce_v3_0_ip_funcs, }; -const struct amdgpu_ip_block_version vce_v3_1_ip_block = -{ +const struct amdgpu_ip_block_version vce_v3_1_ip_block = { .type = AMD_IP_BLOCK_TYPE_VCE, .major = 3, .minor = 1, @@ -1016,8 +1014,7 @@ const struct amdgpu_ip_block_version vce_v3_1_ip_block = .funcs = &vce_v3_0_ip_funcs, }; -const struct amdgpu_ip_block_version vce_v3_4_ip_block = -{ +const struct amdgpu_ip_block_version vce_v3_4_ip_block = { .type = AMD_IP_BLOCK_TYPE_VCE, .major = 3, .minor = 4, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index 16feb491adf5..25ba27151ac0 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -473,7 +473,7 @@ static void vcn_v1_0_disable_clock_gating(struct amdgpu_device *adev) if (adev->cg_flags & AMD_CG_SUPPORT_VCN_MGCG) data |= 1 << UVD_CGC_CTRL__DYN_CLOCK_MODE__SHIFT; else - data &= ~ UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK; + data &= ~UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK; data |= 1 << UVD_CGC_CTRL__CLK_GATE_DLY_TIMER__SHIFT; data |= 4 << UVD_CGC_CTRL__CLK_OFF_DELAY__SHIFT; @@ -1772,7 +1772,7 @@ static int vcn_v1_0_set_powergating_state(void *handle, int ret; struct amdgpu_device *adev = (struct amdgpu_device *)handle; - if(state == adev->vcn.cur_state) + if (state == adev->vcn.cur_state) return 0; if (state == AMD_PG_STATE_GATE) @@ -1780,7 +1780,7 @@ static int vcn_v1_0_set_powergating_state(void *handle, else ret = vcn_v1_0_start(adev); - if(!ret) + if (!ret) adev->vcn.cur_state = state; return ret; } @@ -2065,8 +2065,7 @@ static void vcn_v1_0_set_irq_funcs(struct amdgpu_device *adev) adev->vcn.inst->irq.funcs = &vcn_v1_0_irq_funcs; } -const struct amdgpu_ip_block_version vcn_v1_0_ip_block = -{ +const struct amdgpu_ip_block_version vcn_v1_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_VCN, .major = 1, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index c975aed2f6c7..18794394c5a0 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -881,9 +881,7 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_device *adev, bool indirect) UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); if (indirect) - psp_update_vcn_sram(adev, 0, adev->vcn.inst->dpg_sram_gpu_addr, - (uint32_t)((uintptr_t)adev->vcn.inst->dpg_sram_curr_addr - - (uintptr_t)adev->vcn.inst->dpg_sram_cpu_addr)); + amdgpu_vcn_psp_update_sram(adev, 0, 0); /* force RBC into idle state */ rb_bufsz = order_base_2(ring->ring_size); diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index bb1875f926f1..6fbea38f4d3e 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -912,9 +912,7 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); if (indirect) - psp_update_vcn_sram(adev, inst_idx, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr, - (uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr - - (uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr)); + amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); ring = &adev->vcn.inst[inst_idx].ring_dec; /* force RBC into idle state */ diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index c8f63b3c6f69..a61ecefdafc5 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -1037,9 +1037,7 @@ static int vcn_v3_0_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo VCN, inst_idx, mmUVD_VCPU_CNTL), tmp, 0, indirect); if (indirect) - psp_update_vcn_sram(adev, inst_idx, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr, - (uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr - - (uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr)); + amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); ring = &adev->vcn.inst[inst_idx].ring_dec; /* force RBC into idle state */ @@ -1107,7 +1105,7 @@ static int vcn_v3_0_start(struct amdgpu_device *adev) if (adev->vcn.harvest_config & (1 << i)) continue; - if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG){ + if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { r = vcn_v3_0_start_dpg_mode(adev, i, adev->vcn.indirect_sram); continue; } @@ -1791,7 +1789,7 @@ static int vcn_v3_0_dec_msg(struct amdgpu_cs_parser *p, struct amdgpu_job *job, struct amdgpu_bo *bo; uint64_t start, end; unsigned int i; - void * ptr; + void *ptr; int r; addr &= AMDGPU_GMC_HOLE_MASK; @@ -2097,7 +2095,7 @@ static int vcn_v3_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = state == AMD_CG_STATE_GATE; int i; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -2131,7 +2129,7 @@ static int vcn_v3_0_set_powergating_state(void *handle, return 0; } - if(state == adev->vcn.cur_state) + if (state == adev->vcn.cur_state) return 0; if (state == AMD_PG_STATE_GATE) @@ -2139,7 +2137,7 @@ static int vcn_v3_0_set_powergating_state(void *handle, else ret = vcn_v3_0_start(adev); - if(!ret) + if (!ret) adev->vcn.cur_state = state; return ret; @@ -2230,8 +2228,7 @@ static const struct amd_ip_funcs vcn_v3_0_ip_funcs = { .set_powergating_state = vcn_v3_0_set_powergating_state, }; -const struct amdgpu_ip_block_version vcn_v3_0_ip_block = -{ +const struct amdgpu_ip_block_version vcn_v3_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_VCN, .major = 3, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index 259795098173..29164289c5f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -169,6 +169,12 @@ static int vcn_v4_0_sw_init(void *handle) fw_shared->smu_dpm_interface.smu_interface_type = (adev->flags & AMD_IS_APU) ? AMDGPU_VCN_SMU_DPM_INTERFACE_APU : AMDGPU_VCN_SMU_DPM_INTERFACE_DGPU; + if (adev->ip_versions[VCN_HWIP][0] == IP_VERSION(4, 0, 2)) { + fw_shared->present_flag_0 |= AMDGPU_FW_SHARED_FLAG_0_DRM_KEY_INJECT; + fw_shared->drm_key_wa.method = + AMDGPU_DRM_KEY_INJECT_WORKAROUND_VCNFW_ASD_HANDSHAKING; + } + if (amdgpu_sriov_vf(adev)) fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_VF_RB_SETUP_FLAG); @@ -993,9 +999,7 @@ static int vcn_v4_0_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, boo if (indirect) - psp_update_vcn_sram(adev, inst_idx, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr, - (uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr - - (uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr)); + amdgpu_vcn_psp_update_sram(adev, inst_idx, 0); ring = &adev->vcn.inst[inst_idx].ring_enc[0]; @@ -1135,11 +1139,11 @@ static int vcn_v4_0_start(struct amdgpu_device *adev) if (status & 2) break; mdelay(10); - if (amdgpu_emu_mode==1) + if (amdgpu_emu_mode == 1) msleep(1); } - if (amdgpu_emu_mode==1) { + if (amdgpu_emu_mode == 1) { r = -1; if (status & 2) { r = 0; @@ -1800,7 +1804,7 @@ static int vcn_v4_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p, return 0; } -static const struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = { +static struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = { .type = AMDGPU_RING_TYPE_VCN_ENC, .align_mask = 0x3f, .nop = VCN_ENC_CMD_NO_OP, @@ -1845,7 +1849,11 @@ static void vcn_v4_0_set_unified_ring_funcs(struct amdgpu_device *adev) if (adev->vcn.harvest_config & (1 << i)) continue; - adev->vcn.inst[i].ring_enc[0].funcs = &vcn_v4_0_unified_ring_vm_funcs; + if (adev->ip_versions[VCN_HWIP][0] == IP_VERSION(4, 0, 2)) + vcn_v4_0_unified_ring_vm_funcs.secure_submission_supported = true; + + adev->vcn.inst[i].ring_enc[0].funcs = + (const struct amdgpu_ring_funcs *)&vcn_v4_0_unified_ring_vm_funcs; adev->vcn.inst[i].ring_enc[0].me = i; DRM_INFO("VCN(%d) encode/decode are enabled in VM mode\n", i); @@ -1910,7 +1918,7 @@ static int vcn_v4_0_wait_for_idle(void *handle) static int vcn_v4_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = state == AMD_CG_STATE_GATE; int i; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -1951,7 +1959,7 @@ static int vcn_v4_0_set_powergating_state(void *handle, enum amd_powergating_sta return 0; } - if(state == adev->vcn.cur_state) + if (state == adev->vcn.cur_state) return 0; if (state == AMD_PG_STATE_GATE) @@ -1959,7 +1967,7 @@ static int vcn_v4_0_set_powergating_state(void *handle, enum amd_powergating_sta else ret = vcn_v4_0_start(adev); - if(!ret) + if (!ret) adev->vcn.cur_state = state; return ret; @@ -2093,8 +2101,7 @@ static const struct amd_ip_funcs vcn_v4_0_ip_funcs = { .set_powergating_state = vcn_v4_0_set_powergating_state, }; -const struct amdgpu_ip_block_version vcn_v4_0_ip_block = -{ +const struct amdgpu_ip_block_version vcn_v4_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_VCN, .major = 4, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 5d67b8b8a3d6..f85d18cd74ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -31,6 +31,7 @@ #include "soc15d.h" #include "soc15_hw_ip.h" #include "vcn_v2_0.h" +#include "mmsch_v4_0_3.h" #include "vcn/vcn_4_0_3_offset.h" #include "vcn/vcn_4_0_3_sh_mask.h" @@ -44,6 +45,7 @@ #define VCN_VID_SOC_ADDRESS_2_0 0x1fb00 #define VCN1_VID_SOC_ADDRESS_3_0 0x48300 +static int vcn_v4_0_3_start_sriov(struct amdgpu_device *adev); static void vcn_v4_0_3_set_unified_ring_funcs(struct amdgpu_device *adev); static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev); static int vcn_v4_0_3_set_powergating_state(void *handle, @@ -111,9 +113,16 @@ static int vcn_v4_0_3_sw_init(void *handle) ring = &adev->vcn.inst[i].ring_enc[0]; ring->use_doorbell = true; - ring->doorbell_index = - (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + - 9 * vcn_inst; + + if (!amdgpu_sriov_vf(adev)) + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 9 * vcn_inst; + else + ring->doorbell_index = + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 32 * vcn_inst; + ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id); sprintf(ring->name, "vcn_unified_%d", adev->vcn.inst[i].aid_id); r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0, @@ -130,6 +139,12 @@ static int vcn_v4_0_3_sw_init(void *handle) amdgpu_vcn_fwlog_init(&adev->vcn.inst[i]); } + if (amdgpu_sriov_vf(adev)) { + r = amdgpu_virt_alloc_mm_table(adev); + if (r) + return r; + } + if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) adev->vcn.pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode; @@ -167,6 +182,9 @@ static int vcn_v4_0_3_sw_fini(void *handle) drm_dev_exit(idx); } + if (amdgpu_sriov_vf(adev)) + amdgpu_virt_free_mm_table(adev); + r = amdgpu_vcn_suspend(adev); if (r) return r; @@ -189,33 +207,47 @@ static int vcn_v4_0_3_hw_init(void *handle) struct amdgpu_ring *ring; int i, r, vcn_inst; - for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { - vcn_inst = GET_INST(VCN, i); - ring = &adev->vcn.inst[i].ring_enc[0]; + if (amdgpu_sriov_vf(adev)) { + r = vcn_v4_0_3_start_sriov(adev); + if (r) + goto done; - if (ring->use_doorbell) { - adev->nbio.funcs->vcn_doorbell_range( - adev, ring->use_doorbell, - (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + - 9 * vcn_inst, - adev->vcn.inst[i].aid_id); - - WREG32_SOC15( - VCN, GET_INST(VCN, ring->me), - regVCN_RB1_DB_CTRL, - ring->doorbell_index - << VCN_RB1_DB_CTRL__OFFSET__SHIFT | - VCN_RB1_DB_CTRL__EN_MASK); - - /* Read DB_CTRL to flush the write DB_CTRL command. */ - RREG32_SOC15( - VCN, GET_INST(VCN, ring->me), - regVCN_RB1_DB_CTRL); + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + ring = &adev->vcn.inst[i].ring_enc[0]; + ring->wptr = 0; + ring->wptr_old = 0; + vcn_v4_0_3_unified_ring_set_wptr(ring); + ring->sched.ready = true; } + } else { + for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { + vcn_inst = GET_INST(VCN, i); + ring = &adev->vcn.inst[i].ring_enc[0]; + + if (ring->use_doorbell) { + adev->nbio.funcs->vcn_doorbell_range( + adev, ring->use_doorbell, + (adev->doorbell_index.vcn.vcn_ring0_1 << 1) + + 9 * vcn_inst, + adev->vcn.inst[i].aid_id); + + WREG32_SOC15( + VCN, GET_INST(VCN, ring->me), + regVCN_RB1_DB_CTRL, + ring->doorbell_index + << VCN_RB1_DB_CTRL__OFFSET__SHIFT | + VCN_RB1_DB_CTRL__EN_MASK); + + /* Read DB_CTRL to flush the write DB_CTRL command. */ + RREG32_SOC15( + VCN, GET_INST(VCN, ring->me), + regVCN_RB1_DB_CTRL); + } - r = amdgpu_ring_test_helper(ring); - if (r) - goto done; + r = amdgpu_ring_test_helper(ring); + if (r) + goto done; + } } done: @@ -778,9 +810,7 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect); if (indirect) - psp_update_vcn_sram(adev, 0, adev->vcn.inst[inst_idx].dpg_sram_gpu_addr, - (uint32_t)((uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_curr_addr - - (uintptr_t)adev->vcn.inst[inst_idx].dpg_sram_cpu_addr)); + amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM); ring = &adev->vcn.inst[inst_idx].ring_enc[0]; @@ -815,6 +845,193 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_device *adev, int inst_idx, b return 0; } +static int vcn_v4_0_3_start_sriov(struct amdgpu_device *adev) +{ + int i, vcn_inst; + struct amdgpu_ring *ring_enc; + uint64_t cache_addr; + uint64_t rb_enc_addr; + uint64_t ctx_addr; + uint32_t param, resp, expected; + uint32_t offset, cache_size; + uint32_t tmp, timeout; + + struct amdgpu_mm_table *table = &adev->virt.mm_table; + uint32_t *table_loc; + uint32_t table_size; + uint32_t size, size_dw; + uint32_t init_status; + uint32_t enabled_vcn; + + struct mmsch_v4_0_cmd_direct_write + direct_wt = { {0} }; + struct mmsch_v4_0_cmd_direct_read_modify_write + direct_rd_mod_wt = { {0} }; + struct mmsch_v4_0_cmd_end end = { {0} }; + struct mmsch_v4_0_3_init_header header; + + volatile struct amdgpu_vcn4_fw_shared *fw_shared; + volatile struct amdgpu_fw_shared_rb_setup *rb_setup; + + direct_wt.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_WRITE; + direct_rd_mod_wt.cmd_header.command_type = + MMSCH_COMMAND__DIRECT_REG_READ_MODIFY_WRITE; + end.cmd_header.command_type = MMSCH_COMMAND__END; + + for (i = 0; i < adev->vcn.num_vcn_inst; i++) { + vcn_inst = GET_INST(VCN, i); + + memset(&header, 0, sizeof(struct mmsch_v4_0_3_init_header)); + header.version = MMSCH_VERSION; + header.total_size = sizeof(struct mmsch_v4_0_3_init_header) >> 2; + + table_loc = (uint32_t *)table->cpu_addr; + table_loc += header.total_size; + + table_size = 0; + + MMSCH_V4_0_INSERT_DIRECT_RD_MOD_WT(SOC15_REG_OFFSET(VCN, 0, regUVD_STATUS), + ~UVD_STATUS__UVD_BUSY, UVD_STATUS__UVD_BUSY); + + cache_size = AMDGPU_GPU_PAGE_ALIGN(adev->vcn.fw->size + 4); + + if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) { + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_lo); + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + adev->firmware.ucode[AMDGPU_UCODE_ID_VCN + i].tmr_mc_addr_hi); + + offset = 0; + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_OFFSET0), 0); + } else { + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[i].gpu_addr)); + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[i].gpu_addr)); + offset = cache_size; + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_OFFSET0), + AMDGPU_UVD_FIRMWARE_OFFSET >> 3); + } + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_SIZE0), + cache_size); + + cache_addr = adev->vcn.inst[vcn_inst].gpu_addr + offset; + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE1_64BIT_BAR_LOW), lower_32_bits(cache_addr)); + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE1_64BIT_BAR_HIGH), upper_32_bits(cache_addr)); + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_OFFSET1), 0); + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_SIZE1), AMDGPU_VCN_STACK_SIZE); + + cache_addr = adev->vcn.inst[vcn_inst].gpu_addr + offset + + AMDGPU_VCN_STACK_SIZE; + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE2_64BIT_BAR_LOW), lower_32_bits(cache_addr)); + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_CACHE2_64BIT_BAR_HIGH), upper_32_bits(cache_addr)); + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_OFFSET2), 0); + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_CACHE_SIZE2), AMDGPU_VCN_CONTEXT_SIZE); + + fw_shared = adev->vcn.inst[vcn_inst].fw_shared.cpu_addr; + rb_setup = &fw_shared->rb_setup; + + ring_enc = &adev->vcn.inst[vcn_inst].ring_enc[0]; + ring_enc->wptr = 0; + rb_enc_addr = ring_enc->gpu_addr; + + rb_setup->is_rb_enabled_flags |= RB_ENABLED; + rb_setup->rb_addr_lo = lower_32_bits(rb_enc_addr); + rb_setup->rb_addr_hi = upper_32_bits(rb_enc_addr); + rb_setup->rb_size = ring_enc->ring_size / 4; + fw_shared->present_flag_0 |= cpu_to_le32(AMDGPU_VCN_VF_RB_SETUP_FLAG); + + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_NC0_64BIT_BAR_LOW), + lower_32_bits(adev->vcn.inst[vcn_inst].fw_shared.gpu_addr)); + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_LMI_VCPU_NC0_64BIT_BAR_HIGH), + upper_32_bits(adev->vcn.inst[vcn_inst].fw_shared.gpu_addr)); + MMSCH_V4_0_INSERT_DIRECT_WT(SOC15_REG_OFFSET(VCN, 0, + regUVD_VCPU_NONCACHE_SIZE0), + AMDGPU_GPU_PAGE_ALIGN(sizeof(struct amdgpu_vcn4_fw_shared))); + MMSCH_V4_0_INSERT_END(); + + header.vcn0.init_status = 0; + header.vcn0.table_offset = header.total_size; + header.vcn0.table_size = table_size; + header.total_size += table_size; + + /* Send init table to mmsch */ + size = sizeof(struct mmsch_v4_0_3_init_header); + table_loc = (uint32_t *)table->cpu_addr; + memcpy((void *)table_loc, &header, size); + + ctx_addr = table->gpu_addr; + WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_CTX_ADDR_LO, lower_32_bits(ctx_addr)); + WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_CTX_ADDR_HI, upper_32_bits(ctx_addr)); + + tmp = RREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_VMID); + tmp &= ~MMSCH_VF_VMID__VF_CTX_VMID_MASK; + tmp |= (0 << MMSCH_VF_VMID__VF_CTX_VMID__SHIFT); + WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_VMID, tmp); + + size = header.total_size; + WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_CTX_SIZE, size); + + WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_MAILBOX_RESP, 0); + + param = 0x00000001; + WREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_MAILBOX_HOST, param); + tmp = 0; + timeout = 1000; + resp = 0; + expected = MMSCH_VF_MAILBOX_RESP__OK; + while (resp != expected) { + resp = RREG32_SOC15(VCN, vcn_inst, regMMSCH_VF_MAILBOX_RESP); + if (resp != 0) + break; + + udelay(10); + tmp = tmp + 10; + if (tmp >= timeout) { + DRM_ERROR("failed to init MMSCH. TIME-OUT after %d usec"\ + " waiting for regMMSCH_VF_MAILBOX_RESP "\ + "(expected=0x%08x, readback=0x%08x)\n", + tmp, expected, resp); + return -EBUSY; + } + } + + enabled_vcn = amdgpu_vcn_is_disabled_vcn(adev, VCN_DECODE_RING, 0) ? 1 : 0; + init_status = ((struct mmsch_v4_0_3_init_header *)(table_loc))->vcn0.init_status; + if (resp != expected && resp != MMSCH_VF_MAILBOX_RESP__INCOMPLETE + && init_status != MMSCH_VF_ENGINE_STATUS__PASS) { + DRM_ERROR("MMSCH init status is incorrect! readback=0x%08x, header init "\ + "status for VCN%x: 0x%x\n", resp, enabled_vcn, init_status); + } + } + + return 0; +} + /** * vcn_v4_0_3_start - VCN start * @@ -1289,7 +1506,7 @@ static int vcn_v4_0_3_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE) ? true : false; + bool enable = state == AMD_CG_STATE_GATE; int i; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -1319,6 +1536,15 @@ static int vcn_v4_0_3_set_powergating_state(void *handle, struct amdgpu_device *adev = (struct amdgpu_device *)handle; int ret; + /* for SRIOV, guest should not control VCN Power-gating + * MMSCH FW should control Power-gating and clock-gating + * guest should avoid touching CGC and PG + */ + if (amdgpu_sriov_vf(adev)) { + adev->vcn.cur_state = AMD_PG_STATE_UNGATE; + return 0; + } + if (state == adev->vcn.cur_state) return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index 1e83db0c5438..d364c6dd152c 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -485,7 +485,7 @@ static int vega10_ih_sw_init(void *handle) if (r) return r; - r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, true); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, true); if (r) return r; @@ -510,7 +510,7 @@ static int vega10_ih_sw_init(void *handle) /* initialize ih control registers offset */ vega10_ih_init_register_offset(adev); - r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, true); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, true); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c index 4d719df376a7..dbc99536440f 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c @@ -500,7 +500,8 @@ static int vega20_ih_self_irq(struct amdgpu_device *adev, case 2: schedule_work(&adev->irq.ih2_work); break; - default: break; + default: + break; } return 0; } @@ -539,7 +540,7 @@ static int vega20_ih_sw_init(void *handle) (adev->ip_versions[OSSSYS_HWIP][0] == IP_VERSION(4, 4, 2))) use_bus_addr = false; - r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr); if (r) return r; @@ -565,7 +566,7 @@ static int vega20_ih_sw_init(void *handle) /* initialize ih control registers offset */ vega20_ih_init_register_offset(adev); - r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, PAGE_SIZE, use_bus_addr); + r = amdgpu_ih_ring_init(adev, &adev->irq.ih_soft, IH_SW_RING_SIZE, use_bus_addr); if (r) return r; @@ -710,8 +711,7 @@ static void vega20_ih_set_interrupt_funcs(struct amdgpu_device *adev) adev->irq.ih_funcs = &vega20_ih_funcs; } -const struct amdgpu_ip_block_version vega20_ih_ip_block = -{ +const struct amdgpu_ip_block_version vega20_ih_ip_block = { .type = AMD_IP_BLOCK_TYPE_IH, .major = 4, .minor = 2, diff --git a/drivers/gpu/drm/amd/amdkfd/Kconfig b/drivers/gpu/drm/amd/amdkfd/Kconfig index 93bd4eda0d94..d3c3d3ab7225 100644 --- a/drivers/gpu/drm/amd/amdkfd/Kconfig +++ b/drivers/gpu/drm/amd/amdkfd/Kconfig @@ -6,7 +6,6 @@ config HSA_AMD bool "HSA kernel driver for AMD GPU devices" depends on DRM_AMDGPU && (X86_64 || ARM64 || PPC64) - imply AMD_IOMMU_V2 if X86_64 select HMM_MIRROR select MMU_NOTIFIER select DRM_AMDGPU_USERPTR diff --git a/drivers/gpu/drm/amd/amdkfd/Makefile b/drivers/gpu/drm/amd/amdkfd/Makefile index 2ec8f27c5366..a5ae7bcf44eb 100644 --- a/drivers/gpu/drm/amd/amdkfd/Makefile +++ b/drivers/gpu/drm/amd/amdkfd/Makefile @@ -59,10 +59,6 @@ AMDKFD_FILES := $(AMDKFD_PATH)/kfd_module.o \ $(AMDKFD_PATH)/kfd_crat.o \ $(AMDKFD_PATH)/kfd_debug.o -ifneq ($(CONFIG_AMD_IOMMU_V2),) -AMDKFD_FILES += $(AMDKFD_PATH)/kfd_iommu.o -endif - ifneq ($(CONFIG_DEBUG_FS),) AMDKFD_FILES += $(AMDKFD_PATH)/kfd_debugfs.o endif diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h index 73ca9aebf086..d7cd5fa313ff 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h @@ -274,16 +274,16 @@ static const uint32_t cwsr_trap_gfx8_hex[] = { static const uint32_t cwsr_trap_gfx9_hex[] = { - 0xbf820001, 0xbf820254, + 0xbf820001, 0xbf820258, 0xb8f8f802, 0x8978ff78, 0x00020006, 0xb8fbf803, 0x866eff78, 0x00002000, 0xbf840009, 0x866eff6d, 0x00ff0000, 0xbf85001e, 0x866eff7b, 0x00000400, - 0xbf850051, 0xbf8e0010, + 0xbf850055, 0xbf8e0010, 0xb8fbf803, 0xbf82fffa, - 0x866eff7b, 0x00000900, + 0x866eff7b, 0x03c00900, 0xbf850015, 0x866eff7b, 0x000071ff, 0xbf840008, 0x866fff7b, 0x00007080, @@ -294,13 +294,15 @@ static const uint32_t cwsr_trap_gfx9_hex[] = { 0xbf850007, 0xb8eef801, 0x866eff6e, 0x00000800, 0xbf850003, 0x866eff7b, - 0x00000400, 0xbf850036, + 0x00000400, 0xbf85003a, 0xb8faf807, 0x867aff7a, 0x001f8000, 0x8e7a8b7a, 0x8977ff77, 0xfc000000, 0x87777a77, 0xba7ff807, 0x00000000, 0xb8faf812, 0xb8fbf813, 0x8efa887a, + 0xbf0d8f7b, 0xbf840002, + 0x877bff7b, 0xffff0000, 0xc0031bbd, 0x00000010, 0xbf8cc07f, 0x8e6e976e, 0x8977ff77, 0x00800000, @@ -676,14 +678,14 @@ static const uint32_t cwsr_trap_gfx9_hex[] = { }; static const uint32_t cwsr_trap_nv1x_hex[] = { - 0xbf820001, 0xbf8201f1, + 0xbf820001, 0xbf8201f5, 0xb0804004, 0xb978f802, 0x8a78ff78, 0x00020006, 0xb97bf803, 0x876eff78, 0x00002000, 0xbf840009, 0x876eff6d, 0x00ff0000, 0xbf85001e, 0x876eff7b, - 0x00000400, 0xbf850057, + 0x00000400, 0xbf85005b, 0xbf8e0010, 0xb97bf803, 0xbf82fffa, 0x876eff7b, 0x00000900, 0xbf850015, @@ -697,7 +699,7 @@ static const uint32_t cwsr_trap_nv1x_hex[] = { 0xb96ef801, 0x876eff6e, 0x00000800, 0xbf850003, 0x876eff7b, 0x00000400, - 0xbf85003c, 0x8a77ff77, + 0xbf850040, 0x8a77ff77, 0xff000000, 0xb97af807, 0x877bff7a, 0x02000000, 0x8f7b867b, 0x88777b77, @@ -706,6 +708,8 @@ static const uint32_t cwsr_trap_nv1x_hex[] = { 0x8a7aff7a, 0x023f8000, 0xb9faf807, 0xb97af812, 0xb97bf813, 0x8ffa887a, + 0xbf0d8f7b, 0xbf840002, + 0x887bff7b, 0xffff0000, 0xf4011bbd, 0xfa000010, 0xbf8cc07f, 0x8f6e976e, 0x8a77ff77, 0x00800000, @@ -1094,16 +1098,16 @@ static const uint32_t cwsr_trap_nv1x_hex[] = { }; static const uint32_t cwsr_trap_arcturus_hex[] = { - 0xbf820001, 0xbf8202d0, + 0xbf820001, 0xbf8202d4, 0xb8f8f802, 0x8978ff78, 0x00020006, 0xb8fbf803, 0x866eff78, 0x00002000, 0xbf840009, 0x866eff6d, 0x00ff0000, 0xbf85001e, 0x866eff7b, 0x00000400, - 0xbf850051, 0xbf8e0010, + 0xbf850055, 0xbf8e0010, 0xb8fbf803, 0xbf82fffa, - 0x866eff7b, 0x00000900, + 0x866eff7b, 0x03c00900, 0xbf850015, 0x866eff7b, 0x000071ff, 0xbf840008, 0x866fff7b, 0x00007080, @@ -1114,13 +1118,15 @@ static const uint32_t cwsr_trap_arcturus_hex[] = { 0xbf850007, 0xb8eef801, 0x866eff6e, 0x00000800, 0xbf850003, 0x866eff7b, - 0x00000400, 0xbf850036, + 0x00000400, 0xbf85003a, 0xb8faf807, 0x867aff7a, 0x001f8000, 0x8e7a8b7a, 0x8977ff77, 0xfc000000, 0x87777a77, 0xba7ff807, 0x00000000, 0xb8faf812, 0xb8fbf813, 0x8efa887a, + 0xbf0d8f7b, 0xbf840002, + 0x877bff7b, 0xffff0000, 0xc0031bbd, 0x00000010, 0xbf8cc07f, 0x8e6e976e, 0x8977ff77, 0x00800000, @@ -1572,16 +1578,16 @@ static const uint32_t cwsr_trap_arcturus_hex[] = { }; static const uint32_t cwsr_trap_aldebaran_hex[] = { - 0xbf820001, 0xbf8202db, + 0xbf820001, 0xbf8202df, 0xb8f8f802, 0x8978ff78, 0x00020006, 0xb8fbf803, 0x866eff78, 0x00002000, 0xbf840009, 0x866eff6d, 0x00ff0000, 0xbf85001e, 0x866eff7b, 0x00000400, - 0xbf850051, 0xbf8e0010, + 0xbf850055, 0xbf8e0010, 0xb8fbf803, 0xbf82fffa, - 0x866eff7b, 0x00000900, + 0x866eff7b, 0x03c00900, 0xbf850015, 0x866eff7b, 0x000071ff, 0xbf840008, 0x866fff7b, 0x00007080, @@ -1592,13 +1598,15 @@ static const uint32_t cwsr_trap_aldebaran_hex[] = { 0xbf850007, 0xb8eef801, 0x866eff6e, 0x00000800, 0xbf850003, 0x866eff7b, - 0x00000400, 0xbf850036, + 0x00000400, 0xbf85003a, 0xb8faf807, 0x867aff7a, 0x001f8000, 0x8e7a8b7a, 0x8977ff77, 0xfc000000, 0x87777a77, 0xba7ff807, 0x00000000, 0xb8faf812, 0xb8fbf813, 0x8efa887a, + 0xbf0d8f7b, 0xbf840002, + 0x877bff7b, 0xffff0000, 0xc0031bbd, 0x00000010, 0xbf8cc07f, 0x8e6e976e, 0x8977ff77, 0x00800000, @@ -2061,14 +2069,14 @@ static const uint32_t cwsr_trap_aldebaran_hex[] = { }; static const uint32_t cwsr_trap_gfx10_hex[] = { - 0xbf820001, 0xbf82021c, + 0xbf820001, 0xbf820220, 0xb0804004, 0xb978f802, 0x8a78ff78, 0x00020006, 0xb97bf803, 0x876eff78, 0x00002000, 0xbf840009, 0x876eff6d, 0x00ff0000, 0xbf85001e, 0x876eff7b, - 0x00000400, 0xbf850041, + 0x00000400, 0xbf850045, 0xbf8e0010, 0xb97bf803, 0xbf82fffa, 0x876eff7b, 0x00000900, 0xbf850015, @@ -2082,8 +2090,10 @@ static const uint32_t cwsr_trap_gfx10_hex[] = { 0xb96ef801, 0x876eff6e, 0x00000800, 0xbf850003, 0x876eff7b, 0x00000400, - 0xbf850026, 0xb97af812, + 0xbf85002a, 0xb97af812, 0xb97bf813, 0x8ffa887a, + 0xbf0d8f7b, 0xbf840002, + 0x887bff7b, 0xffff0000, 0xf4011bbd, 0xfa000010, 0xbf8cc07f, 0x8f6e976e, 0x8a77ff77, 0x00800000, @@ -2494,8 +2504,9 @@ static const uint32_t cwsr_trap_gfx10_hex[] = { 0xbf9f0000, 0xbf9f0000, 0xbf9f0000, 0x00000000, }; + static const uint32_t cwsr_trap_gfx11_hex[] = { - 0xbfa00001, 0xbfa00221, + 0xbfa00001, 0xbfa00225, 0xb0804006, 0xb8f8f802, 0x9178ff78, 0x00020006, 0xb8fbf803, 0xbf0d9e6d, @@ -2505,7 +2516,7 @@ static const uint32_t cwsr_trap_gfx11_hex[] = { 0xbfa10009, 0x8b6eff6d, 0x00ff0000, 0xbfa2001e, 0x8b6eff7b, 0x00000400, - 0xbfa20041, 0xbf830010, + 0xbfa20045, 0xbf830010, 0xb8fbf803, 0xbfa0fffa, 0x8b6eff7b, 0x00000900, 0xbfa20015, 0x8b6eff7b, @@ -2518,9 +2529,11 @@ static const uint32_t cwsr_trap_gfx11_hex[] = { 0xbfa20007, 0xb8eef801, 0x8b6eff6e, 0x00000800, 0xbfa20003, 0x8b6eff7b, - 0x00000400, 0xbfa20026, + 0x00000400, 0xbfa2002a, 0xbefa4d82, 0xbf89fc07, - 0x84fa887a, 0xf4005bbd, + 0x84fa887a, 0xbf0d8f7b, + 0xbfa10002, 0x8c7bff7b, + 0xffff0000, 0xf4005bbd, 0xf8000010, 0xbf89fc07, 0x846e976e, 0x9177ff77, 0x00800000, 0x8c776e77, @@ -2938,211 +2951,257 @@ static const uint32_t cwsr_trap_gfx11_hex[] = { }; static const uint32_t cwsr_trap_gfx9_4_3_hex[] = { - 0xbf820001, 0xbf8202d6, - 0xb8f8f802, 0x89788678, - 0xb8fbf803, 0x866eff78, - 0x00002000, 0xbf840009, - 0x866eff6d, 0x00ff0000, - 0xbf85001a, 0x866eff7b, - 0x00000400, 0xbf85004d, - 0xbf8e0010, 0xb8fbf803, - 0xbf82fffa, 0x866eff7b, - 0x03c00900, 0xbf850011, - 0x866eff7b, 0x000071ff, - 0xbf840008, 0x866fff7b, - 0x00007080, 0xbf840001, - 0xbeee1a87, 0xb8eff801, - 0x8e6e8c6e, 0x866e6f6e, - 0xbf850006, 0x866eff6d, - 0x00ff0000, 0xbf850003, + 0xbf820001, 0xbf8202db, + 0xb8f8f802, 0x8978ff78, + 0x00020006, 0xb8fbf803, + 0x866eff78, 0x00002000, + 0xbf840009, 0x866eff6d, + 0x00ff0000, 0xbf85001a, 0x866eff7b, 0x00000400, - 0xbf850036, 0xb8faf807, + 0xbf850051, 0xbf8e0010, + 0xb8fbf803, 0xbf82fffa, + 0x866eff7b, 0x03c00900, + 0xbf850011, 0x866eff7b, + 0x000071ff, 0xbf840008, + 0x866fff7b, 0x00007080, + 0xbf840001, 0xbeee1a87, + 0xb8eff801, 0x8e6e8c6e, + 0x866e6f6e, 0xbf850006, + 0x866eff6d, 0x00ff0000, + 0xbf850003, 0x866eff7b, + 0x00000400, 0xbf85003a, + 0xb8faf807, 0x867aff7a, + 0x001f8000, 0x8e7a8b7a, + 0x8979ff79, 0xfc000000, + 0x87797a79, 0xba7ff807, + 0x00000000, 0xb8faf812, + 0xb8fbf813, 0x8efa887a, + 0xbf0d8f7b, 0xbf840002, + 0x877bff7b, 0xffff0000, + 0xc0031bbd, 0x00000010, + 0xbf8cc07f, 0x8e6e976e, + 0x8979ff79, 0x00800000, + 0x87796e79, 0xc0071bbd, + 0x00000000, 0xbf8cc07f, + 0xc0071ebd, 0x00000008, + 0xbf8cc07f, 0x86ee6e6e, + 0xbf840001, 0xbe801d6e, + 0x866eff6d, 0x01ff0000, + 0xbf850005, 0x8778ff78, + 0x00002000, 0x80ec886c, + 0x82ed806d, 0xbf820005, + 0x866eff6d, 0x01000000, + 0xbf850002, 0x806c846c, + 0x826d806d, 0x866dff6d, + 0x0000ffff, 0x8f7a8b79, 0x867aff7a, 0x001f8000, - 0x8e7a8b7a, 0x8979ff79, - 0xfc000000, 0x87797a79, - 0xba7ff807, 0x00000000, - 0xb8faf812, 0xb8fbf813, - 0x8efa887a, 0xc0031bbd, - 0x00000010, 0xbf8cc07f, - 0x8e6e976e, 0x8979ff79, - 0x00800000, 0x87796e79, - 0xc0071bbd, 0x00000000, - 0xbf8cc07f, 0xc0071ebd, - 0x00000008, 0xbf8cc07f, - 0x86ee6e6e, 0xbf840001, - 0xbe801d6e, 0x866eff6d, - 0x01ff0000, 0xbf850005, - 0x8778ff78, 0x00002000, - 0x80ec886c, 0x82ed806d, - 0xbf820005, 0x866eff6d, - 0x01000000, 0xbf850002, - 0x806c846c, 0x826d806d, + 0xb97af807, 0x86fe7e7e, + 0x86ea6a6a, 0x8f6e8378, + 0xb96ee0c2, 0xbf800002, + 0xb9780002, 0xbe801f6c, 0x866dff6d, 0x0000ffff, - 0x8f7a8b79, 0x867aff7a, - 0x001f8000, 0xb97af807, - 0x86fe7e7e, 0x86ea6a6a, - 0x8f6e8378, 0xb96ee0c2, - 0xbf800002, 0xb9780002, - 0xbe801f6c, 0x866dff6d, - 0x0000ffff, 0xbefa0080, - 0xb97a0283, 0xb8faf807, - 0x867aff7a, 0x001f8000, - 0x8e7a8b7a, 0x8979ff79, - 0xfc000000, 0x87797a79, - 0xba7ff807, 0x00000000, - 0xbeee007e, 0xbeef007f, - 0xbefe0180, 0xbf900004, - 0x877a8478, 0xb97af802, - 0xbf8e0002, 0xbf88fffe, - 0xb8fa2985, 0x807a817a, - 0x8e7a8a7a, 0x8e7a817a, - 0xb8fb1605, 0x807b817b, - 0x8e7b867b, 0x807a7b7a, - 0x807a7e7a, 0x827b807f, - 0x867bff7b, 0x0000ffff, - 0xc04b1c3d, 0x00000050, - 0xbf8cc07f, 0xc04b1d3d, - 0x00000060, 0xbf8cc07f, - 0xc0431e7d, 0x00000074, - 0xbf8cc07f, 0xbef4007e, - 0x8675ff7f, 0x0000ffff, - 0x8775ff75, 0x00040000, - 0xbef60080, 0xbef700ff, - 0x00807fac, 0xbef1007c, - 0xbef00080, 0xb8f02985, - 0x80708170, 0x8e708a70, - 0x8e708170, 0xb8fa1605, - 0x807a817a, 0x8e7a867a, - 0x80707a70, 0xbef60084, - 0xbef600ff, 0x01000000, - 0xbefe007c, 0xbefc0070, - 0xc0611c7a, 0x0000007c, - 0xbf8cc07f, 0x80708470, - 0xbefc007e, 0xbefe007c, - 0xbefc0070, 0xc0611b3a, + 0xbefa0080, 0xb97a0283, + 0xb8faf807, 0x867aff7a, + 0x001f8000, 0x8e7a8b7a, + 0x8979ff79, 0xfc000000, + 0x87797a79, 0xba7ff807, + 0x00000000, 0xbeee007e, + 0xbeef007f, 0xbefe0180, + 0xbf900004, 0x877a8478, + 0xb97af802, 0xbf8e0002, + 0xbf88fffe, 0xb8fa2985, + 0x807a817a, 0x8e7a8a7a, + 0x8e7a817a, 0xb8fb1605, + 0x807b817b, 0x8e7b867b, + 0x807a7b7a, 0x807a7e7a, + 0x827b807f, 0x867bff7b, + 0x0000ffff, 0xc04b1c3d, + 0x00000050, 0xbf8cc07f, + 0xc04b1d3d, 0x00000060, + 0xbf8cc07f, 0xc0431e7d, + 0x00000074, 0xbf8cc07f, + 0xbef4007e, 0x8675ff7f, + 0x0000ffff, 0x8775ff75, + 0x00040000, 0xbef60080, + 0xbef700ff, 0x00807fac, + 0xbef1007c, 0xbef00080, + 0xb8f02985, 0x80708170, + 0x8e708a70, 0x8e708170, + 0xb8fa1605, 0x807a817a, + 0x8e7a867a, 0x80707a70, + 0xbef60084, 0xbef600ff, + 0x01000000, 0xbefe007c, + 0xbefc0070, 0xc0611c7a, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, 0xbefe007c, 0xbefc0070, - 0xc0611b7a, 0x0000007c, + 0xc0611b3a, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, 0xbefe007c, - 0xbefc0070, 0xc0611bba, + 0xbefc0070, 0xc0611b7a, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, 0xbefe007c, 0xbefc0070, - 0xc0611bfa, 0x0000007c, + 0xc0611bba, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, 0xbefe007c, - 0xbefc0070, 0xc0611e3a, - 0x0000007c, 0xbf8cc07f, - 0x80708470, 0xbefc007e, - 0xb8fbf803, 0xbefe007c, - 0xbefc0070, 0xc0611efa, + 0xbefc0070, 0xc0611bfa, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, 0xbefe007c, 0xbefc0070, - 0xc0611a3a, 0x0000007c, + 0xc0611e3a, 0x0000007c, + 0xbf8cc07f, 0x80708470, + 0xbefc007e, 0xb8fbf803, + 0xbefe007c, 0xbefc0070, + 0xc0611efa, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, 0xbefe007c, - 0xbefc0070, 0xc0611a7a, - 0x0000007c, 0xbf8cc07f, - 0x80708470, 0xbefc007e, - 0xb8f1f801, 0xbefe007c, - 0xbefc0070, 0xc0611c7a, + 0xbefc0070, 0xc0611a3a, 0x0000007c, 0xbf8cc07f, 0x80708470, 0xbefc007e, - 0x867aff7f, 0x04000000, - 0xbeef0080, 0x876f6f7a, - 0xb8f02985, 0x80708170, - 0x8e708a70, 0x8e708170, - 0xb8fb1605, 0x807b817b, - 0x8e7b847b, 0x8e76827b, - 0xbef600ff, 0x01000000, - 0xbef20174, 0x80747074, - 0x82758075, 0xbefc0080, - 0xbf800000, 0xbe802b00, - 0xbe822b02, 0xbe842b04, - 0xbe862b06, 0xbe882b08, - 0xbe8a2b0a, 0xbe8c2b0c, - 0xbe8e2b0e, 0xc06b003a, - 0x00000000, 0xbf8cc07f, - 0xc06b013a, 0x00000010, - 0xbf8cc07f, 0xc06b023a, - 0x00000020, 0xbf8cc07f, - 0xc06b033a, 0x00000030, - 0xbf8cc07f, 0x8074c074, - 0x82758075, 0x807c907c, - 0xbf0a7b7c, 0xbf85ffe7, - 0xbef40172, 0xbef00080, - 0xbefe00c1, 0xbeff00c1, - 0xbee80080, 0xbee90080, - 0xbef600ff, 0x01000000, - 0x867aff78, 0x00400000, - 0xbf850003, 0xb8faf803, - 0x897a7aff, 0x10000000, - 0xbf85004d, 0xbe840080, - 0xd2890000, 0x00000900, - 0x80048104, 0xd2890001, + 0xbefe007c, 0xbefc0070, + 0xc0611a7a, 0x0000007c, + 0xbf8cc07f, 0x80708470, + 0xbefc007e, 0xb8f1f801, + 0xbefe007c, 0xbefc0070, + 0xc0611c7a, 0x0000007c, + 0xbf8cc07f, 0x80708470, + 0xbefc007e, 0x867aff7f, + 0x04000000, 0xbeef0080, + 0x876f6f7a, 0xb8f02985, + 0x80708170, 0x8e708a70, + 0x8e708170, 0xb8fb1605, + 0x807b817b, 0x8e7b847b, + 0x8e76827b, 0xbef600ff, + 0x01000000, 0xbef20174, + 0x80747074, 0x82758075, + 0xbefc0080, 0xbf800000, + 0xbe802b00, 0xbe822b02, + 0xbe842b04, 0xbe862b06, + 0xbe882b08, 0xbe8a2b0a, + 0xbe8c2b0c, 0xbe8e2b0e, + 0xc06b003a, 0x00000000, + 0xbf8cc07f, 0xc06b013a, + 0x00000010, 0xbf8cc07f, + 0xc06b023a, 0x00000020, + 0xbf8cc07f, 0xc06b033a, + 0x00000030, 0xbf8cc07f, + 0x8074c074, 0x82758075, + 0x807c907c, 0xbf0a7b7c, + 0xbf85ffe7, 0xbef40172, + 0xbef00080, 0xbefe00c1, + 0xbeff00c1, 0xbee80080, + 0xbee90080, 0xbef600ff, + 0x01000000, 0x867aff78, + 0x00400000, 0xbf850003, + 0xb8faf803, 0x897a7aff, + 0x10000000, 0xbf85004d, + 0xbe840080, 0xd2890000, 0x00000900, 0x80048104, - 0xd2890002, 0x00000900, - 0x80048104, 0xd2890003, + 0xd2890001, 0x00000900, + 0x80048104, 0xd2890002, 0x00000900, 0x80048104, + 0xd2890003, 0x00000900, + 0x80048104, 0xc069003a, + 0x00000070, 0xbf8cc07f, + 0x80709070, 0xbf06c004, + 0xbf84ffee, 0xbe840080, + 0xd2890000, 0x00000901, + 0x80048104, 0xd2890001, + 0x00000901, 0x80048104, + 0xd2890002, 0x00000901, + 0x80048104, 0xd2890003, + 0x00000901, 0x80048104, 0xc069003a, 0x00000070, 0xbf8cc07f, 0x80709070, 0xbf06c004, 0xbf84ffee, 0xbe840080, 0xd2890000, - 0x00000901, 0x80048104, - 0xd2890001, 0x00000901, + 0x00000902, 0x80048104, + 0xd2890001, 0x00000902, 0x80048104, 0xd2890002, - 0x00000901, 0x80048104, - 0xd2890003, 0x00000901, + 0x00000902, 0x80048104, + 0xd2890003, 0x00000902, 0x80048104, 0xc069003a, 0x00000070, 0xbf8cc07f, 0x80709070, 0xbf06c004, 0xbf84ffee, 0xbe840080, - 0xd2890000, 0x00000902, + 0xd2890000, 0x00000903, 0x80048104, 0xd2890001, - 0x00000902, 0x80048104, - 0xd2890002, 0x00000902, + 0x00000903, 0x80048104, + 0xd2890002, 0x00000903, 0x80048104, 0xd2890003, - 0x00000902, 0x80048104, + 0x00000903, 0x80048104, + 0xc069003a, 0x00000070, + 0xbf8cc07f, 0x80709070, + 0xbf06c004, 0xbf84ffee, + 0xbf820008, 0xe0724000, + 0x701d0000, 0xe0724100, + 0x701d0100, 0xe0724200, + 0x701d0200, 0xe0724300, + 0x701d0300, 0xbefe00c1, + 0xbeff00c1, 0xb8fb4306, + 0x867bc17b, 0xbf840064, + 0xbf8a0000, 0x867aff6f, + 0x04000000, 0xbf840060, + 0x8e7b867b, 0x8e7b827b, + 0xbef6007b, 0xb8f02985, + 0x80708170, 0x8e708a70, + 0x8e708170, 0xb8fa1605, + 0x807a817a, 0x8e7a867a, + 0x80707a70, 0x8070ff70, + 0x00000080, 0xbef600ff, + 0x01000000, 0xbefc0080, + 0xd28c0002, 0x000100c1, + 0xd28d0003, 0x000204c1, + 0x867aff78, 0x00400000, + 0xbf850003, 0xb8faf803, + 0x897a7aff, 0x10000000, + 0xbf850030, 0x24040682, + 0xd86e4000, 0x00000002, + 0xbf8cc07f, 0xbe840080, + 0xd2890000, 0x00000900, + 0x80048104, 0xd2890001, + 0x00000900, 0x80048104, + 0xd2890002, 0x00000900, + 0x80048104, 0xd2890003, + 0x00000900, 0x80048104, 0xc069003a, 0x00000070, 0xbf8cc07f, 0x80709070, 0xbf06c004, 0xbf84ffee, 0xbe840080, 0xd2890000, - 0x00000903, 0x80048104, - 0xd2890001, 0x00000903, + 0x00000901, 0x80048104, + 0xd2890001, 0x00000901, 0x80048104, 0xd2890002, - 0x00000903, 0x80048104, - 0xd2890003, 0x00000903, + 0x00000901, 0x80048104, + 0xd2890003, 0x00000901, 0x80048104, 0xc069003a, 0x00000070, 0xbf8cc07f, 0x80709070, 0xbf06c004, - 0xbf84ffee, 0xbf820008, - 0xe0724000, 0x701d0000, - 0xe0724100, 0x701d0100, - 0xe0724200, 0x701d0200, - 0xe0724300, 0x701d0300, + 0xbf84ffee, 0x680404ff, + 0x00000200, 0xd0c9006a, + 0x0000f702, 0xbf87ffd2, + 0xbf820015, 0xd1060002, + 0x00011103, 0x7e0602ff, + 0x00000200, 0xbefc00ff, + 0x00010000, 0xbe800077, + 0x8677ff77, 0xff7fffff, + 0x8777ff77, 0x00058000, + 0xd8ec0000, 0x00000002, + 0xbf8cc07f, 0xe0765000, + 0x701d0002, 0x68040702, + 0xd0c9006a, 0x0000f702, + 0xbf87fff7, 0xbef70000, + 0xbef000ff, 0x00000400, 0xbefe00c1, 0xbeff00c1, - 0xb8fb4306, 0x867bc17b, - 0xbf840064, 0xbf8a0000, - 0x867aff6f, 0x04000000, - 0xbf840060, 0x8e7b867b, - 0x8e7b827b, 0xbef6007b, - 0xb8f02985, 0x80708170, - 0x8e708a70, 0x8e708170, - 0xb8fa1605, 0x807a817a, - 0x8e7a867a, 0x80707a70, - 0x8070ff70, 0x00000080, - 0xbef600ff, 0x01000000, - 0xbefc0080, 0xd28c0002, - 0x000100c1, 0xd28d0003, - 0x000204c1, 0x867aff78, + 0xb8fb2b05, 0x807b817b, + 0x8e7b827b, 0xbef600ff, + 0x01000000, 0xbefc0084, + 0xbf0a7b7c, 0xbf84006d, + 0xbf11017c, 0x807bff7b, + 0x00001000, 0x867aff78, 0x00400000, 0xbf850003, 0xb8faf803, 0x897a7aff, - 0x10000000, 0xbf850030, - 0x24040682, 0xd86e4000, - 0x00000002, 0xbf8cc07f, + 0x10000000, 0xbf850051, 0xbe840080, 0xd2890000, 0x00000900, 0x80048104, 0xd2890001, 0x00000900, @@ -3162,31 +3221,51 @@ static const uint32_t cwsr_trap_gfx9_4_3_hex[] = { 0xc069003a, 0x00000070, 0xbf8cc07f, 0x80709070, 0xbf06c004, 0xbf84ffee, - 0x680404ff, 0x00000200, - 0xd0c9006a, 0x0000f702, - 0xbf87ffd2, 0xbf820015, - 0xd1060002, 0x00011103, - 0x7e0602ff, 0x00000200, - 0xbefc00ff, 0x00010000, - 0xbe800077, 0x8677ff77, - 0xff7fffff, 0x8777ff77, - 0x00058000, 0xd8ec0000, - 0x00000002, 0xbf8cc07f, - 0xe0765000, 0x701d0002, - 0x68040702, 0xd0c9006a, - 0x0000f702, 0xbf87fff7, - 0xbef70000, 0xbef000ff, - 0x00000400, 0xbefe00c1, - 0xbeff00c1, 0xb8fb2b05, - 0x807b817b, 0x8e7b827b, - 0xbef600ff, 0x01000000, - 0xbefc0084, 0xbf0a7b7c, - 0xbf84006d, 0xbf11017c, + 0xbe840080, 0xd2890000, + 0x00000902, 0x80048104, + 0xd2890001, 0x00000902, + 0x80048104, 0xd2890002, + 0x00000902, 0x80048104, + 0xd2890003, 0x00000902, + 0x80048104, 0xc069003a, + 0x00000070, 0xbf8cc07f, + 0x80709070, 0xbf06c004, + 0xbf84ffee, 0xbe840080, + 0xd2890000, 0x00000903, + 0x80048104, 0xd2890001, + 0x00000903, 0x80048104, + 0xd2890002, 0x00000903, + 0x80048104, 0xd2890003, + 0x00000903, 0x80048104, + 0xc069003a, 0x00000070, + 0xbf8cc07f, 0x80709070, + 0xbf06c004, 0xbf84ffee, + 0x807c847c, 0xbf0a7b7c, + 0xbf85ffb1, 0xbf9c0000, + 0xbf820012, 0x7e000300, + 0x7e020301, 0x7e040302, + 0x7e060303, 0xe0724000, + 0x701d0000, 0xe0724100, + 0x701d0100, 0xe0724200, + 0x701d0200, 0xe0724300, + 0x701d0300, 0x807c847c, + 0x8070ff70, 0x00000400, + 0xbf0a7b7c, 0xbf85ffef, + 0xbf9c0000, 0xb8fb2985, + 0x807b817b, 0x8e7b837b, + 0xb8fa2b05, 0x807a817a, + 0x8e7a827a, 0x80fb7a7b, + 0x867b7b7b, 0xbf84007a, 0x807bff7b, 0x00001000, + 0xbefc0080, 0xbf11017c, 0x867aff78, 0x00400000, 0xbf850003, 0xb8faf803, 0x897a7aff, 0x10000000, - 0xbf850051, 0xbe840080, + 0xbf850059, 0xd3d84000, + 0x18000100, 0xd3d84001, + 0x18000101, 0xd3d84002, + 0x18000102, 0xd3d84003, + 0x18000103, 0xbe840080, 0xd2890000, 0x00000900, 0x80048104, 0xd2890001, 0x00000900, 0x80048104, @@ -3225,201 +3304,137 @@ static const uint32_t cwsr_trap_gfx9_4_3_hex[] = { 0x00000070, 0xbf8cc07f, 0x80709070, 0xbf06c004, 0xbf84ffee, 0x807c847c, - 0xbf0a7b7c, 0xbf85ffb1, - 0xbf9c0000, 0xbf820012, - 0x7e000300, 0x7e020301, - 0x7e040302, 0x7e060303, + 0xbf0a7b7c, 0xbf85ffa9, + 0xbf9c0000, 0xbf820016, + 0xd3d84000, 0x18000100, + 0xd3d84001, 0x18000101, + 0xd3d84002, 0x18000102, + 0xd3d84003, 0x18000103, 0xe0724000, 0x701d0000, 0xe0724100, 0x701d0100, 0xe0724200, 0x701d0200, 0xe0724300, 0x701d0300, 0x807c847c, 0x8070ff70, 0x00000400, 0xbf0a7b7c, - 0xbf85ffef, 0xbf9c0000, - 0xb8fb2985, 0x807b817b, - 0x8e7b837b, 0xb8fa2b05, - 0x807a817a, 0x8e7a827a, - 0x80fb7a7b, 0x867b7b7b, - 0xbf84007a, 0x807bff7b, - 0x00001000, 0xbefc0080, - 0xbf11017c, 0x867aff78, - 0x00400000, 0xbf850003, - 0xb8faf803, 0x897a7aff, - 0x10000000, 0xbf850059, - 0xd3d84000, 0x18000100, - 0xd3d84001, 0x18000101, - 0xd3d84002, 0x18000102, - 0xd3d84003, 0x18000103, - 0xbe840080, 0xd2890000, - 0x00000900, 0x80048104, - 0xd2890001, 0x00000900, - 0x80048104, 0xd2890002, - 0x00000900, 0x80048104, - 0xd2890003, 0x00000900, - 0x80048104, 0xc069003a, - 0x00000070, 0xbf8cc07f, - 0x80709070, 0xbf06c004, - 0xbf84ffee, 0xbe840080, - 0xd2890000, 0x00000901, - 0x80048104, 0xd2890001, - 0x00000901, 0x80048104, - 0xd2890002, 0x00000901, - 0x80048104, 0xd2890003, - 0x00000901, 0x80048104, - 0xc069003a, 0x00000070, - 0xbf8cc07f, 0x80709070, - 0xbf06c004, 0xbf84ffee, - 0xbe840080, 0xd2890000, - 0x00000902, 0x80048104, - 0xd2890001, 0x00000902, - 0x80048104, 0xd2890002, - 0x00000902, 0x80048104, - 0xd2890003, 0x00000902, - 0x80048104, 0xc069003a, - 0x00000070, 0xbf8cc07f, - 0x80709070, 0xbf06c004, - 0xbf84ffee, 0xbe840080, - 0xd2890000, 0x00000903, - 0x80048104, 0xd2890001, - 0x00000903, 0x80048104, - 0xd2890002, 0x00000903, - 0x80048104, 0xd2890003, - 0x00000903, 0x80048104, - 0xc069003a, 0x00000070, - 0xbf8cc07f, 0x80709070, - 0xbf06c004, 0xbf84ffee, - 0x807c847c, 0xbf0a7b7c, - 0xbf85ffa9, 0xbf9c0000, - 0xbf820016, 0xd3d84000, - 0x18000100, 0xd3d84001, - 0x18000101, 0xd3d84002, - 0x18000102, 0xd3d84003, - 0x18000103, 0xe0724000, - 0x701d0000, 0xe0724100, - 0x701d0100, 0xe0724200, - 0x701d0200, 0xe0724300, - 0x701d0300, 0x807c847c, - 0x8070ff70, 0x00000400, - 0xbf0a7b7c, 0xbf85ffeb, - 0xbf9c0000, 0xbf8200ee, - 0xbef4007e, 0x8675ff7f, - 0x0000ffff, 0x8775ff75, - 0x00040000, 0xbef60080, - 0xbef700ff, 0x00807fac, - 0x866eff7f, 0x04000000, - 0xbf84001f, 0xbefe00c1, - 0xbeff00c1, 0xb8ef4306, - 0x866fc16f, 0xbf84001a, - 0x8e6f866f, 0x8e6f826f, - 0xbef6006f, 0xb8f82985, - 0x80788178, 0x8e788a78, - 0x8e788178, 0xb8ee1605, - 0x806e816e, 0x8e6e866e, - 0x80786e78, 0x8078ff78, - 0x00000080, 0xbef600ff, - 0x01000000, 0xbefc0080, - 0xe0510000, 0x781d0000, - 0xe0510100, 0x781d0000, - 0x807cff7c, 0x00000200, - 0x8078ff78, 0x00000200, - 0xbf0a6f7c, 0xbf85fff6, + 0xbf85ffeb, 0xbf9c0000, + 0xbf8200ee, 0xbef4007e, + 0x8675ff7f, 0x0000ffff, + 0x8775ff75, 0x00040000, + 0xbef60080, 0xbef700ff, + 0x00807fac, 0x866eff7f, + 0x04000000, 0xbf84001f, 0xbefe00c1, 0xbeff00c1, + 0xb8ef4306, 0x866fc16f, + 0xbf84001a, 0x8e6f866f, + 0x8e6f826f, 0xbef6006f, + 0xb8f82985, 0x80788178, + 0x8e788a78, 0x8e788178, + 0xb8ee1605, 0x806e816e, + 0x8e6e866e, 0x80786e78, + 0x8078ff78, 0x00000080, 0xbef600ff, 0x01000000, - 0xb8ef2b05, 0x806f816f, - 0x8e6f826f, 0x806fff6f, - 0x00008000, 0xbef80080, - 0xbeee0078, 0x8078ff78, - 0x00000400, 0xbefc0084, + 0xbefc0080, 0xe0510000, + 0x781d0000, 0xe0510100, + 0x781d0000, 0x807cff7c, + 0x00000200, 0x8078ff78, + 0x00000200, 0xbf0a6f7c, + 0xbf85fff6, 0xbefe00c1, + 0xbeff00c1, 0xbef600ff, + 0x01000000, 0xb8ef2b05, + 0x806f816f, 0x8e6f826f, + 0x806fff6f, 0x00008000, + 0xbef80080, 0xbeee0078, + 0x8078ff78, 0x00000400, + 0xbefc0084, 0xbf11087c, + 0xe0524000, 0x781d0000, + 0xe0524100, 0x781d0100, + 0xe0524200, 0x781d0200, + 0xe0524300, 0x781d0300, + 0xbf8c0f70, 0x7e000300, + 0x7e020301, 0x7e040302, + 0x7e060303, 0x807c847c, + 0x8078ff78, 0x00000400, + 0xbf0a6f7c, 0xbf85ffee, + 0xb8ef2985, 0x806f816f, + 0x8e6f836f, 0xb8f92b05, + 0x80798179, 0x8e798279, + 0x80ef796f, 0x866f6f6f, + 0xbf84001a, 0x806fff6f, + 0x00008000, 0xbefc0080, 0xbf11087c, 0xe0524000, 0x781d0000, 0xe0524100, 0x781d0100, 0xe0524200, 0x781d0200, 0xe0524300, 0x781d0300, 0xbf8c0f70, - 0x7e000300, 0x7e020301, - 0x7e040302, 0x7e060303, + 0xd3d94000, 0x18000100, + 0xd3d94001, 0x18000101, + 0xd3d94002, 0x18000102, + 0xd3d94003, 0x18000103, 0x807c847c, 0x8078ff78, 0x00000400, 0xbf0a6f7c, - 0xbf85ffee, 0xb8ef2985, - 0x806f816f, 0x8e6f836f, - 0xb8f92b05, 0x80798179, - 0x8e798279, 0x80ef796f, - 0x866f6f6f, 0xbf84001a, - 0x806fff6f, 0x00008000, - 0xbefc0080, 0xbf11087c, - 0xe0524000, 0x781d0000, - 0xe0524100, 0x781d0100, - 0xe0524200, 0x781d0200, - 0xe0524300, 0x781d0300, - 0xbf8c0f70, 0xd3d94000, - 0x18000100, 0xd3d94001, - 0x18000101, 0xd3d94002, - 0x18000102, 0xd3d94003, - 0x18000103, 0x807c847c, - 0x8078ff78, 0x00000400, - 0xbf0a6f7c, 0xbf85ffea, - 0xbf9c0000, 0xe0524000, - 0x6e1d0000, 0xe0524100, - 0x6e1d0100, 0xe0524200, - 0x6e1d0200, 0xe0524300, - 0x6e1d0300, 0xbf8c0f70, - 0xb8f82985, 0x80788178, - 0x8e788a78, 0x8e788178, - 0xb8ee1605, 0x806e816e, - 0x8e6e866e, 0x80786e78, - 0x80f8c078, 0xb8ef1605, - 0x806f816f, 0x8e6f846f, - 0x8e76826f, 0xbef600ff, - 0x01000000, 0xbefc006f, - 0xc031003a, 0x00000078, - 0x80f8c078, 0xbf8cc07f, - 0x80fc907c, 0xbf800000, - 0xbe802d00, 0xbe822d02, - 0xbe842d04, 0xbe862d06, - 0xbe882d08, 0xbe8a2d0a, - 0xbe8c2d0c, 0xbe8e2d0e, - 0xbf06807c, 0xbf84fff0, - 0xb8f82985, 0x80788178, - 0x8e788a78, 0x8e788178, - 0xb8ee1605, 0x806e816e, - 0x8e6e866e, 0x80786e78, - 0xbef60084, 0xbef600ff, - 0x01000000, 0xc0211bfa, + 0xbf85ffea, 0xbf9c0000, + 0xe0524000, 0x6e1d0000, + 0xe0524100, 0x6e1d0100, + 0xe0524200, 0x6e1d0200, + 0xe0524300, 0x6e1d0300, + 0xbf8c0f70, 0xb8f82985, + 0x80788178, 0x8e788a78, + 0x8e788178, 0xb8ee1605, + 0x806e816e, 0x8e6e866e, + 0x80786e78, 0x80f8c078, + 0xb8ef1605, 0x806f816f, + 0x8e6f846f, 0x8e76826f, + 0xbef600ff, 0x01000000, + 0xbefc006f, 0xc031003a, + 0x00000078, 0x80f8c078, + 0xbf8cc07f, 0x80fc907c, + 0xbf800000, 0xbe802d00, + 0xbe822d02, 0xbe842d04, + 0xbe862d06, 0xbe882d08, + 0xbe8a2d0a, 0xbe8c2d0c, + 0xbe8e2d0e, 0xbf06807c, + 0xbf84fff0, 0xb8f82985, + 0x80788178, 0x8e788a78, + 0x8e788178, 0xb8ee1605, + 0x806e816e, 0x8e6e866e, + 0x80786e78, 0xbef60084, + 0xbef600ff, 0x01000000, + 0xc0211bfa, 0x00000078, + 0x80788478, 0xc0211b3a, 0x00000078, 0x80788478, - 0xc0211b3a, 0x00000078, - 0x80788478, 0xc0211b7a, + 0xc0211b7a, 0x00000078, + 0x80788478, 0xc0211c3a, 0x00000078, 0x80788478, - 0xc0211c3a, 0x00000078, - 0x80788478, 0xc0211c7a, + 0xc0211c7a, 0x00000078, + 0x80788478, 0xc0211eba, 0x00000078, 0x80788478, - 0xc0211eba, 0x00000078, - 0x80788478, 0xc0211efa, + 0xc0211efa, 0x00000078, + 0x80788478, 0xc0211a3a, 0x00000078, 0x80788478, - 0xc0211a3a, 0x00000078, - 0x80788478, 0xc0211a7a, + 0xc0211a7a, 0x00000078, + 0x80788478, 0xc0211cfa, 0x00000078, 0x80788478, - 0xc0211cfa, 0x00000078, - 0x80788478, 0xbf8cc07f, - 0xbefc006f, 0xbefe0070, - 0xbeff0071, 0x866f7bff, - 0x000003ff, 0xb96f4803, - 0x866f7bff, 0xfffff800, - 0x8f6f8b6f, 0xb96fa2c3, - 0xb973f801, 0xb8ee2985, - 0x806e816e, 0x8e6e8a6e, - 0x8e6e816e, 0xb8ef1605, - 0x806f816f, 0x8e6f866f, - 0x806e6f6e, 0x806e746e, - 0x826f8075, 0x866fff6f, - 0x0000ffff, 0xc00b1c37, - 0x00000050, 0xc00b1d37, - 0x00000060, 0xc0031e77, - 0x00000074, 0xbf8cc07f, - 0x8f6e8b79, 0x866eff6e, - 0x001f8000, 0xb96ef807, - 0x866dff6d, 0x0000ffff, - 0x86fe7e7e, 0x86ea6a6a, - 0x8f6e837a, 0xb96ee0c2, - 0xbf800002, 0xb97a0002, - 0xbf8a0000, 0xbe801f6c, - 0xbf810000, 0x00000000, + 0xbf8cc07f, 0xbefc006f, + 0xbefe0070, 0xbeff0071, + 0x866f7bff, 0x000003ff, + 0xb96f4803, 0x866f7bff, + 0xfffff800, 0x8f6f8b6f, + 0xb96fa2c3, 0xb973f801, + 0xb8ee2985, 0x806e816e, + 0x8e6e8a6e, 0x8e6e816e, + 0xb8ef1605, 0x806f816f, + 0x8e6f866f, 0x806e6f6e, + 0x806e746e, 0x826f8075, + 0x866fff6f, 0x0000ffff, + 0xc00b1c37, 0x00000050, + 0xc00b1d37, 0x00000060, + 0xc0031e77, 0x00000074, + 0xbf8cc07f, 0x8f6e8b79, + 0x866eff6e, 0x001f8000, + 0xb96ef807, 0x866dff6d, + 0x0000ffff, 0x86fe7e7e, + 0x86ea6a6a, 0x8f6e837a, + 0xb96ee0c2, 0xbf800002, + 0xb97a0002, 0xbf8a0000, + 0xbe801f6c, 0xbf810000, }; diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm index 8b92c33c2a7c..fdab64624422 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx10.asm @@ -276,6 +276,11 @@ L_FETCH_2ND_TRAP: #endif s_lshl_b64 [ttmp14, ttmp15], [ttmp14, ttmp15], 0x8 + s_bitcmp1_b32 ttmp15, 0xF + s_cbranch_scc0 L_NO_SIGN_EXTEND_TMA + s_or_b32 ttmp15, ttmp15, 0xFFFF0000 +L_NO_SIGN_EXTEND_TMA: + s_load_dword ttmp2, [ttmp14, ttmp15], 0x10 glc:1 // debug trap enabled flag s_waitcnt lgkmcnt(0) s_lshl_b32 ttmp2, ttmp2, TTMP11_DEBUG_TRAP_ENABLED_SHIFT diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm index f2087cc2e89d..e506411ad28a 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm @@ -283,6 +283,11 @@ L_FETCH_2ND_TRAP: s_getreg_b32 ttmp15, hwreg(HW_REG_SQ_SHADER_TMA_HI) s_lshl_b64 [ttmp14, ttmp15], [ttmp14, ttmp15], 0x8 + s_bitcmp1_b32 ttmp15, 0xF + s_cbranch_scc0 L_NO_SIGN_EXTEND_TMA + s_or_b32 ttmp15, ttmp15, 0xFFFF0000 +L_NO_SIGN_EXTEND_TMA: + s_load_dword ttmp2, [ttmp14, ttmp15], 0x10 glc:1 // debug trap enabled flag s_waitcnt lgkmcnt(0) s_lshl_b32 ttmp2, ttmp2, TTMP_DEBUG_TRAP_ENABLED_SHIFT diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 6a27b000a246..c37f1fcd2165 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -333,10 +333,12 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, goto err_bind_process; } - if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(dev->kfd, &pdd->doorbell_index) < 0) { - err = -ENOMEM; - goto err_alloc_doorbells; + if (!pdd->qpd.proc_doorbells) { + err = kfd_alloc_process_doorbells(dev->kfd, pdd); + if (err) { + pr_debug("failed to allocate process doorbells\n"); + goto err_bind_process; + } } /* Starting with GFX11, wptr BOs must be mapped to GART for MES to determine work @@ -417,7 +419,6 @@ err_create_queue: if (wptr_bo) amdgpu_amdkfd_free_gtt_mem(dev->adev, wptr_bo); err_wptr_map_gart: -err_alloc_doorbells: err_bind_process: err_pdd: mutex_unlock(&p->mutex); @@ -1025,9 +1026,6 @@ bool kfd_dev_is_large_bar(struct kfd_node *dev) return true; } - if (dev->kfd->use_iommu_v2) - return false; - if (dev->local_mem_info.local_mem_size_private == 0 && dev->local_mem_info.local_mem_size_public > 0) return true; @@ -1487,7 +1485,8 @@ static int kfd_ioctl_alloc_queue_gws(struct file *filep, goto out_unlock; } - if (!kfd_dbg_has_gws_support(dev) && p->debug_trap_enabled) { + if (p->debug_trap_enabled && (!kfd_dbg_has_gws_support(dev) || + kfd_dbg_has_cwsr_workaround(dev))) { retval = -EBUSY; goto out_unlock; } @@ -1845,22 +1844,21 @@ static uint32_t get_process_num_bos(struct kfd_process *p) idr_for_each_entry(&pdd->alloc_idr, mem, id) { struct kgd_mem *kgd_mem = (struct kgd_mem *)mem; - if ((uint64_t)kgd_mem->va > pdd->gpuvm_base) + if (!kgd_mem->va || kgd_mem->va > pdd->gpuvm_base) num_of_bos++; } } return num_of_bos; } -static int criu_get_prime_handle(struct drm_gem_object *gobj, int flags, +static int criu_get_prime_handle(struct kgd_mem *mem, int flags, u32 *shared_fd) { struct dma_buf *dmabuf; int ret; - dmabuf = amdgpu_gem_prime_export(gobj, flags); - if (IS_ERR(dmabuf)) { - ret = PTR_ERR(dmabuf); + ret = amdgpu_amdkfd_gpuvm_export_dmabuf(mem, &dmabuf); + if (ret) { pr_err("dmabuf export failed for the BO\n"); return ret; } @@ -1918,7 +1916,11 @@ static int criu_checkpoint_bos(struct kfd_process *p, kgd_mem = (struct kgd_mem *)mem; dumper_bo = kgd_mem->bo; - if ((uint64_t)kgd_mem->va <= pdd->gpuvm_base) + /* Skip checkpointing BOs that are used for Trap handler + * code and state. Currently, these BOs have a VA that + * is less GPUVM Base + */ + if (kgd_mem->va && kgd_mem->va <= pdd->gpuvm_base) continue; bo_bucket = &bo_buckets[bo_index]; @@ -1940,7 +1942,7 @@ static int criu_checkpoint_bos(struct kfd_process *p, } if (bo_bucket->alloc_flags & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) { - ret = criu_get_prime_handle(&dumper_bo->tbo.base, + ret = criu_get_prime_handle(kgd_mem, bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_WRITABLE ? DRM_RDWR : 0, &bo_bucket->dmabuf_fd); @@ -2262,10 +2264,10 @@ static int criu_restore_devices(struct kfd_process *p, goto exit; } - if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) { - ret = -ENOMEM; - goto exit; + if (!pdd->qpd.proc_doorbells) { + ret = kfd_alloc_process_doorbells(dev->kfd, pdd); + if (ret) + goto exit; } } @@ -2402,7 +2404,7 @@ static int criu_restore_bo(struct kfd_process *p, /* create the dmabuf object and export the bo */ if (bo_bucket->alloc_flags & (KFD_IOC_ALLOC_MEM_FLAGS_VRAM | KFD_IOC_ALLOC_MEM_FLAGS_GTT)) { - ret = criu_get_prime_handle(&kgd_mem->bo->tbo.base, DRM_RDWR, + ret = criu_get_prime_handle(kgd_mem, DRM_RDWR, &bo_bucket->dmabuf_fd); if (ret) return ret; @@ -2755,6 +2757,16 @@ static int runtime_enable(struct kfd_process *p, uint64_t r_debug, if (pdd->qpd.queue_count) return -EEXIST; + + /* + * Setup TTMPs by default. + * Note that this call must remain here for MES ADD QUEUE to + * skip_process_ctx_clear unconditionally as the first call to + * SET_SHADER_DEBUGGER clears any stale process context data + * saved in MES. + */ + if (pdd->dev->kfd->shared_resources.enable_mes) + kfd_dbg_set_mes_debug_mode(pdd, !kfd_dbg_has_cwsr_workaround(pdd->dev)); } p->runtime_info.runtime_state = DEBUG_RUNTIME_STATE_ENABLED; @@ -2848,7 +2860,8 @@ static int runtime_disable(struct kfd_process *p) if (!pdd->dev->kfd->shared_resources.enable_mes) debug_refresh_runlist(pdd->dev->dqm); else - kfd_dbg_set_mes_debug_mode(pdd); + kfd_dbg_set_mes_debug_mode(pdd, + !kfd_dbg_has_cwsr_workaround(pdd->dev)); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index 49f40d9f16e8..f76b7aee5c0a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -26,7 +26,6 @@ #include "kfd_crat.h" #include "kfd_priv.h" #include "kfd_topology.h" -#include "kfd_iommu.h" #include "amdgpu.h" #include "amdgpu_amdkfd.h" @@ -1536,76 +1535,6 @@ int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pc return num_of_cache_types; } -static bool kfd_ignore_crat(void) -{ - bool ret; - - if (ignore_crat) - return true; - -#ifndef KFD_SUPPORT_IOMMU_V2 - ret = true; -#else - ret = false; -#endif - - return ret; -} - -/* - * kfd_create_crat_image_acpi - Allocates memory for CRAT image and - * copies CRAT from ACPI (if available). - * NOTE: Call kfd_destroy_crat_image to free CRAT image memory - * - * @crat_image: CRAT read from ACPI. If no CRAT in ACPI then - * crat_image will be NULL - * @size: [OUT] size of crat_image - * - * Return 0 if successful else return error code - */ -int kfd_create_crat_image_acpi(void **crat_image, size_t *size) -{ - struct acpi_table_header *crat_table; - acpi_status status; - void *pcrat_image; - int rc = 0; - - if (!crat_image) - return -EINVAL; - - *crat_image = NULL; - - if (kfd_ignore_crat()) { - pr_info("CRAT table disabled by module option\n"); - return -ENODATA; - } - - /* Fetch the CRAT table from ACPI */ - status = acpi_get_table(CRAT_SIGNATURE, 0, &crat_table); - if (status == AE_NOT_FOUND) { - pr_info("CRAT table not found\n"); - return -ENODATA; - } else if (ACPI_FAILURE(status)) { - const char *err = acpi_format_exception(status); - - pr_err("CRAT table error: %s\n", err); - return -EINVAL; - } - - pcrat_image = kvmalloc(crat_table->length, GFP_KERNEL); - if (!pcrat_image) { - rc = -ENOMEM; - goto out; - } - - memcpy(pcrat_image, crat_table, crat_table->length); - *crat_image = pcrat_image; - *size = crat_table->length; -out: - acpi_put_table(crat_table); - return rc; -} - /* Memory required to create Virtual CRAT. * Since there is no easy way to predict the amount of memory required, the * following amount is allocated for GPU Virtual CRAT. This is @@ -2158,7 +2087,8 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, amdgpu_amdkfd_get_cu_info(kdev->adev, &cu_info); cu->num_simd_per_cu = cu_info.simd_per_cu; - cu->num_simd_cores = cu_info.simd_per_cu * cu_info.cu_active_number; + cu->num_simd_cores = cu_info.simd_per_cu * + (cu_info.cu_active_number / kdev->kfd->num_nodes); cu->max_waves_simd = cu_info.max_waves_per_simd; cu->wave_front_size = cu_info.wave_front_size; @@ -2173,12 +2103,6 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, cu->hsa_capability = 0; - /* Check if this node supports IOMMU. During parsing this flag will - * translate to HSA_CAP_ATS_PRESENT - */ - if (!kfd_iommu_check_device(kdev->kfd)) - cu->hsa_capability |= CRAT_CU_FLAGS_IOMMU_PRESENT; - crat_table->length += sub_type_hdr->length; crat_table->total_entries++; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h index fc719389b5d6..74c2d7a0d628 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h @@ -79,6 +79,10 @@ struct crat_header { #define CRAT_SUBTYPE_IOLINK_AFFINITY 5 #define CRAT_SUBTYPE_MAX 6 +/* + * Do not change the value of CRAT_SIBLINGMAP_SIZE from 32 + * as it breaks the ABI. + */ #define CRAT_SIBLINGMAP_SIZE 32 /* @@ -307,7 +311,6 @@ struct kfd_gpu_cache_info { }; int kfd_get_gpu_cache_info(struct kfd_node *kdev, struct kfd_gpu_cache_info **pcache_info); -int kfd_create_crat_image_acpi(void **crat_image, size_t *size); void kfd_destroy_crat_image(void *crat_image); int kfd_parse_crat_table(void *crat_image, struct list_head *device_list, uint32_t proximity_domain); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c index fff3ccc04fa9..9ec750666382 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c @@ -302,8 +302,7 @@ static int kfd_dbg_set_queue_workaround(struct queue *q, bool enable) if (!q) return 0; - if (KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) || - KFD_GC_VERSION(q->device) >= IP_VERSION(12, 0, 0)) + if (!kfd_dbg_has_cwsr_workaround(q->device)) return 0; if (enable && q->properties.is_user_cu_masked) @@ -345,11 +344,10 @@ unwind: return r; } -int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd) +int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd, bool sq_trap_en) { uint32_t spi_dbg_cntl = pdd->spi_dbg_override | pdd->spi_dbg_launch_mode; uint32_t flags = pdd->process->dbg_flags; - bool sq_trap_en = !!spi_dbg_cntl; if (!kfd_dbg_is_per_vmid_supported(pdd->dev)) return 0; @@ -433,7 +431,7 @@ int kfd_dbg_trap_clear_dev_address_watch(struct kfd_process_device *pdd, if (!pdd->dev->kfd->shared_resources.enable_mes) r = debug_map_and_unlock(pdd->dev->dqm); else - r = kfd_dbg_set_mes_debug_mode(pdd); + r = kfd_dbg_set_mes_debug_mode(pdd, true); kfd_dbg_clear_dev_watch_id(pdd, watch_id); @@ -446,7 +444,8 @@ int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd, uint32_t *watch_id, uint32_t watch_mode) { - int r = kfd_dbg_get_dev_watch_id(pdd, watch_id); + int xcc_id, r = kfd_dbg_get_dev_watch_id(pdd, watch_id); + uint32_t xcc_mask = pdd->dev->xcc_mask; if (r) return r; @@ -460,19 +459,21 @@ int kfd_dbg_trap_set_dev_address_watch(struct kfd_process_device *pdd, } amdgpu_gfx_off_ctrl(pdd->dev->adev, false); - pdd->watch_points[*watch_id] = pdd->dev->kfd2kgd->set_address_watch( + for_each_inst(xcc_id, xcc_mask) + pdd->watch_points[*watch_id] = pdd->dev->kfd2kgd->set_address_watch( pdd->dev->adev, watch_address, watch_address_mask, *watch_id, watch_mode, - pdd->dev->vm_info.last_vmid_kfd); + pdd->dev->vm_info.last_vmid_kfd, + xcc_id); amdgpu_gfx_off_ctrl(pdd->dev->adev, true); if (!pdd->dev->kfd->shared_resources.enable_mes) r = debug_map_and_unlock(pdd->dev->dqm); else - r = kfd_dbg_set_mes_debug_mode(pdd); + r = kfd_dbg_set_mes_debug_mode(pdd, true); /* HWS is broken so no point in HW rollback but release the watchpoint anyways */ if (r) @@ -514,7 +515,7 @@ int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags) if (!pdd->dev->kfd->shared_resources.enable_mes) r = debug_refresh_runlist(pdd->dev->dqm); else - r = kfd_dbg_set_mes_debug_mode(pdd); + r = kfd_dbg_set_mes_debug_mode(pdd, true); if (r) { target->dbg_flags = prev_flags; @@ -537,7 +538,7 @@ int kfd_dbg_trap_set_flags(struct kfd_process *target, uint32_t *flags) if (!pdd->dev->kfd->shared_resources.enable_mes) debug_refresh_runlist(pdd->dev->dqm); else - kfd_dbg_set_mes_debug_mode(pdd); + kfd_dbg_set_mes_debug_mode(pdd, true); } } @@ -599,7 +600,7 @@ void kfd_dbg_trap_deactivate(struct kfd_process *target, bool unwind, int unwind if (!pdd->dev->kfd->shared_resources.enable_mes) debug_refresh_runlist(pdd->dev->dqm); else - kfd_dbg_set_mes_debug_mode(pdd); + kfd_dbg_set_mes_debug_mode(pdd, !kfd_dbg_has_cwsr_workaround(pdd->dev)); } kfd_dbg_set_workaround(target, false); @@ -715,7 +716,7 @@ int kfd_dbg_trap_activate(struct kfd_process *target) if (!pdd->dev->kfd->shared_resources.enable_mes) r = debug_refresh_runlist(pdd->dev->dqm); else - r = kfd_dbg_set_mes_debug_mode(pdd); + r = kfd_dbg_set_mes_debug_mode(pdd, true); if (r) { target->runtime_info.runtime_state = @@ -751,7 +752,8 @@ int kfd_dbg_trap_enable(struct kfd_process *target, uint32_t fd, if (!KFD_IS_SOC15(pdd->dev)) return -ENODEV; - if (!kfd_dbg_has_gws_support(pdd->dev) && pdd->qpd.num_gws) + if (pdd->qpd.num_gws && (!kfd_dbg_has_gws_support(pdd->dev) || + kfd_dbg_has_cwsr_workaround(pdd->dev))) return -EBUSY; } @@ -848,7 +850,7 @@ int kfd_dbg_trap_set_wave_launch_override(struct kfd_process *target, if (!pdd->dev->kfd->shared_resources.enable_mes) r = debug_refresh_runlist(pdd->dev->dqm); else - r = kfd_dbg_set_mes_debug_mode(pdd); + r = kfd_dbg_set_mes_debug_mode(pdd, true); if (r) break; @@ -880,7 +882,7 @@ int kfd_dbg_trap_set_wave_launch_mode(struct kfd_process *target, if (!pdd->dev->kfd->shared_resources.enable_mes) r = debug_refresh_runlist(pdd->dev->dqm); else - r = kfd_dbg_set_mes_debug_mode(pdd); + r = kfd_dbg_set_mes_debug_mode(pdd, true); if (r) break; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h index a289e59ceb79..fd0ff64d4184 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.h @@ -76,8 +76,9 @@ int kfd_dbg_send_exception_to_runtime(struct kfd_process *p, static inline bool kfd_dbg_is_per_vmid_supported(struct kfd_node *dev) { - return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || - KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0); + return (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || + KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3) || + KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0)); } void debug_event_write_work_handler(struct work_struct *work); @@ -100,6 +101,12 @@ static inline bool kfd_dbg_is_rlc_restore_supported(struct kfd_node *dev) KFD_GC_VERSION(dev) == IP_VERSION(10, 1, 1)); } +static inline bool kfd_dbg_has_cwsr_workaround(struct kfd_node *dev) +{ + return KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0) && + KFD_GC_VERSION(dev) <= IP_VERSION(11, 0, 3); +} + static inline bool kfd_dbg_has_gws_support(struct kfd_node *dev) { if ((KFD_GC_VERSION(dev) == IP_VERSION(9, 0, 1) @@ -119,5 +126,14 @@ static inline bool kfd_dbg_has_gws_support(struct kfd_node *dev) return true; } -int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd); +int kfd_dbg_set_mes_debug_mode(struct kfd_process_device *pdd, bool sq_trap_en); + +static inline bool kfd_dbg_has_ttmps_always_setup(struct kfd_node *dev) +{ + return (KFD_GC_VERSION(dev) < IP_VERSION(11, 0, 0) && + KFD_GC_VERSION(dev) != IP_VERSION(9, 4, 2)) || + (KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0) && + KFD_GC_VERSION(dev) < IP_VERSION(12, 0, 0) && + (dev->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 70); +} #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index 0b3dc754e06b..93ce181eb3ba 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -29,7 +29,6 @@ #include "kfd_pm4_headers_vi.h" #include "kfd_pm4_headers_aldebaran.h" #include "cwsr_trap_handler.h" -#include "kfd_iommu.h" #include "amdgpu_amdkfd.h" #include "kfd_smi_events.h" #include "kfd_svm.h" @@ -62,7 +61,6 @@ static int kfd_gtt_sa_init(struct kfd_dev *kfd, unsigned int buf_size, unsigned int chunk_size); static void kfd_gtt_sa_fini(struct kfd_dev *kfd); -static int kfd_resume_iommu(struct kfd_dev *kfd); static int kfd_resume(struct kfd_node *kfd); static void kfd_device_info_set_sdma_info(struct kfd_dev *kfd) @@ -194,11 +192,6 @@ static void kfd_device_info_init(struct kfd_dev *kfd, kfd_device_info_set_event_interrupt_class(kfd); - /* Raven */ - if (gc_version == IP_VERSION(9, 1, 0) || - gc_version == IP_VERSION(9, 2, 2)) - kfd->device_info.needs_iommu_device = true; - if (gc_version < IP_VERSION(11, 0, 0)) { /* Navi2x+, Navi1x+ */ if (gc_version == IP_VERSION(10, 3, 6)) @@ -233,10 +226,6 @@ static void kfd_device_info_init(struct kfd_dev *kfd, asic_type != CHIP_TONGA) kfd->device_info.supports_cwsr = true; - if (asic_type == CHIP_KAVERI || - asic_type == CHIP_CARRIZO) - kfd->device_info.needs_iommu_device = true; - if (asic_type != CHIP_HAWAII && !vf) kfd->device_info.needs_pci_atomics = true; } @@ -249,7 +238,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) uint32_t gfx_target_version = 0; switch (adev->asic_type) { -#ifdef KFD_SUPPORT_IOMMU_V2 #ifdef CONFIG_DRM_AMDGPU_CIK case CHIP_KAVERI: gfx_target_version = 70000; @@ -262,7 +250,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) if (!vf) f2g = &gfx_v8_kfd2kgd; break; -#endif #ifdef CONFIG_DRM_AMDGPU_CIK case CHIP_HAWAII: gfx_target_version = 70001; @@ -298,7 +285,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) gfx_target_version = 90000; f2g = &gfx_v9_kfd2kgd; break; -#ifdef KFD_SUPPORT_IOMMU_V2 /* Raven */ case IP_VERSION(9, 1, 0): case IP_VERSION(9, 2, 2): @@ -306,7 +292,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) if (!vf) f2g = &gfx_v9_kfd2kgd; break; -#endif /* Vega12 */ case IP_VERSION(9, 2, 1): gfx_target_version = 90004; @@ -455,8 +440,6 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) atomic_set(&kfd->compute_profile, 0); mutex_init(&kfd->doorbell_mutex); - memset(&kfd->doorbell_available_index, 0, - sizeof(kfd->doorbell_available_index)); ida_init(&kfd->doorbell_ida); @@ -508,6 +491,7 @@ static int kfd_gws_init(struct kfd_node *node) { int ret = 0; struct kfd_dev *kfd = node->kfd; + uint32_t mes_rev = node->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK; if (node->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) return 0; @@ -524,7 +508,10 @@ static int kfd_gws_init(struct kfd_node *node) (KFD_GC_VERSION(node) == IP_VERSION(9, 4, 3)) || (KFD_GC_VERSION(node) >= IP_VERSION(10, 3, 0) && KFD_GC_VERSION(node) < IP_VERSION(11, 0, 0) - && kfd->mec2_fw_version >= 0x6b)))) + && kfd->mec2_fw_version >= 0x6b) || + (KFD_GC_VERSION(node) >= IP_VERSION(11, 0, 0) + && KFD_GC_VERSION(node) < IP_VERSION(12, 0, 0) + && mes_rev >= 68)))) ret = amdgpu_amdkfd_alloc_gws(node->adev, node->adev->gds.gws_size, &node->gws); @@ -766,15 +753,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, kfd->noretry = kfd->adev->gmc.noretry; - /* If CRAT is broken, won't set iommu enabled */ - kfd_double_confirm_iommu_support(kfd); - - if (kfd_iommu_device_init(kfd)) { - kfd->use_iommu_v2 = false; - dev_err(kfd_device, "Error initializing iommuv2\n"); - goto device_iommu_error; - } - kfd_cwsr_init(kfd); dev_info(kfd_device, "Total number of KFD nodes to be created: %d\n", @@ -849,9 +827,6 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, svm_range_set_max_pages(kfd->adev); - if (kfd_resume_iommu(kfd)) - goto kfd_resume_iommu_error; - spin_lock_init(&kfd->watch_points_lock); kfd->init_complete = true; @@ -863,11 +838,9 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, goto out; -kfd_resume_iommu_error: node_init_error: node_alloc_error: kfd_cleanup_nodes(kfd, i); -device_iommu_error: kfd_doorbell_fini(kfd); kfd_doorbell_error: kfd_gtt_sa_fini(kfd); @@ -982,7 +955,6 @@ void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm) node = kfd->nodes[i]; node->dqm->ops.stop(node->dqm); } - kfd_iommu_suspend(kfd); } int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) @@ -1012,26 +984,6 @@ int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm) return ret; } -int kgd2kfd_resume_iommu(struct kfd_dev *kfd) -{ - if (!kfd->init_complete) - return 0; - - return kfd_resume_iommu(kfd); -} - -static int kfd_resume_iommu(struct kfd_dev *kfd) -{ - int err = 0; - - err = kfd_iommu_resume(kfd); - if (err) - dev_err(kfd_device, - "Failed to resume IOMMU for device %x:%x\n", - kfd->adev->pdev->vendor, kfd->adev->pdev->device); - return err; -} - static int kfd_resume(struct kfd_node *node) { int err = 0; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index f515cb8f30ca..8a6cb41444a4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -226,9 +226,9 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q, queue_input.paging = false; queue_input.tba_addr = qpd->tba_addr; queue_input.tma_addr = qpd->tma_addr; - queue_input.trap_en = KFD_GC_VERSION(q->device) < IP_VERSION(11, 0, 0) || - KFD_GC_VERSION(q->device) > IP_VERSION(11, 0, 3); - queue_input.skip_process_ctx_clear = qpd->pqm->process->debug_trap_enabled; + queue_input.trap_en = !kfd_dbg_has_cwsr_workaround(q->device); + queue_input.skip_process_ctx_clear = qpd->pqm->process->debug_trap_enabled || + kfd_dbg_has_ttmps_always_setup(q->device); queue_type = convert_to_mes_queue_type(q->properties.type); if (queue_type < 0) { @@ -238,10 +238,7 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q, } queue_input.queue_type = (uint32_t)queue_type; - if (q->gws) { - queue_input.gws_base = 0; - queue_input.gws_size = qpd->num_gws; - } + queue_input.exclusively_scheduled = q->properties.is_gws; amdgpu_mes_lock(&adev->mes); r = adev->mes.funcs->add_hw_queue(&adev->mes, &queue_input); @@ -251,7 +248,7 @@ static int add_queue_mes(struct device_queue_manager *dqm, struct queue *q, q->properties.doorbell_off); pr_err("MES might be in unrecoverable state, issue a GPU reset\n"); kfd_hws_hang(dqm); -} + } return r; } @@ -398,7 +395,7 @@ static int allocate_doorbell(struct qcm_process_device *qpd, unsigned int found; found = find_first_zero_bit(qpd->doorbell_bitmap, - KFD_MAX_NUM_OF_QUEUES_PER_PROCESS); + KFD_MAX_NUM_OF_QUEUES_PER_PROCESS); if (found >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) { pr_debug("No doorbells available"); return -EBUSY; @@ -408,9 +405,9 @@ static int allocate_doorbell(struct qcm_process_device *qpd, } } - q->properties.doorbell_off = - kfd_get_doorbell_dw_offset_in_bar(dev->kfd, qpd_to_pdd(qpd), - q->doorbell_id); + q->properties.doorbell_off = amdgpu_doorbell_index_on_bar(dev->adev, + qpd->proc_doorbells, + q->doorbell_id); return 0; } @@ -1621,7 +1618,8 @@ static int initialize_cpsch(struct device_queue_manager *dqm) if (dqm->dev->kfd2kgd->get_iq_wait_times) dqm->dev->kfd2kgd->get_iq_wait_times(dqm->dev->adev, - &dqm->wait_times); + &dqm->wait_times, + ffs(dqm->dev->xcc_mask) - 1); return 0; } @@ -1663,6 +1661,25 @@ static int start_cpsch(struct device_queue_manager *dqm) if (!dqm->dev->kfd->shared_resources.enable_mes) execute_queues_cpsch(dqm, KFD_UNMAP_QUEUES_FILTER_DYNAMIC_QUEUES, 0, USE_DEFAULT_GRACE_PERIOD); + + /* Set CWSR grace period to 1x1000 cycle for GFX9.4.3 APU */ + if (amdgpu_emu_mode == 0 && dqm->dev->adev->gmc.is_app_apu && + (KFD_GC_VERSION(dqm->dev) == IP_VERSION(9, 4, 3))) { + uint32_t reg_offset = 0; + uint32_t grace_period = 1; + + retval = pm_update_grace_period(&dqm->packet_mgr, + grace_period); + if (retval) + pr_err("Setting grace timeout failed\n"); + else if (dqm->dev->kfd2kgd->build_grace_period_packet_info) + /* Update dqm->wait_times maintained in software */ + dqm->dev->kfd2kgd->build_grace_period_packet_info( + dqm->dev->adev, dqm->wait_times, + grace_period, ®_offset, + &dqm->wait_times); + } + dqm_unlock(dqm); return 0; @@ -1806,8 +1823,7 @@ static int create_queue_cpsch(struct device_queue_manager *dqm, struct queue *q, */ q->properties.is_evicted = !!qpd->evicted; q->properties.is_dbg_wa = qpd->pqm->process->debug_trap_enabled && - KFD_GC_VERSION(q->device) >= IP_VERSION(11, 0, 0) && - KFD_GC_VERSION(q->device) <= IP_VERSION(11, 0, 3); + kfd_dbg_has_cwsr_workaround(q->device); if (qd) mqd_mgr->restore_mqd(mqd_mgr, &q->mqd, q->mqd_mem_obj, &q->gart_mqd_addr, @@ -2540,32 +2556,26 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_node *dev) } switch (dev->adev->asic_type) { - case CHIP_CARRIZO: - device_queue_manager_init_vi(&dqm->asic_ops); - break; - case CHIP_KAVERI: - device_queue_manager_init_cik(&dqm->asic_ops); - break; - case CHIP_HAWAII: - device_queue_manager_init_cik_hawaii(&dqm->asic_ops); + device_queue_manager_init_cik(&dqm->asic_ops); break; + case CHIP_CARRIZO: case CHIP_TONGA: case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: case CHIP_POLARIS12: case CHIP_VEGAM: - device_queue_manager_init_vi_tonga(&dqm->asic_ops); + device_queue_manager_init_vi(&dqm->asic_ops); break; default: if (KFD_GC_VERSION(dev) >= IP_VERSION(11, 0, 0)) device_queue_manager_init_v11(&dqm->asic_ops); else if (KFD_GC_VERSION(dev) >= IP_VERSION(10, 1, 1)) - device_queue_manager_init_v10_navi10(&dqm->asic_ops); + device_queue_manager_init_v10(&dqm->asic_ops); else if (KFD_GC_VERSION(dev) >= IP_VERSION(9, 0, 1)) device_queue_manager_init_v9(&dqm->asic_ops); else { @@ -2805,19 +2815,11 @@ static void copy_context_work_handler (struct work_struct *work) static uint32_t *get_queue_ids(uint32_t num_queues, uint32_t *usr_queue_id_array) { size_t array_size = num_queues * sizeof(uint32_t); - uint32_t *queue_ids = NULL; if (!usr_queue_id_array) return NULL; - queue_ids = kzalloc(array_size, GFP_KERNEL); - if (!queue_ids) - return ERR_PTR(-ENOMEM); - - if (copy_from_user(queue_ids, usr_queue_id_array, array_size)) - return ERR_PTR(-EFAULT); - - return queue_ids; + return memdup_user(usr_queue_id_array, array_size); } int resume_queues(struct kfd_process *p, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h index 7dd4b177219d..cf7e182588f8 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.h @@ -269,15 +269,11 @@ struct device_queue_manager { void device_queue_manager_init_cik( struct device_queue_manager_asic_ops *asic_ops); -void device_queue_manager_init_cik_hawaii( - struct device_queue_manager_asic_ops *asic_ops); void device_queue_manager_init_vi( struct device_queue_manager_asic_ops *asic_ops); -void device_queue_manager_init_vi_tonga( - struct device_queue_manager_asic_ops *asic_ops); void device_queue_manager_init_v9( struct device_queue_manager_asic_ops *asic_ops); -void device_queue_manager_init_v10_navi10( +void device_queue_manager_init_v10( struct device_queue_manager_asic_ops *asic_ops); void device_queue_manager_init_v11( struct device_queue_manager_asic_ops *asic_ops); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c index b1ab5b0775e1..d4d95c7f2e5d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_cik.c @@ -34,17 +34,13 @@ static bool set_cache_memory_policy_cik(struct device_queue_manager *dqm, void __user *alternate_aperture_base, uint64_t alternate_aperture_size); static int update_qpd_cik(struct device_queue_manager *dqm, - struct qcm_process_device *qpd); -static int update_qpd_cik_hawaii(struct device_queue_manager *dqm, - struct qcm_process_device *qpd); -static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd); -static void init_sdma_vm_hawaii(struct device_queue_manager *dqm, - struct queue *q, - struct qcm_process_device *qpd); + struct qcm_process_device *qpd); +static void init_sdma_vm(struct device_queue_manager *dqm, + struct queue *q, + struct qcm_process_device *qpd); void device_queue_manager_init_cik( - struct device_queue_manager_asic_ops *asic_ops) + struct device_queue_manager_asic_ops *asic_ops) { asic_ops->set_cache_memory_policy = set_cache_memory_policy_cik; asic_ops->update_qpd = update_qpd_cik; @@ -52,15 +48,6 @@ void device_queue_manager_init_cik( asic_ops->mqd_manager_init = mqd_manager_init_cik; } -void device_queue_manager_init_cik_hawaii( - struct device_queue_manager_asic_ops *asic_ops) -{ - asic_ops->set_cache_memory_policy = set_cache_memory_policy_cik; - asic_ops->update_qpd = update_qpd_cik_hawaii; - asic_ops->init_sdma_vm = init_sdma_vm_hawaii; - asic_ops->mqd_manager_init = mqd_manager_init_cik_hawaii; -} - static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble) { /* In 64-bit mode, we can only control the top 3 bits of the LDS, @@ -115,41 +102,7 @@ static bool set_cache_memory_policy_cik(struct device_queue_manager *dqm, } static int update_qpd_cik(struct device_queue_manager *dqm, - struct qcm_process_device *qpd) -{ - struct kfd_process_device *pdd; - unsigned int temp; - - pdd = qpd_to_pdd(qpd); - - /* check if sh_mem_config register already configured */ - if (qpd->sh_mem_config == 0) { - qpd->sh_mem_config = - ALIGNMENT_MODE(SH_MEM_ALIGNMENT_MODE_UNALIGNED) | - DEFAULT_MTYPE(MTYPE_NONCACHED) | - APE1_MTYPE(MTYPE_NONCACHED); - qpd->sh_mem_ape1_limit = 0; - qpd->sh_mem_ape1_base = 0; - } - - if (qpd->pqm->process->is_32bit_user_mode) { - temp = get_sh_mem_bases_32(pdd); - qpd->sh_mem_bases = SHARED_BASE(temp); - qpd->sh_mem_config |= PTR32; - } else { - temp = get_sh_mem_bases_nybble_64(pdd); - qpd->sh_mem_bases = compute_sh_mem_bases_64bit(temp); - qpd->sh_mem_config |= 1 << SH_MEM_CONFIG__PRIVATE_ATC__SHIFT; - } - - pr_debug("is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n", - qpd->pqm->process->is_32bit_user_mode, temp, qpd->sh_mem_bases); - - return 0; -} - -static int update_qpd_cik_hawaii(struct device_queue_manager *dqm, - struct qcm_process_device *qpd) + struct qcm_process_device *qpd) { struct kfd_process_device *pdd; unsigned int temp; @@ -178,25 +131,9 @@ static int update_qpd_cik_hawaii(struct device_queue_manager *dqm, return 0; } -static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd) -{ - uint32_t value = (1 << SDMA0_RLC0_VIRTUAL_ADDR__ATC__SHIFT); - - if (q->process->is_32bit_user_mode) - value |= (1 << SDMA0_RLC0_VIRTUAL_ADDR__PTR32__SHIFT) | - get_sh_mem_bases_32(qpd_to_pdd(qpd)); - else - value |= ((get_sh_mem_bases_nybble_64(qpd_to_pdd(qpd))) << - SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE__SHIFT) & - SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE_MASK; - - q->properties.sdma_vm_addr = value; -} - -static void init_sdma_vm_hawaii(struct device_queue_manager *dqm, - struct queue *q, - struct qcm_process_device *qpd) +static void init_sdma_vm(struct device_queue_manager *dqm, + struct queue *q, + struct qcm_process_device *qpd) { /* On dGPU we're always in GPUVM64 addressing mode with 64-bit * aperture addresses. diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v10.c index f1a1f5753e65..245a90dfc2f6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v10.c @@ -32,7 +32,7 @@ static int update_qpd_v10(struct device_queue_manager *dqm, static void init_sdma_vm_v10(struct device_queue_manager *dqm, struct queue *q, struct qcm_process_device *qpd); -void device_queue_manager_init_v10_navi10( +void device_queue_manager_init_v10( struct device_queue_manager_asic_ops *asic_ops) { asic_ops->update_qpd = update_qpd_v10; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c index 8af643388768..54eb1bff903c 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v9.c @@ -60,7 +60,7 @@ static int update_qpd_v9(struct device_queue_manager *dqm, qpd->sh_mem_config = SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT; - if (dqm->dev->kfd->noretry && !dqm->dev->kfd->use_iommu_v2) + if (dqm->dev->kfd->noretry) qpd->sh_mem_config |= 1 << SH_MEM_CONFIG__RETRY_DISABLE__SHIFT; if (KFD_GC_VERSION(dqm->dev->kfd) == IP_VERSION(9, 4, 3)) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c index d7d45832df0f..b291ee0fab94 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c @@ -28,29 +28,19 @@ #include "oss/oss_3_0_sh_mask.h" static bool set_cache_memory_policy_vi(struct device_queue_manager *dqm, - struct qcm_process_device *qpd, - enum cache_policy default_policy, - enum cache_policy alternate_policy, - void __user *alternate_aperture_base, - uint64_t alternate_aperture_size); -static bool set_cache_memory_policy_vi_tonga(struct device_queue_manager *dqm, - struct qcm_process_device *qpd, - enum cache_policy default_policy, - enum cache_policy alternate_policy, - void __user *alternate_aperture_base, - uint64_t alternate_aperture_size); + struct qcm_process_device *qpd, + enum cache_policy default_policy, + enum cache_policy alternate_policy, + void __user *alternate_aperture_base, + uint64_t alternate_aperture_size); static int update_qpd_vi(struct device_queue_manager *dqm, - struct qcm_process_device *qpd); -static int update_qpd_vi_tonga(struct device_queue_manager *dqm, - struct qcm_process_device *qpd); -static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd); -static void init_sdma_vm_tonga(struct device_queue_manager *dqm, - struct queue *q, - struct qcm_process_device *qpd); + struct qcm_process_device *qpd); +static void init_sdma_vm(struct device_queue_manager *dqm, + struct queue *q, + struct qcm_process_device *qpd); void device_queue_manager_init_vi( - struct device_queue_manager_asic_ops *asic_ops) + struct device_queue_manager_asic_ops *asic_ops) { asic_ops->set_cache_memory_policy = set_cache_memory_policy_vi; asic_ops->update_qpd = update_qpd_vi; @@ -58,15 +48,6 @@ void device_queue_manager_init_vi( asic_ops->mqd_manager_init = mqd_manager_init_vi; } -void device_queue_manager_init_vi_tonga( - struct device_queue_manager_asic_ops *asic_ops) -{ - asic_ops->set_cache_memory_policy = set_cache_memory_policy_vi_tonga; - asic_ops->update_qpd = update_qpd_vi_tonga; - asic_ops->init_sdma_vm = init_sdma_vm_tonga; - asic_ops->mqd_manager_init = mqd_manager_init_vi_tonga; -} - static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble) { /* In 64-bit mode, we can only control the top 3 bits of the LDS, @@ -96,35 +77,6 @@ static uint32_t compute_sh_mem_bases_64bit(unsigned int top_address_nybble) } static bool set_cache_memory_policy_vi(struct device_queue_manager *dqm, - struct qcm_process_device *qpd, - enum cache_policy default_policy, - enum cache_policy alternate_policy, - void __user *alternate_aperture_base, - uint64_t alternate_aperture_size) -{ - uint32_t default_mtype; - uint32_t ape1_mtype; - - default_mtype = (default_policy == cache_policy_coherent) ? - MTYPE_CC : - MTYPE_NC; - - ape1_mtype = (alternate_policy == cache_policy_coherent) ? - MTYPE_CC : - MTYPE_NC; - - qpd->sh_mem_config = (qpd->sh_mem_config & - SH_MEM_CONFIG__ADDRESS_MODE_MASK) | - SH_MEM_ALIGNMENT_MODE_UNALIGNED << - SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT | - default_mtype << SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT | - ape1_mtype << SH_MEM_CONFIG__APE1_MTYPE__SHIFT | - SH_MEM_CONFIG__PRIVATE_ATC_MASK; - - return true; -} - -static bool set_cache_memory_policy_vi_tonga(struct device_queue_manager *dqm, struct qcm_process_device *qpd, enum cache_policy default_policy, enum cache_policy alternate_policy, @@ -152,48 +104,7 @@ static bool set_cache_memory_policy_vi_tonga(struct device_queue_manager *dqm, } static int update_qpd_vi(struct device_queue_manager *dqm, - struct qcm_process_device *qpd) -{ - struct kfd_process_device *pdd; - unsigned int temp; - - pdd = qpd_to_pdd(qpd); - - /* check if sh_mem_config register already configured */ - if (qpd->sh_mem_config == 0) { - qpd->sh_mem_config = - SH_MEM_ALIGNMENT_MODE_UNALIGNED << - SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT | - MTYPE_CC << SH_MEM_CONFIG__DEFAULT_MTYPE__SHIFT | - MTYPE_CC << SH_MEM_CONFIG__APE1_MTYPE__SHIFT | - SH_MEM_CONFIG__PRIVATE_ATC_MASK; - - qpd->sh_mem_ape1_limit = 0; - qpd->sh_mem_ape1_base = 0; - } - - if (qpd->pqm->process->is_32bit_user_mode) { - temp = get_sh_mem_bases_32(pdd); - qpd->sh_mem_bases = temp << SH_MEM_BASES__SHARED_BASE__SHIFT; - qpd->sh_mem_config |= SH_MEM_ADDRESS_MODE_HSA32 << - SH_MEM_CONFIG__ADDRESS_MODE__SHIFT; - } else { - temp = get_sh_mem_bases_nybble_64(pdd); - qpd->sh_mem_bases = compute_sh_mem_bases_64bit(temp); - qpd->sh_mem_config |= SH_MEM_ADDRESS_MODE_HSA64 << - SH_MEM_CONFIG__ADDRESS_MODE__SHIFT; - qpd->sh_mem_config |= 1 << - SH_MEM_CONFIG__PRIVATE_ATC__SHIFT; - } - - pr_debug("is32bit process: %d sh_mem_bases nybble: 0x%X and register 0x%X\n", - qpd->pqm->process->is_32bit_user_mode, temp, qpd->sh_mem_bases); - - return 0; -} - -static int update_qpd_vi_tonga(struct device_queue_manager *dqm, - struct qcm_process_device *qpd) + struct qcm_process_device *qpd) { struct kfd_process_device *pdd; unsigned int temp; @@ -226,25 +137,9 @@ static int update_qpd_vi_tonga(struct device_queue_manager *dqm, return 0; } -static void init_sdma_vm(struct device_queue_manager *dqm, struct queue *q, - struct qcm_process_device *qpd) -{ - uint32_t value = (1 << SDMA0_RLC0_VIRTUAL_ADDR__ATC__SHIFT); - - if (q->process->is_32bit_user_mode) - value |= (1 << SDMA0_RLC0_VIRTUAL_ADDR__PTR32__SHIFT) | - get_sh_mem_bases_32(qpd_to_pdd(qpd)); - else - value |= ((get_sh_mem_bases_nybble_64(qpd_to_pdd(qpd))) << - SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE__SHIFT) & - SDMA0_RLC0_VIRTUAL_ADDR__SHARED_BASE_MASK; - - q->properties.sdma_vm_addr = value; -} - -static void init_sdma_vm_tonga(struct device_queue_manager *dqm, - struct queue *q, - struct qcm_process_device *qpd) +static void init_sdma_vm(struct device_queue_manager *dqm, + struct queue *q, + struct qcm_process_device *qpd) { /* On dGPU we're always in GPUVM64 addressing mode with 64-bit * aperture addresses. diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c index 6421b620388d..7b38537c7c99 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c @@ -61,81 +61,46 @@ size_t kfd_doorbell_process_slice(struct kfd_dev *kfd) /* Doorbell calculations for device init. */ int kfd_doorbell_init(struct kfd_dev *kfd) { - size_t doorbell_start_offset; - size_t doorbell_aperture_size; - size_t doorbell_process_limit; + int size = PAGE_SIZE; + int r; /* - * With MES enabled, just set the doorbell base as it is needed - * to calculate doorbell physical address. - */ - if (kfd->shared_resources.enable_mes) { - kfd->doorbell_base = - kfd->shared_resources.doorbell_physical_address; - return 0; - } - - /* - * We start with calculations in bytes because the input data might - * only be byte-aligned. - * Only after we have done the rounding can we assume any alignment. + * Todo: KFD kernel level operations need only one doorbell for + * ring test/HWS. So instead of reserving a whole page here for + * kernel, reserve and consume a doorbell from existing KGD kernel + * doorbell page. */ - doorbell_start_offset = - roundup(kfd->shared_resources.doorbell_start_offset, - kfd_doorbell_process_slice(kfd)); - - doorbell_aperture_size = - rounddown(kfd->shared_resources.doorbell_aperture_size, - kfd_doorbell_process_slice(kfd)); - - if (doorbell_aperture_size > doorbell_start_offset) - doorbell_process_limit = - (doorbell_aperture_size - doorbell_start_offset) / - kfd_doorbell_process_slice(kfd); - else - return -ENOSPC; - - if (!kfd->max_doorbell_slices || - doorbell_process_limit < kfd->max_doorbell_slices) - kfd->max_doorbell_slices = doorbell_process_limit; - - kfd->doorbell_base = kfd->shared_resources.doorbell_physical_address + - doorbell_start_offset; - - kfd->doorbell_base_dw_offset = doorbell_start_offset / sizeof(u32); - - kfd->doorbell_kernel_ptr = ioremap(kfd->doorbell_base, - kfd_doorbell_process_slice(kfd)); - - if (!kfd->doorbell_kernel_ptr) + /* Bitmap to dynamically allocate doorbells from kernel page */ + kfd->doorbell_bitmap = bitmap_zalloc(size / sizeof(u32), GFP_KERNEL); + if (!kfd->doorbell_bitmap) { + DRM_ERROR("Failed to allocate kernel doorbell bitmap\n"); return -ENOMEM; + } - pr_debug("Doorbell initialization:\n"); - pr_debug("doorbell base == 0x%08lX\n", - (uintptr_t)kfd->doorbell_base); - - pr_debug("doorbell_base_dw_offset == 0x%08lX\n", - kfd->doorbell_base_dw_offset); - - pr_debug("doorbell_process_limit == 0x%08lX\n", - doorbell_process_limit); - - pr_debug("doorbell_kernel_offset == 0x%08lX\n", - (uintptr_t)kfd->doorbell_base); - - pr_debug("doorbell aperture size == 0x%08lX\n", - kfd->shared_resources.doorbell_aperture_size); - - pr_debug("doorbell kernel address == %p\n", kfd->doorbell_kernel_ptr); + /* Alloc a doorbell page for KFD kernel usages */ + r = amdgpu_bo_create_kernel(kfd->adev, + size, + PAGE_SIZE, + AMDGPU_GEM_DOMAIN_DOORBELL, + &kfd->doorbells, + NULL, + (void **)&kfd->doorbell_kernel_ptr); + if (r) { + pr_err("failed to allocate kernel doorbells\n"); + bitmap_free(kfd->doorbell_bitmap); + return r; + } + pr_debug("Doorbell kernel address == %p\n", kfd->doorbell_kernel_ptr); return 0; } void kfd_doorbell_fini(struct kfd_dev *kfd) { - if (kfd->doorbell_kernel_ptr) - iounmap(kfd->doorbell_kernel_ptr); + bitmap_free(kfd->doorbell_bitmap); + amdgpu_bo_free_kernel(&kfd->doorbells, NULL, + (void **)&kfd->doorbell_kernel_ptr); } int kfd_doorbell_mmap(struct kfd_node *dev, struct kfd_process *process, @@ -188,22 +153,16 @@ void __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd, u32 inx; mutex_lock(&kfd->doorbell_mutex); - inx = find_first_zero_bit(kfd->doorbell_available_index, - KFD_MAX_NUM_OF_QUEUES_PER_PROCESS); + inx = find_first_zero_bit(kfd->doorbell_bitmap, PAGE_SIZE / sizeof(u32)); - __set_bit(inx, kfd->doorbell_available_index); + __set_bit(inx, kfd->doorbell_bitmap); mutex_unlock(&kfd->doorbell_mutex); if (inx >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) return NULL; - inx *= kfd->device_info.doorbell_size / sizeof(u32); - - /* - * Calculating the kernel doorbell offset using the first - * doorbell page. - */ - *doorbell_off = kfd->doorbell_base_dw_offset + inx; + *doorbell_off = amdgpu_doorbell_index_on_bar(kfd->adev, kfd->doorbells, inx); + inx *= 2; pr_debug("Get kernel queue doorbell\n" " doorbell offset == 0x%08X\n" @@ -217,11 +176,11 @@ void kfd_release_kernel_doorbell(struct kfd_dev *kfd, u32 __iomem *db_addr) { unsigned int inx; - inx = (unsigned int)(db_addr - kfd->doorbell_kernel_ptr) - * sizeof(u32) / kfd->device_info.doorbell_size; + inx = (unsigned int)(db_addr - kfd->doorbell_kernel_ptr); + inx /= 2; mutex_lock(&kfd->doorbell_mutex); - __clear_bit(inx, kfd->doorbell_available_index); + __clear_bit(inx, kfd->doorbell_bitmap); mutex_unlock(&kfd->doorbell_mutex); } @@ -243,80 +202,96 @@ void write_kernel_doorbell64(void __iomem *db, u64 value) } } -unsigned int kfd_get_doorbell_dw_offset_in_bar(struct kfd_dev *kfd, - struct kfd_process_device *pdd, - unsigned int doorbell_id) +static int init_doorbell_bitmap(struct qcm_process_device *qpd, + struct kfd_dev *dev) { - /* - * doorbell_base_dw_offset accounts for doorbells taken by KGD. - * index * kfd_doorbell_process_slice/sizeof(u32) adjusts to - * the process's doorbells. The offset returned is in dword - * units regardless of the ASIC-dependent doorbell size. - */ - if (!kfd->shared_resources.enable_mes) - return kfd->doorbell_base_dw_offset + - pdd->doorbell_index - * kfd_doorbell_process_slice(kfd) / sizeof(u32) + - doorbell_id * - kfd->device_info.doorbell_size / sizeof(u32); - else - return amdgpu_mes_get_doorbell_dw_offset_in_bar( - (struct amdgpu_device *)kfd->adev, - pdd->doorbell_index, doorbell_id); -} + unsigned int i; + int range_start = dev->shared_resources.non_cp_doorbells_start; + int range_end = dev->shared_resources.non_cp_doorbells_end; -uint64_t kfd_get_number_elems(struct kfd_dev *kfd) -{ - uint64_t num_of_elems = (kfd->shared_resources.doorbell_aperture_size - - kfd->shared_resources.doorbell_start_offset) / - kfd_doorbell_process_slice(kfd) + 1; + if (!KFD_IS_SOC15(dev)) + return 0; - return num_of_elems; + /* Mask out doorbells reserved for SDMA, IH, and VCN on SOC15. */ + pr_debug("reserved doorbell 0x%03x - 0x%03x\n", range_start, range_end); + pr_debug("reserved doorbell 0x%03x - 0x%03x\n", + range_start + KFD_QUEUE_DOORBELL_MIRROR_OFFSET, + range_end + KFD_QUEUE_DOORBELL_MIRROR_OFFSET); + + for (i = 0; i < KFD_MAX_NUM_OF_QUEUES_PER_PROCESS / 2; i++) { + if (i >= range_start && i <= range_end) { + __set_bit(i, qpd->doorbell_bitmap); + __set_bit(i + KFD_QUEUE_DOORBELL_MIRROR_OFFSET, + qpd->doorbell_bitmap); + } + } + return 0; } phys_addr_t kfd_get_process_doorbells(struct kfd_process_device *pdd) { - if (!pdd->doorbell_index) { - int r = kfd_alloc_process_doorbells(pdd->dev->kfd, - &pdd->doorbell_index); - if (r < 0) + struct amdgpu_device *adev = pdd->dev->adev; + uint32_t first_db_index; + + if (!pdd->qpd.proc_doorbells) { + if (kfd_alloc_process_doorbells(pdd->dev->kfd, pdd)) + /* phys_addr_t 0 is error */ return 0; } - return pdd->dev->kfd->doorbell_base + - pdd->doorbell_index * kfd_doorbell_process_slice(pdd->dev->kfd); + first_db_index = amdgpu_doorbell_index_on_bar(adev, pdd->qpd.proc_doorbells, 0); + return adev->doorbell.base + first_db_index * sizeof(uint32_t); } -int kfd_alloc_process_doorbells(struct kfd_dev *kfd, unsigned int *doorbell_index) +int kfd_alloc_process_doorbells(struct kfd_dev *kfd, struct kfd_process_device *pdd) { - int r = 0; - - if (!kfd->shared_resources.enable_mes) - r = ida_simple_get(&kfd->doorbell_ida, 1, - kfd->max_doorbell_slices, GFP_KERNEL); - else - r = amdgpu_mes_alloc_process_doorbells( - (struct amdgpu_device *)kfd->adev, - doorbell_index); + int r; + struct qcm_process_device *qpd = &pdd->qpd; + + /* Allocate bitmap for dynamic doorbell allocation */ + qpd->doorbell_bitmap = bitmap_zalloc(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, + GFP_KERNEL); + if (!qpd->doorbell_bitmap) { + DRM_ERROR("Failed to allocate process doorbell bitmap\n"); + return -ENOMEM; + } - if (r > 0) - *doorbell_index = r; + r = init_doorbell_bitmap(&pdd->qpd, kfd); + if (r) { + DRM_ERROR("Failed to initialize process doorbells\n"); + r = -ENOMEM; + goto err; + } - if (r < 0) - pr_err("Failed to allocate process doorbells\n"); + /* Allocate doorbells for this process */ + r = amdgpu_bo_create_kernel(kfd->adev, + kfd_doorbell_process_slice(kfd), + PAGE_SIZE, + AMDGPU_GEM_DOMAIN_DOORBELL, + &qpd->proc_doorbells, + NULL, + NULL); + if (r) { + DRM_ERROR("Failed to allocate process doorbells\n"); + goto err; + } + return 0; +err: + bitmap_free(qpd->doorbell_bitmap); + qpd->doorbell_bitmap = NULL; return r; } -void kfd_free_process_doorbells(struct kfd_dev *kfd, unsigned int doorbell_index) +void kfd_free_process_doorbells(struct kfd_dev *kfd, struct kfd_process_device *pdd) { - if (doorbell_index) { - if (!kfd->shared_resources.enable_mes) - ida_simple_remove(&kfd->doorbell_ida, doorbell_index); - else - amdgpu_mes_free_process_doorbells( - (struct amdgpu_device *)kfd->adev, - doorbell_index); + struct qcm_process_device *qpd = &pdd->qpd; + + if (qpd->doorbell_bitmap) { + bitmap_free(qpd->doorbell_bitmap); + qpd->doorbell_bitmap = NULL; } + + amdgpu_bo_free_kernel(&qpd->proc_doorbells, NULL, NULL); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 8081a9408006..0f58be65132f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -31,7 +31,6 @@ #include <linux/memory.h> #include "kfd_priv.h" #include "kfd_events.h" -#include "kfd_iommu.h" #include <linux/device.h> /* @@ -1146,87 +1145,6 @@ static void lookup_events_by_type_and_signal(struct kfd_process *p, rcu_read_unlock(); } -#ifdef KFD_SUPPORT_IOMMU_V2 -void kfd_signal_iommu_event(struct kfd_node *dev, u32 pasid, - unsigned long address, bool is_write_requested, - bool is_execute_requested) -{ - struct kfd_hsa_memory_exception_data memory_exception_data; - struct vm_area_struct *vma; - int user_gpu_id; - - /* - * Because we are called from arbitrary context (workqueue) as opposed - * to process context, kfd_process could attempt to exit while we are - * running so the lookup function increments the process ref count. - */ - struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); - struct mm_struct *mm; - - if (!p) - return; /* Presumably process exited. */ - - /* Take a safe reference to the mm_struct, which may otherwise - * disappear even while the kfd_process is still referenced. - */ - mm = get_task_mm(p->lead_thread); - if (!mm) { - kfd_unref_process(p); - return; /* Process is exiting */ - } - - user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id); - if (unlikely(user_gpu_id == -EINVAL)) { - WARN_ONCE(1, "Could not get user_gpu_id from dev->id:%x\n", dev->id); - return; - } - memset(&memory_exception_data, 0, sizeof(memory_exception_data)); - - mmap_read_lock(mm); - vma = find_vma(mm, address); - - memory_exception_data.gpu_id = user_gpu_id; - memory_exception_data.va = address; - /* Set failure reason */ - memory_exception_data.failure.NotPresent = 1; - memory_exception_data.failure.NoExecute = 0; - memory_exception_data.failure.ReadOnly = 0; - if (vma && address >= vma->vm_start) { - memory_exception_data.failure.NotPresent = 0; - - if (is_write_requested && !(vma->vm_flags & VM_WRITE)) - memory_exception_data.failure.ReadOnly = 1; - else - memory_exception_data.failure.ReadOnly = 0; - - if (is_execute_requested && !(vma->vm_flags & VM_EXEC)) - memory_exception_data.failure.NoExecute = 1; - else - memory_exception_data.failure.NoExecute = 0; - } - - mmap_read_unlock(mm); - mmput(mm); - - pr_debug("notpresent %d, noexecute %d, readonly %d\n", - memory_exception_data.failure.NotPresent, - memory_exception_data.failure.NoExecute, - memory_exception_data.failure.ReadOnly); - - /* Workaround on Raven to not kill the process when memory is freed - * before IOMMU is able to finish processing all the excessive PPRs - */ - - if (KFD_GC_VERSION(dev) != IP_VERSION(9, 1, 0) && - KFD_GC_VERSION(dev) != IP_VERSION(9, 2, 2) && - KFD_GC_VERSION(dev) != IP_VERSION(9, 3, 0)) - lookup_events_by_type_and_signal(p, KFD_EVENT_TYPE_MEMORY, - &memory_exception_data); - - kfd_unref_process(p); -} -#endif /* KFD_SUPPORT_IOMMU_V2 */ - void kfd_signal_hw_exception_event(u32 pasid) { /* diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index da2ca00d79e5..62b205dac63a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -322,22 +322,19 @@ static void kfd_init_apertures_vi(struct kfd_process_device *pdd, uint8_t id) pdd->lds_base = MAKE_LDS_APP_BASE_VI(); pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base); - if (!pdd->dev->kfd->use_iommu_v2) { - /* dGPUs: SVM aperture starting at 0 - * with small reserved space for kernel. - * Set them to CANONICAL addresses. - */ - pdd->gpuvm_base = SVM_USER_BASE; - pdd->gpuvm_limit = - pdd->dev->kfd->shared_resources.gpuvm_size - 1; - } else { - /* set them to non CANONICAL addresses, and no SVM is - * allocated. - */ - pdd->gpuvm_base = MAKE_GPUVM_APP_BASE_VI(id + 1); - pdd->gpuvm_limit = MAKE_GPUVM_APP_LIMIT(pdd->gpuvm_base, - pdd->dev->kfd->shared_resources.gpuvm_size); - } + /* dGPUs: SVM aperture starting at 0 + * with small reserved space for kernel. + * Set them to CANONICAL addresses. + */ + pdd->gpuvm_base = SVM_USER_BASE; + pdd->gpuvm_limit = + pdd->dev->kfd->shared_resources.gpuvm_size - 1; + + /* dGPUs: the reserved space for kernel + * before SVM + */ + pdd->qpd.cwsr_base = SVM_CWSR_BASE; + pdd->qpd.ib_base = SVM_IB_BASE; pdd->scratch_base = MAKE_SCRATCH_APP_BASE_VI(); pdd->scratch_limit = MAKE_SCRATCH_APP_LIMIT(pdd->scratch_base); @@ -348,18 +345,18 @@ static void kfd_init_apertures_v9(struct kfd_process_device *pdd, uint8_t id) pdd->lds_base = MAKE_LDS_APP_BASE_V9(); pdd->lds_limit = MAKE_LDS_APP_LIMIT(pdd->lds_base); - /* Raven needs SVM to support graphic handle, etc. Leave the small - * reserved space before SVM on Raven as well, even though we don't - * have to. - * Set gpuvm_base and gpuvm_limit to CANONICAL addresses so that they - * are used in Thunk to reserve SVM. - */ - pdd->gpuvm_base = SVM_USER_BASE; + pdd->gpuvm_base = PAGE_SIZE; pdd->gpuvm_limit = pdd->dev->kfd->shared_resources.gpuvm_size - 1; pdd->scratch_base = MAKE_SCRATCH_APP_BASE_V9(); pdd->scratch_limit = MAKE_SCRATCH_APP_LIMIT(pdd->scratch_base); + + /* + * Place TBA/TMA on opposite side of VM hole to prevent + * stray faults from triggering SVM on these pages. + */ + pdd->qpd.cwsr_base = pdd->dev->kfd->shared_resources.gpuvm_size; } int kfd_init_apertures(struct kfd_process *process) @@ -416,14 +413,6 @@ int kfd_init_apertures(struct kfd_process *process) return -EINVAL; } } - - if (!dev->kfd->use_iommu_v2) { - /* dGPUs: the reserved space for kernel - * before SVM - */ - pdd->qpd.cwsr_base = SVM_CWSR_BASE; - pdd->qpd.ib_base = SVM_IB_BASE; - } } dev_dbg(kfd_device, "node id %u\n", id); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index f0731a6a5306..830396b1c3b1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -384,7 +384,7 @@ static void event_interrupt_wq_v9(struct kfd_node *dev, default: break; } - kfd_signal_event_interrupt(pasid, context_id0 & 0xffffff, 24); + kfd_signal_event_interrupt(pasid, sq_int_data, 24); } else if (source_id == SOC15_INTSRC_CP_BAD_OPCODE) { kfd_set_dbg_ev_from_interrupt(dev, pasid, KFD_DEBUG_DOORBELL_ID(context_id0), diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c deleted file mode 100644 index 808ee010520a..000000000000 --- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c +++ /dev/null @@ -1,356 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 OR MIT -/* - * Copyright 2018-2022 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - */ - -#include <linux/kconfig.h> - -#if IS_REACHABLE(CONFIG_AMD_IOMMU_V2) - -#include <linux/printk.h> -#include <linux/device.h> -#include <linux/slab.h> -#include <linux/pci.h> -#include <linux/amd-iommu.h> -#include "kfd_priv.h" -#include "kfd_topology.h" -#include "kfd_iommu.h" - -static const u32 required_iommu_flags = AMD_IOMMU_DEVICE_FLAG_ATS_SUP | - AMD_IOMMU_DEVICE_FLAG_PRI_SUP | - AMD_IOMMU_DEVICE_FLAG_PASID_SUP; - -/** kfd_iommu_check_device - Check whether IOMMU is available for device - */ -int kfd_iommu_check_device(struct kfd_dev *kfd) -{ - struct amd_iommu_device_info iommu_info; - int err; - - if (!kfd->use_iommu_v2) - return -ENODEV; - - iommu_info.flags = 0; - err = amd_iommu_device_info(kfd->adev->pdev, &iommu_info); - if (err) - return err; - - if ((iommu_info.flags & required_iommu_flags) != required_iommu_flags) - return -ENODEV; - - return 0; -} - -/** kfd_iommu_device_init - Initialize IOMMU for device - */ -int kfd_iommu_device_init(struct kfd_dev *kfd) -{ - struct amd_iommu_device_info iommu_info; - unsigned int pasid_limit; - int err; - - if (!kfd->use_iommu_v2) - return 0; - - iommu_info.flags = 0; - err = amd_iommu_device_info(kfd->adev->pdev, &iommu_info); - if (err < 0) { - dev_err(kfd_device, - "error getting iommu info. is the iommu enabled?\n"); - return -ENODEV; - } - - if ((iommu_info.flags & required_iommu_flags) != required_iommu_flags) { - dev_err(kfd_device, - "error required iommu flags ats %i, pri %i, pasid %i\n", - (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_ATS_SUP) != 0, - (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PRI_SUP) != 0, - (iommu_info.flags & AMD_IOMMU_DEVICE_FLAG_PASID_SUP) - != 0); - return -ENODEV; - } - - pasid_limit = min_t(unsigned int, - (unsigned int)(1 << kfd->device_info.max_pasid_bits), - iommu_info.max_pasids); - - if (!kfd_set_pasid_limit(pasid_limit)) { - dev_err(kfd_device, "error setting pasid limit\n"); - return -EBUSY; - } - - return 0; -} - -/** kfd_iommu_bind_process_to_device - Have the IOMMU bind a process - * - * Binds the given process to the given device using its PASID. This - * enables IOMMUv2 address translation for the process on the device. - * - * This function assumes that the process mutex is held. - */ -int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd) -{ - struct kfd_node *dev = pdd->dev; - struct kfd_process *p = pdd->process; - int err; - - if (!dev->kfd->use_iommu_v2 || pdd->bound == PDD_BOUND) - return 0; - - if (unlikely(pdd->bound == PDD_BOUND_SUSPENDED)) { - pr_err("Binding PDD_BOUND_SUSPENDED pdd is unexpected!\n"); - return -EINVAL; - } - - if (!kfd_is_first_node(dev)) { - dev_warn_once(kfd_device, - "IOMMU supported only on first node\n"); - return 0; - } - - err = amd_iommu_bind_pasid(dev->adev->pdev, p->pasid, p->lead_thread); - if (!err) - pdd->bound = PDD_BOUND; - - return err; -} - -/** kfd_iommu_unbind_process - Unbind process from all devices - * - * This removes all IOMMU device bindings of the process. To be used - * before process termination. - */ -void kfd_iommu_unbind_process(struct kfd_process *p) -{ - int i; - - for (i = 0; i < p->n_pdds; i++) - if ((p->pdds[i]->bound == PDD_BOUND) && - (kfd_is_first_node((p->pdds[i]->dev)))) - amd_iommu_unbind_pasid(p->pdds[i]->dev->adev->pdev, - p->pasid); -} - -/* Callback for process shutdown invoked by the IOMMU driver */ -static void iommu_pasid_shutdown_callback(struct pci_dev *pdev, u32 pasid) -{ - struct kfd_node *dev = kfd_device_by_pci_dev(pdev); - struct kfd_process *p; - struct kfd_process_device *pdd; - - if (!dev) - return; - - /* - * Look for the process that matches the pasid. If there is no such - * process, we either released it in amdkfd's own notifier, or there - * is a bug. Unfortunately, there is no way to tell... - */ - p = kfd_lookup_process_by_pasid(pasid); - if (!p) - return; - - pr_debug("Unbinding process 0x%x from IOMMU\n", pasid); - - mutex_lock(&p->mutex); - - pdd = kfd_get_process_device_data(dev, p); - if (pdd) - /* For GPU relying on IOMMU, we need to dequeue here - * when PASID is still bound. - */ - kfd_process_dequeue_from_device(pdd); - - mutex_unlock(&p->mutex); - - kfd_unref_process(p); -} - -/* This function called by IOMMU driver on PPR failure */ -static int iommu_invalid_ppr_cb(struct pci_dev *pdev, u32 pasid, - unsigned long address, u16 flags) -{ - struct kfd_node *dev; - - dev_warn_ratelimited(kfd_device, - "Invalid PPR device %x:%x.%x pasid 0x%x address 0x%lX flags 0x%X", - pdev->bus->number, - PCI_SLOT(pdev->devfn), - PCI_FUNC(pdev->devfn), - pasid, - address, - flags); - - dev = kfd_device_by_pci_dev(pdev); - if (!WARN_ON(!dev)) - kfd_signal_iommu_event(dev, pasid, address, - flags & PPR_FAULT_WRITE, flags & PPR_FAULT_EXEC); - - return AMD_IOMMU_INV_PRI_RSP_INVALID; -} - -/* - * Bind processes do the device that have been temporarily unbound - * (PDD_BOUND_SUSPENDED) in kfd_unbind_processes_from_device. - */ -static int kfd_bind_processes_to_device(struct kfd_node *knode) -{ - struct kfd_process_device *pdd; - struct kfd_process *p; - unsigned int temp; - int err = 0; - - int idx = srcu_read_lock(&kfd_processes_srcu); - - hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { - mutex_lock(&p->mutex); - pdd = kfd_get_process_device_data(knode, p); - - if (WARN_ON(!pdd) || pdd->bound != PDD_BOUND_SUSPENDED) { - mutex_unlock(&p->mutex); - continue; - } - - err = amd_iommu_bind_pasid(knode->adev->pdev, p->pasid, - p->lead_thread); - if (err < 0) { - pr_err("Unexpected pasid 0x%x binding failure\n", - p->pasid); - mutex_unlock(&p->mutex); - break; - } - - pdd->bound = PDD_BOUND; - mutex_unlock(&p->mutex); - } - - srcu_read_unlock(&kfd_processes_srcu, idx); - - return err; -} - -/* - * Mark currently bound processes as PDD_BOUND_SUSPENDED. These - * processes will be restored to PDD_BOUND state in - * kfd_bind_processes_to_device. - */ -static void kfd_unbind_processes_from_device(struct kfd_node *knode) -{ - struct kfd_process_device *pdd; - struct kfd_process *p; - unsigned int temp; - - int idx = srcu_read_lock(&kfd_processes_srcu); - - hash_for_each_rcu(kfd_processes_table, temp, p, kfd_processes) { - mutex_lock(&p->mutex); - pdd = kfd_get_process_device_data(knode, p); - - if (WARN_ON(!pdd)) { - mutex_unlock(&p->mutex); - continue; - } - - if (pdd->bound == PDD_BOUND) - pdd->bound = PDD_BOUND_SUSPENDED; - mutex_unlock(&p->mutex); - } - - srcu_read_unlock(&kfd_processes_srcu, idx); -} - -/** kfd_iommu_suspend - Prepare IOMMU for suspend - * - * This unbinds processes from the device and disables the IOMMU for - * the device. - */ -void kfd_iommu_suspend(struct kfd_dev *kfd) -{ - if (!kfd->use_iommu_v2) - return; - - kfd_unbind_processes_from_device(kfd->nodes[0]); - - amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL); - amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL); - amd_iommu_free_device(kfd->adev->pdev); -} - -/** kfd_iommu_resume - Restore IOMMU after resume - * - * This reinitializes the IOMMU for the device and re-binds previously - * suspended processes to the device. - */ -int kfd_iommu_resume(struct kfd_dev *kfd) -{ - unsigned int pasid_limit; - int err; - - if (!kfd->use_iommu_v2) - return 0; - - pasid_limit = kfd_get_pasid_limit(); - - err = amd_iommu_init_device(kfd->adev->pdev, pasid_limit); - if (err) - return -ENXIO; - - amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, - iommu_pasid_shutdown_callback); - amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, - iommu_invalid_ppr_cb); - - err = kfd_bind_processes_to_device(kfd->nodes[0]); - if (err) { - amd_iommu_set_invalidate_ctx_cb(kfd->adev->pdev, NULL); - amd_iommu_set_invalid_ppr_cb(kfd->adev->pdev, NULL); - amd_iommu_free_device(kfd->adev->pdev); - return err; - } - - return 0; -} - -/** kfd_iommu_add_perf_counters - Add IOMMU performance counters to topology - */ -int kfd_iommu_add_perf_counters(struct kfd_topology_device *kdev) -{ - struct kfd_perf_properties *props; - - if (!(kdev->node_props.capability & HSA_CAP_ATS_PRESENT)) - return 0; - - if (!amd_iommu_pc_supported()) - return 0; - - props = kfd_alloc_struct(props); - if (!props) - return -ENOMEM; - strcpy(props->block_name, "iommu"); - props->max_concurrent = amd_iommu_pc_get_max_banks(0) * - amd_iommu_pc_get_max_counters(0); /* assume one iommu */ - list_add_tail(&props->list, &kdev->perf_props); - - return 0; -} - -#endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.h b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.h deleted file mode 100644 index 8cf0fcbe87c2..000000000000 --- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.h +++ /dev/null @@ -1,84 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 OR MIT */ -/* - * Copyright 2018-2022 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - */ - -#ifndef __KFD_IOMMU_H__ -#define __KFD_IOMMU_H__ - -#include <linux/kconfig.h> - -#if IS_REACHABLE(CONFIG_AMD_IOMMU_V2) - -#define KFD_SUPPORT_IOMMU_V2 - -int kfd_iommu_check_device(struct kfd_dev *kfd); -int kfd_iommu_device_init(struct kfd_dev *kfd); - -int kfd_iommu_bind_process_to_device(struct kfd_process_device *pdd); -void kfd_iommu_unbind_process(struct kfd_process *p); - -void kfd_iommu_suspend(struct kfd_dev *kfd); -int kfd_iommu_resume(struct kfd_dev *kfd); - -int kfd_iommu_add_perf_counters(struct kfd_topology_device *kdev); - -#else - -static inline int kfd_iommu_check_device(struct kfd_dev *kfd) -{ - return -ENODEV; -} -static inline int kfd_iommu_device_init(struct kfd_dev *kfd) -{ -#if IS_MODULE(CONFIG_AMD_IOMMU_V2) - WARN_ONCE(1, "iommu_v2 module is not usable by built-in KFD"); -#endif - return 0; -} - -static inline int kfd_iommu_bind_process_to_device( - struct kfd_process_device *pdd) -{ - return 0; -} -static inline void kfd_iommu_unbind_process(struct kfd_process *p) -{ - /* empty */ -} - -static inline void kfd_iommu_suspend(struct kfd_dev *kfd) -{ - /* empty */ -} -static inline int kfd_iommu_resume(struct kfd_dev *kfd) -{ - return 0; -} - -static inline int kfd_iommu_add_perf_counters(struct kfd_topology_device *kdev) -{ - return 0; -} - -#endif /* IS_REACHABLE(CONFIG_AMD_IOMMU_V2) */ - -#endif /* __KFD_IOMMU_H__ */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index 709ac885ca6d..7d82c7da223a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -461,7 +461,6 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, 0, node->id, trigger); svm_range_dma_unmap(adev->dev, scratch, 0, npages); - svm_range_free_dma_mappings(prange); out_free: kvfree(buf); @@ -543,10 +542,12 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc, addr = next; } - if (cpages) + if (cpages) { prange->actual_loc = best_loc; - else + svm_range_free_dma_mappings(prange, true); + } else { svm_range_vram_node_free(prange); + } return r < 0 ? r : 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index 863cf060af48..447829c22295 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -48,7 +48,7 @@ int pipe_priority_map[] = { struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properties *q) { - struct kfd_mem_obj *mqd_mem_obj = NULL; + struct kfd_mem_obj *mqd_mem_obj; mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); if (!mqd_mem_obj) @@ -64,7 +64,7 @@ struct kfd_mem_obj *allocate_hiq_mqd(struct kfd_node *dev, struct queue_properti struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_node *dev, struct queue_properties *q) { - struct kfd_mem_obj *mqd_mem_obj = NULL; + struct kfd_mem_obj *mqd_mem_obj; uint64_t offset; mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL); @@ -97,18 +97,22 @@ void free_mqd_hiq_sdma(struct mqd_manager *mm, void *mqd, void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, const uint32_t *cu_mask, uint32_t cu_mask_count, - uint32_t *se_mask) + uint32_t *se_mask, uint32_t inst) { struct kfd_cu_info cu_info; uint32_t cu_per_sh[KFD_MAX_NUM_SE][KFD_MAX_NUM_SH_PER_SE] = {0}; bool wgp_mode_req = KFD_GC_VERSION(mm->dev) >= IP_VERSION(10, 0, 0); uint32_t en_mask = wgp_mode_req ? 0x3 : 0x1; - int i, se, sh, cu, cu_bitmap_sh_mul, inc = wgp_mode_req ? 2 : 1; + int i, se, sh, cu, cu_bitmap_sh_mul, cu_inc = wgp_mode_req ? 2 : 1; + uint32_t cu_active_per_node; + int inc = cu_inc * NUM_XCC(mm->dev->xcc_mask); + int xcc_inst = inst + ffs(mm->dev->xcc_mask) - 1; amdgpu_amdkfd_get_cu_info(mm->dev->adev, &cu_info); - if (cu_mask_count > cu_info.cu_active_number) - cu_mask_count = cu_info.cu_active_number; + cu_active_per_node = cu_info.cu_active_number / mm->dev->kfd->num_nodes; + if (cu_mask_count > cu_active_per_node) + cu_mask_count = cu_active_per_node; /* Exceeding these bounds corrupts the stack and indicates a coding error. * Returning with no CU's enabled will hang the queue, which should be @@ -141,7 +145,8 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, for (se = 0; se < cu_info.num_shader_engines; se++) for (sh = 0; sh < cu_info.num_shader_arrays_per_engine; sh++) cu_per_sh[se][sh] = hweight32( - cu_info.cu_bitmap[se % 4][sh + (se / 4) * cu_bitmap_sh_mul]); + cu_info.cu_bitmap[xcc_inst][se % 4][sh + (se / 4) * + cu_bitmap_sh_mul]); /* Symmetrically map cu_mask to all SEs & SHs: * se_mask programs up to 2 SH in the upper and lower 16 bits. @@ -164,20 +169,33 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, * cu_mask[0] bit8 -> se_mask[0] bit1 (SE0,SH0,CU1) * ... * + * For GFX 9.4.3, the following code only looks at a + * subset of the cu_mask corresponding to the inst parameter. + * If we have n XCCs under one GPU node + * cu_mask[0] bit0 -> XCC0 se_mask[0] bit0 (XCC0,SE0,SH0,CU0) + * cu_mask[0] bit1 -> XCC1 se_mask[0] bit0 (XCC1,SE0,SH0,CU0) + * .. + * cu_mask[0] bitn -> XCCn se_mask[0] bit0 (XCCn,SE0,SH0,CU0) + * cu_mask[0] bit n+1 -> XCC0 se_mask[1] bit0 (XCC0,SE1,SH0,CU0) + * + * For example, if there are 6 XCCs under 1 KFD node, this code + * running for each inst, will look at the bits as: + * inst, inst + 6, inst + 12... + * * First ensure all CUs are disabled, then enable user specified CUs. */ for (i = 0; i < cu_info.num_shader_engines; i++) se_mask[i] = 0; - i = 0; - for (cu = 0; cu < 16; cu += inc) { + i = inst; + for (cu = 0; cu < 16; cu += cu_inc) { for (sh = 0; sh < cu_info.num_shader_arrays_per_engine; sh++) { for (se = 0; se < cu_info.num_shader_engines; se++) { if (cu_per_sh[se][sh] > cu) { if (cu_mask[i / 32] & (en_mask << (i % 32))) se_mask[se] |= en_mask << (cu + sh * 16); i += inc; - if (i == cu_mask_count) + if (i >= cu_mask_count) return; } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h index 23158db7da03..57bf5e513f4d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.h @@ -138,7 +138,7 @@ void free_mqd_hiq_sdma(struct mqd_manager *mm, void *mqd, void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm, const uint32_t *cu_mask, uint32_t cu_mask_count, - uint32_t *se_mask); + uint32_t *se_mask, uint32_t inst); int kfd_hiq_load_mqd_kiq(struct mqd_manager *mm, void *mqd, uint32_t pipe_id, uint32_t queue_id, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c index 65c9f01a1f86..1a4a69943c71 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_cik.c @@ -52,7 +52,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, return; mqd_symmetrically_map_cu_mask(mm, - minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask); + minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask, 0); m = get_mqd(mqd); m->compute_static_thread_mgmt_se0 = se_mask[0]; @@ -206,13 +206,6 @@ static void __update_mqd(struct mqd_manager *mm, void *mqd, q->is_active = QUEUE_IS_ACTIVE(*q); } -static void update_mqd(struct mqd_manager *mm, void *mqd, - struct queue_properties *q, - struct mqd_update_info *minfo) -{ - __update_mqd(mm, mqd, q, minfo, 1); -} - static uint32_t read_doorbell_id(void *mqd) { struct cik_mqd *m = (struct cik_mqd *)mqd; @@ -220,9 +213,9 @@ static uint32_t read_doorbell_id(void *mqd) return m->queue_doorbell_id0; } -static void update_mqd_hawaii(struct mqd_manager *mm, void *mqd, - struct queue_properties *q, - struct mqd_update_info *minfo) +static void update_mqd(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, + struct mqd_update_info *minfo) { __update_mqd(mm, mqd, q, minfo, 0); } @@ -387,7 +380,6 @@ static int debugfs_show_mqd_sdma(struct seq_file *m, void *data) #endif - struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, struct kfd_node *dev) { @@ -470,16 +462,3 @@ struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, return mqd; } - -struct mqd_manager *mqd_manager_init_cik_hawaii(enum KFD_MQD_TYPE type, - struct kfd_node *dev) -{ - struct mqd_manager *mqd; - - mqd = mqd_manager_init_cik(type, dev); - if (!mqd) - return NULL; - if (type == KFD_MQD_TYPE_CP) - mqd->update_mqd = update_mqd_hawaii; - return mqd; -} diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c index 94c0fc2e57b7..8b7fed913526 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v10.c @@ -52,7 +52,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, return; mqd_symmetrically_map_cu_mask(mm, - minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask); + minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask, 0); m = get_mqd(mqd); m->compute_static_thread_mgmt_se0 = se_mask[0]; @@ -318,6 +318,26 @@ static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_PQ_CONTROL__KMD_QUEUE__SHIFT; } +static int destroy_hiq_mqd(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id) +{ + int err; + struct v10_compute_mqd *m; + u32 doorbell_off; + + m = get_mqd(mqd); + + doorbell_off = m->cp_hqd_pq_doorbell_control >> + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + + err = amdgpu_amdkfd_unmap_hiq(mm->dev->adev, doorbell_off, 0); + if (err) + pr_debug("Destroy HIQ MQD failed: %d\n", err); + + return err; +} + static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -460,7 +480,7 @@ struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, mqd->free_mqd = free_mqd_hiq_sdma; mqd->load_mqd = kfd_hiq_load_mqd_kiq; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->destroy_mqd = destroy_hiq_mqd; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v10_compute_mqd); mqd->mqd_stride = kfd_mqd_stride; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c index 31fec5e70d13..15277f1d5cf0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c @@ -71,7 +71,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, } mqd_symmetrically_map_cu_mask(mm, - minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask); + minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask, 0); m->compute_static_thread_mgmt_se0 = se_mask[0]; m->compute_static_thread_mgmt_se1 = se_mask[1]; @@ -321,6 +321,43 @@ static int get_wave_state(struct mqd_manager *mm, void *mqd, return 0; } +static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, void *ctl_stack_dst) +{ + struct v11_compute_mqd *m; + + m = get_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct v11_compute_mqd)); +} + +static void restore_mqd(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, const u32 ctl_stack_size) +{ + uint64_t addr; + struct v11_compute_mqd *m; + + m = (struct v11_compute_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + m->cp_hqd_pq_doorbell_control = + qp->doorbell_off << + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", + m->cp_hqd_pq_doorbell_control); + + qp->is_active = 0; +} + + static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -335,6 +372,26 @@ static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_PQ_CONTROL__KMD_QUEUE__SHIFT; } +static int destroy_hiq_mqd(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id) +{ + int err; + struct v11_compute_mqd *m; + u32 doorbell_off; + + m = get_mqd(mqd); + + doorbell_off = m->cp_hqd_pq_doorbell_control >> + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + + err = amdgpu_amdkfd_unmap_hiq(mm->dev->adev, doorbell_off, 0); + if (err) + pr_debug("Destroy HIQ MQD failed: %d\n", err); + + return err; +} + static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -437,6 +494,9 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v11_compute_mqd); mqd->get_wave_state = get_wave_state; + mqd->mqd_stride = kfd_mqd_stride; + mqd->checkpoint_mqd = checkpoint_mqd; + mqd->restore_mqd = restore_mqd; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -449,9 +509,10 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, mqd->free_mqd = free_mqd_hiq_sdma; mqd->load_mqd = kfd_hiq_load_mqd_kiq; mqd->update_mqd = update_mqd; - mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->destroy_mqd = destroy_hiq_mqd; mqd->is_occupied = kfd_is_occupied_cp; mqd->mqd_size = sizeof(struct v11_compute_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd; #endif @@ -480,7 +541,10 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, mqd->update_mqd = update_mqd_sdma; mqd->destroy_mqd = kfd_destroy_mqd_sdma; mqd->is_occupied = kfd_is_occupied_sdma; + mqd->checkpoint_mqd = checkpoint_mqd; + mqd->restore_mqd = restore_mqd; mqd->mqd_size = sizeof(struct v11_sdma_mqd); + mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) mqd->debugfs_show_mqd = debugfs_show_mqd_sdma; #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c index 601bb9f68048..42d881809dc7 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c @@ -60,7 +60,7 @@ static inline struct v9_sdma_mqd *get_sdma_mqd(void *mqd) } static void update_cu_mask(struct mqd_manager *mm, void *mqd, - struct mqd_update_info *minfo) + struct mqd_update_info *minfo, uint32_t inst) { struct v9_mqd *m; uint32_t se_mask[KFD_MAX_NUM_SE] = {0}; @@ -69,27 +69,36 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, return; mqd_symmetrically_map_cu_mask(mm, - minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask); + minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask, inst); m = get_mqd(mqd); + m->compute_static_thread_mgmt_se0 = se_mask[0]; m->compute_static_thread_mgmt_se1 = se_mask[1]; m->compute_static_thread_mgmt_se2 = se_mask[2]; m->compute_static_thread_mgmt_se3 = se_mask[3]; - m->compute_static_thread_mgmt_se4 = se_mask[4]; - m->compute_static_thread_mgmt_se5 = se_mask[5]; - m->compute_static_thread_mgmt_se6 = se_mask[6]; - m->compute_static_thread_mgmt_se7 = se_mask[7]; - - pr_debug("update cu mask to %#x %#x %#x %#x %#x %#x %#x %#x\n", - m->compute_static_thread_mgmt_se0, - m->compute_static_thread_mgmt_se1, - m->compute_static_thread_mgmt_se2, - m->compute_static_thread_mgmt_se3, - m->compute_static_thread_mgmt_se4, - m->compute_static_thread_mgmt_se5, - m->compute_static_thread_mgmt_se6, - m->compute_static_thread_mgmt_se7); + if (KFD_GC_VERSION(mm->dev) != IP_VERSION(9, 4, 3)) { + m->compute_static_thread_mgmt_se4 = se_mask[4]; + m->compute_static_thread_mgmt_se5 = se_mask[5]; + m->compute_static_thread_mgmt_se6 = se_mask[6]; + m->compute_static_thread_mgmt_se7 = se_mask[7]; + + pr_debug("update cu mask to %#x %#x %#x %#x %#x %#x %#x %#x\n", + m->compute_static_thread_mgmt_se0, + m->compute_static_thread_mgmt_se1, + m->compute_static_thread_mgmt_se2, + m->compute_static_thread_mgmt_se3, + m->compute_static_thread_mgmt_se4, + m->compute_static_thread_mgmt_se5, + m->compute_static_thread_mgmt_se6, + m->compute_static_thread_mgmt_se7); + } else { + pr_debug("inst: %u, update cu mask to %#x %#x %#x %#x\n", + inst, m->compute_static_thread_mgmt_se0, + m->compute_static_thread_mgmt_se1, + m->compute_static_thread_mgmt_se2, + m->compute_static_thread_mgmt_se3); + } } static void set_priority(struct v9_mqd *m, struct queue_properties *q) @@ -290,7 +299,8 @@ static void update_mqd(struct mqd_manager *mm, void *mqd, if (mm->dev->kfd->cwsr_enabled && q->ctx_save_restore_area_address) m->cp_hqd_ctx_save_control = 0; - update_cu_mask(mm, mqd, minfo); + if (KFD_GC_VERSION(mm->dev) != IP_VERSION(9, 4, 3)) + update_cu_mask(mm, mqd, minfo, 0); set_priority(m, q); q->is_active = QUEUE_IS_ACTIVE(*q); @@ -405,6 +415,25 @@ static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, 1 << CP_HQD_PQ_CONTROL__KMD_QUEUE__SHIFT; } +static int destroy_hiq_mqd(struct mqd_manager *mm, void *mqd, + enum kfd_preempt_type type, unsigned int timeout, + uint32_t pipe_id, uint32_t queue_id) +{ + int err; + struct v9_mqd *m; + u32 doorbell_off; + + m = get_mqd(mqd); + + doorbell_off = m->cp_hqd_pq_doorbell_control >> + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + err = amdgpu_amdkfd_unmap_hiq(mm->dev->adev, doorbell_off, 0); + if (err) + pr_debug("Destroy HIQ MQD failed: %d\n", err); + + return err; +} + static void init_mqd_sdma(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, struct queue_properties *q) @@ -548,16 +577,19 @@ static int destroy_hiq_mqd_v9_4_3(struct mqd_manager *mm, void *mqd, { uint32_t xcc_mask = mm->dev->xcc_mask; int xcc_id, err, inst = 0; - void *xcc_mqd; uint64_t hiq_mqd_size = kfd_hiq_mqd_stride(mm->dev); + struct v9_mqd *m; + u32 doorbell_off; for_each_inst(xcc_id, xcc_mask) { - xcc_mqd = mqd + hiq_mqd_size * inst; - err = mm->dev->kfd2kgd->hqd_destroy(mm->dev->adev, xcc_mqd, - type, timeout, pipe_id, - queue_id, xcc_id); + m = get_mqd(mqd + hiq_mqd_size * inst); + + doorbell_off = m->cp_hqd_pq_doorbell_control >> + CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + + err = amdgpu_amdkfd_unmap_hiq(mm->dev->adev, doorbell_off, xcc_id); if (err) { - pr_debug("Destroy MQD failed for xcc: %d\n", inst); + pr_debug("Destroy HIQ MQD failed for xcc: %d\n", inst); break; } ++inst; @@ -654,6 +686,8 @@ static void update_mqd_v9_4_3(struct mqd_manager *mm, void *mqd, m = get_mqd(mqd + size * xcc); update_mqd(mm, m, q, minfo); + update_cu_mask(mm, mqd, minfo, xcc); + if (q->format == KFD_QUEUE_FORMAT_AQL) { switch (xcc) { case 0: @@ -846,7 +880,7 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, } else { mqd->init_mqd = init_mqd_hiq; mqd->load_mqd = kfd_hiq_load_mqd_kiq; - mqd->destroy_mqd = kfd_destroy_mqd_cp; + mqd->destroy_mqd = destroy_hiq_mqd; } break; case KFD_MQD_TYPE_DIQ: diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c index d1e962da51dd..3e1a574d4ea6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_vi.c @@ -55,7 +55,7 @@ static void update_cu_mask(struct mqd_manager *mm, void *mqd, return; mqd_symmetrically_map_cu_mask(mm, - minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask); + minfo->cu_mask.ptr, minfo->cu_mask.count, se_mask, 0); m = get_mqd(mqd); m->compute_static_thread_mgmt_se0 = se_mask[0]; @@ -237,14 +237,6 @@ static void __update_mqd(struct mqd_manager *mm, void *mqd, q->is_active = QUEUE_IS_ACTIVE(*q); } - -static void update_mqd(struct mqd_manager *mm, void *mqd, - struct queue_properties *q, - struct mqd_update_info *minfo) -{ - __update_mqd(mm, mqd, q, minfo, MTYPE_CC, 1); -} - static uint32_t read_doorbell_id(void *mqd) { struct vi_mqd *m = (struct vi_mqd *)mqd; @@ -252,9 +244,9 @@ static uint32_t read_doorbell_id(void *mqd) return m->queue_doorbell_id0; } -static void update_mqd_tonga(struct mqd_manager *mm, void *mqd, - struct queue_properties *q, - struct mqd_update_info *minfo) +static void update_mqd(struct mqd_manager *mm, void *mqd, + struct queue_properties *q, + struct mqd_update_info *minfo) { __update_mqd(mm, mqd, q, minfo, MTYPE_UC, 0); } @@ -529,16 +521,3 @@ struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, return mqd; } - -struct mqd_manager *mqd_manager_init_vi_tonga(enum KFD_MQD_TYPE type, - struct kfd_node *dev) -{ - struct mqd_manager *mqd; - - mqd = mqd_manager_init_vi(type, dev); - if (!mqd) - return NULL; - if (type == KFD_MQD_TYPE_CP) - mqd->update_mqd = update_mqd_tonga; - return mqd; -} diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c index 29a2d0499b67..1a03173e2313 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_v9.c @@ -121,6 +121,7 @@ static int pm_map_process_aldebaran(struct packet_manager *pm, packet->sh_mem_bases = qpd->sh_mem_bases; if (qpd->tba_addr) { packet->sq_shader_tba_lo = lower_32_bits(qpd->tba_addr >> 8); + packet->sq_shader_tba_hi = upper_32_bits(qpd->tba_addr >> 8); packet->sq_shader_tma_lo = lower_32_bits(qpd->tma_addr >> 8); packet->sq_shader_tma_hi = upper_32_bits(qpd->tma_addr >> 8); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index d4c9ee3f9953..fa24e1852493 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -175,12 +175,6 @@ extern int send_sigterm; */ extern int debug_largebar; -/* - * Ignore CRAT table during KFD initialization, can be used to work around - * broken CRAT tables on some AMD systems - */ -extern int ignore_crat; - /* Set sh_mem_config.retry_disable on GFX v9 */ extern int amdgpu_noretry; @@ -234,7 +228,6 @@ struct kfd_device_info { uint8_t num_of_watch_points; uint16_t mqd_size_aligned; bool supports_cwsr; - bool needs_iommu_device; bool needs_pci_atomics; uint32_t no_atomic_fw_version; unsigned int num_sdma_queues_per_engine; @@ -323,15 +316,6 @@ struct kfd_dev { struct kfd_device_info device_info; - phys_addr_t doorbell_base; /* Start of actual doorbells used by - * KFD. It is aligned for mapping - * into user mode - */ - size_t doorbell_base_dw_offset; /* Offset from the start of the PCI - * doorbell BAR to the first KFD - * doorbell in dwords. GFX reserves - * the segment before this offset. - */ u32 __iomem *doorbell_kernel_ptr; /* This is a pointer for a doorbells * page used by kernel queue */ @@ -340,8 +324,6 @@ struct kfd_dev { const struct kfd2kgd_calls *kfd2kgd; struct mutex doorbell_mutex; - DECLARE_BITMAP(doorbell_available_index, - KFD_MAX_NUM_OF_QUEUES_PER_PROCESS); void *gtt_mem; uint64_t gtt_start_gpu_addr; @@ -368,9 +350,6 @@ struct kfd_dev { bool pci_atomic_requested; - /* Use IOMMU v2 flag */ - bool use_iommu_v2; - /* Compute Profile ref. count */ atomic_t compute_profile; @@ -385,6 +364,12 @@ struct kfd_dev { /* Track per device allocated watch points */ uint32_t alloc_watch_ids; spinlock_t watch_points_lock; + + /* Kernel doorbells for KFD device */ + struct amdgpu_bo *doorbells; + + /* bitmap for dynamic doorbell allocation from doorbell object */ + unsigned long *doorbell_bitmap; }; enum kfd_mempool { @@ -702,7 +687,10 @@ struct qcm_process_device { uint64_t ib_base; void *ib_kaddr; - /* doorbell resources per process per device */ + /* doorbells for kfd process */ + struct amdgpu_bo *proc_doorbells; + + /* bitmap for dynamic doorbell allocation from the bo */ unsigned long *doorbell_bitmap; }; @@ -792,7 +780,6 @@ struct kfd_process_device { struct attribute attr_evict; struct kobject *kobj_stats; - unsigned int doorbell_index; /* * @cu_occupancy: Reports occupancy of Compute Units (CU) of a process @@ -1100,9 +1087,9 @@ unsigned int kfd_get_doorbell_dw_offset_in_bar(struct kfd_dev *kfd, unsigned int doorbell_id); phys_addr_t kfd_get_process_doorbells(struct kfd_process_device *pdd); int kfd_alloc_process_doorbells(struct kfd_dev *kfd, - unsigned int *doorbell_index); + struct kfd_process_device *pdd); void kfd_free_process_doorbells(struct kfd_dev *kfd, - unsigned int doorbell_index); + struct kfd_process_device *pdd); /* GTT Sub-Allocator */ int kfd_gtt_sa_allocate(struct kfd_node *node, unsigned int size, @@ -1152,7 +1139,6 @@ static inline struct kfd_node *kfd_node_by_irq_ids(struct amdgpu_device *adev, } int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_node **kdev); int kfd_numa_node_to_apic_id(int numa_node_id); -void kfd_double_confirm_iommu_support(struct kfd_dev *gpu); /* Interrupts */ #define KFD_IRQ_FENCE_CLIENTID 0xff @@ -1299,12 +1285,8 @@ void print_queue(struct queue *q); struct mqd_manager *mqd_manager_init_cik(enum KFD_MQD_TYPE type, struct kfd_node *dev); -struct mqd_manager *mqd_manager_init_cik_hawaii(enum KFD_MQD_TYPE type, - struct kfd_node *dev); struct mqd_manager *mqd_manager_init_vi(enum KFD_MQD_TYPE type, struct kfd_node *dev); -struct mqd_manager *mqd_manager_init_vi_tonga(enum KFD_MQD_TYPE type, - struct kfd_node *dev); struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type, struct kfd_node *dev); struct mqd_manager *mqd_manager_init_v10(enum KFD_MQD_TYPE type, @@ -1459,9 +1441,6 @@ int kfd_wait_on_events(struct kfd_process *p, uint32_t *wait_result); void kfd_signal_event_interrupt(u32 pasid, uint32_t partial_id, uint32_t valid_id_bits); -void kfd_signal_iommu_event(struct kfd_node *dev, - u32 pasid, unsigned long address, - bool is_write_requested, bool is_execute_requested); void kfd_signal_hw_exception_event(u32 pasid); int kfd_set_event(struct kfd_process *p, uint32_t event_id); int kfd_reset_event(struct kfd_process *p, uint32_t event_id); @@ -1487,8 +1466,7 @@ void kfd_flush_tlb(struct kfd_process_device *pdd, enum TLB_FLUSH_TYPE type); static inline bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) { - return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 3) || - KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) || + return KFD_GC_VERSION(dev) > IP_VERSION(9, 4, 2) || (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) && dev->sdma_fw_version >= 18) || KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0); } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index a844e68211ac..fbf053001af9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -28,7 +28,6 @@ #include <linux/sched/task.h> #include <linux/mmu_context.h> #include <linux/slab.h> -#include <linux/amd-iommu.h> #include <linux/notifier.h> #include <linux/compat.h> #include <linux/mman.h> @@ -41,7 +40,6 @@ struct mm_struct; #include "kfd_priv.h" #include "kfd_device_queue_manager.h" -#include "kfd_iommu.h" #include "kfd_svm.h" #include "kfd_smi_events.h" #include "kfd_debug.h" @@ -1035,10 +1033,9 @@ static void kfd_process_destroy_pdds(struct kfd_process *p) free_pages((unsigned long)pdd->qpd.cwsr_kaddr, get_order(KFD_CWSR_TBA_TMA_SIZE)); - bitmap_free(pdd->qpd.doorbell_bitmap); idr_destroy(&pdd->alloc_idr); - kfd_free_process_doorbells(pdd->dev->kfd, pdd->doorbell_index); + kfd_free_process_doorbells(pdd->dev->kfd, pdd); if (pdd->dev->kfd->shared_resources.enable_mes) amdgpu_amdkfd_free_gtt_mem(pdd->dev->adev, @@ -1123,7 +1120,6 @@ static void kfd_process_wq_release(struct work_struct *work) dma_fence_signal(p->ef); kfd_process_remove_sysfs(p); - kfd_iommu_unbind_process(p); kfd_process_kunmap_signal_bo(p); kfd_process_free_outstanding_kfd_bos(p); @@ -1550,38 +1546,6 @@ err_alloc_process: return ERR_PTR(err); } -static int init_doorbell_bitmap(struct qcm_process_device *qpd, - struct kfd_dev *dev) -{ - unsigned int i; - int range_start = dev->shared_resources.non_cp_doorbells_start; - int range_end = dev->shared_resources.non_cp_doorbells_end; - - if (!KFD_IS_SOC15(dev)) - return 0; - - qpd->doorbell_bitmap = bitmap_zalloc(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, - GFP_KERNEL); - if (!qpd->doorbell_bitmap) - return -ENOMEM; - - /* Mask out doorbells reserved for SDMA, IH, and VCN on SOC15. */ - pr_debug("reserved doorbell 0x%03x - 0x%03x\n", range_start, range_end); - pr_debug("reserved doorbell 0x%03x - 0x%03x\n", - range_start + KFD_QUEUE_DOORBELL_MIRROR_OFFSET, - range_end + KFD_QUEUE_DOORBELL_MIRROR_OFFSET); - - for (i = 0; i < KFD_MAX_NUM_OF_QUEUES_PER_PROCESS / 2; i++) { - if (i >= range_start && i <= range_end) { - __set_bit(i, qpd->doorbell_bitmap); - __set_bit(i + KFD_QUEUE_DOORBELL_MIRROR_OFFSET, - qpd->doorbell_bitmap); - } - } - - return 0; -} - struct kfd_process_device *kfd_get_process_device_data(struct kfd_node *dev, struct kfd_process *p) { @@ -1606,11 +1570,6 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, if (!pdd) return NULL; - if (init_doorbell_bitmap(&pdd->qpd, dev->kfd)) { - pr_err("Failed to init doorbell for process\n"); - goto err_free_pdd; - } - pdd->dev = dev; INIT_LIST_HEAD(&pdd->qpd.queues_list); INIT_LIST_HEAD(&pdd->qpd.priv_queue_list); @@ -1766,10 +1725,6 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev, } } - err = kfd_iommu_bind_process_to_device(pdd); - if (err) - goto out; - /* * make sure that runtime_usage counter is incremented just once * per pdd @@ -1777,15 +1732,6 @@ struct kfd_process_device *kfd_bind_process_to_device(struct kfd_node *dev, pdd->runtime_inuse = true; return pdd; - -out: - /* balance runpm reference count and exit with error */ - if (!pdd->runtime_inuse) { - pm_runtime_mark_last_busy(adev_to_drm(dev->adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(dev->adev)->dev); - } - - return ERR_PTR(err); } /* Create specific handle mapped to mem from process local memory idr diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index ba9d69054119..adb5e4bdc0b2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -123,7 +123,7 @@ int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid, if (!gws && pdd->qpd.num_gws == 0) return -EINVAL; - if (KFD_GC_VERSION(dev) != IP_VERSION(9, 4, 3)) { + if (KFD_GC_VERSION(dev) != IP_VERSION(9, 4, 3) && !dev->kfd->shared_resources.enable_mes) { if (gws) ret = amdgpu_amdkfd_add_gws_to_process(pdd->process->kgd_process_info, gws, &mem); @@ -136,7 +136,9 @@ int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid, } else { /* * Intentionally set GWS to a non-NULL value - * for GFX 9.4.3. + * for devices that do not use GWS for global wave + * synchronization but require the formality + * of setting GWS for cooperative groups. */ pqn->q->gws = gws ? ERR_PTR(-ENOMEM) : NULL; } @@ -173,7 +175,8 @@ void pqm_uninit(struct process_queue_manager *pqm) list_for_each_entry_safe(pqn, next, &pqm->queues, process_queue_list) { if (pqn->q && pqn->q->gws && - KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3)) + KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3) && + !pqn->q->device->kfd->shared_resources.enable_mes) amdgpu_amdkfd_remove_gws_from_process(pqm->process->kgd_process_info, pqn->q->gws); kfd_procfs_del_queue(pqn->q); @@ -365,17 +368,20 @@ int pqm_create_queue(struct process_queue_manager *pqm, goto err_create_queue; } - if (q && p_doorbell_offset_in_process) + if (q && p_doorbell_offset_in_process) { /* Return the doorbell offset within the doorbell page * to the caller so it can be passed up to user mode * (in bytes). - * There are always 1024 doorbells per process, so in case - * of 8-byte doorbells, there are two doorbell pages per - * process. + * relative doorbell index = Absolute doorbell index - + * absolute index of first doorbell in the page. */ - *p_doorbell_offset_in_process = - (q->properties.doorbell_off * sizeof(uint32_t)) & - (kfd_doorbell_process_slice(dev->kfd) - 1); + uint32_t first_db_index = amdgpu_doorbell_index_on_bar(pdd->dev->adev, + pdd->qpd.proc_doorbells, + 0); + + *p_doorbell_offset_in_process = (q->properties.doorbell_off + - first_db_index) * sizeof(uint32_t); + } pr_debug("PQM After DQM create queue\n"); @@ -455,7 +461,8 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid) } if (pqn->q->gws) { - if (KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3)) + if (KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3) && + !dev->kfd->shared_resources.enable_mes) amdgpu_amdkfd_remove_gws_from_process( pqm->process->kgd_process_info, pqn->q->gws); @@ -929,12 +936,6 @@ int kfd_criu_restore_queue(struct kfd_process *p, goto exit; } - if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) { - ret = -ENOMEM; - goto exit; - } - /* data stored in this order: mqd, ctl_stack */ mqd = q_extra_data; ctl_stack = mqd + q_data->mqd_size; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index aa32b06eea77..bb16b795d1bc 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -23,6 +23,7 @@ #include <linux/types.h> #include <linux/sched/task.h> +#include <linux/dynamic_debug.h> #include <drm/ttm/ttm_tt.h> #include <drm/drm_exec.h> @@ -48,6 +49,13 @@ * page table is updated. */ #define AMDGPU_SVM_RANGE_RETRY_FAULT_PENDING (2UL * NSEC_PER_MSEC) +#if IS_ENABLED(CONFIG_DYNAMIC_DEBUG) +#define dynamic_svm_range_dump(svms) \ + _dynamic_func_call_no_desc("svm_range_dump", svm_range_debug_dump, svms) +#else +#define dynamic_svm_range_dump(svms) \ + do { if (0) svm_range_debug_dump(svms); } while (0) +#endif /* Giant svm range split into smaller ranges based on this, it is decided using * minimum of all dGPU/APU 1/32 VRAM size, between 2MB to 1GB and alignment to @@ -241,7 +249,7 @@ void svm_range_dma_unmap(struct device *dev, dma_addr_t *dma_addr, } } -void svm_range_free_dma_mappings(struct svm_range *prange) +void svm_range_free_dma_mappings(struct svm_range *prange, bool unmap_dma) { struct kfd_process_device *pdd; dma_addr_t *dma_addr; @@ -262,13 +270,14 @@ void svm_range_free_dma_mappings(struct svm_range *prange) continue; } dev = &pdd->dev->adev->pdev->dev; - svm_range_dma_unmap(dev, dma_addr, 0, prange->npages); + if (unmap_dma) + svm_range_dma_unmap(dev, dma_addr, 0, prange->npages); kvfree(dma_addr); prange->dma_addr[gpuidx] = NULL; } } -static void svm_range_free(struct svm_range *prange, bool update_mem_usage) +static void svm_range_free(struct svm_range *prange, bool do_unmap) { uint64_t size = (prange->last - prange->start + 1) << PAGE_SHIFT; struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms); @@ -277,9 +286,9 @@ static void svm_range_free(struct svm_range *prange, bool update_mem_usage) prange->start, prange->last); svm_range_vram_node_free(prange); - svm_range_free_dma_mappings(prange); + svm_range_free_dma_mappings(prange, do_unmap); - if (update_mem_usage && !p->xnack_enabled) { + if (do_unmap && !p->xnack_enabled) { pr_debug("unreserve prange 0x%p size: 0x%llx\n", prange, size); amdgpu_amdkfd_unreserve_mem_limit(NULL, size, KFD_IOC_ALLOC_MEM_FLAGS_USERPTR, 0); @@ -851,6 +860,37 @@ static void svm_range_debug_dump(struct svm_range_list *svms) } } +static void * +svm_range_copy_array(void *psrc, size_t size, uint64_t num_elements, + uint64_t offset) +{ + unsigned char *dst; + + dst = kvmalloc_array(num_elements, size, GFP_KERNEL); + if (!dst) + return NULL; + memcpy(dst, (unsigned char *)psrc + offset, num_elements * size); + + return (void *)dst; +} + +static int +svm_range_copy_dma_addrs(struct svm_range *dst, struct svm_range *src) +{ + int i; + + for (i = 0; i < MAX_GPU_INSTANCE; i++) { + if (!src->dma_addr[i]) + continue; + dst->dma_addr[i] = svm_range_copy_array(src->dma_addr[i], + sizeof(*src->dma_addr[i]), src->npages, 0); + if (!dst->dma_addr[i]) + return -ENOMEM; + } + + return 0; +} + static int svm_range_split_array(void *ppnew, void *ppold, size_t size, uint64_t old_start, uint64_t old_n, @@ -865,22 +905,16 @@ svm_range_split_array(void *ppnew, void *ppold, size_t size, if (!pold) return 0; - new = kvmalloc_array(new_n, size, GFP_KERNEL); + d = (new_start - old_start) * size; + new = svm_range_copy_array(pold, size, new_n, d); if (!new) return -ENOMEM; - - d = (new_start - old_start) * size; - memcpy(new, pold + d, new_n * size); - - old = kvmalloc_array(old_n, size, GFP_KERNEL); + d = (new_start == old_start) ? new_n * size : 0; + old = svm_range_copy_array(pold, size, old_n, d); if (!old) { kvfree(new); return -ENOMEM; } - - d = (new_start == old_start) ? new_n * size : 0; - memcpy(old, pold + d, old_n * size); - kvfree(pold); *(void **)ppold = old; *(void **)ppnew = new; @@ -1521,6 +1555,8 @@ static void *kfd_svm_page_owner(struct kfd_process *p, int32_t gpuidx) struct kfd_process_device *pdd; pdd = kfd_process_device_from_gpuidx(p, gpuidx); + if (!pdd) + return NULL; return SVM_ADEV_PGMAP_OWNER(pdd->dev->adev); } @@ -1595,12 +1631,12 @@ static int svm_range_validate_and_map(struct mm_struct *mm, } if (bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) { - if (!prange->mapped_to_gpu) { + bitmap_copy(ctx->bitmap, prange->bitmap_access, MAX_GPU_INSTANCE); + if (!prange->mapped_to_gpu || + bitmap_empty(ctx->bitmap, MAX_GPU_INSTANCE)) { r = 0; goto free_ctx; } - - bitmap_copy(ctx->bitmap, prange->bitmap_access, MAX_GPU_INSTANCE); } if (prange->actual_loc && !prange->ttm_res) { @@ -1650,6 +1686,8 @@ static int svm_range_validate_and_map(struct mm_struct *mm, WRITE_ONCE(p->svms.faulting_task, NULL); if (r) { pr_debug("failed %d to get svm range pages\n", r); + if (r == -EBUSY) + r = -EAGAIN; goto unreserve_out; } @@ -1924,7 +1962,10 @@ static struct svm_range *svm_range_clone(struct svm_range *old) new = svm_range_new(old->svms, old->start, old->last, false); if (!new) return NULL; - + if (svm_range_copy_dma_addrs(new, old)) { + svm_range_free(new, false); + return NULL; + } if (old->svm_bo) { new->ttm_res = old->ttm_res; new->offset = old->offset; @@ -2620,10 +2661,7 @@ svm_range_get_range_boundaries(struct kfd_process *p, int64_t addr, return -EFAULT; } - *is_heap_stack = (vma->vm_start <= vma->vm_mm->brk && - vma->vm_end >= vma->vm_mm->start_brk) || - (vma->vm_start <= vma->vm_mm->start_stack && - vma->vm_end >= vma->vm_mm->start_stack); + *is_heap_stack = vma_is_initial_heap(vma) || vma_is_initial_stack(vma); start_limit = max(vma->vm_start >> PAGE_SHIFT, (unsigned long)ALIGN_DOWN(addr, 2UL << 8)); @@ -3560,7 +3598,7 @@ out_unlock_range: break; } - svm_range_debug_dump(svms); + dynamic_svm_range_dump(svms); mutex_unlock(&svms->lock); mmap_read_unlock(mm); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h index 21b14510882b..9e668eeefb32 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.h @@ -183,7 +183,7 @@ void svm_range_add_list_work(struct svm_range_list *svms, void schedule_deferred_list_work(struct svm_range_list *svms); void svm_range_dma_unmap(struct device *dev, dma_addr_t *dma_addr, unsigned long offset, unsigned long npages); -void svm_range_free_dma_mappings(struct svm_range *prange); +void svm_range_free_dma_mappings(struct svm_range *prange, bool unmap_dma); int svm_range_get_info(struct kfd_process *p, uint32_t *num_svm_ranges, uint64_t *svm_priv_data_size); int kfd_criu_checkpoint_svm(struct kfd_process *p, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 61fc62f3e003..c8c75ff7cea8 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -36,8 +36,8 @@ #include "kfd_crat.h" #include "kfd_topology.h" #include "kfd_device_queue_manager.h" -#include "kfd_iommu.h" #include "kfd_svm.h" +#include "kfd_debug.h" #include "amdgpu_amdkfd.h" #include "amdgpu_ras.h" #include "amdgpu.h" @@ -450,8 +450,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, sysfs_show_32bit_prop(buffer, offs, "cpu_cores_count", dev->node_props.cpu_cores_count); sysfs_show_32bit_prop(buffer, offs, "simd_count", - dev->gpu ? (dev->node_props.simd_count * - NUM_XCC(dev->gpu->xcc_mask)) : 0); + dev->gpu ? dev->node_props.simd_count : 0); sysfs_show_32bit_prop(buffer, offs, "mem_banks_count", dev->node_props.mem_banks_count); sysfs_show_32bit_prop(buffer, offs, "caches_count", @@ -988,17 +987,6 @@ static void find_system_memory(const struct dmi_header *dm, } } -/* - * Performance counters information is not part of CRAT but we would like to - * put them in the sysfs under topology directory for Thunk to get the data. - * This function is called before updating the sysfs. - */ -static int kfd_add_perf_to_topology(struct kfd_topology_device *kdev) -{ - /* These are the only counters supported so far */ - return kfd_iommu_add_perf_counters(kdev); -} - /* kfd_add_non_crat_information - Add information that is not currently * defined in CRAT but is necessary for KFD topology * @dev - topology device to which addition info is added @@ -1013,25 +1001,6 @@ static void kfd_add_non_crat_information(struct kfd_topology_device *kdev) /* TODO: For GPU node, rearrange code from kfd_topology_add_device */ } -/* kfd_is_acpi_crat_invalid - CRAT from ACPI is valid only for AMD APU devices. - * Ignore CRAT for all other devices. AMD APU is identified if both CPU - * and GPU cores are present. - * @device_list - topology device list created by parsing ACPI CRAT table. - * @return - TRUE if invalid, FALSE is valid. - */ -static bool kfd_is_acpi_crat_invalid(struct list_head *device_list) -{ - struct kfd_topology_device *dev; - - list_for_each_entry(dev, device_list, list) { - if (dev->node_props.cpu_cores_count && - dev->node_props.simd_count) - return false; - } - pr_info("Ignoring ACPI CRAT on non-APU system\n"); - return true; -} - int kfd_topology_init(void) { void *crat_image = NULL; @@ -1062,48 +1031,25 @@ int kfd_topology_init(void) */ proximity_domain = 0; - /* - * Get the CRAT image from the ACPI. If ACPI doesn't have one - * or if ACPI CRAT is invalid create a virtual CRAT. - * NOTE: The current implementation expects all AMD APUs to have - * CRAT. If no CRAT is available, it is assumed to be a CPU - */ - ret = kfd_create_crat_image_acpi(&crat_image, &image_size); - if (!ret) { - ret = kfd_parse_crat_table(crat_image, - &temp_topology_device_list, - proximity_domain); - if (ret || - kfd_is_acpi_crat_invalid(&temp_topology_device_list)) { - kfd_release_topology_device_list( - &temp_topology_device_list); - kfd_destroy_crat_image(crat_image); - crat_image = NULL; - } + ret = kfd_create_crat_image_virtual(&crat_image, &image_size, + COMPUTE_UNIT_CPU, NULL, + proximity_domain); + cpu_only_node = 1; + if (ret) { + pr_err("Error creating VCRAT table for CPU\n"); + return ret; } - if (!crat_image) { - ret = kfd_create_crat_image_virtual(&crat_image, &image_size, - COMPUTE_UNIT_CPU, NULL, - proximity_domain); - cpu_only_node = 1; - if (ret) { - pr_err("Error creating VCRAT table for CPU\n"); - return ret; - } - - ret = kfd_parse_crat_table(crat_image, - &temp_topology_device_list, - proximity_domain); - if (ret) { - pr_err("Error parsing VCRAT table for CPU\n"); - goto err; - } + ret = kfd_parse_crat_table(crat_image, + &temp_topology_device_list, + proximity_domain); + if (ret) { + pr_err("Error parsing VCRAT table for CPU\n"); + goto err; } kdev = list_first_entry(&temp_topology_device_list, struct kfd_topology_device, list); - kfd_add_perf_to_topology(kdev); down_write(&topology_lock); kfd_topology_update_device_list(&temp_topology_device_list, @@ -1189,8 +1135,7 @@ static struct kfd_topology_device *kfd_assign_gpu(struct kfd_node *gpu) /* Discrete GPUs need their own topology device list * entries. Don't assign them to CPU/APU nodes. */ - if (!gpu->kfd->use_iommu_v2 && - dev->node_props.cpu_cores_count) + if (dev->node_props.cpu_cores_count) continue; if (!dev->gpu && (dev->node_props.simd_count > 0)) { @@ -1651,14 +1596,17 @@ static int fill_in_l1_pcache(struct kfd_cache_properties **props_ext, static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, struct kfd_gpu_cache_info *pcache_info, struct kfd_cu_info *cu_info, - int cache_type, unsigned int cu_processor_id) + int cache_type, unsigned int cu_processor_id, + struct kfd_node *knode) { unsigned int cu_sibling_map_mask; int first_active_cu; - int i, j, k; + int i, j, k, xcc, start, end; struct kfd_cache_properties *pcache = NULL; - cu_sibling_map_mask = cu_info->cu_bitmap[0][0]; + start = ffs(knode->xcc_mask) - 1; + end = start + NUM_XCC(knode->xcc_mask); + cu_sibling_map_mask = cu_info->cu_bitmap[start][0][0]; cu_sibling_map_mask &= ((1 << pcache_info[cache_type].num_cu_shared) - 1); first_active_cu = ffs(cu_sibling_map_mask); @@ -1693,16 +1641,18 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, cu_sibling_map_mask = cu_sibling_map_mask >> (first_active_cu - 1); k = 0; - for (i = 0; i < cu_info->num_shader_engines; i++) { - for (j = 0; j < cu_info->num_shader_arrays_per_engine; j++) { - pcache->sibling_map[k] = (uint8_t)(cu_sibling_map_mask & 0xFF); - pcache->sibling_map[k+1] = (uint8_t)((cu_sibling_map_mask >> 8) & 0xFF); - pcache->sibling_map[k+2] = (uint8_t)((cu_sibling_map_mask >> 16) & 0xFF); - pcache->sibling_map[k+3] = (uint8_t)((cu_sibling_map_mask >> 24) & 0xFF); - k += 4; - - cu_sibling_map_mask = cu_info->cu_bitmap[i % 4][j + i / 4]; - cu_sibling_map_mask &= ((1 << pcache_info[cache_type].num_cu_shared) - 1); + for (xcc = start; xcc < end; xcc++) { + for (i = 0; i < cu_info->num_shader_engines; i++) { + for (j = 0; j < cu_info->num_shader_arrays_per_engine; j++) { + pcache->sibling_map[k] = (uint8_t)(cu_sibling_map_mask & 0xFF); + pcache->sibling_map[k+1] = (uint8_t)((cu_sibling_map_mask >> 8) & 0xFF); + pcache->sibling_map[k+2] = (uint8_t)((cu_sibling_map_mask >> 16) & 0xFF); + pcache->sibling_map[k+3] = (uint8_t)((cu_sibling_map_mask >> 24) & 0xFF); + k += 4; + + cu_sibling_map_mask = cu_info->cu_bitmap[xcc][i % 4][j + i / 4]; + cu_sibling_map_mask &= ((1 << pcache_info[cache_type].num_cu_shared) - 1); + } } } pcache->sibling_map_size = k; @@ -1720,7 +1670,7 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct kfd_node *kdev) { struct kfd_gpu_cache_info *pcache_info = NULL; - int i, j, k; + int i, j, k, xcc, start, end; int ct = 0; unsigned int cu_processor_id; int ret; @@ -1754,37 +1704,42 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct * then it will consider only one CU from * the shared unit */ + start = ffs(kdev->xcc_mask) - 1; + end = start + NUM_XCC(kdev->xcc_mask); + for (ct = 0; ct < num_of_cache_types; ct++) { cu_processor_id = gpu_processor_id; if (pcache_info[ct].cache_level == 1) { - for (i = 0; i < pcu_info->num_shader_engines; i++) { - for (j = 0; j < pcu_info->num_shader_arrays_per_engine; j++) { - for (k = 0; k < pcu_info->num_cu_per_sh; k += pcache_info[ct].num_cu_shared) { + for (xcc = start; xcc < end; xcc++) { + for (i = 0; i < pcu_info->num_shader_engines; i++) { + for (j = 0; j < pcu_info->num_shader_arrays_per_engine; j++) { + for (k = 0; k < pcu_info->num_cu_per_sh; k += pcache_info[ct].num_cu_shared) { - ret = fill_in_l1_pcache(&props_ext, pcache_info, pcu_info, - pcu_info->cu_bitmap[i % 4][j + i / 4], ct, + ret = fill_in_l1_pcache(&props_ext, pcache_info, pcu_info, + pcu_info->cu_bitmap[xcc][i % 4][j + i / 4], ct, cu_processor_id, k); - if (ret < 0) - break; + if (ret < 0) + break; - if (!ret) { - num_of_entries++; - list_add_tail(&props_ext->list, &dev->cache_props); - } + if (!ret) { + num_of_entries++; + list_add_tail(&props_ext->list, &dev->cache_props); + } - /* Move to next CU block */ - num_cu_shared = ((k + pcache_info[ct].num_cu_shared) <= - pcu_info->num_cu_per_sh) ? - pcache_info[ct].num_cu_shared : - (pcu_info->num_cu_per_sh - k); - cu_processor_id += num_cu_shared; + /* Move to next CU block */ + num_cu_shared = ((k + pcache_info[ct].num_cu_shared) <= + pcu_info->num_cu_per_sh) ? + pcache_info[ct].num_cu_shared : + (pcu_info->num_cu_per_sh - k); + cu_processor_id += num_cu_shared; + } } } } } else { ret = fill_in_l2_l3_pcache(&props_ext, pcache_info, - pcu_info, ct, cu_processor_id); + pcu_info, ct, cu_processor_id, kdev); if (ret < 0) break; @@ -1931,23 +1886,27 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev) HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_TRAP_OVERRIDE_SUPPORTED | HSA_CAP_TRAP_DEBUG_WAVE_LAUNCH_MODE_SUPPORTED; - if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0)) { - dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9 | - HSA_DBG_WATCH_ADDR_MASK_HI_BIT; + if (kfd_dbg_has_ttmps_always_setup(dev->gpu)) + dev->node_props.debug_prop |= HSA_DBG_DISPATCH_INFO_ALWAYS_VALID; - if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(9, 4, 2)) + if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(10, 0, 0)) { + if (KFD_GC_VERSION(dev->gpu) == IP_VERSION(9, 4, 3)) dev->node_props.debug_prop |= - HSA_DBG_DISPATCH_INFO_ALWAYS_VALID; + HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9_4_3 | + HSA_DBG_WATCH_ADDR_MASK_HI_BIT_GFX9_4_3; else + dev->node_props.debug_prop |= + HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9 | + HSA_DBG_WATCH_ADDR_MASK_HI_BIT; + + if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(9, 4, 2)) dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED; } else { dev->node_props.debug_prop |= HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 | HSA_DBG_WATCH_ADDR_MASK_HI_BIT; - if (KFD_GC_VERSION(dev->gpu) < IP_VERSION(11, 0, 0)) - dev->node_props.debug_prop |= HSA_DBG_DISPATCH_INFO_ALWAYS_VALID; - else + if (KFD_GC_VERSION(dev->gpu) >= IP_VERSION(11, 0, 0)) dev->node_props.capability |= HSA_CAP_TRAP_DEBUG_PRECISE_MEMORY_OPERATIONS_SUPPORTED; } @@ -1965,7 +1924,14 @@ int kfd_topology_add_device(struct kfd_node *gpu) const char *asic_name = amdgpu_asic_name[gpu->adev->asic_type]; gpu_id = kfd_generate_gpu_id(gpu); - pr_debug("Adding new GPU (ID: 0x%x) to topology\n", gpu_id); + if (gpu->xcp && !gpu->xcp->ddev) { + dev_warn(gpu->adev->dev, + "Won't add GPU (ID: 0x%x) to topology since it has no drm node assigned.", + gpu_id); + return 0; + } else { + pr_debug("Adding new GPU (ID: 0x%x) to topology\n", gpu_id); + } /* Check to see if this gpu device exists in the topology_device_list. * If so, assign the gpu to that device, @@ -2076,10 +2042,7 @@ int kfd_topology_add_device(struct kfd_node *gpu) * Overwrite ATS capability according to needs_iommu_device to fix * potential missing corresponding bit in CRAT of BIOS. */ - if (dev->gpu->kfd->use_iommu_v2) - dev->node_props.capability |= HSA_CAP_ATS_PRESENT; - else - dev->node_props.capability &= ~HSA_CAP_ATS_PRESENT; + dev->node_props.capability &= ~HSA_CAP_ATS_PRESENT; /* Fix errors in CZ CRAT. * simd_count: Carrizo CRAT reports wrong simd_count, probably @@ -2274,29 +2237,6 @@ int kfd_numa_node_to_apic_id(int numa_node_id) return kfd_cpumask_to_apic_id(cpumask_of_node(numa_node_id)); } -void kfd_double_confirm_iommu_support(struct kfd_dev *gpu) -{ - struct kfd_topology_device *dev; - - gpu->use_iommu_v2 = false; - - if (!gpu->device_info.needs_iommu_device) - return; - - down_read(&topology_lock); - - /* Only use IOMMUv2 if there is an APU topology node with no GPU - * assigned yet. This GPU will be assigned to it. - */ - list_for_each_entry(dev, &topology_device_list, list) - if (dev->node_props.cpu_cores_count && - dev->node_props.simd_count && - !dev->gpu) - gpu->use_iommu_v2 = true; - - up_read(&topology_lock); -} - #if defined(CONFIG_DEBUG_FS) int kfd_debugfs_hqds_by_device(struct seq_file *m, void *data) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h index cba2cd5ed9d1..27386ce9a021 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.h @@ -32,9 +32,12 @@ #define KFD_TOPOLOGY_PUBLIC_NAME_SIZE 32 #define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9 6 +#define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX9_4_3 7 #define HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 7 #define HSA_DBG_WATCH_ADDR_MASK_HI_BIT \ (29 << HSA_DBG_WATCH_ADDR_MASK_HI_BIT_SHIFT) +#define HSA_DBG_WATCH_ADDR_MASK_HI_BIT_GFX9_4_3 \ + (30 << HSA_DBG_WATCH_ADDR_MASK_HI_BIT_SHIFT) struct kfd_node_properties { uint64_t hive_id; @@ -86,7 +89,7 @@ struct kfd_mem_properties { struct attribute attr; }; -#define CACHE_SIBLINGMAP_SIZE 64 +#define CACHE_SIBLINGMAP_SIZE 128 struct kfd_cache_properties { struct list_head list; diff --git a/drivers/gpu/drm/amd/display/Kconfig b/drivers/gpu/drm/amd/display/Kconfig index bf0a655d009e..901d1961b739 100644 --- a/drivers/gpu/drm/amd/display/Kconfig +++ b/drivers/gpu/drm/amd/display/Kconfig @@ -5,7 +5,7 @@ menu "Display Engine Configuration" config DRM_AMD_DC bool "AMD DC - Enable new display engine" default y - depends on BROKEN || !CC_IS_CLANG || X86_64 || SPARC64 || ARM64 + depends on BROKEN || !CC_IS_CLANG || ARM64 || RISCV || SPARC64 || X86_64 select SND_HDA_COMPONENT if SND_HDA_CORE # !CC_IS_CLANG: https://github.com/ClangBuiltLinux/linux/issues/1752 select DRM_AMD_DC_FP if (X86 || LOONGARCH || (PPC64 && ALTIVEC) || (ARM64 && KERNEL_MODE_NEON && !CC_IS_CLANG)) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile b/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile index 249b073f6a23..8bf94920d23e 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/Makefile @@ -38,7 +38,7 @@ AMDGPUDM += dc_fpu.o endif ifneq ($(CONFIG_DRM_AMD_DC),) -AMDGPUDM += amdgpu_dm_services.o amdgpu_dm_helpers.o amdgpu_dm_pp_smu.o amdgpu_dm_psr.o +AMDGPUDM += amdgpu_dm_services.o amdgpu_dm_helpers.o amdgpu_dm_pp_smu.o amdgpu_dm_psr.o amdgpu_dm_replay.o endif AMDGPUDM += amdgpu_dm_hdcp.o diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index ff0a217b9d56..c6fd34bab358 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -65,6 +65,7 @@ #include "amdgpu_dm_debugfs.h" #endif #include "amdgpu_dm_psr.h" +#include "amdgpu_dm_replay.h" #include "ivsrcid/ivsrcid_vislands30.h" @@ -245,51 +246,52 @@ is_timing_unchanged_for_freesync(struct drm_crtc_state *old_crtc_state, */ static u32 dm_vblank_get_counter(struct amdgpu_device *adev, int crtc) { + struct amdgpu_crtc *acrtc = NULL; + if (crtc >= adev->mode_info.num_crtc) return 0; - else { - struct amdgpu_crtc *acrtc = adev->mode_info.crtcs[crtc]; - if (acrtc->dm_irq_params.stream == NULL) { - DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n", - crtc); - return 0; - } + acrtc = adev->mode_info.crtcs[crtc]; - return dc_stream_get_vblank_counter(acrtc->dm_irq_params.stream); + if (!acrtc->dm_irq_params.stream) { + DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n", + crtc); + return 0; } + + return dc_stream_get_vblank_counter(acrtc->dm_irq_params.stream); } static int dm_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc, u32 *vbl, u32 *position) { u32 v_blank_start, v_blank_end, h_position, v_position; + struct amdgpu_crtc *acrtc = NULL; if ((crtc < 0) || (crtc >= adev->mode_info.num_crtc)) return -EINVAL; - else { - struct amdgpu_crtc *acrtc = adev->mode_info.crtcs[crtc]; - if (acrtc->dm_irq_params.stream == NULL) { - DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n", - crtc); - return 0; - } + acrtc = adev->mode_info.crtcs[crtc]; - /* - * TODO rework base driver to use values directly. - * for now parse it back into reg-format - */ - dc_stream_get_scanoutpos(acrtc->dm_irq_params.stream, - &v_blank_start, - &v_blank_end, - &h_position, - &v_position); - - *position = v_position | (h_position << 16); - *vbl = v_blank_start | (v_blank_end << 16); + if (!acrtc->dm_irq_params.stream) { + DRM_ERROR("dc_stream_state is NULL for crtc '%d'!\n", + crtc); + return 0; } + /* + * TODO rework base driver to use values directly. + * for now parse it back into reg-format + */ + dc_stream_get_scanoutpos(acrtc->dm_irq_params.stream, + &v_blank_start, + &v_blank_end, + &h_position, + &v_position); + + *position = v_position | (h_position << 16); + *vbl = v_blank_start | (v_blank_end << 16); + return 0; } @@ -424,12 +426,12 @@ static void dm_pflip_high_irq(void *interrupt_params) spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags); - if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){ - DC_LOG_PFLIP("amdgpu_crtc->pflip_status = %d !=AMDGPU_FLIP_SUBMITTED(%d) on crtc:%d[%p] \n", - amdgpu_crtc->pflip_status, - AMDGPU_FLIP_SUBMITTED, - amdgpu_crtc->crtc_id, - amdgpu_crtc); + if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED) { + DC_LOG_PFLIP("amdgpu_crtc->pflip_status = %d !=AMDGPU_FLIP_SUBMITTED(%d) on crtc:%d[%p]\n", + amdgpu_crtc->pflip_status, + AMDGPU_FLIP_SUBMITTED, + amdgpu_crtc->crtc_id, + amdgpu_crtc); spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags); return; } @@ -883,7 +885,7 @@ static int dm_set_powergating_state(void *handle, } /* Prototypes of private functions */ -static int dm_early_init(void* handle); +static int dm_early_init(void *handle); /* Allocate memory for FBC compressed data */ static void amdgpu_dm_fbc_init(struct drm_connector *connector) @@ -1272,17 +1274,21 @@ static void mmhub_read_system_context(struct amdgpu_device *adev, struct dc_phy_ pt_base = amdgpu_gmc_pd_addr(adev->gart.bo); - page_table_start.high_part = (u32)(adev->gmc.gart_start >> 44) & 0xF; - page_table_start.low_part = (u32)(adev->gmc.gart_start >> 12); - page_table_end.high_part = (u32)(adev->gmc.gart_end >> 44) & 0xF; - page_table_end.low_part = (u32)(adev->gmc.gart_end >> 12); - page_table_base.high_part = upper_32_bits(pt_base) & 0xF; + page_table_start.high_part = upper_32_bits(adev->gmc.gart_start >> + AMDGPU_GPU_PAGE_SHIFT); + page_table_start.low_part = lower_32_bits(adev->gmc.gart_start >> + AMDGPU_GPU_PAGE_SHIFT); + page_table_end.high_part = upper_32_bits(adev->gmc.gart_end >> + AMDGPU_GPU_PAGE_SHIFT); + page_table_end.low_part = lower_32_bits(adev->gmc.gart_end >> + AMDGPU_GPU_PAGE_SHIFT); + page_table_base.high_part = upper_32_bits(pt_base); page_table_base.low_part = lower_32_bits(pt_base); pa_config->system_aperture.start_addr = (uint64_t)logical_addr_low << 18; pa_config->system_aperture.end_addr = (uint64_t)logical_addr_high << 18; - pa_config->system_aperture.agp_base = (uint64_t)agp_base << 24 ; + pa_config->system_aperture.agp_base = (uint64_t)agp_base << 24; pa_config->system_aperture.agp_bot = (uint64_t)agp_bot << 24; pa_config->system_aperture.agp_top = (uint64_t)agp_top << 24; @@ -1347,6 +1353,15 @@ static void dm_handle_hpd_rx_offload_work(struct work_struct *work) if (amdgpu_in_reset(adev)) goto skip; + if (offload_work->data.bytes.device_service_irq.bits.UP_REQ_MSG_RDY || + offload_work->data.bytes.device_service_irq.bits.DOWN_REP_MSG_RDY) { + dm_handle_mst_sideband_msg_ready_event(&aconnector->mst_mgr, DOWN_OR_UP_MSG_RDY_EVENT); + spin_lock_irqsave(&offload_work->offload_wq->offload_lock, flags); + offload_work->offload_wq->is_handling_mst_msg_rdy_event = false; + spin_unlock_irqrestore(&offload_work->offload_wq->offload_lock, flags); + goto skip; + } + mutex_lock(&adev->dm.dc_lock); if (offload_work->data.bytes.device_service_irq.bits.AUTOMATED_TEST) { dc_link_dp_handle_automated_test(dc_link); @@ -1365,8 +1380,7 @@ static void dm_handle_hpd_rx_offload_work(struct work_struct *work) DP_TEST_RESPONSE, &test_response.raw, sizeof(test_response)); - } - else if ((dc_link->connector_signal != SIGNAL_TYPE_EDP) && + } else if ((dc_link->connector_signal != SIGNAL_TYPE_EDP) && dc_link_check_link_loss_status(dc_link, &offload_work->data) && dc_link_dp_allow_hpd_rx_irq(dc_link)) { /* offload_work->data is from handle_hpd_rx_irq-> @@ -1554,7 +1568,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) mutex_init(&adev->dm.dc_lock); mutex_init(&adev->dm.audio_lock); - if(amdgpu_dm_irq_init(adev)) { + if (amdgpu_dm_irq_init(adev)) { DRM_ERROR("amdgpu: failed to initialize DM IRQ support.\n"); goto error; } @@ -1696,9 +1710,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (amdgpu_dc_debug_mask & DC_DISABLE_STUTTER) adev->dm.dc->debug.disable_stutter = true; - if (amdgpu_dc_debug_mask & DC_DISABLE_DSC) { + if (amdgpu_dc_debug_mask & DC_DISABLE_DSC) adev->dm.dc->debug.disable_dsc = true; - } if (amdgpu_dc_debug_mask & DC_DISABLE_CLOCK_GATING) adev->dm.dc->debug.disable_clock_gate = true; @@ -1815,9 +1828,14 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) * It is expected that DMUB will resend any pending notifications at this point, for * example HPD from DPIA. */ - if (dc_is_dmub_outbox_supported(adev->dm.dc)) + if (dc_is_dmub_outbox_supported(adev->dm.dc)) { dc_enable_dmub_outbox(adev->dm.dc); + /* DPIA trace goes to dmesg logs only if outbox is enabled */ + if (amdgpu_dc_debug_mask & DC_ENABLE_DPIA_TRACE) + dc_dmub_srv_enable_dpia_trace(adev->dm.dc); + } + if (amdgpu_dm_initialize_drm_device(adev)) { DRM_ERROR( "amdgpu: failed to initialize sw for display support.\n"); @@ -1942,8 +1960,6 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev) mutex_destroy(&adev->dm.audio_lock); mutex_destroy(&adev->dm.dc_lock); mutex_destroy(&adev->dm.dpia_aux_lock); - - return; } static int load_dmcu_fw(struct amdgpu_device *adev) @@ -1952,7 +1968,7 @@ static int load_dmcu_fw(struct amdgpu_device *adev) int r; const struct dmcu_firmware_header_v1_0 *hdr; - switch(adev->asic_type) { + switch (adev->asic_type) { #if defined(CONFIG_DRM_AMD_DC_SI) case CHIP_TAHITI: case CHIP_PITCAIRN: @@ -2324,14 +2340,62 @@ static int dm_late_init(void *handle) return detect_mst_link_for_all_connectors(adev_to_drm(adev)); } +static void resume_mst_branch_status(struct drm_dp_mst_topology_mgr *mgr) +{ + int ret; + u8 guid[16]; + u64 tmp64; + + mutex_lock(&mgr->lock); + if (!mgr->mst_primary) + goto out_fail; + + if (drm_dp_read_dpcd_caps(mgr->aux, mgr->dpcd) < 0) { + drm_dbg_kms(mgr->dev, "dpcd read failed - undocked during suspend?\n"); + goto out_fail; + } + + ret = drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL, + DP_MST_EN | + DP_UP_REQ_EN | + DP_UPSTREAM_IS_SRC); + if (ret < 0) { + drm_dbg_kms(mgr->dev, "mst write failed - undocked during suspend?\n"); + goto out_fail; + } + + /* Some hubs forget their guids after they resume */ + ret = drm_dp_dpcd_read(mgr->aux, DP_GUID, guid, 16); + if (ret != 16) { + drm_dbg_kms(mgr->dev, "dpcd read failed - undocked during suspend?\n"); + goto out_fail; + } + + if (memchr_inv(guid, 0, 16) == NULL) { + tmp64 = get_jiffies_64(); + memcpy(&guid[0], &tmp64, sizeof(u64)); + memcpy(&guid[8], &tmp64, sizeof(u64)); + + ret = drm_dp_dpcd_write(mgr->aux, DP_GUID, guid, 16); + + if (ret != 16) { + drm_dbg_kms(mgr->dev, "check mstb guid failed - undocked during suspend?\n"); + goto out_fail; + } + } + + memcpy(mgr->mst_primary->guid, guid, 16); + +out_fail: + mutex_unlock(&mgr->lock); +} + static void s3_handle_mst(struct drm_device *dev, bool suspend) { struct amdgpu_dm_connector *aconnector; struct drm_connector *connector; struct drm_connector_list_iter iter; struct drm_dp_mst_topology_mgr *mgr; - int ret; - bool need_hotplug = false; drm_connector_list_iter_begin(dev, &iter); drm_for_each_connector_iter(connector, &iter) { @@ -2353,18 +2417,15 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend) if (!dp_is_lttpr_present(aconnector->dc_link)) try_to_configure_aux_timeout(aconnector->dc_link->ddc, LINK_AUX_DEFAULT_TIMEOUT_PERIOD); - ret = drm_dp_mst_topology_mgr_resume(mgr, true); - if (ret < 0) { - dm_helpers_dp_mst_stop_top_mgr(aconnector->dc_link->ctx, - aconnector->dc_link); - need_hotplug = true; - } + /* TODO: move resume_mst_branch_status() into drm mst resume again + * once topology probing work is pulled out from mst resume into mst + * resume 2nd step. mst resume 2nd step should be called after old + * state getting restored (i.e. drm_atomic_helper_resume()). + */ + resume_mst_branch_status(mgr); } } drm_connector_list_iter_end(&iter); - - if (need_hotplug) - drm_kms_helper_hotplug_event(dev); } static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) @@ -2709,7 +2770,7 @@ static void dm_gpureset_commit_state(struct dc_state *dc_state, struct dc_scaling_info scaling_infos[MAX_SURFACES]; struct dc_flip_addrs flip_addrs[MAX_SURFACES]; struct dc_stream_update stream_update; - } * bundle; + } *bundle; int k, m; bundle = kzalloc(sizeof(*bundle), GFP_KERNEL); @@ -2739,8 +2800,6 @@ static void dm_gpureset_commit_state(struct dc_state *dc_state, cleanup: kfree(bundle); - - return; } static int dm_resume(void *handle) @@ -2760,7 +2819,8 @@ static int dm_resume(void *handle) struct dm_atomic_state *dm_state = to_dm_atomic_state(dm->atomic_obj.state); enum dc_connection_type new_connection_type = dc_connection_none; struct dc_state *dc_state; - int i, r, j; + int i, r, j, ret; + bool need_hotplug = false; if (amdgpu_in_reset(adev)) { dc_state = dm->cached_dc_state; @@ -2858,7 +2918,7 @@ static int dm_resume(void *handle) continue; /* - * this is the case when traversing through already created + * this is the case when traversing through already created end sink * MST connectors, should be skipped */ if (aconnector && aconnector->mst_root) @@ -2918,6 +2978,27 @@ static int dm_resume(void *handle) dm->cached_state = NULL; + /* Do mst topology probing after resuming cached state*/ + drm_connector_list_iter_begin(ddev, &iter); + drm_for_each_connector_iter(connector, &iter) { + aconnector = to_amdgpu_dm_connector(connector); + if (aconnector->dc_link->type != dc_connection_mst_branch || + aconnector->mst_root) + continue; + + ret = drm_dp_mst_topology_mgr_resume(&aconnector->mst_mgr, true); + + if (ret < 0) { + dm_helpers_dp_mst_stop_top_mgr(aconnector->dc_link->ctx, + aconnector->dc_link); + need_hotplug = true; + } + } + drm_connector_list_iter_end(&iter); + + if (need_hotplug) + drm_kms_helper_hotplug_event(ddev); + amdgpu_dm_irq_resume_late(adev); amdgpu_dm_smu_write_watermarks_table(adev); @@ -2954,8 +3035,7 @@ static const struct amd_ip_funcs amdgpu_dm_funcs = { .set_powergating_state = dm_set_powergating_state, }; -const struct amdgpu_ip_block_version dm_ip_block = -{ +const struct amdgpu_ip_block_version dm_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 1, .minor = 0, @@ -3000,9 +3080,12 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector) caps->ext_caps = &aconnector->dc_link->dpcd_sink_ext_caps; caps->aux_support = false; - if (caps->ext_caps->bits.oled == 1 /*|| - caps->ext_caps->bits.sdr_aux_backlight_control == 1 || - caps->ext_caps->bits.hdr_aux_backlight_control == 1*/) + if (caps->ext_caps->bits.oled == 1 + /* + * || + * caps->ext_caps->bits.sdr_aux_backlight_control == 1 || + * caps->ext_caps->bits.hdr_aux_backlight_control == 1 + */) caps->aux_support = true; if (amdgpu_backlight == 0) @@ -3236,86 +3319,6 @@ static void handle_hpd_irq(void *param) } -static void dm_handle_mst_sideband_msg(struct amdgpu_dm_connector *aconnector) -{ - u8 esi[DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI] = { 0 }; - u8 dret; - bool new_irq_handled = false; - int dpcd_addr; - int dpcd_bytes_to_read; - - const int max_process_count = 30; - int process_count = 0; - - const struct dc_link_status *link_status = dc_link_get_status(aconnector->dc_link); - - if (link_status->dpcd_caps->dpcd_rev.raw < 0x12) { - dpcd_bytes_to_read = DP_LANE0_1_STATUS - DP_SINK_COUNT; - /* DPCD 0x200 - 0x201 for downstream IRQ */ - dpcd_addr = DP_SINK_COUNT; - } else { - dpcd_bytes_to_read = DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI; - /* DPCD 0x2002 - 0x2005 for downstream IRQ */ - dpcd_addr = DP_SINK_COUNT_ESI; - } - - dret = drm_dp_dpcd_read( - &aconnector->dm_dp_aux.aux, - dpcd_addr, - esi, - dpcd_bytes_to_read); - - while (dret == dpcd_bytes_to_read && - process_count < max_process_count) { - u8 ack[DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI] = {}; - u8 retry; - dret = 0; - - process_count++; - - DRM_DEBUG_DRIVER("ESI %02x %02x %02x\n", esi[0], esi[1], esi[2]); - /* handle HPD short pulse irq */ - if (aconnector->mst_mgr.mst_state) - drm_dp_mst_hpd_irq_handle_event(&aconnector->mst_mgr, - esi, - ack, - &new_irq_handled); - - if (new_irq_handled) { - /* ACK at DPCD to notify down stream */ - for (retry = 0; retry < 3; retry++) { - ssize_t wret; - - wret = drm_dp_dpcd_writeb(&aconnector->dm_dp_aux.aux, - dpcd_addr + 1, - ack[1]); - if (wret == 1) - break; - } - - if (retry == 3) { - DRM_ERROR("Failed to ack MST event.\n"); - return; - } - - drm_dp_mst_hpd_irq_send_new_request(&aconnector->mst_mgr); - /* check if there is new irq to be handled */ - dret = drm_dp_dpcd_read( - &aconnector->dm_dp_aux.aux, - dpcd_addr, - esi, - dpcd_bytes_to_read); - - new_irq_handled = false; - } else { - break; - } - } - - if (process_count == max_process_count) - DRM_DEBUG_DRIVER("Loop exceeded max iterations\n"); -} - static void schedule_hpd_rx_offload_work(struct hpd_rx_irq_offload_work_queue *offload_wq, union hpd_irq_data hpd_irq_data) { @@ -3377,7 +3380,23 @@ static void handle_hpd_rx_irq(void *param) if (dc_link_dp_allow_hpd_rx_irq(dc_link)) { if (hpd_irq_data.bytes.device_service_irq.bits.UP_REQ_MSG_RDY || hpd_irq_data.bytes.device_service_irq.bits.DOWN_REP_MSG_RDY) { - dm_handle_mst_sideband_msg(aconnector); + bool skip = false; + + /* + * DOWN_REP_MSG_RDY is also handled by polling method + * mgr->cbs->poll_hpd_irq() + */ + spin_lock(&offload_wq->offload_lock); + skip = offload_wq->is_handling_mst_msg_rdy_event; + + if (!skip) + offload_wq->is_handling_mst_msg_rdy_event = true; + + spin_unlock(&offload_wq->offload_lock); + + if (!skip) + schedule_hpd_rx_offload_work(offload_wq, hpd_irq_data); + goto out; } @@ -3468,7 +3487,7 @@ static void register_hpd_handlers(struct amdgpu_device *adev) aconnector = to_amdgpu_dm_connector(connector); dc_link = aconnector->dc_link; - if (DC_IRQ_SOURCE_INVALID != dc_link->irq_source_hpd) { + if (dc_link->irq_source_hpd != DC_IRQ_SOURCE_INVALID) { int_params.int_context = INTERRUPT_LOW_IRQ_CONTEXT; int_params.irq_source = dc_link->irq_source_hpd; @@ -3477,7 +3496,7 @@ static void register_hpd_handlers(struct amdgpu_device *adev) (void *) aconnector); } - if (DC_IRQ_SOURCE_INVALID != dc_link->irq_source_hpd_rx) { + if (dc_link->irq_source_hpd_rx != DC_IRQ_SOURCE_INVALID) { /* Also register for DP short pulse (hpd_rx). */ int_params.int_context = INTERRUPT_LOW_IRQ_CONTEXT; @@ -3486,11 +3505,11 @@ static void register_hpd_handlers(struct amdgpu_device *adev) amdgpu_dm_irq_register_interrupt(adev, &int_params, handle_hpd_rx_irq, (void *) aconnector); - - if (adev->dm.hpd_rx_offload_wq) - adev->dm.hpd_rx_offload_wq[dc_link->link_index].aconnector = - aconnector; } + + if (adev->dm.hpd_rx_offload_wq) + adev->dm.hpd_rx_offload_wq[connector->index].aconnector = + aconnector; } } @@ -3503,7 +3522,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev) struct dc_interrupt_params int_params = {0}; int r; int i; - unsigned client_id = AMDGPU_IRQ_CLIENTID_LEGACY; + unsigned int client_id = AMDGPU_IRQ_CLIENTID_LEGACY; int_params.requested_polarity = INTERRUPT_POLARITY_DEFAULT; int_params.current_polarity = INTERRUPT_POLARITY_DEFAULT; @@ -3517,11 +3536,12 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev) * Base driver will call amdgpu_dm_irq_handler() for ALL interrupts * coming from DC hardware. * amdgpu_dm_irq_handler() will re-direct the interrupt to DC - * for acknowledging and handling. */ + * for acknowledging and handling. + */ /* Use VBLANK interrupt */ for (i = 0; i < adev->mode_info.num_crtc; i++) { - r = amdgpu_irq_add_id(adev, client_id, i+1 , &adev->crtc_irq); + r = amdgpu_irq_add_id(adev, client_id, i + 1, &adev->crtc_irq); if (r) { DRM_ERROR("Failed to add crtc irq id!\n"); return r; @@ -3529,7 +3549,7 @@ static int dce60_register_irq_handlers(struct amdgpu_device *adev) int_params.int_context = INTERRUPT_HIGH_IRQ_CONTEXT; int_params.irq_source = - dc_interrupt_to_irq_source(dc, i+1 , 0); + dc_interrupt_to_irq_source(dc, i + 1, 0); c_irq_params = &adev->dm.vblank_params[int_params.irq_source - DC_IRQ_SOURCE_VBLANK1]; @@ -3585,7 +3605,7 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev) struct dc_interrupt_params int_params = {0}; int r; int i; - unsigned client_id = AMDGPU_IRQ_CLIENTID_LEGACY; + unsigned int client_id = AMDGPU_IRQ_CLIENTID_LEGACY; if (adev->family >= AMDGPU_FAMILY_AI) client_id = SOC15_IH_CLIENTID_DCE; @@ -3602,7 +3622,8 @@ static int dce110_register_irq_handlers(struct amdgpu_device *adev) * Base driver will call amdgpu_dm_irq_handler() for ALL interrupts * coming from DC hardware. * amdgpu_dm_irq_handler() will re-direct the interrupt to DC - * for acknowledging and handling. */ + * for acknowledging and handling. + */ /* Use VBLANK interrupt */ for (i = VISLANDS30_IV_SRCID_D1_VERTICAL_INTERRUPT0; i <= VISLANDS30_IV_SRCID_D6_VERTICAL_INTERRUPT0; i++) { @@ -4049,7 +4070,7 @@ static void amdgpu_dm_update_backlight_caps(struct amdgpu_display_manager *dm, } static int get_brightness_range(const struct amdgpu_dm_backlight_caps *caps, - unsigned *min, unsigned *max) + unsigned int *min, unsigned int *max) { if (!caps) return 0; @@ -4069,7 +4090,7 @@ static int get_brightness_range(const struct amdgpu_dm_backlight_caps *caps, static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *caps, uint32_t brightness) { - unsigned min, max; + unsigned int min, max; if (!get_brightness_range(caps, &min, &max)) return brightness; @@ -4082,7 +4103,7 @@ static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *c static u32 convert_brightness_to_user(const struct amdgpu_dm_backlight_caps *caps, uint32_t brightness) { - unsigned min, max; + unsigned int min, max; if (!get_brightness_range(caps, &min, &max)) return brightness; @@ -4148,6 +4169,7 @@ static int amdgpu_dm_backlight_update_status(struct backlight_device *bd) static u32 amdgpu_dm_backlight_get_level(struct amdgpu_display_manager *dm, int bl_idx) { + int ret; struct amdgpu_dm_backlight_caps caps; struct dc_link *link = (struct dc_link *)dm->backlight_link[bl_idx]; @@ -4162,13 +4184,14 @@ static u32 amdgpu_dm_backlight_get_level(struct amdgpu_display_manager *dm, if (!rc) return dm->brightness[bl_idx]; return convert_brightness_to_user(&caps, avg); - } else { - int ret = dc_link_get_backlight_level(link); - - if (ret == DC_ERROR_UNEXPECTED) - return dm->brightness[bl_idx]; - return convert_brightness_to_user(&caps, ret); } + + ret = dc_link_get_backlight_level(link); + + if (ret == DC_ERROR_UNEXPECTED) + return dm->brightness[bl_idx]; + + return convert_brightness_to_user(&caps, ret); } static int amdgpu_dm_backlight_get_brightness(struct backlight_device *bd) @@ -4315,6 +4338,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) enum dc_connection_type new_connection_type = dc_connection_none; const struct dc_plane_cap *plane; bool psr_feature_enabled = false; + bool replay_feature_enabled = false; int max_overlay = dm->dc->caps.max_slave_planes; dm->display_indexes_num = dm->dc->caps.max_streams; @@ -4424,6 +4448,20 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) } } + if (!(amdgpu_dc_debug_mask & DC_DISABLE_REPLAY)) { + switch (adev->ip_versions[DCE_HWIP][0]) { + case IP_VERSION(3, 1, 4): + case IP_VERSION(3, 1, 5): + case IP_VERSION(3, 1, 6): + case IP_VERSION(3, 2, 0): + case IP_VERSION(3, 2, 1): + replay_feature_enabled = true; + break; + default: + replay_feature_enabled = amdgpu_dc_feature_mask & DC_REPLAY_MASK; + break; + } + } /* loops over all connectors on the board */ for (i = 0; i < link_cnt; i++) { struct dc_link *link = NULL; @@ -4472,6 +4510,12 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) amdgpu_dm_update_connector_after_detect(aconnector); setup_backlight_device(dm, aconnector); + /* + * Disable psr if replay can be enabled + */ + if (replay_feature_enabled && amdgpu_dm_setup_replay(link, aconnector)) + psr_feature_enabled = false; + if (psr_feature_enabled) amdgpu_dm_set_psr_caps(link); @@ -4562,7 +4606,6 @@ fail: static void amdgpu_dm_destroy_drm_device(struct amdgpu_display_manager *dm) { drm_atomic_private_obj_fini(&dm->atomic_obj); - return; } /****************************************************************************** @@ -5394,6 +5437,7 @@ static bool adjust_colour_depth_from_display_info( { enum dc_color_depth depth = timing_out->display_color_depth; int normalized_clk; + do { normalized_clk = timing_out->pix_clk_100hz / 10; /* YCbCr 4:2:0 requires additional adjustment of 1/2 */ @@ -5609,6 +5653,7 @@ create_fake_sink(struct amdgpu_dm_connector *aconnector) { struct dc_sink_init_data sink_init_data = { 0 }; struct dc_sink *sink = NULL; + sink_init_data.link = aconnector->dc_link; sink_init_data.sink_signal = aconnector->dc_link->connector_signal; @@ -5732,7 +5777,7 @@ get_highest_refresh_rate_mode(struct amdgpu_dm_connector *aconnector, return &aconnector->freesync_vid_base; /* Find the preferred mode */ - list_for_each_entry (m, list_head, head) { + list_for_each_entry(m, list_head, head) { if (m->type & DRM_MODE_TYPE_PREFERRED) { m_pref = m; break; @@ -5756,7 +5801,7 @@ get_highest_refresh_rate_mode(struct amdgpu_dm_connector *aconnector, * For some monitors, preferred mode is not the mode with highest * supported refresh rate. */ - list_for_each_entry (m, list_head, head) { + list_for_each_entry(m, list_head, head) { current_refresh = drm_mode_vrefresh(m); if (m->hdisplay == m_pref->hdisplay && @@ -5849,6 +5894,7 @@ static void apply_dsc_policy_for_edp(struct amdgpu_dm_connector *aconnector, edp_min_bpp_x16, edp_max_bpp_x16, dsc_caps, &stream->timing, + dc_link_get_highest_encoding_format(aconnector->dc_link), &bw_range)) { if (bw_range.max_kbps < link_bw_in_kbps) { @@ -5857,6 +5903,7 @@ static void apply_dsc_policy_for_edp(struct amdgpu_dm_connector *aconnector, &dsc_options, 0, &stream->timing, + dc_link_get_highest_encoding_format(aconnector->dc_link), &dsc_cfg)) { stream->timing.dsc_cfg = dsc_cfg; stream->timing.flags.DSC = 1; @@ -5871,6 +5918,7 @@ static void apply_dsc_policy_for_edp(struct amdgpu_dm_connector *aconnector, &dsc_options, link_bw_in_kbps, &stream->timing, + dc_link_get_highest_encoding_format(aconnector->dc_link), &dsc_cfg)) { stream->timing.dsc_cfg = dsc_cfg; stream->timing.flags.DSC = 1; @@ -5914,12 +5962,14 @@ static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector, &dsc_options, link_bandwidth_kbps, &stream->timing, + dc_link_get_highest_encoding_format(aconnector->dc_link), &stream->timing.dsc_cfg)) { stream->timing.flags.DSC = 1; DRM_DEBUG_DRIVER("%s: [%s] DSC is selected from SST RX\n", __func__, drm_connector->name); } } else if (sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER) { - timing_bw_in_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing); + timing_bw_in_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing, + dc_link_get_highest_encoding_format(aconnector->dc_link)); max_supported_bw_in_kbps = link_bandwidth_kbps; dsc_max_supported_bw_in_kbps = link_bandwidth_kbps; @@ -5931,6 +5981,7 @@ static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector, &dsc_options, dsc_max_supported_bw_in_kbps, &stream->timing, + dc_link_get_highest_encoding_format(aconnector->dc_link), &stream->timing.dsc_cfg)) { stream->timing.flags.DSC = 1; DRM_DEBUG_DRIVER("%s: [%s] DSC is selected from DP-HDMI PCON\n", @@ -6028,7 +6079,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, * This may not be an error, the use case is when we have no * usermode calls to reset and set mode upon hotplug. In this * case, we call set mode ourselves to restore the previous mode - * and the modelist may not be filled in in time. + * and the modelist may not be filled in time. */ DRM_DEBUG_DRIVER("No preferred mode found\n"); } else { @@ -6047,13 +6098,13 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, if (recalculate_timing) drm_mode_set_crtcinfo(&saved_mode, 0); - else + else if (!old_stream) drm_mode_set_crtcinfo(&mode, 0); /* - * If scaling is enabled and refresh rate didn't change - * we copy the vic and polarities of the old timings - */ + * If scaling is enabled and refresh rate didn't change + * we copy the vic and polarities of the old timings + */ if (!scale || mode_refresh != preferred_refresh) fill_stream_properties_from_drm_display_mode( stream, &mode, &aconnector->base, con_state, NULL, @@ -6088,7 +6139,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, if (stream->signal == SIGNAL_TYPE_HDMI_TYPE_A) mod_build_hf_vsif_infopacket(stream, &stream->vsp_infopacket); - if (stream->link->psr_settings.psr_feature_enabled) { + if (stream->link->psr_settings.psr_feature_enabled || stream->link->replay_settings.replay_feature_enabled) { // // should decide stream support vsc sdp colorimetry capability // before building vsc info packet @@ -6817,6 +6868,7 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder, if (!state->duplicated) { int max_bpc = conn_state->max_requested_bpc; + is_y420 = drm_mode_is_420_also(&connector->display_info, adjusted_mode) && aconnector->force_yuv420_output; color_depth = convert_color_depth_from_display_info(connector, @@ -7135,7 +7187,7 @@ static bool is_duplicate_mode(struct amdgpu_dm_connector *aconnector, { struct drm_display_mode *m; - list_for_each_entry (m, &aconnector->base.probed_modes, head) { + list_for_each_entry(m, &aconnector->base.probed_modes, head) { if (drm_mode_equal(m, mode)) return true; } @@ -7295,6 +7347,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, aconnector->as_type = ADAPTIVE_SYNC_TYPE_NONE; memset(&aconnector->vsdb_info, 0, sizeof(aconnector->vsdb_info)); mutex_init(&aconnector->hpd_lock); + mutex_init(&aconnector->handle_mst_msg_ready); /* * configure support HPD hot plug connector_>polled default value is 0 @@ -7350,7 +7403,7 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, if (connector_type == DRM_MODE_CONNECTOR_HDMIA) { if (!drm_mode_create_hdmi_colorspace_property(&aconnector->base, supported_colorspaces)) drm_connector_attach_colorspace_property(&aconnector->base); - } else if (connector_type == DRM_MODE_CONNECTOR_DisplayPort || + } else if ((connector_type == DRM_MODE_CONNECTOR_DisplayPort && !aconnector->mst_root) || connector_type == DRM_MODE_CONNECTOR_eDP) { if (!drm_mode_create_dp_colorspace_property(&aconnector->base, supported_colorspaces)) drm_connector_attach_colorspace_property(&aconnector->base); @@ -7454,7 +7507,6 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm, link->priv = aconnector; - DRM_DEBUG_DRIVER("%s()\n", __func__); i2c = create_i2c(link->ddc, link->link_index, &res); if (!i2c) { @@ -7827,7 +7879,7 @@ static void update_freesync_state_on_stream( aconn = (struct amdgpu_dm_connector *)new_stream->dm_stream_context; - if (aconn && aconn->as_type == FREESYNC_TYPE_PCON_IN_WHITELIST) { + if (aconn && (aconn->as_type == FREESYNC_TYPE_PCON_IN_WHITELIST || aconn->vsdb_info.replay_mode)) { pack_sdp_v1_3 = aconn->pack_sdp_v1_3; if (aconn->vsdb_info.amd_vsdb_version == 1) @@ -7982,7 +8034,6 @@ static inline uint32_t get_mem_type(struct drm_framebuffer *fb) } static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, - struct dc_state *dc_state, struct drm_device *dev, struct amdgpu_display_manager *dm, struct drm_crtc *pcrtc, @@ -8094,7 +8145,8 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, bundle->surface_updates[planes_count].plane_info = &bundle->plane_infos[planes_count]; - if (acrtc_state->stream->link->psr_settings.psr_feature_enabled) { + if (acrtc_state->stream->link->psr_settings.psr_feature_enabled || + acrtc_state->stream->link->replay_settings.replay_feature_enabled) { fill_dc_dirty_rects(plane, old_plane_state, new_plane_state, new_crtc_state, &bundle->flip_addrs[planes_count], @@ -8125,7 +8177,17 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, * Only allow immediate flips for fast updates that don't * change memory domain, FB pitch, DCC state, rotation or * mirroring. + * + * dm_crtc_helper_atomic_check() only accepts async flips with + * fast updates. */ + if (crtc->state->async_flip && + (acrtc_state->update_type != UPDATE_TYPE_FAST || + get_mem_type(old_plane_state->fb) != get_mem_type(fb))) + drm_warn_once(state->dev, + "[PLANE:%d:%s] async flip with non-fast update\n", + plane->base.id, plane->name); + bundle->flip_addrs[planes_count].flip_immediate = crtc->state->async_flip && acrtc_state->update_type == UPDATE_TYPE_FAST && @@ -8168,8 +8230,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, * DRI3/Present extension with defined target_msc. */ last_flip_vblank = amdgpu_get_vblank_counter_kms(pcrtc); - } - else { + } else { /* For variable refresh rate mode only: * Get vblank of last completed flip to avoid > 1 vrr * flips per video frame by use of throttling, but allow @@ -8455,55 +8516,20 @@ static void amdgpu_dm_crtc_copy_transient_flags(struct drm_crtc_state *crtc_stat stream_state->mode_changed = drm_atomic_crtc_needs_modeset(crtc_state); } -/** - * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation. - * @state: The atomic state to commit - * - * This will tell DC to commit the constructed DC state from atomic_check, - * programming the hardware. Any failures here implies a hardware failure, since - * atomic check should have filtered anything non-kosher. - */ -static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) +static void amdgpu_dm_commit_streams(struct drm_atomic_state *state, + struct dc_state *dc_state) { struct drm_device *dev = state->dev; struct amdgpu_device *adev = drm_to_adev(dev); struct amdgpu_display_manager *dm = &adev->dm; - struct dm_atomic_state *dm_state; - struct dc_state *dc_state = NULL, *dc_state_temp = NULL; - u32 i, j; struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state, *new_crtc_state; - unsigned long flags; - bool wait_for_vblank = true; - struct drm_connector *connector; - struct drm_connector_state *old_con_state, *new_con_state; struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state; - int crtc_disable_count = 0; bool mode_set_reset_required = false; - int r; - - trace_amdgpu_dm_atomic_commit_tail_begin(state); - - r = drm_atomic_helper_wait_for_fences(dev, state, false); - if (unlikely(r)) - DRM_ERROR("Waiting for fences timed out!"); - - drm_atomic_helper_update_legacy_modeset_state(dev, state); - drm_dp_mst_atomic_wait_for_dependencies(state); - - dm_state = dm_atomic_get_new_state(state); - if (dm_state && dm_state->context) { - dc_state = dm_state->context; - } else { - /* No state changes, retain current state. */ - dc_state_temp = dc_create_state(dm->dc); - ASSERT(dc_state_temp); - dc_state = dc_state_temp; - dc_resource_state_copy_construct_current(dm->dc, dc_state); - } + u32 i; - for_each_oldnew_crtc_in_state (state, crtc, old_crtc_state, - new_crtc_state, i) { + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, + new_crtc_state, i) { struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); dm_old_crtc_state = to_dm_crtc_state(old_crtc_state); @@ -8526,9 +8552,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) dm_old_crtc_state = to_dm_crtc_state(old_crtc_state); drm_dbg_state(state->dev, - "amdgpu_crtc id:%d crtc_state_flags: enable:%d, active:%d, " - "planes_changed:%d, mode_changed:%d,active_changed:%d," - "connectors_changed:%d\n", + "amdgpu_crtc id:%d crtc_state_flags: enable:%d, active:%d, planes_changed:%d, mode_changed:%d,active_changed:%d,connectors_changed:%d\n", acrtc->crtc_id, new_crtc_state->enable, new_crtc_state->active, @@ -8601,24 +8625,22 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) } } /* for_each_crtc_in_state() */ - if (dc_state) { - /* if there mode set or reset, disable eDP PSR */ - if (mode_set_reset_required) { - if (dm->vblank_control_workqueue) - flush_workqueue(dm->vblank_control_workqueue); + /* if there mode set or reset, disable eDP PSR */ + if (mode_set_reset_required) { + if (dm->vblank_control_workqueue) + flush_workqueue(dm->vblank_control_workqueue); - amdgpu_dm_psr_disable_all(dm); - } + amdgpu_dm_psr_disable_all(dm); + } - dm_enable_per_frame_crtc_master_sync(dc_state); - mutex_lock(&dm->dc_lock); - WARN_ON(!dc_commit_streams(dm->dc, dc_state->streams, dc_state->stream_count)); + dm_enable_per_frame_crtc_master_sync(dc_state); + mutex_lock(&dm->dc_lock); + WARN_ON(!dc_commit_streams(dm->dc, dc_state->streams, dc_state->stream_count)); - /* Allow idle optimization when vblank count is 0 for display off */ - if (dm->active_vblank_irq_count == 0) - dc_allow_idle_optimizations(dm->dc, true); - mutex_unlock(&dm->dc_lock); - } + /* Allow idle optimization when vblank count is 0 for display off */ + if (dm->active_vblank_irq_count == 0) + dc_allow_idle_optimizations(dm->dc, true); + mutex_unlock(&dm->dc_lock); for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); @@ -8638,6 +8660,44 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) acrtc->otg_inst = status->primary_otg_inst; } } +} + +/** + * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation. + * @state: The atomic state to commit + * + * This will tell DC to commit the constructed DC state from atomic_check, + * programming the hardware. Any failures here implies a hardware failure, since + * atomic check should have filtered anything non-kosher. + */ +static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) +{ + struct drm_device *dev = state->dev; + struct amdgpu_device *adev = drm_to_adev(dev); + struct amdgpu_display_manager *dm = &adev->dm; + struct dm_atomic_state *dm_state; + struct dc_state *dc_state = NULL; + u32 i, j; + struct drm_crtc *crtc; + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + unsigned long flags; + bool wait_for_vblank = true; + struct drm_connector *connector; + struct drm_connector_state *old_con_state, *new_con_state; + struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state; + int crtc_disable_count = 0; + + trace_amdgpu_dm_atomic_commit_tail_begin(state); + + drm_atomic_helper_update_legacy_modeset_state(dev, state); + drm_dp_mst_atomic_wait_for_dependencies(state); + + dm_state = dm_atomic_get_new_state(state); + if (dm_state && dm_state->context) { + dc_state = dm_state->context; + amdgpu_dm_commit_streams(state, dc_state); + } + for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc); @@ -8760,13 +8820,12 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); struct dm_connector_state *dm_old_con_state = to_dm_connector_state(old_con_state); struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc); - struct dc_surface_update dummy_updates[MAX_SURFACES]; + struct dc_surface_update *dummy_updates; struct dc_stream_update stream_update; struct dc_info_packet hdr_packet; struct dc_stream_status *status = NULL; bool abm_changed, hdr_changed, scaling_changed; - memset(&dummy_updates, 0, sizeof(dummy_updates)); memset(&stream_update, 0, sizeof(stream_update)); if (acrtc) { @@ -8825,6 +8884,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) * Here we create an empty update on each plane. * To fix this, DC should permit updating only stream properties. */ + dummy_updates = kzalloc(sizeof(struct dc_surface_update) * MAX_SURFACES, GFP_ATOMIC); for (j = 0; j < status->plane_count; j++) dummy_updates[j].surface = status->plane_states[0]; @@ -8836,6 +8896,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) dm_new_crtc_state->stream, &stream_update); mutex_unlock(&dm->dc_lock); + kfree(dummy_updates); } /** @@ -8914,8 +8975,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); if (dm_new_crtc_state->stream) - amdgpu_dm_commit_planes(state, dc_state, dev, - dm, crtc, wait_for_vblank); + amdgpu_dm_commit_planes(state, dev, dm, crtc, wait_for_vblank); } /* Update audio instances for each connector. */ @@ -8970,9 +9030,6 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) for (i = 0; i < crtc_disable_count; i++) pm_runtime_put_autosuspend(dev->dev); pm_runtime_mark_last_busy(dev->dev); - - if (dc_state_temp) - dc_release_state(dc_state_temp); } static int dm_force_atomic_commit(struct drm_connector *connector) @@ -9104,8 +9161,8 @@ static int do_aquire_global_lock(struct drm_device *dev, &commit->flip_done, 10*HZ); if (ret == 0) - DRM_ERROR("[CRTC:%d:%s] hw_done or flip_done " - "timed out\n", crtc->base.id, crtc->name); + DRM_ERROR("[CRTC:%d:%s] hw_done or flip_done timed out\n", + crtc->base.id, crtc->name); drm_crtc_commit_put(commit); } @@ -9190,7 +9247,8 @@ is_timing_unchanged_for_freesync(struct drm_crtc_state *old_crtc_state, return false; } -static void set_freesync_fixed_config(struct dm_crtc_state *dm_new_crtc_state) { +static void set_freesync_fixed_config(struct dm_crtc_state *dm_new_crtc_state) +{ u64 num, den, res; struct drm_crtc_state *new_crtc_state = &dm_new_crtc_state->base; @@ -9312,9 +9370,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, goto skip_modeset; drm_dbg_state(state->dev, - "amdgpu_crtc id:%d crtc_state_flags: enable:%d, active:%d, " - "planes_changed:%d, mode_changed:%d,active_changed:%d," - "connectors_changed:%d\n", + "amdgpu_crtc id:%d crtc_state_flags: enable:%d, active:%d, planes_changed:%d, mode_changed:%d,active_changed:%d,connectors_changed:%d\n", acrtc->crtc_id, new_crtc_state->enable, new_crtc_state->active, @@ -9343,8 +9399,7 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, old_crtc_state)) { new_crtc_state->mode_changed = false; DRM_DEBUG_DRIVER( - "Mode change not required for front porch change, " - "setting mode_changed to %d", + "Mode change not required for front porch change, setting mode_changed to %d", new_crtc_state->mode_changed); set_freesync_fixed_config(dm_new_crtc_state); @@ -9356,9 +9411,8 @@ static int dm_update_crtc_state(struct amdgpu_display_manager *dm, struct drm_display_mode *high_mode; high_mode = get_highest_refresh_rate_mode(aconnector, false); - if (!drm_mode_equal(&new_crtc_state->mode, high_mode)) { + if (!drm_mode_equal(&new_crtc_state->mode, high_mode)) set_freesync_fixed_config(dm_new_crtc_state); - } } ret = dm_atomic_get_state(state, &dm_state); @@ -9526,6 +9580,7 @@ static bool should_reset_plane(struct drm_atomic_state *state, */ for_each_oldnew_plane_in_state(state, other, old_other_state, new_other_state, i) { struct amdgpu_framebuffer *old_afb, *new_afb; + if (other->type == DRM_PLANE_TYPE_CURSOR) continue; @@ -9624,11 +9679,12 @@ static int dm_check_cursor_fb(struct amdgpu_crtc *new_acrtc, } /* Core DRM takes care of checking FB modifiers, so we only need to - * check tiling flags when the FB doesn't have a modifier. */ + * check tiling flags when the FB doesn't have a modifier. + */ if (!(fb->flags & DRM_MODE_FB_MODIFIERS)) { if (adev->family < AMDGPU_FAMILY_AI) { linear = AMDGPU_TILING_GET(afb->tiling_flags, ARRAY_MODE) != DC_ARRAY_2D_TILED_THIN1 && - AMDGPU_TILING_GET(afb->tiling_flags, ARRAY_MODE) != DC_ARRAY_1D_TILED_THIN1 && + AMDGPU_TILING_GET(afb->tiling_flags, ARRAY_MODE) != DC_ARRAY_1D_TILED_THIN1 && AMDGPU_TILING_GET(afb->tiling_flags, MICRO_TILE_MODE) == 0; } else { linear = AMDGPU_TILING_GET(afb->tiling_flags, SWIZZLE_MODE) == 0; @@ -9762,8 +9818,8 @@ static int dm_update_plane_state(struct dc *dc, if (plane->type == DRM_PLANE_TYPE_OVERLAY) { if (is_video_format(new_plane_state->fb->format->format) && *is_top_most_overlay) return -EINVAL; - else - *is_top_most_overlay = false; + + *is_top_most_overlay = false; } DRM_DEBUG_ATOMIC("Enabling DRM plane: %d on DRM crtc %d\n", @@ -9850,12 +9906,12 @@ static int dm_check_crtc_cursor(struct drm_atomic_state *state, /* On DCE and DCN there is no dedicated hardware cursor plane. We get a * cursor per pipe but it's going to inherit the scaling and * positioning from the underlying pipe. Check the cursor plane's - * blending properties match the underlying planes'. */ + * blending properties match the underlying planes'. + */ new_cursor_state = drm_atomic_get_new_plane_state(state, cursor); - if (!new_cursor_state || !new_cursor_state->fb) { + if (!new_cursor_state || !new_cursor_state->fb) return 0; - } dm_get_oriented_plane_size(new_cursor_state, &cursor_src_w, &cursor_src_h); cursor_scale_w = new_cursor_state->crtc_w * 1000 / cursor_src_w; @@ -9900,6 +9956,7 @@ static int add_affected_mst_dsc_crtcs(struct drm_atomic_state *state, struct drm struct drm_connector_state *conn_state, *old_conn_state; struct amdgpu_dm_connector *aconnector = NULL; int i; + for_each_oldnew_connector_in_state(state, connector, old_conn_state, conn_state, i) { if (!conn_state->crtc) conn_state = old_conn_state; @@ -10091,6 +10148,11 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, /* Remove exiting planes if they are modified */ for_each_oldnew_plane_in_state_reverse(state, plane, old_plane_state, new_plane_state, i) { + if (old_plane_state->fb && new_plane_state->fb && + get_mem_type(old_plane_state->fb) != + get_mem_type(new_plane_state->fb)) + lock_and_validation_needed = true; + ret = dm_update_plane_state(dc, state, plane, old_plane_state, new_plane_state, @@ -10334,13 +10396,24 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, } /* Store the overall update type for use later in atomic check. */ - for_each_new_crtc_in_state (state, crtc, new_crtc_state, i) { + for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { struct dm_crtc_state *dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + /* + * Only allow async flips for fast updates that don't change + * the FB pitch, the DCC state, rotation, etc. + */ + if (new_crtc_state->async_flip && lock_and_validation_needed) { + drm_dbg_atomic(crtc->dev, + "[CRTC:%d:%s] async flips are only supported for fast updates\n", + crtc->base.id, crtc->name); + ret = -EINVAL; + goto fail; + } + dm_new_crtc_state->update_type = lock_and_validation_needed ? - UPDATE_TYPE_FULL : - UPDATE_TYPE_FAST; + UPDATE_TYPE_FULL : UPDATE_TYPE_FAST; } /* Must be success */ @@ -10356,7 +10429,7 @@ fail: else if (ret == -EINTR || ret == -EAGAIN || ret == -ERESTARTSYS) DRM_DEBUG_DRIVER("Atomic check stopped due to signal.\n"); else - DRM_DEBUG_DRIVER("Atomic check failed with err: %d \n", ret); + DRM_DEBUG_DRIVER("Atomic check failed with err: %d\n", ret); trace_amdgpu_dm_atomic_check_finish(state, ret); @@ -10512,6 +10585,41 @@ static bool parse_edid_cea(struct amdgpu_dm_connector *aconnector, return ret; } +static int parse_amd_vsdb(struct amdgpu_dm_connector *aconnector, + struct edid *edid, struct amdgpu_hdmi_vsdb_info *vsdb_info) +{ + u8 *edid_ext = NULL; + int i; + int j = 0; + + if (edid == NULL || edid->extensions == 0) + return -ENODEV; + + /* Find DisplayID extension */ + for (i = 0; i < edid->extensions; i++) { + edid_ext = (void *)(edid + (i + 1)); + if (edid_ext[0] == DISPLAYID_EXT) + break; + } + + while (j < EDID_LENGTH) { + struct amd_vsdb_block *amd_vsdb = (struct amd_vsdb_block *)&edid_ext[j]; + unsigned int ieeeId = (amd_vsdb->ieee_id[2] << 16) | (amd_vsdb->ieee_id[1] << 8) | (amd_vsdb->ieee_id[0]); + + if (ieeeId == HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_IEEE_REGISTRATION_ID && + amd_vsdb->version == HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_VERSION_3) { + vsdb_info->replay_mode = (amd_vsdb->feature_caps & AMD_VSDB_VERSION_3_FEATURECAP_REPLAYMODE) ? true : false; + vsdb_info->amd_vsdb_version = HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_VERSION_3; + DRM_DEBUG_KMS("Panel supports Replay Mode: %d\n", vsdb_info->replay_mode); + + return true; + } + j++; + } + + return false; +} + static int parse_hdmi_amd_vsdb(struct amdgpu_dm_connector *aconnector, struct edid *edid, struct amdgpu_hdmi_vsdb_info *vsdb_info) { @@ -10647,6 +10755,14 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, freesync_capable = true; } } + parse_amd_vsdb(amdgpu_dm_connector, edid, &vsdb_info); + + if (vsdb_info.replay_mode) { + amdgpu_dm_connector->vsdb_info.replay_mode = vsdb_info.replay_mode; + amdgpu_dm_connector->vsdb_info.amd_vsdb_version = vsdb_info.amd_vsdb_version; + amdgpu_dm_connector->as_type = ADAPTIVE_SYNC_TYPE_EDP; + } + } else if (edid && sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A) { i = parse_hdmi_amd_vsdb(amdgpu_dm_connector, edid, &vsdb_info); if (i >= 0 && vsdb_info.freesync_supported) { diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 4561f55afa99..9e4cc5eeda76 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -51,6 +51,9 @@ #define AMDGPU_DMUB_NOTIFICATION_MAX 5 +#define HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_IEEE_REGISTRATION_ID 0x00001A +#define AMD_VSDB_VERSION_3_FEATURECAP_REPLAYMODE 0x40 +#define HDMI_AMD_VENDOR_SPECIFIC_DATA_BLOCK_VERSION_3 0x3 /* #include "include/amdgpu_dal_power_if.h" #include "amdgpu_dm_irq.h" @@ -75,6 +78,12 @@ struct dmub_srv; struct dc_plane_state; struct dmub_notification; +struct amd_vsdb_block { + unsigned char ieee_id[3]; + unsigned char version; + unsigned char feature_caps; +}; + struct common_irq_params { struct amdgpu_device *adev; enum dc_irq_source irq_src; @@ -195,6 +204,11 @@ struct hpd_rx_irq_offload_work_queue { */ bool is_handling_link_loss; /** + * @is_handling_mst_msg_rdy_event: Used to prevent inserting mst message + * ready event when we're already handling mst message ready event + */ + bool is_handling_mst_msg_rdy_event; + /** * @aconnector: The aconnector that this work queue is attached to */ struct amdgpu_dm_connector *aconnector; @@ -604,6 +618,11 @@ struct amdgpu_hdmi_vsdb_info { * @max_refresh_rate_hz: FreeSync Maximum Refresh Rate in Hz */ unsigned int max_refresh_rate_hz; + + /** + * @replay_mode: Replay supported + */ + bool replay_mode; }; struct amdgpu_dm_connector { @@ -638,6 +657,8 @@ struct amdgpu_dm_connector { struct drm_dp_mst_port *mst_output_port; struct amdgpu_dm_connector *mst_root; struct drm_dp_aux *dsc_aux; + struct mutex handle_mst_msg_ready; + /* TODO see if we can merge with ddc_bus or make a dm_connector */ struct amdgpu_i2c_adapter *i2c; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c index 0802f8e8fac5..52ecfa746b54 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c @@ -123,9 +123,8 @@ static void amdgpu_dm_crtc_notify_ta_to_read(struct work_struct *work) secure_display_ctx = container_of(work, struct secure_display_context, notify_ta_work); crtc = secure_display_ctx->crtc; - if (!crtc) { + if (!crtc) return; - } psp = &drm_to_adev(crtc->dev)->psp; @@ -151,9 +150,8 @@ static void amdgpu_dm_crtc_notify_ta_to_read(struct work_struct *work) ret = psp_securedisplay_invoke(psp, TA_SECUREDISPLAY_COMMAND__SEND_ROI_CRC); if (!ret) { - if (securedisplay_cmd->status != TA_SECUREDISPLAY_STATUS__SUCCESS) { + if (securedisplay_cmd->status != TA_SECUREDISPLAY_STATUS__SUCCESS) psp_securedisplay_parse_resp_status(psp, securedisplay_cmd->status); - } } mutex_unlock(&psp->securedisplay_context.mutex); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index 440fc0869a34..97b7a0b8a1c2 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -29,6 +29,7 @@ #include "dc.h" #include "amdgpu.h" #include "amdgpu_dm_psr.h" +#include "amdgpu_dm_replay.h" #include "amdgpu_dm_crtc.h" #include "amdgpu_dm_plane.h" #include "amdgpu_dm_trace.h" @@ -123,7 +124,12 @@ static void vblank_control_worker(struct work_struct *work) * fill_dc_dirty_rects(). */ if (vblank_work->stream && vblank_work->stream->link) { - if (vblank_work->enable) { + /* + * Prioritize replay, instead of psr + */ + if (vblank_work->stream->link->replay_settings.replay_feature_enabled) + amdgpu_dm_replay_enable(vblank_work->stream, false); + else if (vblank_work->enable) { if (vblank_work->stream->link->psr_settings.psr_version < DC_PSR_VERSION_SU_1 && vblank_work->stream->link->psr_settings.psr_allow_active) amdgpu_dm_psr_disable(vblank_work->stream); @@ -132,6 +138,7 @@ static void vblank_control_worker(struct work_struct *work) #ifdef CONFIG_DRM_AMD_SECURE_DISPLAY !amdgpu_dm_crc_window_is_activated(&vblank_work->acrtc->base) && #endif + vblank_work->stream->link->panel_config.psr.disallow_replay && vblank_work->acrtc->dm_irq_params.allow_psr_entry) { amdgpu_dm_psr_enable(vblank_work->stream); } @@ -398,6 +405,18 @@ static int dm_crtc_helper_atomic_check(struct drm_crtc *crtc, return -EINVAL; } + /* + * Only allow async flips for fast updates that don't change the FB + * pitch, the DCC state, rotation, etc. + */ + if (crtc_state->async_flip && + dm_crtc_state->update_type != UPDATE_TYPE_FAST) { + drm_dbg_atomic(crtc->dev, + "[CRTC:%d:%s] async flips are only supported for fast updates\n", + crtc->base.id, crtc->name); + return -EINVAL; + } + /* In some use cases, like reset, no stream is attached */ if (!dm_crtc_state->stream) return 0; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index d63ee636483b..7c21e21bcc51 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -1075,24 +1075,24 @@ static int amdgpu_current_colorspace_show(struct seq_file *m, void *data) switch (dm_crtc_state->stream->output_color_space) { case COLOR_SPACE_SRGB: - seq_printf(m, "sRGB"); + seq_puts(m, "sRGB"); break; case COLOR_SPACE_YCBCR601: case COLOR_SPACE_YCBCR601_LIMITED: - seq_printf(m, "BT601_YCC"); + seq_puts(m, "BT601_YCC"); break; case COLOR_SPACE_YCBCR709: case COLOR_SPACE_YCBCR709_LIMITED: - seq_printf(m, "BT709_YCC"); + seq_puts(m, "BT709_YCC"); break; case COLOR_SPACE_ADOBERGB: - seq_printf(m, "opRGB"); + seq_puts(m, "opRGB"); break; case COLOR_SPACE_2020_RGB_FULLRANGE: - seq_printf(m, "BT2020_RGB"); + seq_puts(m, "BT2020_RGB"); break; case COLOR_SPACE_2020_YCBCR: - seq_printf(m, "BT2020_YCC"); + seq_puts(m, "BT2020_YCC"); break; default: goto unlock; @@ -3022,7 +3022,7 @@ static int edp_ilr_show(struct seq_file *m, void *unused) seq_printf(m, "[%d] %d kHz\n", entry/2, link_rate_in_khz); } } else { - seq_printf(m, "ILR is not supported by this eDP panel.\n"); + seq_puts(m, "ILR is not supported by this eDP panel.\n"); } return 0; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c index 5536d17306d0..20cfc5be21a4 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c @@ -39,10 +39,10 @@ static bool lp_write_i2c(void *handle, uint32_t address, const uint8_t *data, uint32_t size) { - struct dc_link *link = handle; struct i2c_payload i2c_payloads[] = {{true, address, size, (void *)data} }; - struct i2c_command cmd = {i2c_payloads, 1, I2C_COMMAND_ENGINE_HW, link->dc->caps.i2c_speed_in_khz}; + struct i2c_command cmd = {i2c_payloads, 1, I2C_COMMAND_ENGINE_HW, + link->dc->caps.i2c_speed_in_khz}; return dm_helpers_submit_i2c(link->ctx, link, &cmd); } @@ -52,8 +52,10 @@ lp_read_i2c(void *handle, uint32_t address, uint8_t offset, uint8_t *data, uint3 { struct dc_link *link = handle; - struct i2c_payload i2c_payloads[] = {{true, address, 1, &offset}, {false, address, size, data} }; - struct i2c_command cmd = {i2c_payloads, 2, I2C_COMMAND_ENGINE_HW, link->dc->caps.i2c_speed_in_khz}; + struct i2c_payload i2c_payloads[] = {{true, address, 1, &offset}, + {false, address, size, data} }; + struct i2c_command cmd = {i2c_payloads, 2, I2C_COMMAND_ENGINE_HW, + link->dc->caps.i2c_speed_in_khz}; return dm_helpers_submit_i2c(link->ctx, link, &cmd); } @@ -76,7 +78,6 @@ lp_read_dpcd(void *handle, uint32_t address, uint8_t *data, uint32_t size) static uint8_t *psp_get_srm(struct psp_context *psp, uint32_t *srm_version, uint32_t *srm_size) { - struct ta_hdcp_shared_memory *hdcp_cmd; if (!psp->hdcp_context.context.initialized) { @@ -96,13 +97,12 @@ static uint8_t *psp_get_srm(struct psp_context *psp, uint32_t *srm_version, uint *srm_version = hdcp_cmd->out_msg.hdcp_get_srm.srm_version; *srm_size = hdcp_cmd->out_msg.hdcp_get_srm.srm_buf_size; - return hdcp_cmd->out_msg.hdcp_get_srm.srm_buf; } -static int psp_set_srm(struct psp_context *psp, uint8_t *srm, uint32_t srm_size, uint32_t *srm_version) +static int psp_set_srm(struct psp_context *psp, + u8 *srm, uint32_t srm_size, uint32_t *srm_version) { - struct ta_hdcp_shared_memory *hdcp_cmd; if (!psp->hdcp_context.context.initialized) { @@ -119,7 +119,8 @@ static int psp_set_srm(struct psp_context *psp, uint8_t *srm, uint32_t srm_size, psp_hdcp_invoke(psp, hdcp_cmd->cmd_id); - if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS || hdcp_cmd->out_msg.hdcp_set_srm.valid_signature != 1 || + if (hdcp_cmd->hdcp_status != TA_HDCP_STATUS__SUCCESS || + hdcp_cmd->out_msg.hdcp_set_srm.valid_signature != 1 || hdcp_cmd->out_msg.hdcp_set_srm.srm_version == PSP_SRM_VERSION_MAX) return -EINVAL; @@ -150,7 +151,6 @@ static void process_output(struct hdcp_workqueue *hdcp_work) static void link_lock(struct hdcp_workqueue *work, bool lock) { - int i = 0; for (i = 0; i < work->max_link; i++) { @@ -160,66 +160,60 @@ static void link_lock(struct hdcp_workqueue *work, bool lock) mutex_unlock(&work[i].mutex); } } + void hdcp_update_display(struct hdcp_workqueue *hdcp_work, unsigned int link_index, struct amdgpu_dm_connector *aconnector, - uint8_t content_type, + u8 content_type, bool enable_encryption) { struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; - struct mod_hdcp_display *display = &hdcp_work[link_index].display; - struct mod_hdcp_link *link = &hdcp_work[link_index].link; - struct mod_hdcp_display_query query; + struct mod_hdcp_link_adjustment link_adjust; + struct mod_hdcp_display_adjustment display_adjust; unsigned int conn_index = aconnector->base.index; mutex_lock(&hdcp_w->mutex); hdcp_w->aconnector[conn_index] = aconnector; - query.display = NULL; - mod_hdcp_query_display(&hdcp_w->hdcp, aconnector->base.index, &query); - - if (query.display != NULL) { - memcpy(display, query.display, sizeof(struct mod_hdcp_display)); - mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output); - - hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; - - if (enable_encryption) { - /* Explicitly set the saved SRM as sysfs call will be after we already enabled hdcp - * (s3 resume case) - */ - if (hdcp_work->srm_size > 0) - psp_set_srm(hdcp_work->hdcp.config.psp.handle, hdcp_work->srm, hdcp_work->srm_size, - &hdcp_work->srm_version); - - display->adjust.disable = MOD_HDCP_DISPLAY_NOT_DISABLE; - if (content_type == DRM_MODE_HDCP_CONTENT_TYPE0) { - hdcp_w->link.adjust.hdcp1.disable = 0; - hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; - } else if (content_type == DRM_MODE_HDCP_CONTENT_TYPE1) { - hdcp_w->link.adjust.hdcp1.disable = 1; - hdcp_w->link.adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_1; - } + memset(&link_adjust, 0, sizeof(link_adjust)); + memset(&display_adjust, 0, sizeof(display_adjust)); - schedule_delayed_work(&hdcp_w->property_validate_dwork, - msecs_to_jiffies(DRM_HDCP_CHECK_PERIOD_MS)); - } else { - display->adjust.disable = MOD_HDCP_DISPLAY_DISABLE_AUTHENTICATION; - hdcp_w->encryption_status[conn_index] = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; - cancel_delayed_work(&hdcp_w->property_validate_dwork); + if (enable_encryption) { + /* Explicitly set the saved SRM as sysfs call will be after we already enabled hdcp + * (s3 resume case) + */ + if (hdcp_work->srm_size > 0) + psp_set_srm(hdcp_work->hdcp.config.psp.handle, hdcp_work->srm, + hdcp_work->srm_size, + &hdcp_work->srm_version); + + display_adjust.disable = MOD_HDCP_DISPLAY_NOT_DISABLE; + + link_adjust.auth_delay = 2; + + if (content_type == DRM_MODE_HDCP_CONTENT_TYPE0) { + link_adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0; + } else if (content_type == DRM_MODE_HDCP_CONTENT_TYPE1) { + link_adjust.hdcp1.disable = 1; + link_adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_1; } - display->state = MOD_HDCP_DISPLAY_ACTIVE; + schedule_delayed_work(&hdcp_w->property_validate_dwork, + msecs_to_jiffies(DRM_HDCP_CHECK_PERIOD_MS)); + } else { + display_adjust.disable = MOD_HDCP_DISPLAY_DISABLE_AUTHENTICATION; + hdcp_w->encryption_status[conn_index] = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; + cancel_delayed_work(&hdcp_w->property_validate_dwork); } - mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output); + mod_hdcp_update_display(&hdcp_w->hdcp, conn_index, &link_adjust, &display_adjust, &hdcp_w->output); process_output(hdcp_w); mutex_unlock(&hdcp_w->mutex); } static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work, - unsigned int link_index, + unsigned int link_index, struct amdgpu_dm_connector *aconnector) { struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; @@ -238,7 +232,8 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work, conn_state->content_protection = DRM_MODE_CONTENT_PROTECTION_DESIRED; DRM_DEBUG_DRIVER("[HDCP_DM] display %d, CP 2 -> 1, type %u, DPMS %u\n", - aconnector->base.index, conn_state->hdcp_content_type, aconnector->base.dpms); + aconnector->base.index, conn_state->hdcp_content_type, + aconnector->base.dpms); } mod_hdcp_remove_display(&hdcp_w->hdcp, aconnector->base.index, &hdcp_w->output); @@ -246,6 +241,7 @@ static void hdcp_remove_display(struct hdcp_workqueue *hdcp_work, process_output(hdcp_w); mutex_unlock(&hdcp_w->mutex); } + void hdcp_reset_display(struct hdcp_workqueue *hdcp_work, unsigned int link_index) { struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; @@ -274,15 +270,12 @@ void hdcp_handle_cpirq(struct hdcp_workqueue *hdcp_work, unsigned int link_index schedule_work(&hdcp_w->cpirq_work); } - - - static void event_callback(struct work_struct *work) { struct hdcp_workqueue *hdcp_work; hdcp_work = container_of(to_delayed_work(work), struct hdcp_workqueue, - callback_dwork); + callback_dwork); mutex_lock(&hdcp_work->mutex); @@ -294,13 +287,12 @@ static void event_callback(struct work_struct *work) process_output(hdcp_work); mutex_unlock(&hdcp_work->mutex); - - } static void event_property_update(struct work_struct *work) { - struct hdcp_workqueue *hdcp_work = container_of(work, struct hdcp_workqueue, property_update_work); + struct hdcp_workqueue *hdcp_work = container_of(work, struct hdcp_workqueue, + property_update_work); struct amdgpu_dm_connector *aconnector = NULL; struct drm_device *dev; long ret; @@ -334,11 +326,10 @@ static void event_property_update(struct work_struct *work) mutex_lock(&hdcp_work->mutex); if (conn_state->commit) { - ret = wait_for_completion_interruptible_timeout( - &conn_state->commit->hw_done, 10 * HZ); + ret = wait_for_completion_interruptible_timeout(&conn_state->commit->hw_done, + 10 * HZ); if (ret == 0) { - DRM_ERROR( - "HDCP state unknown! Setting it to DESIRED"); + DRM_ERROR("HDCP state unknown! Setting it to DESIRED\n"); hdcp_work->encryption_status[conn_index] = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; } @@ -349,24 +340,20 @@ static void event_property_update(struct work_struct *work) DRM_MODE_HDCP_CONTENT_TYPE0 && hdcp_work->encryption_status[conn_index] <= MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE0_ON) { - DRM_DEBUG_DRIVER("[HDCP_DM] DRM_MODE_CONTENT_PROTECTION_ENABLED\n"); - drm_hdcp_update_content_protection( - connector, - DRM_MODE_CONTENT_PROTECTION_ENABLED); + drm_hdcp_update_content_protection(connector, + DRM_MODE_CONTENT_PROTECTION_ENABLED); } else if (conn_state->hdcp_content_type == DRM_MODE_HDCP_CONTENT_TYPE1 && hdcp_work->encryption_status[conn_index] == MOD_HDCP_ENCRYPTION_STATUS_HDCP2_TYPE1_ON) { - drm_hdcp_update_content_protection( - connector, - DRM_MODE_CONTENT_PROTECTION_ENABLED); + drm_hdcp_update_content_protection(connector, + DRM_MODE_CONTENT_PROTECTION_ENABLED); } } else { DRM_DEBUG_DRIVER("[HDCP_DM] DRM_MODE_CONTENT_PROTECTION_DESIRED\n"); - drm_hdcp_update_content_protection( - connector, DRM_MODE_CONTENT_PROTECTION_DESIRED); - + drm_hdcp_update_content_protection(connector, + DRM_MODE_CONTENT_PROTECTION_DESIRED); } mutex_unlock(&hdcp_work->mutex); drm_modeset_unlock(&dev->mode_config.connection_mutex); @@ -402,7 +389,7 @@ static void event_property_validate(struct work_struct *work) &query); DRM_DEBUG_DRIVER("[HDCP_DM] disp %d, connector->CP %u, (query, work): (%d, %d)\n", - aconnector->base.index, + aconnector->base.index, aconnector->base.state->content_protection, query.encryption_status, hdcp_work->encryption_status[conn_index]); @@ -410,7 +397,8 @@ static void event_property_validate(struct work_struct *work) if (query.encryption_status != hdcp_work->encryption_status[conn_index]) { DRM_DEBUG_DRIVER("[HDCP_DM] encryption_status change from %x to %x\n", - hdcp_work->encryption_status[conn_index], query.encryption_status); + hdcp_work->encryption_status[conn_index], + query.encryption_status); hdcp_work->encryption_status[conn_index] = query.encryption_status; @@ -429,7 +417,7 @@ static void event_watchdog_timer(struct work_struct *work) struct hdcp_workqueue *hdcp_work; hdcp_work = container_of(to_delayed_work(work), - struct hdcp_workqueue, + struct hdcp_workqueue, watchdog_timer_dwork); mutex_lock(&hdcp_work->mutex); @@ -443,7 +431,6 @@ static void event_watchdog_timer(struct work_struct *work) process_output(hdcp_work); mutex_unlock(&hdcp_work->mutex); - } static void event_cpirq(struct work_struct *work) @@ -459,10 +446,8 @@ static void event_cpirq(struct work_struct *work) process_output(hdcp_work); mutex_unlock(&hdcp_work->mutex); - } - void hdcp_destroy(struct kobject *kobj, struct hdcp_workqueue *hdcp_work) { int i = 0; @@ -478,10 +463,8 @@ void hdcp_destroy(struct kobject *kobj, struct hdcp_workqueue *hdcp_work) kfree(hdcp_work); } - static bool enable_assr(void *handle, struct dc_link *link) { - struct hdcp_workqueue *hdcp_work = handle; struct mod_hdcp hdcp = hdcp_work->hdcp; struct psp_context *psp = hdcp.config.psp.handle; @@ -499,7 +482,8 @@ static bool enable_assr(void *handle, struct dc_link *link) memset(dtm_cmd, 0, sizeof(struct ta_dtm_shared_memory)); dtm_cmd->cmd_id = TA_DTM_COMMAND__TOPOLOGY_ASSR_ENABLE; - dtm_cmd->dtm_in_message.topology_assr_enable.display_topology_dig_be_index = link->link_enc_hw_inst; + dtm_cmd->dtm_in_message.topology_assr_enable.display_topology_dig_be_index = + link->link_enc_hw_inst; dtm_cmd->dtm_status = TA_DTM_STATUS__GENERIC_FAILURE; psp_dtm_invoke(psp, dtm_cmd->cmd_id); @@ -521,7 +505,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) int link_index = aconnector->dc_link->link_index; struct mod_hdcp_display *display = &hdcp_work[link_index].display; struct mod_hdcp_link *link = &hdcp_work[link_index].link; - struct drm_connector_state *conn_state; + struct hdcp_workqueue *hdcp_w = &hdcp_work[link_index]; struct dc_sink *sink = NULL; bool link_is_hdcp14 = false; @@ -541,7 +525,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) else if (aconnector->dc_em_sink) sink = aconnector->dc_em_sink; - if (sink != NULL) + if (sink) link->mode = mod_hdcp_signal_type_to_operation_mode(sink->sink_signal); display->controller = CONTROLLER_ID_D0 + config->otg_inst; @@ -564,19 +548,27 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) display->adjust.disable = MOD_HDCP_DISPLAY_DISABLE_AUTHENTICATION; link->adjust.auth_delay = 2; link->adjust.hdcp1.disable = 0; - conn_state = aconnector->base.state; + hdcp_w->encryption_status[display->index] = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF; DRM_DEBUG_DRIVER("[HDCP_DM] display %d, CP %d, type %d\n", aconnector->base.index, - (!!aconnector->base.state) ? aconnector->base.state->content_protection : -1, - (!!aconnector->base.state) ? aconnector->base.state->hdcp_content_type : -1); + (!!aconnector->base.state) ? + aconnector->base.state->content_protection : -1, + (!!aconnector->base.state) ? + aconnector->base.state->hdcp_content_type : -1); - if (conn_state) - hdcp_update_display(hdcp_work, link_index, aconnector, - conn_state->hdcp_content_type, false); -} + mutex_lock(&hdcp_w->mutex); + + mod_hdcp_add_display(&hdcp_w->hdcp, link, display, &hdcp_w->output); + + process_output(hdcp_w); + mutex_unlock(&hdcp_w->mutex); +} -/* NOTE: From the usermodes prospective you only need to call write *ONCE*, the kernel +/** + * DOC: Add sysfs interface for set/get srm + * + * NOTE: From the usermodes prospective you only need to call write *ONCE*, the kernel * will automatically call once or twice depending on the size * * call: "cat file > /sys/class/drm/card0/device/hdcp_srm" from usermode no matter what the size is @@ -587,23 +579,23 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) * sysfs interface doesn't tell us the size we will get so we are sending partial SRMs to psp and on * the last call we will send the full SRM. PSP will fail on every call before the last. * - * This means we don't know if the SRM is good until the last call. And because of this limitation we - * cannot throw errors early as it will stop the kernel from writing to sysfs + * This means we don't know if the SRM is good until the last call. And because of this + * limitation we cannot throw errors early as it will stop the kernel from writing to sysfs * * Example 1: - * Good SRM size = 5096 - * first call to write 4096 -> PSP fails - * Second call to write 1000 -> PSP Pass -> SRM is set + * Good SRM size = 5096 + * first call to write 4096 -> PSP fails + * Second call to write 1000 -> PSP Pass -> SRM is set * * Example 2: - * Bad SRM size = 4096 - * first call to write 4096 -> PSP fails (This is the same as above, but we don't know if this - * is the last call) + * Bad SRM size = 4096 + * first call to write 4096 -> PSP fails (This is the same as above, but we don't know if this + * is the last call) * * Solution?: - * 1: Parse the SRM? -> It is signed so we don't know the EOF - * 2: We can have another sysfs that passes the size before calling set. -> simpler solution - * below + * 1: Parse the SRM? -> It is signed so we don't know the EOF + * 2: We can have another sysfs that passes the size before calling set. -> simpler solution + * below * * Easy Solution: * Always call get after Set to verify if set was successful. @@ -612,20 +604,21 @@ static void update_config(void *handle, struct cp_psp_stream_config *config) * +----------------------+ * PSP will only update its srm if its older than the one we are trying to load. * Always do set first than get. - * -if we try to "1. SET" a older version PSP will reject it and we can "2. GET" the newer - * version and save it + * -if we try to "1. SET" a older version PSP will reject it and we can "2. GET" the newer + * version and save it * - * -if we try to "1. SET" a newer version PSP will accept it and we can "2. GET" the - * same(newer) version back and save it + * -if we try to "1. SET" a newer version PSP will accept it and we can "2. GET" the + * same(newer) version back and save it * - * -if we try to "1. SET" a newer version and PSP rejects it. That means the format is - * incorrect/corrupted and we should correct our SRM by getting it from PSP + * -if we try to "1. SET" a newer version and PSP rejects it. That means the format is + * incorrect/corrupted and we should correct our SRM by getting it from PSP */ -static ssize_t srm_data_write(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, +static ssize_t srm_data_write(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buffer, loff_t pos, size_t count) { struct hdcp_workqueue *work; - uint32_t srm_version = 0; + u32 srm_version = 0; work = container_of(bin_attr, struct hdcp_workqueue, attr); link_lock(work, true); @@ -639,19 +632,19 @@ static ssize_t srm_data_write(struct file *filp, struct kobject *kobj, struct bi work->srm_version = srm_version; } - link_lock(work, false); return count; } -static ssize_t srm_data_read(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buffer, +static ssize_t srm_data_read(struct file *filp, struct kobject *kobj, + struct bin_attribute *bin_attr, char *buffer, loff_t pos, size_t count) { struct hdcp_workqueue *work; - uint8_t *srm = NULL; - uint32_t srm_version; - uint32_t srm_size; + u8 *srm = NULL; + u32 srm_version; + u32 srm_size; size_t ret = count; work = container_of(bin_attr, struct hdcp_workqueue, attr); @@ -684,12 +677,12 @@ ret: /* From the hdcp spec (5.Renewability) SRM needs to be stored in a non-volatile memory. * * For example, - * if Application "A" sets the SRM (ver 2) and we reboot/suspend and later when Application "B" - * needs to use HDCP, the version in PSP should be SRM(ver 2). So SRM should be persistent - * across boot/reboots/suspend/resume/shutdown + * if Application "A" sets the SRM (ver 2) and we reboot/suspend and later when Application "B" + * needs to use HDCP, the version in PSP should be SRM(ver 2). So SRM should be persistent + * across boot/reboots/suspend/resume/shutdown * - * Currently when the system goes down (suspend/shutdown) the SRM is cleared from PSP. For HDCP we need - * to make the SRM persistent. + * Currently when the system goes down (suspend/shutdown) the SRM is cleared from PSP. For HDCP + * we need to make the SRM persistent. * * -PSP owns the checking of SRM but doesn't have the ability to store it in a non-volatile memory. * -The kernel cannot write to the file systems. @@ -699,8 +692,8 @@ ret: * * Usermode can read/write to/from PSP using the sysfs interface * For example: - * to save SRM from PSP to storage : cat /sys/class/drm/card0/device/hdcp_srm > srmfile - * to load from storage to PSP: cat srmfile > /sys/class/drm/card0/device/hdcp_srm + * to save SRM from PSP to storage : cat /sys/class/drm/card0/device/hdcp_srm > srmfile + * to load from storage to PSP: cat srmfile > /sys/class/drm/card0/device/hdcp_srm */ static const struct bin_attribute data_attr = { .attr = {.name = "hdcp_srm", .mode = 0664}, @@ -709,10 +702,9 @@ static const struct bin_attribute data_attr = { .read = srm_data_read, }; - -struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev, struct cp_psp *cp_psp, struct dc *dc) +struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev, + struct cp_psp *cp_psp, struct dc *dc) { - int max_caps = dc->caps.max_links; struct hdcp_workqueue *hdcp_work; int i = 0; @@ -721,14 +713,16 @@ struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev, struct if (ZERO_OR_NULL_PTR(hdcp_work)) return NULL; - hdcp_work->srm = kcalloc(PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, sizeof(*hdcp_work->srm), GFP_KERNEL); + hdcp_work->srm = kcalloc(PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, + sizeof(*hdcp_work->srm), GFP_KERNEL); - if (hdcp_work->srm == NULL) + if (!hdcp_work->srm) goto fail_alloc_context; - hdcp_work->srm_temp = kcalloc(PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, sizeof(*hdcp_work->srm_temp), GFP_KERNEL); + hdcp_work->srm_temp = kcalloc(PSP_HDCP_SRM_FIRST_GEN_MAX_SIZE, + sizeof(*hdcp_work->srm_temp), GFP_KERNEL); - if (hdcp_work->srm_temp == NULL) + if (!hdcp_work->srm_temp) goto fail_alloc_context; hdcp_work->max_link = max_caps; @@ -781,10 +775,5 @@ fail_alloc_context: kfree(hdcp_work); return NULL; - - - } - - diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 9ad509279b0a..cbef4ff28cd8 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -68,15 +68,15 @@ static void apply_edid_quirks(struct edid *edid, struct dc_edid_caps *edid_caps) } } -/* dm_helpers_parse_edid_caps - * - * Parse edid caps +/** + * dm_helpers_parse_edid_caps() - Parse edid caps * + * @link: current detected link * @edid: [in] pointer to edid - * edid_caps: [in] pointer to edid caps - * @return - * void - * */ + * @edid_caps: [in] pointer to edid caps + * + * Return: void + */ enum dc_edid_status dm_helpers_parse_edid_caps( struct dc_link *link, const struct dc_edid *edid, @@ -117,7 +117,7 @@ enum dc_edid_status dm_helpers_parse_edid_caps( if (sad_count <= 0) return result; - edid_caps->audio_mode_count = sad_count < DC_MAX_AUDIO_DESC_COUNT ? sad_count : DC_MAX_AUDIO_DESC_COUNT; + edid_caps->audio_mode_count = min(sad_count, DC_MAX_AUDIO_DESC_COUNT); for (i = 0; i < edid_caps->audio_mode_count; ++i) { struct cea_sad *sad = &sads[i]; @@ -255,7 +255,8 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( /* Accessing the connector state is required for vcpi_slots allocation * and directly relies on behaviour in commit check * that blocks before commit guaranteeing that the state - * is not gonna be swapped while still in use in commit tail */ + * is not gonna be swapped while still in use in commit tail + */ if (!aconnector || !aconnector->mst_root) return false; @@ -281,7 +282,8 @@ bool dm_helpers_dp_mst_write_payload_allocation_table( /* mst_mgr->->payloads are VC payload notify MST branch using DPCD or * AUX message. The sequence is slot 1-63 allocated sequence for each * stream. AMD ASIC stream slot allocation should follow the same - * sequence. copy DRM MST allocation to dc */ + * sequence. copy DRM MST allocation to dc + */ fill_dc_mst_payload_table_from_drm(stream->link, enable, target_payload, proposed_table); return true; @@ -430,7 +432,7 @@ void dm_dtn_log_append_v(struct dc_context *ctx, total = log_ctx->pos + n + 1; if (total > log_ctx->size) { - char *buf = (char *)kvcalloc(total, sizeof(char), GFP_KERNEL); + char *buf = kvcalloc(total, sizeof(char), GFP_KERNEL); if (buf) { memcpy(buf, log_ctx->buf, log_ctx->pos); @@ -637,7 +639,7 @@ static bool execute_synaptics_rc_command(struct drm_dp_aux *aux, ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_COMMAND, &rc_cmd, sizeof(rc_cmd)); if (ret < 0) { - DRM_ERROR(" execute_synaptics_rc_command - write cmd ..., err = %d\n", ret); + DRM_ERROR("%s: write cmd ..., err = %d\n", __func__, ret); return false; } @@ -659,7 +661,7 @@ static bool execute_synaptics_rc_command(struct drm_dp_aux *aux, drm_dp_dpcd_read(aux, SYNAPTICS_RC_DATA, data, length); } - DC_LOG_DC(" execute_synaptics_rc_command - success = %d\n", success); + DC_LOG_DC("%s: success = %d\n", __func__, success); return success; } @@ -668,7 +670,7 @@ static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux) { unsigned char data[16] = {0}; - DC_LOG_DC("Start apply_synaptics_fifo_reset_wa\n"); + DC_LOG_DC("Start %s\n", __func__); // Step 2 data[0] = 'P'; @@ -726,7 +728,7 @@ static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux) if (!execute_synaptics_rc_command(aux, true, 0x02, 0, 0, NULL)) return; - DC_LOG_DC("Done apply_synaptics_fifo_reset_wa\n"); + DC_LOG_DC("Done %s\n", __func__); } /* MST Dock */ @@ -999,9 +1001,8 @@ void dm_helpers_override_panel_settings( struct dc_panel_config *panel_config) { // Feature DSC - if (amdgpu_dc_debug_mask & DC_DISABLE_DSC) { + if (amdgpu_dc_debug_mask & DC_DISABLE_DSC) panel_config->dsc.disable_dsc_edp = true; - } } void *dm_helpers_allocate_gpu_mem( diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c index 19f543ba7205..51467f132c26 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c @@ -120,7 +120,8 @@ static void dm_irq_work_func(struct work_struct *work) /* Call a DAL subcomponent which registered for interrupt notification * at INTERRUPT_LOW_IRQ_CONTEXT. - * (The most common use is HPD interrupt) */ + * (The most common use is HPD interrupt) + */ } /* @@ -172,7 +173,8 @@ static struct list_head *remove_irq_handler(struct amdgpu_device *adev, if (handler_removed == false) { /* Not necessarily an error - caller may not - * know the context. */ + * know the context. + */ return NULL; } @@ -261,7 +263,7 @@ validate_irq_registration_params(struct dc_interrupt_params *int_params, static bool validate_irq_unregistration_params(enum dc_irq_source irq_source, irq_handler_idx handler_idx) { - if (DAL_INVALID_IRQ_HANDLER_IDX == handler_idx) { + if (handler_idx == DAL_INVALID_IRQ_HANDLER_IDX) { DRM_ERROR("DM_IRQ: invalid handler_idx==NULL!\n"); return false; } @@ -343,7 +345,8 @@ void *amdgpu_dm_irq_register_interrupt(struct amdgpu_device *adev, /* This pointer will be stored by code which requested interrupt * registration. * The same pointer will be needed in order to unregister the - * interrupt. */ + * interrupt. + */ DRM_DEBUG_KMS( "DM_IRQ: added irq handler: %p for: dal_src=%d, irq context=%d\n", @@ -390,7 +393,8 @@ void amdgpu_dm_irq_unregister_interrupt(struct amdgpu_device *adev, if (handler_list == NULL) { /* If we got here, it means we searched all irq contexts - * for this irq source, but the handler was not found. */ + * for this irq source, but the handler was not found. + */ DRM_ERROR( "DM_IRQ: failed to find irq handler:%p for irq_source:%d!\n", ih, irq_source); @@ -450,7 +454,8 @@ void amdgpu_dm_irq_fini(struct amdgpu_device *adev) DM_IRQ_TABLE_LOCK(adev, irq_table_flags); /* The handler was removed from the table, * it means it is safe to flush all the 'work' - * (because no code can schedule a new one). */ + * (because no code can schedule a new one). + */ lh = &adev->dm.irq_handler_list_low_tab[src]; DM_IRQ_TABLE_UNLOCK(adev, irq_table_flags); @@ -494,7 +499,7 @@ int amdgpu_dm_irq_suspend(struct amdgpu_device *adev) DM_IRQ_TABLE_UNLOCK(adev, irq_table_flags); if (!list_empty(hnd_list_l)) { - list_for_each_safe (entry, tmp, hnd_list_l) { + list_for_each_safe(entry, tmp, hnd_list_l) { handler = list_entry( entry, struct amdgpu_dm_irq_handler_data, @@ -571,7 +576,7 @@ static void amdgpu_dm_irq_schedule_work(struct amdgpu_device *adev, if (list_empty(handler_list)) return; - list_for_each_entry (handler_data, handler_list, list) { + list_for_each_entry(handler_data, handler_list, list) { if (queue_work(system_highpri_wq, &handler_data->work)) { work_queued = true; break; @@ -627,7 +632,8 @@ static void amdgpu_dm_irq_immediate_work(struct amdgpu_device *adev, &adev->dm.irq_handler_list_high_tab[irq_source], list) { /* Call a subcomponent which registered for immediate - * interrupt notification */ + * interrupt notification + */ handler_data->handler(handler_data->handler_arg); } @@ -664,7 +670,7 @@ static int amdgpu_dm_irq_handler(struct amdgpu_device *adev, return 0; } -static enum dc_irq_source amdgpu_dm_hpd_to_dal_irq_source(unsigned type) +static enum dc_irq_source amdgpu_dm_hpd_to_dal_irq_source(unsigned int type) { switch (type) { case AMDGPU_HPD_1: @@ -686,7 +692,7 @@ static enum dc_irq_source amdgpu_dm_hpd_to_dal_irq_source(unsigned type) static int amdgpu_dm_set_hpd_irq_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, - unsigned type, + unsigned int type, enum amdgpu_interrupt_state state) { enum dc_irq_source src = amdgpu_dm_hpd_to_dal_irq_source(type); @@ -698,7 +704,7 @@ static int amdgpu_dm_set_hpd_irq_state(struct amdgpu_device *adev, static inline int dm_irq_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, - unsigned crtc_id, + unsigned int crtc_id, enum amdgpu_interrupt_state state, const enum irq_type dal_irq_type, const char *func) @@ -729,7 +735,7 @@ static inline int dm_irq_state(struct amdgpu_device *adev, static int amdgpu_dm_set_pflip_irq_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, - unsigned crtc_id, + unsigned int crtc_id, enum amdgpu_interrupt_state state) { return dm_irq_state( @@ -743,7 +749,7 @@ static int amdgpu_dm_set_pflip_irq_state(struct amdgpu_device *adev, static int amdgpu_dm_set_crtc_irq_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, - unsigned crtc_id, + unsigned int crtc_id, enum amdgpu_interrupt_state state) { return dm_irq_state( @@ -893,13 +899,13 @@ void amdgpu_dm_hpd_init(struct amdgpu_device *adev) const struct dc_link *dc_link = amdgpu_dm_connector->dc_link; - if (DC_IRQ_SOURCE_INVALID != dc_link->irq_source_hpd) { + if (dc_link->irq_source_hpd != DC_IRQ_SOURCE_INVALID) { dc_interrupt_set(adev->dm.dc, dc_link->irq_source_hpd, true); } - if (DC_IRQ_SOURCE_INVALID != dc_link->irq_source_hpd_rx) { + if (dc_link->irq_source_hpd_rx != DC_IRQ_SOURCE_INVALID) { dc_interrupt_set(adev->dm.dc, dc_link->irq_source_hpd_rx, true); @@ -928,13 +934,13 @@ void amdgpu_dm_hpd_fini(struct amdgpu_device *adev) to_amdgpu_dm_connector(connector); const struct dc_link *dc_link = amdgpu_dm_connector->dc_link; - if (DC_IRQ_SOURCE_INVALID != dc_link->irq_source_hpd) { + if (dc_link->irq_source_hpd != DC_IRQ_SOURCE_INVALID) { dc_interrupt_set(adev->dm.dc, dc_link->irq_source_hpd, false); } - if (DC_IRQ_SOURCE_INVALID != dc_link->irq_source_hpd_rx) { + if (dc_link->irq_source_hpd_rx != DC_IRQ_SOURCE_INVALID) { dc_interrupt_set(adev->dm.dc, dc_link->irq_source_hpd_rx, false); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 46d0a8f57e55..57230661132b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -296,6 +296,7 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) if (!aconnector->edid) { struct edid *edid; + edid = drm_dp_mst_get_edid(connector, &aconnector->mst_root->mst_mgr, aconnector->mst_output_port); if (!edid) { @@ -619,8 +620,118 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, return connector; } +void dm_handle_mst_sideband_msg_ready_event( + struct drm_dp_mst_topology_mgr *mgr, + enum mst_msg_ready_type msg_rdy_type) +{ + uint8_t esi[DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI] = { 0 }; + uint8_t dret; + bool new_irq_handled = false; + int dpcd_addr; + uint8_t dpcd_bytes_to_read; + const uint8_t max_process_count = 30; + uint8_t process_count = 0; + u8 retry; + struct amdgpu_dm_connector *aconnector = + container_of(mgr, struct amdgpu_dm_connector, mst_mgr); + + + const struct dc_link_status *link_status = dc_link_get_status(aconnector->dc_link); + + if (link_status->dpcd_caps->dpcd_rev.raw < 0x12) { + dpcd_bytes_to_read = DP_LANE0_1_STATUS - DP_SINK_COUNT; + /* DPCD 0x200 - 0x201 for downstream IRQ */ + dpcd_addr = DP_SINK_COUNT; + } else { + dpcd_bytes_to_read = DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI; + /* DPCD 0x2002 - 0x2005 for downstream IRQ */ + dpcd_addr = DP_SINK_COUNT_ESI; + } + + mutex_lock(&aconnector->handle_mst_msg_ready); + + while (process_count < max_process_count) { + u8 ack[DP_PSR_ERROR_STATUS - DP_SINK_COUNT_ESI] = {}; + + process_count++; + + dret = drm_dp_dpcd_read( + &aconnector->dm_dp_aux.aux, + dpcd_addr, + esi, + dpcd_bytes_to_read); + + if (dret != dpcd_bytes_to_read) { + DRM_DEBUG_KMS("DPCD read and acked number is not as expected!"); + break; + } + + DRM_DEBUG_DRIVER("ESI %02x %02x %02x\n", esi[0], esi[1], esi[2]); + + switch (msg_rdy_type) { + case DOWN_REP_MSG_RDY_EVENT: + /* Only handle DOWN_REP_MSG_RDY case*/ + esi[1] &= DP_DOWN_REP_MSG_RDY; + break; + case UP_REQ_MSG_RDY_EVENT: + /* Only handle UP_REQ_MSG_RDY case*/ + esi[1] &= DP_UP_REQ_MSG_RDY; + break; + default: + /* Handle both cases*/ + esi[1] &= (DP_DOWN_REP_MSG_RDY | DP_UP_REQ_MSG_RDY); + break; + } + + if (!esi[1]) + break; + + /* handle MST irq */ + if (aconnector->mst_mgr.mst_state) + drm_dp_mst_hpd_irq_handle_event(&aconnector->mst_mgr, + esi, + ack, + &new_irq_handled); + + if (new_irq_handled) { + /* ACK at DPCD to notify down stream */ + for (retry = 0; retry < 3; retry++) { + ssize_t wret; + + wret = drm_dp_dpcd_writeb(&aconnector->dm_dp_aux.aux, + dpcd_addr + 1, + ack[1]); + if (wret == 1) + break; + } + + if (retry == 3) { + DRM_ERROR("Failed to ack MST event.\n"); + break; + } + + drm_dp_mst_hpd_irq_send_new_request(&aconnector->mst_mgr); + + new_irq_handled = false; + } else { + break; + } + } + + mutex_unlock(&aconnector->handle_mst_msg_ready); + + if (process_count == max_process_count) + DRM_DEBUG_DRIVER("Loop exceeded max iterations\n"); +} + +static void dm_handle_mst_down_rep_msg_ready(struct drm_dp_mst_topology_mgr *mgr) +{ + dm_handle_mst_sideband_msg_ready_event(mgr, DOWN_REP_MSG_RDY_EVENT); +} + static const struct drm_dp_mst_topology_cbs dm_mst_cbs = { .add_connector = dm_dp_add_mst_connector, + .poll_hpd_irq = dm_handle_mst_down_rep_msg_ready, }; void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, @@ -717,6 +828,7 @@ static void set_dsc_configs_from_fairness_vars(struct dsc_mst_fairness_params *p &dsc_options, 0, params[i].timing, + dc_link_get_highest_encoding_format(params[i].aconnector->dc_link), ¶ms[i].timing->dsc_cfg)) { params[i].timing->flags.DSC = 1; @@ -767,7 +879,9 @@ static int bpp_x16_from_pbn(struct dsc_mst_fairness_params param, int pbn) param.sink->ctx->dc->res_pool->dscs[0], ¶m.sink->dsc_caps.dsc_dec_caps, &dsc_options, - (int) kbps, param.timing, &dsc_config); + (int) kbps, param.timing, + dc_link_get_highest_encoding_format(param.aconnector->dc_link), + &dsc_config); return dsc_config.bits_per_pixel; } @@ -1005,8 +1119,11 @@ static int compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, dsc_policy.min_target_bpp * 16, dsc_policy.max_target_bpp * 16, &stream->sink->dsc_caps.dsc_dec_caps, - &stream->timing, ¶ms[count].bw_range)) - params[count].bw_range.stream_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing); + &stream->timing, + dc_link_get_highest_encoding_format(dc_link), + ¶ms[count].bw_range)) + params[count].bw_range.stream_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing, + dc_link_get_highest_encoding_format(dc_link)); count++; } @@ -1210,7 +1327,7 @@ int compute_mst_dsc_configs_for_state(struct drm_atomic_state *state, if (computed_streams[i]) continue; - if (!res_pool->funcs->remove_stream_from_ctx || + if (res_pool->funcs->remove_stream_from_ctx && res_pool->funcs->remove_stream_from_ctx(stream->ctx->dc, dc_state, stream) != DC_OK) return -EINVAL; @@ -1466,7 +1583,7 @@ static bool is_dsc_common_config_possible(struct dc_stream_state *stream, dsc_policy.min_target_bpp * 16, dsc_policy.max_target_bpp * 16, &stream->sink->dsc_caps.dsc_dec_caps, - &stream->timing, bw_range); + &stream->timing, dc_link_get_highest_encoding_format(stream->link), bw_range); return bw_range->max_target_bpp_x16 && bw_range->min_target_bpp_x16; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h index 1e4ede1e57ab..37c820ab0fdb 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h @@ -49,6 +49,13 @@ #define PBN_FEC_OVERHEAD_MULTIPLIER_8B_10B 1031 #define PBN_FEC_OVERHEAD_MULTIPLIER_128B_132B 1000 +enum mst_msg_ready_type { + NONE_MSG_RDY_EVENT = 0, + DOWN_REP_MSG_RDY_EVENT = 1, + UP_REQ_MSG_RDY_EVENT = 2, + DOWN_OR_UP_MSG_RDY_EVENT = 3 +}; + struct amdgpu_display_manager; struct amdgpu_dm_connector; @@ -61,6 +68,10 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm, void dm_dp_create_fake_mst_encoders(struct amdgpu_device *adev); +void dm_handle_mst_sideband_msg_ready_event( + struct drm_dp_mst_topology_mgr *mgr, + enum mst_msg_ready_type msg_rdy_type); + struct dsc_mst_fairness_vars { int pbn; bool dsc_enabled; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c index 322668973747..cc74dd69acf2 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c @@ -113,6 +113,11 @@ void amdgpu_dm_plane_fill_blending_from_plane_state(const struct drm_plane_state DRM_FORMAT_ARGB8888, DRM_FORMAT_RGBA8888, DRM_FORMAT_ABGR8888, + DRM_FORMAT_ARGB2101010, + DRM_FORMAT_ABGR2101010, + DRM_FORMAT_ARGB16161616, + DRM_FORMAT_ABGR16161616, + DRM_FORMAT_ARGB16161616F, }; uint32_t format = plane_state->fb->format->format; unsigned int i; @@ -164,7 +169,7 @@ static bool modifier_has_dcc(uint64_t modifier) return IS_AMD_FMT_MOD(modifier) && AMD_FMT_MOD_GET(DCC, modifier); } -static unsigned modifier_gfx9_swizzle_mode(uint64_t modifier) +static unsigned int modifier_gfx9_swizzle_mode(uint64_t modifier) { if (modifier == DRM_FORMAT_MOD_LINEAR) return 0; @@ -581,7 +586,7 @@ static void add_gfx11_modifiers(struct amdgpu_device *adev, int pkrs = 0; u32 gb_addr_config; u8 i = 0; - unsigned swizzle_r_x; + unsigned int swizzle_r_x; uint64_t modifier_r_x; uint64_t modifier_dcc_best; uint64_t modifier_dcc_4k; @@ -698,8 +703,8 @@ static int get_plane_formats(const struct drm_plane *plane, * caps list. */ - switch (plane->type) { - case DRM_PLANE_TYPE_PRIMARY: + if (plane->type == DRM_PLANE_TYPE_PRIMARY || + (plane_cap && plane_cap->type == DC_PLANE_TYPE_DCN_UNIVERSAL && plane->type != DRM_PLANE_TYPE_CURSOR)) { for (i = 0; i < ARRAY_SIZE(rgb_formats); ++i) { if (num_formats >= max_formats) break; @@ -717,25 +722,29 @@ static int get_plane_formats(const struct drm_plane *plane, formats[num_formats++] = DRM_FORMAT_XBGR16161616F; formats[num_formats++] = DRM_FORMAT_ABGR16161616F; } - break; + } else { + switch (plane->type) { + case DRM_PLANE_TYPE_OVERLAY: + for (i = 0; i < ARRAY_SIZE(overlay_formats); ++i) { + if (num_formats >= max_formats) + break; - case DRM_PLANE_TYPE_OVERLAY: - for (i = 0; i < ARRAY_SIZE(overlay_formats); ++i) { - if (num_formats >= max_formats) - break; + formats[num_formats++] = overlay_formats[i]; + } + break; - formats[num_formats++] = overlay_formats[i]; - } - break; + case DRM_PLANE_TYPE_CURSOR: + for (i = 0; i < ARRAY_SIZE(cursor_formats); ++i) { + if (num_formats >= max_formats) + break; - case DRM_PLANE_TYPE_CURSOR: - for (i = 0; i < ARRAY_SIZE(cursor_formats); ++i) { - if (num_formats >= max_formats) - break; + formats[num_formats++] = cursor_formats[i]; + } + break; - formats[num_formats++] = cursor_formats[i]; + default: + break; } - break; } return num_formats; @@ -1260,6 +1269,13 @@ void amdgpu_dm_plane_handle_cursor_update(struct drm_plane *plane, attributes.rotation_angle = 0; attributes.attribute_flags.value = 0; + /* Enable cursor degamma ROM on DCN3+ for implicit sRGB degamma in DRM + * legacy gamma setup. + */ + if (crtc_state->cm_is_degamma_srgb && + adev->dm.dc->caps.color.dpp.gamma_corr) + attributes.attribute_flags.bits.ENABLE_CURSOR_DEGAMMA = 1; + attributes.pitch = afb->base.pitches[0] / afb->base.format->cpp[0]; if (crtc_state->stream) { @@ -1459,6 +1475,15 @@ int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm, drm_plane_create_blend_mode_property(plane, blend_caps); } + if (plane->type == DRM_PLANE_TYPE_PRIMARY) { + drm_plane_create_zpos_immutable_property(plane, 0); + } else if (plane->type == DRM_PLANE_TYPE_OVERLAY) { + unsigned int zpos = 1 + drm_plane_index(plane); + drm_plane_create_zpos_property(plane, zpos, 1, 254); + } else if (plane->type == DRM_PLANE_TYPE_CURSOR) { + drm_plane_create_zpos_immutable_property(plane, 255); + } + if (plane->type == DRM_PLANE_TYPE_PRIMARY && plane_cap && (plane_cap->pixel_format_support.nv12 || diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c index 75284e2cec74..848c5b4bb301 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c @@ -334,7 +334,8 @@ bool dm_pp_get_clock_levels_by_type( if (dc_clks->clocks_in_khz[i] > validation_clks.engine_max_clock) { /* This clock is higher the validation clock. * Than means the previous one is the highest - * non-boosted one. */ + * non-boosted one. + */ DRM_INFO("DM_PPLIB: reducing engine clock level from %d to %d\n", dc_clks->num_levels, i); dc_clks->num_levels = i > 0 ? i : 1; @@ -406,10 +407,10 @@ bool dm_pp_notify_wm_clock_changes( * TODO: expand this to other ASICs */ if ((adev->asic_type >= CHIP_POLARIS10) && - (adev->asic_type <= CHIP_VEGAM) && - !amdgpu_dpm_set_watermarks_for_clocks_ranges(adev, - (void *)wm_with_clock_ranges)) - return true; + (adev->asic_type <= CHIP_VEGAM) && + !amdgpu_dpm_set_watermarks_for_clocks_ranges(adev, + (void *)wm_with_clock_ranges)) + return true; return false; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c index 4f61d4f257cd..08ce3bb8f640 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c @@ -166,6 +166,7 @@ bool amdgpu_dm_psr_enable(struct dc_stream_state *stream) */ if (vsync_rate_hz != 0) { unsigned int frame_time_microsec = 1000000 / vsync_rate_hz; + num_frames_static = (30000 / frame_time_microsec) + 1; } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c new file mode 100644 index 000000000000..32d3086c4cb7 --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c @@ -0,0 +1,183 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "amdgpu_dm_replay.h" +#include "dc.h" +#include "dm_helpers.h" +#include "amdgpu_dm.h" +#include "modules/power/power_helpers.h" +#include "dmub/inc/dmub_cmd.h" +#include "dc/inc/link.h" + +/* + * link_supports_replay() - check if the link supports replay + * @link: link + * @aconnector: aconnector + * + */ +static bool link_supports_replay(struct dc_link *link, struct amdgpu_dm_connector *aconnector) +{ + struct dm_connector_state *state = to_dm_connector_state(aconnector->base.state); + struct dpcd_caps *dpcd_caps = &link->dpcd_caps; + struct adaptive_sync_caps *as_caps = &link->dpcd_caps.adaptive_sync_caps; + + if (!state->freesync_capable) + return false; + + if (!aconnector->vsdb_info.replay_mode) + return false; + + // Check the eDP version + if (dpcd_caps->edp_rev < EDP_REVISION_13) + return false; + + if (!dpcd_caps->alpm_caps.bits.AUX_WAKE_ALPM_CAP) + return false; + + // Check adaptive sync support cap + if (!as_caps->dp_adap_sync_caps.bits.ADAPTIVE_SYNC_SDP_SUPPORT) + return false; + + return true; +} + +/* + * amdgpu_dm_setup_replay() - setup replay configuration + * @link: link + * @aconnector: aconnector + * + */ +bool amdgpu_dm_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *aconnector) +{ + struct replay_config pr_config; + union replay_debug_flags *debug_flags = NULL; + + // For eDP, if Replay is supported, return true to skip checks + if (link->replay_settings.config.replay_supported) + return true; + + if (!dc_is_embedded_signal(link->connector_signal)) + return false; + + if (link->panel_config.psr.disallow_replay) + return false; + + if (!link_supports_replay(link, aconnector)) + return false; + + // Mark Replay is supported in link and update related attributes + pr_config.replay_supported = true; + pr_config.replay_power_opt_supported = 0; + pr_config.replay_enable_option |= pr_enable_option_static_screen; + pr_config.replay_timing_sync_supported = aconnector->max_vfreq >= 2 * aconnector->min_vfreq ? true : false; + + if (!pr_config.replay_timing_sync_supported) + pr_config.replay_enable_option &= ~pr_enable_option_general_ui; + + debug_flags = (union replay_debug_flags *)&pr_config.debug_flags; + debug_flags->u32All = 0; + debug_flags->bitfields.visual_confirm = + link->ctx->dc->debug.visual_confirm == VISUAL_CONFIRM_REPLAY ? true : false; + + link->replay_settings.replay_feature_enabled = true; + + init_replay_config(link, &pr_config); + + return true; +} + + +/* + * amdgpu_dm_replay_enable() - enable replay f/w + * @stream: stream state + * + * Return: true if success + */ +bool amdgpu_dm_replay_enable(struct dc_stream_state *stream, bool wait) +{ + uint64_t state; + unsigned int retry_count; + bool replay_active = true; + const unsigned int max_retry = 1000; + bool force_static = true; + struct dc_link *link = NULL; + + + if (stream == NULL) + return false; + + link = stream->link; + + if (link == NULL) + return false; + + link->dc->link_srv->edp_setup_replay(link, stream); + + link->dc->link_srv->edp_set_replay_allow_active(link, NULL, false, false, NULL); + + link->dc->link_srv->edp_set_replay_allow_active(link, &replay_active, false, true, NULL); + + if (wait == true) { + + for (retry_count = 0; retry_count <= max_retry; retry_count++) { + dc_link_get_replay_state(link, &state); + if (replay_active) { + if (state != REPLAY_STATE_0 && + (!force_static || state == REPLAY_STATE_3)) + break; + } else { + if (state == REPLAY_STATE_0) + break; + } + udelay(500); + } + + /* assert if max retry hit */ + if (retry_count >= max_retry) + ASSERT(0); + } else { + /* To-do: Add trace log */ + } + + return true; +} + +/* + * amdgpu_dm_replay_disable() - disable replay f/w + * @stream: stream state + * + * Return: true if success + */ +bool amdgpu_dm_replay_disable(struct dc_stream_state *stream) +{ + + if (stream->link) { + DRM_DEBUG_DRIVER("Disabling replay...\n"); + stream->link->dc->link_srv->edp_set_replay_allow_active(stream->link, NULL, false, false, NULL); + return true; + } + + return false; +} diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h new file mode 100644 index 000000000000..01cba3cd6246 --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h @@ -0,0 +1,46 @@ +/* + * Copyright 2021 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef AMDGPU_DM_AMDGPU_DM_REPLAY_H_ +#define AMDGPU_DM_AMDGPU_DM_REPLAY_H_ + +#include "amdgpu.h" + +enum replay_enable_option { + pr_enable_option_static_screen = 0x1, + pr_enable_option_mpo_video = 0x2, + pr_enable_option_full_screen_video = 0x4, + pr_enable_option_general_ui = 0x8, + pr_enable_option_static_screen_coasting = 0x10000, + pr_enable_option_mpo_video_coasting = 0x20000, + pr_enable_option_full_screen_video_coasting = 0x40000, +}; + + +bool amdgpu_dm_replay_enable(struct dc_stream_state *stream, bool enable); +bool amdgpu_dm_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *aconnector); +bool amdgpu_dm_replay_disable(struct dc_stream_state *stream); + +#endif /* AMDGPU_DM_AMDGPU_DM_REPLAY_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile index 69ffd4424dc7..1b8c2aef4633 100644 --- a/drivers/gpu/drm/amd/display/dc/Makefile +++ b/drivers/gpu/drm/amd/display/dc/Makefile @@ -78,3 +78,4 @@ DC_EDID += dc_edid_parser.o AMD_DISPLAY_DMUB = $(addprefix $(AMDDALPATH)/dc/,$(DC_DMUB)) AMD_DISPLAY_EDID = $(addprefix $(AMDDALPATH)/dc/,$(DC_EDID)) AMD_DISPLAY_FILES += $(AMD_DISPLAY_DMUB) $(AMD_DISPLAY_EDID) + diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.c b/drivers/gpu/drm/amd/display/dc/basics/conversion.c index 352e9afb85c6..e295a839ab47 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.c +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.c @@ -24,7 +24,7 @@ */ #include "dm_services.h" -#include "conversion.h" +#include "basics/conversion.h" #define DIVIDER 10000 diff --git a/drivers/gpu/drm/amd/display/dc/basics/vector.c b/drivers/gpu/drm/amd/display/dc/basics/vector.c index 84aeccf36b4b..6d2924114a3e 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/vector.c +++ b/drivers/gpu/drm/amd/display/dc/basics/vector.c @@ -50,12 +50,11 @@ bool dal_vector_construct( return true; } -static bool dal_vector_presized_costruct( - struct vector *vector, - struct dc_context *ctx, - uint32_t count, - void *initial_value, - uint32_t struct_size) +static bool dal_vector_presized_costruct(struct vector *vector, + struct dc_context *ctx, + uint32_t count, + void *initial_value, + uint32_t struct_size) { uint32_t i; diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c index 27af9d3c2b73..6b3190447581 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c @@ -96,7 +96,7 @@ struct dc_bios *bios_parser_create( struct bp_init_data *init, enum dce_version dce_version) { - struct bios_parser *bp = NULL; + struct bios_parser *bp; bp = kzalloc(sizeof(struct bios_parser), GFP_KERNEL); if (!bp) @@ -2576,7 +2576,7 @@ static struct integrated_info *bios_parser_create_integrated_info( struct dc_bios *dcb) { struct bios_parser *bp = BP_FROM_DCB(dcb); - struct integrated_info *info = NULL; + struct integrated_info *info; info = kzalloc(sizeof(struct integrated_info), GFP_KERNEL); @@ -2593,11 +2593,10 @@ static struct integrated_info *bios_parser_create_integrated_info( return NULL; } -static enum bp_result update_slot_layout_info( - struct dc_bios *dcb, - unsigned int i, - struct slot_layout_info *slot_layout_info, - unsigned int record_offset) +static enum bp_result update_slot_layout_info(struct dc_bios *dcb, + unsigned int i, + struct slot_layout_info *slot_layout_info, + unsigned int record_offset) { unsigned int j; struct bios_parser *bp; @@ -2696,10 +2695,9 @@ static enum bp_result update_slot_layout_info( } -static enum bp_result get_bracket_layout_record( - struct dc_bios *dcb, - unsigned int bracket_layout_id, - struct slot_layout_info *slot_layout_info) +static enum bp_result get_bracket_layout_record(struct dc_bios *dcb, + unsigned int bracket_layout_id, + struct slot_layout_info *slot_layout_info) { unsigned int i; unsigned int record_offset; diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index cce47d3f1a13..484d62bcf2c2 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -340,9 +340,8 @@ static struct atom_display_object_path_v2 *get_bios_object( } /* from graphics_object_id, find display path which includes the object_id */ -static struct atom_display_object_path_v3 *get_bios_object_from_path_v3( - struct bios_parser *bp, - struct graphics_object_id id) +static struct atom_display_object_path_v3 *get_bios_object_from_path_v3(struct bios_parser *bp, + struct graphics_object_id id) { unsigned int i; struct graphics_object_id obj_id = {0}; @@ -521,9 +520,8 @@ static enum bp_result get_gpio_i2c_info( return BP_RESULT_OK; } -static struct atom_hpd_int_record *get_hpd_record_for_path_v3( - struct bios_parser *bp, - struct atom_display_object_path_v3 *object) +static struct atom_hpd_int_record *get_hpd_record_for_path_v3(struct bios_parser *bp, + struct atom_display_object_path_v3 *object) { struct atom_common_record_header *header; uint32_t offset; @@ -774,20 +772,20 @@ static enum bp_result bios_parser_get_device_tag( return BP_RESULT_BADINPUT; switch (bp->object_info_tbl.revision.minor) { - case 4: - default: + case 4: + default: /* getBiosObject will return MXM object */ - object = get_bios_object(bp, connector_object_id); + object = get_bios_object(bp, connector_object_id); if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object id */ return BP_RESULT_BADINPUT; } - info->acpi_device = 0; /* BIOS no longer provides this */ - info->dev_id = device_type_from_device_id(object->device_tag); - break; - case 5: + info->acpi_device = 0; /* BIOS no longer provides this */ + info->dev_id = device_type_from_device_id(object->device_tag); + break; + case 5: object_path_v3 = get_bios_object_from_path_v3(bp, connector_object_id); if (!object_path_v3) { @@ -1582,13 +1580,13 @@ static bool bios_parser_is_device_id_supported( uint32_t mask = get_support_mask_for_device_id(id); switch (bp->object_info_tbl.revision.minor) { - case 4: - default: - return (le16_to_cpu(bp->object_info_tbl.v1_4->supporteddevices) & mask) != 0; - break; - case 5: - return (le16_to_cpu(bp->object_info_tbl.v1_5->supporteddevices) & mask) != 0; - break; + case 4: + default: + return (le16_to_cpu(bp->object_info_tbl.v1_4->supporteddevices) & mask) != 0; + break; + case 5: + return (le16_to_cpu(bp->object_info_tbl.v1_5->supporteddevices) & mask) != 0; + break; } return false; @@ -1757,7 +1755,7 @@ static enum bp_result bios_parser_get_firmware_info( case 2: case 3: result = get_firmware_info_v3_2(bp, info); - break; + break; case 4: result = get_firmware_info_v3_4(bp, info); break; @@ -2175,9 +2173,8 @@ static struct atom_disp_connector_caps_record *get_disp_connector_caps_record( return NULL; } -static struct atom_connector_caps_record *get_connector_caps_record( - struct bios_parser *bp, - struct atom_display_object_path_v3 *object) +static struct atom_connector_caps_record *get_connector_caps_record(struct bios_parser *bp, + struct atom_display_object_path_v3 *object) { struct atom_common_record_header *header; uint32_t offset; @@ -2228,7 +2225,7 @@ static enum bp_result bios_parser_get_disp_connector_caps_info( return BP_RESULT_BADINPUT; switch (bp->object_info_tbl.revision.minor) { - case 4: + case 4: default: object = get_bios_object(bp, object_id); @@ -2264,9 +2261,8 @@ static enum bp_result bios_parser_get_disp_connector_caps_info( return BP_RESULT_OK; } -static struct atom_connector_speed_record *get_connector_speed_cap_record( - struct bios_parser *bp, - struct atom_display_object_path_v3 *object) +static struct atom_connector_speed_record *get_connector_speed_cap_record(struct bios_parser *bp, + struct atom_display_object_path_v3 *object) { struct atom_common_record_header *header; uint32_t offset; @@ -3090,7 +3086,7 @@ static struct integrated_info *bios_parser_create_integrated_info( struct dc_bios *dcb) { struct bios_parser *bp = BP_FROM_DCB(dcb); - struct integrated_info *info = NULL; + struct integrated_info *info; info = kzalloc(sizeof(struct integrated_info), GFP_KERNEL); @@ -3679,7 +3675,7 @@ struct dc_bios *firmware_parser_create( struct bp_init_data *init, enum dce_version dce_version) { - struct bios_parser *bp = NULL; + struct bios_parser *bp; bp = kzalloc(sizeof(struct bios_parser), GFP_KERNEL); if (!bp) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index 6127d6045336..dcedf9645161 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -117,6 +117,7 @@ void clk_mgr_exit_optimized_pwr_state(const struct dc *dc, struct clk_mgr *clk_m continue; clk_mgr->psr_allow_active_cache = edp_link->psr_settings.psr_allow_active; dc->link_srv->edp_set_psr_allow_active(edp_link, &allow_active, false, false, NULL); + dc->link_srv->edp_set_replay_allow_active(edp_link, &allow_active, false, false, NULL); } } @@ -137,6 +138,8 @@ void clk_mgr_optimize_pwr_state(const struct dc *dc, struct clk_mgr *clk_mgr) continue; dc->link_srv->edp_set_psr_allow_active(edp_link, &clk_mgr->psr_allow_active_cache, false, false, NULL); + dc->link_srv->edp_set_replay_allow_active(edp_link, + &clk_mgr->psr_allow_active_cache, false, false, NULL); } } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c index 7ccd96959256..3db4ef564b99 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c @@ -87,6 +87,11 @@ static int dcn31_get_active_display_cnt_wa( stream->signal == SIGNAL_TYPE_DVI_SINGLE_LINK || stream->signal == SIGNAL_TYPE_DVI_DUAL_LINK) tmds_present = true; + + /* Checking stream / link detection ensuring that PHY is active*/ + if (dc_is_dp_signal(stream->signal) && !stream->dpms_off) + display_count++; + } for (i = 0; i < dc->link_count; i++) { diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c index 2f7c8996b19d..7326b7565846 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c @@ -783,7 +783,6 @@ void dcn314_clk_mgr_construct( clk_mgr->base.base.clks.ref_dtbclk_khz = 600000; dce_clock_read_ss_info(&clk_mgr->base); /*if bios enabled SS, driver needs to adjust dtb clock, only enable with correct bios*/ - //clk_mgr->base.dccg->ref_dtbclk_khz = dce_adjust_dp_ref_freq_for_ss(clk_mgr_internal, clk_mgr->base.base.dprefclk_khz); clk_mgr->base.base.bw_params = &dcn314_bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c index 925d6e13620e..1042cf1a3ab0 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_smu.c @@ -32,29 +32,28 @@ #define MAX_INSTANCE 6 #define MAX_SEGMENT 6 +#define SMU_REGISTER_WRITE_RETRY_COUNT 5 -struct IP_BASE_INSTANCE -{ +struct IP_BASE_INSTANCE { unsigned int segment[MAX_SEGMENT]; }; -struct IP_BASE -{ +struct IP_BASE { struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; }; static const struct IP_BASE MP0_BASE = { { { { 0x00016000, 0x00DC0000, 0x00E00000, 0x00E40000, 0x0243FC00, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } } } }; + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; static const struct IP_BASE NBIO_BASE = { { { { 0x00000000, 0x00000014, 0x00000D20, 0x00010400, 0x0241B000, 0x04040000 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } } } }; + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; #define regBIF_BX_PF2_RSMU_INDEX 0x0000 #define regBIF_BX_PF2_RSMU_INDEX_BASE_IDX 1 @@ -134,6 +133,8 @@ static int dcn315_smu_send_msg_with_param( unsigned int msg_id, unsigned int param) { uint32_t result; + uint32_t i = 0; + uint32_t read_back_data; result = dcn315_smu_wait_for_response(clk_mgr, 10, 200000); @@ -150,10 +151,19 @@ static int dcn315_smu_send_msg_with_param( /* Set the parameter register for the SMU message, unit is Mhz */ REG_WRITE(MP1_SMN_C2PMSG_37, param); - /* Trigger the message transaction by writing the message ID */ - generic_write_indirect_reg(CTX, - REG_NBIO(RSMU_INDEX), REG_NBIO(RSMU_DATA), - mmMP1_C2PMSG_3, msg_id); + for (i = 0; i < SMU_REGISTER_WRITE_RETRY_COUNT; i++) { + /* Trigger the message transaction by writing the message ID */ + generic_write_indirect_reg(CTX, + REG_NBIO(RSMU_INDEX), REG_NBIO(RSMU_DATA), + mmMP1_C2PMSG_3, msg_id); + read_back_data = generic_read_indirect_reg(CTX, + REG_NBIO(RSMU_INDEX), REG_NBIO(RSMU_DATA), + mmMP1_C2PMSG_3); + if (read_back_data == msg_id) + break; + udelay(2); + smu_print("SMU msg id write fail %x times. \n", i + 1); + } result = dcn315_smu_wait_for_response(clk_mgr, 10, 200000); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c index d7de756301cf..09151cc56ce4 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c @@ -45,24 +45,14 @@ #define MAX_INSTANCE 7 #define MAX_SEGMENT 6 -struct IP_BASE_INSTANCE -{ +struct IP_BASE_INSTANCE { unsigned int segment[MAX_SEGMENT]; }; -struct IP_BASE -{ +struct IP_BASE { struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; }; -static const struct IP_BASE CLK_BASE = { { { { 0x00016C00, 0x02401800, 0, 0, 0, 0 } }, - { { 0x00016E00, 0x02401C00, 0, 0, 0, 0 } }, - { { 0x00017000, 0x02402000, 0, 0, 0, 0 } }, - { { 0x00017200, 0x02402400, 0, 0, 0, 0 } }, - { { 0x0001B000, 0x0242D800, 0, 0, 0, 0 } }, - { { 0x0001B200, 0x0242DC00, 0, 0, 0, 0 } }, - { { 0x0001B400, 0x0242E000, 0, 0, 0, 0 } } } }; - #define regCLK1_CLK_PLL_REQ 0x0237 #define regCLK1_CLK_PLL_REQ_BASE_IDX 0 @@ -73,9 +63,6 @@ static const struct IP_BASE CLK_BASE = { { { { 0x00016C00, 0x02401800, 0, 0, 0, #define CLK1_CLK_PLL_REQ__PllSpineDiv_MASK 0x0000F000L #define CLK1_CLK_PLL_REQ__FbMult_frac_MASK 0xFFFF0000L -#define REG(reg_name) \ - (CLK_BASE.instance[0].segment[reg ## reg_name ## _BASE_IDX] + reg ## reg_name) - #define TO_CLK_MGR_DCN316(clk_mgr)\ container_of(clk_mgr, struct clk_mgr_dcn316, base) @@ -577,36 +564,6 @@ static struct clk_mgr_funcs dcn316_funcs = { }; extern struct clk_mgr_funcs dcn3_fpga_funcs; -static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr) -{ - /* get FbMult value */ - struct fixed31_32 pll_req; - unsigned int fbmult_frac_val = 0; - unsigned int fbmult_int_val = 0; - - /* - * Register value of fbmult is in 8.16 format, we are converting to 31.32 - * to leverage the fix point operations available in driver - */ - - REG_GET(CLK1_CLK_PLL_REQ, FbMult_frac, &fbmult_frac_val); /* 16 bit fractional part*/ - REG_GET(CLK1_CLK_PLL_REQ, FbMult_int, &fbmult_int_val); /* 8 bit integer part */ - - pll_req = dc_fixpt_from_int(fbmult_int_val); - - /* - * since fractional part is only 16 bit in register definition but is 32 bit - * in our fix point definiton, need to shift left by 16 to obtain correct value - */ - pll_req.value |= fbmult_frac_val << 16; - - /* multiply by REFCLK period */ - pll_req = dc_fixpt_mul_int(pll_req, clk_mgr->dfs_ref_freq_khz); - - /* integer part is now VCO frequency in kHz */ - return dc_fixpt_floor(pll_req); -} - void dcn316_clk_mgr_construct( struct dc_context *ctx, struct clk_mgr_dcn316 *clk_mgr, @@ -660,7 +617,8 @@ void dcn316_clk_mgr_construct( clk_mgr->base.smu_present = true; // Skip this for now as it did not work on DCN315, renable during bring up - clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); + //clk_mgr->base.base.dentist_vco_freq_khz = get_vco_frequency_from_reg(&clk_mgr->base); + clk_mgr->base.base.dentist_vco_freq_khz = 2500000; /* in case we don't get a value from the register, use default */ if (clk_mgr->base.base.dentist_vco_freq_khz == 0) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c index 457a9254ae1c..3ed19197a755 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_smu.c @@ -34,23 +34,21 @@ #define MAX_INSTANCE 7 #define MAX_SEGMENT 6 -struct IP_BASE_INSTANCE -{ +struct IP_BASE_INSTANCE { unsigned int segment[MAX_SEGMENT]; }; -struct IP_BASE -{ +struct IP_BASE { struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; }; static const struct IP_BASE MP0_BASE = { { { { 0x00016000, 0x00DC0000, 0x00E00000, 0x00E40000, 0x0243FC00, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } }, - { { 0, 0, 0, 0, 0, 0 } } } }; + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } }, + { { 0, 0, 0, 0, 0, 0 } } } }; #define REG(reg_name) \ (MP0_BASE.instance[0].segment[reg ## reg_name ## _BASE_IDX] + reg ## reg_name) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c index cb992aca760d..984b52923534 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c @@ -297,7 +297,7 @@ void dcn32_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, clk_mgr->dccg->ref_dppclk = clk_mgr->base.clks.dppclk_khz; for (i = 0; i < clk_mgr->base.ctx->dc->res_pool->pipe_count; i++) { - int dpp_inst, dppclk_khz, prev_dppclk_khz; + int dpp_inst = 0, dppclk_khz, prev_dppclk_khz; dppclk_khz = context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz; @@ -555,6 +555,11 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, } } + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) + dcn32_smu_wait_for_dmub_ack_mclk(clk_mgr, true); + else + dcn32_smu_wait_for_dmub_ack_mclk(clk_mgr, false); + /* Always update saved value, even if new value not set due to P-State switching unsupported. Also check safe_to_lower for FCLK */ if (safe_to_lower && (clk_mgr_base->clks.fclk_p_state_change_support != clk_mgr_base->clks.fclk_prev_p_state_change_support)) { update_fclk = true; @@ -802,7 +807,7 @@ static void dcn32_set_hard_min_memclk(struct clk_mgr *clk_mgr_base, bool current khz_to_mhz_ceil(clk_mgr_base->clks.dramclk_khz)); else dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, - clk_mgr_base->bw_params->clk_table.entries[clk_mgr_base->bw_params->clk_table.num_entries_per_clk.num_memclk_levels - 1].memclk_mhz); + clk_mgr_base->bw_params->max_memclk_mhz); } else { dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, clk_mgr_base->bw_params->clk_table.entries[0].memclk_mhz); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c index fb524fe4ab26..700ce42036d7 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c @@ -139,3 +139,10 @@ unsigned int dcn32_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, ui return response; } + +void dcn32_smu_wait_for_dmub_ack_mclk(struct clk_mgr_internal *clk_mgr, bool enable) +{ + smu_print("PMFW to wait for DMCUB ack for MCLK : %d\n", enable); + + dcn32_smu_send_msg_with_param(clk_mgr, 0x14, enable ? 1 : 0, NULL); +} diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.h index a68038a41972..a34c258c19dc 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.h +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.h @@ -43,5 +43,6 @@ void dcn32_smu_set_pme_workaround(struct clk_mgr_internal *clk_mgr); void dcn32_smu_send_cab_for_uclk_message(struct clk_mgr_internal *clk_mgr, unsigned int num_ways); void dcn32_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr); unsigned int dcn32_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz); +void dcn32_smu_wait_for_dmub_ack_mclk(struct clk_mgr_internal *clk_mgr, bool enable); #endif /* __DCN32_CLK_MGR_SMU_MSG_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index d133e4186a52..3a9077b60029 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -586,18 +586,15 @@ dc_stream_forward_crc_window(struct dc_stream_state *stream, bool dc_stream_configure_crc(struct dc *dc, struct dc_stream_state *stream, struct crc_params *crc_window, bool enable, bool continuous) { - int i; struct pipe_ctx *pipe; struct crc_params param; struct timing_generator *tg; - for (i = 0; i < MAX_PIPES; i++) { - pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe->stream == stream && !pipe->top_pipe && !pipe->prev_odm_pipe) - break; - } + pipe = resource_get_otg_master_for_stream( + &dc->current_state->res_ctx, stream); + /* Stream not found */ - if (i == MAX_PIPES) + if (pipe == NULL) return false; /* By default, capture the full frame */ @@ -1047,8 +1044,10 @@ static void disable_all_writeback_pipes_for_stream( stream->writeback_info[i].wb_enabled = false; } -static void apply_ctx_interdependent_lock(struct dc *dc, struct dc_state *context, - struct dc_stream_state *stream, bool lock) +static void apply_ctx_interdependent_lock(struct dc *dc, + struct dc_state *context, + struct dc_stream_state *stream, + bool lock) { int i; @@ -1062,7 +1061,7 @@ static void apply_ctx_interdependent_lock(struct dc *dc, struct dc_state *contex // Copied conditions that were previously in dce110_apply_ctx_for_surface if (stream == pipe_ctx->stream) { - if (!pipe_ctx->top_pipe && + if (resource_is_pipe_type(pipe_ctx, OPP_HEAD) && (pipe_ctx->plane_state || old_pipe_ctx->plane_state)) dc->hwss.pipe_control_lock(dc, pipe_ctx, lock); } @@ -1923,6 +1922,14 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c dc_trigger_sync(dc, context); + /* Full update should unconditionally be triggered when dc_commit_state_no_check is called */ + for (i = 0; i < context->stream_count; i++) { + uint32_t prev_dsc_changed = context->streams[i]->update_flags.bits.dsc_changed; + + context->streams[i]->update_flags.raw = 0xFFFFFFFF; + context->streams[i]->update_flags.bits.dsc_changed = prev_dsc_changed; + } + /* Program all planes within new context*/ if (dc->hwss.program_front_end_for_ctx) { dc->hwss.interdependent_update_lock(dc, context, true); @@ -2001,6 +2008,11 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c for (i = 0; i < context->stream_count; i++) context->streams[i]->mode_changed = false; + /* Clear update flags that were set earlier to avoid redundant programming */ + for (i = 0; i < context->stream_count; i++) { + context->streams[i]->update_flags.raw = 0x0; + } + old_state = dc->current_state; dc->current_state = context; @@ -2061,12 +2073,12 @@ enum dc_status dc_commit_streams(struct dc *dc, } } - /* Check for case where we are going from odm 2:1 to max - * pipe scenario. For these cases, we will call - * commit_minimal_transition_state() to exit out of odm 2:1 - * first before processing new streams + /* ODM Combine 2:1 power optimization is only applied for single stream + * scenario, it uses extra pipes than needed to reduce power consumption + * We need to switch off this feature to make room for new streams. */ - if (stream_count == dc->res_pool->pipe_count) { + if (stream_count > dc->current_state->stream_count && + dc->current_state->stream_count == 1) { for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->next_odm_pipe) @@ -2480,9 +2492,7 @@ static enum surface_update_type get_scaling_info_update_type( if (!u->scaling_info) return UPDATE_TYPE_FAST; - if (u->scaling_info->clip_rect.width != u->surface->clip_rect.width - || u->scaling_info->clip_rect.height != u->surface->clip_rect.height - || u->scaling_info->dst_rect.width != u->surface->dst_rect.width + if (u->scaling_info->dst_rect.width != u->surface->dst_rect.width || u->scaling_info->dst_rect.height != u->surface->dst_rect.height || u->scaling_info->scaling_quality.integer_scaling != u->surface->scaling_quality.integer_scaling @@ -2686,96 +2696,6 @@ static enum surface_update_type check_update_surfaces_for_stream( return overall_type; } -static bool dc_check_is_fullscreen_video(struct rect src, struct rect clip_rect) -{ - int view_height, view_width, clip_x, clip_y, clip_width, clip_height; - - view_height = src.height; - view_width = src.width; - - clip_x = clip_rect.x; - clip_y = clip_rect.y; - - clip_width = clip_rect.width; - clip_height = clip_rect.height; - - /* check for centered video accounting for off by 1 scaling truncation */ - if ((view_height - clip_y - clip_height <= clip_y + 1) && - (view_width - clip_x - clip_width <= clip_x + 1) && - (view_height - clip_y - clip_height >= clip_y - 1) && - (view_width - clip_x - clip_width >= clip_x - 1)) { - - /* when OS scales up/down to letter box, it may end up - * with few blank pixels on the border due to truncating. - * Add offset margin to account for this - */ - if (clip_x <= 4 || clip_y <= 4) - return true; - } - - return false; -} - -static enum surface_update_type check_boundary_crossing_for_windowed_mpo_with_odm(struct dc *dc, - struct dc_surface_update *srf_updates, int surface_count, - enum surface_update_type update_type) -{ - enum surface_update_type new_update_type = update_type; - int i, j; - struct pipe_ctx *pipe = NULL; - struct dc_stream_state *stream; - - /* Check that we are in windowed MPO with ODM - * - look for MPO pipe by scanning pipes for first pipe matching - * surface that has moved ( position change ) - * - MPO pipe will have top pipe - * - check that top pipe has ODM pointer - */ - if ((surface_count > 1) && dc->config.enable_windowed_mpo_odm) { - for (i = 0; i < surface_count; i++) { - if (srf_updates[i].surface && srf_updates[i].scaling_info - && srf_updates[i].surface->update_flags.bits.position_change) { - - for (j = 0; j < dc->res_pool->pipe_count; j++) { - if (srf_updates[i].surface == dc->current_state->res_ctx.pipe_ctx[j].plane_state) { - pipe = &dc->current_state->res_ctx.pipe_ctx[j]; - stream = pipe->stream; - break; - } - } - - if (pipe && pipe->top_pipe && (get_num_odm_splits(pipe->top_pipe) > 0) && stream - && !dc_check_is_fullscreen_video(stream->src, srf_updates[i].scaling_info->clip_rect)) { - struct rect old_clip_rect, new_clip_rect; - bool old_clip_rect_left, old_clip_rect_right, old_clip_rect_middle; - bool new_clip_rect_left, new_clip_rect_right, new_clip_rect_middle; - - old_clip_rect = srf_updates[i].surface->clip_rect; - new_clip_rect = srf_updates[i].scaling_info->clip_rect; - - old_clip_rect_left = ((old_clip_rect.x + old_clip_rect.width) <= (stream->src.x + (stream->src.width/2))); - old_clip_rect_right = (old_clip_rect.x >= (stream->src.x + (stream->src.width/2))); - old_clip_rect_middle = !old_clip_rect_left && !old_clip_rect_right; - - new_clip_rect_left = ((new_clip_rect.x + new_clip_rect.width) <= (stream->src.x + (stream->src.width/2))); - new_clip_rect_right = (new_clip_rect.x >= (stream->src.x + (stream->src.width/2))); - new_clip_rect_middle = !new_clip_rect_left && !new_clip_rect_right; - - if (old_clip_rect_left && new_clip_rect_middle) - new_update_type = UPDATE_TYPE_FULL; - else if (old_clip_rect_middle && new_clip_rect_right) - new_update_type = UPDATE_TYPE_FULL; - else if (old_clip_rect_right && new_clip_rect_middle) - new_update_type = UPDATE_TYPE_FULL; - else if (old_clip_rect_middle && new_clip_rect_left) - new_update_type = UPDATE_TYPE_FULL; - } - } - } - } - return new_update_type; -} - /* * dc_check_update_surfaces_for_stream() - Determine update type (fast, med, or full) * @@ -2807,10 +2727,6 @@ enum surface_update_type dc_check_update_surfaces_for_stream( updates[i].surface->update_flags.raw = 0xFFFFFFFF; } - if (type == UPDATE_TYPE_MED) - type = check_boundary_crossing_for_windowed_mpo_with_odm(dc, - updates, surface_count, type); - if (type == UPDATE_TYPE_FAST) { // If there's an available clock comparator, we use that. if (dc->clk_mgr->funcs->are_clock_states_equal) { @@ -3245,7 +3161,7 @@ static void commit_planes_do_stream_update(struct dc *dc, for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - if (!pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe && pipe_ctx->stream == stream) { + if (resource_is_pipe_type(pipe_ctx, OTG_MASTER) && pipe_ctx->stream == stream) { if (stream_update->periodic_interrupt && dc->hwss.setup_periodic_interrupt) dc->hwss.setup_periodic_interrupt(dc, pipe_ctx); @@ -3370,6 +3286,9 @@ static bool dc_dmub_should_send_dirty_rect_cmd(struct dc *dc, struct dc_stream_s && stream->ctx->dce_version >= DCN_VERSION_3_1) return true; + if (stream->link->replay_settings.config.replay_supported) + return true; + return false; } @@ -3524,16 +3443,9 @@ static void commit_planes_for_stream_fast(struct dc *dc, struct pipe_ctx *top_pipe_to_program = NULL; dc_z10_restore(dc); - for (j = 0; j < dc->res_pool->pipe_count; j++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - - if (!pipe_ctx->top_pipe && - !pipe_ctx->prev_odm_pipe && - pipe_ctx->stream && - pipe_ctx->stream == stream) { - top_pipe_to_program = pipe_ctx; - } - } + top_pipe_to_program = resource_get_otg_master_for_stream( + &context->res_ctx, + stream); if (dc->debug.visual_confirm) { for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -3582,13 +3494,52 @@ static void commit_planes_for_stream_fast(struct dc *dc, context->block_sequence_steps); /* Clear update flags so next flip doesn't have redundant programming * (if there's no stream update, the update flags are not cleared). + * Surface updates are cleared unconditionally at the beginning of each flip, + * so no need to clear here. */ - if (top_pipe_to_program->plane_state) - top_pipe_to_program->plane_state->update_flags.raw = 0; if (top_pipe_to_program->stream) top_pipe_to_program->stream->update_flags.raw = 0; } +static void wait_for_outstanding_hw_updates(struct dc *dc, const struct dc_state *dc_context) +{ +/* + * This function calls HWSS to wait for any potentially double buffered + * operations to complete. It should be invoked as a pre-amble prior + * to full update programming before asserting any HW locks. + */ + int pipe_idx; + int opp_inst; + int opp_count = dc->res_pool->pipe_count; + struct hubp *hubp; + int mpcc_inst; + const struct pipe_ctx *pipe_ctx; + + for (pipe_idx = 0; pipe_idx < dc->res_pool->pipe_count; pipe_idx++) { + pipe_ctx = &dc_context->res_ctx.pipe_ctx[pipe_idx]; + + if (!pipe_ctx->stream) + continue; + + if (pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear) + pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear(pipe_ctx->stream_res.tg); + + hubp = pipe_ctx->plane_res.hubp; + if (!hubp) + continue; + + mpcc_inst = hubp->inst; + // MPCC inst is equal to pipe index in practice + for (opp_inst = 0; opp_inst < opp_count; opp_inst++) { + if (dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst]) { + dc->res_pool->mpc->funcs->wait_for_idle(dc->res_pool->mpc, mpcc_inst); + dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst] = false; + break; + } + } + } +} + static void commit_planes_for_stream(struct dc *dc, struct dc_surface_update *srf_updates, int surface_count, @@ -3607,24 +3558,9 @@ static void commit_planes_for_stream(struct dc *dc, // dc->current_state anymore, so we have to cache it before we apply // the new SubVP context subvp_prev_use = false; - - dc_z10_restore(dc); - - if (update_type == UPDATE_TYPE_FULL) { - /* wait for all double-buffer activity to clear on all pipes */ - int pipe_idx; - - for (pipe_idx = 0; pipe_idx < dc->res_pool->pipe_count; pipe_idx++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[pipe_idx]; - - if (!pipe_ctx->stream) - continue; - - if (pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear) - pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear(pipe_ctx->stream_res.tg); - } - } + if (update_type == UPDATE_TYPE_FULL) + wait_for_outstanding_hw_updates(dc, context); if (update_type == UPDATE_TYPE_FULL) { dc_allow_idle_optimizations(dc, false); @@ -3638,16 +3574,9 @@ static void commit_planes_for_stream(struct dc *dc, context_clock_trace(dc, context); } - for (j = 0; j < dc->res_pool->pipe_count; j++) { - struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; - - if (!pipe_ctx->top_pipe && - !pipe_ctx->prev_odm_pipe && - pipe_ctx->stream && - pipe_ctx->stream == stream) { - top_pipe_to_program = pipe_ctx; - } - } + top_pipe_to_program = resource_get_otg_master_for_stream( + &context->res_ctx, + stream); for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -4088,9 +4017,9 @@ static bool commit_minimal_transition_state(struct dc *dc, struct dc_state *transition_base_context) { struct dc_state *transition_context = dc_create_state(dc); - enum pipe_split_policy tmp_mpc_policy; - bool temp_dynamic_odm_policy; - bool temp_subvp_policy; + enum pipe_split_policy tmp_mpc_policy = 0; + bool temp_dynamic_odm_policy = 0; + bool temp_subvp_policy = 0; enum dc_status ret = DC_ERROR_UNEXPECTED; unsigned int i, j; unsigned int pipe_in_use = 0; @@ -4284,7 +4213,8 @@ static bool fast_updates_exist(struct dc_fast_update *fast_update, int surface_c return false; } -static bool full_update_required(struct dc_surface_update *srf_updates, +static bool full_update_required(struct dc *dc, + struct dc_surface_update *srf_updates, int surface_count, struct dc_stream_update *stream_update, struct dc_stream_state *stream) @@ -4292,6 +4222,7 @@ static bool full_update_required(struct dc_surface_update *srf_updates, int i; struct dc_stream_status *stream_status; + const struct dc_state *context = dc->current_state; for (i = 0; i < surface_count; i++) { if (srf_updates && @@ -4302,7 +4233,11 @@ static bool full_update_required(struct dc_surface_update *srf_updates, srf_updates[i].in_transfer_func || srf_updates[i].func_shaper || srf_updates[i].lut3d_func || - srf_updates[i].blend_tf)) + srf_updates[i].blend_tf || + srf_updates[i].surface->force_full_update || + (srf_updates[i].flip_addr && + srf_updates[i].flip_addr->address.tmz_surface != srf_updates[i].surface->address.tmz_surface) || + !is_surface_in_context(context, srf_updates[i].surface))) return true; } @@ -4340,18 +4275,21 @@ static bool full_update_required(struct dc_surface_update *srf_updates, if (stream_status == NULL || stream_status->plane_count != surface_count) return true; } + if (dc->idle_optimizations_allowed) + return true; return false; } -static bool fast_update_only(struct dc_fast_update *fast_update, +static bool fast_update_only(struct dc *dc, + struct dc_fast_update *fast_update, struct dc_surface_update *srf_updates, int surface_count, struct dc_stream_update *stream_update, struct dc_stream_state *stream) { return fast_updates_exist(fast_update, surface_count) - && !full_update_required(srf_updates, surface_count, stream_update, stream); + && !full_update_required(dc, srf_updates, surface_count, stream_update, stream); } bool dc_update_planes_and_stream(struct dc *dc, @@ -4369,8 +4307,8 @@ bool dc_update_planes_and_stream(struct dc *dc, * cause underflow. Apply stream configuration with minimal pipe * split first to avoid unsupported transitions for active pipes. */ - bool force_minimal_pipe_splitting; - bool is_plane_addition; + bool force_minimal_pipe_splitting = 0; + bool is_plane_addition = 0; populate_fast_updates(fast_update, srf_updates, surface_count, stream_update); force_minimal_pipe_splitting = could_mpcc_tree_change_for_active_pipes( @@ -4423,7 +4361,7 @@ bool dc_update_planes_and_stream(struct dc *dc, } update_seamless_boot_flags(dc, context, surface_count, stream); - if (fast_update_only(fast_update, srf_updates, surface_count, stream_update, stream) && + if (fast_update_only(dc, fast_update, srf_updates, surface_count, stream_update, stream) && !dc->debug.enable_legacy_fast_update) { commit_planes_for_stream_fast(dc, srf_updates, @@ -4569,7 +4507,7 @@ void dc_commit_updates_for_stream(struct dc *dc, TRACE_DC_PIPE_STATE(pipe_ctx, i, MAX_PIPES); update_seamless_boot_flags(dc, context, surface_count, stream); - if (fast_update_only(fast_update, srf_updates, surface_count, stream_update, stream) && + if (fast_update_only(dc, fast_update, srf_updates, surface_count, stream_update, stream) && !dc->debug.enable_legacy_fast_update) { commit_planes_for_stream_fast(dc, srf_updates, @@ -5245,6 +5183,9 @@ void dc_notify_vsync_int_state(struct dc *dc, struct dc_stream_state *stream, bo if (link->psr_settings.psr_feature_enabled) return; + if (link->replay_settings.replay_feature_enabled) + return; + /*find primary pipe associated with stream*/ for (i = 0; i < MAX_PIPES; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -5273,3 +5214,70 @@ void dc_notify_vsync_int_state(struct dc *dc, struct dc_stream_state *stream, bo if (pipe->stream_res.abm && pipe->stream_res.abm->funcs->set_abm_pause) pipe->stream_res.abm->funcs->set_abm_pause(pipe->stream_res.abm, !enable, i, pipe->stream_res.tg->inst); } + +/***************************************************************************** + * dc_abm_save_restore() - Interface to DC for save+pause and restore+un-pause + * ABM + * @dc: dc structure + * @stream: stream where vsync int state changed + * @pData: abm hw states + * + ****************************************************************************/ +bool dc_abm_save_restore( + struct dc *dc, + struct dc_stream_state *stream, + struct abm_save_restore *pData) +{ + int i; + int edp_num; + struct pipe_ctx *pipe = NULL; + struct dc_link *link = stream->sink->link; + struct dc_link *edp_links[MAX_NUM_EDP]; + + + /*find primary pipe associated with stream*/ + for (i = 0; i < MAX_PIPES; i++) { + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; + + if (pipe->stream == stream && pipe->stream_res.tg) + break; + } + + if (i == MAX_PIPES) { + ASSERT(0); + return false; + } + + dc_get_edp_links(dc, edp_links, &edp_num); + + /* Determine panel inst */ + for (i = 0; i < edp_num; i++) + if (edp_links[i] == link) + break; + + if (i == edp_num) + return false; + + if (pipe->stream_res.abm && + pipe->stream_res.abm->funcs->save_restore) + return pipe->stream_res.abm->funcs->save_restore( + pipe->stream_res.abm, + i, + pData); + return false; +} + +void dc_query_current_properties(struct dc *dc, struct dc_current_properties *properties) +{ + unsigned int i; + bool subvp_in_use = false; + + for (i = 0; i < dc->current_state->stream_count; i++) { + if (dc->current_state->streams[i]->mall_stream_config.type != SUBVP_NONE) { + subvp_in_use = true; + break; + } + } + properties->cursor_size_limit = subvp_in_use ? 64 : dc->caps.max_cursor_size; +} + diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c index cb2bf9a466f5..f99ec1b0efaf 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c @@ -187,6 +187,7 @@ static bool is_ycbcr709_limited_type( ret = true; return ret; } + static enum dc_color_space_type get_color_space_type(enum dc_color_space color_space) { enum dc_color_space_type type = COLOR_SPACE_RGB_TYPE; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c index 30c0644d4418..be5a6d008b29 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c @@ -169,11 +169,23 @@ static void add_link_enc_assignment( /* Return first available DIG link encoder. */ static enum engine_id find_first_avail_link_enc( const struct dc_context *ctx, - const struct dc_state *state) + const struct dc_state *state, + enum engine_id eng_id_requested) { enum engine_id eng_id = ENGINE_ID_UNKNOWN; int i; + if (eng_id_requested != ENGINE_ID_UNKNOWN) { + + for (i = 0; i < ctx->dc->res_pool->res_cap->num_dig_link_enc; i++) { + eng_id = state->res_ctx.link_enc_cfg_ctx.link_enc_avail[i]; + if (eng_id == eng_id_requested) + return eng_id; + } + } + + eng_id = ENGINE_ID_UNKNOWN; + for (i = 0; i < ctx->dc->res_pool->res_cap->num_dig_link_enc; i++) { eng_id = state->res_ctx.link_enc_cfg_ctx.link_enc_avail[i]; if (eng_id != ENGINE_ID_UNKNOWN) @@ -287,7 +299,7 @@ void link_enc_cfg_link_encs_assign( struct dc_stream_state *streams[], uint8_t stream_count) { - enum engine_id eng_id = ENGINE_ID_UNKNOWN; + enum engine_id eng_id = ENGINE_ID_UNKNOWN, eng_id_req = ENGINE_ID_UNKNOWN; int i; int j; @@ -377,8 +389,14 @@ void link_enc_cfg_link_encs_assign( * assigned to that endpoint. */ link_enc = get_link_enc_used_by_link(state, stream->link); - if (link_enc == NULL) - eng_id = find_first_avail_link_enc(stream->ctx, state); + if (link_enc == NULL) { + + if (stream->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && + stream->link->dpia_preferred_eng_id != ENGINE_ID_UNKNOWN) + eng_id_req = stream->link->dpia_preferred_eng_id; + + eng_id = find_first_avail_link_enc(stream->ctx, state, eng_id_req); + } else eng_id = link_enc->preferred_engine; @@ -402,7 +420,9 @@ void link_enc_cfg_link_encs_assign( DC_LOG_DEBUG("%s: CUR %s(%d) - enc_id(%d)\n", __func__, assignment.ep_id.ep_type == DISPLAY_ENDPOINT_PHY ? "PHY" : "DPIA", - assignment.ep_id.link_id.enum_id - 1, + assignment.ep_id.ep_type == DISPLAY_ENDPOINT_PHY ? + assignment.ep_id.link_id.enum_id : + assignment.ep_id.link_id.enum_id - 1, assignment.eng_id); } for (i = 0; i < MAX_PIPES; i++) { @@ -413,7 +433,9 @@ void link_enc_cfg_link_encs_assign( DC_LOG_DEBUG("%s: NEW %s(%d) - enc_id(%d)\n", __func__, assignment.ep_id.ep_type == DISPLAY_ENDPOINT_PHY ? "PHY" : "DPIA", - assignment.ep_id.link_id.enum_id - 1, + assignment.ep_id.ep_type == DISPLAY_ENDPOINT_PHY ? + assignment.ep_id.link_id.enum_id : + assignment.ep_id.link_id.enum_id - 1, assignment.eng_id); } @@ -478,7 +500,6 @@ struct dc_link *link_enc_cfg_get_link_using_link_enc( if (stream) link = stream->link; - // dm_output_to_console("%s: No link using DIG(%d).\n", __func__, eng_id); return link; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c index 18e098568cb4..ed94187c2afa 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c @@ -314,6 +314,24 @@ const struct dc_link_settings *dc_link_get_link_cap(const struct dc_link *link) return link->dc->link_srv->dp_get_verified_link_cap(link); } +enum dc_link_encoding_format dc_link_get_highest_encoding_format(const struct dc_link *link) +{ + if (dc_is_dp_signal(link->connector_signal)) { + if (link->dpcd_caps.dongle_type >= DISPLAY_DONGLE_DP_DVI_DONGLE && + link->dpcd_caps.dongle_type <= DISPLAY_DONGLE_DP_HDMI_MISMATCHED_DONGLE) + return DC_LINK_ENCODING_HDMI_TMDS; + else if (link->dc->link_srv->dp_get_encoding_format(&link->verified_link_cap) == + DP_8b_10b_ENCODING) + return DC_LINK_ENCODING_DP_8b_10b; + else if (link->dc->link_srv->dp_get_encoding_format(&link->verified_link_cap) == + DP_128b_132b_ENCODING) + return DC_LINK_ENCODING_DP_128b_132b; + } else if (dc_is_hdmi_signal(link->connector_signal)) { + } + + return DC_LINK_ENCODING_UNSPECIFIED; +} + bool dc_link_is_dp_sink_present(struct dc_link *link) { return link->dc->link_srv->dp_is_sink_present(link); @@ -449,6 +467,11 @@ bool dc_link_setup_psr(struct dc_link *link, return link->dc->link_srv->edp_setup_psr(link, stream, psr_config, psr_context); } +bool dc_link_get_replay_state(const struct dc_link *link, uint64_t *state) +{ + return link->dc->link_srv->edp_get_replay_state(link, state); +} + bool dc_link_wait_for_t12(struct dc_link *link) { return link->dc->link_srv->edp_wait_for_t12(link); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index 2f3d9a698486..f7b51aca6020 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -45,6 +45,8 @@ #include "link/hwss/link_hwss_dio.h" #include "link/hwss/link_hwss_dpia.h" #include "link/hwss/link_hwss_hpo_dp.h" +#include "link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h" +#include "link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h" #if defined(CONFIG_DRM_AMD_DC_SI) #include "dce60/dce60_resource.h" @@ -69,9 +71,20 @@ #include "../dcn32/dcn32_resource.h" #include "../dcn321/dcn321_resource.h" +#define VISUAL_CONFIRM_BASE_DEFAULT 3 +#define VISUAL_CONFIRM_BASE_MIN 1 +#define VISUAL_CONFIRM_BASE_MAX 10 +/* we choose 240 because it is a common denominator of common v addressable + * such as 2160, 1440, 1200, 960. So we take 1/240 portion of v addressable as + * the visual confirm dpp offset height. So visual confirm height can stay + * relatively the same independent from timing used. + */ +#define VISUAL_CONFIRM_DPP_OFFSET_DENO 240 #define DC_LOGGER_INIT(logger) +#define UNABLE_TO_SPLIT -1 + enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) { enum dce_version dc_version = DCE_VERSION_UNKNOWN; @@ -719,10 +732,10 @@ static inline void get_vp_scan_direction( *flip_horz_scan_dir = !*flip_horz_scan_dir; } -int get_num_mpc_splits(struct pipe_ctx *pipe) +int resource_get_num_mpc_splits(const struct pipe_ctx *pipe) { int mpc_split_count = 0; - struct pipe_ctx *other_pipe = pipe->bottom_pipe; + const struct pipe_ctx *other_pipe = pipe->bottom_pipe; while (other_pipe && other_pipe->plane_state == pipe->plane_state) { mpc_split_count++; @@ -737,48 +750,46 @@ int get_num_mpc_splits(struct pipe_ctx *pipe) return mpc_split_count; } -int get_num_odm_splits(struct pipe_ctx *pipe) +int resource_get_num_odm_splits(const struct pipe_ctx *pipe) { int odm_split_count = 0; - struct pipe_ctx *next_pipe = pipe->next_odm_pipe; - while (next_pipe) { - odm_split_count++; - next_pipe = next_pipe->next_odm_pipe; - } - pipe = pipe->prev_odm_pipe; - while (pipe) { + + pipe = resource_get_otg_master(pipe); + + while (pipe->next_odm_pipe) { odm_split_count++; - pipe = pipe->prev_odm_pipe; + pipe = pipe->next_odm_pipe; } return odm_split_count; } -static void calculate_split_count_and_index(struct pipe_ctx *pipe_ctx, int *split_count, int *split_idx) +static int get_odm_split_index(struct pipe_ctx *pipe_ctx) { - *split_count = get_num_odm_splits(pipe_ctx); - *split_idx = 0; - if (*split_count == 0) { - /*Check for mpc split*/ - struct pipe_ctx *split_pipe = pipe_ctx->top_pipe; - - *split_count = get_num_mpc_splits(pipe_ctx); - while (split_pipe && split_pipe->plane_state == pipe_ctx->plane_state) { - (*split_idx)++; - split_pipe = split_pipe->top_pipe; - } + int index = 0; - /* MPO window on right side of ODM split */ - if (split_pipe && split_pipe->prev_odm_pipe && !pipe_ctx->prev_odm_pipe) - (*split_idx)++; - } else { - /*Get odm split index*/ - struct pipe_ctx *split_pipe = pipe_ctx->prev_odm_pipe; + pipe_ctx = resource_get_opp_head(pipe_ctx); + if (!pipe_ctx) + return 0; - while (split_pipe) { - (*split_idx)++; - split_pipe = split_pipe->prev_odm_pipe; - } + while (pipe_ctx->prev_odm_pipe) { + index++; + pipe_ctx = pipe_ctx->prev_odm_pipe; + } + + return index; +} + +static int get_mpc_split_index(struct pipe_ctx *pipe_ctx) +{ + struct pipe_ctx *split_pipe = pipe_ctx->top_pipe; + int index = 0; + + while (split_pipe && split_pipe->plane_state == pipe_ctx->plane_state) { + index++; + split_pipe = split_pipe->top_pipe; } + + return index; } /* @@ -800,82 +811,366 @@ static void calculate_viewport_size(struct pipe_ctx *pipe_ctx) } } -static void calculate_recout(struct pipe_ctx *pipe_ctx) +static struct rect intersect_rec(const struct rect *r0, const struct rect *r1) { - const struct dc_plane_state *plane_state = pipe_ctx->plane_state; - const struct dc_stream_state *stream = pipe_ctx->stream; - struct scaler_data *data = &pipe_ctx->plane_res.scl_data; - struct rect surf_clip = plane_state->clip_rect; - bool split_tb = stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM; - int split_count, split_idx; + struct rect rec; + int r0_x_end = r0->x + r0->width; + int r1_x_end = r1->x + r1->width; + int r0_y_end = r0->y + r0->height; + int r1_y_end = r1->y + r1->height; + + rec.x = r0->x > r1->x ? r0->x : r1->x; + rec.width = r0_x_end > r1_x_end ? r1_x_end - rec.x : r0_x_end - rec.x; + rec.y = r0->y > r1->y ? r0->y : r1->y; + rec.height = r0_y_end > r1_y_end ? r1_y_end - rec.y : r0_y_end - rec.y; + + /* in case that there is no intersection */ + if (rec.width < 0 || rec.height < 0) + memset(&rec, 0, sizeof(rec)); + + return rec; +} - calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx); - if (stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE) - split_idx = 0; +static struct rect shift_rec(const struct rect *rec_in, int x, int y) +{ + struct rect rec_out = *rec_in; + rec_out.x += x; + rec_out.y += y; + + return rec_out; +} + +static struct rect calculate_odm_slice_in_timing_active(struct pipe_ctx *pipe_ctx) +{ + const struct dc_stream_state *stream = pipe_ctx->stream; + int odm_slice_count = resource_get_num_odm_splits(pipe_ctx) + 1; + int odm_slice_idx = get_odm_split_index(pipe_ctx); + bool is_last_odm_slice = (odm_slice_idx + 1) == odm_slice_count; + int h_active = stream->timing.h_addressable + + stream->timing.h_border_left + + stream->timing.h_border_right; + int odm_slice_width = h_active / odm_slice_count; + struct rect odm_rec; + + odm_rec.x = odm_slice_width * odm_slice_idx; + odm_rec.width = is_last_odm_slice ? + /* last slice width is the reminder of h_active */ + h_active - odm_slice_width * (odm_slice_count - 1) : + /* odm slice width is the floor of h_active / count */ + odm_slice_width; + odm_rec.y = 0; + odm_rec.height = stream->timing.v_addressable + + stream->timing.v_border_bottom + + stream->timing.v_border_top; + + return odm_rec; +} + +static struct rect calculate_plane_rec_in_timing_active( + struct pipe_ctx *pipe_ctx, + const struct rect *rec_in) +{ /* - * Only the leftmost ODM pipe should be offset by a nonzero distance + * The following diagram shows an example where we map a 1920x1200 + * desktop to a 2560x1440 timing with a plane rect in the middle + * of the screen. To map a plane rect from Stream Source to Timing + * Active space, we first multiply stream scaling ratios (i.e 2304/1920 + * horizontal and 1440/1200 vertical) to the plane's x and y, then + * we add stream destination offsets (i.e 128 horizontal, 0 vertical). + * This will give us a plane rect's position in Timing Active. However + * we have to remove the fractional. The rule is that we find left/right + * and top/bottom positions and round the value to the adjacent integer. + * + * Stream Source Space + * ------------ + * __________________________________________________ + * |Stream Source (1920 x 1200) ^ | + * | y | + * | <------- w --------|> | + * | __________________V | + * |<-- x -->|Plane//////////////| ^ | + * | |(pre scale)////////| | | + * | |///////////////////| | | + * | |///////////////////| h | + * | |///////////////////| | | + * | |///////////////////| | | + * | |///////////////////| V | + * | | + * | | + * |__________________________________________________| + * + * + * Timing Active Space + * --------------------------------- + * + * Timing Active (2560 x 1440) + * __________________________________________________ + * |*****| Stteam Destination (2304 x 1440) |*****| + * |*****| |*****| + * |<128>| |*****| + * |*****| __________________ |*****| + * |*****| |Plane/////////////| |*****| + * |*****| |(post scale)//////| |*****| + * |*****| |//////////////////| |*****| + * |*****| |//////////////////| |*****| + * |*****| |//////////////////| |*****| + * |*****| |//////////////////| |*****| + * |*****| |*****| + * |*****| |*****| + * |*****| |*****| + * |*****|______________________________________|*****| + * + * So the resulting formulas are shown below: + * + * recout_x = 128 + round(plane_x * 2304 / 1920) + * recout_w = 128 + round((plane_x + plane_w) * 2304 / 1920) - recout_x + * recout_y = 0 + round(plane_y * 1440 / 1280) + * recout_h = 0 + round((plane_y + plane_h) * 1440 / 1200) - recout_y + * + * NOTE: fixed point division is not error free. To reduce errors + * introduced by fixed point division, we divide only after + * multiplication is complete. */ - if (pipe_ctx->top_pipe && pipe_ctx->top_pipe->prev_odm_pipe && !pipe_ctx->prev_odm_pipe) { - /* MPO window on right side of ODM split */ - data->recout.x = stream->dst.x + (surf_clip.x - stream->src.x - stream->src.width/2) * - stream->dst.width / stream->src.width; - } else if (!pipe_ctx->prev_odm_pipe || split_idx == split_count) { - data->recout.x = stream->dst.x; - if (stream->src.x < surf_clip.x) - data->recout.x += (surf_clip.x - stream->src.x) * stream->dst.width - / stream->src.width; - } else - data->recout.x = 0; - - if (stream->src.x > surf_clip.x) - surf_clip.width -= stream->src.x - surf_clip.x; - data->recout.width = surf_clip.width * stream->dst.width / stream->src.width; - if (data->recout.width + data->recout.x > stream->dst.x + stream->dst.width) - data->recout.width = stream->dst.x + stream->dst.width - data->recout.x; - - data->recout.y = stream->dst.y; - if (stream->src.y < surf_clip.y) - data->recout.y += (surf_clip.y - stream->src.y) * stream->dst.height - / stream->src.height; - else if (stream->src.y > surf_clip.y) - surf_clip.height -= stream->src.y - surf_clip.y; - - data->recout.height = surf_clip.height * stream->dst.height / stream->src.height; - if (data->recout.height + data->recout.y > stream->dst.y + stream->dst.height) - data->recout.height = stream->dst.y + stream->dst.height - data->recout.y; - - /* Handle h & v split */ - if (split_tb) { - ASSERT(data->recout.height % 2 == 0); - data->recout.height /= 2; - } else if (split_count) { - if (!pipe_ctx->next_odm_pipe && !pipe_ctx->prev_odm_pipe) { - /* extra pixels in the division remainder need to go to pipes after - * the extra pixel index minus one(epimo) defined here as: - */ - int epimo = split_count - data->recout.width % (split_count + 1); + const struct dc_stream_state *stream = pipe_ctx->stream; + struct rect rec_out = {0}; + struct fixed31_32 temp; - data->recout.x += (data->recout.width / (split_count + 1)) * split_idx; - if (split_idx > epimo) - data->recout.x += split_idx - epimo - 1; - ASSERT(stream->view_format != VIEW_3D_FORMAT_SIDE_BY_SIDE || data->recout.width % 2 == 0); - data->recout.width = data->recout.width / (split_count + 1) + (split_idx > epimo ? 1 : 0); - } else { - /* odm */ - if (split_idx == split_count) { - /* rightmost pipe is the remainder recout */ - data->recout.width -= data->h_active * split_count - data->recout.x; - - /* ODM combine cases with MPO we can get negative widths */ - if (data->recout.width < 0) - data->recout.width = 0; - - data->recout.x = 0; - } else - data->recout.width = data->h_active - data->recout.x; - } + temp = dc_fixpt_from_fraction(rec_in->x * stream->dst.width, + stream->src.width); + rec_out.x = stream->dst.x + dc_fixpt_round(temp); + + temp = dc_fixpt_from_fraction( + (rec_in->x + rec_in->width) * stream->dst.width, + stream->src.width); + rec_out.width = stream->dst.x + dc_fixpt_round(temp) - rec_out.x; + + temp = dc_fixpt_from_fraction(rec_in->y * stream->dst.height, + stream->src.height); + rec_out.y = stream->dst.y + dc_fixpt_round(temp); + + temp = dc_fixpt_from_fraction( + (rec_in->y + rec_in->height) * stream->dst.height, + stream->src.height); + rec_out.height = stream->dst.y + dc_fixpt_round(temp) - rec_out.y; + + return rec_out; +} + +static struct rect calculate_mpc_slice_in_timing_active( + struct pipe_ctx *pipe_ctx, + struct rect *plane_clip_rec) +{ + const struct dc_stream_state *stream = pipe_ctx->stream; + int mpc_slice_count = resource_get_num_mpc_splits(pipe_ctx) + 1; + int mpc_slice_idx = get_mpc_split_index(pipe_ctx); + int epimo = mpc_slice_count - plane_clip_rec->width % mpc_slice_count - 1; + struct rect mpc_rec; + + mpc_rec.width = plane_clip_rec->width / mpc_slice_count; + mpc_rec.x = plane_clip_rec->x + mpc_rec.width * mpc_slice_idx; + mpc_rec.height = plane_clip_rec->height; + mpc_rec.y = plane_clip_rec->y; + ASSERT(mpc_slice_count == 1 || + stream->view_format != VIEW_3D_FORMAT_SIDE_BY_SIDE || + mpc_rec.width % 2 == 0); + + /* extra pixels in the division remainder need to go to pipes after + * the extra pixel index minus one(epimo) defined here as: + */ + if (mpc_slice_idx > epimo) { + mpc_rec.x += mpc_slice_idx - epimo - 1; + mpc_rec.width += 1; + } + + if (stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM) { + ASSERT(mpc_rec.height % 2 == 0); + mpc_rec.height /= 2; + } + return mpc_rec; +} + +static void adjust_recout_for_visual_confirm(struct rect *recout, + struct pipe_ctx *pipe_ctx) +{ + struct dc *dc = pipe_ctx->stream->ctx->dc; + int dpp_offset, base_offset; + + if (dc->debug.visual_confirm == VISUAL_CONFIRM_DISABLE) + return; + + dpp_offset = pipe_ctx->stream->timing.v_addressable / VISUAL_CONFIRM_DPP_OFFSET_DENO; + dpp_offset *= pipe_ctx->plane_res.dpp->inst; + + if ((dc->debug.visual_confirm_rect_height >= VISUAL_CONFIRM_BASE_MIN) && + dc->debug.visual_confirm_rect_height <= VISUAL_CONFIRM_BASE_MAX) + base_offset = dc->debug.visual_confirm_rect_height; + else + base_offset = VISUAL_CONFIRM_BASE_DEFAULT; + + recout->height -= base_offset; + recout->height -= dpp_offset; +} + +/* + * The function maps a plane clip from Stream Source Space to ODM Slice Space + * and calculates the rec of the overlapping area of MPC slice of the plane + * clip, ODM slice associated with the pipe context and stream destination rec. + */ +static void calculate_recout(struct pipe_ctx *pipe_ctx) +{ + /* + * A plane clip represents the desired plane size and position in Stream + * Source Space. Stream Source is the destination where all planes are + * blended (i.e. positioned, scaled and overlaid). It is a canvas where + * all planes associated with the current stream are drawn together. + * After Stream Source is completed, we will further scale and + * reposition the entire canvas of the stream source to Stream + * Destination in Timing Active Space. This could be due to display + * overscan adjustment where we will need to rescale and reposition all + * the planes so they can fit into a TV with overscan or downscale + * upscale features such as GPU scaling or VSR. + * + * This two step blending is a virtual procedure in software. In + * hardware there is no such thing as Stream Source. all planes are + * blended once in Timing Active Space. Software virtualizes a Stream + * Source space to decouple the math complicity so scaling param + * calculation focuses on one step at a time. + * + * In the following two diagrams, user applied 10% overscan adjustment + * so the Stream Source needs to be scaled down a little before mapping + * to Timing Active Space. As a result the Plane Clip is also scaled + * down by the same ratio, Plane Clip position (i.e. x and y) with + * respect to Stream Source is also scaled down. To map it in Timing + * Active Space additional x and y offsets from Stream Destination are + * added to Plane Clip as well. + * + * Stream Source Space + * ------------ + * __________________________________________________ + * |Stream Source (3840 x 2160) ^ | + * | y | + * | | | + * | __________________V | + * |<-- x -->|Plane Clip/////////| | + * | |(pre scale)////////| | + * | |///////////////////| | + * | |///////////////////| | + * | |///////////////////| | + * | |///////////////////| | + * | |///////////////////| | + * | | + * | | + * |__________________________________________________| + * + * + * Timing Active Space (3840 x 2160) + * --------------------------------- + * + * Timing Active + * __________________________________________________ + * | y_____________________________________________ | + * |x |Stream Destination (3456 x 1944) | | + * | | | | + * | | __________________ | | + * | | |Plane Clip////////| | | + * | | |(post scale)//////| | | + * | | |//////////////////| | | + * | | |//////////////////| | | + * | | |//////////////////| | | + * | | |//////////////////| | | + * | | | | + * | | | | + * | |____________________________________________| | + * |__________________________________________________| + * + * + * In Timing Active Space a plane clip could be further sliced into + * pieces called MPC slices. Each Pipe Context is responsible for + * processing only one MPC slice so the plane processing workload can be + * distributed to multiple DPP Pipes. MPC slices could be blended + * together to a single ODM slice. Each ODM slice is responsible for + * processing a portion of Timing Active divided horizontally so the + * output pixel processing workload can be distributed to multiple OPP + * pipes. All ODM slices are mapped together in ODM block so all MPC + * slices belong to different ODM slices could be pieced together to + * form a single image in Timing Active. MPC slices must belong to + * single ODM slice. If an MPC slice goes across ODM slice boundary, it + * needs to be divided into two MPC slices one for each ODM slice. + * + * In the following diagram the output pixel processing workload is + * divided horizontally into two ODM slices one for each OPP blend tree. + * OPP0 blend tree is responsible for processing left half of Timing + * Active, while OPP2 blend tree is responsible for processing right + * half. + * + * The plane has two MPC slices. However since the right MPC slice goes + * across ODM boundary, two DPP pipes are needed one for each OPP blend + * tree. (i.e. DPP1 for OPP0 blend tree and DPP2 for OPP2 blend tree). + * + * Assuming that we have a Pipe Context associated with OPP0 and DPP1 + * working on processing the plane in the diagram. We want to know the + * width and height of the shaded rectangle and its relative position + * with respect to the ODM slice0. This is called the recout of the pipe + * context. + * + * Planes can be at arbitrary size and position and there could be an + * arbitrary number of MPC and ODM slices. The algorithm needs to take + * all scenarios into account. + * + * Timing Active Space (3840 x 2160) + * --------------------------------- + * + * Timing Active + * __________________________________________________ + * |OPP0(ODM slice0)^ |OPP2(ODM slice1) | + * | y | | + * | | <- w -> | + * | _____V________|____ | + * | |DPP0 ^ |DPP1 |DPP2| | + * |<------ x |-----|->|/////| | | + * | | | |/////| | | + * | | h |/////| | | + * | | | |/////| | | + * | |_____V__|/////|____| | + * | | | + * | | | + * | | | + * |_________________________|________________________| + * + * + */ + struct rect plane_clip; + struct rect mpc_slice_of_plane_clip; + struct rect odm_slice; + struct rect overlapping_area; + + plane_clip = calculate_plane_rec_in_timing_active(pipe_ctx, + &pipe_ctx->plane_state->clip_rect); + /* guard plane clip from drawing beyond stream dst here */ + plane_clip = intersect_rec(&plane_clip, + &pipe_ctx->stream->dst); + mpc_slice_of_plane_clip = calculate_mpc_slice_in_timing_active( + pipe_ctx, &plane_clip); + odm_slice = calculate_odm_slice_in_timing_active(pipe_ctx); + overlapping_area = intersect_rec(&mpc_slice_of_plane_clip, &odm_slice); + if (overlapping_area.height > 0 && + overlapping_area.width > 0) { + /* shift the overlapping area so it is with respect to current + * ODM slice's position + */ + pipe_ctx->plane_res.scl_data.recout = shift_rec( + &overlapping_area, + -odm_slice.x, -odm_slice.y); + adjust_recout_for_visual_confirm( + &pipe_ctx->plane_res.scl_data.recout, + pipe_ctx); + } else { + /* if there is no overlap, zero recout */ + memset(&pipe_ctx->plane_res.scl_data.recout, 0, + sizeof(struct rect)); } + } static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) @@ -997,33 +1292,30 @@ static void calculate_init_and_vp( static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; - const struct dc_stream_state *stream = pipe_ctx->stream; struct scaler_data *data = &pipe_ctx->plane_res.scl_data; struct rect src = plane_state->src_rect; + struct rect recout_dst_in_active_timing; + struct rect recout_clip_in_active_timing; + struct rect recout_clip_in_recout_dst; + struct rect overlap_in_active_timing; + struct rect odm_slice = calculate_odm_slice_in_timing_active(pipe_ctx); int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; - int split_count, split_idx, ro_lb, ro_tb, recout_full_x, recout_full_y; bool orthogonal_rotation, flip_vert_scan_dir, flip_horz_scan_dir; - calculate_split_count_and_index(pipe_ctx, &split_count, &split_idx); - /* - * recout full is what the recout would have been if we didnt clip - * the source plane at all. We only care about left(ro_lb) and top(ro_tb) - * offsets of recout within recout full because those are the directions - * we scan from and therefore the only ones that affect inits. - */ - recout_full_x = stream->dst.x + (plane_state->dst_rect.x - stream->src.x) - * stream->dst.width / stream->src.width; - recout_full_y = stream->dst.y + (plane_state->dst_rect.y - stream->src.y) - * stream->dst.height / stream->src.height; - if (pipe_ctx->prev_odm_pipe && split_idx) - ro_lb = data->h_active * split_idx - recout_full_x; - else if (pipe_ctx->top_pipe && pipe_ctx->top_pipe->prev_odm_pipe) - ro_lb = data->h_active * split_idx - recout_full_x + data->recout.x; + recout_clip_in_active_timing = shift_rec( + &data->recout, odm_slice.x, odm_slice.y); + recout_dst_in_active_timing = calculate_plane_rec_in_timing_active( + pipe_ctx, &plane_state->dst_rect); + overlap_in_active_timing = intersect_rec(&recout_clip_in_active_timing, + &recout_dst_in_active_timing); + if (overlap_in_active_timing.width > 0 && + overlap_in_active_timing.height > 0) + recout_clip_in_recout_dst = shift_rec(&overlap_in_active_timing, + -recout_dst_in_active_timing.x, + -recout_dst_in_active_timing.y); else - ro_lb = data->recout.x - recout_full_x; - ro_tb = data->recout.y - recout_full_y; - ASSERT(ro_lb >= 0 && ro_tb >= 0); + memset(&recout_clip_in_recout_dst, 0, sizeof(struct rect)); /* * Work in recout rotation since that requires less transformations @@ -1042,7 +1334,7 @@ static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx) calculate_init_and_vp( flip_horz_scan_dir, - ro_lb, + recout_clip_in_recout_dst.x, data->recout.width, src.width, data->taps.h_taps, @@ -1052,7 +1344,7 @@ static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx) &data->viewport.width); calculate_init_and_vp( flip_horz_scan_dir, - ro_lb, + recout_clip_in_recout_dst.x, data->recout.width, src.width / vpc_div, data->taps.h_taps_c, @@ -1062,7 +1354,7 @@ static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx) &data->viewport_c.width); calculate_init_and_vp( flip_vert_scan_dir, - ro_tb, + recout_clip_in_recout_dst.y, data->recout.height, src.height, data->taps.v_taps, @@ -1072,7 +1364,7 @@ static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx) &data->viewport.height); calculate_init_and_vp( flip_vert_scan_dir, - ro_tb, + recout_clip_in_recout_dst.y, data->recout.height, src.height / vpc_div, data->taps.v_taps_c, @@ -1097,6 +1389,7 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; + const struct rect odm_slice_rec = calculate_odm_slice_in_timing_active(pipe_ctx); bool res = false; DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); @@ -1121,30 +1414,9 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) pipe_ctx->stream->dst.y += timing->v_border_top; /* Calculate H and V active size */ - pipe_ctx->plane_res.scl_data.h_active = timing->h_addressable + - timing->h_border_left + timing->h_border_right; - pipe_ctx->plane_res.scl_data.v_active = timing->v_addressable + - timing->v_border_top + timing->v_border_bottom; - if (pipe_ctx->next_odm_pipe || pipe_ctx->prev_odm_pipe) { - pipe_ctx->plane_res.scl_data.h_active /= get_num_odm_splits(pipe_ctx) + 1; - - DC_LOG_SCALER("%s pipe %d: next_odm_pipe:%d prev_odm_pipe:%d\n", - __func__, - pipe_ctx->pipe_idx, - pipe_ctx->next_odm_pipe ? pipe_ctx->next_odm_pipe->pipe_idx : -1, - pipe_ctx->prev_odm_pipe ? pipe_ctx->prev_odm_pipe->pipe_idx : -1); - } /* ODM + windows MPO, where window is on either right or left ODM half */ - else if (pipe_ctx->top_pipe && (pipe_ctx->top_pipe->next_odm_pipe || pipe_ctx->top_pipe->prev_odm_pipe)) { - - pipe_ctx->plane_res.scl_data.h_active /= get_num_odm_splits(pipe_ctx->top_pipe) + 1; - - DC_LOG_SCALER("%s ODM + windows MPO: pipe:%d top_pipe:%d top_pipe->next_odm_pipe:%d top_pipe->prev_odm_pipe:%d\n", - __func__, - pipe_ctx->pipe_idx, - pipe_ctx->top_pipe->pipe_idx, - pipe_ctx->top_pipe->next_odm_pipe ? pipe_ctx->top_pipe->next_odm_pipe->pipe_idx : -1, - pipe_ctx->top_pipe->prev_odm_pipe ? pipe_ctx->top_pipe->prev_odm_pipe->pipe_idx : -1); - } + pipe_ctx->plane_res.scl_data.h_active = odm_slice_rec.width; + pipe_ctx->plane_res.scl_data.v_active = odm_slice_rec.height; + /* depends on h_active */ calculate_recout(pipe_ctx); /* depends on pixel format */ @@ -1226,17 +1498,12 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.scl_data.recout.x += pipe_ctx->plane_res.scl_data.recout.width; } - if (!pipe_ctx->stream->ctx->dc->config.enable_windowed_mpo_odm) { - if (pipe_ctx->plane_res.scl_data.viewport.height < MIN_VIEWPORT_SIZE || - pipe_ctx->plane_res.scl_data.viewport.width < MIN_VIEWPORT_SIZE) - res = false; - } else { - /* Clamp minimum viewport size */ - if (pipe_ctx->plane_res.scl_data.viewport.height < MIN_VIEWPORT_SIZE) - pipe_ctx->plane_res.scl_data.viewport.height = MIN_VIEWPORT_SIZE; - if (pipe_ctx->plane_res.scl_data.viewport.width < MIN_VIEWPORT_SIZE) - pipe_ctx->plane_res.scl_data.viewport.width = MIN_VIEWPORT_SIZE; - } + /* Clamp minimum viewport size */ + if (pipe_ctx->plane_res.scl_data.viewport.height < MIN_VIEWPORT_SIZE) + pipe_ctx->plane_res.scl_data.viewport.height = MIN_VIEWPORT_SIZE; + if (pipe_ctx->plane_res.scl_data.viewport.width < MIN_VIEWPORT_SIZE) + pipe_ctx->plane_res.scl_data.viewport.width = MIN_VIEWPORT_SIZE; + DC_LOG_SCALER("%s pipe %d:\nViewport: height:%d width:%d x:%d y:%d Recout: height:%d width:%d x:%d y:%d HACTIVE:%d VACTIVE:%d\n" "src_rect: height:%d width:%d x:%d y:%d dst_rect: height:%d width:%d x:%d y:%d clip_rect: height:%d width:%d x:%d y:%d\n", @@ -1288,7 +1555,7 @@ enum dc_status resource_build_scaling_params_for_context( return DC_OK; } -struct pipe_ctx *find_idle_secondary_pipe( +struct pipe_ctx *resource_find_free_secondary_pipe_legacy( struct resource_context *res_ctx, const struct resource_pool *pool, const struct pipe_ctx *primary_pipe) @@ -1348,73 +1615,182 @@ struct pipe_ctx *find_idle_secondary_pipe( return secondary_pipe; } -struct pipe_ctx *resource_get_head_pipe_for_stream( - struct resource_context *res_ctx, - struct dc_stream_state *stream) +int resource_find_free_pipe_used_in_cur_mpc_blending_tree( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct pipe_ctx *cur_opp_head) { + const struct pipe_ctx *cur_sec_dpp = cur_opp_head->bottom_pipe; + struct pipe_ctx *new_pipe; + int free_pipe_idx = FREE_PIPE_INDEX_NOT_FOUND; + + while (cur_sec_dpp) { + /* find a free pipe used in current opp blend tree, + * this is to avoid MPO pipe switching to different opp blending + * tree + */ + new_pipe = &new_res_ctx->pipe_ctx[cur_sec_dpp->pipe_idx]; + if (resource_is_pipe_type(new_pipe, FREE_PIPE)) { + free_pipe_idx = cur_sec_dpp->pipe_idx; + break; + } + cur_sec_dpp = cur_sec_dpp->bottom_pipe; + } + + return free_pipe_idx; +} + +int recource_find_free_pipe_not_used_in_cur_res_ctx( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct resource_pool *pool) +{ + int free_pipe_idx = FREE_PIPE_INDEX_NOT_FOUND; + const struct pipe_ctx *new_pipe, *cur_pipe; int i; - for (i = 0; i < MAX_PIPES; i++) { - if (res_ctx->pipe_ctx[i].stream == stream - && !res_ctx->pipe_ctx[i].top_pipe - && !res_ctx->pipe_ctx[i].prev_odm_pipe) - return &res_ctx->pipe_ctx[i]; + for (i = 0; i < pool->pipe_count; i++) { + cur_pipe = &cur_res_ctx->pipe_ctx[i]; + new_pipe = &new_res_ctx->pipe_ctx[i]; + + if (resource_is_pipe_type(cur_pipe, FREE_PIPE) && + resource_is_pipe_type(new_pipe, FREE_PIPE)) { + free_pipe_idx = i; + break; + } } - return NULL; + + return free_pipe_idx; } -static struct pipe_ctx *resource_get_tail_pipe( - struct resource_context *res_ctx, - struct pipe_ctx *head_pipe) +int resource_find_free_pipe_used_as_cur_sec_dpp_in_mpcc_combine( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct resource_pool *pool) { - struct pipe_ctx *tail_pipe; - - tail_pipe = head_pipe->bottom_pipe; + int free_pipe_idx = FREE_PIPE_INDEX_NOT_FOUND; + const struct pipe_ctx *new_pipe, *cur_pipe; + int i; - while (tail_pipe) { - head_pipe = tail_pipe; - tail_pipe = tail_pipe->bottom_pipe; + for (i = 0; i < pool->pipe_count; i++) { + cur_pipe = &cur_res_ctx->pipe_ctx[i]; + new_pipe = &new_res_ctx->pipe_ctx[i]; + + if (resource_is_pipe_type(cur_pipe, DPP_PIPE) && + !resource_is_pipe_type(cur_pipe, OPP_HEAD) && + resource_is_for_mpcc_combine(cur_pipe) && + resource_is_pipe_type(new_pipe, FREE_PIPE)) { + free_pipe_idx = i; + break; + } } - return head_pipe; + return free_pipe_idx; } -/* - * A free_pipe for a stream is defined here as a pipe - * that has no surface attached yet - */ -static struct pipe_ctx *acquire_free_pipe_for_head( - struct dc_state *context, - const struct resource_pool *pool, - struct pipe_ctx *head_pipe) +int resource_find_any_free_pipe(struct resource_context *new_res_ctx, + const struct resource_pool *pool) { + int free_pipe_idx = FREE_PIPE_INDEX_NOT_FOUND; + const struct pipe_ctx *new_pipe; int i; - struct resource_context *res_ctx = &context->res_ctx; - if (!head_pipe->plane_state) - return head_pipe; + for (i = 0; i < pool->pipe_count; i++) { + new_pipe = &new_res_ctx->pipe_ctx[i]; - /* Re-use pipe already acquired for this stream if available*/ - for (i = pool->pipe_count - 1; i >= 0; i--) { - if (res_ctx->pipe_ctx[i].stream == head_pipe->stream && - !res_ctx->pipe_ctx[i].plane_state) { - return &res_ctx->pipe_ctx[i]; + if (resource_is_pipe_type(new_pipe, FREE_PIPE)) { + free_pipe_idx = i; + break; } } - /* - * At this point we have no re-useable pipe for this stream and we need - * to acquire an idle one to satisfy the request + return free_pipe_idx; +} + +bool resource_is_pipe_type(const struct pipe_ctx *pipe_ctx, enum pipe_type type) +{ +#ifdef DBG + if (pipe_ctx->stream == NULL) { + /* a free pipe with dangling states */ + ASSERT(!pipe_ctx->plane_state); + ASSERT(!pipe_ctx->prev_odm_pipe); + ASSERT(!pipe_ctx->next_odm_pipe); + ASSERT(!pipe_ctx->top_pipe); + ASSERT(!pipe_ctx->bottom_pipe); + } else if (pipe_ctx->top_pipe) { + /* a secondary DPP pipe must be signed to a plane */ + ASSERT(pipe_ctx->plane_state) + } + /* Add more checks here to prevent corrupted pipe ctx. It is very hard + * to debug this issue afterwards because we can't pinpoint the code + * location causing inconsistent pipe context states. */ +#endif + switch (type) { + case OTG_MASTER: + return !pipe_ctx->prev_odm_pipe && + !pipe_ctx->top_pipe && + pipe_ctx->stream; + case OPP_HEAD: + return !pipe_ctx->top_pipe && pipe_ctx->stream; + case DPP_PIPE: + return pipe_ctx->plane_state && pipe_ctx->stream; + case FREE_PIPE: + return !pipe_ctx->plane_state && !pipe_ctx->stream; + default: + return false; + } +} - if (!pool->funcs->acquire_idle_pipe_for_layer) { - if (!pool->funcs->acquire_idle_pipe_for_head_pipe_in_layer) - return NULL; - else - return pool->funcs->acquire_idle_pipe_for_head_pipe_in_layer(context, pool, head_pipe->stream, head_pipe); +bool resource_is_for_mpcc_combine(const struct pipe_ctx *pipe_ctx) +{ + return resource_get_num_mpc_splits(pipe_ctx) > 0; +} + +struct pipe_ctx *resource_get_otg_master_for_stream( + struct resource_context *res_ctx, + struct dc_stream_state *stream) +{ + int i; + + for (i = 0; i < MAX_PIPES; i++) { + if (res_ctx->pipe_ctx[i].stream == stream && + resource_is_pipe_type(&res_ctx->pipe_ctx[i], OTG_MASTER)) + return &res_ctx->pipe_ctx[i]; + } + return NULL; +} + +struct pipe_ctx *resource_get_otg_master(const struct pipe_ctx *pipe_ctx) +{ + struct pipe_ctx *otg_master = resource_get_opp_head(pipe_ctx); + + while (otg_master->prev_odm_pipe) + otg_master = otg_master->prev_odm_pipe; + return otg_master; +} + +struct pipe_ctx *resource_get_opp_head(const struct pipe_ctx *pipe_ctx) +{ + struct pipe_ctx *opp_head = (struct pipe_ctx *) pipe_ctx; + + ASSERT(!resource_is_pipe_type(opp_head, FREE_PIPE)); + while (opp_head->top_pipe) + opp_head = opp_head->top_pipe; + return opp_head; +} + +static struct pipe_ctx *get_tail_pipe( + struct pipe_ctx *head_pipe) +{ + struct pipe_ctx *tail_pipe = head_pipe->bottom_pipe; + + while (tail_pipe) { + head_pipe = tail_pipe; + tail_pipe = tail_pipe->bottom_pipe; } - return pool->funcs->acquire_idle_pipe_for_layer(context, pool, head_pipe->stream); + return head_pipe; } static int acquire_first_split_pipe( @@ -1447,275 +1823,126 @@ static int acquire_first_split_pipe( split_pipe->stream = stream; return i; - } else if (split_pipe->prev_odm_pipe && - split_pipe->prev_odm_pipe->plane_state == split_pipe->plane_state) { - split_pipe->prev_odm_pipe->next_odm_pipe = split_pipe->next_odm_pipe; - if (split_pipe->next_odm_pipe) - split_pipe->next_odm_pipe->prev_odm_pipe = split_pipe->prev_odm_pipe; - - if (split_pipe->prev_odm_pipe->plane_state) - resource_build_scaling_params(split_pipe->prev_odm_pipe); - - memset(split_pipe, 0, sizeof(*split_pipe)); - split_pipe->stream_res.tg = pool->timing_generators[i]; - split_pipe->plane_res.hubp = pool->hubps[i]; - split_pipe->plane_res.ipp = pool->ipps[i]; - split_pipe->plane_res.dpp = pool->dpps[i]; - split_pipe->stream_res.opp = pool->opps[i]; - split_pipe->plane_res.mpcc_inst = pool->dpps[i]->inst; - split_pipe->pipe_idx = i; - - split_pipe->stream = stream; - return i; } } - return -1; + return UNABLE_TO_SPLIT; } -bool dc_add_plane_to_context( - const struct dc *dc, - struct dc_stream_state *stream, +static bool add_plane_to_opp_head_pipes(struct pipe_ctx *otg_master_pipe, struct dc_plane_state *plane_state, struct dc_state *context) { - int i; - struct resource_pool *pool = dc->res_pool; - struct pipe_ctx *head_pipe, *tail_pipe, *free_pipe; - struct dc_stream_status *stream_status = NULL; - struct pipe_ctx *prev_right_head = NULL; - struct pipe_ctx *free_right_pipe = NULL; - struct pipe_ctx *prev_left_head = NULL; + struct pipe_ctx *opp_head_pipe = otg_master_pipe; - DC_LOGGER_INIT(stream->ctx->logger); - for (i = 0; i < context->stream_count; i++) - if (context->streams[i] == stream) { - stream_status = &context->stream_status[i]; - break; + while (opp_head_pipe) { + if (opp_head_pipe->plane_state) { + ASSERT(0); + return false; } - if (stream_status == NULL) { - dm_error("Existing stream not found; failed to attach surface!\n"); - return false; + opp_head_pipe->plane_state = plane_state; + opp_head_pipe = opp_head_pipe->next_odm_pipe; } + return true; +} - if (stream_status->plane_count == MAX_SURFACE_NUM) { - dm_error("Surface: can not attach plane_state %p! Maximum is: %d\n", - plane_state, MAX_SURFACE_NUM); - return false; +static void insert_secondary_dpp_pipe_with_plane(struct pipe_ctx *opp_head_pipe, + struct pipe_ctx *sec_pipe, struct dc_plane_state *plane_state) +{ + struct pipe_ctx *tail_pipe = get_tail_pipe(opp_head_pipe); + + tail_pipe->bottom_pipe = sec_pipe; + sec_pipe->top_pipe = tail_pipe; + if (tail_pipe->prev_odm_pipe) { + ASSERT(tail_pipe->prev_odm_pipe->bottom_pipe); + sec_pipe->prev_odm_pipe = tail_pipe->prev_odm_pipe->bottom_pipe; + tail_pipe->prev_odm_pipe->bottom_pipe->next_odm_pipe = sec_pipe; } + sec_pipe->plane_state = plane_state; +} - head_pipe = resource_get_head_pipe_for_stream(&context->res_ctx, stream); +/* for each opp head pipe of an otg master pipe, acquire a secondary dpp pipe + * and add the plane. So the plane is added to all MPC blend trees associated + * with the otg master pipe. + */ +static bool acquire_secondary_dpp_pipes_and_add_plane( + struct pipe_ctx *otg_master_pipe, + struct dc_plane_state *plane_state, + struct dc_state *new_ctx, + struct dc_state *cur_ctx, + struct resource_pool *pool) +{ + struct pipe_ctx *opp_head_pipe, *sec_pipe; - if (!head_pipe) { - dm_error("Head pipe not found for stream_state %p !\n", stream); + if (!pool->funcs->acquire_free_pipe_as_secondary_dpp_pipe) return false; - } - /* retain new surface, but only once per stream */ - dc_plane_state_retain(plane_state); - - while (head_pipe) { - free_pipe = acquire_free_pipe_for_head(context, pool, head_pipe); + opp_head_pipe = otg_master_pipe; + while (opp_head_pipe) { + sec_pipe = pool->funcs->acquire_free_pipe_as_secondary_dpp_pipe( + cur_ctx, + new_ctx, + pool, + opp_head_pipe); + if (!sec_pipe) { + /* try tearing down MPCC combine */ + int pipe_idx = acquire_first_split_pipe( + &new_ctx->res_ctx, pool, + otg_master_pipe->stream); - if (!free_pipe) { - int pipe_idx = acquire_first_split_pipe(&context->res_ctx, pool, stream); if (pipe_idx >= 0) - free_pipe = &context->res_ctx.pipe_ctx[pipe_idx]; + sec_pipe = &new_ctx->res_ctx.pipe_ctx[pipe_idx]; } - if (!free_pipe) { - dc_plane_state_release(plane_state); + if (!sec_pipe) return false; - } - - free_pipe->plane_state = plane_state; - - if (head_pipe != free_pipe) { - tail_pipe = resource_get_tail_pipe(&context->res_ctx, head_pipe); - ASSERT(tail_pipe); - - /* ODM + window MPO, where MPO window is on right half only */ - if (free_pipe->plane_state && - (free_pipe->plane_state->clip_rect.x >= free_pipe->stream->src.x + free_pipe->stream->src.width/2) && - tail_pipe->next_odm_pipe) { - - /* For ODM + window MPO, in 3 plane case, if we already have a MPO window on - * the right side, then we will invalidate a 2nd one on the right side - */ - if (head_pipe->next_odm_pipe && tail_pipe->next_odm_pipe->bottom_pipe) { - dc_plane_state_release(plane_state); - return false; - } - - DC_LOG_SCALER("%s - ODM + window MPO(right). free_pipe:%d tail_pipe->next_odm_pipe:%d\n", - __func__, - free_pipe->pipe_idx, - tail_pipe->next_odm_pipe ? tail_pipe->next_odm_pipe->pipe_idx : -1); - - /* - * We want to avoid the case where the right side already has a pipe assigned to - * it and is different from free_pipe ( which would cause trigger a pipe - * reallocation ). - * Check the old context to see if the right side already has a pipe allocated - * - If not, continue to use free_pipe - * - If the right side already has a pipe, use that pipe instead if its available - */ - - /* - * We also want to avoid the case where with three plane ( 2 MPO videos ), we have - * both videos on the left side so one of the videos is invalidated. Then we - * move the invalidated video back to the right side. If the order of the plane - * states is such that the right MPO plane is processed first, the free pipe - * selected by the head will be the left MPO pipe. But since there was no right - * MPO pipe, it will assign the free pipe to the right MPO pipe instead and - * a pipe reallocation will occur. - * Check the old context to see if the left side already has a pipe allocated - * - If not, continue to use free_pipe - * - If the left side is already using this pipe, then pick another pipe for right - */ - - prev_right_head = &dc->current_state->res_ctx.pipe_ctx[tail_pipe->next_odm_pipe->pipe_idx]; - if ((prev_right_head->bottom_pipe) && - (free_pipe->pipe_idx != prev_right_head->bottom_pipe->pipe_idx)) { - free_right_pipe = acquire_free_pipe_for_head(context, pool, tail_pipe->next_odm_pipe); - } else { - prev_left_head = &dc->current_state->res_ctx.pipe_ctx[head_pipe->pipe_idx]; - if ((prev_left_head->bottom_pipe) && - (free_pipe->pipe_idx == prev_left_head->bottom_pipe->pipe_idx)) { - free_right_pipe = acquire_free_pipe_for_head(context, pool, head_pipe); - } - } - - if (free_right_pipe) { - free_pipe->stream = NULL; - memset(&free_pipe->stream_res, 0, sizeof(struct stream_resource)); - memset(&free_pipe->plane_res, 0, sizeof(struct plane_resource)); - free_pipe->plane_state = NULL; - free_pipe->pipe_idx = 0; - free_right_pipe->plane_state = plane_state; - free_pipe = free_right_pipe; - } - - free_pipe->stream_res.tg = tail_pipe->next_odm_pipe->stream_res.tg; - free_pipe->stream_res.abm = tail_pipe->next_odm_pipe->stream_res.abm; - free_pipe->stream_res.opp = tail_pipe->next_odm_pipe->stream_res.opp; - free_pipe->stream_res.stream_enc = tail_pipe->next_odm_pipe->stream_res.stream_enc; - free_pipe->stream_res.audio = tail_pipe->next_odm_pipe->stream_res.audio; - free_pipe->clock_source = tail_pipe->next_odm_pipe->clock_source; - - free_pipe->top_pipe = tail_pipe->next_odm_pipe; - tail_pipe->next_odm_pipe->bottom_pipe = free_pipe; - } else if (free_pipe->plane_state && - (free_pipe->plane_state->clip_rect.x >= free_pipe->stream->src.x + free_pipe->stream->src.width/2) - && head_pipe->next_odm_pipe) { - - /* For ODM + window MPO, support 3 plane ( 2 MPO ) case. - * Here we have a desktop ODM + left window MPO and a new MPO window appears - * on the right side only. It fails the first case, because tail_pipe is the - * left window MPO, so it has no next_odm_pipe. So in this scenario, we check - * for head_pipe->next_odm_pipe instead - */ - DC_LOG_SCALER("%s - ODM + win MPO (left) + win MPO (right). free_pipe:%d head_pipe->next_odm:%d\n", - __func__, - free_pipe->pipe_idx, - head_pipe->next_odm_pipe ? head_pipe->next_odm_pipe->pipe_idx : -1); - - /* - * We want to avoid the case where the right side already has a pipe assigned to - * it and is different from free_pipe ( which would cause trigger a pipe - * reallocation ). - * Check the old context to see if the right side already has a pipe allocated - * - If not, continue to use free_pipe - * - If the right side already has a pipe, use that pipe instead if its available - */ - prev_right_head = &dc->current_state->res_ctx.pipe_ctx[head_pipe->next_odm_pipe->pipe_idx]; - if ((prev_right_head->bottom_pipe) && - (free_pipe->pipe_idx != prev_right_head->bottom_pipe->pipe_idx)) { - free_right_pipe = acquire_free_pipe_for_head(context, pool, head_pipe->next_odm_pipe); - if (free_right_pipe) { - free_pipe->stream = NULL; - memset(&free_pipe->stream_res, 0, sizeof(struct stream_resource)); - memset(&free_pipe->plane_res, 0, sizeof(struct plane_resource)); - free_pipe->plane_state = NULL; - free_pipe->pipe_idx = 0; - free_right_pipe->plane_state = plane_state; - free_pipe = free_right_pipe; - } - } - free_pipe->stream_res.tg = head_pipe->next_odm_pipe->stream_res.tg; - free_pipe->stream_res.abm = head_pipe->next_odm_pipe->stream_res.abm; - free_pipe->stream_res.opp = head_pipe->next_odm_pipe->stream_res.opp; - free_pipe->stream_res.stream_enc = head_pipe->next_odm_pipe->stream_res.stream_enc; - free_pipe->stream_res.audio = head_pipe->next_odm_pipe->stream_res.audio; - free_pipe->clock_source = head_pipe->next_odm_pipe->clock_source; - - free_pipe->top_pipe = head_pipe->next_odm_pipe; - head_pipe->next_odm_pipe->bottom_pipe = free_pipe; - } else { - - /* For ODM + window MPO, in 3 plane case, if we already have a MPO window on - * the left side, then we will invalidate a 2nd one on the left side - */ - if (head_pipe->next_odm_pipe && tail_pipe->top_pipe) { - dc_plane_state_release(plane_state); - return false; - } - - free_pipe->stream_res.tg = tail_pipe->stream_res.tg; - free_pipe->stream_res.abm = tail_pipe->stream_res.abm; - free_pipe->stream_res.opp = tail_pipe->stream_res.opp; - free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; - free_pipe->stream_res.audio = tail_pipe->stream_res.audio; - free_pipe->clock_source = tail_pipe->clock_source; - - free_pipe->top_pipe = tail_pipe; - tail_pipe->bottom_pipe = free_pipe; - - /* Connect MPO pipes together if MPO window is in the centre */ - if (!(free_pipe->plane_state && - (free_pipe->plane_state->clip_rect.x + free_pipe->plane_state->clip_rect.width <= - free_pipe->stream->src.x + free_pipe->stream->src.width/2))) { - if (!free_pipe->next_odm_pipe && - tail_pipe->next_odm_pipe && tail_pipe->next_odm_pipe->bottom_pipe) { - free_pipe->next_odm_pipe = tail_pipe->next_odm_pipe->bottom_pipe; - tail_pipe->next_odm_pipe->bottom_pipe->prev_odm_pipe = free_pipe; - } - if (!free_pipe->prev_odm_pipe && - tail_pipe->prev_odm_pipe && tail_pipe->prev_odm_pipe->bottom_pipe) { - free_pipe->prev_odm_pipe = tail_pipe->prev_odm_pipe->bottom_pipe; - tail_pipe->prev_odm_pipe->bottom_pipe->next_odm_pipe = free_pipe; - } - } - } - } + insert_secondary_dpp_pipe_with_plane(opp_head_pipe, sec_pipe, + plane_state); + opp_head_pipe = opp_head_pipe->next_odm_pipe; + } + return true; +} - /* ODM + window MPO, where MPO window is on left half only */ - if (free_pipe->plane_state && - (free_pipe->plane_state->clip_rect.x + free_pipe->plane_state->clip_rect.width <= - free_pipe->stream->src.x + free_pipe->stream->src.width/2)) { - DC_LOG_SCALER("%s - ODM + window MPO(left). free_pipe:%d\n", - __func__, - free_pipe->pipe_idx); - break; - } - /* ODM + window MPO, where MPO window is on right half only */ - if (free_pipe->plane_state && - (free_pipe->plane_state->clip_rect.x >= free_pipe->stream->src.x + free_pipe->stream->src.width/2)) { - DC_LOG_SCALER("%s - ODM + window MPO(right). free_pipe:%d\n", - __func__, - free_pipe->pipe_idx); - break; - } +bool dc_add_plane_to_context( + const struct dc *dc, + struct dc_stream_state *stream, + struct dc_plane_state *plane_state, + struct dc_state *context) +{ + struct resource_pool *pool = dc->res_pool; + struct pipe_ctx *otg_master_pipe; + struct dc_stream_status *stream_status = NULL; + bool added = false; - head_pipe = head_pipe->next_odm_pipe; + stream_status = dc_stream_get_status_from_state(context, stream); + if (stream_status == NULL) { + dm_error("Existing stream not found; failed to attach surface!\n"); + goto out; + } else if (stream_status->plane_count == MAX_SURFACE_NUM) { + dm_error("Surface: can not attach plane_state %p! Maximum is: %d\n", + plane_state, MAX_SURFACE_NUM); + goto out; } - /* assign new surfaces*/ - stream_status->plane_states[stream_status->plane_count] = plane_state; - stream_status->plane_count++; + otg_master_pipe = resource_get_otg_master_for_stream( + &context->res_ctx, stream); + if (otg_master_pipe->plane_state == NULL) + added = add_plane_to_opp_head_pipes(otg_master_pipe, + plane_state, context); + else + added = acquire_secondary_dpp_pipes_and_add_plane( + otg_master_pipe, plane_state, context, + dc->current_state, pool); + if (added) { + stream_status->plane_states[stream_status->plane_count] = + plane_state; + stream_status->plane_count++; + dc_plane_state_retain(plane_state); + } - return true; +out: + return added; } bool dc_remove_plane_from_context( @@ -2219,7 +2446,7 @@ enum dc_status dc_remove_stream_from_ctx( { int i; struct dc_context *dc_ctx = dc->ctx; - struct pipe_ctx *del_pipe = resource_get_head_pipe_for_stream(&new_ctx->res_ctx, stream); + struct pipe_ctx *del_pipe = resource_get_otg_master_for_stream(&new_ctx->res_ctx, stream); struct pipe_ctx *odm_pipe; if (!del_pipe) { @@ -3473,7 +3700,7 @@ enum dc_status resource_map_clock_resources( { /* acquire new resources */ const struct resource_pool *pool = dc->res_pool; - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream( + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream( &context->res_ctx, stream); if (!pipe_ctx) @@ -3863,10 +4090,7 @@ void reset_syncd_pipes_from_disabled_pipes(struct dc *dc, pipe_ctx_old = &dc->current_state->res_ctx.pipe_ctx[i]; pipe_ctx = &context->res_ctx.pipe_ctx[i]; - if (!pipe_ctx_old->stream) - continue; - - if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) + if (!resource_is_pipe_type(pipe_ctx_old, OTG_MASTER)) continue; if (!pipe_ctx->stream || @@ -3990,11 +4214,13 @@ const struct link_hwss *get_link_hwss(const struct dc_link *link, * with an hpo encoder. Or we can return a very dummy one that doesn't * do work for all functions */ - return get_hpo_dp_link_hwss(); + return (requires_fixed_vs_pe_retimer_hpo_link_hwss(link) ? + get_hpo_fixed_vs_pe_retimer_dp_link_hwss() : get_hpo_dp_link_hwss()); else if (can_use_dpia_link_hwss(link, link_res)) return get_dpia_link_hwss(); else if (can_use_dio_link_hwss(link, link_res)) - return get_dio_link_hwss(); + return (requires_fixed_vs_pe_retimer_dio_link_hwss(link)) ? + get_dio_fixed_vs_pe_retimer_link_hwss() : get_dio_link_hwss(); else return get_virtual_link_hwss(); } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 6e11d2b701f8..01fe2d2fd241 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -71,8 +71,7 @@ static bool dc_stream_construct(struct dc_stream_state *stream, /* Copy audio modes */ /* TODO - Remove this translation */ - for (i = 0; i < (dc_sink_data->edid_caps.audio_mode_count); i++) - { + for (i = 0; i < (dc_sink_data->edid_caps.audio_mode_count); i++) { stream->audio_info.modes[i].channel_count = dc_sink_data->edid_caps.audio_modes[i].channel_count; stream->audio_info.modes[i].format_code = dc_sink_data->edid_caps.audio_modes[i].format_code; stream->audio_info.modes[i].sample_rates.all = dc_sink_data->edid_caps.audio_modes[i].sample_rate; @@ -306,6 +305,32 @@ bool dc_optimize_timing_for_fsft( } #endif +static bool is_subvp_high_refresh_candidate(struct dc_stream_state *stream) +{ + uint32_t refresh_rate; + struct dc *dc = stream->ctx->dc; + + refresh_rate = (stream->timing.pix_clk_100hz * (uint64_t)100 + + stream->timing.v_total * stream->timing.h_total - (uint64_t)1); + refresh_rate = div_u64(refresh_rate, stream->timing.v_total); + refresh_rate = div_u64(refresh_rate, stream->timing.h_total); + + /* If there's any stream that fits the SubVP high refresh criteria, + * we must return true. This is because cursor updates are asynchronous + * with full updates, so we could transition into a SubVP config and + * remain in HW cursor mode if there's no cursor update which will + * then cause corruption. + */ + if ((refresh_rate >= 120 && refresh_rate <= 175 && + stream->timing.v_addressable >= 1440 && + stream->timing.v_addressable <= 2160) && + (dc->current_state->stream_count > 1 || + (dc->current_state->stream_count == 1 && !stream->allow_freesync))) + return true; + + return false; +} + /* * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address */ @@ -334,12 +359,13 @@ bool dc_stream_set_cursor_attributes( /* SubVP is not compatible with HW cursor larger than 64 x 64 x 4. * Therefore, if cursor is greater than 64 x 64 x 4, fallback to SW cursor in the following case: - * 1. For single display cases, if resolution is >= 5K and refresh rate < 120hz - * 2. For multi display cases, if resolution is >= 4K and refresh rate < 120hz - * - * [< 120hz is a requirement for SubVP configs] + * 1. If the config is a candidate for SubVP high refresh (both single an dual display configs) + * 2. If not subvp high refresh, for single display cases, if resolution is >= 5K and refresh rate < 120hz + * 3. If not subvp high refresh, for multi display cases, if resolution is >= 4K and refresh rate < 120hz */ if (dc->debug.allow_sw_cursor_fallback && attributes->height * attributes->width * 4 > 16384) { + if (!dc->debug.disable_subvp_high_refresh && is_subvp_high_refresh_candidate(stream)) + return false; if (dc->current_state->stream_count == 1 && stream->timing.v_addressable >= 2880 && ((stream->timing.pix_clk_100hz * 100) / stream->timing.v_total / stream->timing.h_total) < 120) return false; diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 63948170fd6d..31e3183497a7 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -40,12 +40,14 @@ #include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" +struct abm_save_restore; + /* forward declaration */ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.241" +#define DC_VER "3.2.247" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -428,6 +430,7 @@ enum visual_confirm { VISUAL_CONFIRM_SWAPCHAIN = 6, VISUAL_CONFIRM_FAMS = 7, VISUAL_CONFIRM_SWIZZLE = 9, + VISUAL_CONFIRM_REPLAY = 12, VISUAL_CONFIRM_SUBVP = 14, VISUAL_CONFIRM_MCLK_SWITCH = 16, }; @@ -506,7 +509,7 @@ enum dcn_zstate_support_state { DCN_ZSTATE_SUPPORT_DISALLOW, }; -/** +/* * struct dc_clocks - DC pipe clocks * * For any clocks that may differ per pipe only the max is stored in this @@ -728,7 +731,7 @@ struct resource_pool; struct dce_hwseq; struct link_service; -/** +/* * struct dc_debug_options - DC debug struct * * This struct provides a simple mechanism for developers to change some @@ -756,7 +759,7 @@ struct dc_debug_options { bool use_max_lb; enum dcc_option disable_dcc; - /** + /* * @pipe_split_policy: Define which pipe split policy is used by the * display core. */ @@ -861,6 +864,7 @@ struct dc_debug_options { bool psr_skip_crtc_disable; union dpia_debug_options dpia_debug; bool disable_fixed_vs_aux_timeout_wa; + uint32_t fixed_vs_aux_delay_config_wa; bool force_disable_subvp; bool force_subvp_mclk_switch; bool allow_sw_cursor_fallback; @@ -902,9 +906,18 @@ struct dc_debug_options { uint32_t fpo_vactive_max_blank_us; bool enable_legacy_fast_update; bool disable_dc_mode_overwrite; + bool replay_skip_crtc_disabled; }; struct gpu_info_soc_bounding_box_v1_0; + +/* Generic structure that can be used to query properties of DC. More fields + * can be added as required. + */ +struct dc_current_properties { + unsigned int cursor_size_limit; +}; + struct dc { struct dc_debug_options debug; struct dc_versions versions; @@ -1334,7 +1347,7 @@ struct dc_validation_set { struct dc_stream_state *stream; /** - * @plane_state: Surface state + * @plane_states: Surface state */ struct dc_plane_state *plane_states[MAX_SURFACES]; @@ -1409,10 +1422,14 @@ struct dc_plane_state *dc_get_surface_for_mpcc(struct dc *dc, uint32_t dc_get_opp_for_plane(struct dc *dc, struct dc_plane_state *plane); +void dc_set_disable_128b_132b_stream_overhead(bool disable); + /* The function returns minimum bandwidth required to drive a given timing * return - minimum required timing bandwidth in kbps. */ -uint32_t dc_bandwidth_in_kbps_from_timing(const struct dc_crtc_timing *timing); +uint32_t dc_bandwidth_in_kbps_from_timing( + const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding); /* Link Interfaces */ /* @@ -1479,8 +1496,10 @@ struct dc_link { * object creation. */ enum engine_id eng_id; + enum engine_id dpia_preferred_eng_id; bool test_pattern_enabled; + enum dp_test_pattern current_test_pattern; union compliance_test_state compliance_test_state; void *priv; @@ -1514,8 +1533,11 @@ struct dc_link { enum edp_revision edp_revision; union dpcd_sink_ext_caps dpcd_sink_ext_caps; + struct backlight_settings backlight_settings; struct psr_settings psr_settings; + struct replay_settings replay_settings; + /* Drive settings read from integrated info table */ struct dc_lane_settings bios_forced_drive_settings; @@ -1849,6 +1871,14 @@ enum dp_link_encoding dc_link_dp_mst_decide_link_encoding_format( */ const struct dc_link_settings *dc_link_get_link_cap(const struct dc_link *link); +/* Get the highest encoding format that the link supports; highest meaning the + * encoding format which supports the maximum bandwidth. + * + * @link - a link with DP RX connection + * return - highest encoding format link supports. + */ +enum dc_link_encoding_format dc_link_get_highest_encoding_format(const struct dc_link *link); + /* Check if a RX (ex. DP sink, MST hub, passive or active dongle) is connected * to a link with dp connector signal type. * @link - a link with dp connector signal type @@ -1983,6 +2013,8 @@ bool dc_link_setup_psr(struct dc_link *dc_link, const struct dc_stream_state *stream, struct psr_config *psr_config, struct psr_context *psr_context); +bool dc_link_get_replay_state(const struct dc_link *dc_link, uint64_t *state); + /* On eDP links this function call will stall until T12 has elapsed. * If the panel is not in power off state, this function will return * immediately. @@ -2230,6 +2262,11 @@ void dc_z10_save_init(struct dc *dc); bool dc_is_dmub_outbox_supported(struct dc *dc); bool dc_enable_dmub_notifications(struct dc *dc); +bool dc_abm_save_restore( + struct dc *dc, + struct dc_stream_state *stream, + struct abm_save_restore *pData); + void dc_enable_dmub_outbox(struct dc *dc); bool dc_process_dmub_aux_transfer_async(struct dc *dc, @@ -2255,6 +2292,8 @@ void dc_process_dmub_dpia_hpd_int_enable(const struct dc *dc, void dc_print_dmub_diagnostic_data(const struct dc *dc); +void dc_query_current_properties(struct dc *dc, struct dc_current_properties *properties); + /* DSC Interfaces */ #include "dc_dsc.h" diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index c753c6f30dd7..4c5ef3ef8dbd 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -31,6 +31,7 @@ #include "core_types.h" #include "../basics/conversion.h" #include "cursor_reg_cache.h" +#include "resource.h" #define CTX dc_dmub_srv->ctx #define DC_LOGGER CTX->logger @@ -356,7 +357,7 @@ bool dc_dmub_srv_p_state_delegate(struct dc *dc, bool should_manage_pstate, stru for (i = 0, k = 0; context && i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->top_pipe && !pipe->prev_odm_pipe && pipe->stream && pipe->stream->fpo_in_use) { + if (resource_is_pipe_type(pipe, OTG_MASTER) && pipe->stream->fpo_in_use) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; uint8_t min_refresh_in_hz = (pipe->stream->timing.min_refresh_in_uhz + 999999) / 1000000; @@ -381,6 +382,9 @@ void dc_dmub_srv_query_caps_cmd(struct dc_dmub_srv *dc_dmub_srv) { union dmub_rb_cmd cmd = { 0 }; + if (dc_dmub_srv->ctx->dc->debug.dmcub_emulation) + return; + memset(&cmd, 0, sizeof(cmd)); /* Prepare fw command */ @@ -528,7 +532,8 @@ static void populate_subvp_cmd_vblank_pipe_info(struct dc *dc, // We check for master pipe, but it shouldn't matter since we only need // the pipe for timing info (stream should be same for any pipe splits) - if (!pipe->stream || !pipe->plane_state || pipe->top_pipe || pipe->prev_odm_pipe) + if (!resource_is_pipe_type(pipe, OTG_MASTER) || + !resource_is_pipe_type(pipe, DPP_PIPE)) continue; // Find the SubVP pipe @@ -725,12 +730,10 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->stream) - continue; - /* For SubVP pipe count, only count the top most (ODM / MPC) pipe */ - if (pipe->plane_state && !pipe->top_pipe && !pipe->prev_odm_pipe && + if (resource_is_pipe_type(pipe, OTG_MASTER) && + resource_is_pipe_type(pipe, DPP_PIPE) && pipe->stream->mall_stream_config.type == SUBVP_MAIN) subvp_pipes[subvp_count++] = pipe; } @@ -747,12 +750,14 @@ void dc_dmub_setup_subvp_dmub_command(struct dc *dc, * Any ODM or MPC splits being used in SubVP will be handled internally in * populate_subvp_cmd_pipe_info */ - if (pipe->plane_state && pipe->stream->mall_stream_config.paired_stream && - !pipe->top_pipe && !pipe->prev_odm_pipe && + if (resource_is_pipe_type(pipe, OTG_MASTER) && + resource_is_pipe_type(pipe, DPP_PIPE) && + pipe->stream->mall_stream_config.paired_stream && pipe->stream->mall_stream_config.type == SUBVP_MAIN) { populate_subvp_cmd_pipe_info(dc, context, &cmd, pipe, cmd_pipe_index++); - } else if (pipe->plane_state && pipe->stream->mall_stream_config.type == SUBVP_NONE && - !pipe->top_pipe && !pipe->prev_odm_pipe) { + } else if (resource_is_pipe_type(pipe, OTG_MASTER) && + resource_is_pipe_type(pipe, DPP_PIPE) && + pipe->stream->mall_stream_config.type == SUBVP_NONE) { // Don't need to check for ActiveDRAMClockChangeMargin < 0, not valid in cases where // we run through DML without calculating "natural" P-state support populate_subvp_cmd_vblank_pipe_info(dc, context, &cmd, pipe, cmd_pipe_index++); @@ -894,6 +899,9 @@ static bool dc_dmub_should_update_cursor_data(struct pipe_ctx *pipe_ctx) pipe_ctx->stream->ctx->dce_version >= DCN_VERSION_3_1) return true; + if (pipe_ctx->stream->link->replay_settings.config.replay_supported) + return true; + return false; } @@ -1018,3 +1026,32 @@ bool dc_dmub_check_min_version(struct dmub_srv *srv) return true; return srv->hw_funcs.is_psrsu_supported(srv); } + +void dc_dmub_srv_enable_dpia_trace(const struct dc *dc) +{ + struct dc_dmub_srv *dc_dmub_srv = dc->ctx->dmub_srv; + struct dmub_srv *dmub; + enum dmub_status status; + static const uint32_t timeout_us = 30; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) { + DC_LOG_ERROR("%s: invalid parameters.", __func__); + return; + } + + dmub = dc_dmub_srv->dmub; + + status = dmub_srv_send_gpint_command(dmub, DMUB_GPINT__SET_TRACE_BUFFER_MASK_WORD1, 0x0010, timeout_us); + if (status != DMUB_STATUS_OK) { + DC_LOG_ERROR("timeout updating trace buffer mask word\n"); + return; + } + + status = dmub_srv_send_gpint_command(dmub, DMUB_GPINT__UPDATE_TRACE_BUFFER_MASK, 0x0000, timeout_us); + if (status != DMUB_STATUS_OK) { + DC_LOG_ERROR("timeout updating trace buffer mask word\n"); + return; + } + + DC_LOG_DEBUG("Enabled DPIA trace\n"); +}
\ No newline at end of file diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index 099f94b6107c..bb3fe162dd93 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -87,4 +87,7 @@ void dc_dmub_srv_log_diagnostic_data(struct dc_dmub_srv *dc_dmub_srv); void dc_send_update_cursor_info_to_dmu(struct pipe_ctx *pCtx, uint8_t pipe_idx); bool dc_dmub_check_min_version(struct dmub_srv *srv); + +void dc_dmub_srv_enable_dpia_trace(const struct dc *dc); + #endif /* _DMUB_DC_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 55139d7bf422..cfaa39c5dd16 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -1117,6 +1117,11 @@ struct edp_psr_info { uint8_t force_psrsu_cap; }; +struct replay_info { + uint8_t pixel_deviation_per_line; + uint8_t max_deviation_line; +}; + struct dprx_states { bool cable_id_written; }; @@ -1236,6 +1241,8 @@ struct dpcd_caps { uint8_t edp_rev; union edp_alpm_caps alpm_caps; struct edp_psr_info psr_info; + + struct replay_info pr_info; }; union dpcd_sink_ext_caps { @@ -1276,6 +1283,28 @@ union dpcd_psr_configuration { unsigned char raw; }; +union replay_enable_and_configuration { + struct { + unsigned char FREESYNC_PANEL_REPLAY_MODE :1; + unsigned char TIMING_DESYNC_ERROR_VERIFICATION :1; + unsigned char STATE_TRANSITION_ERROR_DETECTION :1; + unsigned char RESERVED0 :1; + unsigned char RESERVED1 :4; + } bits; + unsigned char raw; +}; + +union dpcd_replay_configuration { + struct { + unsigned char STATE_TRANSITION_ERROR_STATUS : 1; + unsigned char DESYNC_ERROR_STATUS : 1; + unsigned char SINK_DEVICE_REPLAY_STATUS : 3; + unsigned char SINK_FRAME_LOCKED : 2; + unsigned char RESERVED : 1; + } bits; + unsigned char raw; +}; + union dpcd_alpm_configuration { struct { unsigned char ENABLE : 1; diff --git a/drivers/gpu/drm/amd/display/dc/dc_dsc.h b/drivers/gpu/drm/amd/display/dc/dc_dsc.h index 9491b76d61f5..fe3078b8789e 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dsc.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dsc.h @@ -73,6 +73,7 @@ bool dc_dsc_compute_bandwidth_range( uint32_t max_bpp_x16, const struct dsc_dec_dpcd_caps *dsc_sink_caps, const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_bw_range *range); bool dc_dsc_compute_config( @@ -81,6 +82,7 @@ bool dc_dsc_compute_config( const struct dc_dsc_config_options *options, uint32_t target_bandwidth_kbps, const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_config *dsc_cfg); uint32_t dc_dsc_stream_bandwidth_in_kbps(const struct dc_crtc_timing *timing, diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 0ce7728a5a4b..445ad79001ce 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -189,7 +189,6 @@ struct dc_panel_patch { unsigned int disable_fams; unsigned int skip_avmute; unsigned int mst_start_top_delay; - unsigned int delay_disable_aux_intercept_ms; }; struct dc_edid_caps { @@ -879,7 +878,7 @@ struct dsc_dec_dpcd_caps { uint32_t branch_overall_throughput_0_mps; /* In MPs */ uint32_t branch_overall_throughput_1_mps; /* In MPs */ uint32_t branch_max_line_width; - bool is_dp; + bool is_dp; /* Decoded format */ }; struct dc_golden_table { @@ -902,6 +901,14 @@ enum dc_gpu_mem_alloc_type { DC_MEM_ALLOC_TYPE_AGP }; +enum dc_link_encoding_format { + DC_LINK_ENCODING_UNSPECIFIED = 0, + DC_LINK_ENCODING_DP_8b_10b, + DC_LINK_ENCODING_DP_128b_132b, + DC_LINK_ENCODING_HDMI_TMDS, + DC_LINK_ENCODING_HDMI_FRL +}; + enum dc_psr_version { DC_PSR_VERSION_1 = 0, DC_PSR_VERSION_SU_1 = 1, @@ -995,6 +1002,10 @@ struct link_mst_stream_allocation_table { struct link_mst_stream_allocation stream_allocations[MAX_CONTROLLER_NUM]; }; +struct backlight_settings { + uint32_t backlight_millinits; +}; + /* PSR feature flags */ struct psr_settings { bool psr_feature_enabled; // PSR is supported by sink @@ -1014,6 +1025,45 @@ struct psr_settings { unsigned int psr_power_opt; }; +enum replay_coasting_vtotal_type { + PR_COASTING_TYPE_NOM = 0, + PR_COASTING_TYPE_STATIC, + PR_COASTING_TYPE_FULL_SCREEN_VIDEO, + PR_COASTING_TYPE_TEST_HARNESS, + PR_COASTING_TYPE_NUM, +}; + +union replay_error_status { + struct { + unsigned char STATE_TRANSITION_ERROR :1; + unsigned char LINK_CRC_ERROR :1; + unsigned char DESYNC_ERROR :1; + unsigned char RESERVED :5; + } bits; + unsigned char raw; +}; + +struct replay_config { + bool replay_supported; // Replay feature is supported + unsigned int replay_power_opt_supported; // Power opt flags that are supported + bool replay_smu_opt_supported; // SMU optimization is supported + unsigned int replay_enable_option; // Replay enablement option + uint32_t debug_flags; // Replay debug flags + bool replay_timing_sync_supported; // Replay desync is supported + union replay_error_status replay_error_status; // Replay error status +}; + +/* Replay feature flags */ +struct replay_settings { + struct replay_config config; // Replay configuration + bool replay_feature_enabled; // Replay feature is ready for activating + bool replay_allow_active; // Replay is currently active + unsigned int replay_power_opt_active; // Power opt flags that are activated currently + bool replay_smu_opt_enable; // SMU optimization is enabled + uint16_t coasting_vtotal; // Current Coasting vtotal + uint16_t coasting_vtotal_table[PR_COASTING_TYPE_NUM]; // Coasting vtotal table +}; + /* To split out "global" and "per-panel" config settings. * Add a struct dc_panel_config under dc_link */ @@ -1040,9 +1090,11 @@ struct dc_panel_config { struct psr { bool disable_psr; bool disallow_psrsu; + bool disallow_replay; bool rc_disable; bool rc_allow_static_screen; bool rc_allow_fullscreen_VPB; + unsigned int replay_enable_option; } psr; /* ABM */ struct varib { diff --git a/drivers/gpu/drm/amd/display/dc/dce/Makefile b/drivers/gpu/drm/amd/display/dc/dce/Makefile index 01490c9ba958..15b64c26d5a2 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dce/Makefile @@ -30,7 +30,7 @@ DCE = dce_audio.o dce_stream_encoder.o dce_link_encoder.o dce_hwseq.o \ dce_mem_input.o dce_clock_source.o dce_scl_filters.o dce_transform.o \ dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \ dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o dmub_psr.o dmub_abm.o dmub_abm_lcd.o dce_panel_cntl.o \ -dmub_hw_lock_mgr.o dmub_outbox.o +dmub_hw_lock_mgr.o dmub_outbox.o dmub_replay.o AMD_DAL_DCE = $(addprefix $(AMDDALPATH)/dc/dce/,$(DCE)) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c index 07359eb89efc..e7acd6eec1fd 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c @@ -640,7 +640,7 @@ static void dce11_pplib_apply_display_requirements( * on power saving. * */ - pp_display_cfg->min_dcfclock_khz = (context->stream_count > 4)? + pp_display_cfg->min_dcfclock_khz = (context->stream_count > 4) ? pp_display_cfg->min_engine_clock_khz : 0; pp_display_cfg->min_engine_clock_deep_sleep_khz diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c index 63009db8b5a7..b87bfecb7755 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_dmcu.c @@ -76,9 +76,9 @@ static bool dce_dmcu_init(struct dmcu *dmcu) } static bool dce_dmcu_load_iram(struct dmcu *dmcu, - unsigned int start_offset, - const char *src, - unsigned int bytes) + unsigned int start_offset, + const char *src, + unsigned int bytes) { struct dce_dmcu *dmcu_dce = TO_DCE_DMCU(dmcu); unsigned int count = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h index a3fee929cd12..86233f94db4a 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h @@ -98,6 +98,29 @@ SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 4), \ SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 5) +#define HWSEQ_PIXEL_RATE_REG_LIST_302(blk) \ + SRII(PIXEL_RATE_CNTL, blk, 0), \ + SRII(PIXEL_RATE_CNTL, blk, 1),\ + SRII(PIXEL_RATE_CNTL, blk, 2),\ + SRII(PIXEL_RATE_CNTL, blk, 3), \ + SRII(PIXEL_RATE_CNTL, blk, 4) + +#define HWSEQ_PHYPLL_REG_LIST_302(blk) \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 0), \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 1),\ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 2),\ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 3), \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 4) + +#define HWSEQ_PIXEL_RATE_REG_LIST_303(blk) \ + SRII(PIXEL_RATE_CNTL, blk, 0), \ + SRII(PIXEL_RATE_CNTL, blk, 1) + +#define HWSEQ_PHYPLL_REG_LIST_303(blk) \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 0), \ + SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 1) + + #define HWSEQ_PHYPLL_REG_LIST_201(blk) \ SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 0), \ SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 1) @@ -387,7 +410,11 @@ SR(MPC_CRC_RESULT_C), \ SR(MPC_CRC_RESULT_AR), \ SR(AZALIA_AUDIO_DTO), \ - SR(AZALIA_CONTROLLER_CLOCK_GATING) + SR(AZALIA_CONTROLLER_CLOCK_GATING), \ + SR(HPO_TOP_CLOCK_CONTROL), \ + SR(ODM_MEM_PWR_CTRL3), \ + SR(DMU_MEM_PWR_CNTL), \ + SR(MMHUBBUB_MEM_PWR_CNTL) #define HWSEQ_DCN301_REG_LIST()\ SR(REFCLK_CNTL), \ @@ -508,8 +535,11 @@ SR(D5VGA_CONTROL), \ SR(D6VGA_CONTROL), \ SR(DC_IP_REQUEST_CNTL), \ + HWSEQ_PIXEL_RATE_REG_LIST_302(OTG), \ + HWSEQ_PHYPLL_REG_LIST_302(OTG), \ SR(AZALIA_AUDIO_DTO), \ - SR(AZALIA_CONTROLLER_CLOCK_GATING) + SR(AZALIA_CONTROLLER_CLOCK_GATING), \ + SR(HPO_TOP_CLOCK_CONTROL) #define HWSEQ_DCN303_REG_LIST() \ HWSEQ_DCN_REG_LIST(), \ @@ -540,28 +570,6 @@ SR(AZALIA_CONTROLLER_CLOCK_GATING), \ SR(HPO_TOP_CLOCK_CONTROL) -#define HWSEQ_PIXEL_RATE_REG_LIST_302(blk) \ - SRII(PIXEL_RATE_CNTL, blk, 0), \ - SRII(PIXEL_RATE_CNTL, blk, 1),\ - SRII(PIXEL_RATE_CNTL, blk, 2),\ - SRII(PIXEL_RATE_CNTL, blk, 3), \ - SRII(PIXEL_RATE_CNTL, blk, 4) - -#define HWSEQ_PHYPLL_REG_LIST_302(blk) \ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 0), \ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 1),\ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 2),\ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 3), \ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 4) - -#define HWSEQ_PIXEL_RATE_REG_LIST_303(blk) \ - SRII(PIXEL_RATE_CNTL, blk, 0), \ - SRII(PIXEL_RATE_CNTL, blk, 1) - -#define HWSEQ_PHYPLL_REG_LIST_303(blk) \ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 0), \ - SRII(PHYPLL_PIXEL_RATE_CNTL, blk, 1) - struct dce_hwseq_registers { uint32_t DCFE_CLOCK_CONTROL[6]; uint32_t DCFEV_CLOCK_CONTROL; @@ -663,14 +671,15 @@ struct dce_hwseq_registers { uint32_t MC_VM_XGMI_LFB_CNTL; uint32_t AZALIA_AUDIO_DTO; uint32_t AZALIA_CONTROLLER_CLOCK_GATING; + /* MMHUB VM */ + uint32_t MC_VM_FB_LOCATION_BASE; + uint32_t MC_VM_FB_LOCATION_TOP; + uint32_t MC_VM_FB_OFFSET; + uint32_t MMHUBBUB_MEM_PWR_CNTL; uint32_t HPO_TOP_CLOCK_CONTROL; uint32_t ODM_MEM_PWR_CTRL3; uint32_t DMU_MEM_PWR_CNTL; - uint32_t MMHUBBUB_MEM_PWR_CNTL; uint32_t DCHUBBUB_ARB_HOSTVM_CNTL; - uint32_t MC_VM_FB_LOCATION_BASE; - uint32_t MC_VM_FB_LOCATION_TOP; - uint32_t MC_VM_FB_OFFSET; uint32_t HPO_TOP_HW_CONTROL; }; /* set field name */ @@ -915,6 +924,7 @@ struct dce_hwseq_registers { #define HWSEQ_DCN30_MASK_SH_LIST(mask_sh)\ HWSEQ_DCN2_MASK_SH_LIST(mask_sh), \ HWS_SF(, AZALIA_AUDIO_DTO, AZALIA_AUDIO_DTO_MODULE, mask_sh), \ + HWS_SF(, HPO_TOP_CLOCK_CONTROL, HPO_HDMISTREAMCLK_GATE_DIS, mask_sh), \ HWS_SF(, ODM_MEM_PWR_CTRL3, ODM_MEM_UNASSIGNED_PWR_MODE, mask_sh), \ HWS_SF(, ODM_MEM_PWR_CTRL3, ODM_MEM_VBLANK_PWR_MODE, mask_sh), \ HWS_SF(, DMU_MEM_PWR_CNTL, DMCU_ERAM_MEM_PWR_FORCE, mask_sh), \ @@ -1012,7 +1022,8 @@ struct dce_hwseq_registers { HWS_SF(, DOMAIN19_PG_STATUS, DOMAIN19_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DOMAIN20_PG_STATUS, DOMAIN20_PGFSM_PWR_STATUS, mask_sh), \ HWS_SF(, DC_IP_REQUEST_CNTL, IP_REQUEST_EN, mask_sh), \ - HWS_SF(, AZALIA_AUDIO_DTO, AZALIA_AUDIO_DTO_MODULE, mask_sh) + HWS_SF(, AZALIA_AUDIO_DTO, AZALIA_AUDIO_DTO_MODULE, mask_sh), \ + HWS_SF(, HPO_TOP_CLOCK_CONTROL, HPO_HDMISTREAMCLK_GATE_DIS, mask_sh) #define HWSEQ_DCN303_MASK_SH_LIST(mask_sh) \ HWSEQ_DCN_MASK_SH_LIST(mask_sh), \ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c index 6d1b01c267b7..4f552c3e7663 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c @@ -442,10 +442,9 @@ struct dce_i2c_hw *acquire_i2c_hw_engine( return dce_i2c_hw; } -static enum i2c_channel_operation_result dce_i2c_hw_engine_wait_on_operation_result( - struct dce_i2c_hw *dce_i2c_hw, - uint32_t timeout, - enum i2c_channel_operation_result expected_result) +static enum i2c_channel_operation_result dce_i2c_hw_engine_wait_on_operation_result(struct dce_i2c_hw *dce_i2c_hw, + uint32_t timeout, + enum i2c_channel_operation_result expected_result) { enum i2c_channel_operation_result result; uint32_t i = 0; @@ -509,11 +508,10 @@ static uint32_t get_transaction_timeout_hw( return period_timeout * num_of_clock_stretches; } -static bool dce_i2c_hw_engine_submit_payload( - struct dce_i2c_hw *dce_i2c_hw, - struct i2c_payload *payload, - bool middle_of_transaction, - uint32_t speed) +static bool dce_i2c_hw_engine_submit_payload(struct dce_i2c_hw *dce_i2c_hw, + struct i2c_payload *payload, + bool middle_of_transaction, + uint32_t speed) { struct i2c_request_transaction_data request; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c index f1aeb6d1967c..e188447c8156 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c @@ -367,6 +367,7 @@ static bool dce_i2c_sw_engine_acquire_engine( return true; } + bool dce_i2c_engine_acquire_sw( struct dce_i2c_sw *dce_i2c_sw, struct ddc *ddc_handle) @@ -392,12 +393,8 @@ bool dce_i2c_engine_acquire_sw( return result; } - - - -static void dce_i2c_sw_engine_submit_channel_request( - struct dce_i2c_sw *engine, - struct i2c_request_transaction_data *req) +static void dce_i2c_sw_engine_submit_channel_request(struct dce_i2c_sw *engine, + struct i2c_request_transaction_data *req) { struct ddc *ddc = engine->ddc; uint16_t clock_delay_div_4 = engine->clock_delay >> 2; @@ -439,10 +436,9 @@ static void dce_i2c_sw_engine_submit_channel_request( I2C_CHANNEL_OPERATION_FAILED; } -static bool dce_i2c_sw_engine_submit_payload( - struct dce_i2c_sw *engine, - struct i2c_payload *payload, - bool middle_of_transaction) +static bool dce_i2c_sw_engine_submit_payload(struct dce_i2c_sw *engine, + struct i2c_payload *payload, + bool middle_of_transaction) { struct i2c_request_transaction_data request; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c index fa314493ffc5..136bd93c3b65 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c @@ -941,9 +941,7 @@ bool dce110_link_encoder_validate_output_with_stream( break; case SIGNAL_TYPE_EDP: case SIGNAL_TYPE_LVDS: - is_valid = - (stream->timing. - pixel_encoding == PIXEL_ENCODING_RGB) ? true : false; + is_valid = stream->timing.pixel_encoding == PIXEL_ENCODING_RGB; break; case SIGNAL_TYPE_VIRTUAL: is_valid = true; @@ -1645,7 +1643,7 @@ void dce110_link_encoder_enable_hpd(struct link_encoder *enc) uint32_t hpd_enable = 0; uint32_t value = dm_read_reg(ctx, addr); - get_reg_field_value(hpd_enable, DC_HPD_CONTROL, DC_HPD_EN); + hpd_enable = get_reg_field_value(hpd_enable, DC_HPD_CONTROL, DC_HPD_EN); if (hpd_enable == 0) set_reg_field_value(value, 1, DC_HPD_CONTROL, DC_HPD_EN); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c index 2fb9572ce25d..d3e6544022b7 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm.c @@ -27,6 +27,7 @@ #include "dmub_abm_lcd.h" #include "dc.h" #include "core_types.h" +#include "dmub_cmd.h" #define TO_DMUB_ABM(abm)\ container_of(abm, struct dce_abm, base) @@ -118,6 +119,32 @@ static bool dmub_abm_set_pause_ex(struct abm *abm, bool pause, unsigned int pane return ret; } +/***************************************************************************** + * dmub_abm_save_restore_ex() - calls dmub_abm_save_restore for preserving DMUB's + * Varibright states for LCD only. OLED is TBD + * @abm: used to check get dc context + * @panel_inst: panel instance index + * @pData: contains command to pause/un-pause abm and abm parameters + * + * + ***************************************************************************/ +static bool dmub_abm_save_restore_ex( + struct abm *abm, + unsigned int panel_inst, + struct abm_save_restore *pData) +{ + bool ret = false; + unsigned int feature_support; + struct dc_context *dc = abm->ctx; + + feature_support = abm_feature_support(abm, panel_inst); + + if (feature_support == ABM_LCD_SUPPORT) + ret = dmub_abm_save_restore(dc, panel_inst, pData); + + return ret; +} + static bool dmub_abm_set_pipe_ex(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst) { bool ret = false; @@ -155,6 +182,7 @@ static const struct abm_funcs abm_funcs = { .get_target_backlight = dmub_abm_get_target_backlight_ex, .init_abm_config = dmub_abm_init_config_ex, .set_abm_pause = dmub_abm_set_pause_ex, + .save_restore = dmub_abm_save_restore_ex, .set_pipe_ex = dmub_abm_set_pipe_ex, .set_backlight_level_pwm = dmub_abm_set_backlight_level_pwm_ex, }; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c index 39da73eba86e..592a8f7a1c6d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.c @@ -208,6 +208,52 @@ bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, un return true; } + +/***************************************************************************** + * dmub_abm_save_restore() - dmub interface for abm save+pause and restore+ + * un-pause + * @dc: dc context + * @panel_inst: panel instance index + * @pData: contains command to pause/un-pause abm and exchange abm parameters + * + * When called Pause will get abm data and store in pData, and un-pause will + * set/apply abm data stored in pData. + * + *****************************************************************************/ +bool dmub_abm_save_restore( + struct dc_context *dc, + unsigned int panel_inst, + struct abm_save_restore *pData) +{ + union dmub_rb_cmd cmd; + uint8_t panel_mask = 0x01 << panel_inst; + unsigned int bytes = sizeof(struct abm_save_restore); + + // TODO: Optimize by only reading back final 4 bytes + dmub_flush_buffer_mem(&dc->dmub_srv->dmub->scratch_mem_fb); + + // Copy iramtable into cw7 + memcpy(dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, (void *)pData, bytes); + + memset(&cmd, 0, sizeof(cmd)); + cmd.abm_save_restore.header.type = DMUB_CMD__ABM; + cmd.abm_save_restore.header.sub_type = DMUB_CMD__ABM_SAVE_RESTORE; + + cmd.abm_save_restore.abm_init_config_data.src.quad_part = dc->dmub_srv->dmub->scratch_mem_fb.gpu_addr; + cmd.abm_save_restore.abm_init_config_data.bytes = bytes; + cmd.abm_save_restore.abm_init_config_data.version = DMUB_CMD_ABM_CONTROL_VERSION_1; + cmd.abm_save_restore.abm_init_config_data.panel_mask = panel_mask; + + cmd.abm_save_restore.header.payload_bytes = sizeof(struct dmub_rb_cmd_abm_save_restore); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + // Copy iramtable data into local structure + memcpy((void *)pData, dc->dmub_srv->dmub->scratch_mem_fb.cpu_addr, bytes); + + return true; +} + bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst) { union dmub_rb_cmd cmd; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h index 00b4e268768e..853564d7f471 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_abm_lcd.h @@ -28,6 +28,8 @@ #include "abm.h" +struct abm_save_restore; + void dmub_abm_init(struct abm *abm, uint32_t backlight); bool dmub_abm_set_level(struct abm *abm, uint32_t level, uint8_t panel_mask); unsigned int dmub_abm_get_current_backlight(struct abm *abm); @@ -38,6 +40,10 @@ void dmub_abm_init_config(struct abm *abm, unsigned int inst); bool dmub_abm_set_pause(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int stream_inst); +bool dmub_abm_save_restore( + struct dc_context *dc, + unsigned int panel_inst, + struct abm_save_restore *pData); bool dmub_abm_set_pipe(struct abm *abm, uint32_t otg_inst, uint32_t option, uint32_t panel_inst); bool dmub_abm_set_backlight_level(struct abm *abm, unsigned int backlight_pwm_u16_16, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c new file mode 100644 index 000000000000..28149e53c2a6 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c @@ -0,0 +1,303 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "dc.h" +#include "dc_dmub_srv.h" +#include "dmub/dmub_srv.h" +#include "core_types.h" +#include "dmub_replay.h" + +#define DC_TRACE_LEVEL_MESSAGE(...) /* do nothing */ + +#define MAX_PIPES 6 + +/* + * Get Replay state from firmware. + */ +static void dmub_replay_get_state(struct dmub_replay *dmub, enum replay_state *state, uint8_t panel_inst) +{ + struct dmub_srv *srv = dmub->ctx->dmub_srv->dmub; + /* uint32_t raw_state = 0; */ + uint32_t retry_count = 0; + enum dmub_status status; + + do { + // Send gpint command and wait for ack + status = dmub_srv_send_gpint_command(srv, DMUB_GPINT__GET_REPLAY_STATE, panel_inst, 30); + + if (status == DMUB_STATUS_OK) { + // GPINT was executed, get response + dmub_srv_get_gpint_response(srv, (uint32_t *)state); + } else + // Return invalid state when GPINT times out + *state = REPLAY_STATE_INVALID; + } while (++retry_count <= 1000 && *state == REPLAY_STATE_INVALID); + + // Assert if max retry hit + if (retry_count >= 1000 && *state == REPLAY_STATE_INVALID) { + ASSERT(0); + /* To-do: Add retry fail log */ + } +} + +/* + * Enable/Disable Replay. + */ +static void dmub_replay_enable(struct dmub_replay *dmub, bool enable, bool wait, uint8_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + uint32_t retry_count; + enum replay_state state = REPLAY_STATE_0; + + memset(&cmd, 0, sizeof(cmd)); + cmd.replay_enable.header.type = DMUB_CMD__REPLAY; + cmd.replay_enable.data.panel_inst = panel_inst; + + cmd.replay_enable.header.sub_type = DMUB_CMD__REPLAY_ENABLE; + if (enable) + cmd.replay_enable.data.enable = REPLAY_ENABLE; + else + cmd.replay_enable.data.enable = REPLAY_DISABLE; + + cmd.replay_enable.header.payload_bytes = sizeof(struct dmub_rb_cmd_replay_enable_data); + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + /* Below loops 1000 x 500us = 500 ms. + * Exit REPLAY may need to wait 1-2 frames to power up. Timeout after at + * least a few frames. Should never hit the max retry assert below. + */ + if (wait) { + for (retry_count = 0; retry_count <= 1000; retry_count++) { + dmub_replay_get_state(dmub, &state, panel_inst); + + if (enable) { + if (state != REPLAY_STATE_0) + break; + } else { + if (state == REPLAY_STATE_0) + break; + } + + fsleep(500); + } + + /* assert if max retry hit */ + if (retry_count >= 1000) + ASSERT(0); + } + +} + +/* + * Set REPLAY power optimization flags. + */ +static void dmub_replay_set_power_opt(struct dmub_replay *dmub, unsigned int power_opt, uint8_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + + memset(&cmd, 0, sizeof(cmd)); + cmd.replay_set_power_opt.header.type = DMUB_CMD__REPLAY; + cmd.replay_set_power_opt.header.sub_type = DMUB_CMD__SET_REPLAY_POWER_OPT; + cmd.replay_set_power_opt.header.payload_bytes = sizeof(struct dmub_cmd_replay_set_power_opt_data); + cmd.replay_set_power_opt.replay_set_power_opt_data.power_opt = power_opt; + cmd.replay_set_power_opt.replay_set_power_opt_data.panel_inst = panel_inst; + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + +/* + * Setup Replay by programming phy registers and sending replay hw context values to firmware. + */ +static bool dmub_replay_copy_settings(struct dmub_replay *dmub, + struct dc_link *link, + struct replay_context *replay_context, + uint8_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + struct dmub_cmd_replay_copy_settings_data *copy_settings_data + = &cmd.replay_copy_settings.replay_copy_settings_data; + struct pipe_ctx *pipe_ctx = NULL; + struct resource_context *res_ctx = &link->ctx->dc->current_state->res_ctx; + int i = 0; + + for (i = 0; i < MAX_PIPES; i++) { + if (res_ctx && + res_ctx->pipe_ctx[i].stream && + res_ctx->pipe_ctx[i].stream->link && + res_ctx->pipe_ctx[i].stream->link == link && + res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) { + pipe_ctx = &res_ctx->pipe_ctx[i]; + //TODO: refactor for multi edp support + break; + } + } + + if (!pipe_ctx) + return false; + + memset(&cmd, 0, sizeof(cmd)); + cmd.replay_copy_settings.header.type = DMUB_CMD__REPLAY; + cmd.replay_copy_settings.header.sub_type = DMUB_CMD__REPLAY_COPY_SETTINGS; + cmd.replay_copy_settings.header.payload_bytes = sizeof(struct dmub_cmd_replay_copy_settings_data); + + // HW insts + copy_settings_data->aux_inst = replay_context->aux_inst; + copy_settings_data->digbe_inst = replay_context->digbe_inst; + copy_settings_data->digfe_inst = replay_context->digfe_inst; + + if (pipe_ctx->plane_res.dpp) + copy_settings_data->dpp_inst = pipe_ctx->plane_res.dpp->inst; + else + copy_settings_data->dpp_inst = 0; + if (pipe_ctx->stream_res.tg) + copy_settings_data->otg_inst = pipe_ctx->stream_res.tg->inst; + else + copy_settings_data->otg_inst = 0; + + copy_settings_data->dpphy_inst = link->link_enc->transmitter; + + // Misc + copy_settings_data->line_time_in_ns = replay_context->line_time_in_ns; + copy_settings_data->panel_inst = panel_inst; + copy_settings_data->debug.u32All = link->replay_settings.config.debug_flags; + copy_settings_data->pixel_deviation_per_line = link->dpcd_caps.pr_info.pixel_deviation_per_line; + copy_settings_data->max_deviation_line = link->dpcd_caps.pr_info.max_deviation_line; + copy_settings_data->smu_optimizations_en = link->replay_settings.replay_smu_opt_enable; + copy_settings_data->replay_timing_sync_supported = link->replay_settings.config.replay_timing_sync_supported; + + copy_settings_data->flags.u32All = 0; + copy_settings_data->flags.bitfields.fec_enable_status = (link->fec_state == dc_link_fec_enabled); + copy_settings_data->flags.bitfields.dsc_enable_status = (pipe_ctx->stream->timing.flags.DSC == 1); + // WA for PSRSU+DSC on specific TCON, if DSC is enabled, force PSRSU as ffu mode(full frame update) + if (((link->dpcd_caps.fec_cap.bits.FEC_CAPABLE && + !link->dc->debug.disable_fec) && + (link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT && + !link->panel_config.dsc.disable_dsc_edp && + link->dc->caps.edp_dsc_support)) && + link->dpcd_caps.sink_dev_id == DP_DEVICE_ID_38EC11 /*&& + (!memcmp(link->dpcd_caps.sink_dev_id_str, DP_SINK_DEVICE_STR_ID_1, + sizeof(DP_SINK_DEVICE_STR_ID_1)) || + !memcmp(link->dpcd_caps.sink_dev_id_str, DP_SINK_DEVICE_STR_ID_2, + sizeof(DP_SINK_DEVICE_STR_ID_2)))*/) + copy_settings_data->flags.bitfields.force_wakeup_by_tps3 = 1; + else + copy_settings_data->flags.bitfields.force_wakeup_by_tps3 = 0; + + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); + + return true; +} + +/* + * Set coasting vtotal. + */ +static void dmub_replay_set_coasting_vtotal(struct dmub_replay *dmub, + uint16_t coasting_vtotal, + uint8_t panel_inst) +{ + union dmub_rb_cmd cmd; + struct dc_context *dc = dmub->ctx; + + memset(&cmd, 0, sizeof(cmd)); + cmd.replay_set_coasting_vtotal.header.type = DMUB_CMD__REPLAY; + cmd.replay_set_coasting_vtotal.header.sub_type = DMUB_CMD__REPLAY_SET_COASTING_VTOTAL; + cmd.replay_set_coasting_vtotal.header.payload_bytes = sizeof(struct dmub_cmd_replay_set_coasting_vtotal_data); + cmd.replay_set_coasting_vtotal.replay_set_coasting_vtotal_data.coasting_vtotal = coasting_vtotal; + + dm_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); +} + +/* + * Get Replay residency from firmware. + */ +static void dmub_replay_residency(struct dmub_replay *dmub, uint8_t panel_inst, + uint32_t *residency, const bool is_start, const bool is_alpm) +{ + struct dmub_srv *srv = dmub->ctx->dmub_srv->dmub; + uint16_t param = (uint16_t)(panel_inst << 8); + + if (is_alpm) + param |= REPLAY_RESIDENCY_MODE_ALPM; + + if (is_start) + param |= REPLAY_RESIDENCY_ENABLE; + + // Send gpint command and wait for ack + dmub_srv_send_gpint_command(srv, DMUB_GPINT__REPLAY_RESIDENCY, param, 30); + + if (!is_start) + dmub_srv_get_gpint_response(srv, residency); + else + *residency = 0; +} + +static const struct dmub_replay_funcs replay_funcs = { + .replay_copy_settings = dmub_replay_copy_settings, + .replay_enable = dmub_replay_enable, + .replay_get_state = dmub_replay_get_state, + .replay_set_power_opt = dmub_replay_set_power_opt, + .replay_set_coasting_vtotal = dmub_replay_set_coasting_vtotal, + .replay_residency = dmub_replay_residency, +}; + +/* + * Construct Replay object. + */ +static void dmub_replay_construct(struct dmub_replay *replay, struct dc_context *ctx) +{ + replay->ctx = ctx; + replay->funcs = &replay_funcs; +} + +/* + * Allocate and initialize Replay object. + */ +struct dmub_replay *dmub_replay_create(struct dc_context *ctx) +{ + struct dmub_replay *replay = kzalloc(sizeof(struct dmub_replay), GFP_KERNEL); + + if (replay == NULL) { + BREAK_TO_DEBUGGER(); + return NULL; + } + + dmub_replay_construct(replay, ctx); + + return replay; +} + +/* + * Deallocate Replay object. + */ +void dmub_replay_destroy(struct dmub_replay **dmub) +{ + kfree(*dmub); + *dmub = NULL; +} diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h new file mode 100644 index 000000000000..e8385bbf51fc --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h @@ -0,0 +1,58 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef _DMUB_REPLAY_H_ +#define _DMUB_REPLAY_H_ + +#include "dc_types.h" +#include "dmub_cmd.h" +struct dc_link; +struct dmub_replay_funcs; + +struct dmub_replay { + struct dc_context *ctx; + const struct dmub_replay_funcs *funcs; +}; + +struct dmub_replay_funcs { + void (*replay_get_state)(struct dmub_replay *dmub, enum replay_state *state, + uint8_t panel_inst); + void (*replay_enable)(struct dmub_replay *dmub, bool enable, bool wait, + uint8_t panel_inst); + bool (*replay_copy_settings)(struct dmub_replay *dmub, struct dc_link *link, + struct replay_context *replay_context, uint8_t panel_inst); + void (*replay_set_power_opt)(struct dmub_replay *dmub, unsigned int power_opt, + uint8_t panel_inst); + void (*replay_set_coasting_vtotal)(struct dmub_replay *dmub, uint16_t coasting_vtotal, + uint8_t panel_inst); + void (*replay_residency)(struct dmub_replay *dmub, + uint8_t panel_inst, uint32_t *residency, const bool is_start, const bool is_alpm); +}; + +struct dmub_replay *dmub_replay_create(struct dc_context *ctx); +void dmub_replay_destroy(struct dmub_replay **dmub); + + +#endif /* _DMUB_REPLAY_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c index 42e9b6a529f6..899b25b0bad8 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c @@ -824,7 +824,7 @@ static enum dc_status build_mapped_resource( struct dc_state *context, struct dc_stream_state *stream) { - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream(&context->res_ctx, stream); if (!pipe_ctx) return DC_ERROR_UNEXPECTED; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 6c9ca43d1040..478281f2a5ba 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -219,7 +219,7 @@ static bool dce110_enable_display_power_gating( if (controller_id == underlay_idx) controller_id = CONTROLLER_ID_UNDERLAY0 - 1; - if (power_gating != PIPE_GATING_CONTROL_INIT || controller_id == 0){ + if (power_gating != PIPE_GATING_CONTROL_INIT || controller_id == 0) { bp_result = dcb->funcs->enable_disp_power_gating( dcb, controller_id + 1, cntl); @@ -777,7 +777,8 @@ void dce110_edp_wait_for_hpd_ready( dal_gpio_destroy_irq(&hpd); /* ensure that the panel is detected */ - ASSERT(edp_hpd_high); + if (!edp_hpd_high) + DC_LOG_DC("%s: wait timed out!\n", __func__); } void dce110_edp_power_control( @@ -963,7 +964,9 @@ void dce110_edp_backlight_control( return; } - if (link->panel_cntl) { + if (link->panel_cntl && !(link->dpcd_sink_ext_caps.bits.oled || + link->dpcd_sink_ext_caps.bits.hdr_aux_backlight_control == 1 || + link->dpcd_sink_ext_caps.bits.sdr_aux_backlight_control == 1)) { bool is_backlight_on = link->panel_cntl->funcs->is_panel_backlight_on(link->panel_cntl); if ((enable && is_backlight_on) || (!enable && !is_backlight_on)) { @@ -1150,6 +1153,8 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx) struct timing_generator *tg = pipe_ctx->stream_res.tg; struct dtbclk_dto_params dto_params = {0}; int dp_hpo_inst; + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(pipe_ctx->stream->link); + struct stream_encoder *stream_enc = pipe_ctx->stream_res.stream_enc; if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) { pipe_ctx->stream_res.stream_enc->funcs->stop_hdmi_info_packets( @@ -1176,7 +1181,9 @@ void dce110_disable_stream(struct pipe_ctx *pipe_ctx) dccg->funcs->set_dtbclk_dto(dccg, &dto_params); dccg->funcs->disable_symclk32_se(dccg, dp_hpo_inst); dccg->funcs->set_dpstreamclk(dccg, REFCLK, tg->inst, dp_hpo_inst); - } + } else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST && dccg->funcs->disable_symclk_se) + dccg->funcs->disable_symclk_se(dccg, stream_enc->stream_enc_inst, + link_enc->transmitter - TRANSMITTER_UNIPHY_A); if (dc->link_srv->dp_is_128b_132b_signal(pipe_ctx)) { /* TODO: This looks like a bug to me as we are disabling HPO IO when @@ -1585,6 +1592,7 @@ static enum dc_status apply_single_controller_ctx_to_hw( */ if (pipe_ctx->stream->mall_stream_config.type != SUBVP_PHANTOM) { pipe_ctx->stream->link->psr_settings.psr_feature_enabled = false; + pipe_ctx->stream->link->replay_settings.replay_feature_enabled = false; } return DC_OK; } @@ -1792,10 +1800,13 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context) hws->funcs.edp_backlight_control(edp_link_with_sink, false); } /*resume from S3, no vbios posting, no need to power down again*/ + clk_mgr_exit_optimized_pwr_state(dc, dc->clk_mgr); + power_down_all_hw_blocks(dc); disable_vga_and_power_gate_all_controllers(dc); if (edp_link_with_sink && !keep_edp_vdd_on) dc->hwss.edp_power_control(edp_link_with_sink, false); + clk_mgr_optimize_pwr_state(dc, dc->clk_mgr); } bios_set_scratch_acc_mode_change(dc->ctx->dc_bios, 1); } @@ -2013,6 +2024,10 @@ static bool should_enable_fbc(struct dc *dc, if (pipe_ctx->stream->link->psr_settings.psr_feature_enabled) return false; + /* Replay should not be enabled */ + if (pipe_ctx->stream->link->replay_settings.replay_feature_enabled) + return false; + /* Nothing to compress */ if (!pipe_ctx->plane_state) return false; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c index 46eca5a21e1c..1289b9418877 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c @@ -942,7 +942,7 @@ static enum dc_status build_mapped_resource( struct dc_state *context, struct dc_stream_state *stream) { - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream(&context->res_ctx, stream); if (!pipe_ctx) return DC_ERROR_UNEXPECTED; @@ -1119,13 +1119,15 @@ static enum dc_status dce110_add_stream_to_ctx( } static struct pipe_ctx *dce110_acquire_underlay( - struct dc_state *context, + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream) + const struct pipe_ctx *opp_head_pipe) { + struct dc_stream_state *stream = opp_head_pipe->stream; struct dc *dc = stream->ctx->dc; struct dce_hwseq *hws = dc->hwseq; - struct resource_context *res_ctx = &context->res_ctx; + struct resource_context *res_ctx = &new_ctx->res_ctx; unsigned int underlay_idx = pool->underlay_pipe_index; struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[underlay_idx]; @@ -1173,7 +1175,7 @@ static struct pipe_ctx *dce110_acquire_underlay( stream->timing.h_total, stream->timing.v_total, stream->timing.pix_clk_100hz / 10, - context->stream_count); + new_ctx->stream_count); color_space_to_black_color(dc, COLOR_SPACE_YCBCR601, &black_color); @@ -1233,7 +1235,7 @@ static const struct resource_funcs dce110_res_pool_funcs = { .panel_cntl_create = dce110_panel_cntl_create, .validate_bandwidth = dce110_validate_bandwidth, .validate_plane = dce110_validate_plane, - .acquire_idle_pipe_for_layer = dce110_acquire_underlay, + .acquire_free_pipe_as_secondary_dpp_pipe = dce110_acquire_underlay, .add_stream_to_ctx = dce110_add_stream_to_ctx, .validate_global = dce110_validate_global, .find_first_free_match_stream_enc_for_link = dce110_find_first_free_match_stream_enc_for_link diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c index 27cbb5b42c7e..6424e7f279dc 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_timing_generator.c @@ -288,7 +288,7 @@ bool dce110_timing_generator_program_timing_generator( uint32_t vsync_offset = dc_crtc_timing->v_border_bottom + dc_crtc_timing->v_front_porch; - uint32_t v_sync_start =dc_crtc_timing->v_addressable + vsync_offset; + uint32_t v_sync_start = dc_crtc_timing->v_addressable + vsync_offset; uint32_t hsync_offset = dc_crtc_timing->h_border_right + dc_crtc_timing->h_front_porch; @@ -603,7 +603,7 @@ void dce110_timing_generator_program_blanking( { uint32_t vsync_offset = timing->v_border_bottom + timing->v_front_porch; - uint32_t v_sync_start =timing->v_addressable + vsync_offset; + uint32_t v_sync_start = timing->v_addressable + vsync_offset; uint32_t hsync_offset = timing->h_border_right + timing->h_front_porch; diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c index 690caaaff019..0ef9ebb3c1e2 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_hw_sequencer.c @@ -127,7 +127,7 @@ static bool dce112_enable_display_power_gating( else cntl = ASIC_PIPE_DISABLE; - if (power_gating != PIPE_GATING_CONTROL_INIT || controller_id == 0){ + if (power_gating != PIPE_GATING_CONTROL_INIT || controller_id == 0) { bp_result = dcb->funcs->enable_disp_power_gating( dcb, controller_id + 1, cntl); diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c index e115ff91aaaa..2b20180f1a32 100644 --- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c @@ -873,7 +873,7 @@ static enum dc_status build_mapped_resource( struct dc_state *context, struct dc_stream_state *stream) { - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream(&context->res_ctx, stream); if (!pipe_ctx) return DC_ERROR_UNEXPECTED; @@ -964,7 +964,7 @@ enum dc_status resource_map_phy_clock_resources( { /* acquire new resources */ - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream( + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream( &context->res_ctx, stream); if (!pipe_ctx) diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c index 3935fd455f0f..061221394ce0 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c @@ -58,13 +58,13 @@ #include "dce/dce_i2c.h" /* TODO remove this include */ -#include "dce80_resource.h" - #ifndef mmMC_HUB_RDREQ_DMIF_LIMIT #include "gmc/gmc_7_1_d.h" #include "gmc/gmc_7_1_sh_mask.h" #endif +#include "dce80/dce80_resource.h" + #ifndef mmDP_DPHY_INTERNAL_CTRL #define mmDP_DPHY_INTERNAL_CTRL 0x1CDE #define mmDP0_DP_DPHY_INTERNAL_CTRL 0x1CDE diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c index b33955928bd0..5ca9ab8a76e8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c @@ -39,9 +39,6 @@ #define BLACK_OFFSET_RGB_Y 0x0 #define BLACK_OFFSET_CBCR 0x8000 -#define VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT 3 -#define VISUAL_CONFIRM_RECT_HEIGHT_MIN 1 -#define VISUAL_CONFIRM_RECT_HEIGHT_MAX 10 #define REG(reg)\ dpp->tf_regs->reg @@ -200,8 +197,7 @@ static void dpp1_dscl_set_lb( DITHER_EN, 0, /* Dithering enable: Disabled */ INTERLEAVE_EN, lb_params->interleave_en, /* Interleave source enable */ LB_DATA_FORMAT__ALPHA_EN, lb_params->alpha_en); /* Alpha enable */ - } - else { + } else { /* DSCL caps: pixel data processed in float format */ REG_SET_2(LB_DATA_FORMAT, 0, INTERLEAVE_EN, lb_params->interleave_en, /* Interleave source enable */ @@ -591,18 +587,6 @@ static void dpp1_dscl_set_manual_ratio_init( static void dpp1_dscl_set_recout(struct dcn10_dpp *dpp, const struct rect *recout) { - int visual_confirm_on = 0; - unsigned short visual_confirm_rect_height = VISUAL_CONFIRM_RECT_HEIGHT_DEFAULT; - - if (dpp->base.ctx->dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE) - visual_confirm_on = 1; - - /* Check bounds to ensure the VC bar height was set to a sane value */ - if ((dpp->base.ctx->dc->debug.visual_confirm_rect_height >= VISUAL_CONFIRM_RECT_HEIGHT_MIN) && - (dpp->base.ctx->dc->debug.visual_confirm_rect_height <= VISUAL_CONFIRM_RECT_HEIGHT_MAX)) { - visual_confirm_rect_height = dpp->base.ctx->dc->debug.visual_confirm_rect_height; - } - REG_SET_2(RECOUT_START, 0, /* First pixel of RECOUT in the active OTG area */ RECOUT_START_X, recout->x, @@ -613,8 +597,7 @@ static void dpp1_dscl_set_recout(struct dcn10_dpp *dpp, /* Number of RECOUT horizontal pixels */ RECOUT_WIDTH, recout->width, /* Number of RECOUT vertical lines */ - RECOUT_HEIGHT, recout->height - - visual_confirm_on * 2 * (dpp->base.inst + visual_confirm_rect_height)); + RECOUT_HEIGHT, recout->height); } /** diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index a50309039d08..9834b75f1837 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -3278,7 +3278,8 @@ void dcn10_wait_for_mpcc_disconnect( if (pipe_ctx->stream_res.opp->mpcc_disconnect_pending[mpcc_inst]) { struct hubp *hubp = get_hubp_by_inst(res_pool, mpcc_inst); - if (pipe_ctx->stream_res.tg->funcs->is_tg_enabled(pipe_ctx->stream_res.tg)) + if (pipe_ctx->stream_res.tg && + pipe_ctx->stream_res.tg->funcs->is_tg_enabled(pipe_ctx->stream_res.tg)) res_pool->mpc->funcs->wait_for_idle(res_pool->mpc, mpcc_inst); pipe_ctx->stream_res.opp->mpcc_disconnect_pending[mpcc_inst] = false; hubp->funcs->set_blank(hubp, true); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index ee08b545aaea..377f1ba1a81b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -1056,7 +1056,7 @@ void dcn10_link_encoder_disable_output( struct bp_transmitter_control cntl = { 0 }; enum bp_result result; - if (!dcn10_is_dig_enabled(enc)) { + if (enc->funcs->is_dig_enabled && !enc->funcs->is_dig_enabled(enc)) { /* OF_SKIP_POWER_DOWN_INACTIVE_ENCODER */ /*in DP_Alt_No_Connect case, we turn off the dig already, after excuation the PHY w/a sequence, not allow touch PHY any more*/ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c index 8e9384094f6d..f2f55565e98a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_mpc.c @@ -212,8 +212,9 @@ struct mpcc *mpc1_insert_plane( /* check insert_above_mpcc exist in tree->opp_list */ struct mpcc *temp_mpcc = tree->opp_list; - while (temp_mpcc && temp_mpcc->mpcc_bot != insert_above_mpcc) - temp_mpcc = temp_mpcc->mpcc_bot; + if (temp_mpcc != insert_above_mpcc) + while (temp_mpcc && temp_mpcc->mpcc_bot != insert_above_mpcc) + temp_mpcc = temp_mpcc->mpcc_bot; if (temp_mpcc == NULL) return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 4b02f8443534..9f9145742f14 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -1055,7 +1055,7 @@ static enum dc_status build_mapped_resource( struct dc_state *context, struct dc_stream_state *stream) { - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream(&context->res_ctx, stream); if (!pipe_ctx) return DC_ERROR_UNEXPECTED; @@ -1083,14 +1083,15 @@ static enum dc_status dcn10_add_stream_to_ctx( return result; } -static struct pipe_ctx *dcn10_acquire_idle_pipe_for_layer( - struct dc_state *context, +static struct pipe_ctx *dcn10_acquire_free_pipe_for_layer( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream) + const struct pipe_ctx *opp_head_pipe) { - struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - struct pipe_ctx *idle_pipe = find_idle_secondary_pipe(res_ctx, pool, head_pipe); + struct resource_context *res_ctx = &new_ctx->res_ctx; + struct pipe_ctx *head_pipe = resource_get_otg_master_for_stream(res_ctx, opp_head_pipe->stream); + struct pipe_ctx *idle_pipe = resource_find_free_secondary_pipe_legacy(res_ctx, pool, head_pipe); if (!head_pipe) { ASSERT(0); @@ -1271,7 +1272,7 @@ static const struct resource_funcs dcn10_res_pool_funcs = { .link_enc_create = dcn10_link_encoder_create, .panel_cntl_create = dcn10_panel_cntl_create, .validate_bandwidth = dcn10_validate_bandwidth, - .acquire_idle_pipe_for_layer = dcn10_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn10_acquire_free_pipe_for_layer, .validate_plane = dcn10_validate_plane, .validate_global = dcn10_validate_global, .add_stream_to_ctx = dcn10_add_stream_to_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index 4492bc2392b6..e72f15ac0048 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -1054,9 +1054,9 @@ void dcn20_blank_pixel_data( enum controller_dp_color_space test_pattern_color_space = CONTROLLER_DP_COLOR_SPACE_UDEFINED; struct pipe_ctx *odm_pipe; int odm_cnt = 1; - - int width = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; - int height = stream->timing.v_addressable + stream->timing.v_border_bottom + stream->timing.v_border_top; + int h_active = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right; + int v_active = stream->timing.v_addressable + stream->timing.v_border_bottom + stream->timing.v_border_top; + int odm_slice_width, last_odm_slice_width, offset = 0; if (stream->link->test_pattern_enabled) return; @@ -1066,8 +1066,8 @@ void dcn20_blank_pixel_data( for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) odm_cnt++; - - width = width / odm_cnt; + odm_slice_width = h_active / odm_cnt; + last_odm_slice_width = h_active - odm_slice_width * (odm_cnt - 1); if (blank) { dc->hwss.set_abm_immediate_disable(pipe_ctx); @@ -1080,51 +1080,31 @@ void dcn20_blank_pixel_data( test_pattern = CONTROLLER_DP_TEST_PATTERN_VIDEOMODE; } - dc->hwss.set_disp_pattern_generator(dc, - pipe_ctx, - test_pattern, - test_pattern_color_space, - stream->timing.display_color_depth, - &black_color, - width, - height, - 0); + odm_pipe = pipe_ctx; - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { + while (odm_pipe->next_odm_pipe) { dc->hwss.set_disp_pattern_generator(dc, odm_pipe, - dc->debug.visual_confirm != VISUAL_CONFIRM_DISABLE && blank ? - CONTROLLER_DP_TEST_PATTERN_COLORRAMP : test_pattern, + test_pattern, test_pattern_color_space, stream->timing.display_color_depth, &black_color, - width, - height, - 0); + odm_slice_width, + v_active, + offset); + offset += odm_slice_width; + odm_pipe = odm_pipe->next_odm_pipe; } - if (!blank && dc->debug.enable_single_display_2to1_odm_policy) { - /* when exiting dynamic ODM need to reinit DPG state for unused pipes */ - struct pipe_ctx *old_odm_pipe = dc->current_state->res_ctx.pipe_ctx[pipe_ctx->pipe_idx].next_odm_pipe; - - odm_pipe = pipe_ctx->next_odm_pipe; - - while (old_odm_pipe) { - if (!odm_pipe || old_odm_pipe->pipe_idx != odm_pipe->pipe_idx) - dc->hwss.set_disp_pattern_generator(dc, - old_odm_pipe, - CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, - CONTROLLER_DP_COLOR_SPACE_UDEFINED, - COLOR_DEPTH_888, - NULL, - 0, - 0, - 0); - old_odm_pipe = old_odm_pipe->next_odm_pipe; - if (odm_pipe) - odm_pipe = odm_pipe->next_odm_pipe; - } - } + dc->hwss.set_disp_pattern_generator(dc, + odm_pipe, + test_pattern, + test_pattern_color_space, + stream->timing.display_color_depth, + &black_color, + last_odm_slice_width, + v_active, + offset); if (!blank) if (stream_res->abm) { @@ -1266,20 +1246,21 @@ void dcn20_pipe_control_lock( } if (flip_immediate && lock) { - const int TIMEOUT_FOR_FLIP_PENDING = 100000; + const int TIMEOUT_FOR_FLIP_PENDING_US = 100000; + unsigned int polling_interval_us = 1; int i; temp_pipe = pipe; while (temp_pipe) { if (temp_pipe->plane_state && temp_pipe->plane_state->flip_immediate) { - for (i = 0; i < TIMEOUT_FOR_FLIP_PENDING; ++i) { + for (i = 0; i < TIMEOUT_FOR_FLIP_PENDING_US / polling_interval_us; ++i) { if (!temp_pipe->plane_res.hubp->funcs->hubp_is_flip_pending(temp_pipe->plane_res.hubp)) break; - udelay(1); + udelay(polling_interval_us); } /* no reason it should take this long for immediate flips */ - ASSERT(i != TIMEOUT_FOR_FLIP_PENDING); + ASSERT(i != TIMEOUT_FOR_FLIP_PENDING_US); } temp_pipe = temp_pipe->bottom_pipe; } @@ -1580,17 +1561,6 @@ static void dcn20_update_dchubp_dpp( || plane_state->update_flags.bits.global_alpha_change || plane_state->update_flags.bits.per_pixel_alpha_change) { // MPCC inst is equal to pipe index in practice - int mpcc_inst = hubp->inst; - int opp_inst; - int opp_count = dc->res_pool->pipe_count; - - for (opp_inst = 0; opp_inst < opp_count; opp_inst++) { - if (dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst]) { - dc->res_pool->mpc->funcs->wait_for_idle(dc->res_pool->mpc, mpcc_inst); - dc->res_pool->opps[opp_inst]->mpcc_disconnect_pending[mpcc_inst] = false; - break; - } - } hws->funcs.update_mpcc(dc, pipe_ctx); } @@ -1634,6 +1604,7 @@ static void dcn20_update_dchubp_dpp( if (pipe_ctx->update_flags.bits.enable || pipe_ctx->update_flags.bits.opp_changed || pipe_ctx->update_flags.bits.plane_changed || pipe_ctx->stream->update_flags.bits.gamut_remap + || plane_state->update_flags.bits.gamut_remap_change || pipe_ctx->stream->update_flags.bits.out_csc) { /* dpp/cm gamut remap*/ dc->hwss.program_gamut_remap(pipe_ctx); @@ -1717,11 +1688,16 @@ static void dcn20_program_pipe( struct dc_state *context) { struct dce_hwseq *hws = dc->hwseq; - /* Only need to unblank on top pipe */ - if ((pipe_ctx->update_flags.bits.enable || pipe_ctx->stream->update_flags.bits.abm_level) - && !pipe_ctx->top_pipe && !pipe_ctx->prev_odm_pipe) - hws->funcs.blank_pixel_data(dc, pipe_ctx, !pipe_ctx->plane_state->visible); + /* Only need to unblank on top pipe */ + if (resource_is_pipe_type(pipe_ctx, OTG_MASTER)) { + if (pipe_ctx->update_flags.bits.enable || + pipe_ctx->update_flags.bits.odm || + pipe_ctx->stream->update_flags.bits.abm_level) + hws->funcs.blank_pixel_data(dc, pipe_ctx, + !pipe_ctx->plane_state || + !pipe_ctx->plane_state->visible); + } /* Only update TG on top pipe */ if (pipe_ctx->update_flags.bits.global_sync && !pipe_ctx->top_pipe @@ -1949,7 +1925,8 @@ void dcn20_post_unlock_program_front_end( struct dc_state *context) { int i; - const unsigned int TIMEOUT_FOR_PIPE_ENABLE_MS = 100; + const unsigned int TIMEOUT_FOR_PIPE_ENABLE_US = 100000; + unsigned int polling_interval_us = 1; struct dce_hwseq *hwseq = dc->hwseq; DC_LOGGER_INIT(dc->ctx->logger); @@ -1971,10 +1948,9 @@ void dcn20_post_unlock_program_front_end( pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) { struct hubp *hubp = pipe->plane_res.hubp; int j = 0; - - for (j = 0; j < TIMEOUT_FOR_PIPE_ENABLE_MS*1000 + for (j = 0; j < TIMEOUT_FOR_PIPE_ENABLE_US / polling_interval_us && hubp->funcs->hubp_is_flip_pending(hubp); j++) - udelay(1); + udelay(polling_interval_us); } } @@ -2123,6 +2099,15 @@ void dcn20_optimize_bandwidth( if (hubbub->funcs->program_compbuf_size) hubbub->funcs->program_compbuf_size(hubbub, context->bw_ctx.bw.dcn.compbuf_size_kb, true); + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) { + dc_dmub_srv_p_state_delegate(dc, + true, context); + context->bw_ctx.bw.dcn.clk.p_state_change_support = true; + dc->clk_mgr->clks.fw_based_mclk_switching = true; + } else { + dc->clk_mgr->clks.fw_based_mclk_switching = false; + } + dc->clk_mgr->funcs->update_clocks( dc->clk_mgr, context, @@ -2707,6 +2692,8 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) struct dce_hwseq *hws = dc->hwseq; unsigned int k1_div = PIXEL_RATE_DIV_NA; unsigned int k2_div = PIXEL_RATE_DIV_NA; + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(pipe_ctx->stream->link); + struct stream_encoder *stream_enc = pipe_ctx->stream_res.stream_enc; if (dc->link_srv->dp_is_128b_132b_signal(pipe_ctx)) { if (dc->hwseq->funcs.setup_hpo_hw_control) @@ -2726,7 +2713,9 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx) dto_params.timing = &pipe_ctx->stream->timing; dto_params.ref_dtbclk_khz = dc->clk_mgr->funcs->get_dtb_ref_clk_frequency(dc->clk_mgr); dccg->funcs->set_dtbclk_dto(dccg, &dto_params); - } + } else if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST && dccg->funcs->enable_symclk_se) + dccg->funcs->enable_symclk_se(dccg, + stream_enc->stream_enc_inst, link_enc->transmitter - TRANSMITTER_UNIPHY_A); if (hws->funcs.calculate_dccg_k1_k2_values && dc->res_pool->dccg->funcs->set_pixel_rate_div) { hws->funcs.calculate_dccg_k1_k2_values(pipe_ctx, &k1_div, &k2_div); diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 4cc8de2627ce..d587f807dfd7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -712,7 +712,7 @@ static const struct dc_debug_options debug_defaults_drv = { .timing_trace = false, .clock_trace = true, .disable_pplib_clock_request = true, - .pipe_split_policy = MPC_SPLIT_DYNAMIC, + .pipe_split_policy = MPC_SPLIT_AVOID_MULT_DISP, .force_single_disp_pipe_split = false, .disable_dcc = DCC_ENABLE, .vsr_support = true, @@ -1294,7 +1294,7 @@ static enum dc_status build_pipe_hw_param(struct pipe_ctx *pipe_ctx) enum dc_status dcn20_build_mapped_resource(const struct dc *dc, struct dc_state *context, struct dc_stream_state *stream) { enum dc_status status = DC_OK; - struct pipe_ctx *pipe_ctx = resource_get_head_pipe_for_stream(&context->res_ctx, stream); + struct pipe_ctx *pipe_ctx = resource_get_otg_master_for_stream(&context->res_ctx, stream); if (!pipe_ctx) return DC_ERROR_UNEXPECTED; @@ -1948,7 +1948,7 @@ int dcn20_validate_apply_pipe_split_flags( v->ODMCombineEnablePerState[vlevel][pipe_plane]; if (v->ODMCombineEnabled[pipe_plane] == dm_odm_combine_mode_disabled) { - if (get_num_mpc_splits(pipe) == 1) { + if (resource_get_num_mpc_splits(pipe) == 1) { /*If need split for mpc but 2 way split already*/ if (split[i] == 4) split[i] = 2; /* 2 -> 4 MPC */ @@ -1956,7 +1956,7 @@ int dcn20_validate_apply_pipe_split_flags( split[i] = 0; /* 2 -> 2 MPC */ else if (pipe->top_pipe && pipe->top_pipe->plane_state == pipe->plane_state) merge[i] = true; /* 2 -> 1 MPC */ - } else if (get_num_mpc_splits(pipe) == 3) { + } else if (resource_get_num_mpc_splits(pipe) == 3) { /*If need split for mpc but 4 way split already*/ if (split[i] == 2 && ((pipe->top_pipe && !pipe->top_pipe->top_pipe) || !pipe->bottom_pipe)) { @@ -1965,7 +1965,7 @@ int dcn20_validate_apply_pipe_split_flags( pipe->top_pipe->plane_state == pipe->plane_state) merge[i] = true; /* 4 -> 1 MPC */ split[i] = 0; - } else if (get_num_odm_splits(pipe)) { + } else if (resource_get_num_odm_splits(pipe)) { /* ODM -> MPC transition */ if (pipe->prev_odm_pipe) { split[i] = 0; @@ -1973,7 +1973,7 @@ int dcn20_validate_apply_pipe_split_flags( } } } else { - if (get_num_odm_splits(pipe) == 1) { + if (resource_get_num_odm_splits(pipe) == 1) { /*If need split for odm but 2 way split already*/ if (split[i] == 4) split[i] = 2; /* 2 -> 4 ODM */ @@ -1983,7 +1983,7 @@ int dcn20_validate_apply_pipe_split_flags( ASSERT(0); /* NOT expected yet */ merge[i] = true; /* exit ODM */ } - } else if (get_num_odm_splits(pipe) == 3) { + } else if (resource_get_num_odm_splits(pipe) == 3) { /*If need split for odm but 4 way split already*/ if (split[i] == 2 && ((pipe->prev_odm_pipe && !pipe->prev_odm_pipe->prev_odm_pipe) || !pipe->next_odm_pipe)) { @@ -1993,7 +1993,7 @@ int dcn20_validate_apply_pipe_split_flags( merge[i] = true; /* exit ODM */ } split[i] = 0; - } else if (get_num_mpc_splits(pipe)) { + } else if (resource_get_num_mpc_splits(pipe)) { /* MPC -> ODM transition */ ASSERT(0); /* NOT expected yet */ if (pipe->top_pipe && pipe->top_pipe->plane_state == pipe->plane_state) { @@ -2147,31 +2147,31 @@ bool dcn20_validate_bandwidth(struct dc *dc, struct dc_state *context, return voltage_supported; } -struct pipe_ctx *dcn20_acquire_idle_pipe_for_layer( - struct dc_state *state, +struct pipe_ctx *dcn20_acquire_free_pipe_for_layer( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream) + const struct pipe_ctx *opp_head) { - struct resource_context *res_ctx = &state->res_ctx; - struct pipe_ctx *head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - struct pipe_ctx *idle_pipe = find_idle_secondary_pipe(res_ctx, pool, head_pipe); + struct resource_context *res_ctx = &new_ctx->res_ctx; + struct pipe_ctx *otg_master = resource_get_otg_master_for_stream(res_ctx, opp_head->stream); + struct pipe_ctx *sec_dpp_pipe = resource_find_free_secondary_pipe_legacy(res_ctx, pool, otg_master); - if (!head_pipe) - ASSERT(0); + ASSERT(otg_master); - if (!idle_pipe) + if (!sec_dpp_pipe) return NULL; - idle_pipe->stream = head_pipe->stream; - idle_pipe->stream_res.tg = head_pipe->stream_res.tg; - idle_pipe->stream_res.opp = head_pipe->stream_res.opp; + sec_dpp_pipe->stream = opp_head->stream; + sec_dpp_pipe->stream_res.tg = opp_head->stream_res.tg; + sec_dpp_pipe->stream_res.opp = opp_head->stream_res.opp; - idle_pipe->plane_res.hubp = pool->hubps[idle_pipe->pipe_idx]; - idle_pipe->plane_res.ipp = pool->ipps[idle_pipe->pipe_idx]; - idle_pipe->plane_res.dpp = pool->dpps[idle_pipe->pipe_idx]; - idle_pipe->plane_res.mpcc_inst = pool->dpps[idle_pipe->pipe_idx]->inst; + sec_dpp_pipe->plane_res.hubp = pool->hubps[sec_dpp_pipe->pipe_idx]; + sec_dpp_pipe->plane_res.ipp = pool->ipps[sec_dpp_pipe->pipe_idx]; + sec_dpp_pipe->plane_res.dpp = pool->dpps[sec_dpp_pipe->pipe_idx]; + sec_dpp_pipe->plane_res.mpcc_inst = pool->dpps[sec_dpp_pipe->pipe_idx]->inst; - return idle_pipe; + return sec_dpp_pipe; } bool dcn20_get_dcc_compression_cap(const struct dc *dc, @@ -2216,7 +2216,7 @@ static const struct resource_funcs dcn20_res_pool_funcs = { .link_enc_create = dcn20_link_encoder_create, .panel_cntl_create = dcn20_panel_cntl_create, .validate_bandwidth = dcn20_validate_bandwidth, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn20_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h index da0241e8c255..6d1a8924e57b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.h @@ -58,10 +58,11 @@ unsigned int dcn20_calc_max_scaled_time( enum mmhubbub_wbif_mode mode, unsigned int urgent_watermark); -struct pipe_ctx *dcn20_acquire_idle_pipe_for_layer( - struct dc_state *state, +struct pipe_ctx *dcn20_acquire_free_pipe_for_layer( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream); + const struct pipe_ctx *opp_head_pipe); struct stream_encoder *dcn20_stream_encoder_create( enum engine_id eng_id, diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c index fdba8a9f5c30..2dc4d2c1410b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c @@ -992,14 +992,15 @@ static struct hubp *dcn201_hubp_create( return NULL; } -static struct pipe_ctx *dcn201_acquire_idle_pipe_for_layer( - struct dc_state *context, +static struct pipe_ctx *dcn201_acquire_free_pipe_for_layer( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream) + const struct pipe_ctx *opp_head_pipe) { - struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *head_pipe = resource_get_head_pipe_for_stream(res_ctx, stream); - struct pipe_ctx *idle_pipe = find_idle_secondary_pipe(res_ctx, pool, head_pipe); + struct resource_context *res_ctx = &new_ctx->res_ctx; + struct pipe_ctx *head_pipe = resource_get_otg_master_for_stream(res_ctx, opp_head_pipe->stream); + struct pipe_ctx *idle_pipe = resource_find_free_secondary_pipe_legacy(res_ctx, pool, head_pipe); if (!head_pipe) ASSERT(0); @@ -1067,7 +1068,7 @@ static struct resource_funcs dcn201_res_pool_funcs = { .add_stream_to_ctx = dcn20_add_stream_to_ctx, .add_dsc_to_stream_resource = NULL, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, - .acquire_idle_pipe_for_layer = dcn201_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn201_acquire_free_pipe_for_layer, .populate_dml_writeback_from_context = dcn201_populate_dml_writeback_from_context, .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.c index 33fc9aa8621b..d07c04458d31 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.c @@ -43,7 +43,7 @@ #define DC_LOGGER \ dccg->ctx->logger -void dccg21_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) +static void dccg21_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.h index e44a37491c1e..b7efa777ec73 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_dccg.h @@ -32,6 +32,5 @@ struct dccg *dccg21_create( const struct dccg_shift *dccg_shift, const struct dccg_mask *dccg_mask); -void dccg21_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk); #endif /* __DCN21_DCCG_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index d693ea42d033..d1a25fe6c44f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -660,6 +660,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, .ilr = { .optimize_edp_link_rate = true, @@ -854,8 +855,8 @@ bool dcn21_fast_validate_bw(struct dc *dc, /* We only support full screen mpo with ODM */ if (vba->ODMCombineEnabled[vba->pipe_plane[pipe_idx]] != dm_odm_combine_mode_disabled && pipe->plane_state && mpo_pipe - && memcmp(&mpo_pipe->plane_res.scl_data.recout, - &pipe->plane_res.scl_data.recout, + && memcmp(&mpo_pipe->plane_state->clip_rect, + &pipe->stream->src, sizeof(struct rect)) != 0) { ASSERT(mpo_pipe->plane_state != pipe->plane_state); goto validate_fail; @@ -1387,7 +1388,7 @@ static const struct resource_funcs dcn21_res_pool_funcs = { .add_stream_to_ctx = dcn20_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .populate_dml_writeback_from_context = dcn20_populate_dml_writeback_from_context, .patch_unknown_plane_state = dcn21_patch_unknown_plane_state, .set_mcif_arb_params = dcn20_set_mcif_arb_params, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c index e5b7ef7422b8..50dc83404644 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c @@ -357,8 +357,11 @@ void dpp3_set_cursor_attributes( int cur_rom_en = 0; if (color_format == CURSOR_MODE_COLOR_PRE_MULTIPLIED_ALPHA || - color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) - cur_rom_en = 1; + color_format == CURSOR_MODE_COLOR_UN_PRE_MULTIPLIED_ALPHA) { + if (cursor_attributes->attribute_flags.bits.ENABLE_CURSOR_DEGAMMA) { + cur_rom_en = 1; + } + } REG_UPDATE_3(CURSOR0_CONTROL, CUR0_MODE, color_format, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c index bf8864bc8a99..255713ec29bb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c @@ -461,6 +461,11 @@ void dcn30_init_hw(struct dc *dc) REG_SET_2(ODM_MEM_PWR_CTRL3, 0, ODM_MEM_UNASSIGNED_PWR_MODE, 3, ODM_MEM_VBLANK_PWR_MODE, 1); } + if (dc->debug.enable_mem_low_power.bits.vga) { + // Power down VGA memory + REG_UPDATE(MMHUBBUB_MEM_PWR_CNTL, VGA_MEM_PWR_FORCE, 1); + } + if (dc->ctx->dc_bios->fw_info_valid) { res_pool->ref_clocks.xtalin_clock_inKhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency; @@ -949,13 +954,53 @@ void dcn30_set_disp_pattern_generator(const struct dc *dc, } void dcn30_prepare_bandwidth(struct dc *dc, - struct dc_state *context) + struct dc_state *context) { + bool p_state_change_support = context->bw_ctx.bw.dcn.clk.p_state_change_support; + /* Any transition into an FPO config should disable MCLK switching first to avoid + * driver and FW P-State synchronization issues. + */ + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) { + dc->optimized_required = true; + context->bw_ctx.bw.dcn.clk.p_state_change_support = false; + } + if (dc->clk_mgr->dc_mode_softmax_enabled) if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 && context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000) dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz); dcn20_prepare_bandwidth(dc, context); + /* + * enabled -> enabled: do not disable + * enabled -> disabled: disable + * disabled -> enabled: don't care + * disabled -> disabled: don't care + */ + if (!context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching) + dc_dmub_srv_p_state_delegate(dc, false, context); + + if (context->bw_ctx.bw.dcn.clk.fw_based_mclk_switching || dc->clk_mgr->clks.fw_based_mclk_switching) { + /* After disabling P-State, restore the original value to ensure we get the correct P-State + * on the next optimize. */ + context->bw_ctx.bw.dcn.clk.p_state_change_support = p_state_change_support; + } } +void dcn30_set_static_screen_control(struct pipe_ctx **pipe_ctx, + int num_pipes, const struct dc_static_screen_params *params) +{ + unsigned int i; + unsigned int triggers = 0; + + if (params->triggers.surface_update) + triggers |= 0x100; + if (params->triggers.cursor_update) + triggers |= 0x8; + if (params->triggers.force_trigger) + triggers |= 0x1; + + for (i = 0; i < num_pipes; i++) + pipe_ctx[i]->stream_res.tg->funcs->set_static_screen_control(pipe_ctx[i]->stream_res.tg, + triggers, params->num_frames); +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h index a24a8e33a3d2..ce19c54097f8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h @@ -87,5 +87,7 @@ void dcn30_set_hubp_blank(const struct dc *dc, void dcn30_prepare_bandwidth(struct dc *dc, struct dc_state *context); +void dcn30_set_static_screen_control(struct pipe_ctx **pipe_ctx, + int num_pipes, const struct dc_static_screen_params *params); #endif /* __DC_HWSS_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c index 3d19acaa12f3..0de8b2783cf6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c @@ -64,7 +64,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = { .update_bandwidth = dcn20_update_bandwidth, .set_drr = dcn10_set_drr, .get_position = dcn10_get_position, - .set_static_screen_control = dcn10_set_static_screen_control, + .set_static_screen_control = dcn30_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, .set_avmute = dcn30_set_avmute, .log_hw_state = dcn10_log_hw_state, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c index dfb8f62765f2..5bf4d0aa6230 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.c @@ -215,7 +215,7 @@ void optc3_set_odm_bypass(struct timing_generator *optc, optc1->opp_count = 1; } -static void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, +void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, struct dc_crtc_timing *timing) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -293,7 +293,7 @@ static void optc3_set_timing_double_buffer(struct timing_generator *optc, bool e OTG_DRR_TIMING_DBUF_UPDATE_MODE, mode); } -static void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *optc) +void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h index fb06dc9a4893..d3a056c12b0d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_optc.h @@ -351,6 +351,9 @@ void optc3_set_timing_db_mode(struct timing_generator *optc, bool enable); void optc3_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); +void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_cnt, + struct dc_crtc_timing *timing); +void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *optc); void optc3_tg_init(struct timing_generator *optc); void optc3_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max); #endif /* __DC_OPTC_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index abe4c12a10b5..88c0b24a3249 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -733,6 +733,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, }; @@ -1705,8 +1706,8 @@ noinline bool dcn30_internal_validate_bw( /* We only support full screen mpo with ODM */ if (vba->ODMCombineEnabled[vba->pipe_plane[pipe_idx]] != dm_odm_combine_mode_disabled && pipe->plane_state && mpo_pipe - && memcmp(&mpo_pipe->plane_res.scl_data.recout, - &pipe->plane_res.scl_data.recout, + && memcmp(&mpo_pipe->plane_state->clip_rect, + &pipe->stream->src, sizeof(struct rect)) != 0) { ASSERT(mpo_pipe->plane_state != pipe->plane_state); goto validate_fail; @@ -2062,7 +2063,8 @@ bool dcn30_validate_bandwidth(struct dc *dc, } DC_FP_START(); - dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); + if (dc->res_pool->funcs->calculate_wm_and_dlg) + dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); DC_FP_END(); BW_VAL_TRACE_END_WATERMARKS(); @@ -2214,7 +2216,7 @@ static const struct resource_funcs dcn30_res_pool_funcs = { .calculate_wm_and_dlg = dcn30_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn30_update_soc_for_wm_a, .populate_dml_pipes = dcn30_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/Makefile b/drivers/gpu/drm/amd/display/dc/dcn301/Makefile index 7aa628c21973..9002cb10a6ae 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dcn301/Makefile @@ -11,7 +11,8 @@ # Makefile for dcn30. DCN301 = dcn301_init.o dcn301_resource.o dcn301_dccg.o \ - dcn301_dio_link_encoder.o dcn301_hwseq.o dcn301_panel_cntl.o dcn301_hubbub.o + dcn301_dio_link_encoder.o dcn301_hwseq.o dcn301_panel_cntl.o dcn301_hubbub.o \ + dcn301_optc.o AMD_DAL_DCN301 = $(addprefix $(AMDDALPATH)/dc/dcn301/,$(DCN301)) diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c index 257df8660b4c..61205cdbe2d5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c @@ -75,6 +75,7 @@ static const struct hw_sequencer_funcs dcn301_funcs = { .get_hw_state = dcn10_get_hw_state, .clear_status_bits = dcn10_clear_status_bits, .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, + .edp_backlight_control = dce110_edp_backlight_control, .edp_power_control = dce110_edp_power_control, .edp_wait_for_hpd_ready = dce110_edp_wait_for_hpd_ready, .set_cursor_position = dcn10_set_cursor_position, diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_optc.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_optc.c new file mode 100644 index 000000000000..b3cfcb887905 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_optc.c @@ -0,0 +1,185 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "reg_helper.h" +#include "dcn301_optc.h" +#include "dc.h" +#include "dcn_calc_math.h" +#include "dc_dmub_srv.h" + +#include "dml/dcn30/dcn30_fpu.h" +#include "dc_trace.h" + +#define REG(reg)\ + optc1->tg_regs->reg + +#define CTX \ + optc1->base.ctx + +#undef FN +#define FN(reg_name, field_name) \ + optc1->tg_shift->field_name, optc1->tg_mask->field_name + + +/** + * optc301_set_drr() - Program dynamic refresh rate registers m_OTGx_OTG_V_TOTAL_*. + * + * @optc: timing_generator instance. + * @params: parameters used for Dynamic Refresh Rate. + */ +void optc301_set_drr( + struct timing_generator *optc, + const struct drr_params *params) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + + if (params != NULL && + params->vertical_total_max > 0 && + params->vertical_total_min > 0) { + + if (params->vertical_total_mid != 0) { + + REG_SET(OTG_V_TOTAL_MID, 0, + OTG_V_TOTAL_MID, params->vertical_total_mid - 1); + + REG_UPDATE_2(OTG_V_TOTAL_CONTROL, + OTG_VTOTAL_MID_REPLACING_MAX_EN, 1, + OTG_VTOTAL_MID_FRAME_NUM, + (uint8_t)params->vertical_total_mid_frame_num); + + } + + optc->funcs->set_vtotal_min_max(optc, params->vertical_total_min - 1, params->vertical_total_max - 1); + + REG_UPDATE_5(OTG_V_TOTAL_CONTROL, + OTG_V_TOTAL_MIN_SEL, 1, + OTG_V_TOTAL_MAX_SEL, 1, + OTG_FORCE_LOCK_ON_EVENT, 0, + OTG_SET_V_TOTAL_MIN_MASK_EN, 0, + OTG_SET_V_TOTAL_MIN_MASK, 0); + // Setup manual flow control for EOF via TRIG_A + optc->funcs->setup_manual_trigger(optc); + + } else { + REG_UPDATE_4(OTG_V_TOTAL_CONTROL, + OTG_SET_V_TOTAL_MIN_MASK, 0, + OTG_V_TOTAL_MIN_SEL, 0, + OTG_V_TOTAL_MAX_SEL, 0, + OTG_FORCE_LOCK_ON_EVENT, 0); + + optc->funcs->set_vtotal_min_max(optc, 0, 0); + } +} + + +void optc301_setup_manual_trigger(struct timing_generator *optc) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + + REG_SET_8(OTG_TRIGA_CNTL, 0, + OTG_TRIGA_SOURCE_SELECT, 21, + OTG_TRIGA_SOURCE_PIPE_SELECT, optc->inst, + OTG_TRIGA_RISING_EDGE_DETECT_CNTL, 1, + OTG_TRIGA_FALLING_EDGE_DETECT_CNTL, 0, + OTG_TRIGA_POLARITY_SELECT, 0, + OTG_TRIGA_FREQUENCY_SELECT, 0, + OTG_TRIGA_DELAY, 0, + OTG_TRIGA_CLEAR, 1); +} + +static struct timing_generator_funcs dcn30_tg_funcs = { + .validate_timing = optc1_validate_timing, + .program_timing = optc1_program_timing, + .setup_vertical_interrupt0 = optc1_setup_vertical_interrupt0, + .setup_vertical_interrupt1 = optc1_setup_vertical_interrupt1, + .setup_vertical_interrupt2 = optc1_setup_vertical_interrupt2, + .program_global_sync = optc1_program_global_sync, + .enable_crtc = optc2_enable_crtc, + .disable_crtc = optc1_disable_crtc, + /* used by enable_timing_synchronization. Not need for FPGA */ + .is_counter_moving = optc1_is_counter_moving, + .get_position = optc1_get_position, + .get_frame_count = optc1_get_vblank_counter, + .get_scanoutpos = optc1_get_crtc_scanoutpos, + .get_otg_active_size = optc1_get_otg_active_size, + .set_early_control = optc1_set_early_control, + /* used by enable_timing_synchronization. Not need for FPGA */ + .wait_for_state = optc1_wait_for_state, + .set_blank_color = optc3_program_blank_color, + .did_triggered_reset_occur = optc1_did_triggered_reset_occur, + .triplebuffer_lock = optc3_triplebuffer_lock, + .triplebuffer_unlock = optc2_triplebuffer_unlock, + .enable_reset_trigger = optc1_enable_reset_trigger, + .enable_crtc_reset = optc1_enable_crtc_reset, + .disable_reset_trigger = optc1_disable_reset_trigger, + .lock = optc3_lock, + .unlock = optc1_unlock, + .lock_doublebuffer_enable = optc3_lock_doublebuffer_enable, + .lock_doublebuffer_disable = optc3_lock_doublebuffer_disable, + .enable_optc_clock = optc1_enable_optc_clock, + .set_drr = optc301_set_drr, + .get_last_used_drr_vtotal = optc2_get_last_used_drr_vtotal, + .set_vtotal_min_max = optc3_set_vtotal_min_max, + .set_static_screen_control = optc1_set_static_screen_control, + .program_stereo = optc1_program_stereo, + .is_stereo_left_eye = optc1_is_stereo_left_eye, + .tg_init = optc3_tg_init, + .is_tg_enabled = optc1_is_tg_enabled, + .is_optc_underflow_occurred = optc1_is_optc_underflow_occurred, + .clear_optc_underflow = optc1_clear_optc_underflow, + .setup_global_swap_lock = NULL, + .get_crc = optc1_get_crc, + .configure_crc = optc2_configure_crc, + .set_dsc_config = optc3_set_dsc_config, + .get_dsc_status = optc2_get_dsc_status, + .set_dwb_source = NULL, + .set_odm_bypass = optc3_set_odm_bypass, + .set_odm_combine = optc3_set_odm_combine, + .get_optc_source = optc2_get_optc_source, + .set_out_mux = optc3_set_out_mux, + .set_drr_trigger_window = optc3_set_drr_trigger_window, + .set_vtotal_change_limit = optc3_set_vtotal_change_limit, + .set_gsl = optc2_set_gsl, + .set_gsl_source_select = optc2_set_gsl_source_select, + .set_vtg_params = optc1_set_vtg_params, + .program_manual_trigger = optc2_program_manual_trigger, + .setup_manual_trigger = optc301_setup_manual_trigger, + .get_hw_timing = optc1_get_hw_timing, + .wait_drr_doublebuffer_pending_clear = optc3_wait_drr_doublebuffer_pending_clear, +}; + +void dcn301_timing_generator_init(struct optc *optc1) +{ + optc1->base.funcs = &dcn30_tg_funcs; + + optc1->max_h_total = optc1->tg_mask->OTG_H_TOTAL + 1; + optc1->max_v_total = optc1->tg_mask->OTG_V_TOTAL + 1; + + optc1->min_h_blank = 32; + optc1->min_v_blank = 3; + optc1->min_v_blank_interlace = 5; + optc1->min_h_sync_width = 4; + optc1->min_v_sync_width = 1; +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_optc.h b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_optc.h new file mode 100644 index 000000000000..b49585682a15 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_optc.h @@ -0,0 +1,36 @@ +/* + * Copyright 2020 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef __DC_OPTC_DCN301_H__ +#define __DC_OPTC_DCN301_H__ + +#include "dcn20/dcn20_optc.h" +#include "dcn30/dcn30_optc.h" + +void dcn301_timing_generator_init(struct optc *optc1); +void optc301_setup_manual_trigger(struct timing_generator *optc); +void optc301_set_drr(struct timing_generator *optc, const struct drr_params *params); + +#endif /* __DC_OPTC_DCN301_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c index 3485fbb1093e..79d6697d13b6 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c @@ -42,7 +42,7 @@ #include "dcn30/dcn30_hubp.h" #include "irq/dcn30/irq_service_dcn30.h" #include "dcn30/dcn30_dpp.h" -#include "dcn30/dcn30_optc.h" +#include "dcn301/dcn301_optc.h" #include "dcn20/dcn20_hwseq.h" #include "dcn30/dcn30_hwseq.h" #include "dce110/dce110_hw_sequencer.h" @@ -855,7 +855,7 @@ static struct timing_generator *dcn301_timing_generator_create( tgn10->tg_shift = &optc_shift; tgn10->tg_mask = &optc_mask; - dcn30_timing_generator_init(tgn10); + dcn301_timing_generator_init(tgn10); return &tgn10->base; } @@ -1379,7 +1379,7 @@ static struct resource_funcs dcn301_res_pool_funcs = { .calculate_wm_and_dlg = dcn301_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn30_update_soc_for_wm_a, .populate_dml_pipes = dcn30_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, @@ -1425,9 +1425,9 @@ static bool dcn301_resource_construct( dc->caps.max_cursor_size = 256; dc->caps.min_horizontal_blanking_period = 80; dc->caps.dmdata_alloc_size = 2048; - dc->caps.max_slave_planes = 1; - dc->caps.max_slave_yuv_planes = 1; - dc->caps.max_slave_rgb_planes = 1; + dc->caps.max_slave_planes = 2; + dc->caps.max_slave_yuv_planes = 2; + dc->caps.max_slave_rgb_planes = 2; dc->caps.is_apu = true; dc->caps.post_blend_color_processing = true; dc->caps.force_dp_tps4_for_cp2520 = true; diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c index 5ad6a22ee47d..447abcd593be 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c @@ -103,6 +103,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, }; @@ -1135,7 +1136,7 @@ static struct resource_funcs dcn302_res_pool_funcs = { .calculate_wm_and_dlg = dcn30_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn30_update_soc_for_wm_a, .populate_dml_pipes = dcn30_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 45956ef6f3f9..adf4989177f7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -65,7 +65,7 @@ static const struct dc_debug_options debug_defaults_drv = { .timing_trace = false, .clock_trace = true, .disable_pplib_clock_request = true, - .pipe_split_policy = MPC_SPLIT_DYNAMIC, + .pipe_split_policy = MPC_SPLIT_AVOID, .force_single_disp_pipe_split = false, .disable_dcc = DCC_ENABLE, .vsr_support = true, @@ -85,6 +85,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, }; @@ -1061,7 +1062,7 @@ static struct resource_funcs dcn303_res_pool_funcs = { .calculate_wm_and_dlg = dcn30_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn30_update_soc_for_wm_a, .populate_dml_pipes = dcn30_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c index 65c1d754e2d6..8664f0c4c9b7 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c @@ -84,7 +84,8 @@ static enum phyd32clk_clock_source get_phy_mux_symclk( struct dcn_dccg *dccg_dcn, enum phyd32clk_clock_source src) { - if (dccg_dcn->base.ctx->asic_id.hw_internal_rev == YELLOW_CARP_B0) { + if (dccg_dcn->base.ctx->asic_id.chip_family == FAMILY_YELLOW_CARP && + dccg_dcn->base.ctx->asic_id.hw_internal_rev == YELLOW_CARP_B0) { if (src == PHYD32CLKC) src = PHYD32CLKF; if (src == PHYD32CLKD) @@ -284,19 +285,11 @@ void dccg31_enable_symclk32_le( /* select one of the PHYD32CLKs as the source for symclk32_le */ switch (hpo_le_inst) { case 0: - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, - SYMCLK32_LE0_GATE_DISABLE, 1, - SYMCLK32_ROOT_LE0_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_LE_CNTL, SYMCLK32_LE0_SRC_SEL, phyd32clk, SYMCLK32_LE0_EN, 1); break; case 1: - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, - SYMCLK32_LE1_GATE_DISABLE, 1, - SYMCLK32_ROOT_LE1_GATE_DISABLE, 1); REG_UPDATE_2(SYMCLK32_LE_CNTL, SYMCLK32_LE1_SRC_SEL, phyd32clk, SYMCLK32_LE1_EN, 1); @@ -319,19 +312,38 @@ void dccg31_disable_symclk32_le( REG_UPDATE_2(SYMCLK32_LE_CNTL, SYMCLK32_LE0_SRC_SEL, 0, SYMCLK32_LE0_EN, 0); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, - SYMCLK32_LE0_GATE_DISABLE, 0, - SYMCLK32_ROOT_LE0_GATE_DISABLE, 0); break; case 1: REG_UPDATE_2(SYMCLK32_LE_CNTL, SYMCLK32_LE1_SRC_SEL, 0, SYMCLK32_LE1_EN, 0); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) - REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, - SYMCLK32_LE1_GATE_DISABLE, 0, - SYMCLK32_ROOT_LE1_GATE_DISABLE, 0); + break; + default: + BREAK_TO_DEBUGGER(); + return; + } +} + +void dccg31_set_symclk32_le_root_clock_gating( + struct dccg *dccg, + int hpo_le_inst, + bool enable) +{ + struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + + if (!dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) + return; + + switch (hpo_le_inst) { + case 0: + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_LE0_GATE_DISABLE, enable ? 1 : 0, + SYMCLK32_ROOT_LE0_GATE_DISABLE, enable ? 1 : 0); + break; + case 1: + REG_UPDATE_2(DCCG_GATE_DISABLE_CNTL3, + SYMCLK32_LE1_GATE_DISABLE, enable ? 1 : 0, + SYMCLK32_ROOT_LE1_GATE_DISABLE, enable ? 1 : 0); break; default: BREAK_TO_DEBUGGER(); @@ -660,10 +672,8 @@ void dccg31_init(struct dccg *dccg) dccg31_disable_symclk32_se(dccg, 2); dccg31_disable_symclk32_se(dccg, 3); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_le) { - dccg31_disable_symclk32_le(dccg, 0); - dccg31_disable_symclk32_le(dccg, 1); - } + dccg31_set_symclk32_le_root_clock_gating(dccg, 0, false); + dccg31_set_symclk32_le_root_clock_gating(dccg, 1, false); if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpstream) { dccg31_disable_dpstreamclk(dccg, 0); diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h index 0902ce5eb8a1..e3caaacf7493 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.h @@ -179,6 +179,11 @@ void dccg31_disable_symclk32_le( struct dccg *dccg, int hpo_le_inst); +void dccg31_set_symclk32_le_root_clock_gating( + struct dccg *dccg, + int hpo_le_inst, + bool enable); + void dccg31_set_physymclk( struct dccg *dccg, int phy_inst, diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c index bd62502380d8..4596f3bac1b4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c @@ -558,7 +558,7 @@ void dcn31_link_encoder_disable_output( struct dmub_cmd_dig_dpia_control_data dpia_control = { 0 }; struct dc_link *link; - if (!dcn10_is_dig_enabled(enc)) + if (enc->funcs->is_dig_enabled && !enc->funcs->is_dig_enabled(enc)) return; link = link_enc_cfg_get_link_using_link_enc(enc->ctx->dc, enc->preferred_engine); diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c index 0278bae50a9d..45143459eedd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hpo_dp_stream_encoder.c @@ -154,7 +154,7 @@ static void dcn31_hpo_dp_stream_enc_dp_blank( VID_STREAM_STATUS, 0, 10, 5000); - /* Disable SDP tranmission */ + /* Disable SDP transmission */ REG_UPDATE(DP_SYM32_ENC_SDP_CONTROL, SDP_STREAM_ENABLE, 0); diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c index 7445ed27852a..1f4e0b6261ad 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_hubbub.c @@ -1018,8 +1018,8 @@ void hubbub31_init(struct hubbub *hubbub) /*done in hwseq*/ /*REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);*/ REG_UPDATE_2(DCHUBBUB_CLOCK_CNTL, - DISPCLK_R_DCHUBBUB_GATE_DIS, 0, - DCFCLK_R_DCHUBBUB_GATE_DIS, 0); + DISPCLK_R_DCHUBBUB_GATE_DIS, 1, + DCFCLK_R_DCHUBBUB_GATE_DIS, 1); } /* diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c index fc25cc300a17..1d7bc1e39afe 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c @@ -67,7 +67,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = { .update_bandwidth = dcn20_update_bandwidth, .set_drr = dcn10_set_drr, .get_position = dcn10_get_position, - .set_static_screen_control = dcn10_set_static_screen_control, + .set_static_screen_control = dcn30_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, .set_avmute = dcn30_set_avmute, .log_hw_state = dcn10_log_hw_state, diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index fc33b5fcabe1..82de4fe2637f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -96,6 +96,7 @@ #include "dce/dmub_psr.h" #include "dce/dce_aux.h" #include "dce/dce_i2c.h" +#include "dce/dmub_replay.h" #include "dml/dcn30/display_mode_vba_30.h" #include "vm_helper.h" @@ -896,6 +897,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, .ilr = { .optimize_edp_link_rate = true, @@ -1479,6 +1481,9 @@ static void dcn31_resource_destruct(struct dcn31_resource_pool *pool) if (pool->base.psr != NULL) dmub_psr_destroy(&pool->base.psr); + if (pool->base.replay != NULL) + dmub_replay_destroy(&pool->base.replay); + if (pool->base.dccg != NULL) dcn_dccg_destroy(&pool->base.dccg); } @@ -1776,8 +1781,8 @@ bool dcn31_validate_bandwidth(struct dc *dc, BW_VAL_TRACE_SKIP(fast); goto validate_out; } - - dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); + if (dc->res_pool->funcs->calculate_wm_and_dlg) + dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); BW_VAL_TRACE_END_WATERMARKS(); @@ -1818,7 +1823,7 @@ static struct resource_funcs dcn31_res_pool_funcs = { .calculate_wm_and_dlg = dcn31_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn31_update_soc_for_wm_a, .populate_dml_pipes = dcn31_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, @@ -2085,6 +2090,14 @@ static bool dcn31_resource_construct( goto create_fail; } + /* Replay */ + pool->base.replay = dmub_replay_create(ctx); + if (pool->base.replay == NULL) { + dm_error("DC: failed to create replay obj!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + /* ABM */ for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) { pool->base.multiple_abms[i] = dmub_abm_create(ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c index 0746ed31d1d1..ad3f019a784f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_dccg.c @@ -362,6 +362,7 @@ static const struct dccg_funcs dccg314_funcs = { .disable_symclk32_se = dccg31_disable_symclk32_se, .enable_symclk32_le = dccg31_enable_symclk32_le, .disable_symclk32_le = dccg31_disable_symclk32_le, + .set_symclk32_le_root_clock_gating = dccg31_set_symclk32_le_root_clock_gating, .set_physymclk = dccg31_set_physymclk, .set_dtbclk_dto = dccg314_set_dtbclk_dto, .set_audio_dtbclk_dto = dccg31_set_audio_dtbclk_dto, diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c index ca8fe55c33b8..4ef85c3a0688 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_init.c @@ -69,7 +69,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = { .update_bandwidth = dcn20_update_bandwidth, .set_drr = dcn10_set_drr, .get_position = dcn10_get_position, - .set_static_screen_control = dcn10_set_static_screen_control, + .set_static_screen_control = dcn30_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, .set_avmute = dcn30_set_avmute, .log_hw_state = dcn10_log_hw_state, diff --git a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c index 6a9024aa3285..004beed9bd44 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn314/dcn314_resource.c @@ -93,6 +93,7 @@ #include "reg_helper.h" #include "dce/dmub_abm.h" #include "dce/dmub_psr.h" +#include "dce/dmub_replay.h" #include "dce/dce_aux.h" #include "dce/dce_i2c.h" #include "dml/dcn314/display_mode_vba_314.h" @@ -870,6 +871,7 @@ static const struct dc_debug_options debug_defaults_drv = { .enable_z9_disable_interface = true, .minimum_z8_residency_time = 2000, .psr_skip_crtc_disable = true, + .replay_skip_crtc_disabled = true, .disable_dmcu = true, .force_abm_enable = false, .timing_trace = false, @@ -908,15 +910,15 @@ static const struct dc_debug_options debug_defaults_drv = { .root_clock_optimization = { .bits = { .dpp = true, - .dsc = false, - .hdmistream = false, - .hdmichar = false, - .dpstream = false, - .symclk32_se = false, - .symclk32_le = false, - .symclk_fe = false, - .physymclk = false, - .dpiasymclk = false, + .dsc = true, + .hdmistream = true, + .hdmichar = true, + .dpstream = true, + .symclk32_se = true, + .symclk32_le = true, + .symclk_fe = true, + .physymclk = true, + .dpiasymclk = true, } }, @@ -945,6 +947,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, .ilr = { .optimize_edp_link_rate = true, @@ -1029,6 +1032,28 @@ static const struct dce_i2c_mask i2c_masks = { I2C_COMMON_MASK_SH_LIST_DCN30(_MASK) }; +/* ========================================================== */ + +/* + * DPIA index | Preferred Encoder | Host Router + * 0 | C | 0 + * 1 | First Available | 0 + * 2 | D | 1 + * 3 | First Available | 1 + */ +/* ========================================================== */ +static const enum engine_id dpia_to_preferred_enc_id_table[] = { + ENGINE_ID_DIGC, + ENGINE_ID_DIGC, + ENGINE_ID_DIGD, + ENGINE_ID_DIGD +}; + +static enum engine_id dcn314_get_preferred_eng_id_dpia(unsigned int dpia_index) +{ + return dpia_to_preferred_enc_id_table[dpia_index]; +} + static struct dce_i2c_hw *dcn31_i2c_hw_create( struct dc_context *ctx, uint32_t inst) @@ -1528,6 +1553,9 @@ static void dcn314_resource_destruct(struct dcn314_resource_pool *pool) if (pool->base.psr != NULL) dmub_psr_destroy(&pool->base.psr); + if (pool->base.replay != NULL) + dmub_replay_destroy(&pool->base.replay); + if (pool->base.dccg != NULL) dcn_dccg_destroy(&pool->base.dccg); } @@ -1683,7 +1711,9 @@ static bool filter_modes_for_single_channel_workaround(struct dc *dc, struct dc_state *context) { // Filter 2K@240Hz+8K@24fps above combination timing if memory only has single dimm LPDDR - if (dc->clk_mgr->bw_params->vram_type == 34 && dc->clk_mgr->bw_params->num_channels < 2) { + if (dc->clk_mgr->bw_params->vram_type == 34 && + dc->clk_mgr->bw_params->num_channels < 2 && + context->stream_count > 1) { int total_phy_pix_clk = 0; for (int i = 0; i < context->stream_count; i++) @@ -1732,8 +1762,8 @@ bool dcn314_validate_bandwidth(struct dc *dc, BW_VAL_TRACE_SKIP(fast); goto validate_out; } - - dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); + if (dc->res_pool->funcs->calculate_wm_and_dlg) + dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); BW_VAL_TRACE_END_WATERMARKS(); @@ -1765,7 +1795,7 @@ static struct resource_funcs dcn314_res_pool_funcs = { .calculate_wm_and_dlg = dcn31_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn31_update_soc_for_wm_a, .populate_dml_pipes = dcn314_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, @@ -1777,6 +1807,7 @@ static struct resource_funcs dcn314_res_pool_funcs = { .update_bw_bounding_box = dcn314_update_bw_bounding_box, .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn314_get_panel_config_defaults, + .get_preferred_eng_id_dpia = dcn314_get_preferred_eng_id_dpia, }; static struct clock_source *dcn30_clock_source_create( @@ -2025,6 +2056,14 @@ static bool dcn314_resource_construct( goto create_fail; } + /* Replay */ + pool->base.replay = dmub_replay_create(ctx); + if (pool->base.replay == NULL) { + dm_error("DC: failed to create replay obj!\n"); + BREAK_TO_DEBUGGER(); + goto create_fail; + } + /* ABM */ for (i = 0; i < pool->base.res_cap->num_timing_generator; i++) { pool->base.multiple_abms[i] = dmub_abm_create(ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c index df3a438abda8..127487ea3d7d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn315/dcn315_resource.c @@ -895,6 +895,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, .ilr = { .optimize_edp_link_rate = true, @@ -1659,7 +1660,7 @@ static int dcn315_populate_dml_pipes_from_context( { int i, pipe_cnt, crb_idx, crb_pipes; struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *pipe; + struct pipe_ctx *pipe = NULL; const int max_usable_det = context->bw_ctx.dml.ip.config_return_buffer_size_in_kbytes - DCN3_15_MIN_COMPBUF_SIZE_KB; int remaining_det_segs = max_usable_det / DCN3_15_CRB_SEGMENT_SIZE_KB; bool pixel_rate_crb = allow_pixel_rate_crb(dc, context); @@ -1817,7 +1818,7 @@ static struct resource_funcs dcn315_res_pool_funcs = { .calculate_wm_and_dlg = dcn31_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn315_update_soc_for_wm_a, .populate_dml_pipes = dcn315_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c index 707cf28bbceb..5fe2c61527df 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn316/dcn316_resource.c @@ -891,6 +891,7 @@ static const struct dc_panel_config panel_config_defaults = { .psr = { .disable_psr = false, .disallow_psrsu = false, + .disallow_replay = false, }, .ilr = { .optimize_edp_link_rate = true, @@ -1704,7 +1705,7 @@ static struct resource_funcs dcn316_res_pool_funcs = { .calculate_wm_and_dlg = dcn31_calculate_wm_and_dlg, .update_soc_for_wm_a = dcn31_update_soc_for_wm_a, .populate_dml_pipes = dcn316_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_layer = dcn20_acquire_idle_pipe_for_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn20_acquire_free_pipe_for_layer, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c index 11e28e056cf7..921f58c0c729 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dccg.c @@ -49,7 +49,10 @@ static void dccg32_trigger_dio_fifo_resync( uint32_t dispclk_rdivider_value = 0; REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_RDIVIDER, &dispclk_rdivider_value); - REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value); + + /* Not valid for the WDIVIDER to be set to 0 */ + if (dispclk_rdivider_value != 0) + REG_UPDATE(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, dispclk_rdivider_value); } static void dccg32_get_pixel_rate_div( @@ -278,7 +281,8 @@ static void dccg32_set_dpstreamclk( struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); /* set the dtbclk_p source */ - dccg32_set_dtbclk_p_src(dccg, src, otg_inst); + /* always program refclk as DTBCLK. No use-case expected to require DPREFCLK as refclk */ + dccg32_set_dtbclk_p_src(dccg, DTBCLK0, otg_inst); /* enabled to select one of the DTBCLKs for pipe */ switch (dp_hpo_inst) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c index a18b9c0c5709..8bfef6d095b2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hubbub.c @@ -955,8 +955,8 @@ void hubbub32_init(struct hubbub *hubbub) /*REG_UPDATE(DCFCLK_CNTL, DCFCLK_GATE_DIS, 0);*/ REG_UPDATE_2(DCHUBBUB_CLOCK_CNTL, - DISPCLK_R_DCHUBBUB_GATE_DIS, 0, - DCFCLK_R_DCHUBBUB_GATE_DIS, 0); + DISPCLK_R_DCHUBBUB_GATE_DIS, 1, + DCFCLK_R_DCHUBBUB_GATE_DIS, 1); } /* ignore the "df_pre_cstate_req" from the SDP port control. diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c index d52d5feeb311..680e7fa8d18a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_hwseq.c @@ -47,11 +47,9 @@ #include "clk_mgr.h" #include "dsc.h" #include "dcn20/dcn20_optc.h" -#include "dmub_subvp_state.h" #include "dce/dmub_hw_lock_mgr.h" #include "dcn32_resource.h" #include "link.h" -#include "dmub/inc/dmub_subvp_state.h" #define DC_LOGGER_INIT(logger) @@ -569,7 +567,7 @@ bool dcn32_set_output_transfer_func(struct dc *dc, bool ret = false; /* program OGAM or 3DLUT only for the top pipe*/ - if (pipe_ctx->top_pipe == NULL) { + if (resource_is_pipe_type(pipe_ctx, OPP_HEAD)) { /*program shaper and 3dlut in MPC*/ ret = dcn32_set_mpc_shaper_3dlut(pipe_ctx, stream); if (ret == false && mpc->funcs->set_output_gamma && stream->out_transfer_func) { @@ -1204,10 +1202,10 @@ void dcn32_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_ for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe->top_pipe || pipe->prev_odm_pipe) + if (!resource_is_pipe_type(pipe, OTG_MASTER)) continue; - if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal)) + if ((pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal)) && pipe->stream->mall_stream_config.type != SUBVP_PHANTOM) { pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg); reset_sync_context_for_pipe(dc, context, i); @@ -1301,7 +1299,7 @@ static void apply_symclk_on_tx_off_wa(struct dc_link *link) if (link->phy_state.symclk_ref_cnts.otg > 0) { for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; - if (pipe_ctx->stream && pipe_ctx->stream->link == link && pipe_ctx->top_pipe == NULL) { + if (resource_is_pipe_type(pipe_ctx, OPP_HEAD) && pipe_ctx->stream->link == link) { pipe_ctx->clock_source->funcs->program_pix_clk( pipe_ctx->clock_source, &pipe_ctx->stream_res.pix_clk_params, @@ -1384,7 +1382,7 @@ void dcn32_apply_update_flags_for_phantom(struct pipe_ctx *phantom_pipe) { phantom_pipe->update_flags.raw = 0; if (phantom_pipe->stream && phantom_pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { - if (phantom_pipe->stream && phantom_pipe->plane_state) { + if (resource_is_pipe_type(phantom_pipe, DPP_PIPE)) { phantom_pipe->update_flags.bits.enable = 1; phantom_pipe->update_flags.bits.mpcc = 1; phantom_pipe->update_flags.bits.dppclk = 1; @@ -1394,7 +1392,7 @@ void dcn32_apply_update_flags_for_phantom(struct pipe_ctx *phantom_pipe) phantom_pipe->update_flags.bits.scaler = 1; phantom_pipe->update_flags.bits.viewport = 1; phantom_pipe->update_flags.bits.det_size = 1; - if (!phantom_pipe->top_pipe && !phantom_pipe->prev_odm_pipe) { + if (resource_is_pipe_type(phantom_pipe, OTG_MASTER)) { phantom_pipe->update_flags.bits.odm = 1; phantom_pipe->update_flags.bits.global_sync = 1; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c index 777b2fac20c4..c7417147dff1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_init.c @@ -65,7 +65,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = { .update_bandwidth = dcn20_update_bandwidth, .set_drr = dcn10_set_drr, .get_position = dcn10_get_position, - .set_static_screen_control = dcn10_set_static_screen_control, + .set_static_screen_control = dcn30_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, .set_avmute = dcn30_set_avmute, .log_hw_state = dcn10_log_hw_state, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.c index c8041cfd594d..3082da04a63d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.c @@ -64,7 +64,7 @@ void mpc32_mpc_init(struct mpc *mpc) } } -static void mpc32_power_on_blnd_lut( +void mpc32_power_on_blnd_lut( struct mpc *mpc, uint32_t mpcc_id, bool power_on) @@ -120,7 +120,7 @@ static enum dc_lut_mode mpc32_get_post1dlut_current(struct mpc *mpc, uint32_t mp return mode; } -static void mpc32_configure_post1dlut( +void mpc32_configure_post1dlut( struct mpc *mpc, uint32_t mpcc_id, bool is_ram_a) @@ -163,7 +163,7 @@ static void mpc32_post1dlut_get_reg_field( } /*program blnd lut RAM A*/ -static void mpc32_program_post1dluta_settings( +void mpc32_program_post1dluta_settings( struct mpc *mpc, uint32_t mpcc_id, const struct pwl_params *params) @@ -192,7 +192,7 @@ static void mpc32_program_post1dluta_settings( } /*program blnd lut RAM B*/ -static void mpc32_program_post1dlutb_settings( +void mpc32_program_post1dlutb_settings( struct mpc *mpc, uint32_t mpcc_id, const struct pwl_params *params) @@ -220,7 +220,7 @@ static void mpc32_program_post1dlutb_settings( cm_helper_program_gamcor_xfer_func(mpc->ctx, params, &gam_regs); } -static void mpc32_program_post1dlut_pwl( +void mpc32_program_post1dlut_pwl( struct mpc *mpc, uint32_t mpcc_id, const struct pwl_result_data *rgb, @@ -321,7 +321,7 @@ static enum dc_lut_mode mpc32_get_shaper_current(struct mpc *mpc, uint32_t mpcc_ } -static void mpc32_configure_shaper_lut( +void mpc32_configure_shaper_lut( struct mpc *mpc, bool is_ram_a, uint32_t mpcc_id) @@ -336,7 +336,7 @@ static void mpc32_configure_shaper_lut( } -static void mpc32_program_shaper_luta_settings( +void mpc32_program_shaper_luta_settings( struct mpc *mpc, const struct pwl_params *params, uint32_t mpcc_id) @@ -486,7 +486,7 @@ static void mpc32_program_shaper_luta_settings( } -static void mpc32_program_shaper_lutb_settings( +void mpc32_program_shaper_lutb_settings( struct mpc *mpc, const struct pwl_params *params, uint32_t mpcc_id) @@ -637,7 +637,7 @@ static void mpc32_program_shaper_lutb_settings( } -static void mpc32_program_shaper_lut( +void mpc32_program_shaper_lut( struct mpc *mpc, const struct pwl_result_data *rgb, uint32_t num, @@ -671,7 +671,7 @@ static void mpc32_program_shaper_lut( } -static void mpc32_power_on_shaper_3dlut( +void mpc32_power_on_shaper_3dlut( struct mpc *mpc, uint32_t mpcc_id, bool power_on) @@ -789,7 +789,7 @@ static enum dc_lut_mode get3dlut_config( } -static void mpc32_select_3dlut_ram( +void mpc32_select_3dlut_ram( struct mpc *mpc, enum dc_lut_mode mode, bool is_color_channel_12bits, @@ -803,7 +803,7 @@ static void mpc32_select_3dlut_ram( } -static void mpc32_select_3dlut_ram_mask( +void mpc32_select_3dlut_ram_mask( struct mpc *mpc, uint32_t ram_selection_mask, uint32_t mpcc_id) @@ -816,7 +816,7 @@ static void mpc32_select_3dlut_ram_mask( } -static void mpc32_set3dlut_ram12( +void mpc32_set3dlut_ram12( struct mpc *mpc, const struct dc_rgb *lut, uint32_t entries, @@ -848,7 +848,7 @@ static void mpc32_set3dlut_ram12( } -static void mpc32_set3dlut_ram10( +void mpc32_set3dlut_ram10( struct mpc *mpc, const struct dc_rgb *lut, uint32_t entries, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.h index 2c2ecd053806..9ac584fa89ce 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_mpc.h @@ -332,4 +332,65 @@ void dcn32_mpc_construct(struct dcn30_mpc *mpc30, int num_mpcc, int num_rmu); +void mpc32_power_on_blnd_lut( + struct mpc *mpc, + uint32_t mpcc_id, + bool power_on); +void mpc32_program_post1dlut_pwl( + struct mpc *mpc, + uint32_t mpcc_id, + const struct pwl_result_data *rgb, + uint32_t num); +void mpc32_program_post1dlutb_settings( + struct mpc *mpc, + uint32_t mpcc_id, + const struct pwl_params *params); +void mpc32_program_post1dluta_settings( + struct mpc *mpc, + uint32_t mpcc_id, + const struct pwl_params *params); +void mpc32_configure_post1dlut( + struct mpc *mpc, + uint32_t mpcc_id, + bool is_ram_a); +void mpc32_program_shaper_lut( + struct mpc *mpc, + const struct pwl_result_data *rgb, + uint32_t num, + uint32_t mpcc_id); +void mpc32_program_shaper_lutb_settings( + struct mpc *mpc, + const struct pwl_params *params, + uint32_t mpcc_id); +void mpc32_program_shaper_luta_settings( + struct mpc *mpc, + const struct pwl_params *params, + uint32_t mpcc_id); +void mpc32_configure_shaper_lut( + struct mpc *mpc, + bool is_ram_a, + uint32_t mpcc_id); +void mpc32_power_on_shaper_3dlut( + struct mpc *mpc, + uint32_t mpcc_id, + bool power_on); +void mpc32_set3dlut_ram10( + struct mpc *mpc, + const struct dc_rgb *lut, + uint32_t entries, + uint32_t mpcc_id); +void mpc32_set3dlut_ram12( + struct mpc *mpc, + const struct dc_rgb *lut, + uint32_t entries, + uint32_t mpcc_id); +void mpc32_select_3dlut_ram_mask( + struct mpc *mpc, + uint32_t ram_selection_mask, + uint32_t mpcc_id); +void mpc32_select_3dlut_ram( + struct mpc *mpc, + enum dc_lut_mode mode, + bool is_color_channel_12bits, + uint32_t mpcc_id); #endif //__DC_MPCC_DCN32_H__ diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c index 1cc09799f92d..f9d601c8c721 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.c @@ -1709,8 +1709,8 @@ void dcn32_retain_phantom_pipes(struct dc *dc, struct dc_state *context) for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->top_pipe && !pipe->prev_odm_pipe && - pipe->plane_state && pipe->stream && + if (resource_is_pipe_type(pipe, OTG_MASTER) && + resource_is_pipe_type(pipe, DPP_PIPE) && pipe->stream->mall_stream_config.type == SUBVP_PHANTOM) { phantom_plane = pipe->plane_state; phantom_stream = pipe->stream; @@ -1892,7 +1892,7 @@ int dcn32_populate_dml_pipes_from_context( { int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; - struct pipe_ctx *pipe; + struct pipe_ctx *pipe = NULL; bool subvp_in_use = false; struct dc_crtc_timing *timing; bool vsr_odm_support = false; @@ -2038,7 +2038,7 @@ static struct resource_funcs dcn32_res_pool_funcs = { .validate_bandwidth = dcn32_validate_bandwidth, .calculate_wm_and_dlg = dcn32_calculate_wm_and_dlg, .populate_dml_pipes = dcn32_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_head_pipe_in_layer = dcn32_acquire_idle_pipe_for_head_pipe_in_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn32_acquire_free_pipe_as_secondary_dpp_pipe, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, @@ -2485,6 +2485,85 @@ struct resource_pool *dcn32_create_resource_pool( return NULL; } +/* + * Find the most optimal free pipe from res_ctx, which could be used as a + * secondary dpp pipe for input opp head pipe. + * + * a free pipe - a pipe in input res_ctx not yet used for any streams or + * planes. + * secondary dpp pipe - a pipe gets inserted to a head OPP pipe's MPC blending + * tree. This is typical used for rendering MPO planes or additional offset + * areas in MPCC combine. + * + * Hardware Transition Minimization Algorithm for Finding a Secondary DPP Pipe + * ------------------------------------------------------------------------- + * + * PROBLEM: + * + * 1. There is a hardware limitation that a secondary DPP pipe cannot be + * transferred from one MPC blending tree to the other in a single frame. + * Otherwise it could cause glitches on the screen. + * + * For instance, we cannot transition from state 1 to state 2 in one frame. This + * is because PIPE1 is transferred from PIPE0's MPC blending tree over to + * PIPE2's MPC blending tree, which is not supported by hardware. + * To support this transition we need to first remove PIPE1 from PIPE0's MPC + * blending tree in one frame and then insert PIPE1 to PIPE2's MPC blending tree + * in the next frame. This is not optimal as it will delay the flip for two + * frames. + * + * State 1: + * PIPE0 -- secondary DPP pipe --> (PIPE1) + * PIPE2 -- secondary DPP pipe --> NONE + * + * State 2: + * PIPE0 -- secondary DPP pipe --> NONE + * PIPE2 -- secondary DPP pipe --> (PIPE1) + * + * 2. We want to in general minimize the unnecessary changes in pipe topology. + * If a pipe is already added in current blending tree and there are no changes + * to plane topology, we don't want to swap it with another free pipe + * unnecessarily in every update. Powering up and down a pipe would require a + * full update which delays the flip for 1 frame. If we use the original pipe + * we don't have to toggle its power. So we can flip faster. + */ +static int find_optimal_free_pipe_as_secondary_dpp_pipe( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct resource_pool *pool, + const struct pipe_ctx *new_opp_head) +{ + const struct pipe_ctx *cur_opp_head; + int free_pipe_idx; + + cur_opp_head = &cur_res_ctx->pipe_ctx[new_opp_head->pipe_idx]; + free_pipe_idx = resource_find_free_pipe_used_in_cur_mpc_blending_tree( + cur_res_ctx, new_res_ctx, cur_opp_head); + + /* Up until here if we have not found a free secondary pipe, we will + * need to wait for at least one frame to complete the transition + * sequence. + */ + if (free_pipe_idx == FREE_PIPE_INDEX_NOT_FOUND) + free_pipe_idx = recource_find_free_pipe_not_used_in_cur_res_ctx( + cur_res_ctx, new_res_ctx, pool); + + /* Up until here if we have not found a free secondary pipe, we will + * need to wait for at least two frames to complete the transition + * sequence. It really doesn't matter which pipe we decide take from + * current enabled pipes. It won't save our frame time when we swap only + * one pipe or more pipes. + */ + if (free_pipe_idx == FREE_PIPE_INDEX_NOT_FOUND) + free_pipe_idx = resource_find_free_pipe_used_as_cur_sec_dpp_in_mpcc_combine( + cur_res_ctx, new_res_ctx, pool); + + if (free_pipe_idx == FREE_PIPE_INDEX_NOT_FOUND) + free_pipe_idx = resource_find_any_free_pipe(new_res_ctx, pool); + + return free_pipe_idx; +} + static struct pipe_ctx *find_idle_secondary_pipe_check_mpo( struct resource_context *res_ctx, const struct resource_pool *pool, @@ -2547,11 +2626,11 @@ static struct pipe_ctx *find_idle_secondary_pipe_check_mpo( return secondary_pipe; } -struct pipe_ctx *dcn32_acquire_idle_pipe_for_head_pipe_in_layer( +static struct pipe_ctx *dcn32_acquire_idle_pipe_for_head_pipe_in_layer( struct dc_state *state, const struct resource_pool *pool, struct dc_stream_state *stream, - struct pipe_ctx *head_pipe) + const struct pipe_ctx *head_pipe) { struct resource_context *res_ctx = &state->res_ctx; struct pipe_ctx *idle_pipe, *pipe; @@ -2590,6 +2669,43 @@ struct pipe_ctx *dcn32_acquire_idle_pipe_for_head_pipe_in_layer( return idle_pipe; } +struct pipe_ctx *dcn32_acquire_free_pipe_as_secondary_dpp_pipe( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, + const struct resource_pool *pool, + const struct pipe_ctx *opp_head_pipe) +{ + + int free_pipe_idx; + struct pipe_ctx *free_pipe; + + if (!opp_head_pipe->stream->ctx->dc->config.enable_windowed_mpo_odm) + return dcn32_acquire_idle_pipe_for_head_pipe_in_layer( + new_ctx, pool, opp_head_pipe->stream, opp_head_pipe); + + free_pipe_idx = find_optimal_free_pipe_as_secondary_dpp_pipe( + &cur_ctx->res_ctx, &new_ctx->res_ctx, + pool, opp_head_pipe); + if (free_pipe_idx >= 0) { + free_pipe = &new_ctx->res_ctx.pipe_ctx[free_pipe_idx]; + free_pipe->pipe_idx = free_pipe_idx; + free_pipe->stream = opp_head_pipe->stream; + free_pipe->stream_res.tg = opp_head_pipe->stream_res.tg; + free_pipe->stream_res.opp = opp_head_pipe->stream_res.opp; + + free_pipe->plane_res.hubp = pool->hubps[free_pipe->pipe_idx]; + free_pipe->plane_res.ipp = pool->ipps[free_pipe->pipe_idx]; + free_pipe->plane_res.dpp = pool->dpps[free_pipe->pipe_idx]; + free_pipe->plane_res.mpcc_inst = + pool->dpps[free_pipe->pipe_idx]->inst; + } else { + ASSERT(opp_head_pipe); + free_pipe = NULL; + } + + return free_pipe; +} + unsigned int dcn32_calc_num_avail_chans_for_mall(struct dc *dc, int num_chans) { /* diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h index 81e443170829..103a2b54d025 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource.h @@ -136,11 +136,11 @@ bool dcn32_any_surfaces_rotated(struct dc *dc, struct dc_state *context); bool dcn32_is_center_timing(struct pipe_ctx *pipe); bool dcn32_is_psr_capable(struct pipe_ctx *pipe); -struct pipe_ctx *dcn32_acquire_idle_pipe_for_head_pipe_in_layer( - struct dc_state *state, +struct pipe_ctx *dcn32_acquire_free_pipe_as_secondary_dpp_pipe( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream, - struct pipe_ctx *head_pipe); + const struct pipe_ctx *opp_head_pipe); void dcn32_determine_det_override(struct dc *dc, struct dc_state *context, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c index 5be242a1b82c..3ad2b48954e0 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c @@ -641,16 +641,21 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context) uint8_t non_subvp_pipes = 0; bool drr_pipe_found = false; bool drr_psr_capable = false; + uint64_t refresh_rate = 0; for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->stream) - continue; - - if (pipe->plane_state && !pipe->top_pipe) { - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (resource_is_pipe_type(pipe, OPP_HEAD) && + resource_is_pipe_type(pipe, DPP_PIPE)) { + if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) { subvp_count++; + + refresh_rate = (pipe->stream->timing.pix_clk_100hz * (uint64_t)100 + + pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1); + refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); + refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total); + } if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { non_subvp_pipes++; drr_psr_capable = (drr_psr_capable || dcn32_is_psr_capable(pipe)); @@ -662,7 +667,8 @@ bool dcn32_subvp_drr_admissable(struct dc *dc, struct dc_state *context) } } - if (subvp_count == 1 && non_subvp_pipes == 1 && drr_pipe_found && !drr_psr_capable) + if (subvp_count == 1 && non_subvp_pipes == 1 && drr_pipe_found && !drr_psr_capable && + ((uint32_t)refresh_rate < 120)) result = true; return result; @@ -693,16 +699,21 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int bool drr_pipe_found = false; struct vba_vars_st *vba = &context->bw_ctx.dml.vba; bool vblank_psr_capable = false; + uint64_t refresh_rate = 0; for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; - if (!pipe->stream) - continue; - - if (pipe->plane_state && !pipe->top_pipe) { - if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) + if (resource_is_pipe_type(pipe, OPP_HEAD) && + resource_is_pipe_type(pipe, DPP_PIPE)) { + if (pipe->stream->mall_stream_config.type == SUBVP_MAIN) { subvp_count++; + + refresh_rate = (pipe->stream->timing.pix_clk_100hz * (uint64_t)100 + + pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1); + refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); + refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total); + } if (pipe->stream->mall_stream_config.type == SUBVP_NONE) { non_subvp_pipes++; vblank_psr_capable = (vblank_psr_capable || dcn32_is_psr_capable(pipe)); @@ -715,7 +726,8 @@ bool dcn32_subvp_vblank_admissable(struct dc *dc, struct dc_state *context, int } if (subvp_count == 1 && non_subvp_pipes == 1 && !drr_pipe_found && !vblank_psr_capable && - vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp) + ((uint32_t)refresh_rate < 120) && + vba->DRAMClockChangeSupport[vlevel][vba->maxMpcComb] == dm_dram_clock_change_vblank_w_mall_sub_vp) result = true; return result; diff --git a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c index a53478e15ce3..8d73cceb485b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn321/dcn321_resource.c @@ -1588,7 +1588,7 @@ static struct resource_funcs dcn321_res_pool_funcs = { .validate_bandwidth = dcn32_validate_bandwidth, .calculate_wm_and_dlg = dcn32_calculate_wm_and_dlg, .populate_dml_pipes = dcn32_populate_dml_pipes_from_context, - .acquire_idle_pipe_for_head_pipe_in_layer = dcn32_acquire_idle_pipe_for_head_pipe_in_layer, + .acquire_free_pipe_as_secondary_dpp_pipe = dcn32_acquire_free_pipe_as_secondary_dpp_pipe, .add_stream_to_ctx = dcn30_add_stream_to_ctx, .add_dsc_to_stream_resource = dcn20_add_dsc_to_stream_resource, .remove_stream_from_ctx = dcn20_remove_stream_from_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dml/calcs/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dce_calcs.c index 0100a6053ab6..f2dfa96f9ef5 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/calcs/dce_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dce_calcs.c @@ -3015,7 +3015,7 @@ static bool all_displays_in_sync(const struct pipe_ctx pipe[], int i, num_active_pipes = 0; for (i = 0; i < pipe_count; i++) { - if (!pipe[i].stream || pipe[i].top_pipe) + if (!resource_is_pipe_type(&pipe[i], OPP_HEAD)) continue; active_pipes[num_active_pipes++] = &pipe[i]; diff --git a/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c index e73f089c84bb..50b0434354f8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/dml/calcs/dcn_calcs.c @@ -1258,7 +1258,7 @@ bool dcn_validate_bandwidth( hsplit_pipe->pipe_dlg_param.vblank_end = pipe->pipe_dlg_param.vblank_end; } else { /* pipe not split previously needs split */ - hsplit_pipe = find_idle_secondary_pipe(&context->res_ctx, pool, pipe); + hsplit_pipe = resource_find_free_secondary_pipe_legacy(&context->res_ctx, pool, pipe); ASSERT(hsplit_pipe); split_stream_across_pipes(&context->res_ctx, pool, pipe, hsplit_pipe); } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c index 8ae5ddbd1b27..5805fb02af14 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c @@ -1099,6 +1099,11 @@ void dcn20_calculate_dlg_params(struct dc *dc, context->res_ctx.pipe_ctx[i].plane_res.bw.dppclk_khz = pipes[pipe_idx].clks_cfg.dppclk_mhz * 1000; context->res_ctx.pipe_ctx[i].pipe_dlg_param = pipes[pipe_idx].pipe.dest; + if (dc->ctx->dce_version < DCN_VERSION_3_1 && + context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid) + dcn20_adjust_freesync_v_startup( + &context->res_ctx.pipe_ctx[i].stream->timing, + &context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start); pipe_idx++; } @@ -1305,7 +1310,7 @@ int dcn20_populate_dml_pipes_from_context(struct dc *dc, pipes[pipe_cnt].dout.is_virtual = 0; pipes[pipe_cnt].pipe.dest.vtotal_min = res_ctx->pipe_ctx[i].stream->adjust.v_total_min; pipes[pipe_cnt].pipe.dest.vtotal_max = res_ctx->pipe_ctx[i].stream->adjust.v_total_max; - switch (get_num_odm_splits(&res_ctx->pipe_ctx[i])) { + switch (resource_get_num_odm_splits(&res_ctx->pipe_ctx[i])) { case 1: pipes[pipe_cnt].pipe.dest.odm_combine = dm_odm_combine_mode_2to1; break; @@ -1927,7 +1932,6 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co int vlevel = 0; int pipe_split_from[MAX_PIPES]; int pipe_cnt = 0; - int i = 0; display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); DC_LOGGER_INIT(dc->ctx->logger); @@ -1951,15 +1955,6 @@ static bool dcn20_validate_bandwidth_internal(struct dc *dc, struct dc_state *co dcn20_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); - for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid) - dcn20_adjust_freesync_v_startup( - &context->res_ctx.pipe_ctx[i].stream->timing, - &context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start); - } - BW_VAL_TRACE_END_WATERMARKS(); goto validate_out; @@ -2232,7 +2227,6 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, int vlevel = 0; int pipe_split_from[MAX_PIPES]; int pipe_cnt = 0; - int i = 0; display_e2e_pipe_params_st *pipes = kzalloc(dc->res_pool->pipe_count * sizeof(display_e2e_pipe_params_st), GFP_ATOMIC); DC_LOGGER_INIT(dc->ctx->logger); @@ -2261,15 +2255,6 @@ bool dcn21_validate_bandwidth_fp(struct dc *dc, dcn21_calculate_wm(dc, context, pipes, &pipe_cnt, pipe_split_from, vlevel, fast_validate); dcn20_calculate_dlg_params(dc, context, pipes, pipe_cnt, vlevel); - for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (!context->res_ctx.pipe_ctx[i].stream) - continue; - if (context->res_ctx.pipe_ctx[i].stream->adaptive_sync_infopacket.valid) - dcn20_adjust_freesync_v_startup( - &context->res_ctx.pipe_ctx[i].stream->timing, - &context->res_ctx.pipe_ctx[i].pipe_dlg_param.vstartup_start); - } - BW_VAL_TRACE_END_WATERMARKS(); goto validate_out; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c index f294f2f8c75b..57cf0358cc43 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c @@ -3194,7 +3194,7 @@ static void CalculateFlipSchedule( unsigned int HostVMDynamicLevels; double TimeForFetchingMetaPTEImmediateFlip; double TimeForFetchingRowInVBlankImmediateFlip; - double ImmediateFlipBW; + double ImmediateFlipBW = 1.0; double HostVMInefficiencyFactor; double VRatioClamped; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c index 9af1a43c042b..ad741a723c0e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c @@ -784,8 +784,7 @@ static unsigned int dscComputeDelay(enum output_format_class pixelFormat, enum o Delay = Delay + 1; // sft Delay = Delay + 1; - } - else { + } else { // sfr Delay = Delay + 2; // dsccif @@ -3489,8 +3488,7 @@ static double TruncToValidBPP( if (Format == dm_n422) { MinDSCBPP = 7; MaxDSCBPP = 2 * DSCInputBitPerComponent - 1.0 / 16.0; - } - else { + } else { MinDSCBPP = 8; MaxDSCBPP = 3 * DSCInputBitPerComponent - 1.0 / 16.0; } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index 43016c462251..adea459e7d36 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -3505,7 +3505,7 @@ static void CalculateFlipSchedule( unsigned int HostVMDynamicLevelsTrips; double TimeForFetchingMetaPTEImmediateFlip; double TimeForFetchingRowInVBlankImmediateFlip; - double ImmediateFlipBW; + double ImmediateFlipBW = 1.0; double LineTime = v->HTotal[k] / v->PixelClock[k]; if (v->GPUVMEnable == true && v->HostVMEnable == true) { @@ -4135,7 +4135,9 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } if (v->OutputFormat[k] == dm_420 && v->HActive[k] > DCN31_MAX_FMT_420_BUFFER_WIDTH && v->ODMCombineEnablePerState[i][k] != dm_odm_combine_mode_4to1) { - if (v->HActive[k] / 2 > DCN31_MAX_FMT_420_BUFFER_WIDTH) { + if (v->Output[k] == dm_hdmi) { + FMTBufferExceeded = true; + } else if (v->HActive[k] / 2 > DCN31_MAX_FMT_420_BUFFER_WIDTH) { v->ODMCombineEnablePerState[i][k] = dm_odm_combine_mode_4to1; v->PlaneRequiredDISPCLK = v->PlaneRequiredDISPCLKWithODMCombine4To1; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c index d9e049e7ff0a..fb21572750e8 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/dcn314_fpu.c @@ -31,6 +31,7 @@ #include "dml/dcn20/dcn20_fpu.h" #include "dml/dcn31/dcn31_fpu.h" #include "dml/display_mode_vba.h" +#include "dml/dml_inline_defs.h" struct _vcs_dpi_ip_params_st dcn3_14_ip = { .VBlankNomDefaultUS = 668, @@ -273,6 +274,36 @@ static bool is_dual_plane(enum surface_pixel_format format) return format >= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN || format == SURFACE_PIXEL_FORMAT_GRPH_RGBE_ALPHA; } +/* + * micro_sec_to_vert_lines () - converts time to number of vertical lines for a given timing + * + * @param: num_us: number of microseconds + * @return: number of vertical lines. If exact number of vertical lines is not found then + * it will round up to next number of lines to guarantee num_us + */ +static unsigned int micro_sec_to_vert_lines(unsigned int num_us, struct dc_crtc_timing *timing) +{ + unsigned int num_lines = 0; + unsigned int lines_time_in_ns = 1000.0 * + (((float)timing->h_total * 1000.0) / + ((float)timing->pix_clk_100hz / 10.0)); + + num_lines = dml_ceil(1000.0 * num_us / lines_time_in_ns, 1.0); + + return num_lines; +} + +static unsigned int get_vertical_back_porch(struct dc_crtc_timing *timing) +{ + unsigned int v_active = 0, v_blank = 0, v_back_porch = 0; + + v_active = timing->v_border_top + timing->v_addressable + timing->v_border_bottom; + v_blank = timing->v_total - v_active; + v_back_porch = v_blank - timing->v_front_porch - timing->v_sync_width; + + return v_back_porch; +} + int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, bool fast_validate) @@ -289,16 +320,31 @@ int dcn314_populate_dml_pipes_from_context_fpu(struct dc *dc, struct dc_state *c for (i = 0, pipe_cnt = 0; i < dc->res_pool->pipe_count; i++) { struct dc_crtc_timing *timing; + unsigned int num_lines = 0; + unsigned int v_back_porch = 0; if (!res_ctx->pipe_ctx[i].stream) continue; pipe = &res_ctx->pipe_ctx[i]; timing = &pipe->stream->timing; - pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min; + num_lines = micro_sec_to_vert_lines(dcn3_14_ip.VBlankNomDefaultUS, timing); + + if (pipe->stream->adjust.v_total_min != 0) + pipes[pipe_cnt].pipe.dest.vtotal = pipe->stream->adjust.v_total_min; + else + pipes[pipe_cnt].pipe.dest.vtotal = timing->v_total; + + v_back_porch = get_vertical_back_porch(timing); + pipes[pipe_cnt].pipe.dest.vblank_nom = timing->v_total - pipes[pipe_cnt].pipe.dest.vactive; - pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, dcn3_14_ip.VBlankNomDefaultUS); - pipes[pipe_cnt].pipe.dest.vblank_nom = max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width); + pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, num_lines); + // vblank_nom should not smaller than (VSync (timing->v_sync_width + v_back_porch) + 2) + // + 2 is because + // 1 -> VStartup_start should be 1 line before VSync + // 1 -> always reserve 1 line between start of vblank to vstartup signal + pipes[pipe_cnt].pipe.dest.vblank_nom = + max(pipes[pipe_cnt].pipe.dest.vblank_nom, timing->v_sync_width + v_back_porch + 2); pipes[pipe_cnt].pipe.dest.vblank_nom = min(pipes[pipe_cnt].pipe.dest.vblank_nom, max_allowed_vblank_nom); if (pipe->plane_state && diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c index 9010c47476e9..a94aa0f21a7f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c @@ -3613,7 +3613,7 @@ static void CalculateFlipSchedule( unsigned int HostVMDynamicLevelsTrips; double TimeForFetchingMetaPTEImmediateFlip; double TimeForFetchingRowInVBlankImmediateFlip; - double ImmediateFlipBW; + double ImmediateFlipBW = 1.0; double LineTime = v->HTotal[k] / v->PixelClock[k]; if (v->GPUVMEnable == true && v->HostVMEnable == true) { @@ -4227,7 +4227,9 @@ void dml314_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_ } if (v->OutputFormat[k] == dm_420 && v->HActive[k] > DCN314_MAX_FMT_420_BUFFER_WIDTH && v->ODMCombineEnablePerState[i][k] != dm_odm_combine_mode_4to1) { - if (v->HActive[k] / 2 > DCN314_MAX_FMT_420_BUFFER_WIDTH) { + if (v->Output[k] == dm_hdmi) { + FMTBufferExceeded = true; + } else if (v->HActive[k] / 2 > DCN314_MAX_FMT_420_BUFFER_WIDTH) { v->ODMCombineEnablePerState[i][k] = dm_odm_combine_mode_4to1; v->PlaneRequiredDISPCLK = v->PlaneRequiredDISPCLKWithODMCombine4To1; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c index a95034801712..711d4085b33b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c @@ -37,7 +37,7 @@ static const struct subvp_high_refresh_list subvp_high_refresh_list = { .min_refresh = 120, - .max_refresh = 165, + .max_refresh = 175, .res = { {.width = 3840, .height = 2160, }, {.width = 3440, .height = 1440, }, @@ -756,7 +756,7 @@ static bool dcn32_enough_pipes_for_subvp(struct dc *dc, struct dc_state *context struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; // Find the minimum pipe split count for non SubVP pipes - if (pipe->stream && !pipe->top_pipe && + if (resource_is_pipe_type(pipe, OPP_HEAD) && pipe->stream->mall_stream_config.type == SUBVP_NONE) { split_cnt = 0; while (pipe) { @@ -886,7 +886,8 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context) // We check for master pipe, but it shouldn't matter since we only need // the pipe for timing info (stream should be same for any pipe splits) - if (!pipe->stream || !pipe->plane_state || pipe->top_pipe || pipe->prev_odm_pipe) + if (!resource_is_pipe_type(pipe, OTG_MASTER) || + !resource_is_pipe_type(pipe, DPP_PIPE)) continue; // Find the SubVP pipe @@ -899,7 +900,8 @@ static bool subvp_drr_schedulable(struct dc *dc, struct dc_state *context) drr_pipe = &context->res_ctx.pipe_ctx[i]; // We check for master pipe only - if (!drr_pipe->stream || !drr_pipe->plane_state || drr_pipe->top_pipe || drr_pipe->prev_odm_pipe) + if (!resource_is_pipe_type(pipe, OTG_MASTER) || + !resource_is_pipe_type(pipe, DPP_PIPE)) continue; if (drr_pipe->stream->mall_stream_config.type == SUBVP_NONE && drr_pipe->stream->ignore_msa_timing_param && @@ -980,7 +982,8 @@ static bool subvp_vblank_schedulable(struct dc *dc, struct dc_state *context) // We check for master pipe, but it shouldn't matter since we only need // the pipe for timing info (stream should be same for any pipe splits) - if (!pipe->stream || !pipe->plane_state || pipe->top_pipe || pipe->prev_odm_pipe) + if (!resource_is_pipe_type(pipe, OTG_MASTER) || + !resource_is_pipe_type(pipe, DPP_PIPE)) continue; if (!found && pipe->stream->mall_stream_config.type == SUBVP_NONE) { @@ -1040,7 +1043,7 @@ static bool subvp_subvp_admissable(struct dc *dc, uint32_t i; uint8_t subvp_count = 0; uint32_t min_refresh = subvp_high_refresh_list.min_refresh, max_refresh = 0; - uint32_t refresh_rate = 0; + uint64_t refresh_rate = 0; for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; @@ -1050,19 +1053,22 @@ static bool subvp_subvp_admissable(struct dc *dc, if (pipe->plane_state && !pipe->top_pipe && pipe->stream->mall_stream_config.type == SUBVP_MAIN) { - refresh_rate = (pipe->stream->timing.pix_clk_100hz * 100 + - pipe->stream->timing.v_total * pipe->stream->timing.h_total - 1) - / (double)(pipe->stream->timing.v_total * pipe->stream->timing.h_total); - if (refresh_rate < min_refresh) - min_refresh = refresh_rate; - if (refresh_rate > max_refresh) - max_refresh = refresh_rate; + refresh_rate = (pipe->stream->timing.pix_clk_100hz * (uint64_t)100 + + pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1); + refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total); + refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total); + + if ((uint32_t)refresh_rate < min_refresh) + min_refresh = (uint32_t)refresh_rate; + if ((uint32_t)refresh_rate > max_refresh) + max_refresh = (uint32_t)refresh_rate; subvp_count++; } } if (subvp_count == 2 && ((min_refresh < 120 && max_refresh < 120) || - (min_refresh >= 120 && max_refresh >= 120))) + (min_refresh >= subvp_high_refresh_list.min_refresh && + max_refresh <= subvp_high_refresh_list.max_refresh))) result = true; return result; @@ -1715,8 +1721,8 @@ bool dcn32_internal_validate_bw(struct dc *dc, if (vba->ODMCombineEnabled[vba->pipe_plane[pipe_idx]] != dm_odm_combine_mode_disabled && !dc->config.enable_windowed_mpo_odm && pipe->plane_state && mpo_pipe - && memcmp(&mpo_pipe->plane_res.scl_data.recout, - &pipe->plane_res.scl_data.recout, + && memcmp(&mpo_pipe->plane_state->clip_rect, + &pipe->stream->src, sizeof(struct rect)) != 0) { ASSERT(mpo_pipe->plane_state != pipe->plane_state); goto validate_fail; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c index a50e7f4dce42..ecea008f19d3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c @@ -3459,6 +3459,7 @@ bool dml32_CalculatePrefetchSchedule( double TimeForFetchingMetaPTE = 0; double TimeForFetchingRowInVBlank = 0; double LinesToRequestPrefetchPixelData = 0; + double LinesForPrefetchBandwidth = 0; unsigned int HostVMDynamicLevelsTrips; double trip_to_mem; double Tvm_trips; @@ -3888,11 +3889,15 @@ bool dml32_CalculatePrefetchSchedule( TimeForFetchingMetaPTE = Tvm_oto; TimeForFetchingRowInVBlank = Tr0_oto; *PrefetchBandwidth = prefetch_bw_oto; + /* Clamp to oto for bandwidth calculation */ + LinesForPrefetchBandwidth = dst_y_prefetch_oto; } else { *DestinationLinesForPrefetch = dst_y_prefetch_equ; TimeForFetchingMetaPTE = Tvm_equ; TimeForFetchingRowInVBlank = Tr0_equ; *PrefetchBandwidth = prefetch_bw_equ; + /* Clamp to equ for bandwidth calculation */ + LinesForPrefetchBandwidth = dst_y_prefetch_equ; } *DestinationLinesToRequestVMInVBlank = dml_ceil(4.0 * TimeForFetchingMetaPTE / LineTime, 1.0) / 4.0; @@ -3900,7 +3905,7 @@ bool dml32_CalculatePrefetchSchedule( *DestinationLinesToRequestRowInVBlank = dml_ceil(4.0 * TimeForFetchingRowInVBlank / LineTime, 1.0) / 4.0; - LinesToRequestPrefetchPixelData = *DestinationLinesForPrefetch - + LinesToRequestPrefetchPixelData = LinesForPrefetchBandwidth - *DestinationLinesToRequestVMInVBlank - 2 * *DestinationLinesToRequestRowInVBlank; #ifdef __DML_VBA_DEBUG__ @@ -4124,7 +4129,7 @@ void dml32_CalculateFlipSchedule( unsigned int HostVMDynamicLevelsTrips; double TimeForFetchingMetaPTEImmediateFlip; double TimeForFetchingRowInVBlankImmediateFlip; - double ImmediateFlipBW; + double ImmediateFlipBW = 1.0; if (GPUVMEnable == true && HostVMEnable == true) HostVMDynamicLevelsTrips = HostVMMaxNonCachedPageTableLevels; diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c index 58dd62cce4bb..3966845c7694 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -40,6 +40,8 @@ static bool dsc_policy_enable_dsc_when_not_needed; static bool dsc_policy_disable_dsc_stream_overhead; +static bool disable_128b_132b_stream_overhead; + #ifndef MAX #define MAX(X, Y) ((X) > (Y) ? (X) : (Y)) #endif @@ -47,8 +49,44 @@ static bool dsc_policy_disable_dsc_stream_overhead; #define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) #endif +/* Need to account for padding due to pixel-to-symbol packing + * for uncompressed 128b/132b streams. + */ +static uint32_t apply_128b_132b_stream_overhead( + const struct dc_crtc_timing *timing, const uint32_t kbps) +{ + uint32_t total_kbps = kbps; + + if (disable_128b_132b_stream_overhead) + return kbps; + + if (!timing->flags.DSC) { + struct fixed31_32 bpp; + struct fixed31_32 overhead_factor; + + bpp = dc_fixpt_from_int(kbps); + bpp = dc_fixpt_div_int(bpp, timing->pix_clk_100hz / 10); + + /* Symbols_per_HActive = HActive * bpp / (4 lanes * 32-bit symbol size) + * Overhead_factor = ceil(Symbols_per_HActive) / Symbols_per_HActive + */ + overhead_factor = dc_fixpt_from_int(timing->h_addressable); + overhead_factor = dc_fixpt_mul(overhead_factor, bpp); + overhead_factor = dc_fixpt_div_int(overhead_factor, 128); + overhead_factor = dc_fixpt_div( + dc_fixpt_from_int(dc_fixpt_ceil(overhead_factor)), + overhead_factor); + + total_kbps = dc_fixpt_ceil( + dc_fixpt_mul_int(overhead_factor, total_kbps)); + } + + return total_kbps; +} + uint32_t dc_bandwidth_in_kbps_from_timing( - const struct dc_crtc_timing *timing) + const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding) { uint32_t bits_per_channel = 0; uint32_t kbps; @@ -96,6 +134,9 @@ uint32_t dc_bandwidth_in_kbps_from_timing( kbps = kbps * 2 / 3; } + if (link_encoding == DC_LINK_ENCODING_DP_128b_132b) + kbps = apply_128b_132b_stream_overhead(timing, kbps); + return kbps; } @@ -107,6 +148,7 @@ static bool decide_dsc_bandwidth_range( const uint32_t num_slices_h, const struct dsc_enc_caps *dsc_caps, const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_bw_range *range); static uint32_t compute_bpp_x16_from_target_bandwidth( @@ -133,6 +175,7 @@ static bool setup_dsc_config( int target_bandwidth_kbps, const struct dc_crtc_timing *timing, const struct dc_dsc_config_options *options, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_config *dsc_cfg); static bool dsc_buff_block_size_from_dpcd(int dpcd_buff_block_size, int *buff_block_size) @@ -398,6 +441,7 @@ bool dc_dsc_compute_bandwidth_range( uint32_t max_bpp_x16, const struct dsc_dec_dpcd_caps *dsc_sink_caps, const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_bw_range *range) { bool is_dsc_possible = false; @@ -417,11 +461,11 @@ bool dc_dsc_compute_bandwidth_range( if (is_dsc_possible) is_dsc_possible = setup_dsc_config(dsc_sink_caps, &dsc_enc_caps, 0, timing, - &options, &config); + &options, link_encoding, &config); if (is_dsc_possible) is_dsc_possible = decide_dsc_bandwidth_range(min_bpp_x16, max_bpp_x16, - config.num_slices_h, &dsc_common_caps, timing, range); + config.num_slices_h, &dsc_common_caps, timing, link_encoding, range); return is_dsc_possible; } @@ -557,6 +601,7 @@ static bool decide_dsc_bandwidth_range( const uint32_t num_slices_h, const struct dsc_enc_caps *dsc_caps, const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_bw_range *range) { uint32_t preferred_bpp_x16 = timing->dsc_fixed_bits_per_pixel_x16; @@ -586,7 +631,7 @@ static bool decide_dsc_bandwidth_range( /* populate output structure */ if (range->max_target_bpp_x16 >= range->min_target_bpp_x16 && range->min_target_bpp_x16 > 0) { /* native stream bandwidth */ - range->stream_kbps = dc_bandwidth_in_kbps_from_timing(timing); + range->stream_kbps = dc_bandwidth_in_kbps_from_timing(timing, link_encoding); /* max dsc target bpp */ range->max_kbps = dc_dsc_stream_bandwidth_in_kbps(timing, @@ -612,6 +657,7 @@ static bool decide_dsc_target_bpp_x16( const int target_bandwidth_kbps, const struct dc_crtc_timing *timing, const int num_slices_h, + const enum dc_link_encoding_format link_encoding, int *target_bpp_x16) { struct dc_dsc_bw_range range; @@ -619,7 +665,7 @@ static bool decide_dsc_target_bpp_x16( *target_bpp_x16 = 0; if (decide_dsc_bandwidth_range(policy->min_target_bpp * 16, policy->max_target_bpp * 16, - num_slices_h, dsc_common_caps, timing, &range)) { + num_slices_h, dsc_common_caps, timing, link_encoding, &range)) { if (target_bandwidth_kbps >= range.stream_kbps) { if (policy->enable_dsc_when_not_needed) /* enable max bpp even dsc is not needed */ @@ -796,6 +842,7 @@ static bool setup_dsc_config( int target_bandwidth_kbps, const struct dc_crtc_timing *timing, const struct dc_dsc_config_options *options, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_config *dsc_cfg) { struct dsc_enc_caps dsc_common_caps; @@ -995,6 +1042,7 @@ static bool setup_dsc_config( target_bandwidth_kbps, timing, num_slices_h, + link_encoding, &target_bpp); dsc_cfg->bits_per_pixel = target_bpp; } @@ -1023,6 +1071,7 @@ bool dc_dsc_compute_config( const struct dc_dsc_config_options *options, uint32_t target_bandwidth_kbps, const struct dc_crtc_timing *timing, + const enum dc_link_encoding_format link_encoding, struct dc_dsc_config *dsc_cfg) { bool is_dsc_possible = false; @@ -1032,7 +1081,7 @@ bool dc_dsc_compute_config( is_dsc_possible = setup_dsc_config(dsc_sink_caps, &dsc_enc_caps, target_bandwidth_kbps, - timing, options, dsc_cfg); + timing, options, link_encoding, dsc_cfg); return is_dsc_possible; } @@ -1165,6 +1214,11 @@ void dc_dsc_policy_set_disable_dsc_stream_overhead(bool disable) dsc_policy_disable_dsc_stream_overhead = disable; } +void dc_set_disable_128b_132b_stream_overhead(bool disable) +{ + disable_128b_132b_stream_overhead = disable; +} + void dc_dsc_get_default_config_option(const struct dc *dc, struct dc_dsc_config_options *options) { options->dsc_min_slice_height_override = dc->debug.dsc_min_slice_height_override; diff --git a/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h b/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h index 59884ef651b3..4a2bf81286d8 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/ddc_regs.h @@ -31,21 +31,21 @@ /****************************** new register headers */ /*** following in header */ -#define DDC_GPIO_REG_LIST_ENTRY(type,cd,id) \ +#define DDC_GPIO_REG_LIST_ENTRY(type, cd, id) \ .type ## _reg = REG(DC_GPIO_DDC ## id ## _ ## type),\ .type ## _mask = DC_GPIO_DDC ## id ## _ ## type ## __DC_GPIO_DDC ## id ## cd ## _ ## type ## _MASK,\ .type ## _shift = DC_GPIO_DDC ## id ## _ ## type ## __DC_GPIO_DDC ## id ## cd ## _ ## type ## __SHIFT -#define DDC_GPIO_REG_LIST(cd,id) \ +#define DDC_GPIO_REG_LIST(cd, id) \ {\ - DDC_GPIO_REG_LIST_ENTRY(MASK,cd,id),\ - DDC_GPIO_REG_LIST_ENTRY(A,cd,id),\ - DDC_GPIO_REG_LIST_ENTRY(EN,cd,id),\ - DDC_GPIO_REG_LIST_ENTRY(Y,cd,id)\ + DDC_GPIO_REG_LIST_ENTRY(MASK, cd, id),\ + DDC_GPIO_REG_LIST_ENTRY(A, cd, id),\ + DDC_GPIO_REG_LIST_ENTRY(EN, cd, id),\ + DDC_GPIO_REG_LIST_ENTRY(Y, cd, id)\ } -#define DDC_REG_LIST(cd,id) \ - DDC_GPIO_REG_LIST(cd,id),\ +#define DDC_REG_LIST(cd, id) \ + DDC_GPIO_REG_LIST(cd, id),\ .ddc_setup = REG(DC_I2C_DDC ## id ## _SETUP) #define DDC_REG_LIST_DCN2(cd, id) \ @@ -54,34 +54,34 @@ .phy_aux_cntl = REG(PHY_AUX_CNTL), \ .dc_gpio_aux_ctrl_5 = REG(DC_GPIO_AUX_CTRL_5) -#define DDC_GPIO_VGA_REG_LIST_ENTRY(type,cd)\ +#define DDC_GPIO_VGA_REG_LIST_ENTRY(type, cd)\ .type ## _reg = REG(DC_GPIO_DDCVGA_ ## type),\ .type ## _mask = DC_GPIO_DDCVGA_ ## type ## __DC_GPIO_DDCVGA ## cd ## _ ## type ## _MASK,\ .type ## _shift = DC_GPIO_DDCVGA_ ## type ## __DC_GPIO_DDCVGA ## cd ## _ ## type ## __SHIFT #define DDC_GPIO_VGA_REG_LIST(cd) \ {\ - DDC_GPIO_VGA_REG_LIST_ENTRY(MASK,cd),\ - DDC_GPIO_VGA_REG_LIST_ENTRY(A,cd),\ - DDC_GPIO_VGA_REG_LIST_ENTRY(EN,cd),\ - DDC_GPIO_VGA_REG_LIST_ENTRY(Y,cd)\ + DDC_GPIO_VGA_REG_LIST_ENTRY(MASK, cd),\ + DDC_GPIO_VGA_REG_LIST_ENTRY(A, cd),\ + DDC_GPIO_VGA_REG_LIST_ENTRY(EN, cd),\ + DDC_GPIO_VGA_REG_LIST_ENTRY(Y, cd)\ } #define DDC_VGA_REG_LIST(cd) \ DDC_GPIO_VGA_REG_LIST(cd),\ .ddc_setup = mmDC_I2C_DDCVGA_SETUP -#define DDC_GPIO_I2C_REG_LIST_ENTRY(type,cd) \ +#define DDC_GPIO_I2C_REG_LIST_ENTRY(type, cd) \ .type ## _reg = REG(DC_GPIO_I2CPAD_ ## type),\ .type ## _mask = DC_GPIO_I2CPAD_ ## type ## __DC_GPIO_ ## cd ## _ ## type ## _MASK,\ .type ## _shift = DC_GPIO_I2CPAD_ ## type ## __DC_GPIO_ ## cd ## _ ## type ## __SHIFT #define DDC_GPIO_I2C_REG_LIST(cd) \ {\ - DDC_GPIO_I2C_REG_LIST_ENTRY(MASK,cd),\ - DDC_GPIO_I2C_REG_LIST_ENTRY(A,cd),\ - DDC_GPIO_I2C_REG_LIST_ENTRY(EN,cd),\ - DDC_GPIO_I2C_REG_LIST_ENTRY(Y,cd)\ + DDC_GPIO_I2C_REG_LIST_ENTRY(MASK, cd),\ + DDC_GPIO_I2C_REG_LIST_ENTRY(A, cd),\ + DDC_GPIO_I2C_REG_LIST_ENTRY(EN, cd),\ + DDC_GPIO_I2C_REG_LIST_ENTRY(Y, cd)\ } #define DDC_I2C_REG_LIST(cd) \ @@ -150,12 +150,12 @@ struct ddc_sh_mask { #define ddc_data_regs(id) \ {\ - DDC_REG_LIST(DATA,id)\ + DDC_REG_LIST(DATA, id)\ } #define ddc_clk_regs(id) \ {\ - DDC_REG_LIST(CLK,id)\ + DDC_REG_LIST(CLK, id)\ } #define ddc_vga_data_regs \ diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hpd_regs.h b/drivers/gpu/drm/amd/display/dc/gpio/hpd_regs.h index dcfdd71b2304..debb363cfcf4 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hpd_regs.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/hpd_regs.h @@ -36,17 +36,17 @@ #define ONE_MORE_5 6 -#define HPD_GPIO_REG_LIST_ENTRY(type,cd,id) \ +#define HPD_GPIO_REG_LIST_ENTRY(type, cd, id) \ .type ## _reg = REG(DC_GPIO_HPD_## type),\ .type ## _mask = DC_GPIO_HPD_ ## type ## __DC_GPIO_HPD ## id ## _ ## type ## _MASK,\ .type ## _shift = DC_GPIO_HPD_ ## type ## __DC_GPIO_HPD ## id ## _ ## type ## __SHIFT #define HPD_GPIO_REG_LIST(id) \ {\ - HPD_GPIO_REG_LIST_ENTRY(MASK,cd,id),\ - HPD_GPIO_REG_LIST_ENTRY(A,cd,id),\ - HPD_GPIO_REG_LIST_ENTRY(EN,cd,id),\ - HPD_GPIO_REG_LIST_ENTRY(Y,cd,id)\ + HPD_GPIO_REG_LIST_ENTRY(MASK, cd, id),\ + HPD_GPIO_REG_LIST_ENTRY(A, cd, id),\ + HPD_GPIO_REG_LIST_ENTRY(EN, cd, id),\ + HPD_GPIO_REG_LIST_ENTRY(Y, cd, id)\ } #define HPD_REG_LIST(id) \ diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 034610b74a37..eaad1260bfd1 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -65,6 +65,7 @@ struct resource_context; struct clk_bw_params; struct resource_funcs { + enum engine_id (*get_preferred_eng_id_dpia)(unsigned int dpia_index); void (*destroy)(struct resource_pool **pool); void (*link_init)(struct dc_link *link); struct panel_cntl*(*panel_cntl_create)( @@ -125,39 +126,15 @@ struct resource_funcs { struct dc *dc, struct dc_state *context); - /* - * Acquires a free pipe for the head pipe. - * The head pipe is first pipe in the current context that matches the stream - * and does not have a top pipe or prev_odm_pipe. - */ - struct pipe_ctx *(*acquire_idle_pipe_for_layer)( - struct dc_state *context, - const struct resource_pool *pool, - struct dc_stream_state *stream); - - /* - * Acquires a free pipe for the head pipe with some additional checks for odm. - * The head pipe is passed in as an argument unlike acquire_idle_pipe_for_layer - * where it is read from the context. So this allows us look for different - * idle_pipe if the head_pipes are different ( ex. in odm 2:1 when we have - * a left and right pipe ). - * - * It also checks the old context to see if: - * - * 1. a pipe has already been allocated for the head pipe. If so, it will - * try to select that pipe as the idle pipe if it is available in the current - * context. - * 2. if the head_pipe is on the left, it will check if the right pipe has - * a pipe already allocated. If so, it will not use that pipe if it is - * selected as the idle pipe. - */ - struct pipe_ctx *(*acquire_idle_pipe_for_head_pipe_in_layer)( - struct dc_state *context, + struct pipe_ctx *(*acquire_free_pipe_as_secondary_dpp_pipe)( + const struct dc_state *cur_ctx, + struct dc_state *new_ctx, const struct resource_pool *pool, - struct dc_stream_state *stream, - struct pipe_ctx *head_pipe); + const struct pipe_ctx *opp_head_pipe); - enum dc_status (*validate_plane)(const struct dc_plane_state *plane_state, struct dc_caps *caps); + enum dc_status (*validate_plane)( + const struct dc_plane_state *plane_state, + struct dc_caps *caps); enum dc_status (*add_stream_to_ctx)( struct dc *dc, @@ -304,6 +281,8 @@ struct resource_pool { struct dmcu *dmcu; struct dmub_psr *psr; + struct dmub_replay *replay; + struct abm *multiple_abms[MAX_PIPES]; const struct resource_funcs *funcs; @@ -572,6 +551,23 @@ struct dc_state { } perf_params; }; +struct replay_context { + /* ddc line */ + enum channel_id aux_inst; + /* Transmitter id */ + enum transmitter digbe_inst; + /* Engine Id is used for Dig Be source select */ + enum engine_id digfe_inst; + /* Controller Id used for Dig Fe source select */ + enum controller_id controllerId; + unsigned int line_time_in_ns; +}; + +enum dc_replay_enable { + DC_REPLAY_DISABLE = 0, + DC_REPLAY_ENABLE = 1, +}; + struct dc_bounding_box_max_clk { int max_dcfclk_mhz; int max_dispclk_mhz; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h index d2190a3320f6..33db15d69f23 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h @@ -27,6 +27,8 @@ #include "dm_services_types.h" +struct abm_save_restore; + struct abm { struct dc_context *ctx; const struct abm_funcs *funcs; @@ -55,6 +57,10 @@ struct abm_funcs { unsigned int bytes, unsigned int inst); bool (*set_abm_pause)(struct abm *abm, bool pause, unsigned int panel_inst, unsigned int otg_inst); + bool (*save_restore)( + struct abm *abm, + unsigned int panel_inst, + struct abm_save_restore *pData); bool (*set_pipe_ex)(struct abm *abm, unsigned int otg_inst, unsigned int option, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h b/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h index 7254182b7c72..af6b9509d09d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/aux_engine.h @@ -172,8 +172,6 @@ struct aux_engine_funcs { struct aux_engine *engine, uint8_t *returned_bytes); bool (*is_engine_available)(struct aux_engine *engine); - enum i2caux_engine_type (*get_engine_type)( - const struct aux_engine *engine); bool (*acquire)( struct aux_engine *engine, struct ddc *ddc); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h index 8dc804bbe98b..3e2f0f64c98c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h @@ -123,6 +123,11 @@ struct dccg_funcs { struct dccg *dccg, int hpo_le_inst); + void (*set_symclk32_le_root_clock_gating)( + struct dccg *dccg, + int hpo_le_inst, + bool enable); + void (*set_physymclk)( struct dccg *dccg, int phy_inst, @@ -167,6 +172,16 @@ struct dccg_funcs { struct dccg *dccg, unsigned int dpp_inst, bool clock_on); + + void (*enable_symclk_se)( + struct dccg *dccg, + uint32_t stream_enc_inst, + uint32_t link_enc_inst); + + void (*disable_symclk_se)( + struct dccg *dccg, + uint32_t stream_enc_inst, + uint32_t link_enc_inst); }; #endif //__DAL_DCCG_H__ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h index aaa293613846..f5677dbb4e7d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -49,6 +49,8 @@ struct dcn_hubbub_wm_set { uint32_t dram_clk_change; uint32_t usr_retrain; uint32_t fclk_pstate_change; + uint32_t sr_enter_exit_Z8; + uint32_t sr_enter_Z8; }; struct dcn_hubbub_wm { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h index c4fbbf08ef86..a6dedf3c7d74 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/stream_encoder.h @@ -269,6 +269,7 @@ struct stream_encoder_funcs { struct stream_encoder *enc, unsigned int pix_per_container); void (*enable_fifo)(struct stream_encoder *enc); void (*disable_fifo)(struct stream_encoder *enc); + void (*map_stream_to_link)(struct stream_encoder *enc, uint32_t stream_enc_inst, uint32_t link_enc_inst); }; struct hpo_dp_stream_encoder_state { diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h index f839494d59d8..e3e8c76c17cf 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/link.h +++ b/drivers/gpu/drm/amd/display/dc/inc/link.h @@ -179,6 +179,10 @@ struct link_service { int (*aux_transfer_raw)(struct ddc_service *ddc, struct aux_payload *payload, enum aux_return_code_type *operation_result); + bool (*configure_fixed_vs_pe_retimer)( + struct ddc_service *ddc, + const uint8_t *data, + uint32_t len); bool (*aux_transfer_with_retries_no_mutex)(struct ddc_service *ddc, struct aux_payload *payload); bool (*is_in_aux_transaction_mode)(struct ddc_service *ddc); @@ -269,6 +273,20 @@ struct link_service { uint16_t psr_vtotal_su); void (*edp_get_psr_residency)( const struct dc_link *link, uint32_t *residency); + + bool (*edp_get_replay_state)( + const struct dc_link *link, uint64_t *state); + bool (*edp_set_replay_allow_active)(struct dc_link *dc_link, + const bool *enable, bool wait, bool force_static, + const unsigned int *power_opts); + bool (*edp_setup_replay)(struct dc_link *link, + const struct dc_stream_state *stream); + bool (*edp_set_coasting_vtotal)( + struct dc_link *link, uint16_t coasting_vtotal); + bool (*edp_replay_residency)(const struct dc_link *link, + unsigned int *residency, const bool is_start, + const bool is_alpm); + bool (*edp_wait_for_t12)(struct dc_link *link); bool (*edp_is_ilr_optimization_required)(struct dc_link *link, struct dc_crtc_timing *crtc_timing); diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index eaeb684c8a48..e546b9c506c1 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -142,10 +142,6 @@ struct clock_source *dc_resource_find_first_free_pll( struct resource_context *res_ctx, const struct resource_pool *pool); -struct pipe_ctx *resource_get_head_pipe_for_stream( - struct resource_context *res_ctx, - struct dc_stream_state *stream); - bool resource_attach_surfaces_to_context( struct dc_plane_state *const *plane_state, int surface_count, @@ -153,11 +149,232 @@ bool resource_attach_surfaces_to_context( struct dc_state *context, const struct resource_pool *pool); -struct pipe_ctx *find_idle_secondary_pipe( +#define FREE_PIPE_INDEX_NOT_FOUND -1 + +/* + * pipe types are identified based on MUXes in DCN front end that are capable + * of taking input from one DCN pipeline to another DCN pipeline. The name is + * in a form of XXXX_YYYY, where XXXX is the DCN front end hardware block the + * pipeline ends with and YYYY is the rendering role that the pipe is in. + * + * For instance OTG_MASTER is a pipe ending with OTG hardware block in its + * pipeline and it is in a role of a master pipe for timing generation. + * + * For quick reference a diagram of each pipe type's areas of responsibility + * for outputting timings on the screen is shown below: + * + * Timing Active for Stream 0 + * __________________________________________________ + * |OTG master 0 (OPP head 0)|OPP head 2 (DPP pipe 2) | + * | (DPP pipe 0)| | + * | Top Plane 0 | | + * | ______________|____ | + * | |DPP pipe 1 |DPP | | + * | | |pipe| | + * | | Bottom |3 | | + * | | Plane 1 | | | + * | | | | | + * | |______________|____| | + * | | | + * | | | + * | ODM slice 0 | ODM slice 1 | + * |_________________________|________________________| + * + * Timing Active for Stream 1 + * __________________________________________________ + * |OTG master 4 (OPP head 4) | + * | | + * | | + * | | + * | | + * | | + * | Blank Pixel Data | + * | (generated by DPG4) | + * | | + * | | + * | | + * | | + * | | + * |__________________________________________________| + * + * Inter-pipe Relation + * __________________________________________________ + * |PIPE IDX| DPP PIPES | OPP HEADS | OTG MASTER | + * | | plane 0 | slice 0 | | + * | 0 | -------------MPC---------ODM----------- | + * | | plane 1 | | | | | + * | 1 | ------------- | | | | + * | | plane 0 | slice 1 | | | + * | 2 | -------------MPC--------- | | + * | | plane 1 | | | | + * | 3 | ------------- | | | + * | | | blank | | + * | 4 | | ----------------------- | + * | | | | | + * | 5 | (FREE) | | | + * |________|_______________|___________|_____________| + */ +enum pipe_type { + /* free pipe - free pipe is an uninitialized pipe without a stream + * associated with it. It is a free DCN pipe resource. It can be + * acquired as any type of pipe. + */ + FREE_PIPE, + + /* OTG master pipe - the master pipe of its OPP head pipes with a + * functional OTG. It merges all its OPP head pipes pixel data in ODM + * block and output to backend DIG. OTG master pipe is responsible for + * generating entire crtc timing to backend DIG. An OTG master pipe may + * or may not have a plane. If it has a plane it blends it as the left + * most MPC slice of the top most layer. If it doesn't have a plane it + * can output pixel data from its OPP head pipes' test pattern + * generators (DPG) such as solid black pixel data to blank the screen. + */ + OTG_MASTER, + + /* OPP head pipe - the head pipe of an MPC blending tree with a + * functional OPP outputting to an OTG. OPP head pipe is responsible for + * processing output pixels in its own ODM slice. It may or may not have + * a plane. If it has a plane it blends it as the top most layer within + * its own ODM slice. If it doesn't have a plane it can output pixel + * data from its DPG such as solid black pixel data to blank the pixel + * data in its own ODM slice. OTG master pipe is also an OPP head pipe + * but with more responsibility. + */ + OPP_HEAD, + + /* DPP pipe - the pipe with a functional DPP outputting to an OPP head + * pipe's MPC. DPP pipe is responsible for processing pixel data from + * its own MPC slice of a plane. It must be connected to an OPP head + * pipe and it must have a plane associated with it. + */ + DPP_PIPE, +}; + +/* + * Determine if the input pipe ctx is of a pipe type. + * return - true if pipe ctx is of the input type. + */ +bool resource_is_pipe_type(const struct pipe_ctx *pipe_ctx, enum pipe_type type); + +/* + * Determine if the input pipe ctx is used for rendering a plane with MPCC + * combine. MPCC combine is a hardware feature to combine multiple DPP pipes + * into a single plane. It is typically used for bypassing pipe bandwidth + * limitation for rendering a very large plane or saving power by reducing UCLK + * and DPPCLK speeds. + * + * For instance in the Inter-pipe Relation diagram shown below, both PIPE 0 and + * 1 are for MPCC combine for plane 0 + * + * Inter-pipe Relation + * __________________________________________________ + * |PIPE IDX| DPP PIPES | OPP HEADS | OTG MASTER | + * | | plane 0 | | | + * | 0 | -------------MPC----------------------- | + * | | plane 0 | | | | + * | 1 | ------------- | | | + * |________|_______________|___________|_____________| + * + * return - true if pipe ctx is used for mpcc combine. + */ +bool resource_is_for_mpcc_combine(const struct pipe_ctx *pipe_ctx); + +/* + * Look for a free pipe in new resource context that is used as a secondary DPP + * pipe in MPC blending tree associated with input OPP head pipe. + * + * return - FREE_PIPE_INDEX_NOT_FOUND if free pipe is not found, otherwise + * pipe idx of the free pipe + */ +int resource_find_free_pipe_used_in_cur_mpc_blending_tree( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct pipe_ctx *cur_opp_head); + +/* + * Look for a free pipe in new resource context that is not used in current + * resource context. + * + * return - FREE_PIPE_INDEX_NOT_FOUND if free pipe is not found, otherwise + * pipe idx of the free pipe + */ +int recource_find_free_pipe_not_used_in_cur_res_ctx( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct resource_pool *pool); + +/* + * Look for a free pipe in new resource context that is used as a secondary DPP + * pipe in any MPCC combine in current resource context. + * return - FREE_PIPE_INDEX_NOT_FOUND if free pipe is not found, otherwise + * pipe idx of the free pipe + */ +int resource_find_free_pipe_used_as_cur_sec_dpp_in_mpcc_combine( + const struct resource_context *cur_res_ctx, + struct resource_context *new_res_ctx, + const struct resource_pool *pool); + +/* + * Look for any free pipe in new resource context. + * return - FREE_PIPE_INDEX_NOT_FOUND if free pipe is not found, otherwise + * pipe idx of the free pipe + */ +int resource_find_any_free_pipe(struct resource_context *new_res_ctx, + const struct resource_pool *pool); + +/* + * Legacy find free secondary pipe logic deprecated for newer DCNs as it doesn't + * find the most optimal free pipe to prevent from time consuming hardware state + * transitions. + */ +struct pipe_ctx *resource_find_free_secondary_pipe_legacy( struct resource_context *res_ctx, const struct resource_pool *pool, const struct pipe_ctx *primary_pipe); +/* + * Get number of MPC "cuts" of the plane associated with the pipe. MPC slice + * count is equal to MPC splits + 1. For example if a plane is cut 3 times, it + * will have 4 pieces of slice. + * return - 0 if pipe is not used for a plane with MPCC combine. otherwise + * the number of MPC "cuts" for the plane. + */ +int resource_get_num_mpc_splits(const struct pipe_ctx *pipe); + +/* + * Get number of ODM "cuts" of the timing associated with the pipe. ODM slice + * count is equal to ODM splits + 1. For example if a timing is cut 3 times, it + * will have 4 pieces of slice. + * return - 0 if pipe is not used for ODM combine. otherwise + * the number of ODM "cuts" for the timing. + */ +int resource_get_num_odm_splits(const struct pipe_ctx *pipe); + +/* + * Get the OTG master pipe in resource context associated with the stream. + * return - NULL if not found. Otherwise the OTG master pipe associated with the + * stream. + */ +struct pipe_ctx *resource_get_otg_master_for_stream( + struct resource_context *res_ctx, + struct dc_stream_state *stream); + +/* + * Get the OTG master pipe for the input pipe context. + * return - the OTG master pipe for the input pipe + * context. + */ +struct pipe_ctx *resource_get_otg_master(const struct pipe_ctx *pipe_ctx); + +/* + * Get the OPP head pipe for the input pipe context. + * return - the OPP head pipe for the input pipe + * context. + */ +struct pipe_ctx *resource_get_opp_head(const struct pipe_ctx *pipe_ctx); + + bool resource_validate_attach_surfaces( const struct dc_validation_set set[], int set_count, @@ -193,10 +410,6 @@ unsigned int resource_pixel_format_to_bpp(enum surface_pixel_format format); void get_audio_check(struct audio_info *aud_modes, struct audio_check *aud_chk); -int get_num_mpc_splits(struct pipe_ctx *pipe); - -int get_num_odm_splits(struct pipe_ctx *pipe); - bool get_temp_dp_link_res(struct dc_link *link, struct link_resource *link_res, struct dc_link_settings *link_settings); diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c b/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c index c923b2af8510..37bc98faa7a0 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn314/irq_service_dcn314.c @@ -38,10 +38,9 @@ #define DCN_BASE__INST0_SEG2 0x000034C0 -static enum dc_irq_source to_dal_irq_source_dcn314( - struct irq_service *irq_service, - uint32_t src_id, - uint32_t ext_id) +static enum dc_irq_source to_dal_irq_source_dcn314(struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) { switch (src_id) { case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: diff --git a/drivers/gpu/drm/amd/display/dc/link/Makefile b/drivers/gpu/drm/amd/display/dc/link/Makefile index a52b56e2859e..6af8a97d4a77 100644 --- a/drivers/gpu/drm/amd/display/dc/link/Makefile +++ b/drivers/gpu/drm/amd/display/dc/link/Makefile @@ -42,7 +42,8 @@ AMD_DISPLAY_FILES += $(AMD_DAL_LINK_ACCESSORIES) ############################################################################### # hwss ############################################################################### -LINK_HWSS = link_hwss_dio.o link_hwss_dpia.o link_hwss_hpo_dp.o +LINK_HWSS = link_hwss_dio.o link_hwss_dpia.o link_hwss_hpo_dp.o \ +link_hwss_dio_fixed_vs_pe_retimer.o link_hwss_hpo_fixed_vs_pe_retimer_dp.o AMD_DAL_LINK_HWSS = $(addprefix $(AMDDALPATH)/dc/link/hwss/, \ $(LINK_HWSS)) diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c index db9f1baa27e5..fe4282771cd0 100644 --- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c +++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c @@ -428,15 +428,24 @@ static void set_crtc_test_pattern(struct dc_link *link, stream->timing.display_color_depth; struct bit_depth_reduction_params params; struct output_pixel_processor *opp = pipe_ctx->stream_res.opp; - int width = pipe_ctx->stream->timing.h_addressable + + struct pipe_ctx *odm_pipe; + int odm_cnt = 1; + int h_active = pipe_ctx->stream->timing.h_addressable + pipe_ctx->stream->timing.h_border_left + pipe_ctx->stream->timing.h_border_right; - int height = pipe_ctx->stream->timing.v_addressable + + int v_active = pipe_ctx->stream->timing.v_addressable + pipe_ctx->stream->timing.v_border_bottom + pipe_ctx->stream->timing.v_border_top; + int odm_slice_width, last_odm_slice_width, offset = 0; memset(¶ms, 0, sizeof(params)); + for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) + odm_cnt++; + + odm_slice_width = h_active / odm_cnt; + last_odm_slice_width = h_active - odm_slice_width * (odm_cnt - 1); + switch (test_pattern) { case DP_TEST_PATTERN_COLOR_SQUARES: controller_test_pattern = @@ -473,16 +482,13 @@ static void set_crtc_test_pattern(struct dc_link *link, { /* disable bit depth reduction */ pipe_ctx->stream->bit_depth_params = params; - opp->funcs->opp_program_bit_depth_reduction(opp, ¶ms); - if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) + if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) { + opp->funcs->opp_program_bit_depth_reduction(opp, ¶ms); pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg, controller_test_pattern, color_depth); - else if (link->dc->hwss.set_disp_pattern_generator) { - struct pipe_ctx *odm_pipe; + } else if (link->dc->hwss.set_disp_pattern_generator) { enum controller_dp_color_space controller_color_space; - int opp_cnt = 1; - int offset = 0; - int dpg_width = width; + struct output_pixel_processor *odm_opp; switch (test_pattern_color_space) { case DP_TEST_PATTERN_COLOR_SPACE_RGB: @@ -502,24 +508,9 @@ static void set_crtc_test_pattern(struct dc_link *link, break; } - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) - opp_cnt++; - dpg_width = width / opp_cnt; - offset = dpg_width; - - link->dc->hwss.set_disp_pattern_generator(link->dc, - pipe_ctx, - controller_test_pattern, - controller_color_space, - color_depth, - NULL, - dpg_width, - height, - 0); - - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { - struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; - + odm_pipe = pipe_ctx; + while (odm_pipe->next_odm_pipe) { + odm_opp = odm_pipe->stream_res.opp; odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); link->dc->hwss.set_disp_pattern_generator(link->dc, odm_pipe, @@ -527,11 +518,23 @@ static void set_crtc_test_pattern(struct dc_link *link, controller_color_space, color_depth, NULL, - dpg_width, - height, + odm_slice_width, + v_active, offset); - offset += offset; + offset += odm_slice_width; + odm_pipe = odm_pipe->next_odm_pipe; } + odm_opp = odm_pipe->stream_res.opp; + odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); + link->dc->hwss.set_disp_pattern_generator(link->dc, + odm_pipe, + controller_test_pattern, + controller_color_space, + color_depth, + NULL, + last_odm_slice_width, + v_active, + offset); } } break; @@ -540,23 +543,17 @@ static void set_crtc_test_pattern(struct dc_link *link, /* restore bitdepth reduction */ resource_build_bit_depth_reduction_params(pipe_ctx->stream, ¶ms); pipe_ctx->stream->bit_depth_params = params; - opp->funcs->opp_program_bit_depth_reduction(opp, ¶ms); - if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) + if (pipe_ctx->stream_res.tg->funcs->set_test_pattern) { + opp->funcs->opp_program_bit_depth_reduction(opp, ¶ms); pipe_ctx->stream_res.tg->funcs->set_test_pattern(pipe_ctx->stream_res.tg, - CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, - color_depth); - else if (link->dc->hwss.set_disp_pattern_generator) { - struct pipe_ctx *odm_pipe; - int opp_cnt = 1; - int dpg_width; - - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) - opp_cnt++; - - dpg_width = width / opp_cnt; - for (odm_pipe = pipe_ctx->next_odm_pipe; odm_pipe; odm_pipe = odm_pipe->next_odm_pipe) { - struct output_pixel_processor *odm_opp = odm_pipe->stream_res.opp; + CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, + color_depth); + } else if (link->dc->hwss.set_disp_pattern_generator) { + struct output_pixel_processor *odm_opp; + odm_pipe = pipe_ctx; + while (odm_pipe->next_odm_pipe) { + odm_opp = odm_pipe->stream_res.opp; odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); link->dc->hwss.set_disp_pattern_generator(link->dc, odm_pipe, @@ -564,19 +561,23 @@ static void set_crtc_test_pattern(struct dc_link *link, CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, - dpg_width, - height, - 0); + odm_slice_width, + v_active, + offset); + offset += odm_slice_width; + odm_pipe = odm_pipe->next_odm_pipe; } + odm_opp = odm_pipe->stream_res.opp; + odm_opp->funcs->opp_program_bit_depth_reduction(odm_opp, ¶ms); link->dc->hwss.set_disp_pattern_generator(link->dc, - pipe_ctx, + odm_pipe, CONTROLLER_DP_TEST_PATTERN_VIDEOMODE, CONTROLLER_DP_COLOR_SPACE_UDEFINED, color_depth, NULL, - dpg_width, - height, - 0); + last_odm_slice_width, + v_active, + offset); } } break; @@ -674,7 +675,8 @@ bool dp_set_test_pattern( if (pipes[i].stream == NULL) continue; - if (pipes[i].stream->link == link && !pipes[i].top_pipe && !pipes[i].prev_odm_pipe) { + if (resource_is_pipe_type(&pipes[i], OTG_MASTER) && + pipes[i].stream->link == link) { pipe_ctx = &pipes[i]; break; } @@ -702,6 +704,7 @@ bool dp_set_test_pattern( /* Reset Test Pattern state */ link->test_pattern_enabled = false; + link->current_test_pattern = test_pattern; return true; } @@ -739,6 +742,7 @@ bool dp_set_test_pattern( if (test_pattern != DP_TEST_PATTERN_VIDEO_MODE) { /* Set Test Pattern state */ link->test_pattern_enabled = true; + link->current_test_pattern = test_pattern; if (p_link_settings != NULL) dpcd_set_link_settings(link, p_link_settings); @@ -937,6 +941,7 @@ bool dp_set_test_pattern( /* Set Test Pattern state */ link->test_pattern_enabled = true; + link->current_test_pattern = test_pattern; } return true; diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c index bebf9c4c8702..1328a0ade342 100644 --- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c @@ -46,6 +46,9 @@ void setup_dio_stream_encoder(struct pipe_ctx *pipe_ctx) if (dc_is_dp_signal(pipe_ctx->stream->signal)) pipe_ctx->stream->ctx->dc->link_srv->dp_trace_source_sequence(pipe_ctx->stream->link, DPCD_SOURCE_SEQ_AFTER_CONNECT_DIG_FE_BE); + if (stream_enc->funcs->map_stream_to_link) + stream_enc->funcs->map_stream_to_link(stream_enc, + stream_enc->stream_enc_inst, link_enc->transmitter - TRANSMITTER_UNIPHY_A); if (stream_enc->funcs->enable_fifo) stream_enc->funcs->enable_fifo(stream_enc); } @@ -163,7 +166,7 @@ void set_dio_dp_lane_settings(struct dc_link *link, link_enc->funcs->dp_set_lane_settings(link_enc, link_settings, lane_settings); } -static void update_dio_stream_allocation_table(struct dc_link *link, +void update_dio_stream_allocation_table(struct dc_link *link, const struct link_resource *link_res, const struct link_mst_stream_allocation_table *table) { diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h index 8b8a099feeb0..f4633d3cf9b9 100644 --- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h @@ -55,5 +55,8 @@ void setup_dio_audio_output(struct pipe_ctx *pipe_ctx, struct audio_output *audio_output, uint32_t audio_inst); void enable_dio_audio_packet(struct pipe_ctx *pipe_ctx); void disable_dio_audio_packet(struct pipe_ctx *pipe_ctx); +void update_dio_stream_allocation_table(struct dc_link *link, + const struct link_resource *link_res, + const struct link_mst_stream_allocation_table *table); #endif /* __LINK_HWSS_DIO_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.c new file mode 100644 index 000000000000..b659baa23147 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.c @@ -0,0 +1,200 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "link_hwss_dio.h" +#include "link_hwss_dio_fixed_vs_pe_retimer.h" +#include "link_enc_cfg.h" + +uint8_t dp_dio_fixed_vs_pe_retimer_lane_cfg_to_hw_cfg(struct dc_link *link) +{ + // TODO: Get USB-C cable orientation + if (link->cur_link_settings.lane_count == LANE_COUNT_FOUR) + return 0xF2; + else + return 0x12; +} + +void dp_dio_fixed_vs_pe_retimer_exit_manual_automation(struct dc_link *link) +{ + const uint8_t dp_type = dp_dio_fixed_vs_pe_retimer_lane_cfg_to_hw_cfg(link); + const uint8_t vendor_lttpr_exit_manual_automation_0[4] = {0x1, 0x11, 0x0, 0x06}; + const uint8_t vendor_lttpr_exit_manual_automation_1[4] = {0x1, 0x50, dp_type, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_2[4] = {0x1, 0x50, 0x50, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_3[4] = {0x1, 0x51, 0x50, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_4[4] = {0x1, 0x10, 0x58, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_5[4] = {0x1, 0x10, 0x59, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_6[4] = {0x1, 0x30, 0x51, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_7[4] = {0x1, 0x30, 0x52, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_8[4] = {0x1, 0x30, 0x54, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_9[4] = {0x1, 0x30, 0x55, 0x0}; + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_0[0], sizeof(vendor_lttpr_exit_manual_automation_0)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_1[0], sizeof(vendor_lttpr_exit_manual_automation_1)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_2[0], sizeof(vendor_lttpr_exit_manual_automation_2)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_3[0], sizeof(vendor_lttpr_exit_manual_automation_3)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_4[0], sizeof(vendor_lttpr_exit_manual_automation_4)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_5[0], sizeof(vendor_lttpr_exit_manual_automation_5)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_6[0], sizeof(vendor_lttpr_exit_manual_automation_6)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_7[0], sizeof(vendor_lttpr_exit_manual_automation_7)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_8[0], sizeof(vendor_lttpr_exit_manual_automation_8)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_9[0], sizeof(vendor_lttpr_exit_manual_automation_9)); +} + +static bool set_dio_fixed_vs_pe_retimer_dp_link_test_pattern_override(struct dc_link *link, + const struct link_resource *link_res, struct encoder_set_dp_phy_pattern_param *tp_params, + const struct link_hwss *link_hwss) +{ + struct encoder_set_dp_phy_pattern_param hw_tp_params = { 0 }; + const uint8_t pltpat_custom[10] = {0x1F, 0x7C, 0xF0, 0xC1, 0x07, 0x1F, 0x7C, 0xF0, 0xC1, 0x07}; + const uint8_t vendor_lttpr_write_data_pg0[4] = {0x1, 0x11, 0x0, 0x0}; + const uint8_t vendor_lttpr_exit_manual_automation_0[4] = {0x1, 0x11, 0x0, 0x06}; + + + if (tp_params == NULL) + return false; + + if (link->current_test_pattern >= DP_TEST_PATTERN_SQUARE_BEGIN && + link->current_test_pattern <= DP_TEST_PATTERN_SQUARE_END) { + // Deprogram overrides from previous test pattern + dp_dio_fixed_vs_pe_retimer_exit_manual_automation(link); + } + + switch (tp_params->dp_phy_pattern) { + case DP_TEST_PATTERN_80BIT_CUSTOM: + if (tp_params->custom_pattern_size == 0 || memcmp(tp_params->custom_pattern, + pltpat_custom, tp_params->custom_pattern_size) != 0) + return false; + break; + case DP_TEST_PATTERN_D102: + break; + default: + if (link->current_test_pattern == DP_TEST_PATTERN_80BIT_CUSTOM || + link->current_test_pattern == DP_TEST_PATTERN_D102) + // Deprogram overrides from previous test pattern + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_0[0], + sizeof(vendor_lttpr_exit_manual_automation_0)); + + return false; + } + + hw_tp_params.dp_phy_pattern = tp_params->dp_phy_pattern; + hw_tp_params.dp_panel_mode = tp_params->dp_panel_mode; + + if (link_hwss->ext.set_dp_link_test_pattern) + link_hwss->ext.set_dp_link_test_pattern(link, link_res, &hw_tp_params); + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg0[0], sizeof(vendor_lttpr_write_data_pg0)); + + return true; +} + +static void set_dio_fixed_vs_pe_retimer_dp_link_test_pattern(struct dc_link *link, + const struct link_resource *link_res, + struct encoder_set_dp_phy_pattern_param *tp_params) +{ + struct link_encoder *link_enc = link_enc_cfg_get_link_enc(link); + + if (!set_dio_fixed_vs_pe_retimer_dp_link_test_pattern_override( + link, link_res, tp_params, get_dio_link_hwss())) { + link_enc->funcs->dp_set_phy_pattern(link_enc, tp_params); + } + link->dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_SET_SOURCE_PATTERN); +} + +void enable_dio_fixed_vs_pe_retimer_program_4lane_output(struct dc_link *link) +{ + const uint8_t vendor_lttpr_write_data_4lane_1[4] = {0x1, 0x6E, 0xF2, 0x19}; + const uint8_t vendor_lttpr_write_data_4lane_2[4] = {0x1, 0x6B, 0xF2, 0x01}; + const uint8_t vendor_lttpr_write_data_4lane_3[4] = {0x1, 0x6D, 0xF2, 0x18}; + const uint8_t vendor_lttpr_write_data_4lane_4[4] = {0x1, 0x6C, 0xF2, 0x03}; + const uint8_t vendor_lttpr_write_data_4lane_5[4] = {0x1, 0x03, 0xF3, 0x06}; + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_1[0], sizeof(vendor_lttpr_write_data_4lane_1)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_2[0], sizeof(vendor_lttpr_write_data_4lane_2)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_3[0], sizeof(vendor_lttpr_write_data_4lane_3)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_4[0], sizeof(vendor_lttpr_write_data_4lane_4)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_5[0], sizeof(vendor_lttpr_write_data_4lane_5)); +} + +static void enable_dio_fixed_vs_pe_retimer_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings) +{ + if (link_settings->lane_count == LANE_COUNT_FOUR) + enable_dio_fixed_vs_pe_retimer_program_4lane_output(link); + + enable_dio_dp_link_output(link, link_res, signal, clock_source, link_settings); +} + +static const struct link_hwss dio_fixed_vs_pe_retimer_link_hwss = { + .setup_stream_encoder = setup_dio_stream_encoder, + .reset_stream_encoder = reset_dio_stream_encoder, + .setup_stream_attribute = setup_dio_stream_attribute, + .disable_link_output = disable_dio_link_output, + .setup_audio_output = setup_dio_audio_output, + .enable_audio_packet = enable_dio_audio_packet, + .disable_audio_packet = disable_dio_audio_packet, + .ext = { + .set_throttled_vcp_size = set_dio_throttled_vcp_size, + .enable_dp_link_output = enable_dio_fixed_vs_pe_retimer_dp_link_output, + .set_dp_link_test_pattern = set_dio_fixed_vs_pe_retimer_dp_link_test_pattern, + .set_dp_lane_settings = set_dio_dp_lane_settings, + .update_stream_allocation_table = update_dio_stream_allocation_table, + }, +}; + +bool requires_fixed_vs_pe_retimer_dio_link_hwss(const struct dc_link *link) +{ + if (!(link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN)) + return false; + + if (!link->dpcd_caps.lttpr_caps.main_link_channel_coding.bits.DP_128b_132b_SUPPORTED) + return false; + + return true; +} + +const struct link_hwss *get_dio_fixed_vs_pe_retimer_link_hwss(void) +{ + return &dio_fixed_vs_pe_retimer_link_hwss; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h new file mode 100644 index 000000000000..9ac08a332540 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h @@ -0,0 +1,37 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_HWSS_DIO_FIXED_VS_PE_RETIMER_H__ +#define __LINK_HWSS_DIO_FIXED_VS_PE_RETIMER_H__ + +#include "link.h" + +uint32_t dp_dio_fixed_vs_pe_retimer_get_lttpr_write_address(struct dc_link *link); +uint8_t dp_dio_fixed_vs_pe_retimer_lane_cfg_to_hw_cfg(struct dc_link *link); +void dp_dio_fixed_vs_pe_retimer_exit_manual_automation(struct dc_link *link); +void enable_dio_fixed_vs_pe_retimer_program_4lane_output(struct dc_link *link); +bool requires_fixed_vs_pe_retimer_dio_link_hwss(const struct dc_link *link); +const struct link_hwss *get_dio_fixed_vs_pe_retimer_link_hwss(void); + +#endif /* __LINK_HWSS_DIO_FIXED_VS_PE_RETIMER_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c index 586fe25c1702..e1257404357b 100644 --- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.c @@ -28,7 +28,7 @@ #include "dccg.h" #include "clk_mgr.h" -static void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx, +void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx, struct fixed31_32 throttled_vcp_size) { struct hpo_dp_stream_encoder *hpo_dp_stream_encoder = @@ -41,7 +41,7 @@ static void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx, throttled_vcp_size); } -static void set_hpo_dp_hblank_min_symbol_width(struct pipe_ctx *pipe_ctx, +void set_hpo_dp_hblank_min_symbol_width(struct pipe_ctx *pipe_ctx, const struct dc_link_settings *link_settings, struct fixed31_32 throttled_vcp_size) { @@ -69,7 +69,7 @@ static void set_hpo_dp_hblank_min_symbol_width(struct pipe_ctx *pipe_ctx, hblank_min_symbol_width); } -static void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) +void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) { struct hpo_dp_stream_encoder *stream_enc = pipe_ctx->stream_res.hpo_dp_stream_enc; struct hpo_dp_link_encoder *link_enc = pipe_ctx->link_res.hpo_dp_link_enc; @@ -78,14 +78,14 @@ static void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) stream_enc->funcs->map_stream_to_link(stream_enc, stream_enc->inst, link_enc->inst); } -static void reset_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) +void reset_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx) { struct hpo_dp_stream_encoder *stream_enc = pipe_ctx->stream_res.hpo_dp_stream_enc; stream_enc->funcs->disable(stream_enc); } -static void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx) +void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx) { struct hpo_dp_stream_encoder *stream_enc = pipe_ctx->stream_res.hpo_dp_stream_enc; struct dc_stream_state *stream = pipe_ctx->stream; @@ -102,12 +102,17 @@ static void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx) DPCD_SOURCE_SEQ_AFTER_DP_STREAM_ATTR); } -static void enable_hpo_dp_link_output(struct dc_link *link, +void enable_hpo_dp_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal, enum clock_source_id clock_source, const struct dc_link_settings *link_settings) { + if (link->dc->res_pool->dccg->funcs->set_symclk32_le_root_clock_gating) + link->dc->res_pool->dccg->funcs->set_symclk32_le_root_clock_gating( + link->dc->res_pool->dccg, + link_res->hpo_dp_link_enc->inst, + true); link_res->hpo_dp_link_enc->funcs->enable_link_phy( link_res->hpo_dp_link_enc, link_settings, @@ -115,13 +120,18 @@ static void enable_hpo_dp_link_output(struct dc_link *link, link->link_enc->hpd_source); } -static void disable_hpo_dp_link_output(struct dc_link *link, +void disable_hpo_dp_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal) { link_res->hpo_dp_link_enc->funcs->link_disable(link_res->hpo_dp_link_enc); link_res->hpo_dp_link_enc->funcs->disable_link_phy( link_res->hpo_dp_link_enc, signal); + if (link->dc->res_pool->dccg->funcs->set_symclk32_le_root_clock_gating) + link->dc->res_pool->dccg->funcs->set_symclk32_le_root_clock_gating( + link->dc->res_pool->dccg, + link_res->hpo_dp_link_enc->inst, + false); } static void set_hpo_dp_link_test_pattern(struct dc_link *link, @@ -144,7 +154,7 @@ static void set_hpo_dp_lane_settings(struct dc_link *link, lane_settings[0].FFE_PRESET.raw); } -static void update_hpo_dp_stream_allocation_table(struct dc_link *link, +void update_hpo_dp_stream_allocation_table(struct dc_link *link, const struct link_resource *link_res, const struct link_mst_stream_allocation_table *table) { @@ -153,7 +163,7 @@ static void update_hpo_dp_stream_allocation_table(struct dc_link *link, table); } -static void setup_hpo_dp_audio_output(struct pipe_ctx *pipe_ctx, +void setup_hpo_dp_audio_output(struct pipe_ctx *pipe_ctx, struct audio_output *audio_output, uint32_t audio_inst) { pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_audio_setup( @@ -162,13 +172,13 @@ static void setup_hpo_dp_audio_output(struct pipe_ctx *pipe_ctx, &pipe_ctx->stream->audio_info); } -static void enable_hpo_dp_audio_packet(struct pipe_ctx *pipe_ctx) +void enable_hpo_dp_audio_packet(struct pipe_ctx *pipe_ctx) { pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_audio_enable( pipe_ctx->stream_res.hpo_dp_stream_enc); } -static void disable_hpo_dp_audio_packet(struct pipe_ctx *pipe_ctx) +void disable_hpo_dp_audio_packet(struct pipe_ctx *pipe_ctx) { if (pipe_ctx->stream_res.audio) pipe_ctx->stream_res.hpo_dp_stream_enc->funcs->dp_audio_disable( diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h index 3cbb94b41a23..1d3ed8ca83b5 100644 --- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h @@ -28,9 +28,35 @@ #include "link_hwss.h" #include "link.h" +void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx, + struct fixed31_32 throttled_vcp_size); +void set_hpo_dp_hblank_min_symbol_width(struct pipe_ctx *pipe_ctx, + const struct dc_link_settings *link_settings, + struct fixed31_32 throttled_vcp_size); +void set_hpo_dp_hblank_min_symbol_width(struct pipe_ctx *pipe_ctx, + const struct dc_link_settings *link_settings, + struct fixed31_32 throttled_vcp_size); +void setup_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx); +void reset_hpo_dp_stream_encoder(struct pipe_ctx *pipe_ctx); +void setup_hpo_dp_stream_attribute(struct pipe_ctx *pipe_ctx); +void enable_hpo_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings); +void disable_hpo_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal); +void update_hpo_dp_stream_allocation_table(struct dc_link *link, + const struct link_resource *link_res, + const struct link_mst_stream_allocation_table *table); +void setup_hpo_dp_audio_output(struct pipe_ctx *pipe_ctx, + struct audio_output *audio_output, uint32_t audio_inst); +void enable_hpo_dp_audio_packet(struct pipe_ctx *pipe_ctx); +void disable_hpo_dp_audio_packet(struct pipe_ctx *pipe_ctx); +const struct link_hwss *get_hpo_dp_link_hwss(void); bool can_use_hpo_dp_link_hwss(const struct dc_link *link, const struct link_resource *link_res); -const struct link_hwss *get_hpo_dp_link_hwss(void); #endif /* __LINK_HWSS_HPO_DP_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.c new file mode 100644 index 000000000000..b621b97711b6 --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.c @@ -0,0 +1,229 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#include "link_hwss_hpo_dp.h" +#include "link_hwss_hpo_fixed_vs_pe_retimer_dp.h" +#include "link_hwss_dio_fixed_vs_pe_retimer.h" + +static void dp_hpo_fixed_vs_pe_retimer_set_tx_ffe(struct dc_link *link, + const struct dc_lane_settings *hw_lane_settings) +{ + const uint8_t vendor_ffe_preset_table[16] = { + 0x01, 0x41, 0x61, 0x81, + 0xB1, 0x05, 0x35, 0x65, + 0x85, 0xA5, 0x09, 0x39, + 0x59, 0x89, 0x0F, 0x24}; + + const uint8_t ffe_mask[4] = { + (hw_lane_settings[0].FFE_PRESET.settings.no_deemphasis != 0 ? 0x0F : 0xFF) + & (hw_lane_settings[0].FFE_PRESET.settings.no_preshoot != 0 ? 0xF1 : 0xFF), + (hw_lane_settings[1].FFE_PRESET.settings.no_deemphasis != 0 ? 0x0F : 0xFF) + & (hw_lane_settings[1].FFE_PRESET.settings.no_preshoot != 0 ? 0xF1 : 0xFF), + (hw_lane_settings[2].FFE_PRESET.settings.no_deemphasis != 0 ? 0x0F : 0xFF) + & (hw_lane_settings[2].FFE_PRESET.settings.no_preshoot != 0 ? 0xF1 : 0xFF), + (hw_lane_settings[3].FFE_PRESET.settings.no_deemphasis != 0 ? 0x0F : 0xFF) + & (hw_lane_settings[3].FFE_PRESET.settings.no_preshoot != 0 ? 0xF1 : 0xFF)}; + + const uint8_t ffe_cfg[4] = { + vendor_ffe_preset_table[hw_lane_settings[0].FFE_PRESET.settings.level] & ffe_mask[0], + vendor_ffe_preset_table[hw_lane_settings[1].FFE_PRESET.settings.level] & ffe_mask[1], + vendor_ffe_preset_table[hw_lane_settings[2].FFE_PRESET.settings.level] & ffe_mask[2], + vendor_ffe_preset_table[hw_lane_settings[3].FFE_PRESET.settings.level] & ffe_mask[3]}; + + const uint8_t dp_type = dp_dio_fixed_vs_pe_retimer_lane_cfg_to_hw_cfg(link); + + const uint8_t vendor_lttpr_write_data_ffe1[4] = {0x01, 0x50, dp_type, 0x0F}; + const uint8_t vendor_lttpr_write_data_ffe2[4] = {0x01, 0x55, dp_type, ffe_cfg[0]}; + const uint8_t vendor_lttpr_write_data_ffe3[4] = {0x01, 0x56, dp_type, ffe_cfg[1]}; + const uint8_t vendor_lttpr_write_data_ffe4[4] = {0x01, 0x57, dp_type, ffe_cfg[2]}; + const uint8_t vendor_lttpr_write_data_ffe5[4] = {0x01, 0x58, dp_type, ffe_cfg[3]}; + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_ffe1[0], sizeof(vendor_lttpr_write_data_ffe1)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_ffe2[0], sizeof(vendor_lttpr_write_data_ffe2)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_ffe3[0], sizeof(vendor_lttpr_write_data_ffe3)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_ffe4[0], sizeof(vendor_lttpr_write_data_ffe4)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_ffe5[0], sizeof(vendor_lttpr_write_data_ffe5)); +} + +static void dp_hpo_fixed_vs_pe_retimer_program_override_test_pattern(struct dc_link *link, + struct encoder_set_dp_phy_pattern_param *tp_params) +{ + const uint8_t vendor_lttpr_write_data_pg0[4] = {0x1, 0x11, 0x0, 0x0}; + const uint8_t vendor_lttpr_write_data_pg1[4] = {0x1, 0x50, 0x50, 0x0}; + const uint8_t vendor_lttpr_write_data_pg2[4] = {0x1, 0x51, 0x50, 0x0}; + const uint8_t vendor_lttpr_write_data_pg3[4] = {0x1, 0x10, 0x58, 0x21}; + const uint8_t vendor_lttpr_write_data_pg4[4] = {0x1, 0x10, 0x59, 0x21}; + const uint8_t vendor_lttpr_write_data_pg5[4] = {0x1, 0x1C, 0x58, 0x4F}; + const uint8_t vendor_lttpr_write_data_pg6[4] = {0x1, 0x1C, 0x59, 0x4F}; + const uint8_t vendor_lttpr_write_data_pg7[4] = {0x1, 0x30, 0x51, 0x20}; + const uint8_t vendor_lttpr_write_data_pg8[4] = {0x1, 0x30, 0x52, 0x20}; + const uint8_t vendor_lttpr_write_data_pg9[4] = {0x1, 0x30, 0x54, 0x20}; + const uint8_t vendor_lttpr_write_data_pg10[4] = {0x1, 0x30, 0x55, 0x20}; + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg0[0], sizeof(vendor_lttpr_write_data_pg0)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg1[0], sizeof(vendor_lttpr_write_data_pg1)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg2[0], sizeof(vendor_lttpr_write_data_pg2)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg3[0], sizeof(vendor_lttpr_write_data_pg3)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg4[0], sizeof(vendor_lttpr_write_data_pg4)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg5[0], sizeof(vendor_lttpr_write_data_pg5)); + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg6[0], sizeof(vendor_lttpr_write_data_pg6)); + + if (link->cur_link_settings.lane_count == LANE_COUNT_FOUR) + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg7[0], sizeof(vendor_lttpr_write_data_pg7)); + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg8[0], sizeof(vendor_lttpr_write_data_pg8)); + + if (link->cur_link_settings.lane_count == LANE_COUNT_FOUR) + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg9[0], sizeof(vendor_lttpr_write_data_pg9)); + + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pg10[0], sizeof(vendor_lttpr_write_data_pg10)); +} + +static bool dp_hpo_fixed_vs_pe_retimer_set_override_test_pattern(struct dc_link *link, + const struct link_resource *link_res, struct encoder_set_dp_phy_pattern_param *tp_params, + const struct link_hwss *link_hwss) +{ + struct encoder_set_dp_phy_pattern_param hw_tp_params = { 0 }; + const uint8_t vendor_lttpr_exit_manual_automation_0[4] = {0x1, 0x11, 0x0, 0x06}; + + if (tp_params == NULL) + return false; + + if (tp_params->dp_phy_pattern < DP_TEST_PATTERN_SQUARE_BEGIN || + tp_params->dp_phy_pattern > DP_TEST_PATTERN_SQUARE_END) { + // Deprogram overrides from previously set square wave override + if (link->current_test_pattern == DP_TEST_PATTERN_80BIT_CUSTOM || + link->current_test_pattern == DP_TEST_PATTERN_D102) + link->dc->link_srv->configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_exit_manual_automation_0[0], + sizeof(vendor_lttpr_exit_manual_automation_0)); + else + dp_dio_fixed_vs_pe_retimer_exit_manual_automation(link); + + return false; + } + + hw_tp_params.dp_phy_pattern = DP_TEST_PATTERN_PRBS31; + hw_tp_params.dp_panel_mode = tp_params->dp_panel_mode; + + if (link_hwss->ext.set_dp_link_test_pattern) + link_hwss->ext.set_dp_link_test_pattern(link, link_res, &hw_tp_params); + + dp_hpo_fixed_vs_pe_retimer_program_override_test_pattern(link, tp_params); + + dp_hpo_fixed_vs_pe_retimer_set_tx_ffe(link, &link->cur_lane_setting[0]); + + return true; +} + +static void set_hpo_fixed_vs_pe_retimer_dp_link_test_pattern(struct dc_link *link, + const struct link_resource *link_res, + struct encoder_set_dp_phy_pattern_param *tp_params) +{ + if (!dp_hpo_fixed_vs_pe_retimer_set_override_test_pattern( + link, link_res, tp_params, get_hpo_dp_link_hwss())) { + link_res->hpo_dp_link_enc->funcs->set_link_test_pattern( + link_res->hpo_dp_link_enc, tp_params); + } + link->dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_SET_SOURCE_PATTERN); +} + +static void set_hpo_fixed_vs_pe_retimer_dp_lane_settings(struct dc_link *link, + const struct link_resource *link_res, + const struct dc_link_settings *link_settings, + const struct dc_lane_settings lane_settings[LANE_COUNT_DP_MAX]) +{ + link_res->hpo_dp_link_enc->funcs->set_ffe( + link_res->hpo_dp_link_enc, + link_settings, + lane_settings[0].FFE_PRESET.raw); + + // FFE is programmed when retimer is programmed for SQ128, but explicit + // programming needed here as well in case FFE-only update is requested + if (link->current_test_pattern >= DP_TEST_PATTERN_SQUARE_BEGIN && + link->current_test_pattern <= DP_TEST_PATTERN_SQUARE_END) + dp_hpo_fixed_vs_pe_retimer_set_tx_ffe(link, &lane_settings[0]); +} + +static void enable_hpo_fixed_vs_pe_retimer_dp_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal, + enum clock_source_id clock_source, + const struct dc_link_settings *link_settings) +{ + if (link_settings->lane_count == LANE_COUNT_FOUR) + enable_dio_fixed_vs_pe_retimer_program_4lane_output(link); + + enable_hpo_dp_link_output(link, link_res, signal, clock_source, link_settings); +} + +static const struct link_hwss hpo_fixed_vs_pe_retimer_dp_link_hwss = { + .setup_stream_encoder = setup_hpo_dp_stream_encoder, + .reset_stream_encoder = reset_hpo_dp_stream_encoder, + .setup_stream_attribute = setup_hpo_dp_stream_attribute, + .disable_link_output = disable_hpo_dp_link_output, + .setup_audio_output = setup_hpo_dp_audio_output, + .enable_audio_packet = enable_hpo_dp_audio_packet, + .disable_audio_packet = disable_hpo_dp_audio_packet, + .ext = { + .set_throttled_vcp_size = set_hpo_dp_throttled_vcp_size, + .set_hblank_min_symbol_width = set_hpo_dp_hblank_min_symbol_width, + .enable_dp_link_output = enable_hpo_fixed_vs_pe_retimer_dp_link_output, + .set_dp_link_test_pattern = set_hpo_fixed_vs_pe_retimer_dp_link_test_pattern, + .set_dp_lane_settings = set_hpo_fixed_vs_pe_retimer_dp_lane_settings, + .update_stream_allocation_table = update_hpo_dp_stream_allocation_table, + }, +}; + +bool requires_fixed_vs_pe_retimer_hpo_link_hwss(const struct dc_link *link) +{ + if (!(link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN)) + return false; + + if (!link->dpcd_caps.lttpr_caps.main_link_channel_coding.bits.DP_128b_132b_SUPPORTED) + return false; + + return true; +} + +const struct link_hwss *get_hpo_fixed_vs_pe_retimer_dp_link_hwss(void) +{ + return &hpo_fixed_vs_pe_retimer_dp_link_hwss; +} diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h new file mode 100644 index 000000000000..82301187bc7c --- /dev/null +++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h @@ -0,0 +1,33 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ +#ifndef __LINK_HWSS_HPO_FIXED_VS_PE_RETIMER_DP_H__ +#define __LINK_HWSS_HPO_FIXED_VS_PE_RETIMER_DP_H__ + +#include "link.h" + +bool requires_fixed_vs_pe_retimer_hpo_link_hwss(const struct dc_link *link); +const struct link_hwss *get_hpo_fixed_vs_pe_retimer_dp_link_hwss(void); + +#endif /* __LINK_HWSS_HPO_FIXED_VS_PE_RETIMER_DP_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c index 8041b8369e45..c9b6676eaf53 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c @@ -876,8 +876,7 @@ static bool detect_link_and_local_sink(struct dc_link *link, (link->dpcd_sink_ext_caps.bits.oled == 1)) { dpcd_set_source_specific_data(link); msleep(post_oui_delay); - set_default_brightness_aux(link); - //TODO: use cached + set_cached_brightness_aux(link); } return true; diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c index 1a7b93e41e35..79aef205598b 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -182,11 +182,8 @@ void link_resume(struct dc_link *link) static bool is_master_pipe_for_link(const struct dc_link *link, const struct pipe_ctx *pipe) { - return (pipe->stream && - pipe->stream->link && - pipe->stream->link == link && - pipe->top_pipe == NULL && - pipe->prev_odm_pipe == NULL); + return resource_is_pipe_type(pipe, OTG_MASTER) && + pipe->stream->link == link; } /* @@ -1079,8 +1076,14 @@ static struct fixed31_32 get_pbn_from_bw_in_kbps(uint64_t kbps) static struct fixed31_32 get_pbn_from_timing(struct pipe_ctx *pipe_ctx) { uint64_t kbps; + enum dc_link_encoding_format link_encoding; - kbps = dc_bandwidth_in_kbps_from_timing(&pipe_ctx->stream->timing); + if (dp_is_128b_132b_signal(pipe_ctx)) + link_encoding = DC_LINK_ENCODING_DP_128b_132b; + else + link_encoding = DC_LINK_ENCODING_DP_8b_10b; + + kbps = dc_bandwidth_in_kbps_from_timing(&pipe_ctx->stream->timing, link_encoding); return get_pbn_from_bw_in_kbps(kbps); } @@ -1538,7 +1541,8 @@ struct fixed31_32 link_calculate_sst_avg_time_slots_per_mtp( dc_fixpt_div_int(link_bw_effective, MAX_MTP_SLOT_COUNT); struct fixed31_32 timing_bw = dc_fixpt_from_int( - dc_bandwidth_in_kbps_from_timing(&stream->timing)); + dc_bandwidth_in_kbps_from_timing(&stream->timing, + dc_link_get_highest_encoding_format(link))); struct fixed31_32 avg_time_slots_per_mtp = dc_fixpt_div(timing_bw, timeslot_bw_effective); @@ -1971,6 +1975,7 @@ static void enable_link_hdmi(struct pipe_ctx *pipe_ctx) bool is_vga_mode = (stream->timing.h_addressable == 640) && (stream->timing.v_addressable == 480); struct dc *dc = pipe_ctx->stream->ctx->dc; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); if (stream->phy_pix_clk == 0) stream->phy_pix_clk = stream->timing.pix_clk_100hz / 10; @@ -2010,6 +2015,12 @@ static void enable_link_hdmi(struct pipe_ctx *pipe_ctx) if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422) display_color_depth = COLOR_DEPTH_888; + /* We need to enable stream encoder for TMDS first to apply 1/4 TMDS + * character clock in case that beyond 340MHz. + */ + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + link_hwss->setup_stream_encoder(pipe_ctx); + dc->hwss.enable_tmds_link_output( link, &pipe_ctx->link_res, @@ -2129,7 +2140,8 @@ static enum dc_status enable_link_dp(struct dc_state *state, if (link->dpcd_sink_ext_caps.bits.oled == 1 || link->dpcd_sink_ext_caps.bits.sdr_aux_backlight_control == 1 || link->dpcd_sink_ext_caps.bits.hdr_aux_backlight_control == 1) { - set_default_brightness_aux(link); // TODO: use cached if known + set_cached_brightness_aux(link); + if (link->dpcd_sink_ext_caps.bits.oled == 1) msleep(bl_oled_enable_delay); edp_backlight_enable_aux(link, true); diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c index ac1c3e2e7c1d..0895742a3102 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c @@ -132,6 +132,7 @@ static void construct_link_service_ddc(struct link_service *link_srv) link_srv->destroy_ddc_service = link_destroy_ddc_service; link_srv->query_ddc_data = link_query_ddc_data; link_srv->aux_transfer_raw = link_aux_transfer_raw; + link_srv->configure_fixed_vs_pe_retimer = link_configure_fixed_vs_pe_retimer; link_srv->aux_transfer_with_retries_no_mutex = link_aux_transfer_with_retries_no_mutex; link_srv->is_in_aux_transaction_mode = link_is_in_aux_transaction_mode; @@ -207,6 +208,13 @@ static void construct_link_service_edp_panel_control(struct link_service *link_s link_srv->edp_set_sink_vtotal_in_psr_active = edp_set_sink_vtotal_in_psr_active; link_srv->edp_get_psr_residency = edp_get_psr_residency; + + link_srv->edp_get_replay_state = edp_get_replay_state; + link_srv->edp_set_replay_allow_active = edp_set_replay_allow_active; + link_srv->edp_setup_replay = edp_setup_replay; + link_srv->edp_set_coasting_vtotal = edp_set_coasting_vtotal; + link_srv->edp_replay_residency = edp_replay_residency; + link_srv->edp_wait_for_t12 = edp_wait_for_t12; link_srv->edp_is_ilr_optimization_required = edp_is_ilr_optimization_required; @@ -783,6 +791,10 @@ static bool construct_dpia(struct dc_link *link, /* Set dpia port index : 0 to number of dpia ports */ link->ddc_hw_inst = init_params->connector_index; + // Assign Dpia preferred eng_id + if (link->dc->res_pool->funcs->get_preferred_eng_id_dpia) + link->dpia_preferred_eng_id = link->dc->res_pool->funcs->get_preferred_eng_id_dpia(link->ddc_hw_inst); + /* TODO: Create link encoder */ link->psr_settings.psr_version = DC_PSR_VERSION_UNSUPPORTED; diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c index e8b2fc4002a5..b45fda96eaf6 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -130,7 +130,8 @@ static bool dp_active_dongle_validate_timing( /* DP input has DSC, HDMI FRL output doesn't have DSC, remove DSC from output timing */ outputTiming.flags.DSC = 0; #endif - if (dc_bandwidth_in_kbps_from_timing(&outputTiming) > dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps) + if (dc_bandwidth_in_kbps_from_timing(&outputTiming, DC_LINK_ENCODING_HDMI_FRL) > + dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps) return false; } else { // DP to HDMI TMDS converter if (get_tmds_output_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10)) @@ -285,7 +286,7 @@ static bool dp_validate_mode_timing( link_setting = &link->verified_link_cap; */ - req_bw = dc_bandwidth_in_kbps_from_timing(timing); + req_bw = dc_bandwidth_in_kbps_from_timing(timing, dc_link_get_highest_encoding_format(link)); max_bw = dp_link_bandwidth_kbps(link, link_setting); if (req_bw <= max_bw) { @@ -357,7 +358,8 @@ bool link_validate_dpia_bandwidth(const struct dc_stream_state *stream, const un for (uint8_t i = 0; i < num_streams; ++i) { link[i] = stream[i].link; - bw_needed[i] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing); + bw_needed[i] = dc_bandwidth_in_kbps_from_timing(&stream[i].timing, + dc_link_get_highest_encoding_format(link[i])); } ret = dpia_validate_usb4_bw(link, bw_needed, num_streams); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.c index 0fa1228bc178..ecfd83299e75 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.c @@ -412,6 +412,88 @@ int link_aux_transfer_raw(struct ddc_service *ddc, } } +uint32_t link_get_fixed_vs_pe_retimer_write_address(struct dc_link *link) +{ + uint32_t vendor_lttpr_write_address = 0xF004F; + uint8_t offset; + + switch (link->dpcd_caps.lttpr_caps.phy_repeater_cnt) { + case 0x80: // 1 lttpr repeater + offset = 1; + break; + case 0x40: // 2 lttpr repeaters + offset = 2; + break; + case 0x20: // 3 lttpr repeaters + offset = 3; + break; + case 0x10: // 4 lttpr repeaters + offset = 4; + break; + case 0x08: // 5 lttpr repeaters + offset = 5; + break; + case 0x04: // 6 lttpr repeaters + offset = 6; + break; + case 0x02: // 7 lttpr repeaters + offset = 7; + break; + case 0x01: // 8 lttpr repeaters + offset = 8; + break; + default: + offset = 0xFF; + } + + if (offset != 0xFF) { + vendor_lttpr_write_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + } + return vendor_lttpr_write_address; +} + +uint32_t link_get_fixed_vs_pe_retimer_read_address(struct dc_link *link) +{ + return link_get_fixed_vs_pe_retimer_write_address(link) + 4; +} + +bool link_configure_fixed_vs_pe_retimer(struct ddc_service *ddc, const uint8_t *data, uint32_t length) +{ + struct aux_payload write_payload = { + .i2c_over_aux = false, + .write = true, + .address = link_get_fixed_vs_pe_retimer_write_address(ddc->link), + .length = length, + .data = (uint8_t *) data, + .reply = NULL, + .mot = I2C_MOT_UNDEF, + .write_status_update = false, + .defer_delay = 0, + }; + + return link_aux_transfer_with_retries_no_mutex(ddc, + &write_payload); +} + +bool link_query_fixed_vs_pe_retimer(struct ddc_service *ddc, uint8_t *data, uint32_t length) +{ + struct aux_payload read_payload = { + .i2c_over_aux = false, + .write = false, + .address = link_get_fixed_vs_pe_retimer_read_address(ddc->link), + .length = length, + .data = data, + .reply = NULL, + .mot = I2C_MOT_UNDEF, + .write_status_update = false, + .defer_delay = 0, + }; + + return link_aux_transfer_with_retries_no_mutex(ddc, + &read_payload); +} + bool link_aux_transfer_with_retries_no_mutex(struct ddc_service *ddc, struct aux_payload *payload) { @@ -427,7 +509,7 @@ bool try_to_configure_aux_timeout(struct ddc_service *ddc, if ((ddc->link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && !ddc->link->dc->debug.disable_fixed_vs_aux_timeout_wa && - ASICREV_IS_YELLOW_CARP(ddc->ctx->asic_id.hw_internal_rev)) { + ddc->ctx->dce_version == DCN_VERSION_3_1) { /* Fixed VS workaround for AUX timeout */ const uint32_t fixed_vs_address = 0xF004F; const uint8_t fixed_vs_data[4] = {0x1, 0x22, 0x63, 0xc}; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h index 860ef15d7f1b..a3e25e55bed6 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h @@ -72,6 +72,20 @@ bool link_query_ddc_data( bool link_aux_transfer_with_retries_no_mutex(struct ddc_service *ddc, struct aux_payload *payload); +bool link_configure_fixed_vs_pe_retimer( + struct ddc_service *ddc, + const uint8_t *data, + uint32_t length); + +bool link_query_fixed_vs_pe_retimer( + struct ddc_service *ddc, + uint8_t *data, + uint32_t length); + +uint32_t link_get_fixed_vs_pe_retimer_read_address(struct dc_link *link); +uint32_t link_get_fixed_vs_pe_retimer_write_address(struct dc_link *link); + + void write_scdc_data( struct ddc_service *ddc_service, uint32_t pix_clk, diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index 3a5e80b57711..237e0ff955f3 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -906,7 +906,7 @@ bool link_decide_link_settings(struct dc_stream_state *stream, struct dc_link_settings *link_setting) { struct dc_link *link = stream->link; - uint32_t req_bw = dc_bandwidth_in_kbps_from_timing(&stream->timing); + uint32_t req_bw = dc_bandwidth_in_kbps_from_timing(&stream->timing, dc_link_get_highest_encoding_format(link)); memset(link_setting, 0, sizeof(*link_setting)); @@ -939,7 +939,8 @@ bool link_decide_link_settings(struct dc_stream_state *stream, tmp_link_setting.link_rate = LINK_RATE_UNKNOWN; tmp_timing.flags.DSC = 0; - orig_req_bw = dc_bandwidth_in_kbps_from_timing(&tmp_timing); + orig_req_bw = dc_bandwidth_in_kbps_from_timing(&tmp_timing, + dc_link_get_highest_encoding_format(link)); edp_decide_link_settings(link, &tmp_link_setting, orig_req_bw); max_link_rate = tmp_link_setting.link_rate; } @@ -2008,6 +2009,16 @@ void detect_edp_sink_caps(struct dc_link *link) core_link_read_dpcd(link, DP_RECEIVER_ALPM_CAP, &link->dpcd_caps.alpm_caps.raw, sizeof(link->dpcd_caps.alpm_caps.raw)); + + /* + * Read REPLAY info + */ + core_link_read_dpcd(link, DP_SINK_PR_PIXEL_DEVIATION_PER_LINE, + &link->dpcd_caps.pr_info.pixel_deviation_per_line, + sizeof(link->dpcd_caps.pr_info.pixel_deviation_per_line)); + core_link_read_dpcd(link, DP_SINK_PR_MAX_NUMBER_OF_DEVIATION_LINE, + &link->dpcd_caps.pr_info.max_deviation_line, + sizeof(link->dpcd_caps.pr_info.max_deviation_line)); } bool dp_get_max_link_enc_cap(const struct dc_link *link, struct dc_link_settings *max_link_enc_cap) @@ -2165,7 +2176,9 @@ static bool dp_verify_link_cap( link, &irq_data)) (*fail_count)++; - + } else if (status == LINK_TRAINING_LINK_LOSS) { + success = true; + (*fail_count)++; } else { (*fail_count)++; } @@ -2188,6 +2201,7 @@ bool dp_verify_link_cap_with_retries( int i = 0; bool success = false; int fail_count = 0; + struct dc_link_settings last_verified_link_cap = fail_safe_link_settings; dp_trace_detect_lt_init(link); @@ -2204,10 +2218,14 @@ bool dp_verify_link_cap_with_retries( if (!link_detect_connection_type(link, &type) || type == dc_connection_none) { link->verified_link_cap = fail_safe_link_settings; break; - } else if (dp_verify_link_cap(link, known_limit_link_setting, - &fail_count) && fail_count == 0) { - success = true; - break; + } else if (dp_verify_link_cap(link, known_limit_link_setting, &fail_count)) { + last_verified_link_cap = link->verified_link_cap; + if (fail_count == 0) { + success = true; + break; + } + } else { + link->verified_link_cap = last_verified_link_cap; } fsleep(10 * 1000); } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c index ef8739df91bc..e047bbeaa49a 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.c @@ -182,6 +182,68 @@ static bool handle_hpd_irq_psr_sink(struct dc_link *link) return false; } +static bool handle_hpd_irq_replay_sink(struct dc_link *link) +{ + union dpcd_replay_configuration replay_configuration; + /*AMD Replay version reuse DP_PSR_ERROR_STATUS for REPLAY_ERROR status.*/ + union psr_error_status replay_error_status; + + if (!link->replay_settings.replay_feature_enabled) + return false; + + dm_helpers_dp_read_dpcd( + link->ctx, + link, + DP_SINK_PR_REPLAY_STATUS, + &replay_configuration.raw, + sizeof(replay_configuration.raw)); + + dm_helpers_dp_read_dpcd( + link->ctx, + link, + DP_PSR_ERROR_STATUS, + &replay_error_status.raw, + sizeof(replay_error_status.raw)); + + link->replay_settings.config.replay_error_status.bits.LINK_CRC_ERROR = + replay_error_status.bits.LINK_CRC_ERROR; + link->replay_settings.config.replay_error_status.bits.DESYNC_ERROR = + replay_configuration.bits.DESYNC_ERROR_STATUS; + link->replay_settings.config.replay_error_status.bits.STATE_TRANSITION_ERROR = + replay_configuration.bits.STATE_TRANSITION_ERROR_STATUS; + + if (link->replay_settings.config.replay_error_status.bits.LINK_CRC_ERROR || + link->replay_settings.config.replay_error_status.bits.DESYNC_ERROR || + link->replay_settings.config.replay_error_status.bits.STATE_TRANSITION_ERROR) { + bool allow_active; + + /* Acknowledge and clear configuration bits */ + dm_helpers_dp_write_dpcd( + link->ctx, + link, + DP_SINK_PR_REPLAY_STATUS, + &replay_configuration.raw, + sizeof(replay_configuration.raw)); + + /* Acknowledge and clear error bits */ + dm_helpers_dp_write_dpcd( + link->ctx, + link, + DP_PSR_ERROR_STATUS,/*DpcdAddress_REPLAY_Error_Status*/ + &replay_error_status.raw, + sizeof(replay_error_status.raw)); + + /* Replay error, disable and re-enable Replay */ + if (link->replay_settings.replay_allow_active) { + allow_active = false; + edp_set_replay_allow_active(link, &allow_active, true, false, NULL); + allow_active = true; + edp_set_replay_allow_active(link, &allow_active, true, false, NULL); + } + } + return true; +} + void dp_handle_link_loss(struct dc_link *link) { struct pipe_ctx *pipes[MAX_PIPES]; @@ -360,6 +422,10 @@ bool dp_handle_hpd_rx_irq(struct dc_link *link, /* PSR-related error was detected and handled */ return true; + if (handle_hpd_irq_replay_sink(link)) + /* Replay-related error was detected and handled */ + return true; + /* If PSR-related error handled, Main link may be off, * so do not handle as a normal sink status change interrupt. */ diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c index e011df4bdaf2..90339c2dfd84 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c @@ -1699,13 +1699,20 @@ bool perform_link_training_with_retries( } else if (do_fallback) { /* Try training at lower link bandwidth if doing fallback. */ uint32_t req_bw; uint32_t link_bw; + enum dc_link_encoding_format link_encoding = DC_LINK_ENCODING_UNSPECIFIED; decide_fallback_link_setting(link, &max_link_settings, &cur_link_settings, status); + + if (link_dp_get_encoding_format(&cur_link_settings) == DP_8b_10b_ENCODING) + link_encoding = DC_LINK_ENCODING_DP_8b_10b; + else if (link_dp_get_encoding_format(&cur_link_settings) == DP_128b_132b_ENCODING) + link_encoding = DC_LINK_ENCODING_DP_128b_132b; + /* Flag if reduced link bandwidth no longer meets stream requirements or fallen back to * minimum link bandwidth. */ - req_bw = dc_bandwidth_in_kbps_from_timing(&stream->timing); + req_bw = dc_bandwidth_in_kbps_from_timing(&stream->timing, link_encoding); link_bw = dp_link_bandwidth_kbps(link, &cur_link_settings); is_link_bw_low = (req_bw > link_bw); is_link_bw_min = ((cur_link_settings.link_rate <= LINK_RATE_LOW) && diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c index 15faaf645b14..fd8f6f198146 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c @@ -36,6 +36,7 @@ #include "link_dpcd.h" #include "link_dp_phy.h" #include "link_dp_capability.h" +#include "link_ddc.h" #define DC_LOGGER \ link->ctx->logger @@ -46,42 +47,20 @@ void dp_fixed_vs_pe_read_lane_adjust( { const uint8_t vendor_lttpr_write_data_vs[3] = {0x0, 0x53, 0x63}; const uint8_t vendor_lttpr_write_data_pe[3] = {0x0, 0x54, 0x63}; - const uint8_t offset = dp_parse_lttpr_repeater_count( - link->dpcd_caps.lttpr_caps.phy_repeater_cnt); - uint32_t vendor_lttpr_write_address = 0xF004F; - uint32_t vendor_lttpr_read_address = 0xF0053; uint8_t dprx_vs = 0; uint8_t dprx_pe = 0; uint8_t lane; - if (offset != 0xFF) { - vendor_lttpr_write_address += - ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); - vendor_lttpr_read_address += - ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); - } - /* W/A to read lane settings requested by DPRX */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_read_dpcd( - link, - vendor_lttpr_read_address, - &dprx_vs, - 1); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); - core_link_read_dpcd( - link, - vendor_lttpr_read_address, - &dprx_pe, - 1); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + + link_query_fixed_vs_pe_retimer(link->ddc, &dprx_vs, 1); + + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); + + link_query_fixed_vs_pe_retimer(link->ddc, &dprx_pe, 1); for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET = (dprx_vs >> (2 * lane)) & 0x3; @@ -95,19 +74,11 @@ void dp_fixed_vs_pe_set_retimer_lane_settings( const union dpcd_training_lane dpcd_lane_adjust[LANE_COUNT_DP_MAX], uint8_t lane_count) { - const uint8_t offset = dp_parse_lttpr_repeater_count( - link->dpcd_caps.lttpr_caps.phy_repeater_cnt); const uint8_t vendor_lttpr_write_data_reset[4] = {0x1, 0x50, 0x63, 0xFF}; - uint32_t vendor_lttpr_write_address = 0xF004F; uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; uint8_t lane = 0; - if (offset != 0xFF) { - vendor_lttpr_write_address += - ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); - } - for (lane = 0; lane < lane_count; lane++) { vendor_lttpr_write_data_vs[3] |= dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET << (2 * lane); @@ -116,21 +87,14 @@ void dp_fixed_vs_pe_set_retimer_lane_settings( } /* Force LTTPR to output desired VS and PE */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_reset[0], - sizeof(vendor_lttpr_write_data_reset)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_reset[0], sizeof(vendor_lttpr_write_data_reset)); + + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); } static enum link_training_result perform_fixed_vs_pe_nontransparent_training_sequence( @@ -236,7 +200,11 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( uint32_t pre_disable_intercept_delay_ms = 0; uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; - uint32_t vendor_lttpr_write_address = 0xF004F; + const uint8_t vendor_lttpr_write_data_4lane_1[4] = {0x1, 0x6E, 0xF2, 0x19}; + const uint8_t vendor_lttpr_write_data_4lane_2[4] = {0x1, 0x6B, 0xF2, 0x01}; + const uint8_t vendor_lttpr_write_data_4lane_3[4] = {0x1, 0x6D, 0xF2, 0x18}; + const uint8_t vendor_lttpr_write_data_4lane_4[4] = {0x1, 0x6C, 0xF2, 0x03}; + const uint8_t vendor_lttpr_write_data_4lane_5[4] = {0x1, 0x03, 0xF3, 0x06}; enum link_training_result status = LINK_TRAINING_SUCCESS; uint8_t lane = 0; union down_spread_ctrl downspread = {0}; @@ -244,10 +212,6 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( uint8_t toggle_rate; uint8_t rate; - if (link->local_sink) - pre_disable_intercept_delay_ms = - link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms; - /* Only 8b/10b is supported */ ASSERT(link_dp_get_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING); @@ -258,37 +222,27 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( } if (offset != 0xFF) { - vendor_lttpr_write_address += - ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + if (offset == 2) { + pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa; /* Certain display and cable configuration require extra delay */ - if (offset > 2) - pre_disable_intercept_delay_ms = pre_disable_intercept_delay_ms * 2; + } else if (offset > 2) { + pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2; + } } /* Vendor specific: Reset lane settings */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_reset[0], - sizeof(vendor_lttpr_write_data_reset)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_reset[0], sizeof(vendor_lttpr_write_data_reset)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); /* Vendor specific: Enable intercept */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_intercept_en[0], - sizeof(vendor_lttpr_write_data_intercept_en)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_intercept_en[0], sizeof(vendor_lttpr_write_data_intercept_en)); + /* 1. set link rate, lane count and spread. */ @@ -339,6 +293,19 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( DP_DOWNSPREAD_CTRL, lt_settings->link_settings.link_spread); + if (lt_settings->link_settings.lane_count == LANE_COUNT_FOUR) { + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_1[0], sizeof(vendor_lttpr_write_data_4lane_1)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_2[0], sizeof(vendor_lttpr_write_data_4lane_2)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_3[0], sizeof(vendor_lttpr_write_data_4lane_3)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_4[0], sizeof(vendor_lttpr_write_data_4lane_4)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_5[0], sizeof(vendor_lttpr_write_data_4lane_5)); + } + /* 2. Perform link training */ /* Perform Clock Recovery Sequence */ @@ -351,7 +318,6 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX]; union lane_align_status_updated dpcd_lane_status_updated; union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0}; - enum dc_status dpcd_status = DC_OK; uint8_t i = 0; retries_cr = 0; @@ -386,18 +352,12 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( for (i = 0; i < max_vendor_dpcd_retries; i++) { if (pre_disable_intercept_delay_ms != 0) msleep(pre_disable_intercept_delay_ms); - dpcd_status = core_link_write_dpcd( - link, - vendor_lttpr_write_address, + if (link_configure_fixed_vs_pe_retimer(link->ddc, &vendor_lttpr_write_data_intercept_dis[0], - sizeof(vendor_lttpr_write_data_intercept_dis)); - - if (dpcd_status == DC_OK) + sizeof(vendor_lttpr_write_data_intercept_dis))) break; - core_link_write_dpcd( - link, - vendor_lttpr_write_address, + link_configure_fixed_vs_pe_retimer(link->ddc, &vendor_lttpr_write_data_intercept_en[0], sizeof(vendor_lttpr_write_data_intercept_en)); } @@ -413,16 +373,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( } /* Vendor specific: Update VS and PE to DPRX requested value */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); dpcd_set_lane_settings( link, @@ -518,16 +472,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( } /* Vendor specific: Update VS and PE to DPRX requested value */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); /* 2. update DPCD*/ if (!retries_ch_eq) @@ -596,10 +544,14 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( const uint8_t vendor_lttpr_write_data_adicora_eq1[4] = {0x1, 0x55, 0x63, 0x2E}; const uint8_t vendor_lttpr_write_data_adicora_eq2[4] = {0x1, 0x55, 0x63, 0x01}; const uint8_t vendor_lttpr_write_data_adicora_eq3[4] = {0x1, 0x55, 0x63, 0x68}; + uint32_t pre_disable_intercept_delay_ms = 0; uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; - uint32_t pre_disable_intercept_delay_ms = 0; - uint32_t vendor_lttpr_write_address = 0xF004F; + const uint8_t vendor_lttpr_write_data_4lane_1[4] = {0x1, 0x6E, 0xF2, 0x19}; + const uint8_t vendor_lttpr_write_data_4lane_2[4] = {0x1, 0x6B, 0xF2, 0x01}; + const uint8_t vendor_lttpr_write_data_4lane_3[4] = {0x1, 0x6D, 0xF2, 0x18}; + const uint8_t vendor_lttpr_write_data_4lane_4[4] = {0x1, 0x6C, 0xF2, 0x03}; + const uint8_t vendor_lttpr_write_data_4lane_5[4] = {0x1, 0x03, 0xF3, 0x06}; enum link_training_result status = LINK_TRAINING_SUCCESS; uint8_t lane = 0; union down_spread_ctrl downspread = {0}; @@ -607,10 +559,6 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( uint8_t toggle_rate; uint8_t rate; - if (link->local_sink) - pre_disable_intercept_delay_ms = - link->local_sink->edid_caps.panel_patch.delay_disable_aux_intercept_ms; - /* Only 8b/10b is supported */ ASSERT(link_dp_get_encoding_format(<_settings->link_settings) == DP_8b_10b_ENCODING); @@ -621,37 +569,26 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( } if (offset != 0xFF) { - vendor_lttpr_write_address += - ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + if (offset == 2) { + pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa; /* Certain display and cable configuration require extra delay */ - if (offset > 2) - pre_disable_intercept_delay_ms = pre_disable_intercept_delay_ms * 2; + } else if (offset > 2) { + pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2; + } } /* Vendor specific: Reset lane settings */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_reset[0], - sizeof(vendor_lttpr_write_data_reset)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_reset[0], sizeof(vendor_lttpr_write_data_reset)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); /* Vendor specific: Enable intercept */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_intercept_en[0], - sizeof(vendor_lttpr_write_data_intercept_en)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_intercept_en[0], sizeof(vendor_lttpr_write_data_intercept_en)); /* 1. set link rate, lane count and spread. */ @@ -702,6 +639,19 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( DP_DOWNSPREAD_CTRL, lt_settings->link_settings.link_spread); + if (lt_settings->link_settings.lane_count == LANE_COUNT_FOUR) { + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_1[0], sizeof(vendor_lttpr_write_data_4lane_1)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_2[0], sizeof(vendor_lttpr_write_data_4lane_2)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_3[0], sizeof(vendor_lttpr_write_data_4lane_3)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_4[0], sizeof(vendor_lttpr_write_data_4lane_4)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_4lane_5[0], sizeof(vendor_lttpr_write_data_4lane_5)); + } + /* 2. Perform link training */ /* Perform Clock Recovery Sequence */ @@ -714,7 +664,6 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX]; union lane_align_status_updated dpcd_lane_status_updated; union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0}; - enum dc_status dpcd_status = DC_OK; uint8_t i = 0; retries_cr = 0; @@ -749,18 +698,12 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( for (i = 0; i < max_vendor_dpcd_retries; i++) { if (pre_disable_intercept_delay_ms != 0) msleep(pre_disable_intercept_delay_ms); - dpcd_status = core_link_write_dpcd( - link, - vendor_lttpr_write_address, + if (link_configure_fixed_vs_pe_retimer(link->ddc, &vendor_lttpr_write_data_intercept_dis[0], - sizeof(vendor_lttpr_write_data_intercept_dis)); - - if (dpcd_status == DC_OK) + sizeof(vendor_lttpr_write_data_intercept_dis))) break; - core_link_write_dpcd( - link, - vendor_lttpr_write_address, + link_configure_fixed_vs_pe_retimer(link->ddc, &vendor_lttpr_write_data_intercept_en[0], sizeof(vendor_lttpr_write_data_intercept_en)); } @@ -776,16 +719,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( } /* Vendor specific: Update VS and PE to DPRX requested value */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); dpcd_set_lane_settings( link, @@ -858,17 +795,14 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = {0}; union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0}; - core_link_write_dpcd( - link, - vendor_lttpr_write_address, + link_configure_fixed_vs_pe_retimer(link->ddc, &vendor_lttpr_write_data_adicora_eq1[0], sizeof(vendor_lttpr_write_data_adicora_eq1)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, + link_configure_fixed_vs_pe_retimer(link->ddc, &vendor_lttpr_write_data_adicora_eq2[0], sizeof(vendor_lttpr_write_data_adicora_eq2)); + /* Note: also check that TPS4 is a supported feature*/ tr_pattern = lt_settings->pattern_for_eq; @@ -892,16 +826,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( } /* Vendor specific: Update VS and PE to DPRX requested value */ - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_vs[0], - sizeof(vendor_lttpr_write_data_vs)); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_pe[0], - sizeof(vendor_lttpr_write_data_pe)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_vs[0], sizeof(vendor_lttpr_write_data_vs)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_pe[0], sizeof(vendor_lttpr_write_data_pe)); /* 2. update DPCD*/ if (!retries_ch_eq) { @@ -914,11 +842,10 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( lt_settings, tr_pattern, 0); - core_link_write_dpcd( - link, - vendor_lttpr_write_address, - &vendor_lttpr_write_data_adicora_eq3[0], - sizeof(vendor_lttpr_write_data_adicora_eq3)); + link_configure_fixed_vs_pe_retimer(link->ddc, + &vendor_lttpr_write_data_adicora_eq3[0], + sizeof(vendor_lttpr_write_data_adicora_eq3)); + } else dpcd_set_lane_settings(link, lt_settings, 0); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index 2039a345f23a..98e715aa6d8e 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -34,9 +34,13 @@ #include "dm_helpers.h" #include "dal_asic_id.h" #include "dce/dmub_psr.h" +#include "dc/dc_dmub_srv.h" +#include "dce/dmub_replay.h" #include "abm.h" #define DC_LOGGER_INIT(logger) +#define DP_SINK_PR_ENABLE_AND_CONFIGURATION 0x37B + /* Travis */ static const uint8_t DP_VGA_LVDS_CONVERTER_ID_2[] = "sivarT"; /* Nutmeg */ @@ -46,43 +50,42 @@ void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode) { union dpcd_edp_config edp_config_set; bool panel_mode_edp = false; + enum dc_status result; memset(&edp_config_set, '\0', sizeof(union dpcd_edp_config)); - if (panel_mode != DP_PANEL_MODE_DEFAULT) { + switch (panel_mode) { + case DP_PANEL_MODE_EDP: + case DP_PANEL_MODE_SPECIAL: + panel_mode_edp = true; + break; - switch (panel_mode) { - case DP_PANEL_MODE_EDP: - case DP_PANEL_MODE_SPECIAL: - panel_mode_edp = true; - break; + default: + break; + } - default: - break; - } + /*set edp panel mode in receiver*/ + result = core_link_read_dpcd( + link, + DP_EDP_CONFIGURATION_SET, + &edp_config_set.raw, + sizeof(edp_config_set.raw)); - /*set edp panel mode in receiver*/ - core_link_read_dpcd( + if (result == DC_OK && + edp_config_set.bits.PANEL_MODE_EDP + != panel_mode_edp) { + + edp_config_set.bits.PANEL_MODE_EDP = + panel_mode_edp; + result = core_link_write_dpcd( link, DP_EDP_CONFIGURATION_SET, &edp_config_set.raw, sizeof(edp_config_set.raw)); - if (edp_config_set.bits.PANEL_MODE_EDP - != panel_mode_edp) { - enum dc_status result; - - edp_config_set.bits.PANEL_MODE_EDP = - panel_mode_edp; - result = core_link_write_dpcd( - link, - DP_EDP_CONFIGURATION_SET, - &edp_config_set.raw, - sizeof(edp_config_set.raw)); - - ASSERT(result == DC_OK); - } + ASSERT(result == DC_OK); } + link->panel_mode = panel_mode; DC_LOG_DETECTION_DP_CAPS("Link: %d eDP panel mode supported: %d " "eDP panel mode enabled: %d \n", @@ -164,6 +167,7 @@ bool edp_set_backlight_level_nits(struct dc_link *link, *(uint32_t *)&dpcd_backlight_set.backlight_level_millinits = backlight_millinits; *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; + link->backlight_settings.backlight_millinits = backlight_millinits; if (!link->dpcd_caps.panel_luminance_control) { if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, @@ -251,10 +255,20 @@ static bool read_default_bl_aux(struct dc_link *link, uint32_t *backlight_millin link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT)) return false; - if (!core_link_read_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, - (uint8_t *) backlight_millinits, - sizeof(uint32_t))) - return false; + if (!link->dpcd_caps.panel_luminance_control) { + if (!core_link_read_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, + (uint8_t *)backlight_millinits, + sizeof(uint32_t))) + return false; + } else { + //setting to 0 as a precaution, since target_luminance_value is 3 bytes + memset(backlight_millinits, 0, sizeof(uint32_t)); + + if (!core_link_read_dpcd(link, DP_EDP_PANEL_TARGET_LUMINANCE_VALUE, + (uint8_t *)backlight_millinits, + sizeof(struct target_luminance_value))) + return false; + } return true; } @@ -276,6 +290,16 @@ bool set_default_brightness_aux(struct dc_link *link) return false; } +bool set_cached_brightness_aux(struct dc_link *link) +{ + if (link->backlight_settings.backlight_millinits) + return edp_set_backlight_level_nits(link, true, + link->backlight_settings.backlight_millinits, 0); + else + return set_default_brightness_aux(link); + return false; +} + bool edp_is_ilr_optimization_required(struct dc_link *link, struct dc_crtc_timing *crtc_timing) { @@ -309,7 +333,7 @@ bool edp_is_ilr_optimization_required(struct dc_link *link, core_link_read_dpcd(link, DP_LANE_COUNT_SET, &lane_count_set.raw, sizeof(lane_count_set)); - req_bw = dc_bandwidth_in_kbps_from_timing(crtc_timing); + req_bw = dc_bandwidth_in_kbps_from_timing(crtc_timing, dc_link_get_highest_encoding_format(link)); if (!crtc_timing->flags.DSC) edp_decide_link_settings(link, &link_setting, req_bw); @@ -807,6 +831,167 @@ bool edp_set_sink_vtotal_in_psr_active(const struct dc_link *link, uint16_t psr_ return true; } +bool edp_set_replay_allow_active(struct dc_link *link, const bool *allow_active, + bool wait, bool force_static, const unsigned int *power_opts) +{ + struct dc *dc = link->ctx->dc; + struct dmub_replay *replay = dc->res_pool->replay; + unsigned int panel_inst; + + if (replay == NULL && force_static) + return false; + + if (!dc_get_edp_link_panel_inst(dc, link, &panel_inst)) + return false; + + /* Set power optimization flag */ + if (power_opts && link->replay_settings.replay_power_opt_active != *power_opts) { + if (link->replay_settings.replay_feature_enabled && replay->funcs->replay_set_power_opt) { + replay->funcs->replay_set_power_opt(replay, *power_opts, panel_inst); + link->replay_settings.replay_power_opt_active = *power_opts; + } + } + + /* Activate or deactivate Replay */ + if (allow_active && link->replay_settings.replay_allow_active != *allow_active) { + // TODO: Handle mux change case if force_static is set + // If force_static is set, just change the replay_allow_active state directly + if (replay != NULL && link->replay_settings.replay_feature_enabled) + replay->funcs->replay_enable(replay, *allow_active, wait, panel_inst); + link->replay_settings.replay_allow_active = *allow_active; + } + + return true; +} + +bool edp_get_replay_state(const struct dc_link *link, uint64_t *state) +{ + struct dc *dc = link->ctx->dc; + struct dmub_replay *replay = dc->res_pool->replay; + unsigned int panel_inst; + enum replay_state pr_state = REPLAY_STATE_0; + + if (!dc_get_edp_link_panel_inst(dc, link, &panel_inst)) + return false; + + if (replay != NULL && link->replay_settings.replay_feature_enabled) + replay->funcs->replay_get_state(replay, &pr_state, panel_inst); + *state = pr_state; + + return true; +} + +bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream) +{ + /* To-do: Setup Replay */ + struct dc *dc = link->ctx->dc; + struct dmub_replay *replay = dc->res_pool->replay; + int i; + unsigned int panel_inst; + struct replay_context replay_context = { 0 }; + unsigned int lineTimeInNs = 0; + + + union replay_enable_and_configuration replay_config; + + union dpcd_alpm_configuration alpm_config; + + replay_context.controllerId = CONTROLLER_ID_UNDEFINED; + + if (!link) + return false; + + if (!replay) + return false; + + if (!dc_get_edp_link_panel_inst(dc, link, &panel_inst)) + return false; + + replay_context.aux_inst = link->ddc->ddc_pin->hw_info.ddc_channel; + replay_context.digbe_inst = link->link_enc->transmitter; + replay_context.digfe_inst = link->link_enc->preferred_engine; + + for (i = 0; i < MAX_PIPES; i++) { + if (dc->current_state->res_ctx.pipe_ctx[i].stream + == stream) { + /* dmcu -1 for all controller id values, + * therefore +1 here + */ + replay_context.controllerId = + dc->current_state->res_ctx.pipe_ctx[i].stream_res.tg->inst + 1; + break; + } + } + + lineTimeInNs = + ((stream->timing.h_total * 1000000) / + (stream->timing.pix_clk_100hz / 10)) + 1; + + replay_context.line_time_in_ns = lineTimeInNs; + + if (replay) + link->replay_settings.replay_feature_enabled = + replay->funcs->replay_copy_settings(replay, link, &replay_context, panel_inst); + if (link->replay_settings.replay_feature_enabled) { + + replay_config.bits.FREESYNC_PANEL_REPLAY_MODE = 1; + replay_config.bits.TIMING_DESYNC_ERROR_VERIFICATION = + link->replay_settings.config.replay_timing_sync_supported; + replay_config.bits.STATE_TRANSITION_ERROR_DETECTION = 1; + dm_helpers_dp_write_dpcd(link->ctx, link, + DP_SINK_PR_ENABLE_AND_CONFIGURATION, + (uint8_t *)&(replay_config.raw), sizeof(uint8_t)); + + memset(&alpm_config, 0, sizeof(alpm_config)); + alpm_config.bits.ENABLE = 1; + dm_helpers_dp_write_dpcd( + link->ctx, + link, + DP_RECEIVER_ALPM_CONFIG, + &alpm_config.raw, + sizeof(alpm_config.raw)); + } + return true; +} + +bool edp_set_coasting_vtotal(struct dc_link *link, uint16_t coasting_vtotal) +{ + struct dc *dc = link->ctx->dc; + struct dmub_replay *replay = dc->res_pool->replay; + unsigned int panel_inst; + + if (!replay) + return false; + + if (!dc_get_edp_link_panel_inst(dc, link, &panel_inst)) + return false; + + if (coasting_vtotal && link->replay_settings.coasting_vtotal != coasting_vtotal) { + replay->funcs->replay_set_coasting_vtotal(replay, coasting_vtotal, panel_inst); + link->replay_settings.coasting_vtotal = coasting_vtotal; + } + + return true; +} + +bool edp_replay_residency(const struct dc_link *link, + unsigned int *residency, const bool is_start, const bool is_alpm) +{ + struct dc *dc = link->ctx->dc; + struct dmub_replay *replay = dc->res_pool->replay; + unsigned int panel_inst; + + if (!dc_get_edp_link_panel_inst(dc, link, &panel_inst)) + return false; + + if (replay != NULL && link->replay_settings.replay_feature_enabled) + replay->funcs->replay_residency(replay, panel_inst, residency, is_start, is_alpm); + else + *residency = 0; + + return true; +} + static struct abm *get_abm_from_stream_res(const struct dc_link *link) { int i; diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h index 28f552080558..0a5bbda8c739 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h @@ -30,6 +30,7 @@ enum dp_panel_mode dp_get_panel_mode(struct dc_link *link); void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode); bool set_default_brightness_aux(struct dc_link *link); +bool set_cached_brightness_aux(struct dc_link *link); void edp_panel_backlight_power_on(struct dc_link *link, bool wait_for_hpd); int edp_get_backlight_level(const struct dc_link *link); bool edp_get_backlight_level_nits(struct dc_link *link, @@ -52,6 +53,14 @@ bool edp_setup_psr(struct dc_link *link, bool edp_set_sink_vtotal_in_psr_active(const struct dc_link *link, uint16_t psr_vtotal_idle, uint16_t psr_vtotal_su); void edp_get_psr_residency(const struct dc_link *link, uint32_t *residency); +bool edp_set_replay_allow_active(struct dc_link *dc_link, const bool *enable, + bool wait, bool force_static, const unsigned int *power_opts); +bool edp_setup_replay(struct dc_link *link, + const struct dc_stream_state *stream); +bool edp_set_coasting_vtotal(struct dc_link *link, uint16_t coasting_vtotal); +bool edp_replay_residency(const struct dc_link *link, + unsigned int *residency, const bool is_start, const bool is_alpm); +bool edp_get_replay_state(const struct dc_link *link, uint64_t *state); bool edp_wait_for_t12(struct dc_link *link); bool edp_is_ilr_optimization_required(struct dc_link *link, struct dc_crtc_timing *crtc_timing); diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index 4585e0419da6..2d995c87fbb9 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -378,6 +378,7 @@ struct dmub_srv_hw_funcs { union dmub_fw_boot_status (*get_fw_status)(struct dmub_srv *dmub); + union dmub_fw_boot_options (*get_fw_boot_option)(struct dmub_srv *dmub); void (*set_gpint)(struct dmub_srv *dmub, union dmub_gpint_data_register reg); @@ -778,9 +779,15 @@ void dmub_flush_buffer_mem(const struct dmub_fb *fb); enum dmub_status dmub_srv_get_fw_boot_status(struct dmub_srv *dmub, union dmub_fw_boot_status *status); +enum dmub_status dmub_srv_get_fw_boot_option(struct dmub_srv *dmub, + union dmub_fw_boot_options *option); + enum dmub_status dmub_srv_cmd_with_reply_data(struct dmub_srv *dmub, union dmub_rb_cmd *cmd); +enum dmub_status dmub_srv_set_skip_panel_power_sequence(struct dmub_srv *dmub, + bool skip); + bool dmub_srv_get_outbox0_msg(struct dmub_srv *dmub, struct dmcub_trace_buf_entry *entry); bool dmub_srv_get_diagnostic_data(struct dmub_srv *dmub, struct dmub_diagnostic_data *diag_data); diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index af1f50742371..7afa78b918b5 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -170,6 +170,95 @@ extern "C" { #endif #pragma pack(push, 1) +#define ABM_NUM_OF_ACE_SEGMENTS 5 + +union abm_flags { + struct { + /** + * @abm_enabled: Indicates if ABM is enabled. + */ + unsigned int abm_enabled : 1; + + /** + * @disable_abm_requested: Indicates if driver has requested ABM to be disabled. + */ + unsigned int disable_abm_requested : 1; + + /** + * @disable_abm_immediately: Indicates if driver has requested ABM to be disabled + * immediately. + */ + unsigned int disable_abm_immediately : 1; + + /** + * @disable_abm_immediate_keep_gain: Indicates if driver has requested ABM + * to be disabled immediately and keep gain. + */ + unsigned int disable_abm_immediate_keep_gain : 1; + + /** + * @fractional_pwm: Indicates if fractional duty cycle for backlight PWM is enabled. + */ + unsigned int fractional_pwm : 1; + + /** + * @abm_gradual_bl_change: Indicates if algorithm has completed gradual adjustment + * of user backlight level. + */ + unsigned int abm_gradual_bl_change : 1; + } bitfields; + + unsigned int u32All; +}; + +struct abm_save_restore { + /** + * @flags: Misc. ABM flags. + */ + union abm_flags flags; + + /** + * @pause: true: pause ABM and get state + * false: unpause ABM after setting state + */ + uint32_t pause; + + /** + * @next_ace_slope: Next ACE slopes to be programmed in HW (u3.13) + */ + uint32_t next_ace_slope[ABM_NUM_OF_ACE_SEGMENTS]; + + /** + * @next_ace_thresh: Next ACE thresholds to be programmed in HW (u10.6) + */ + uint32_t next_ace_thresh[ABM_NUM_OF_ACE_SEGMENTS]; + + /** + * @next_ace_offset: Next ACE offsets to be programmed in HW (u10.6) + */ + uint32_t next_ace_offset[ABM_NUM_OF_ACE_SEGMENTS]; + + + /** + * @knee_threshold: Current x-position of ACE knee (u0.16). + */ + uint32_t knee_threshold; + /** + * @current_gain: Current backlight reduction (u16.16). + */ + uint32_t current_gain; + /** + * @curr_bl_level: Current actual backlight level converging to target backlight level. + */ + uint16_t curr_bl_level; + + /** + * @curr_user_bl_level: Current nominal backlight level converging to level requested by user. + */ + uint16_t curr_user_bl_level; + +}; + /** * union dmub_addr - DMUB physical/virtual 64-bit address. */ @@ -249,6 +338,112 @@ union dmub_psr_debug_flags { }; /** + * Flags that can be set by driver to change some Replay behaviour. + */ +union replay_debug_flags { + struct { + /** + * Enable visual confirm in FW. + */ + uint32_t visual_confirm : 1; + + /** + * @skip_crc: Set if need to skip CRC. + */ + uint32_t skip_crc : 1; + + /** + * @force_link_power_on: Force disable ALPM control + */ + uint32_t force_link_power_on : 1; + + /** + * @force_phy_power_on: Force phy power on + */ + uint32_t force_phy_power_on : 1; + + /** + * @timing_resync_disabled: Disabled Replay normal sleep mode timing resync + */ + uint32_t timing_resync_disabled : 1; + + /** + * @skip_crtc_disabled: CRTC disable skipped + */ + uint32_t skip_crtc_disabled : 1; + + /** + * @force_defer_one_frame_update: Force defer one frame update in ultra sleep mode + */ + uint32_t force_defer_one_frame_update : 1; + /** + * @disable_delay_alpm_on: Force disable delay alpm on + */ + uint32_t disable_delay_alpm_on : 1; + /** + * @disable_desync_error_check: Force disable desync error check + */ + uint32_t disable_desync_error_check : 1; + /** + * @disable_desync_error_check: Force disable desync error check + */ + uint32_t disable_dmub_save_restore : 1; + + uint32_t reserved : 22; + } bitfields; + + uint32_t u32All; +}; + +union replay_hw_flags { + struct { + /** + * @allow_alpm_fw_standby_mode: To indicate whether the + * ALPM FW standby mode is allowed + */ + uint32_t allow_alpm_fw_standby_mode : 1; + + /* + * @dsc_enable_status: DSC enable status in driver + */ + uint32_t dsc_enable_status : 1; + + /** + * @fec_enable_status: receive fec enable/disable status from driver + */ + uint32_t fec_enable_status : 1; + + /* + * @smu_optimizations_en: SMU power optimization. + * Only when active display is Replay capable and display enters Replay. + * Trigger interrupt to SMU to powerup/down. + */ + uint32_t smu_optimizations_en : 1; + + /** + * @otg_powered_down: Flag to keep track of OTG power state. + */ + uint32_t otg_powered_down : 1; + + /** + * @phy_power_state: Indicates current phy power state + */ + uint32_t phy_power_state : 1; + + /** + * @link_power_state: Indicates current link power state + */ + uint32_t link_power_state : 1; + /** + * Use TPS3 signal when restore main link. + */ + uint32_t force_wakeup_by_tps3 : 1; + } bitfields; + + uint32_t u32All; +}; + +/** * DMUB visual confirm color */ struct dmub_feature_caps { @@ -566,9 +761,42 @@ enum dmub_gpint_command { DMUB_GPINT__PSR_RESIDENCY = 9, /** + * DESC: Get REPLAY state from FW. + * RETURN: REPLAY state enum. This enum may need to be converted to the legacy REPLAY state value. + */ + DMUB_GPINT__GET_REPLAY_STATE = 13, + + /** + * DESC: Start REPLAY residency counter. Stop REPLAY resdiency counter and get value. + * ARGS: We can measure residency from various points. The argument will specify the residency mode. + * By default, it is measured from after we powerdown the PHY, to just before we powerup the PHY. + * RETURN: REPLAY residency in milli-percent. + */ + DMUB_GPINT__REPLAY_RESIDENCY = 14, + + + /** * DESC: Notifies DMCUB detection is done so detection required can be cleared. */ DMUB_GPINT__NOTIFY_DETECTION_DONE = 12, + /** + * DESC: Updates the trace buffer lower 32-bit mask. + * ARGS: The new mask + * RETURN: Lower 32-bit mask. + */ + DMUB_GPINT__UPDATE_TRACE_BUFFER_MASK = 101, + /** + * DESC: Updates the trace buffer lower 32-bit mask. + * ARGS: The new mask + * RETURN: Lower 32-bit mask. + */ + DMUB_GPINT__SET_TRACE_BUFFER_MASK_WORD0 = 102, + /** + * DESC: Updates the trace buffer mask bi0~bit15. + * ARGS: The new mask + * RETURN: Lower 32-bit mask. + */ + DMUB_GPINT__SET_TRACE_BUFFER_MASK_WORD1 = 103, }; /** @@ -764,6 +992,11 @@ enum dmub_cmd_type { */ /** + * Command type used for all REPLAY commands. + */ + DMUB_CMD__REPLAY = 83, + + /** * Command type used for all SECURE_DISPLAY commands. */ DMUB_CMD__SECURE_DISPLAY = 85, @@ -1911,6 +2144,10 @@ enum dmub_phy_fsm_state { DMUB_PHY_FSM_PLL_EN, DMUB_PHY_FSM_TX_EN, DMUB_PHY_FSM_FAST_LP, + DMUB_PHY_FSM_P2_PLL_OFF_CPM, + DMUB_PHY_FSM_P2_PLL_OFF_PG, + DMUB_PHY_FSM_P2_PLL_OFF, + DMUB_PHY_FSM_P2_PLL_ON, }; /** @@ -2496,6 +2733,272 @@ struct dmub_cmd_psr_set_power_opt_data { uint32_t power_opt; }; +#define REPLAY_RESIDENCY_MODE_SHIFT (0) +#define REPLAY_RESIDENCY_ENABLE_SHIFT (1) + +#define REPLAY_RESIDENCY_MODE_MASK (0x1 << REPLAY_RESIDENCY_MODE_SHIFT) +# define REPLAY_RESIDENCY_MODE_PHY (0x0 << REPLAY_RESIDENCY_MODE_SHIFT) +# define REPLAY_RESIDENCY_MODE_ALPM (0x1 << REPLAY_RESIDENCY_MODE_SHIFT) + +#define REPLAY_RESIDENCY_ENABLE_MASK (0x1 << REPLAY_RESIDENCY_ENABLE_SHIFT) +# define REPLAY_RESIDENCY_DISABLE (0x0 << REPLAY_RESIDENCY_ENABLE_SHIFT) +# define REPLAY_RESIDENCY_ENABLE (0x1 << REPLAY_RESIDENCY_ENABLE_SHIFT) + +enum replay_state { + REPLAY_STATE_0 = 0x0, + REPLAY_STATE_1 = 0x10, + REPLAY_STATE_1A = 0x11, + REPLAY_STATE_2 = 0x20, + REPLAY_STATE_3 = 0x30, + REPLAY_STATE_3INIT = 0x31, + REPLAY_STATE_4 = 0x40, + REPLAY_STATE_4A = 0x41, + REPLAY_STATE_4B = 0x42, + REPLAY_STATE_4C = 0x43, + REPLAY_STATE_4D = 0x44, + REPLAY_STATE_4B_LOCKED = 0x4A, + REPLAY_STATE_4C_UNLOCKED = 0x4B, + REPLAY_STATE_5 = 0x50, + REPLAY_STATE_5A = 0x51, + REPLAY_STATE_5B = 0x52, + REPLAY_STATE_5A_LOCKED = 0x5A, + REPLAY_STATE_5B_UNLOCKED = 0x5B, + REPLAY_STATE_6 = 0x60, + REPLAY_STATE_6A = 0x61, + REPLAY_STATE_6B = 0x62, + REPLAY_STATE_INVALID = 0xFF, +}; + +/** + * Replay command sub-types. + */ +enum dmub_cmd_replay_type { + /** + * Copy driver-calculated parameters to REPLAY state. + */ + DMUB_CMD__REPLAY_COPY_SETTINGS = 0, + /** + * Enable REPLAY. + */ + DMUB_CMD__REPLAY_ENABLE = 1, + /** + * Set Replay power option. + */ + DMUB_CMD__SET_REPLAY_POWER_OPT = 2, + /** + * Set coasting vtotal. + */ + DMUB_CMD__REPLAY_SET_COASTING_VTOTAL = 3, +}; + +/** + * Data passed from driver to FW in a DMUB_CMD__REPLAY_COPY_SETTINGS command. + */ +struct dmub_cmd_replay_copy_settings_data { + /** + * Flags that can be set by driver to change some replay behaviour. + */ + union replay_debug_flags debug; + + /** + * @flags: Flags used to determine feature functionality. + */ + union replay_hw_flags flags; + + /** + * DPP HW instance. + */ + uint8_t dpp_inst; + /** + * OTG HW instance. + */ + uint8_t otg_inst; + /** + * DIG FE HW instance. + */ + uint8_t digfe_inst; + /** + * DIG BE HW instance. + */ + uint8_t digbe_inst; + /** + * AUX HW instance. + */ + uint8_t aux_inst; + /** + * Panel Instance. + * Panel isntance to identify which psr_state to use + * Currently the support is only for 0 or 1 + */ + uint8_t panel_inst; + /** + * @pixel_deviation_per_line: Indicate the maximum pixel deviation per line compare + * to Source timing when Sink maintains coasting vtotal during the Replay normal sleep mode + */ + uint8_t pixel_deviation_per_line; + /** + * @max_deviation_line: The max number of deviation line that can keep the timing + * synchronized between the Source and Sink during Replay normal sleep mode. + */ + uint8_t max_deviation_line; + /** + * Length of each horizontal line in ns. + */ + uint32_t line_time_in_ns; + /** + * PHY instance. + */ + uint8_t dpphy_inst; + /** + * Determines if SMU optimzations are enabled/disabled. + */ + uint8_t smu_optimizations_en; + /** + * Determines if timing sync are enabled/disabled. + */ + uint8_t replay_timing_sync_supported; + /* + * Use FSM state for Replay power up/down + */ + uint8_t use_phy_fsm; +}; + +/** + * Definition of a DMUB_CMD__REPLAY_COPY_SETTINGS command. + */ +struct dmub_rb_cmd_replay_copy_settings { + /** + * Command header. + */ + struct dmub_cmd_header header; + /** + * Data passed from driver to FW in a DMUB_CMD__REPLAY_COPY_SETTINGS command. + */ + struct dmub_cmd_replay_copy_settings_data replay_copy_settings_data; +}; + +/** + * Replay disable / enable state for dmub_rb_cmd_replay_enable_data.enable + */ +enum replay_enable { + /** + * Disable REPLAY. + */ + REPLAY_DISABLE = 0, + /** + * Enable REPLAY. + */ + REPLAY_ENABLE = 1, +}; + +/** + * Data passed from driver to FW in a DMUB_CMD__REPLAY_ENABLE command. + */ +struct dmub_rb_cmd_replay_enable_data { + /** + * Replay enable or disable. + */ + uint8_t enable; + /** + * Panel Instance. + * Panel isntance to identify which replay_state to use + * Currently the support is only for 0 or 1 + */ + uint8_t panel_inst; + /** + * Phy state to enter. + * Values to use are defined in dmub_phy_fsm_state + */ + uint8_t phy_fsm_state; + /** + * Phy rate for DP - RBR/HBR/HBR2/HBR3. + * Set this using enum phy_link_rate. + * This does not support HDMI/DP2 for now. + */ + uint8_t phy_rate; +}; + +/** + * Definition of a DMUB_CMD__REPLAY_ENABLE command. + * Replay enable/disable is controlled using action in data. + */ +struct dmub_rb_cmd_replay_enable { + /** + * Command header. + */ + struct dmub_cmd_header header; + + struct dmub_rb_cmd_replay_enable_data data; +}; + +/** + * Data passed from driver to FW in a DMUB_CMD__SET_REPLAY_POWER_OPT command. + */ +struct dmub_cmd_replay_set_power_opt_data { + /** + * Panel Instance. + * Panel isntance to identify which replay_state to use + * Currently the support is only for 0 or 1 + */ + uint8_t panel_inst; + /** + * Explicit padding to 4 byte boundary. + */ + uint8_t pad[3]; + /** + * REPLAY power option + */ + uint32_t power_opt; +}; + +/** + * Definition of a DMUB_CMD__SET_REPLAY_POWER_OPT command. + */ +struct dmub_rb_cmd_replay_set_power_opt { + /** + * Command header. + */ + struct dmub_cmd_header header; + /** + * Definition of a DMUB_CMD__SET_REPLAY_POWER_OPT command. + */ + struct dmub_cmd_replay_set_power_opt_data replay_set_power_opt_data; +}; + +/** + * Data passed from driver to FW in a DMUB_CMD__REPLAY_SET_COASTING_VTOTAL command. + */ +struct dmub_cmd_replay_set_coasting_vtotal_data { + /** + * 16-bit value dicated by driver that indicates the coasting vtotal. + */ + uint16_t coasting_vtotal; + /** + * REPLAY control version. + */ + uint8_t cmd_version; + /** + * Panel Instance. + * Panel isntance to identify which replay_state to use + * Currently the support is only for 0 or 1 + */ + uint8_t panel_inst; +}; + +/** + * Definition of a DMUB_CMD__REPLAY_SET_COASTING_VTOTAL command. + */ +struct dmub_rb_cmd_replay_set_coasting_vtotal { + /** + * Command header. + */ + struct dmub_cmd_header header; + /** + * Definition of a DMUB_CMD__REPLAY_SET_COASTING_VTOTAL command. + */ + struct dmub_cmd_replay_set_coasting_vtotal_data replay_set_coasting_vtotal_data; +}; + /** * Definition of a DMUB_CMD__SET_PSR_POWER_OPT command. */ @@ -2587,6 +3090,10 @@ enum hw_lock_client { */ HW_LOCK_CLIENT_PSR_SU = 1, /** + * Replay is the client of HW Lock Manager. + */ + HW_LOCK_CLIENT_REPLAY = 4, + /** * Invalid client. */ HW_LOCK_CLIENT_INVALID = 0xFFFFFFFF, @@ -2672,6 +3179,12 @@ enum dmub_cmd_abm_type { * unregister vertical interrupt after steady state is reached */ DMUB_CMD__ABM_PAUSE = 6, + + /** + * Save and Restore ABM state. On save we save parameters, and + * on restore we update state with passed in data. + */ + DMUB_CMD__ABM_SAVE_RESTORE = 7, }; /** @@ -3056,6 +3569,7 @@ struct dmub_cmd_abm_pause_data { uint8_t pad[1]; }; + /** * Definition of a DMUB_CMD__ABM_PAUSE command. */ @@ -3072,6 +3586,36 @@ struct dmub_rb_cmd_abm_pause { }; /** + * Definition of a DMUB_CMD__ABM_SAVE_RESTORE command. + */ +struct dmub_rb_cmd_abm_save_restore { + /** + * Command header. + */ + struct dmub_cmd_header header; + + /** + * OTG hw instance + */ + uint8_t otg_inst; + + /** + * Enable or disable ABM pause + */ + uint8_t freeze; + + /** + * Explicit padding to 4 byte boundary. + */ + uint8_t debug; + + /** + * Data passed from driver to FW in a DMUB_CMD__ABM_INIT_CONFIG command. + */ + struct dmub_cmd_abm_init_config_data abm_init_config_data; +}; + +/** * Data passed from driver to FW in a DMUB_CMD__QUERY_FEATURE_CAPS command. */ struct dmub_cmd_query_feature_caps_data { @@ -3509,6 +4053,11 @@ union dmub_rb_cmd { struct dmub_rb_cmd_abm_pause abm_pause; /** + * Definition of a DMUB_CMD__ABM_SAVE_RESTORE command. + */ + struct dmub_rb_cmd_abm_save_restore abm_save_restore; + + /** * Definition of a DMUB_CMD__DP_AUX_ACCESS command. */ struct dmub_rb_cmd_dp_aux_access dp_aux_access; @@ -3576,6 +4125,22 @@ union dmub_rb_cmd { * Definition of a DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE command. */ struct dmub_rb_cmd_idle_opt_dcn_notify_idle idle_opt_notify_idle; + /* + * Definition of a DMUB_CMD__REPLAY_COPY_SETTINGS command. + */ + struct dmub_rb_cmd_replay_copy_settings replay_copy_settings; + /** + * Definition of a DMUB_CMD__REPLAY_ENABLE command. + */ + struct dmub_rb_cmd_replay_enable replay_enable; + /** + * Definition of a DMUB_CMD__SET_REPLAY_POWER_OPT command. + */ + struct dmub_rb_cmd_replay_set_power_opt replay_set_power_opt; + /** + * Definition of a DMUB_CMD__REPLAY_SET_COASTING_VTOTAL command. + */ + struct dmub_rb_cmd_replay_set_coasting_vtotal replay_set_coasting_vtotal; }; /** diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_subvp_state.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_subvp_state.h deleted file mode 100644 index 21b02bad696f..000000000000 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_subvp_state.h +++ /dev/null @@ -1,183 +0,0 @@ -/* - * Copyright 2019 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - * Authors: AMD - * - */ - -#ifndef DMUB_SUBVP_STATE_H -#define DMUB_SUBVP_STATE_H - -#include "dmub_cmd.h" - -#define DMUB_SUBVP_INST0 0 -#define DMUB_SUBVP_INST1 1 -#define SUBVP_MAX_WATERMARK 0xFFFF - -struct dmub_subvp_hubp_state { - uint32_t CURSOR0_0_CURSOR_POSITION; - uint32_t CURSOR0_0_CURSOR_HOT_SPOT; - uint32_t CURSOR0_0_CURSOR_DST_OFFSET; - uint32_t CURSOR0_0_CURSOR_SURFACE_ADDRESS_HIGH; - uint32_t CURSOR0_0_CURSOR_SURFACE_ADDRESS; - uint32_t CURSOR0_0_CURSOR_SIZE; - uint32_t CURSOR0_0_CURSOR_CONTROL; - uint32_t HUBPREQ0_CURSOR_SETTINGS; - uint32_t HUBPREQ0_DCSURF_SURFACE_EARLIEST_INUSE_HIGH; - uint32_t HUBPREQ0_DCSURF_SURFACE_EARLIEST_INUSE; - uint32_t HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH; - uint32_t HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS; - uint32_t HUBPREQ0_DCSURF_PRIMARY_META_SURFACE_ADDRESS; - uint32_t HUBPREQ0_DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH; - uint32_t HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; - uint32_t HUBPREQ0_DCSURF_PRIMARY_SURFACE_ADDRESS_C; - uint32_t HUBPREQ0_DCSURF_PRIMARY_META_SURFACE_ADDRESS_HIGH_C; - uint32_t HUBPREQ0_DCSURF_PRIMARY_META_SURFACE_ADDRESS_C; -}; - -enum subvp_error_code { - DMUB_SUBVP_INVALID_STATE, - DMUB_SUBVP_INVALID_TRANSITION, -}; - -enum subvp_state { - DMUB_SUBVP_DISABLED, - DMUB_SUBVP_IDLE, - DMUB_SUBVP_TRY_ACQUIRE_LOCKS, - DMUB_SUBVP_WAIT_FOR_LOCKS, - DMUB_SUBVP_PRECONFIGURE, - DMUB_SUBVP_PREPARE, - DMUB_SUBVP_ENABLE, - DMUB_SUBVP_SWITCHING, - DMUB_SUBVP_END, - DMUB_SUBVP_RESTORE, -}; - -/* Defines information for SUBVP to handle vertical interrupts. */ -struct dmub_subvp_vertical_interrupt_event { - /** - * @inst: Hardware instance of vertical interrupt. - */ - uint8_t otg_inst; - - /** - * @pad: Align structure to 4 byte boundary. - */ - uint8_t pad[3]; - - enum subvp_state curr_state; -}; - -struct dmub_subvp_vertical_interrupt_state { - /** - * @events: Event list. - */ - struct dmub_subvp_vertical_interrupt_event events[DMUB_MAX_STREAMS]; -}; - -struct dmub_subvp_vline_interrupt_event { - - uint8_t hubp_inst; - uint8_t pad[3]; -}; - -struct dmub_subvp_vline_interrupt_state { - struct dmub_subvp_vline_interrupt_event events[DMUB_MAX_PLANES]; -}; - -struct dmub_subvp_interrupt_ctx { - struct dmub_subvp_vertical_interrupt_state vertical_int; - struct dmub_subvp_vline_interrupt_state vline_int; -}; - -struct dmub_subvp_pipe_state { - uint32_t pix_clk_100hz; - uint16_t main_vblank_start; - uint16_t main_vblank_end; - uint16_t mall_region_lines; - uint16_t prefetch_lines; - uint16_t prefetch_to_mall_start_lines; - uint16_t processing_delay_lines; - uint8_t main_pipe_index; - uint8_t phantom_pipe_index; - uint16_t htotal; // htotal for main / phantom pipe - uint16_t vtotal; - uint16_t optc_underflow_count; - uint16_t hubp_underflow_count; - uint8_t pad[2]; -}; - -/** - * struct dmub_subvp_vblank_drr_info - Store DRR state when handling - * SubVP + VBLANK with DRR multi-display case. - * - * The info stored in this struct is only valid if drr_in_use = 1. - */ -struct dmub_subvp_vblank_drr_info { - uint8_t drr_in_use; - uint8_t drr_window_size_ms; // DRR window size -- indicates largest VMIN/VMAX adjustment per frame - uint16_t min_vtotal_supported; // Min VTOTAL that supports switching in VBLANK - uint16_t max_vtotal_supported; // Max VTOTAL that can still support SubVP static scheduling requirements - uint16_t prev_vmin; // Store VMIN value before MCLK switch (used to restore after MCLK end) - uint16_t prev_vmax; // Store VMAX value before MCLK switch (used to restore after MCLK end) - uint8_t use_ramping; // Use ramping or not - uint8_t pad[1]; -}; - -struct dmub_subvp_vblank_pipe_info { - uint32_t pix_clk_100hz; - uint16_t vblank_start; - uint16_t vblank_end; - uint16_t vstartup_start; - uint16_t vtotal; - uint16_t htotal; - uint8_t pipe_index; - uint8_t pad[1]; - struct dmub_subvp_vblank_drr_info drr_info; // DRR considered as part of SubVP + VBLANK case -}; - -enum subvp_switch_type { - DMUB_SUBVP_ONLY, // Used for SubVP only, and SubVP + VACTIVE - DMUB_SUBVP_AND_SUBVP, // 2 SubVP displays - DMUB_SUBVP_AND_VBLANK, - DMUB_SUBVP_AND_FPO, -}; - -/* SubVP state. */ -struct dmub_subvp_state { - struct dmub_subvp_pipe_state pipe_state[DMUB_MAX_SUBVP_STREAMS]; - struct dmub_subvp_interrupt_ctx int_ctx; - struct dmub_subvp_vblank_pipe_info vblank_info; - enum subvp_state state; // current state - enum subvp_switch_type switch_type; // enum take up 4 bytes (?) - uint8_t mclk_pending; - uint8_t num_subvp_streams; - uint8_t vertical_int_margin_us; - uint8_t pstate_allow_width_us; - uint32_t subvp_mclk_switch_count; - uint32_t subvp_wait_lock_count; - uint32_t driver_wait_lock_count; - uint32_t subvp_vblank_frame_count; - uint16_t watermark_a_cache; - uint8_t pad[2]; -}; - -#endif /* _DMUB_SUBVP_STATE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c index 5e952541e72d..094e9f864557 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c @@ -352,6 +352,14 @@ union dmub_fw_boot_status dmub_dcn31_get_fw_boot_status(struct dmub_srv *dmub) return status; } +union dmub_fw_boot_options dmub_dcn31_get_fw_boot_option(struct dmub_srv *dmub) +{ + union dmub_fw_boot_options option; + + option.all = REG_READ(DMCUB_SCRATCH14); + return option; +} + void dmub_dcn31_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmub_srv_hw_params *params) { union dmub_fw_boot_options boot_options = {0}; diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h index 89c5a948b67d..4d520a893c7b 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.h @@ -239,6 +239,8 @@ void dmub_dcn31_skip_dmub_panel_power_sequence(struct dmub_srv *dmub, bool skip) union dmub_fw_boot_status dmub_dcn31_get_fw_boot_status(struct dmub_srv *dmub); +union dmub_fw_boot_options dmub_dcn31_get_fw_boot_option(struct dmub_srv *dmub); + void dmub_dcn31_setup_outbox0(struct dmub_srv *dmub, const struct dmub_region *outbox0); diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c index bdaf43892f47..93624ffe4eb8 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -255,6 +255,7 @@ static bool dmub_srv_hw_setup(struct dmub_srv *dmub, enum dmub_asic asic) funcs->get_gpint_response = dmub_dcn31_get_gpint_response; funcs->get_gpint_dataout = dmub_dcn31_get_gpint_dataout; funcs->get_fw_status = dmub_dcn31_get_fw_boot_status; + funcs->get_fw_boot_option = dmub_dcn31_get_fw_boot_option; funcs->enable_dmub_boot_options = dmub_dcn31_enable_dmub_boot_options; funcs->skip_dmub_panel_power_sequence = dmub_dcn31_skip_dmub_panel_power_sequence; //outbox0 call stacks @@ -639,11 +640,11 @@ enum dmub_status dmub_srv_hw_init(struct dmub_srv *dmub, if (dmub->hw_funcs.enable_dmub_boot_options) dmub->hw_funcs.enable_dmub_boot_options(dmub, params); - if (dmub->hw_funcs.skip_dmub_panel_power_sequence) + if (dmub->hw_funcs.skip_dmub_panel_power_sequence && !dmub->is_virtual) dmub->hw_funcs.skip_dmub_panel_power_sequence(dmub, params->skip_panel_power_sequence); - if (dmub->hw_funcs.reset_release) + if (dmub->hw_funcs.reset_release && !dmub->is_virtual) dmub->hw_funcs.reset_release(dmub); dmub->hw_init = true; @@ -846,6 +847,32 @@ enum dmub_status dmub_srv_get_fw_boot_status(struct dmub_srv *dmub, return DMUB_STATUS_OK; } +enum dmub_status dmub_srv_get_fw_boot_option(struct dmub_srv *dmub, + union dmub_fw_boot_options *option) +{ + option->all = 0; + + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (dmub->hw_funcs.get_fw_boot_option) + *option = dmub->hw_funcs.get_fw_boot_option(dmub); + + return DMUB_STATUS_OK; +} + +enum dmub_status dmub_srv_set_skip_panel_power_sequence(struct dmub_srv *dmub, + bool skip) +{ + if (!dmub->sw_init) + return DMUB_STATUS_INVALID; + + if (dmub->hw_funcs.skip_dmub_panel_power_sequence && !dmub->is_virtual) + dmub->hw_funcs.skip_dmub_panel_power_sequence(dmub, skip); + + return DMUB_STATUS_OK; +} + enum dmub_status dmub_srv_cmd_with_reply_data(struct dmub_srv *dmub, union dmub_rb_cmd *cmd) { diff --git a/drivers/gpu/drm/amd/display/include/ddc_service_types.h b/drivers/gpu/drm/amd/display/include/ddc_service_types.h index f843fc497855..68dfc7968017 100644 --- a/drivers/gpu/drm/amd/display/include/ddc_service_types.h +++ b/drivers/gpu/drm/amd/display/include/ddc_service_types.h @@ -40,6 +40,7 @@ #define DP_BRANCH_HW_REV_20 0x20 #define DP_DEVICE_ID_38EC11 0x38EC11 +#define DP_DEVICE_ID_BA4159 0xBA4159 #define DP_FORCE_PSRSU_CAPABILITY 0x40F #define DP_SINK_PSR_ACTIVE_VTOTAL 0x373 diff --git a/drivers/gpu/drm/amd/display/include/dpcd_defs.h b/drivers/gpu/drm/amd/display/include/dpcd_defs.h index c062a44db078..914f28e9f224 100644 --- a/drivers/gpu/drm/amd/display/include/dpcd_defs.h +++ b/drivers/gpu/drm/amd/display/include/dpcd_defs.h @@ -172,6 +172,9 @@ enum dpcd_psr_sink_states { #define DP_SOURCE_BACKLIGHT_CURRENT_PEAK 0x326 #define DP_SOURCE_BACKLIGHT_CONTROL 0x32E #define DP_SOURCE_BACKLIGHT_ENABLE 0x32F -#define DP_SOURCE_MINIMUM_HBLANK_SUPPORTED 0x340 +#define DP_SOURCE_MINIMUM_HBLANK_SUPPORTED 0x340 +#define DP_SINK_PR_REPLAY_STATUS 0x378 +#define DP_SINK_PR_PIXEL_DEVIATION_PER_LINE 0x379 +#define DP_SINK_PR_MAX_NUMBER_OF_DEVIATION_LINE 0x37A #endif /* __DAL_DPCD_DEFS_H__ */ diff --git a/drivers/gpu/drm/amd/display/include/link_service_types.h b/drivers/gpu/drm/amd/display/include/link_service_types.h index cd870af5fd25..1b8ab20f1715 100644 --- a/drivers/gpu/drm/amd/display/include/link_service_types.h +++ b/drivers/gpu/drm/amd/display/include/link_service_types.h @@ -53,7 +53,7 @@ enum { BITS_PER_DP_BYTE = 10, DATA_EFFICIENCY_8b_10b_x10000 = 8000, /* 80% data efficiency */ DATA_EFFICIENCY_8b_10b_FEC_EFFICIENCY_x100 = 97, /* 97% data efficiency when FEC is enabled */ - DATA_EFFICIENCY_128b_132b_x10000 = 9646, /* 96.71% data efficiency x 99.75% downspread factor */ + DATA_EFFICIENCY_128b_132b_x10000 = 9641, /* 96.71% data efficiency x 99.7% downspread factor */ }; enum lttpr_mode { diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c index 67a062af3ab0..ff8e5708735d 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c @@ -359,7 +359,7 @@ static struct fixed31_32 translate_from_linear_space( scratch_1 = dc_fixpt_add(one, args->a3); /* In the first region (first 16 points) and in the * region delimited by START/END we calculate with - * full precision to avoid error accumulation. + * full precision to avoid error accumulation. */ if ((cal_buffer->buffer_index >= PRECISE_LUT_REGION_START && cal_buffer->buffer_index <= PRECISE_LUT_REGION_END) || @@ -379,8 +379,7 @@ static struct fixed31_32 translate_from_linear_space( scratch_1 = dc_fixpt_sub(scratch_1, args->a2); return scratch_1; - } - else + } else return dc_fixpt_mul(args->arg, args->a1); } diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index dbd60811f95d..ef3a67409021 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -338,7 +338,9 @@ static void apply_below_the_range(struct core_freesync *core_freesync, * - Delta for CEIL: delta_from_mid_point_in_us_1 * - Delta for FLOOR: delta_from_mid_point_in_us_2 */ - if ((last_render_time_in_us / mid_point_frames_ceil) < in_out_vrr->min_duration_in_us) { + if (mid_point_frames_ceil && + (last_render_time_in_us / mid_point_frames_ceil) < + in_out_vrr->min_duration_in_us) { /* Check for out of range. * If using CEIL produces a value that is out of range, * then we are forced to use FLOOR. @@ -385,8 +387,9 @@ static void apply_below_the_range(struct core_freesync *core_freesync, /* Either we've calculated the number of frames to insert, * or we need to insert min duration frames */ - if (last_render_time_in_us / frames_to_insert < - in_out_vrr->min_duration_in_us){ + if (frames_to_insert && + (last_render_time_in_us / frames_to_insert) < + in_out_vrr->min_duration_in_us){ frames_to_insert -= (frames_to_insert > 1) ? 1 : 0; } diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index ec64f19e1786..84f9b412a4f1 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -149,6 +149,8 @@ void mod_build_vsc_infopacket(const struct dc_stream_state *stream, /* VSC packet set to 4 for PSR-SU, or 2 for PSR1 */ if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_SU_1) vsc_packet_revision = vsc_packet_rev4; + else if (stream->link->replay_settings.config.replay_supported) + vsc_packet_revision = vsc_packet_rev4; else if (stream->link->psr_settings.psr_version == DC_PSR_VERSION_1) vsc_packet_revision = vsc_packet_rev2; @@ -536,6 +538,9 @@ void mod_build_adaptive_sync_infopacket(const struct dc_stream_state *stream, case FREESYNC_TYPE_PCON_IN_WHITELIST: mod_build_adaptive_sync_infopacket_v1(info_packet); break; + case ADAPTIVE_SYNC_TYPE_EDP: + mod_build_adaptive_sync_infopacket_v1(info_packet); + break; case ADAPTIVE_SYNC_TYPE_NONE: case FREESYNC_TYPE_PCON_NOT_IN_WHITELIST: default: diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index 30349881a283..73a2b37fbbd7 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -926,6 +926,11 @@ void mod_power_calc_psr_configs(struct psr_config *psr_config, !link->dpcd_caps.psr_info.psr_dpcd_caps.bits.LINK_TRAINING_ON_EXIT_NOT_REQUIRED; } +void init_replay_config(struct dc_link *link, struct replay_config *pr_config) +{ + link->replay_settings.config = *pr_config; +} + bool mod_power_only_edp(const struct dc_state *context, const struct dc_stream_state *stream) { return context && context->stream_count == 1 && dc_is_embedded_signal(stream->signal); diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h index ffc924c9991b..d9e0d67d67f7 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h @@ -53,6 +53,8 @@ bool dmub_init_abm_config(struct resource_pool *res_pool, struct dmcu_iram_parameters params, unsigned int inst); +void init_replay_config(struct dc_link *link, struct replay_config *pr_config); + bool is_psr_su_specific_panel(struct dc_link *link); void mod_power_calc_psr_configs(struct psr_config *psr_config, struct dc_link *link, diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index f175e65b853a..67d7b7ee8a2a 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -240,6 +240,7 @@ enum DC_FEATURE_MASK { DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default DC_PSR_ALLOW_SMU_OPT = (1 << 7), //0x80, disabled by default DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8), //0x100, disabled by default + DC_REPLAY_MASK = (1 << 9), //0x200, disabled by default for dcn < 3.1.4 }; enum DC_DEBUG_MASK { @@ -250,6 +251,8 @@ enum DC_DEBUG_MASK { DC_DISABLE_PSR = 0x10, DC_FORCE_SUBVP_MCLK_SWITCH = 0x20, DC_DISABLE_MPO = 0x40, + DC_DISABLE_REPLAY = 0x50, + DC_ENABLE_DPIA_TRACE = 0x80, }; enum amd_dpm_forced_level; diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h index 537aee0536d3..f2f8f9b39c6b 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_offset.h @@ -15805,6 +15805,11 @@ #define mmDME6_DME_MEMORY_CONTROL 0x093d #define mmDME6_DME_MEMORY_CONTROL_BASE_IDX 3 +// addressBlock: dce_dc_hpo_hpo_top_dispdec +// base address: 0x0 +#define mmHPO_TOP_CLOCK_CONTROL 0x0e43 +#define mmHPO_TOP_CLOCK_CONTROL_BASE_IDX 3 + // base address: 0x1a698 #define mmDC_PERFMON29_PERFCOUNTER_CNTL 0x0e66 #define mmDC_PERFMON29_PERFCOUNTER_CNTL_BASE_IDX 3 diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h index f9d90b098519..e0a447351623 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_0_sh_mask.h @@ -60666,7 +60666,12 @@ #define DME6_DME_MEMORY_CONTROL__DME_MEM_PWR_STATE_MASK 0x00000300L #define DME6_DME_MEMORY_CONTROL__DME_MEM_DEFAULT_MEM_LOW_POWER_STATE_MASK 0x00003000L +// addressBlock: dce_dc_hpo_hpo_top_dispdec +//HPO_TOP_CLOCK_CONTROL +#define HPO_TOP_CLOCK_CONTROL__HPO_HDMISTREAMCLK_GATE_DIS__SHIFT 0x9 +#define HPO_TOP_CLOCK_CONTROL__HPO_HDMISTREAMCLK_GATE_DIS_MASK 0x00000200L +// addressBlock: dce_dc_hpo_hpo_dcperfmon_dc_perfmon_dispdec //DC_PERFMON29_PERFCOUNTER_CNTL #define DC_PERFMON29_PERFCOUNTER_CNTL__PERFCOUNTER_EVENT_SEL__SHIFT 0x0 #define DC_PERFMON29_PERFCOUNTER_CNTL__PERFCOUNTER_CVALUE_SEL__SHIFT 0x9 diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_offset.h index 476469d41d73..b45a35aae241 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_offset.h @@ -14205,6 +14205,10 @@ +// addressBlock: dce_dc_hpo_hpo_top_dispdec +// base address: 0x0 +#define mmHPO_TOP_CLOCK_CONTROL 0x0e43 +#define mmHPO_TOP_CLOCK_CONTROL_BASE_IDX 3 // base address: 0x1a698 #define mmDC_PERFMON26_PERFCOUNTER_CNTL 0x0e66 diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_sh_mask.h index b9de0ebc8b03..3dae29f9581e 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_0_2_sh_mask.h @@ -52401,7 +52401,10 @@ #define DC_PERFMON25_PERFMON_LOW__PERFMON_LOW__SHIFT 0x0 #define DC_PERFMON25_PERFMON_LOW__PERFMON_LOW_MASK 0xFFFFFFFFL - +// addressBlock: dce_dc_hpo_hpo_top_dispdec +//HPO_TOP_CLOCK_CONTROL +#define HPO_TOP_CLOCK_CONTROL__HPO_HDMISTREAMCLK_GATE_DIS__SHIFT 0x9 +#define HPO_TOP_CLOCK_CONTROL__HPO_HDMISTREAMCLK_GATE_DIS_MASK 0x00000200L // addressBlock: dce_dc_hpo_hpo_dcperfmon_dc_perfmon_dispdec //DC_PERFMON26_PERFCOUNTER_CNTL diff --git a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_sh_mask.h index a22481e7bcdb..e0c28c29ddb0 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/nbio/nbio_7_9_0_sh_mask.h @@ -38896,5 +38896,13 @@ #define RCC_DEV0_EPF0_VF7_GFXMSIX_PBA__MSIX_PENDING_BITS_0_MASK 0x00000001L #define RCC_DEV0_EPF0_VF7_GFXMSIX_PBA__MSIX_PENDING_BITS_1_MASK 0x00000002L +//PCIE_PERF_CNTL_TXCLK3 +#define PCIE_PERF_CNTL_TXCLK3__EVENT0_SEL__SHIFT 0x0 +#define PCIE_PERF_CNTL_TXCLK3__EVENT0_SEL_MASK 0x000000FFL + +//PCIE_PERF_CNTL_TXCLK7 +#define PCIE_PERF_CNTL_TXCLK7__EVENT0_SEL__SHIFT 0x0 +#define PCIE_PERF_CNTL_TXCLK7__EVENT0_SEL_MASK 0x000000FFL + #endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_6_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_6_1_0_offset.h new file mode 100644 index 000000000000..a5e7ba5d99ca --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_6_1_0_offset.h @@ -0,0 +1,279 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef _osssys_6_1_0_OFFSET_HEADER +#define _osssys_6_1_0_OFFSET_HEADER + + + +// addressBlock: osssys_osssysdec +// base address: 0x4280 +#define regIH_VMID_0_LUT 0x0000 +#define regIH_VMID_0_LUT_BASE_IDX 0 +#define regIH_VMID_1_LUT 0x0001 +#define regIH_VMID_1_LUT_BASE_IDX 0 +#define regIH_VMID_2_LUT 0x0002 +#define regIH_VMID_2_LUT_BASE_IDX 0 +#define regIH_VMID_3_LUT 0x0003 +#define regIH_VMID_3_LUT_BASE_IDX 0 +#define regIH_VMID_4_LUT 0x0004 +#define regIH_VMID_4_LUT_BASE_IDX 0 +#define regIH_VMID_5_LUT 0x0005 +#define regIH_VMID_5_LUT_BASE_IDX 0 +#define regIH_VMID_6_LUT 0x0006 +#define regIH_VMID_6_LUT_BASE_IDX 0 +#define regIH_VMID_7_LUT 0x0007 +#define regIH_VMID_7_LUT_BASE_IDX 0 +#define regIH_VMID_8_LUT 0x0008 +#define regIH_VMID_8_LUT_BASE_IDX 0 +#define regIH_VMID_9_LUT 0x0009 +#define regIH_VMID_9_LUT_BASE_IDX 0 +#define regIH_VMID_10_LUT 0x000a +#define regIH_VMID_10_LUT_BASE_IDX 0 +#define regIH_VMID_11_LUT 0x000b +#define regIH_VMID_11_LUT_BASE_IDX 0 +#define regIH_VMID_12_LUT 0x000c +#define regIH_VMID_12_LUT_BASE_IDX 0 +#define regIH_VMID_13_LUT 0x000d +#define regIH_VMID_13_LUT_BASE_IDX 0 +#define regIH_VMID_14_LUT 0x000e +#define regIH_VMID_14_LUT_BASE_IDX 0 +#define regIH_VMID_15_LUT 0x000f +#define regIH_VMID_15_LUT_BASE_IDX 0 +#define regIH_VMID_0_LUT_MM 0x0010 +#define regIH_VMID_0_LUT_MM_BASE_IDX 0 +#define regIH_VMID_1_LUT_MM 0x0011 +#define regIH_VMID_1_LUT_MM_BASE_IDX 0 +#define regIH_VMID_2_LUT_MM 0x0012 +#define regIH_VMID_2_LUT_MM_BASE_IDX 0 +#define regIH_VMID_3_LUT_MM 0x0013 +#define regIH_VMID_3_LUT_MM_BASE_IDX 0 +#define regIH_VMID_4_LUT_MM 0x0014 +#define regIH_VMID_4_LUT_MM_BASE_IDX 0 +#define regIH_VMID_5_LUT_MM 0x0015 +#define regIH_VMID_5_LUT_MM_BASE_IDX 0 +#define regIH_VMID_6_LUT_MM 0x0016 +#define regIH_VMID_6_LUT_MM_BASE_IDX 0 +#define regIH_VMID_7_LUT_MM 0x0017 +#define regIH_VMID_7_LUT_MM_BASE_IDX 0 +#define regIH_VMID_8_LUT_MM 0x0018 +#define regIH_VMID_8_LUT_MM_BASE_IDX 0 +#define regIH_VMID_9_LUT_MM 0x0019 +#define regIH_VMID_9_LUT_MM_BASE_IDX 0 +#define regIH_VMID_10_LUT_MM 0x001a +#define regIH_VMID_10_LUT_MM_BASE_IDX 0 +#define regIH_VMID_11_LUT_MM 0x001b +#define regIH_VMID_11_LUT_MM_BASE_IDX 0 +#define regIH_VMID_12_LUT_MM 0x001c +#define regIH_VMID_12_LUT_MM_BASE_IDX 0 +#define regIH_VMID_13_LUT_MM 0x001d +#define regIH_VMID_13_LUT_MM_BASE_IDX 0 +#define regIH_VMID_14_LUT_MM 0x001e +#define regIH_VMID_14_LUT_MM_BASE_IDX 0 +#define regIH_VMID_15_LUT_MM 0x001f +#define regIH_VMID_15_LUT_MM_BASE_IDX 0 +#define regIH_COOKIE_0 0x0020 +#define regIH_COOKIE_0_BASE_IDX 0 +#define regIH_COOKIE_1 0x0021 +#define regIH_COOKIE_1_BASE_IDX 0 +#define regIH_COOKIE_2 0x0022 +#define regIH_COOKIE_2_BASE_IDX 0 +#define regIH_COOKIE_3 0x0023 +#define regIH_COOKIE_3_BASE_IDX 0 +#define regIH_COOKIE_4 0x0024 +#define regIH_COOKIE_4_BASE_IDX 0 +#define regIH_COOKIE_5 0x0025 +#define regIH_COOKIE_5_BASE_IDX 0 +#define regIH_COOKIE_6 0x0026 +#define regIH_COOKIE_6_BASE_IDX 0 +#define regIH_COOKIE_7 0x0027 +#define regIH_COOKIE_7_BASE_IDX 0 +#define regIH_REGISTER_LAST_PART0 0x003f +#define regIH_REGISTER_LAST_PART0_BASE_IDX 0 +#define regIH_RB_CNTL 0x0080 +#define regIH_RB_CNTL_BASE_IDX 0 +#define regIH_RB_RPTR 0x0081 +#define regIH_RB_RPTR_BASE_IDX 0 +#define regIH_RB_WPTR 0x0082 +#define regIH_RB_WPTR_BASE_IDX 0 +#define regIH_RB_BASE 0x0083 +#define regIH_RB_BASE_BASE_IDX 0 +#define regIH_RB_BASE_HI 0x0084 +#define regIH_RB_BASE_HI_BASE_IDX 0 +#define regIH_RB_WPTR_ADDR_HI 0x0085 +#define regIH_RB_WPTR_ADDR_HI_BASE_IDX 0 +#define regIH_RB_WPTR_ADDR_LO 0x0086 +#define regIH_RB_WPTR_ADDR_LO_BASE_IDX 0 +#define regIH_DOORBELL_RPTR 0x0087 +#define regIH_DOORBELL_RPTR_BASE_IDX 0 +#define regIH_DOORBELL_RETRY_CAM 0x0088 +#define regIH_DOORBELL_RETRY_CAM_BASE_IDX 0 +#define regIH_RB_CNTL_RING1 0x008c +#define regIH_RB_CNTL_RING1_BASE_IDX 0 +#define regIH_RB_RPTR_RING1 0x008d +#define regIH_RB_RPTR_RING1_BASE_IDX 0 +#define regIH_RB_WPTR_RING1 0x008e +#define regIH_RB_WPTR_RING1_BASE_IDX 0 +#define regIH_RB_BASE_RING1 0x008f +#define regIH_RB_BASE_RING1_BASE_IDX 0 +#define regIH_RB_BASE_HI_RING1 0x0090 +#define regIH_RB_BASE_HI_RING1_BASE_IDX 0 +#define regIH_DOORBELL_RPTR_RING1 0x0093 +#define regIH_DOORBELL_RPTR_RING1_BASE_IDX 0 +#define regIH_RETRY_CAM_ACK 0x00a4 +#define regIH_RETRY_CAM_ACK_BASE_IDX 0 +#define regIH_VERSION 0x00a5 +#define regIH_VERSION_BASE_IDX 0 +#define regIH_CNTL 0x00a8 +#define regIH_CNTL_BASE_IDX 0 +#define regIH_CLK_CTRL 0x00a9 +#define regIH_CLK_CTRL_BASE_IDX 0 +#define regIH_STORM_CLIENT_LIST_CNTL 0x00aa +#define regIH_STORM_CLIENT_LIST_CNTL_BASE_IDX 0 +#define regIH_LIMIT_INT_RATE_CNTL 0x00ab +#define regIH_LIMIT_INT_RATE_CNTL_BASE_IDX 0 +#define regIH_RETRY_INT_CAM_CNTL 0x00ac +#define regIH_RETRY_INT_CAM_CNTL_BASE_IDX 0 +#define regIH_MEM_POWER_CTRL 0x00ad +#define regIH_MEM_POWER_CTRL_BASE_IDX 0 +#define regIH_MEM_POWER_CTRL2 0x00ae +#define regIH_MEM_POWER_CTRL2_BASE_IDX 0 +#define regIH_CNTL2 0x00c1 +#define regIH_CNTL2_BASE_IDX 0 +#define regIH_STATUS 0x00c2 +#define regIH_STATUS_BASE_IDX 0 +#define regIH_PERFMON_CNTL 0x00c3 +#define regIH_PERFMON_CNTL_BASE_IDX 0 +#define regIH_PERFCOUNTER0_RESULT 0x00c4 +#define regIH_PERFCOUNTER0_RESULT_BASE_IDX 0 +#define regIH_PERFCOUNTER1_RESULT 0x00c5 +#define regIH_PERFCOUNTER1_RESULT_BASE_IDX 0 +#define regIH_DSM_MATCH_VALUE_BIT_31_0 0x00c7 +#define regIH_DSM_MATCH_VALUE_BIT_31_0_BASE_IDX 0 +#define regIH_DSM_MATCH_VALUE_BIT_63_32 0x00c8 +#define regIH_DSM_MATCH_VALUE_BIT_63_32_BASE_IDX 0 +#define regIH_DSM_MATCH_VALUE_BIT_95_64 0x00c9 +#define regIH_DSM_MATCH_VALUE_BIT_95_64_BASE_IDX 0 +#define regIH_DSM_MATCH_FIELD_CONTROL 0x00ca +#define regIH_DSM_MATCH_FIELD_CONTROL_BASE_IDX 0 +#define regIH_DSM_MATCH_DATA_CONTROL 0x00cb +#define regIH_DSM_MATCH_DATA_CONTROL_BASE_IDX 0 +#define regIH_DSM_MATCH_FCN_ID 0x00cc +#define regIH_DSM_MATCH_FCN_ID_BASE_IDX 0 +#define regIH_VF_RB_STATUS 0x00ce +#define regIH_VF_RB_STATUS_BASE_IDX 0 +#define regIH_VF_RB_STATUS2 0x00cf +#define regIH_VF_RB_STATUS2_BASE_IDX 0 +#define regIH_VF_RB1_STATUS 0x00d0 +#define regIH_VF_RB1_STATUS_BASE_IDX 0 +#define regIH_VF_RB1_STATUS2 0x00d1 +#define regIH_VF_RB1_STATUS2_BASE_IDX 0 +#define regIH_RB_STATUS 0x00d4 +#define regIH_RB_STATUS_BASE_IDX 0 +#define regIH_INT_FLOOD_CNTL 0x00d5 +#define regIH_INT_FLOOD_CNTL_BASE_IDX 0 +#define regIH_RB0_INT_FLOOD_STATUS 0x00d6 +#define regIH_RB0_INT_FLOOD_STATUS_BASE_IDX 0 +#define regIH_RB1_INT_FLOOD_STATUS 0x00d7 +#define regIH_RB1_INT_FLOOD_STATUS_BASE_IDX 0 +#define regIH_INT_FLOOD_STATUS 0x00d9 +#define regIH_INT_FLOOD_STATUS_BASE_IDX 0 +#define regIH_INT_FLAGS 0x00dc +#define regIH_INT_FLAGS_BASE_IDX 0 +#define regIH_LAST_INT_INFO0 0x00dd +#define regIH_LAST_INT_INFO0_BASE_IDX 0 +#define regIH_LAST_INT_INFO1 0x00de +#define regIH_LAST_INT_INFO1_BASE_IDX 0 +#define regIH_LAST_INT_INFO2 0x00df +#define regIH_LAST_INT_INFO2_BASE_IDX 0 +#define regIH_SCRATCH 0x00e0 +#define regIH_SCRATCH_BASE_IDX 0 +#define regIH_CLIENT_CREDIT_ERROR 0x00e1 +#define regIH_CLIENT_CREDIT_ERROR_BASE_IDX 0 +#define regIH_GPU_IOV_VIOLATION_LOG 0x00e2 +#define regIH_GPU_IOV_VIOLATION_LOG_BASE_IDX 0 +#define regIH_GPU_IOV_VIOLATION_LOG2 0x00e3 +#define regIH_GPU_IOV_VIOLATION_LOG2_BASE_IDX 0 +#define regIH_COOKIE_REC_VIOLATION_LOG 0x00e4 +#define regIH_COOKIE_REC_VIOLATION_LOG_BASE_IDX 0 +#define regIH_CREDIT_STATUS 0x00e5 +#define regIH_CREDIT_STATUS_BASE_IDX 0 +#define regIH_MMHUB_ERROR 0x00e6 +#define regIH_MMHUB_ERROR_BASE_IDX 0 +#define regIH_VF_RB_STATUS3 0x00ea +#define regIH_VF_RB_STATUS3_BASE_IDX 0 +#define regIH_VF_RB_STATUS4 0x00eb +#define regIH_VF_RB_STATUS4_BASE_IDX 0 +#define regIH_VF_RB1_STATUS3 0x00ec +#define regIH_VF_RB1_STATUS3_BASE_IDX 0 +#define regIH_MSI_STORM_CTRL 0x00f1 +#define regIH_MSI_STORM_CTRL_BASE_IDX 0 +#define regIH_MSI_STORM_CLIENT_INDEX 0x00f2 +#define regIH_MSI_STORM_CLIENT_INDEX_BASE_IDX 0 +#define regIH_MSI_STORM_CLIENT_DATA 0x00f3 +#define regIH_MSI_STORM_CLIENT_DATA_BASE_IDX 0 +#define regIH_REGISTER_LAST_PART2 0x00ff +#define regIH_REGISTER_LAST_PART2_BASE_IDX 0 +#define regSEM_MAILBOX 0x010a +#define regSEM_MAILBOX_BASE_IDX 0 +#define regSEM_MAILBOX_CLEAR 0x010b +#define regSEM_MAILBOX_CLEAR_BASE_IDX 0 +#define regSEM_REGISTER_LAST_PART2 0x017f +#define regSEM_REGISTER_LAST_PART2_BASE_IDX 0 +#define regIH_ACTIVE_FCN_ID 0x0180 +#define regIH_ACTIVE_FCN_ID_BASE_IDX 0 +#define regIH_VIRT_RESET_REQ 0x0181 +#define regIH_VIRT_RESET_REQ_BASE_IDX 0 +#define regIH_CLIENT_CFG 0x0184 +#define regIH_CLIENT_CFG_BASE_IDX 0 +#define regIH_RING1_CLIENT_CFG_INDEX 0x0185 +#define regIH_RING1_CLIENT_CFG_INDEX_BASE_IDX 0 +#define regIH_RING1_CLIENT_CFG_DATA 0x0186 +#define regIH_RING1_CLIENT_CFG_DATA_BASE_IDX 0 +#define regIH_CLIENT_CFG_INDEX 0x0188 +#define regIH_CLIENT_CFG_INDEX_BASE_IDX 0 +#define regIH_CLIENT_CFG_DATA 0x0189 +#define regIH_CLIENT_CFG_DATA_BASE_IDX 0 +#define regIH_CLIENT_CFG_DATA2 0x018a +#define regIH_CLIENT_CFG_DATA2_BASE_IDX 0 +#define regIH_CID_REMAP_INDEX 0x018b +#define regIH_CID_REMAP_INDEX_BASE_IDX 0 +#define regIH_CID_REMAP_DATA 0x018c +#define regIH_CID_REMAP_DATA_BASE_IDX 0 +#define regIH_CHICKEN 0x018d +#define regIH_CHICKEN_BASE_IDX 0 +#define regIH_MMHUB_CNTL 0x018e +#define regIH_MMHUB_CNTL_BASE_IDX 0 +#define regIH_INT_DROP_CNTL 0x018f +#define regIH_INT_DROP_CNTL_BASE_IDX 0 +#define regIH_INT_DROP_MATCH_VALUE0 0x0190 +#define regIH_INT_DROP_MATCH_VALUE0_BASE_IDX 0 +#define regIH_INT_DROP_MATCH_VALUE1 0x0191 +#define regIH_INT_DROP_MATCH_VALUE1_BASE_IDX 0 +#define regIH_INT_DROP_MATCH_MASK0 0x0192 +#define regIH_INT_DROP_MATCH_MASK0_BASE_IDX 0 +#define regIH_INT_DROP_MATCH_MASK1 0x0193 +#define regIH_INT_DROP_MATCH_MASK1_BASE_IDX 0 +#define regIH_REGISTER_LAST_PART1 0x019f +#define regIH_REGISTER_LAST_PART1_BASE_IDX 0 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_6_1_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_6_1_0_sh_mask.h new file mode 100644 index 000000000000..15d5689dde65 --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/oss/osssys_6_1_0_sh_mask.h @@ -0,0 +1,1019 @@ +/* + * Copyright 2023 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ +#ifndef _osssys_6_1_0_SH_MASK_HEADER +#define _osssys_6_1_0_SH_MASK_HEADER + + +// addressBlock: osssys_osssysdec +//IH_VMID_0_LUT +#define IH_VMID_0_LUT__PASID__SHIFT 0x0 +#define IH_VMID_0_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_1_LUT +#define IH_VMID_1_LUT__PASID__SHIFT 0x0 +#define IH_VMID_1_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_2_LUT +#define IH_VMID_2_LUT__PASID__SHIFT 0x0 +#define IH_VMID_2_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_3_LUT +#define IH_VMID_3_LUT__PASID__SHIFT 0x0 +#define IH_VMID_3_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_4_LUT +#define IH_VMID_4_LUT__PASID__SHIFT 0x0 +#define IH_VMID_4_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_5_LUT +#define IH_VMID_5_LUT__PASID__SHIFT 0x0 +#define IH_VMID_5_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_6_LUT +#define IH_VMID_6_LUT__PASID__SHIFT 0x0 +#define IH_VMID_6_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_7_LUT +#define IH_VMID_7_LUT__PASID__SHIFT 0x0 +#define IH_VMID_7_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_8_LUT +#define IH_VMID_8_LUT__PASID__SHIFT 0x0 +#define IH_VMID_8_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_9_LUT +#define IH_VMID_9_LUT__PASID__SHIFT 0x0 +#define IH_VMID_9_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_10_LUT +#define IH_VMID_10_LUT__PASID__SHIFT 0x0 +#define IH_VMID_10_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_11_LUT +#define IH_VMID_11_LUT__PASID__SHIFT 0x0 +#define IH_VMID_11_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_12_LUT +#define IH_VMID_12_LUT__PASID__SHIFT 0x0 +#define IH_VMID_12_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_13_LUT +#define IH_VMID_13_LUT__PASID__SHIFT 0x0 +#define IH_VMID_13_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_14_LUT +#define IH_VMID_14_LUT__PASID__SHIFT 0x0 +#define IH_VMID_14_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_15_LUT +#define IH_VMID_15_LUT__PASID__SHIFT 0x0 +#define IH_VMID_15_LUT__PASID_MASK 0x0000FFFFL +//IH_VMID_0_LUT_MM +#define IH_VMID_0_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_0_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_1_LUT_MM +#define IH_VMID_1_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_1_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_2_LUT_MM +#define IH_VMID_2_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_2_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_3_LUT_MM +#define IH_VMID_3_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_3_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_4_LUT_MM +#define IH_VMID_4_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_4_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_5_LUT_MM +#define IH_VMID_5_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_5_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_6_LUT_MM +#define IH_VMID_6_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_6_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_7_LUT_MM +#define IH_VMID_7_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_7_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_8_LUT_MM +#define IH_VMID_8_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_8_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_9_LUT_MM +#define IH_VMID_9_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_9_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_10_LUT_MM +#define IH_VMID_10_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_10_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_11_LUT_MM +#define IH_VMID_11_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_11_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_12_LUT_MM +#define IH_VMID_12_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_12_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_13_LUT_MM +#define IH_VMID_13_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_13_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_14_LUT_MM +#define IH_VMID_14_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_14_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_VMID_15_LUT_MM +#define IH_VMID_15_LUT_MM__PASID__SHIFT 0x0 +#define IH_VMID_15_LUT_MM__PASID_MASK 0x0000FFFFL +//IH_COOKIE_0 +#define IH_COOKIE_0__CLIENT_ID__SHIFT 0x0 +#define IH_COOKIE_0__SOURCE_ID__SHIFT 0x8 +#define IH_COOKIE_0__RING_ID__SHIFT 0x10 +#define IH_COOKIE_0__VM_ID__SHIFT 0x18 +#define IH_COOKIE_0__RESERVED__SHIFT 0x1c +#define IH_COOKIE_0__VMID_TYPE__SHIFT 0x1f +#define IH_COOKIE_0__CLIENT_ID_MASK 0x000000FFL +#define IH_COOKIE_0__SOURCE_ID_MASK 0x0000FF00L +#define IH_COOKIE_0__RING_ID_MASK 0x00FF0000L +#define IH_COOKIE_0__VM_ID_MASK 0x0F000000L +#define IH_COOKIE_0__RESERVED_MASK 0x70000000L +#define IH_COOKIE_0__VMID_TYPE_MASK 0x80000000L +//IH_COOKIE_1 +#define IH_COOKIE_1__TIMESTAMP_31_0__SHIFT 0x0 +#define IH_COOKIE_1__TIMESTAMP_31_0_MASK 0xFFFFFFFFL +//IH_COOKIE_2 +#define IH_COOKIE_2__TIMESTAMP_47_32__SHIFT 0x0 +#define IH_COOKIE_2__RESERVED__SHIFT 0x10 +#define IH_COOKIE_2__TIMESTAMP_SRC__SHIFT 0x1f +#define IH_COOKIE_2__TIMESTAMP_47_32_MASK 0x0000FFFFL +#define IH_COOKIE_2__RESERVED_MASK 0x7FFF0000L +#define IH_COOKIE_2__TIMESTAMP_SRC_MASK 0x80000000L +//IH_COOKIE_3 +#define IH_COOKIE_3__PAS_ID__SHIFT 0x0 +#define IH_COOKIE_3__RESERVED__SHIFT 0x10 +#define IH_COOKIE_3__PASID_SRC__SHIFT 0x1f +#define IH_COOKIE_3__PAS_ID_MASK 0x0000FFFFL +#define IH_COOKIE_3__RESERVED_MASK 0x7FFF0000L +#define IH_COOKIE_3__PASID_SRC_MASK 0x80000000L +//IH_COOKIE_4 +#define IH_COOKIE_4__CONTEXT_ID_31_0__SHIFT 0x0 +#define IH_COOKIE_4__CONTEXT_ID_31_0_MASK 0xFFFFFFFFL +//IH_COOKIE_5 +#define IH_COOKIE_5__CONTEXT_ID_63_32__SHIFT 0x0 +#define IH_COOKIE_5__CONTEXT_ID_63_32_MASK 0xFFFFFFFFL +//IH_COOKIE_6 +#define IH_COOKIE_6__CONTEXT_ID_95_64__SHIFT 0x0 +#define IH_COOKIE_6__CONTEXT_ID_95_64_MASK 0xFFFFFFFFL +//IH_COOKIE_7 +#define IH_COOKIE_7__CONTEXT_ID_128_96__SHIFT 0x0 +#define IH_COOKIE_7__CONTEXT_ID_128_96_MASK 0xFFFFFFFFL +//IH_REGISTER_LAST_PART0 +#define IH_REGISTER_LAST_PART0__RESERVED__SHIFT 0x0 +#define IH_REGISTER_LAST_PART0__RESERVED_MASK 0xFFFFFFFFL +//IH_RB_CNTL +#define IH_RB_CNTL__RB_ENABLE__SHIFT 0x0 +#define IH_RB_CNTL__RB_SIZE__SHIFT 0x1 +#define IH_RB_CNTL__WPTR_WRITEBACK_ENABLE__SHIFT 0x8 +#define IH_RB_CNTL__RB_FULL_DRAIN_ENABLE__SHIFT 0x9 +#define IH_RB_CNTL__FULL_DRAIN_CLEAR__SHIFT 0xa +#define IH_RB_CNTL__PAGE_RB_CLEAR__SHIFT 0xb +#define IH_RB_CNTL__RB_USED_INT_THRESHOLD__SHIFT 0xc +#define IH_RB_CNTL__WPTR_OVERFLOW_ENABLE__SHIFT 0x10 +#define IH_RB_CNTL__ENABLE_INTR__SHIFT 0x11 +#define IH_RB_CNTL__MC_SWAP__SHIFT 0x12 +#define IH_RB_CNTL__MC_SNOOP__SHIFT 0x14 +#define IH_RB_CNTL__RPTR_REARM__SHIFT 0x15 +#define IH_RB_CNTL__MC_RO__SHIFT 0x16 +#define IH_RB_CNTL__MC_VMID__SHIFT 0x18 +#define IH_RB_CNTL__MC_SPACE__SHIFT 0x1c +#define IH_RB_CNTL__WPTR_OVERFLOW_CLEAR__SHIFT 0x1f +#define IH_RB_CNTL__RB_ENABLE_MASK 0x00000001L +#define IH_RB_CNTL__RB_SIZE_MASK 0x0000003EL +#define IH_RB_CNTL__WPTR_WRITEBACK_ENABLE_MASK 0x00000100L +#define IH_RB_CNTL__RB_FULL_DRAIN_ENABLE_MASK 0x00000200L +#define IH_RB_CNTL__FULL_DRAIN_CLEAR_MASK 0x00000400L +#define IH_RB_CNTL__PAGE_RB_CLEAR_MASK 0x00000800L +#define IH_RB_CNTL__RB_USED_INT_THRESHOLD_MASK 0x0000F000L +#define IH_RB_CNTL__WPTR_OVERFLOW_ENABLE_MASK 0x00010000L +#define IH_RB_CNTL__ENABLE_INTR_MASK 0x00020000L +#define IH_RB_CNTL__MC_SWAP_MASK 0x000C0000L +#define IH_RB_CNTL__MC_SNOOP_MASK 0x00100000L +#define IH_RB_CNTL__RPTR_REARM_MASK 0x00200000L +#define IH_RB_CNTL__MC_RO_MASK 0x00400000L +#define IH_RB_CNTL__MC_VMID_MASK 0x0F000000L +#define IH_RB_CNTL__MC_SPACE_MASK 0x70000000L +#define IH_RB_CNTL__WPTR_OVERFLOW_CLEAR_MASK 0x80000000L +//IH_RB_RPTR +#define IH_RB_RPTR__OFFSET__SHIFT 0x2 +#define IH_RB_RPTR__OFFSET_MASK 0x0003FFFCL +//IH_RB_WPTR +#define IH_RB_WPTR__RB_OVERFLOW__SHIFT 0x0 +#define IH_RB_WPTR__OFFSET__SHIFT 0x2 +#define IH_RB_WPTR__RB_LEFT_NONE__SHIFT 0x12 +#define IH_RB_WPTR__RB_MAY_OVERFLOW__SHIFT 0x13 +#define IH_RB_WPTR__RB_OVERFLOW_MASK 0x00000001L +#define IH_RB_WPTR__OFFSET_MASK 0x0003FFFCL +#define IH_RB_WPTR__RB_LEFT_NONE_MASK 0x00040000L +#define IH_RB_WPTR__RB_MAY_OVERFLOW_MASK 0x00080000L +//IH_RB_BASE +#define IH_RB_BASE__ADDR__SHIFT 0x0 +#define IH_RB_BASE__ADDR_MASK 0xFFFFFFFFL +//IH_RB_BASE_HI +#define IH_RB_BASE_HI__ADDR__SHIFT 0x0 +#define IH_RB_BASE_HI__ADDR_MASK 0x000000FFL +//IH_RB_WPTR_ADDR_HI +#define IH_RB_WPTR_ADDR_HI__ADDR__SHIFT 0x0 +#define IH_RB_WPTR_ADDR_HI__ADDR_MASK 0x0000FFFFL +//IH_RB_WPTR_ADDR_LO +#define IH_RB_WPTR_ADDR_LO__ADDR__SHIFT 0x2 +#define IH_RB_WPTR_ADDR_LO__ADDR_MASK 0xFFFFFFFCL +//IH_DOORBELL_RPTR +#define IH_DOORBELL_RPTR__OFFSET__SHIFT 0x0 +#define IH_DOORBELL_RPTR__ENABLE__SHIFT 0x1c +#define IH_DOORBELL_RPTR__OFFSET_MASK 0x03FFFFFFL +#define IH_DOORBELL_RPTR__ENABLE_MASK 0x10000000L +//IH_DOORBELL_RETRY_CAM +#define IH_DOORBELL_RETRY_CAM__OFFSET__SHIFT 0x0 +#define IH_DOORBELL_RETRY_CAM__ENABLE__SHIFT 0x1c +#define IH_DOORBELL_RETRY_CAM__OFFSET_MASK 0x03FFFFFFL +#define IH_DOORBELL_RETRY_CAM__ENABLE_MASK 0x10000000L +//IH_RB_CNTL_RING1 +#define IH_RB_CNTL_RING1__RB_ENABLE__SHIFT 0x0 +#define IH_RB_CNTL_RING1__RB_SIZE__SHIFT 0x1 +#define IH_RB_CNTL_RING1__RB_FULL_DRAIN_ENABLE__SHIFT 0x9 +#define IH_RB_CNTL_RING1__FULL_DRAIN_CLEAR__SHIFT 0xa +#define IH_RB_CNTL_RING1__PAGE_RB_CLEAR__SHIFT 0xb +#define IH_RB_CNTL_RING1__RB_USED_INT_THRESHOLD__SHIFT 0xc +#define IH_RB_CNTL_RING1__WPTR_OVERFLOW_ENABLE__SHIFT 0x10 +#define IH_RB_CNTL_RING1__MC_SWAP__SHIFT 0x12 +#define IH_RB_CNTL_RING1__MC_SNOOP__SHIFT 0x14 +#define IH_RB_CNTL_RING1__MC_RO__SHIFT 0x16 +#define IH_RB_CNTL_RING1__MC_VMID__SHIFT 0x18 +#define IH_RB_CNTL_RING1__MC_SPACE__SHIFT 0x1c +#define IH_RB_CNTL_RING1__WPTR_OVERFLOW_CLEAR__SHIFT 0x1f +#define IH_RB_CNTL_RING1__RB_ENABLE_MASK 0x00000001L +#define IH_RB_CNTL_RING1__RB_SIZE_MASK 0x0000003EL +#define IH_RB_CNTL_RING1__RB_FULL_DRAIN_ENABLE_MASK 0x00000200L +#define IH_RB_CNTL_RING1__FULL_DRAIN_CLEAR_MASK 0x00000400L +#define IH_RB_CNTL_RING1__PAGE_RB_CLEAR_MASK 0x00000800L +#define IH_RB_CNTL_RING1__RB_USED_INT_THRESHOLD_MASK 0x0000F000L +#define IH_RB_CNTL_RING1__WPTR_OVERFLOW_ENABLE_MASK 0x00010000L +#define IH_RB_CNTL_RING1__MC_SWAP_MASK 0x000C0000L +#define IH_RB_CNTL_RING1__MC_SNOOP_MASK 0x00100000L +#define IH_RB_CNTL_RING1__MC_RO_MASK 0x00400000L +#define IH_RB_CNTL_RING1__MC_VMID_MASK 0x0F000000L +#define IH_RB_CNTL_RING1__MC_SPACE_MASK 0x70000000L +#define IH_RB_CNTL_RING1__WPTR_OVERFLOW_CLEAR_MASK 0x80000000L +//IH_RB_RPTR_RING1 +#define IH_RB_RPTR_RING1__OFFSET__SHIFT 0x2 +#define IH_RB_RPTR_RING1__OFFSET_MASK 0x0003FFFCL +//IH_RB_WPTR_RING1 +#define IH_RB_WPTR_RING1__RB_OVERFLOW__SHIFT 0x0 +#define IH_RB_WPTR_RING1__OFFSET__SHIFT 0x2 +#define IH_RB_WPTR_RING1__RB_LEFT_NONE__SHIFT 0x12 +#define IH_RB_WPTR_RING1__RB_MAY_OVERFLOW__SHIFT 0x13 +#define IH_RB_WPTR_RING1__RB_OVERFLOW_MASK 0x00000001L +#define IH_RB_WPTR_RING1__OFFSET_MASK 0x0003FFFCL +#define IH_RB_WPTR_RING1__RB_LEFT_NONE_MASK 0x00040000L +#define IH_RB_WPTR_RING1__RB_MAY_OVERFLOW_MASK 0x00080000L +//IH_RB_BASE_RING1 +#define IH_RB_BASE_RING1__ADDR__SHIFT 0x0 +#define IH_RB_BASE_RING1__ADDR_MASK 0xFFFFFFFFL +//IH_RB_BASE_HI_RING1 +#define IH_RB_BASE_HI_RING1__ADDR__SHIFT 0x0 +#define IH_RB_BASE_HI_RING1__ADDR_MASK 0x000000FFL +//IH_DOORBELL_RPTR_RING1 +#define IH_DOORBELL_RPTR_RING1__OFFSET__SHIFT 0x0 +#define IH_DOORBELL_RPTR_RING1__ENABLE__SHIFT 0x1c +#define IH_DOORBELL_RPTR_RING1__OFFSET_MASK 0x03FFFFFFL +#define IH_DOORBELL_RPTR_RING1__ENABLE_MASK 0x10000000L +//IH_RETRY_CAM_ACK +#define IH_RETRY_CAM_ACK__INDEX__SHIFT 0x0 +#define IH_RETRY_CAM_ACK__INDEX_MASK 0x000003FFL +//IH_VERSION +#define IH_VERSION__MINVER__SHIFT 0x0 +#define IH_VERSION__MAJVER__SHIFT 0x8 +#define IH_VERSION__REV__SHIFT 0x10 +#define IH_VERSION__MINVER_MASK 0x0000007FL +#define IH_VERSION__MAJVER_MASK 0x00007F00L +#define IH_VERSION__REV_MASK 0x003F0000L +//IH_CNTL +#define IH_CNTL__WPTR_WRITEBACK_TIMER__SHIFT 0x0 +#define IH_CNTL__IH_IDLE_HYSTERESIS_CNTL__SHIFT 0x6 +#define IH_CNTL__IH_FIFO_HIGHWATER__SHIFT 0x8 +#define IH_CNTL__MC_WR_CLEAN_CNT__SHIFT 0x14 +#define IH_CNTL__WPTR_WRITEBACK_TIMER_MASK 0x0000001FL +#define IH_CNTL__IH_IDLE_HYSTERESIS_CNTL_MASK 0x000000C0L +#define IH_CNTL__IH_FIFO_HIGHWATER_MASK 0x00007F00L +#define IH_CNTL__MC_WR_CLEAN_CNT_MASK 0x01F00000L +//IH_CLK_CTRL +#define IH_CLK_CTRL__IH_PASID_LUT_MEM_CLK_SOFT_OVERRIDE__SHIFT 0x17 +#define IH_CLK_CTRL__MSI_STORM_COUNTER_CLK_SOFT_OVERRIDE__SHIFT 0x18 +#define IH_CLK_CTRL__IH_RETRY_INT_CAM_MEM_CLK_SOFT_OVERRIDE__SHIFT 0x19 +#define IH_CLK_CTRL__IH_BUFFER_MEM_CLK_SOFT_OVERRIDE__SHIFT 0x1a +#define IH_CLK_CTRL__DBUS_MUX_CLK_SOFT_OVERRIDE__SHIFT 0x1b +#define IH_CLK_CTRL__OSSSYS_SHARE_CLK_SOFT_OVERRIDE__SHIFT 0x1c +#define IH_CLK_CTRL__LIMIT_SMN_CLK_SOFT_OVERRIDE__SHIFT 0x1d +#define IH_CLK_CTRL__DYN_CLK_SOFT_OVERRIDE__SHIFT 0x1e +#define IH_CLK_CTRL__REG_CLK_SOFT_OVERRIDE__SHIFT 0x1f +#define IH_CLK_CTRL__IH_PASID_LUT_MEM_CLK_SOFT_OVERRIDE_MASK 0x00800000L +#define IH_CLK_CTRL__MSI_STORM_COUNTER_CLK_SOFT_OVERRIDE_MASK 0x01000000L +#define IH_CLK_CTRL__IH_RETRY_INT_CAM_MEM_CLK_SOFT_OVERRIDE_MASK 0x02000000L +#define IH_CLK_CTRL__IH_BUFFER_MEM_CLK_SOFT_OVERRIDE_MASK 0x04000000L +#define IH_CLK_CTRL__DBUS_MUX_CLK_SOFT_OVERRIDE_MASK 0x08000000L +#define IH_CLK_CTRL__OSSSYS_SHARE_CLK_SOFT_OVERRIDE_MASK 0x10000000L +#define IH_CLK_CTRL__LIMIT_SMN_CLK_SOFT_OVERRIDE_MASK 0x20000000L +#define IH_CLK_CTRL__DYN_CLK_SOFT_OVERRIDE_MASK 0x40000000L +#define IH_CLK_CTRL__REG_CLK_SOFT_OVERRIDE_MASK 0x80000000L +//IH_STORM_CLIENT_LIST_CNTL +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT1_IS_STORM_CLIENT__SHIFT 0x1 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT2_IS_STORM_CLIENT__SHIFT 0x2 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT3_IS_STORM_CLIENT__SHIFT 0x3 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT4_IS_STORM_CLIENT__SHIFT 0x4 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT5_IS_STORM_CLIENT__SHIFT 0x5 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT6_IS_STORM_CLIENT__SHIFT 0x6 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT7_IS_STORM_CLIENT__SHIFT 0x7 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT8_IS_STORM_CLIENT__SHIFT 0x8 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT9_IS_STORM_CLIENT__SHIFT 0x9 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT10_IS_STORM_CLIENT__SHIFT 0xa +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT11_IS_STORM_CLIENT__SHIFT 0xb +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT12_IS_STORM_CLIENT__SHIFT 0xc +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT13_IS_STORM_CLIENT__SHIFT 0xd +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT14_IS_STORM_CLIENT__SHIFT 0xe +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT15_IS_STORM_CLIENT__SHIFT 0xf +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT16_IS_STORM_CLIENT__SHIFT 0x10 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT17_IS_STORM_CLIENT__SHIFT 0x11 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT18_IS_STORM_CLIENT__SHIFT 0x12 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT19_IS_STORM_CLIENT__SHIFT 0x13 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT20_IS_STORM_CLIENT__SHIFT 0x14 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT21_IS_STORM_CLIENT__SHIFT 0x15 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT22_IS_STORM_CLIENT__SHIFT 0x16 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT23_IS_STORM_CLIENT__SHIFT 0x17 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT24_IS_STORM_CLIENT__SHIFT 0x18 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT25_IS_STORM_CLIENT__SHIFT 0x19 +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT26_IS_STORM_CLIENT__SHIFT 0x1a +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT27_IS_STORM_CLIENT__SHIFT 0x1b +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT28_IS_STORM_CLIENT__SHIFT 0x1c +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT29_IS_STORM_CLIENT__SHIFT 0x1d +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT30_IS_STORM_CLIENT__SHIFT 0x1e +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT31_IS_STORM_CLIENT__SHIFT 0x1f +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT1_IS_STORM_CLIENT_MASK 0x00000002L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT2_IS_STORM_CLIENT_MASK 0x00000004L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT3_IS_STORM_CLIENT_MASK 0x00000008L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT4_IS_STORM_CLIENT_MASK 0x00000010L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT5_IS_STORM_CLIENT_MASK 0x00000020L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT6_IS_STORM_CLIENT_MASK 0x00000040L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT7_IS_STORM_CLIENT_MASK 0x00000080L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT8_IS_STORM_CLIENT_MASK 0x00000100L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT9_IS_STORM_CLIENT_MASK 0x00000200L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT10_IS_STORM_CLIENT_MASK 0x00000400L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT11_IS_STORM_CLIENT_MASK 0x00000800L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT12_IS_STORM_CLIENT_MASK 0x00001000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT13_IS_STORM_CLIENT_MASK 0x00002000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT14_IS_STORM_CLIENT_MASK 0x00004000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT15_IS_STORM_CLIENT_MASK 0x00008000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT16_IS_STORM_CLIENT_MASK 0x00010000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT17_IS_STORM_CLIENT_MASK 0x00020000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT18_IS_STORM_CLIENT_MASK 0x00040000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT19_IS_STORM_CLIENT_MASK 0x00080000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT20_IS_STORM_CLIENT_MASK 0x00100000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT21_IS_STORM_CLIENT_MASK 0x00200000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT22_IS_STORM_CLIENT_MASK 0x00400000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT23_IS_STORM_CLIENT_MASK 0x00800000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT24_IS_STORM_CLIENT_MASK 0x01000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT25_IS_STORM_CLIENT_MASK 0x02000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT26_IS_STORM_CLIENT_MASK 0x04000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT27_IS_STORM_CLIENT_MASK 0x08000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT28_IS_STORM_CLIENT_MASK 0x10000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT29_IS_STORM_CLIENT_MASK 0x20000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT30_IS_STORM_CLIENT_MASK 0x40000000L +#define IH_STORM_CLIENT_LIST_CNTL__CLIENT31_IS_STORM_CLIENT_MASK 0x80000000L +//IH_LIMIT_INT_RATE_CNTL +#define IH_LIMIT_INT_RATE_CNTL__LIMIT_ENABLE__SHIFT 0x0 +#define IH_LIMIT_INT_RATE_CNTL__PERF_INTERVAL__SHIFT 0x1 +#define IH_LIMIT_INT_RATE_CNTL__PERF_THRESHOLD__SHIFT 0x5 +#define IH_LIMIT_INT_RATE_CNTL__RETURN_DELAY__SHIFT 0x11 +#define IH_LIMIT_INT_RATE_CNTL__PERF_RESULT__SHIFT 0x15 +#define IH_LIMIT_INT_RATE_CNTL__LIMIT_ENABLE_MASK 0x00000001L +#define IH_LIMIT_INT_RATE_CNTL__PERF_INTERVAL_MASK 0x0000001EL +#define IH_LIMIT_INT_RATE_CNTL__PERF_THRESHOLD_MASK 0x0000FFE0L +#define IH_LIMIT_INT_RATE_CNTL__RETURN_DELAY_MASK 0x001E0000L +#define IH_LIMIT_INT_RATE_CNTL__PERF_RESULT_MASK 0xFFE00000L +//IH_RETRY_INT_CAM_CNTL +#define IH_RETRY_INT_CAM_CNTL__CAM_SIZE__SHIFT 0x0 +#define IH_RETRY_INT_CAM_CNTL__BACK_PRESSURE_SKID_VALUE__SHIFT 0x8 +#define IH_RETRY_INT_CAM_CNTL__PER_VF_ENTRY_SIZE__SHIFT 0x14 +#define IH_RETRY_INT_CAM_CNTL__CAM_SIZE_MASK 0x0000001FL +#define IH_RETRY_INT_CAM_CNTL__BACK_PRESSURE_SKID_VALUE_MASK 0x00003F00L +#define IH_RETRY_INT_CAM_CNTL__PER_VF_ENTRY_SIZE_MASK 0x00300000L +//IH_MEM_POWER_CTRL +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_CTRL_EN__SHIFT 0x0 +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_LS_EN__SHIFT 0x1 +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_DS_EN__SHIFT 0x2 +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_SD_EN__SHIFT 0x3 +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_IDLE_HYSTERESIS__SHIFT 0x4 +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_UP_RECOVER_DELAY__SHIFT 0x8 +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_DOWN_ENTER_DELAY__SHIFT 0xe +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_CTRL_EN__SHIFT 0x10 +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_LS_EN__SHIFT 0x11 +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_DS_EN__SHIFT 0x12 +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_SD_EN__SHIFT 0x13 +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_IDLE_HYSTERESIS__SHIFT 0x14 +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_UP_RECOVER_DELAY__SHIFT 0x18 +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_DOWN_ENTER_DELAY__SHIFT 0x1e +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_CTRL_EN_MASK 0x00000001L +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_LS_EN_MASK 0x00000002L +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_DS_EN_MASK 0x00000004L +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_SD_EN_MASK 0x00000008L +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_IDLE_HYSTERESIS_MASK 0x00000070L +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_UP_RECOVER_DELAY_MASK 0x00003F00L +#define IH_MEM_POWER_CTRL__IH_BUFFER_MEM_POWER_DOWN_ENTER_DELAY_MASK 0x0000C000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_CTRL_EN_MASK 0x00010000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_LS_EN_MASK 0x00020000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_DS_EN_MASK 0x00040000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_SD_EN_MASK 0x00080000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_IDLE_HYSTERESIS_MASK 0x00700000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_UP_RECOVER_DELAY_MASK 0x3F000000L +#define IH_MEM_POWER_CTRL__IH_RETRY_INT_CAM_MEM_POWER_DOWN_ENTER_DELAY_MASK 0xC0000000L +//IH_MEM_POWER_CTRL2 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_CTRL_EN__SHIFT 0x0 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_LS_EN__SHIFT 0x1 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_DS_EN__SHIFT 0x2 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_SD_EN__SHIFT 0x3 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_IDLE_HYSTERESIS__SHIFT 0x4 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_UP_RECOVER_DELAY__SHIFT 0x8 +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_DOWN_ENTER_DELAY__SHIFT 0xe +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_CTRL_EN_MASK 0x00000001L +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_LS_EN_MASK 0x00000002L +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_DS_EN_MASK 0x00000004L +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_SD_EN_MASK 0x00000008L +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_IDLE_HYSTERESIS_MASK 0x00000070L +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_UP_RECOVER_DELAY_MASK 0x00003F00L +#define IH_MEM_POWER_CTRL2__IH_PASID_LUT_MEM_POWER_DOWN_ENTER_DELAY_MASK 0x0000C000L +//IH_CNTL2 +#define IH_CNTL2__SELF_IV_FORCE_WPTR_UPDATE_TIMEOUT__SHIFT 0x0 +#define IH_CNTL2__SELF_IV_FORCE_WPTR_UPDATE_ENABLE__SHIFT 0x8 +#define IH_CNTL2__SELF_IV_FORCE_WPTR_UPDATE_TIMEOUT_MASK 0x0000001FL +#define IH_CNTL2__SELF_IV_FORCE_WPTR_UPDATE_ENABLE_MASK 0x00000100L +//IH_STATUS +#define IH_STATUS__IDLE__SHIFT 0x0 +#define IH_STATUS__INPUT_IDLE__SHIFT 0x1 +#define IH_STATUS__BUFFER_IDLE__SHIFT 0x2 +#define IH_STATUS__RB_FULL__SHIFT 0x3 +#define IH_STATUS__RB_FULL_DRAIN__SHIFT 0x4 +#define IH_STATUS__RB_OVERFLOW__SHIFT 0x5 +#define IH_STATUS__MC_WR_IDLE__SHIFT 0x6 +#define IH_STATUS__MC_WR_STALL__SHIFT 0x7 +#define IH_STATUS__MC_WR_CLEAN_PENDING__SHIFT 0x8 +#define IH_STATUS__MC_WR_CLEAN_STALL__SHIFT 0x9 +#define IH_STATUS__BIF_INTERRUPT_LINE__SHIFT 0xa +#define IH_STATUS__SWITCH_READY__SHIFT 0xb +#define IH_STATUS__RB1_FULL__SHIFT 0xc +#define IH_STATUS__RB1_FULL_DRAIN__SHIFT 0xd +#define IH_STATUS__RB1_OVERFLOW__SHIFT 0xe +#define IH_STATUS__SELF_INT_GEN_IDLE__SHIFT 0x12 +#define IH_STATUS__RETRY_INT_CAM_IDLE__SHIFT 0x13 +#define IH_STATUS__ZSTATES_FENCE__SHIFT 0x14 +#define IH_STATUS__IH_BUFFER_MEM_POWER_GATED__SHIFT 0x15 +#define IH_STATUS__IH_RETRY_INT_CAM_MEM_POWER_GATED__SHIFT 0x16 +#define IH_STATUS__IH_PASID_LUT_MEM_POWER_GATED__SHIFT 0x17 +#define IH_STATUS__IDLE_MASK 0x00000001L +#define IH_STATUS__INPUT_IDLE_MASK 0x00000002L +#define IH_STATUS__BUFFER_IDLE_MASK 0x00000004L +#define IH_STATUS__RB_FULL_MASK 0x00000008L +#define IH_STATUS__RB_FULL_DRAIN_MASK 0x00000010L +#define IH_STATUS__RB_OVERFLOW_MASK 0x00000020L +#define IH_STATUS__MC_WR_IDLE_MASK 0x00000040L +#define IH_STATUS__MC_WR_STALL_MASK 0x00000080L +#define IH_STATUS__MC_WR_CLEAN_PENDING_MASK 0x00000100L +#define IH_STATUS__MC_WR_CLEAN_STALL_MASK 0x00000200L +#define IH_STATUS__BIF_INTERRUPT_LINE_MASK 0x00000400L +#define IH_STATUS__SWITCH_READY_MASK 0x00000800L +#define IH_STATUS__RB1_FULL_MASK 0x00001000L +#define IH_STATUS__RB1_FULL_DRAIN_MASK 0x00002000L +#define IH_STATUS__RB1_OVERFLOW_MASK 0x00004000L +#define IH_STATUS__SELF_INT_GEN_IDLE_MASK 0x00040000L +#define IH_STATUS__RETRY_INT_CAM_IDLE_MASK 0x00080000L +#define IH_STATUS__ZSTATES_FENCE_MASK 0x00100000L +#define IH_STATUS__IH_BUFFER_MEM_POWER_GATED_MASK 0x00200000L +#define IH_STATUS__IH_RETRY_INT_CAM_MEM_POWER_GATED_MASK 0x00400000L +#define IH_STATUS__IH_PASID_LUT_MEM_POWER_GATED_MASK 0x00800000L +//IH_PERFMON_CNTL +#define IH_PERFMON_CNTL__ENABLE0__SHIFT 0x0 +#define IH_PERFMON_CNTL__CLEAR0__SHIFT 0x1 +#define IH_PERFMON_CNTL__PERF_SEL0__SHIFT 0x2 +#define IH_PERFMON_CNTL__ENABLE1__SHIFT 0x10 +#define IH_PERFMON_CNTL__CLEAR1__SHIFT 0x11 +#define IH_PERFMON_CNTL__PERF_SEL1__SHIFT 0x12 +#define IH_PERFMON_CNTL__ENABLE0_MASK 0x00000001L +#define IH_PERFMON_CNTL__CLEAR0_MASK 0x00000002L +#define IH_PERFMON_CNTL__PERF_SEL0_MASK 0x00000FFCL +#define IH_PERFMON_CNTL__ENABLE1_MASK 0x00010000L +#define IH_PERFMON_CNTL__CLEAR1_MASK 0x00020000L +#define IH_PERFMON_CNTL__PERF_SEL1_MASK 0x0FFC0000L +//IH_PERFCOUNTER0_RESULT +#define IH_PERFCOUNTER0_RESULT__PERF_COUNT__SHIFT 0x0 +#define IH_PERFCOUNTER0_RESULT__PERF_COUNT_MASK 0xFFFFFFFFL +//IH_PERFCOUNTER1_RESULT +#define IH_PERFCOUNTER1_RESULT__PERF_COUNT__SHIFT 0x0 +#define IH_PERFCOUNTER1_RESULT__PERF_COUNT_MASK 0xFFFFFFFFL +//IH_DSM_MATCH_VALUE_BIT_31_0 +#define IH_DSM_MATCH_VALUE_BIT_31_0__VALUE__SHIFT 0x0 +#define IH_DSM_MATCH_VALUE_BIT_31_0__VALUE_MASK 0xFFFFFFFFL +//IH_DSM_MATCH_VALUE_BIT_63_32 +#define IH_DSM_MATCH_VALUE_BIT_63_32__VALUE__SHIFT 0x0 +#define IH_DSM_MATCH_VALUE_BIT_63_32__VALUE_MASK 0xFFFFFFFFL +//IH_DSM_MATCH_VALUE_BIT_95_64 +#define IH_DSM_MATCH_VALUE_BIT_95_64__VALUE__SHIFT 0x0 +#define IH_DSM_MATCH_VALUE_BIT_95_64__VALUE_MASK 0xFFFFFFFFL +//IH_DSM_MATCH_FIELD_CONTROL +#define IH_DSM_MATCH_FIELD_CONTROL__SRC_EN__SHIFT 0x0 +#define IH_DSM_MATCH_FIELD_CONTROL__FCNID_EN__SHIFT 0x1 +#define IH_DSM_MATCH_FIELD_CONTROL__TIMESTAMP_EN__SHIFT 0x2 +#define IH_DSM_MATCH_FIELD_CONTROL__RINGID_EN__SHIFT 0x3 +#define IH_DSM_MATCH_FIELD_CONTROL__VMID_EN__SHIFT 0x4 +#define IH_DSM_MATCH_FIELD_CONTROL__PASID_EN__SHIFT 0x5 +#define IH_DSM_MATCH_FIELD_CONTROL__CLIENT_ID_EN__SHIFT 0x6 +#define IH_DSM_MATCH_FIELD_CONTROL__SRC_EN_MASK 0x00000001L +#define IH_DSM_MATCH_FIELD_CONTROL__FCNID_EN_MASK 0x00000002L +#define IH_DSM_MATCH_FIELD_CONTROL__TIMESTAMP_EN_MASK 0x00000004L +#define IH_DSM_MATCH_FIELD_CONTROL__RINGID_EN_MASK 0x00000008L +#define IH_DSM_MATCH_FIELD_CONTROL__VMID_EN_MASK 0x00000010L +#define IH_DSM_MATCH_FIELD_CONTROL__PASID_EN_MASK 0x00000020L +#define IH_DSM_MATCH_FIELD_CONTROL__CLIENT_ID_EN_MASK 0x00000040L +//IH_DSM_MATCH_DATA_CONTROL +#define IH_DSM_MATCH_DATA_CONTROL__VALUE__SHIFT 0x0 +#define IH_DSM_MATCH_DATA_CONTROL__VALUE_MASK 0x0FFFFFFFL +//IH_DSM_MATCH_FCN_ID +#define IH_DSM_MATCH_FCN_ID__VF_ID__SHIFT 0x0 +#define IH_DSM_MATCH_FCN_ID__PF_VF__SHIFT 0x7 +#define IH_DSM_MATCH_FCN_ID__VF_ID_MASK 0x0000000FL +#define IH_DSM_MATCH_FCN_ID__PF_VF_MASK 0x00000080L +//IH_VF_RB_STATUS +#define IH_VF_RB_STATUS__RB_FULL_DRAIN_VF__SHIFT 0x0 +#define IH_VF_RB_STATUS__RB_FULL_DRAIN_VF_MASK 0x0000FFFFL +//IH_VF_RB_STATUS2 +#define IH_VF_RB_STATUS2__RB_FULL_VF__SHIFT 0x0 +#define IH_VF_RB_STATUS2__RB_FULL_VF_MASK 0x0000FFFFL +//IH_VF_RB1_STATUS +#define IH_VF_RB1_STATUS__RB_FULL_DRAIN_VF__SHIFT 0x0 +#define IH_VF_RB1_STATUS__RB_FULL_DRAIN_VF_MASK 0x0000FFFFL +//IH_VF_RB1_STATUS2 +#define IH_VF_RB1_STATUS2__RB_FULL_VF__SHIFT 0x0 +#define IH_VF_RB1_STATUS2__RB_FULL_VF_MASK 0x0000FFFFL +//IH_RB_STATUS +#define IH_RB_STATUS__RB_FULL__SHIFT 0x0 +#define IH_RB_STATUS__RB_FULL_DRAIN__SHIFT 0x1 +#define IH_RB_STATUS__RB_OVERFLOW__SHIFT 0x2 +#define IH_RB_STATUS__RB1_FULL__SHIFT 0x4 +#define IH_RB_STATUS__RB1_FULL_DRAIN__SHIFT 0x5 +#define IH_RB_STATUS__RB1_OVERFLOW__SHIFT 0x6 +#define IH_RB_STATUS__RB_FULL_MASK 0x00000001L +#define IH_RB_STATUS__RB_FULL_DRAIN_MASK 0x00000002L +#define IH_RB_STATUS__RB_OVERFLOW_MASK 0x00000004L +#define IH_RB_STATUS__RB1_FULL_MASK 0x00000010L +#define IH_RB_STATUS__RB1_FULL_DRAIN_MASK 0x00000020L +#define IH_RB_STATUS__RB1_OVERFLOW_MASK 0x00000040L +//IH_INT_FLOOD_CNTL +#define IH_INT_FLOOD_CNTL__HIGHWATER__SHIFT 0x0 +#define IH_INT_FLOOD_CNTL__FLOOD_CNTL_ENABLE__SHIFT 0x3 +#define IH_INT_FLOOD_CNTL__CLEAR_INT_FLOOD_STATUS__SHIFT 0x4 +#define IH_INT_FLOOD_CNTL__HIGHWATER_MASK 0x00000007L +#define IH_INT_FLOOD_CNTL__FLOOD_CNTL_ENABLE_MASK 0x00000008L +#define IH_INT_FLOOD_CNTL__CLEAR_INT_FLOOD_STATUS_MASK 0x00000010L +//IH_RB0_INT_FLOOD_STATUS +#define IH_RB0_INT_FLOOD_STATUS__RB_INT_DROPPED_VF__SHIFT 0x0 +#define IH_RB0_INT_FLOOD_STATUS__RB_INT_DROPPED__SHIFT 0x1f +#define IH_RB0_INT_FLOOD_STATUS__RB_INT_DROPPED_VF_MASK 0x0000FFFFL +#define IH_RB0_INT_FLOOD_STATUS__RB_INT_DROPPED_MASK 0x80000000L +//IH_RB1_INT_FLOOD_STATUS +#define IH_RB1_INT_FLOOD_STATUS__RB_INT_DROPPED_VF__SHIFT 0x0 +#define IH_RB1_INT_FLOOD_STATUS__RB_INT_DROPPED__SHIFT 0x1f +#define IH_RB1_INT_FLOOD_STATUS__RB_INT_DROPPED_VF_MASK 0x0000FFFFL +#define IH_RB1_INT_FLOOD_STATUS__RB_INT_DROPPED_MASK 0x80000000L +//IH_INT_FLOOD_STATUS +#define IH_INT_FLOOD_STATUS__INT_DROP_CNT__SHIFT 0x0 +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_CLIENT_ID__SHIFT 0x8 +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_SOURCE_ID__SHIFT 0x10 +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_VF_ID__SHIFT 0x18 +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_VF__SHIFT 0x1d +#define IH_INT_FLOOD_STATUS__INT_DROPPED__SHIFT 0x1e +#define IH_INT_FLOOD_STATUS__INT_DROP_CNT_MASK 0x000000FFL +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_CLIENT_ID_MASK 0x0000FF00L +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_SOURCE_ID_MASK 0x00FF0000L +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_VF_ID_MASK 0x0F000000L +#define IH_INT_FLOOD_STATUS__FIRST_DROP_INT_VF_MASK 0x20000000L +#define IH_INT_FLOOD_STATUS__INT_DROPPED_MASK 0x40000000L +//IH_INT_FLAGS +#define IH_INT_FLAGS__CLIENT_0_FLAG__SHIFT 0x0 +#define IH_INT_FLAGS__CLIENT_1_FLAG__SHIFT 0x1 +#define IH_INT_FLAGS__CLIENT_2_FLAG__SHIFT 0x2 +#define IH_INT_FLAGS__CLIENT_3_FLAG__SHIFT 0x3 +#define IH_INT_FLAGS__CLIENT_4_FLAG__SHIFT 0x4 +#define IH_INT_FLAGS__CLIENT_5_FLAG__SHIFT 0x5 +#define IH_INT_FLAGS__CLIENT_6_FLAG__SHIFT 0x6 +#define IH_INT_FLAGS__CLIENT_7_FLAG__SHIFT 0x7 +#define IH_INT_FLAGS__CLIENT_8_FLAG__SHIFT 0x8 +#define IH_INT_FLAGS__CLIENT_9_FLAG__SHIFT 0x9 +#define IH_INT_FLAGS__CLIENT_10_FLAG__SHIFT 0xa +#define IH_INT_FLAGS__CLIENT_11_FLAG__SHIFT 0xb +#define IH_INT_FLAGS__CLIENT_12_FLAG__SHIFT 0xc +#define IH_INT_FLAGS__CLIENT_13_FLAG__SHIFT 0xd +#define IH_INT_FLAGS__CLIENT_14_FLAG__SHIFT 0xe +#define IH_INT_FLAGS__CLIENT_15_FLAG__SHIFT 0xf +#define IH_INT_FLAGS__CLIENT_16_FLAG__SHIFT 0x10 +#define IH_INT_FLAGS__CLIENT_17_FLAG__SHIFT 0x11 +#define IH_INT_FLAGS__CLIENT_18_FLAG__SHIFT 0x12 +#define IH_INT_FLAGS__CLIENT_19_FLAG__SHIFT 0x13 +#define IH_INT_FLAGS__CLIENT_20_FLAG__SHIFT 0x14 +#define IH_INT_FLAGS__CLIENT_21_FLAG__SHIFT 0x15 +#define IH_INT_FLAGS__CLIENT_22_FLAG__SHIFT 0x16 +#define IH_INT_FLAGS__CLIENT_23_FLAG__SHIFT 0x17 +#define IH_INT_FLAGS__CLIENT_24_FLAG__SHIFT 0x18 +#define IH_INT_FLAGS__CLIENT_25_FLAG__SHIFT 0x19 +#define IH_INT_FLAGS__CLIENT_26_FLAG__SHIFT 0x1a +#define IH_INT_FLAGS__CLIENT_27_FLAG__SHIFT 0x1b +#define IH_INT_FLAGS__CLIENT_28_FLAG__SHIFT 0x1c +#define IH_INT_FLAGS__CLIENT_29_FLAG__SHIFT 0x1d +#define IH_INT_FLAGS__CLIENT_30_FLAG__SHIFT 0x1e +#define IH_INT_FLAGS__CLIENT_31_FLAG__SHIFT 0x1f +#define IH_INT_FLAGS__CLIENT_0_FLAG_MASK 0x00000001L +#define IH_INT_FLAGS__CLIENT_1_FLAG_MASK 0x00000002L +#define IH_INT_FLAGS__CLIENT_2_FLAG_MASK 0x00000004L +#define IH_INT_FLAGS__CLIENT_3_FLAG_MASK 0x00000008L +#define IH_INT_FLAGS__CLIENT_4_FLAG_MASK 0x00000010L +#define IH_INT_FLAGS__CLIENT_5_FLAG_MASK 0x00000020L +#define IH_INT_FLAGS__CLIENT_6_FLAG_MASK 0x00000040L +#define IH_INT_FLAGS__CLIENT_7_FLAG_MASK 0x00000080L +#define IH_INT_FLAGS__CLIENT_8_FLAG_MASK 0x00000100L +#define IH_INT_FLAGS__CLIENT_9_FLAG_MASK 0x00000200L +#define IH_INT_FLAGS__CLIENT_10_FLAG_MASK 0x00000400L +#define IH_INT_FLAGS__CLIENT_11_FLAG_MASK 0x00000800L +#define IH_INT_FLAGS__CLIENT_12_FLAG_MASK 0x00001000L +#define IH_INT_FLAGS__CLIENT_13_FLAG_MASK 0x00002000L +#define IH_INT_FLAGS__CLIENT_14_FLAG_MASK 0x00004000L +#define IH_INT_FLAGS__CLIENT_15_FLAG_MASK 0x00008000L +#define IH_INT_FLAGS__CLIENT_16_FLAG_MASK 0x00010000L +#define IH_INT_FLAGS__CLIENT_17_FLAG_MASK 0x00020000L +#define IH_INT_FLAGS__CLIENT_18_FLAG_MASK 0x00040000L +#define IH_INT_FLAGS__CLIENT_19_FLAG_MASK 0x00080000L +#define IH_INT_FLAGS__CLIENT_20_FLAG_MASK 0x00100000L +#define IH_INT_FLAGS__CLIENT_21_FLAG_MASK 0x00200000L +#define IH_INT_FLAGS__CLIENT_22_FLAG_MASK 0x00400000L +#define IH_INT_FLAGS__CLIENT_23_FLAG_MASK 0x00800000L +#define IH_INT_FLAGS__CLIENT_24_FLAG_MASK 0x01000000L +#define IH_INT_FLAGS__CLIENT_25_FLAG_MASK 0x02000000L +#define IH_INT_FLAGS__CLIENT_26_FLAG_MASK 0x04000000L +#define IH_INT_FLAGS__CLIENT_27_FLAG_MASK 0x08000000L +#define IH_INT_FLAGS__CLIENT_28_FLAG_MASK 0x10000000L +#define IH_INT_FLAGS__CLIENT_29_FLAG_MASK 0x20000000L +#define IH_INT_FLAGS__CLIENT_30_FLAG_MASK 0x40000000L +#define IH_INT_FLAGS__CLIENT_31_FLAG_MASK 0x80000000L +//IH_LAST_INT_INFO0 +#define IH_LAST_INT_INFO0__CLIENT_ID__SHIFT 0x0 +#define IH_LAST_INT_INFO0__SOURCE_ID__SHIFT 0x8 +#define IH_LAST_INT_INFO0__RING_ID__SHIFT 0x10 +#define IH_LAST_INT_INFO0__VM_ID__SHIFT 0x18 +#define IH_LAST_INT_INFO0__VMID_TYPE__SHIFT 0x1f +#define IH_LAST_INT_INFO0__CLIENT_ID_MASK 0x000000FFL +#define IH_LAST_INT_INFO0__SOURCE_ID_MASK 0x0000FF00L +#define IH_LAST_INT_INFO0__RING_ID_MASK 0x00FF0000L +#define IH_LAST_INT_INFO0__VM_ID_MASK 0x0F000000L +#define IH_LAST_INT_INFO0__VMID_TYPE_MASK 0x80000000L +//IH_LAST_INT_INFO1 +#define IH_LAST_INT_INFO1__CONTEXT_ID__SHIFT 0x0 +#define IH_LAST_INT_INFO1__CONTEXT_ID_MASK 0xFFFFFFFFL +//IH_LAST_INT_INFO2 +#define IH_LAST_INT_INFO2__PAS_ID__SHIFT 0x0 +#define IH_LAST_INT_INFO2__VF_ID__SHIFT 0x10 +#define IH_LAST_INT_INFO2__VF__SHIFT 0x17 +#define IH_LAST_INT_INFO2__PAS_ID_MASK 0x0000FFFFL +#define IH_LAST_INT_INFO2__VF_ID_MASK 0x000F0000L +#define IH_LAST_INT_INFO2__VF_MASK 0x00800000L +//IH_SCRATCH +#define IH_SCRATCH__DATA__SHIFT 0x0 +#define IH_SCRATCH__DATA_MASK 0xFFFFFFFFL +//IH_CLIENT_CREDIT_ERROR +#define IH_CLIENT_CREDIT_ERROR__CLEAR__SHIFT 0x0 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_1_ERROR__SHIFT 0x1 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_2_ERROR__SHIFT 0x2 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_3_ERROR__SHIFT 0x3 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_4_ERROR__SHIFT 0x4 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_5_ERROR__SHIFT 0x5 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_6_ERROR__SHIFT 0x6 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_7_ERROR__SHIFT 0x7 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_8_ERROR__SHIFT 0x8 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_9_ERROR__SHIFT 0x9 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_10_ERROR__SHIFT 0xa +#define IH_CLIENT_CREDIT_ERROR__CLIENT_11_ERROR__SHIFT 0xb +#define IH_CLIENT_CREDIT_ERROR__CLIENT_12_ERROR__SHIFT 0xc +#define IH_CLIENT_CREDIT_ERROR__CLIENT_13_ERROR__SHIFT 0xd +#define IH_CLIENT_CREDIT_ERROR__CLIENT_14_ERROR__SHIFT 0xe +#define IH_CLIENT_CREDIT_ERROR__CLIENT_15_ERROR__SHIFT 0xf +#define IH_CLIENT_CREDIT_ERROR__CLIENT_16_ERROR__SHIFT 0x10 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_17_ERROR__SHIFT 0x11 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_18_ERROR__SHIFT 0x12 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_19_ERROR__SHIFT 0x13 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_20_ERROR__SHIFT 0x14 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_21_ERROR__SHIFT 0x15 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_22_ERROR__SHIFT 0x16 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_23_ERROR__SHIFT 0x17 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_24_ERROR__SHIFT 0x18 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_25_ERROR__SHIFT 0x19 +#define IH_CLIENT_CREDIT_ERROR__CLIENT_26_ERROR__SHIFT 0x1a +#define IH_CLIENT_CREDIT_ERROR__CLIENT_27_ERROR__SHIFT 0x1b +#define IH_CLIENT_CREDIT_ERROR__CLIENT_28_ERROR__SHIFT 0x1c +#define IH_CLIENT_CREDIT_ERROR__CLIENT_29_ERROR__SHIFT 0x1d +#define IH_CLIENT_CREDIT_ERROR__CLIENT_30_ERROR__SHIFT 0x1e +#define IH_CLIENT_CREDIT_ERROR__CLIENT_31_ERROR__SHIFT 0x1f +#define IH_CLIENT_CREDIT_ERROR__CLEAR_MASK 0x00000001L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_1_ERROR_MASK 0x00000002L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_2_ERROR_MASK 0x00000004L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_3_ERROR_MASK 0x00000008L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_4_ERROR_MASK 0x00000010L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_5_ERROR_MASK 0x00000020L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_6_ERROR_MASK 0x00000040L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_7_ERROR_MASK 0x00000080L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_8_ERROR_MASK 0x00000100L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_9_ERROR_MASK 0x00000200L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_10_ERROR_MASK 0x00000400L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_11_ERROR_MASK 0x00000800L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_12_ERROR_MASK 0x00001000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_13_ERROR_MASK 0x00002000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_14_ERROR_MASK 0x00004000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_15_ERROR_MASK 0x00008000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_16_ERROR_MASK 0x00010000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_17_ERROR_MASK 0x00020000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_18_ERROR_MASK 0x00040000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_19_ERROR_MASK 0x00080000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_20_ERROR_MASK 0x00100000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_21_ERROR_MASK 0x00200000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_22_ERROR_MASK 0x00400000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_23_ERROR_MASK 0x00800000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_24_ERROR_MASK 0x01000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_25_ERROR_MASK 0x02000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_26_ERROR_MASK 0x04000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_27_ERROR_MASK 0x08000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_28_ERROR_MASK 0x10000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_29_ERROR_MASK 0x20000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_30_ERROR_MASK 0x40000000L +#define IH_CLIENT_CREDIT_ERROR__CLIENT_31_ERROR_MASK 0x80000000L +//IH_GPU_IOV_VIOLATION_LOG +#define IH_GPU_IOV_VIOLATION_LOG__VIOLATION_STATUS__SHIFT 0x0 +#define IH_GPU_IOV_VIOLATION_LOG__MULTIPLE_VIOLATION_STATUS__SHIFT 0x1 +#define IH_GPU_IOV_VIOLATION_LOG__ADDRESS__SHIFT 0x2 +#define IH_GPU_IOV_VIOLATION_LOG__OPCODE__SHIFT 0x16 +#define IH_GPU_IOV_VIOLATION_LOG__VF__SHIFT 0x17 +#define IH_GPU_IOV_VIOLATION_LOG__VF_ID__SHIFT 0x18 +#define IH_GPU_IOV_VIOLATION_LOG__VIOLATION_STATUS_MASK 0x00000001L +#define IH_GPU_IOV_VIOLATION_LOG__MULTIPLE_VIOLATION_STATUS_MASK 0x00000002L +#define IH_GPU_IOV_VIOLATION_LOG__ADDRESS_MASK 0x000FFFFCL +#define IH_GPU_IOV_VIOLATION_LOG__OPCODE_MASK 0x00400000L +#define IH_GPU_IOV_VIOLATION_LOG__VF_MASK 0x00800000L +#define IH_GPU_IOV_VIOLATION_LOG__VF_ID_MASK 0x0F000000L +//IH_GPU_IOV_VIOLATION_LOG2 +#define IH_GPU_IOV_VIOLATION_LOG2__INITIATOR_ID__SHIFT 0x0 +#define IH_GPU_IOV_VIOLATION_LOG2__INITIATOR_ID_MASK 0x000003FFL +//IH_COOKIE_REC_VIOLATION_LOG +#define IH_COOKIE_REC_VIOLATION_LOG__VIOLATION_STATUS__SHIFT 0x0 +#define IH_COOKIE_REC_VIOLATION_LOG__CLIENT_ID__SHIFT 0x8 +#define IH_COOKIE_REC_VIOLATION_LOG__INITIATOR_ID__SHIFT 0x10 +#define IH_COOKIE_REC_VIOLATION_LOG__VIOLATION_STATUS_MASK 0x00000001L +#define IH_COOKIE_REC_VIOLATION_LOG__CLIENT_ID_MASK 0x0000FF00L +#define IH_COOKIE_REC_VIOLATION_LOG__INITIATOR_ID_MASK 0x03FF0000L +//IH_CREDIT_STATUS +#define IH_CREDIT_STATUS__CLIENT_1_CREDIT_RETURNED__SHIFT 0x1 +#define IH_CREDIT_STATUS__CLIENT_2_CREDIT_RETURNED__SHIFT 0x2 +#define IH_CREDIT_STATUS__CLIENT_3_CREDIT_RETURNED__SHIFT 0x3 +#define IH_CREDIT_STATUS__CLIENT_4_CREDIT_RETURNED__SHIFT 0x4 +#define IH_CREDIT_STATUS__CLIENT_5_CREDIT_RETURNED__SHIFT 0x5 +#define IH_CREDIT_STATUS__CLIENT_6_CREDIT_RETURNED__SHIFT 0x6 +#define IH_CREDIT_STATUS__CLIENT_7_CREDIT_RETURNED__SHIFT 0x7 +#define IH_CREDIT_STATUS__CLIENT_8_CREDIT_RETURNED__SHIFT 0x8 +#define IH_CREDIT_STATUS__CLIENT_9_CREDIT_RETURNED__SHIFT 0x9 +#define IH_CREDIT_STATUS__CLIENT_10_CREDIT_RETURNED__SHIFT 0xa +#define IH_CREDIT_STATUS__CLIENT_11_CREDIT_RETURNED__SHIFT 0xb +#define IH_CREDIT_STATUS__CLIENT_12_CREDIT_RETURNED__SHIFT 0xc +#define IH_CREDIT_STATUS__CLIENT_13_CREDIT_RETURNED__SHIFT 0xd +#define IH_CREDIT_STATUS__CLIENT_14_CREDIT_RETURNED__SHIFT 0xe +#define IH_CREDIT_STATUS__CLIENT_15_CREDIT_RETURNED__SHIFT 0xf +#define IH_CREDIT_STATUS__CLIENT_16_CREDIT_RETURNED__SHIFT 0x10 +#define IH_CREDIT_STATUS__CLIENT_17_CREDIT_RETURNED__SHIFT 0x11 +#define IH_CREDIT_STATUS__CLIENT_18_CREDIT_RETURNED__SHIFT 0x12 +#define IH_CREDIT_STATUS__CLIENT_19_CREDIT_RETURNED__SHIFT 0x13 +#define IH_CREDIT_STATUS__CLIENT_20_CREDIT_RETURNED__SHIFT 0x14 +#define IH_CREDIT_STATUS__CLIENT_21_CREDIT_RETURNED__SHIFT 0x15 +#define IH_CREDIT_STATUS__CLIENT_22_CREDIT_RETURNED__SHIFT 0x16 +#define IH_CREDIT_STATUS__CLIENT_23_CREDIT_RETURNED__SHIFT 0x17 +#define IH_CREDIT_STATUS__CLIENT_24_CREDIT_RETURNED__SHIFT 0x18 +#define IH_CREDIT_STATUS__CLIENT_25_CREDIT_RETURNED__SHIFT 0x19 +#define IH_CREDIT_STATUS__CLIENT_26_CREDIT_RETURNED__SHIFT 0x1a +#define IH_CREDIT_STATUS__CLIENT_27_CREDIT_RETURNED__SHIFT 0x1b +#define IH_CREDIT_STATUS__CLIENT_28_CREDIT_RETURNED__SHIFT 0x1c +#define IH_CREDIT_STATUS__CLIENT_29_CREDIT_RETURNED__SHIFT 0x1d +#define IH_CREDIT_STATUS__CLIENT_30_CREDIT_RETURNED__SHIFT 0x1e +#define IH_CREDIT_STATUS__CLIENT_31_CREDIT_RETURNED__SHIFT 0x1f +#define IH_CREDIT_STATUS__CLIENT_1_CREDIT_RETURNED_MASK 0x00000002L +#define IH_CREDIT_STATUS__CLIENT_2_CREDIT_RETURNED_MASK 0x00000004L +#define IH_CREDIT_STATUS__CLIENT_3_CREDIT_RETURNED_MASK 0x00000008L +#define IH_CREDIT_STATUS__CLIENT_4_CREDIT_RETURNED_MASK 0x00000010L +#define IH_CREDIT_STATUS__CLIENT_5_CREDIT_RETURNED_MASK 0x00000020L +#define IH_CREDIT_STATUS__CLIENT_6_CREDIT_RETURNED_MASK 0x00000040L +#define IH_CREDIT_STATUS__CLIENT_7_CREDIT_RETURNED_MASK 0x00000080L +#define IH_CREDIT_STATUS__CLIENT_8_CREDIT_RETURNED_MASK 0x00000100L +#define IH_CREDIT_STATUS__CLIENT_9_CREDIT_RETURNED_MASK 0x00000200L +#define IH_CREDIT_STATUS__CLIENT_10_CREDIT_RETURNED_MASK 0x00000400L +#define IH_CREDIT_STATUS__CLIENT_11_CREDIT_RETURNED_MASK 0x00000800L +#define IH_CREDIT_STATUS__CLIENT_12_CREDIT_RETURNED_MASK 0x00001000L +#define IH_CREDIT_STATUS__CLIENT_13_CREDIT_RETURNED_MASK 0x00002000L +#define IH_CREDIT_STATUS__CLIENT_14_CREDIT_RETURNED_MASK 0x00004000L +#define IH_CREDIT_STATUS__CLIENT_15_CREDIT_RETURNED_MASK 0x00008000L +#define IH_CREDIT_STATUS__CLIENT_16_CREDIT_RETURNED_MASK 0x00010000L +#define IH_CREDIT_STATUS__CLIENT_17_CREDIT_RETURNED_MASK 0x00020000L +#define IH_CREDIT_STATUS__CLIENT_18_CREDIT_RETURNED_MASK 0x00040000L +#define IH_CREDIT_STATUS__CLIENT_19_CREDIT_RETURNED_MASK 0x00080000L +#define IH_CREDIT_STATUS__CLIENT_20_CREDIT_RETURNED_MASK 0x00100000L +#define IH_CREDIT_STATUS__CLIENT_21_CREDIT_RETURNED_MASK 0x00200000L +#define IH_CREDIT_STATUS__CLIENT_22_CREDIT_RETURNED_MASK 0x00400000L +#define IH_CREDIT_STATUS__CLIENT_23_CREDIT_RETURNED_MASK 0x00800000L +#define IH_CREDIT_STATUS__CLIENT_24_CREDIT_RETURNED_MASK 0x01000000L +#define IH_CREDIT_STATUS__CLIENT_25_CREDIT_RETURNED_MASK 0x02000000L +#define IH_CREDIT_STATUS__CLIENT_26_CREDIT_RETURNED_MASK 0x04000000L +#define IH_CREDIT_STATUS__CLIENT_27_CREDIT_RETURNED_MASK 0x08000000L +#define IH_CREDIT_STATUS__CLIENT_28_CREDIT_RETURNED_MASK 0x10000000L +#define IH_CREDIT_STATUS__CLIENT_29_CREDIT_RETURNED_MASK 0x20000000L +#define IH_CREDIT_STATUS__CLIENT_30_CREDIT_RETURNED_MASK 0x40000000L +#define IH_CREDIT_STATUS__CLIENT_31_CREDIT_RETURNED_MASK 0x80000000L +//IH_MMHUB_ERROR +#define IH_MMHUB_ERROR__IH_BRESP_01__SHIFT 0x1 +#define IH_MMHUB_ERROR__IH_BRESP_10__SHIFT 0x2 +#define IH_MMHUB_ERROR__IH_BRESP_11__SHIFT 0x3 +#define IH_MMHUB_ERROR__IH_BUSER_NACK_01__SHIFT 0x5 +#define IH_MMHUB_ERROR__IH_BUSER_NACK_10__SHIFT 0x6 +#define IH_MMHUB_ERROR__IH_BUSER_NACK_11__SHIFT 0x7 +#define IH_MMHUB_ERROR__IH_BRESP_01_MASK 0x00000002L +#define IH_MMHUB_ERROR__IH_BRESP_10_MASK 0x00000004L +#define IH_MMHUB_ERROR__IH_BRESP_11_MASK 0x00000008L +#define IH_MMHUB_ERROR__IH_BUSER_NACK_01_MASK 0x00000020L +#define IH_MMHUB_ERROR__IH_BUSER_NACK_10_MASK 0x00000040L +#define IH_MMHUB_ERROR__IH_BUSER_NACK_11_MASK 0x00000080L +//IH_VF_RB_STATUS3 +#define IH_VF_RB_STATUS3__RB_OVERFLOW_VF__SHIFT 0x0 +#define IH_VF_RB_STATUS3__RB_OVERFLOW_VF_MASK 0x0000FFFFL +//IH_VF_RB_STATUS4 +#define IH_VF_RB_STATUS4__BIF_INTERRUPT_LINE_VF__SHIFT 0x0 +#define IH_VF_RB_STATUS4__BIF_INTERRUPT_LINE_VF_MASK 0x0000FFFFL +//IH_VF_RB1_STATUS3 +#define IH_VF_RB1_STATUS3__RB_OVERFLOW_VF__SHIFT 0x0 +#define IH_VF_RB1_STATUS3__RB_OVERFLOW_VF_MASK 0x0000FFFFL +//IH_MSI_STORM_CTRL +#define IH_MSI_STORM_CTRL__DELAY__SHIFT 0x0 +#define IH_MSI_STORM_CTRL__DELAY_MASK 0x00000FFFL +//IH_MSI_STORM_CLIENT_INDEX +#define IH_MSI_STORM_CLIENT_INDEX__INDEX__SHIFT 0x0 +#define IH_MSI_STORM_CLIENT_INDEX__INDEX_MASK 0x00000007L +//IH_MSI_STORM_CLIENT_DATA +#define IH_MSI_STORM_CLIENT_DATA__CLIENT_ID__SHIFT 0x0 +#define IH_MSI_STORM_CLIENT_DATA__SOURCE_ID__SHIFT 0x8 +#define IH_MSI_STORM_CLIENT_DATA__SOURCE_ID_MATCH_ENABLE__SHIFT 0x10 +#define IH_MSI_STORM_CLIENT_DATA__UTCL2_PAGE_FAULT_MATCH_ENABLE__SHIFT 0x11 +#define IH_MSI_STORM_CLIENT_DATA__ENTRY_VALID__SHIFT 0x1f +#define IH_MSI_STORM_CLIENT_DATA__CLIENT_ID_MASK 0x000000FFL +#define IH_MSI_STORM_CLIENT_DATA__SOURCE_ID_MASK 0x0000FF00L +#define IH_MSI_STORM_CLIENT_DATA__SOURCE_ID_MATCH_ENABLE_MASK 0x00010000L +#define IH_MSI_STORM_CLIENT_DATA__UTCL2_PAGE_FAULT_MATCH_ENABLE_MASK 0x00020000L +#define IH_MSI_STORM_CLIENT_DATA__ENTRY_VALID_MASK 0x80000000L +//IH_REGISTER_LAST_PART2 +#define IH_REGISTER_LAST_PART2__RESERVED__SHIFT 0x0 +#define IH_REGISTER_LAST_PART2__RESERVED_MASK 0xFFFFFFFFL +//SEM_MAILBOX +#define SEM_MAILBOX__HOSTPORT__SHIFT 0x0 +#define SEM_MAILBOX__RESERVED__SHIFT 0x10 +#define SEM_MAILBOX__HOSTPORT_MASK 0x0000FFFFL +#define SEM_MAILBOX__RESERVED_MASK 0xFFFF0000L +//SEM_MAILBOX_CLEAR +#define SEM_MAILBOX_CLEAR__CLEAR__SHIFT 0x0 +#define SEM_MAILBOX_CLEAR__RESERVED__SHIFT 0x10 +#define SEM_MAILBOX_CLEAR__CLEAR_MASK 0x0000FFFFL +#define SEM_MAILBOX_CLEAR__RESERVED_MASK 0xFFFF0000L +//SEM_REGISTER_LAST_PART2 +#define SEM_REGISTER_LAST_PART2__RESERVED__SHIFT 0x0 +#define SEM_REGISTER_LAST_PART2__RESERVED_MASK 0xFFFFFFFFL +//IH_ACTIVE_FCN_ID +#define IH_ACTIVE_FCN_ID__VF_ID__SHIFT 0x0 +#define IH_ACTIVE_FCN_ID__RESERVED__SHIFT 0x4 +#define IH_ACTIVE_FCN_ID__PF_VF__SHIFT 0x1f +#define IH_ACTIVE_FCN_ID__VF_ID_MASK 0x0000000FL +#define IH_ACTIVE_FCN_ID__RESERVED_MASK 0x7FFFFFF0L +#define IH_ACTIVE_FCN_ID__PF_VF_MASK 0x80000000L +//IH_VIRT_RESET_REQ +#define IH_VIRT_RESET_REQ__VF__SHIFT 0x0 +#define IH_VIRT_RESET_REQ__PF__SHIFT 0x1f +#define IH_VIRT_RESET_REQ__VF_MASK 0x0000FFFFL +#define IH_VIRT_RESET_REQ__PF_MASK 0x80000000L +//IH_CLIENT_CFG +#define IH_CLIENT_CFG__TOTAL_CLIENT_NUM__SHIFT 0x0 +#define IH_CLIENT_CFG__TOTAL_CLIENT_NUM_MASK 0x0000003FL +//IH_RING1_CLIENT_CFG_INDEX +#define IH_RING1_CLIENT_CFG_INDEX__INDEX__SHIFT 0x0 +#define IH_RING1_CLIENT_CFG_INDEX__INDEX_MASK 0x00000007L +//IH_RING1_CLIENT_CFG_DATA +#define IH_RING1_CLIENT_CFG_DATA__CLIENT_ID__SHIFT 0x0 +#define IH_RING1_CLIENT_CFG_DATA__SOURCE_ID__SHIFT 0x8 +#define IH_RING1_CLIENT_CFG_DATA__SOURCE_ID_MATCH_ENABLE__SHIFT 0x10 +#define IH_RING1_CLIENT_CFG_DATA__CLIENT_ID_MASK 0x000000FFL +#define IH_RING1_CLIENT_CFG_DATA__SOURCE_ID_MASK 0x0000FF00L +#define IH_RING1_CLIENT_CFG_DATA__SOURCE_ID_MATCH_ENABLE_MASK 0x00010000L +//IH_CLIENT_CFG_INDEX +#define IH_CLIENT_CFG_INDEX__INDEX__SHIFT 0x0 +#define IH_CLIENT_CFG_INDEX__INDEX_MASK 0x0000001FL +//IH_CLIENT_CFG_DATA +#define IH_CLIENT_CFG_DATA__CLIENT_TYPE__SHIFT 0x12 +#define IH_CLIENT_CFG_DATA__VF_RB_SELECT__SHIFT 0x16 +#define IH_CLIENT_CFG_DATA__OVERWRITE_RING_ID_WITH_ACTIVE_FCN_ID__SHIFT 0x18 +#define IH_CLIENT_CFG_DATA__INTERFACE_TYPE__SHIFT 0x19 +#define IH_CLIENT_CFG_DATA__CLIENT_TYPE_MASK 0x000C0000L +#define IH_CLIENT_CFG_DATA__VF_RB_SELECT_MASK 0x00C00000L +#define IH_CLIENT_CFG_DATA__OVERWRITE_RING_ID_WITH_ACTIVE_FCN_ID_MASK 0x01000000L +#define IH_CLIENT_CFG_DATA__INTERFACE_TYPE_MASK 0x02000000L +//IH_CLIENT_CFG_DATA2 +#define IH_CLIENT_CFG_DATA2__CREDIT_RETURN_ADDR__SHIFT 0x0 +#define IH_CLIENT_CFG_DATA2__CREDIT_RETURN_ADDR_MASK 0xFFFFFFFFL +//IH_CID_REMAP_INDEX +#define IH_CID_REMAP_INDEX__INDEX__SHIFT 0x0 +#define IH_CID_REMAP_INDEX__INDEX_MASK 0x00000003L +//IH_CID_REMAP_DATA +#define IH_CID_REMAP_DATA__CLIENT_ID__SHIFT 0x0 +#define IH_CID_REMAP_DATA__INITIATOR_ID__SHIFT 0x8 +#define IH_CID_REMAP_DATA__CLIENT_ID_REMAP__SHIFT 0x18 +#define IH_CID_REMAP_DATA__CLIENT_ID_MASK 0x000000FFL +#define IH_CID_REMAP_DATA__INITIATOR_ID_MASK 0x0003FF00L +#define IH_CID_REMAP_DATA__CLIENT_ID_REMAP_MASK 0xFF000000L +//IH_CHICKEN +#define IH_CHICKEN__ACTIVE_FCN_ID_PROT_ENABLE__SHIFT 0x0 +#define IH_CHICKEN__DBGU_TRIGGER_ENABLE__SHIFT 0x1 +#define IH_CHICKEN__CROSS_TRIGGER_ENABLE__SHIFT 0x2 +#define IH_CHICKEN__MC_SPACE_FBPA_ENABLE__SHIFT 0x3 +#define IH_CHICKEN__MC_SPACE_GPA_ENABLE__SHIFT 0x4 +#define IH_CHICKEN__REG_FIREWALL_ENABLE__SHIFT 0x5 +#define IH_CHICKEN__ACTIVE_FCN_ID_PROT_ENABLE_MASK 0x00000001L +#define IH_CHICKEN__DBGU_TRIGGER_ENABLE_MASK 0x00000002L +#define IH_CHICKEN__CROSS_TRIGGER_ENABLE_MASK 0x00000004L +#define IH_CHICKEN__MC_SPACE_FBPA_ENABLE_MASK 0x00000008L +#define IH_CHICKEN__MC_SPACE_GPA_ENABLE_MASK 0x00000010L +#define IH_CHICKEN__REG_FIREWALL_ENABLE_MASK 0x00000020L +//IH_MMHUB_CNTL +#define IH_MMHUB_CNTL__UNITID__SHIFT 0x0 +#define IH_MMHUB_CNTL__IV_TLVL__SHIFT 0x8 +#define IH_MMHUB_CNTL__WPTR_WB_TLVL__SHIFT 0xc +#define IH_MMHUB_CNTL__UNITID_MASK 0x0000003FL +#define IH_MMHUB_CNTL__IV_TLVL_MASK 0x00000F00L +#define IH_MMHUB_CNTL__WPTR_WB_TLVL_MASK 0x0000F000L +//IH_INT_DROP_CNTL +#define IH_INT_DROP_CNTL__INT_DROP_EN__SHIFT 0x0 +#define IH_INT_DROP_CNTL__CLIENT_ID_MATCH_EN__SHIFT 0x1 +#define IH_INT_DROP_CNTL__SOURCE_ID_MATCH_EN__SHIFT 0x2 +#define IH_INT_DROP_CNTL__VF_ID_MATCH_EN__SHIFT 0x3 +#define IH_INT_DROP_CNTL__VF_MATCH_EN__SHIFT 0x4 +#define IH_INT_DROP_CNTL__CONTEXT_ID_MATCH_EN__SHIFT 0x5 +#define IH_INT_DROP_CNTL__INT_DROP_MODE__SHIFT 0x6 +#define IH_INT_DROP_CNTL__UTCL2_RETRY_INT_DROP_EN__SHIFT 0x8 +#define IH_INT_DROP_CNTL__INT_DROPPED__SHIFT 0x10 +#define IH_INT_DROP_CNTL__INT_DROP_EN_MASK 0x00000001L +#define IH_INT_DROP_CNTL__CLIENT_ID_MATCH_EN_MASK 0x00000002L +#define IH_INT_DROP_CNTL__SOURCE_ID_MATCH_EN_MASK 0x00000004L +#define IH_INT_DROP_CNTL__VF_ID_MATCH_EN_MASK 0x00000008L +#define IH_INT_DROP_CNTL__VF_MATCH_EN_MASK 0x00000010L +#define IH_INT_DROP_CNTL__CONTEXT_ID_MATCH_EN_MASK 0x00000020L +#define IH_INT_DROP_CNTL__INT_DROP_MODE_MASK 0x000000C0L +#define IH_INT_DROP_CNTL__UTCL2_RETRY_INT_DROP_EN_MASK 0x00000100L +#define IH_INT_DROP_CNTL__INT_DROPPED_MASK 0x00010000L +//IH_INT_DROP_MATCH_VALUE0 +#define IH_INT_DROP_MATCH_VALUE0__CLIENT_ID_MATCH_VALUE__SHIFT 0x0 +#define IH_INT_DROP_MATCH_VALUE0__SOURCE_ID_MATCH_VALUE__SHIFT 0x8 +#define IH_INT_DROP_MATCH_VALUE0__VF_ID_MATCH_VALUE__SHIFT 0x10 +#define IH_INT_DROP_MATCH_VALUE0__VF_MATCH_VALUE__SHIFT 0x17 +#define IH_INT_DROP_MATCH_VALUE0__CONTEXT_ID_39_32_MATCH_VALUE__SHIFT 0x18 +#define IH_INT_DROP_MATCH_VALUE0__CLIENT_ID_MATCH_VALUE_MASK 0x000000FFL +#define IH_INT_DROP_MATCH_VALUE0__SOURCE_ID_MATCH_VALUE_MASK 0x0000FF00L +#define IH_INT_DROP_MATCH_VALUE0__VF_ID_MATCH_VALUE_MASK 0x001F0000L +#define IH_INT_DROP_MATCH_VALUE0__VF_MATCH_VALUE_MASK 0x00800000L +#define IH_INT_DROP_MATCH_VALUE0__CONTEXT_ID_39_32_MATCH_VALUE_MASK 0xFF000000L +//IH_INT_DROP_MATCH_VALUE1 +#define IH_INT_DROP_MATCH_VALUE1__CONTEXT_ID_31_0_MATCH_VALUE__SHIFT 0x0 +#define IH_INT_DROP_MATCH_VALUE1__CONTEXT_ID_31_0_MATCH_VALUE_MASK 0xFFFFFFFFL +//IH_INT_DROP_MATCH_MASK0 +#define IH_INT_DROP_MATCH_MASK0__CLIENT_ID_MATCH_MASK__SHIFT 0x0 +#define IH_INT_DROP_MATCH_MASK0__SOURCE_ID_MATCH_MASK__SHIFT 0x8 +#define IH_INT_DROP_MATCH_MASK0__VF_ID_MATCH_MASK__SHIFT 0x10 +#define IH_INT_DROP_MATCH_MASK0__VF_MATCH_MASK__SHIFT 0x17 +#define IH_INT_DROP_MATCH_MASK0__CONTEXT_ID_39_32_MATCH_MASK__SHIFT 0x18 +#define IH_INT_DROP_MATCH_MASK0__CLIENT_ID_MATCH_MASK_MASK 0x000000FFL +#define IH_INT_DROP_MATCH_MASK0__SOURCE_ID_MATCH_MASK_MASK 0x0000FF00L +#define IH_INT_DROP_MATCH_MASK0__VF_ID_MATCH_MASK_MASK 0x001F0000L +#define IH_INT_DROP_MATCH_MASK0__VF_MATCH_MASK_MASK 0x00800000L +#define IH_INT_DROP_MATCH_MASK0__CONTEXT_ID_39_32_MATCH_MASK_MASK 0xFF000000L +//IH_INT_DROP_MATCH_MASK1 +#define IH_INT_DROP_MATCH_MASK1__CONTEXT_ID_31_0_MATCH_MASK__SHIFT 0x0 +#define IH_INT_DROP_MATCH_MASK1__CONTEXT_ID_31_0_MATCH_MASK_MASK 0xFFFFFFFFL +//IH_REGISTER_LAST_PART1 +#define IH_REGISTER_LAST_PART1__RESERVED__SHIFT 0x0 +#define IH_REGISTER_LAST_PART1__RESERVED_MASK 0xFFFFFFFFL + +#endif diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h index e68c1e280322..fa7d6ced786f 100644 --- a/drivers/gpu/drm/amd/include/atomfirmware.h +++ b/drivers/gpu/drm/amd/include/atomfirmware.h @@ -3117,6 +3117,24 @@ enum atom_umc_config1_def { UMC_CONFIG1__ENABLE_ECC_CAPABLE = 0x00010000, }; +struct atom_umc_info_v4_0 { + struct atom_common_table_header table_header; + uint32_t ucode_reserved[5]; + uint8_t umcip_min_ver; + uint8_t umcip_max_ver; + uint8_t vram_type; + uint8_t umc_config; + uint32_t mem_refclk_10khz; + uint32_t clk_reserved[4]; + uint32_t golden_reserved; + uint32_t umc_config1; + uint32_t reserved[2]; + uint8_t channel_num; + uint8_t channel_width; + uint8_t channel_reserve[2]; + uint8_t umc_info_reserved[16]; +}; + /* *************************************************************************** Data Table vram_info structure diff --git a/drivers/gpu/drm/amd/include/discovery.h b/drivers/gpu/drm/amd/include/discovery.h index f43e29722ef7..7a9d473d0917 100644 --- a/drivers/gpu/drm/amd/include/discovery.h +++ b/drivers/gpu/drm/amd/include/discovery.h @@ -30,7 +30,7 @@ #define GC_TABLE_ID 0x4347 #define HARVEST_TABLE_SIGNATURE 0x56524148 #define VCN_INFO_TABLE_ID 0x004E4356 -#define MALL_INFO_TABLE_ID 0x4D414C4C +#define MALL_INFO_TABLE_ID 0x4C4C414D typedef enum { @@ -280,6 +280,36 @@ struct gc_info_v2_0 { uint32_t gc_num_packer_per_sc; }; +struct gc_info_v2_1 { + struct gpu_info_header header; + + uint32_t gc_num_se; + uint32_t gc_num_cu_per_sh; + uint32_t gc_num_sh_per_se; + uint32_t gc_num_rb_per_se; + uint32_t gc_num_tccs; + uint32_t gc_num_gprs; + uint32_t gc_num_max_gs_thds; + uint32_t gc_gs_table_depth; + uint32_t gc_gsprim_buff_depth; + uint32_t gc_parameter_cache_depth; + uint32_t gc_double_offchip_lds_buffer; + uint32_t gc_wave_size; + uint32_t gc_max_waves_per_simd; + uint32_t gc_max_scratch_slots_per_cu; + uint32_t gc_lds_size; + uint32_t gc_num_sc_per_se; + uint32_t gc_num_packer_per_sc; + /* new for v2_1 */ + uint32_t gc_num_tcp_per_sh; + uint32_t gc_tcp_size_per_cu; + uint32_t gc_num_sdp_interface; + uint32_t gc_num_cu_per_sqc; + uint32_t gc_instruction_cache_size_per_sqc; + uint32_t gc_scalar_data_cache_size_per_sqc; + uint32_t gc_tcc_size; +}; + typedef struct harvest_info_header { uint32_t signature; /* Table Signature */ uint32_t version; /* Table Version */ @@ -312,6 +342,12 @@ struct mall_info_v1_0 { uint32_t reserved[5]; }; +struct mall_info_v2_0 { + struct mall_info_header header; + uint32_t mall_size_per_umc; + uint32_t reserved[8]; +}; + #define VCN_INFO_TABLE_MAX_NUM_INSTANCES 4 struct vcn_info_header { diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index d0df3381539f..3b5a56585c4b 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -31,12 +31,12 @@ #include <linux/types.h> #include <linux/bitmap.h> #include <linux/dma-fence.h> +#include "amdgpu_irq.h" +#include "amdgpu_gfx.h" struct pci_dev; struct amdgpu_device; -#define KGD_MAX_QUEUES 128 - struct kfd_dev; struct kgd_mem; @@ -68,7 +68,7 @@ struct kfd_cu_info { uint32_t wave_front_size; uint32_t max_scratch_slots_per_cu; uint32_t lds_size; - uint32_t cu_bitmap[4][4]; + uint32_t cu_bitmap[AMDGPU_MAX_GC_INSTANCES][4][4]; }; /* For getting GPU local memory information from KGD */ @@ -315,11 +315,13 @@ struct kfd2kgd_calls { uint32_t watch_address_mask, uint32_t watch_id, uint32_t watch_mode, - uint32_t debug_vmid); + uint32_t debug_vmid, + uint32_t inst); uint32_t (*clear_address_watch)(struct amdgpu_device *adev, uint32_t watch_id); void (*get_iq_wait_times)(struct amdgpu_device *adev, - uint32_t *wait_times); + uint32_t *wait_times, + uint32_t inst); void (*build_grace_period_packet_info)(struct amdgpu_device *adev, uint32_t wait_times, uint32_t grace_period, diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index 9f542f6e19ed..84c5224d994c 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -132,7 +132,8 @@ enum amd_pp_sensors { AMDGPU_PP_SENSOR_MEM_TEMP, AMDGPU_PP_SENSOR_VCE_POWER, AMDGPU_PP_SENSOR_UVD_POWER, - AMDGPU_PP_SENSOR_GPU_POWER, + AMDGPU_PP_SENSOR_GPU_AVG_POWER, + AMDGPU_PP_SENSOR_GPU_INPUT_POWER, AMDGPU_PP_SENSOR_SS_APU_SHARE, AMDGPU_PP_SENSOR_SS_DGPU_SHARE, AMDGPU_PP_SENSOR_STABLE_PSTATE_SCLK, @@ -892,4 +893,73 @@ struct gpu_metrics_v2_3 { uint16_t average_temperature_core[8]; // average CPU core temperature on APUs uint16_t average_temperature_l3[2]; }; + +struct gpu_metrics_v2_4 { + struct metrics_table_header common_header; + + /* Temperature (unit: centi-Celsius) */ + uint16_t temperature_gfx; + uint16_t temperature_soc; + uint16_t temperature_core[8]; + uint16_t temperature_l3[2]; + + /* Utilization (unit: centi) */ + uint16_t average_gfx_activity; + uint16_t average_mm_activity; + + /* Driver attached timestamp (in ns) */ + uint64_t system_clock_counter; + + /* Power/Energy (unit: mW) */ + uint16_t average_socket_power; + uint16_t average_cpu_power; + uint16_t average_soc_power; + uint16_t average_gfx_power; + uint16_t average_core_power[8]; + + /* Average clocks (unit: MHz) */ + uint16_t average_gfxclk_frequency; + uint16_t average_socclk_frequency; + uint16_t average_uclk_frequency; + uint16_t average_fclk_frequency; + uint16_t average_vclk_frequency; + uint16_t average_dclk_frequency; + + /* Current clocks (unit: MHz) */ + uint16_t current_gfxclk; + uint16_t current_socclk; + uint16_t current_uclk; + uint16_t current_fclk; + uint16_t current_vclk; + uint16_t current_dclk; + uint16_t current_coreclk[8]; + uint16_t current_l3clk[2]; + + /* Throttle status (ASIC dependent) */ + uint32_t throttle_status; + + /* Fans */ + uint16_t fan_pwm; + + uint16_t padding[3]; + + /* Throttle status (ASIC independent) */ + uint64_t indep_throttle_status; + + /* Average Temperature (unit: centi-Celsius) */ + uint16_t average_temperature_gfx; + uint16_t average_temperature_soc; + uint16_t average_temperature_core[8]; + uint16_t average_temperature_l3[2]; + + /* Power/Voltage (unit: mV) */ + uint16_t average_cpu_voltage; + uint16_t average_soc_voltage; + uint16_t average_gfx_voltage; + + /* Power/Current (unit: mA) */ + uint16_t average_cpu_current; + uint16_t average_soc_current; + uint16_t average_gfx_current; +}; #endif diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h index 0997e999416a..b1db2b190187 100644 --- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h +++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h @@ -275,7 +275,9 @@ union MESAPI__ADD_QUEUE { uint32_t trap_en : 1; uint32_t is_aql_queue : 1; uint32_t skip_process_ctx_clear : 1; - uint32_t reserved : 19; + uint32_t map_legacy_kq : 1; + uint32_t exclusively_scheduled : 1; + uint32_t reserved : 17; }; struct MES_API_STATUS api_status; uint64_t tma_addr; diff --git a/drivers/gpu/drm/amd/include/yellow_carp_offset.h b/drivers/gpu/drm/amd/include/yellow_carp_offset.h index 0fea6a746611..a2c8dca2425e 100644 --- a/drivers/gpu/drm/amd/include/yellow_carp_offset.h +++ b/drivers/gpu/drm/amd/include/yellow_carp_offset.h @@ -7,13 +7,11 @@ #define MAX_SEGMENT 6 -struct IP_BASE_INSTANCE -{ +struct IP_BASE_INSTANCE { unsigned int segment[MAX_SEGMENT]; } __maybe_unused; -struct IP_BASE -{ +struct IP_BASE { struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 9ef88a0b1b57..41147da54458 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -743,7 +743,7 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev, type = PP_OD_EDIT_CCLK_VDDC_TABLE; else if (*buf == 'm') type = PP_OD_EDIT_MCLK_VDDC_TABLE; - else if(*buf == 'r') + else if (*buf == 'r') type = PP_OD_RESTORE_DEFAULT_TABLE; else if (*buf == 'c') type = PP_OD_COMMIT_DPM_TABLE; @@ -1467,6 +1467,32 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev, return -EINVAL; } +static unsigned int amdgpu_hwmon_get_sensor_generic(struct amdgpu_device *adev, + enum amd_pp_sensors sensor, + void *query) +{ + int r, size = sizeof(uint32_t); + + if (amdgpu_in_reset(adev)) + return -EPERM; + if (adev->in_suspend && !adev->in_runpm) + return -EPERM; + + r = pm_runtime_get_sync(adev_to_drm(adev)->dev); + if (r < 0) { + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + return r; + } + + /* get the sensor value */ + r = amdgpu_dpm_read_sensor(adev, sensor, query, &size); + + pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); + pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + + return r; +} + /** * DOC: gpu_busy_percent * @@ -1481,26 +1507,10 @@ static ssize_t amdgpu_get_gpu_busy_percent(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - int r, value, size = sizeof(value); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(ddev->dev); - if (r < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return r; - } - - /* read the IP busy sensor */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_LOAD, - (void *)&value, &size); - - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); + unsigned int value; + int r; + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_LOAD, &value); if (r) return r; @@ -1521,26 +1531,10 @@ static ssize_t amdgpu_get_mem_busy_percent(struct device *dev, { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - int r, value, size = sizeof(value); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(ddev->dev); - if (r < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return r; - } - - /* read the IP busy sensor */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MEM_LOAD, - (void *)&value, &size); - - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); + unsigned int value; + int r; + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MEM_LOAD, &value); if (r) return r; @@ -1814,45 +1808,15 @@ out: return size; } -static int amdgpu_device_read_powershift(struct amdgpu_device *adev, - uint32_t *ss_power, bool dgpu_share) -{ - struct drm_device *ddev = adev_to_drm(adev); - uint32_t size; - int r = 0; - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(ddev->dev); - if (r < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return r; - } - - if (dgpu_share) - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_SS_DGPU_SHARE, - (void *)ss_power, &size); - else - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_SS_APU_SHARE, - (void *)ss_power, &size); - - pm_runtime_mark_last_busy(ddev->dev); - pm_runtime_put_autosuspend(ddev->dev); - return r; -} - static int amdgpu_show_powershift_percent(struct device *dev, - char *buf, bool dgpu_share) + char *buf, enum amd_pp_sensors sensor) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); uint32_t ss_power; int r = 0, i; - r = amdgpu_device_read_powershift(adev, &ss_power, dgpu_share); + r = amdgpu_hwmon_get_sensor_generic(adev, sensor, (void *)&ss_power); if (r == -EOPNOTSUPP) { /* sensor not available on dGPU, try to read from APU */ adev = NULL; @@ -1865,14 +1829,15 @@ static int amdgpu_show_powershift_percent(struct device *dev, } mutex_unlock(&mgpu_info.mutex); if (adev) - r = amdgpu_device_read_powershift(adev, &ss_power, dgpu_share); + r = amdgpu_hwmon_get_sensor_generic(adev, sensor, (void *)&ss_power); } - if (!r) - r = sysfs_emit(buf, "%u%%\n", ss_power); + if (r) + return r; - return r; + return sysfs_emit(buf, "%u%%\n", ss_power); } + /** * DOC: smartshift_apu_power * @@ -1886,7 +1851,7 @@ static int amdgpu_show_powershift_percent(struct device *dev, static ssize_t amdgpu_get_smartshift_apu_power(struct device *dev, struct device_attribute *attr, char *buf) { - return amdgpu_show_powershift_percent(dev, buf, false); + return amdgpu_show_powershift_percent(dev, buf, AMDGPU_PP_SENSOR_SS_APU_SHARE); } /** @@ -1902,7 +1867,7 @@ static ssize_t amdgpu_get_smartshift_apu_power(struct device *dev, struct device static ssize_t amdgpu_get_smartshift_dgpu_power(struct device *dev, struct device_attribute *attr, char *buf) { - return amdgpu_show_powershift_percent(dev, buf, true); + return amdgpu_show_powershift_percent(dev, buf, AMDGPU_PP_SENSOR_SS_DGPU_SHARE); } /** @@ -1965,7 +1930,6 @@ out: return r; } - static int ss_power_attr_update(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, uint32_t mask, enum amdgpu_device_attr_states *states) { @@ -1978,15 +1942,15 @@ static int ss_power_attr_update(struct amdgpu_device *adev, struct amdgpu_device static int ss_bias_attr_update(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, uint32_t mask, enum amdgpu_device_attr_states *states) { - uint32_t ss_power, size; + uint32_t ss_power; if (!amdgpu_device_supports_smart_shift(adev_to_drm(adev))) *states = ATTR_STATE_UNSUPPORTED; - else if (amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_SS_APU_SHARE, - (void *)&ss_power, &size)) + else if (amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_APU_SHARE, + (void *)&ss_power)) *states = ATTR_STATE_UNSUPPORTED; - else if (amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_SS_DGPU_SHARE, - (void *)&ss_power, &size)) + else if (amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_DGPU_SHARE, + (void *)&ss_power)) *states = ATTR_STATE_UNSUPPORTED; return 0; @@ -2049,8 +2013,7 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pp_dpm_dcefclk)) { if (gc_ver < IP_VERSION(9, 0, 0) || - gc_ver == IP_VERSION(9, 4, 1) || - gc_ver == IP_VERSION(9, 4, 2)) + !amdgpu_device_has_display_hardware(adev)) *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pp_dpm_fclk)) { if (mp1_ver < IP_VERSION(10, 0, 0)) @@ -2096,7 +2059,8 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ gc_ver == IP_VERSION(10, 1, 2) || gc_ver == IP_VERSION(11, 0, 0) || gc_ver == IP_VERSION(11, 0, 2) || - gc_ver == IP_VERSION(11, 0, 3))) + gc_ver == IP_VERSION(11, 0, 3) || + gc_ver == IP_VERSION(9, 4, 3))) *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pp_dpm_vclk1)) { if (!((gc_ver == IP_VERSION(10, 3, 1) || @@ -2110,7 +2074,8 @@ static int default_attr_update(struct amdgpu_device *adev, struct amdgpu_device_ gc_ver == IP_VERSION(10, 1, 2) || gc_ver == IP_VERSION(11, 0, 0) || gc_ver == IP_VERSION(11, 0, 2) || - gc_ver == IP_VERSION(11, 0, 3))) + gc_ver == IP_VERSION(11, 0, 3) || + gc_ver == IP_VERSION(9, 4, 3))) *states = ATTR_STATE_UNSUPPORTED; } else if (DEVICE_ATTR_IS(pp_dpm_dclk1)) { if (!((gc_ver == IP_VERSION(10, 3, 1) || @@ -2172,15 +2137,19 @@ static int amdgpu_device_attr_create(struct amdgpu_device *adev, uint32_t mask, struct list_head *attr_list) { int ret = 0; - struct device_attribute *dev_attr = &attr->dev_attr; - const char *name = dev_attr->attr.name; enum amdgpu_device_attr_states attr_states = ATTR_STATE_SUPPORTED; struct amdgpu_device_attr_entry *attr_entry; + struct device_attribute *dev_attr; + const char *name; int (*attr_update)(struct amdgpu_device *adev, struct amdgpu_device_attr *attr, uint32_t mask, enum amdgpu_device_attr_states *states) = default_attr_update; - BUG_ON(!attr); + if (!attr) + return -EINVAL; + + dev_attr = &attr->dev_attr; + name = dev_attr->attr.name; attr_update = attr->attr_update ? attr->attr_update : default_attr_update; @@ -2266,46 +2235,32 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); int channel = to_sensor_dev_attr(attr)->index; - int r, temp = 0, size = sizeof(temp); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; + int r, temp = 0; if (channel >= PP_TEMP_MAX) return -EINVAL; - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } - switch (channel) { case PP_TEMP_JUNCTION: /* get current junction temperature */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_HOTSPOT_TEMP, - (void *)&temp, &size); + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_HOTSPOT_TEMP, + (void *)&temp); break; case PP_TEMP_EDGE: /* get current edge temperature */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_EDGE_TEMP, - (void *)&temp, &size); + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_EDGE_TEMP, + (void *)&temp); break; case PP_TEMP_MEM: /* get current memory temperature */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MEM_TEMP, - (void *)&temp, &size); + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MEM_TEMP, + (void *)&temp); break; default: r = -EINVAL; break; } - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - if (r) return r; @@ -2589,25 +2544,10 @@ static ssize_t amdgpu_hwmon_get_fan1_min(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); u32 min_rpm = 0; - u32 size = sizeof(min_rpm); int r; - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } - - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MIN_FAN_RPM, - (void *)&min_rpm, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MIN_FAN_RPM, + (void *)&min_rpm); if (r) return r; @@ -2621,25 +2561,10 @@ static ssize_t amdgpu_hwmon_get_fan1_max(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); u32 max_rpm = 0; - u32 size = sizeof(max_rpm); int r; - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } - - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_MAX_FAN_RPM, - (void *)&max_rpm, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MAX_FAN_RPM, + (void *)&max_rpm); if (r) return r; @@ -2801,26 +2726,11 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); u32 vddgfx; - int r, size = sizeof(vddgfx); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } + int r; /* get the voltage */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDGFX, - (void *)&vddgfx, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDGFX, + (void *)&vddgfx); if (r) return r; @@ -2840,30 +2750,15 @@ static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); u32 vddnb; - int r, size = sizeof(vddnb); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; + int r; /* only APUs have vddnb */ if (!(adev->flags & AMD_IS_APU)) return -EINVAL; - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } - /* get the voltage */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDNB, - (void *)&vddnb, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDNB, + (void *)&vddnb); if (r) return r; @@ -2877,40 +2772,48 @@ static ssize_t amdgpu_hwmon_show_vddnb_label(struct device *dev, return sysfs_emit(buf, "vddnb\n"); } -static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev, - struct device_attribute *attr, - char *buf) +static unsigned int amdgpu_hwmon_get_power(struct device *dev, + enum amd_pp_sensors sensor) { struct amdgpu_device *adev = dev_get_drvdata(dev); + unsigned int uw; u32 query = 0; - int r, size = sizeof(u32); - unsigned uw; - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } - - /* get the voltage */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_POWER, - (void *)&query, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + int r; + r = amdgpu_hwmon_get_sensor_generic(adev, sensor, (void *)&query); if (r) return r; /* convert to microwatts */ uw = (query >> 8) * 1000000 + (query & 0xff) * 1000; - return sysfs_emit(buf, "%u\n", uw); + return uw; +} + +static ssize_t amdgpu_hwmon_show_power_avg(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + unsigned int val; + + val = amdgpu_hwmon_get_power(dev, AMDGPU_PP_SENSOR_GPU_AVG_POWER); + if (val < 0) + return val; + + return sysfs_emit(buf, "%u\n", val); +} + +static ssize_t amdgpu_hwmon_show_power_input(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + unsigned int val; + + val = amdgpu_hwmon_get_power(dev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER); + if (val < 0) + return val; + + return sysfs_emit(buf, "%u\n", val); } static ssize_t amdgpu_hwmon_show_power_cap_min(struct device *dev, @@ -3045,26 +2948,11 @@ static ssize_t amdgpu_hwmon_show_sclk(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); uint32_t sclk; - int r, size = sizeof(sclk); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } + int r; /* get the sclk */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_SCLK, - (void *)&sclk, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GFX_SCLK, + (void *)&sclk); if (r) return r; @@ -3084,26 +2972,11 @@ static ssize_t amdgpu_hwmon_show_mclk(struct device *dev, { struct amdgpu_device *adev = dev_get_drvdata(dev); uint32_t mclk; - int r, size = sizeof(mclk); - - if (amdgpu_in_reset(adev)) - return -EPERM; - if (adev->in_suspend && !adev->in_runpm) - return -EPERM; - - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } + int r; /* get the sclk */ - r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_MCLK, - (void *)&mclk, &size); - - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - + r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GFX_MCLK, + (void *)&mclk); if (r) return r; @@ -3163,6 +3036,8 @@ static ssize_t amdgpu_hwmon_show_mclk_label(struct device *dev, * * - power1_average: average power used by the SoC in microWatts. On APUs this includes the CPU. * + * - power1_input: instantaneous power used by the SoC in microWatts. On APUs this includes the CPU. + * * - power1_cap_min: minimum cap supported in microWatts * * - power1_cap_max: maximum cap supported in microWatts @@ -3231,6 +3106,7 @@ static SENSOR_DEVICE_ATTR(in0_label, S_IRUGO, amdgpu_hwmon_show_vddgfx_label, NU static SENSOR_DEVICE_ATTR(in1_input, S_IRUGO, amdgpu_hwmon_show_vddnb, NULL, 0); static SENSOR_DEVICE_ATTR(in1_label, S_IRUGO, amdgpu_hwmon_show_vddnb_label, NULL, 0); static SENSOR_DEVICE_ATTR(power1_average, S_IRUGO, amdgpu_hwmon_show_power_avg, NULL, 0); +static SENSOR_DEVICE_ATTR(power1_input, S_IRUGO, amdgpu_hwmon_show_power_input, NULL, 0); static SENSOR_DEVICE_ATTR(power1_cap_max, S_IRUGO, amdgpu_hwmon_show_power_cap_max, NULL, 0); static SENSOR_DEVICE_ATTR(power1_cap_min, S_IRUGO, amdgpu_hwmon_show_power_cap_min, NULL, 0); static SENSOR_DEVICE_ATTR(power1_cap, S_IRUGO | S_IWUSR, amdgpu_hwmon_show_power_cap, amdgpu_hwmon_set_power_cap, 0); @@ -3277,6 +3153,7 @@ static struct attribute *hwmon_attributes[] = { &sensor_dev_attr_in1_input.dev_attr.attr, &sensor_dev_attr_in1_label.dev_attr.attr, &sensor_dev_attr_power1_average.dev_attr.attr, + &sensor_dev_attr_power1_input.dev_attr.attr, &sensor_dev_attr_power1_cap_max.dev_attr.attr, &sensor_dev_attr_power1_cap_min.dev_attr.attr, &sensor_dev_attr_power1_cap.dev_attr.attr, @@ -3302,6 +3179,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, struct amdgpu_device *adev = dev_get_drvdata(dev); umode_t effective_mode = attr->mode; uint32_t gc_ver = adev->ip_versions[GC_HWIP][0]; + uint32_t tmp; /* under multi-vf mode, the hwmon attributes are all not supported */ if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) @@ -3387,6 +3265,14 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, (attr == &sensor_dev_attr_power1_average.dev_attr.attr)) return 0; + /* not all products support both average and instantaneous */ + if (attr == &sensor_dev_attr_power1_average.dev_attr.attr && + amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&tmp) == -EOPNOTSUPP) + return 0; + if (attr == &sensor_dev_attr_power1_input.dev_attr.attr && + amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER, (void *)&tmp) == -EOPNOTSUPP) + return 0; + /* hide max/min values if we can't both query and manage the fan */ if (((amdgpu_dpm_set_fan_speed_pwm(adev, U32_MAX) == -EOPNOTSUPP) && (amdgpu_dpm_get_fan_speed_pwm(adev, NULL) == -EOPNOTSUPP) && @@ -3425,8 +3311,10 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, (gc_ver != IP_VERSION(9, 4, 3)) && (attr == &sensor_dev_attr_temp2_input.dev_attr.attr || attr == &sensor_dev_attr_temp2_label.dev_attr.attr || + attr == &sensor_dev_attr_temp2_crit.dev_attr.attr || attr == &sensor_dev_attr_temp3_input.dev_attr.attr || - attr == &sensor_dev_attr_temp3_label.dev_attr.attr)) + attr == &sensor_dev_attr_temp3_label.dev_attr.attr || + attr == &sensor_dev_attr_temp3_crit.dev_attr.attr)) return 0; /* hotspot temperature for gc 9,4,3*/ @@ -3438,9 +3326,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, /* only SOC15 dGPUs support hotspot and mem temperatures */ if (((adev->flags & AMD_IS_APU) || gc_ver < IP_VERSION(9, 0, 0) || (gc_ver == IP_VERSION(9, 4, 3))) && - (attr == &sensor_dev_attr_temp2_crit.dev_attr.attr || - attr == &sensor_dev_attr_temp2_crit_hyst.dev_attr.attr || - attr == &sensor_dev_attr_temp3_crit.dev_attr.attr || + (attr == &sensor_dev_attr_temp2_crit_hyst.dev_attr.attr || attr == &sensor_dev_attr_temp3_crit_hyst.dev_attr.attr || attr == &sensor_dev_attr_temp1_emergency.dev_attr.attr || attr == &sensor_dev_attr_temp2_emergency.dev_attr.attr || @@ -3533,7 +3419,8 @@ void amdgpu_pm_sysfs_fini(struct amdgpu_device *adev) #if defined(CONFIG_DEBUG_FS) static void amdgpu_debugfs_prints_cpu_info(struct seq_file *m, - struct amdgpu_device *adev) { + struct amdgpu_device *adev) +{ uint16_t *p_val; uint32_t size; int i; @@ -3582,8 +3469,11 @@ static int amdgpu_debugfs_pm_info_pp(struct seq_file *m, struct amdgpu_device *a if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_VDDNB, (void *)&value, &size)) seq_printf(m, "\t%u mV (VDDNB)\n", value); size = sizeof(uint32_t); - if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_POWER, (void *)&query, &size)) + if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&query, &size)) seq_printf(m, "\t%u.%u W (average GPU)\n", query >> 8, query & 0xff); + size = sizeof(uint32_t); + if (!amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER, (void *)&query, &size)) + seq_printf(m, "\t%u.%u W (current GPU)\n", query >> 8, query & 0xff); size = sizeof(value); seq_printf(m, "\n"); diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_pm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_pm.h index 52045ad59bed..eec816f0cbf9 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_pm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_pm.h @@ -24,8 +24,7 @@ #ifndef __AMDGPU_PM_H__ #define __AMDGPU_PM_H__ -struct cg_flag_name -{ +struct cg_flag_name { u64 flag; const char *name; }; diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h b/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h index 1dc7a065a6d4..251ed011b3b0 100644 --- a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h +++ b/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h @@ -41,8 +41,7 @@ #define SMU_13_0_0_PP_OVERDRIVE_VERSION 0x83 // OverDrive 8 Table Version 0.2 #define SMU_13_0_0_PP_POWERSAVINGCLOCK_VERSION 0x01 // Power Saving Clock Table Version 1.00 -enum SMU_13_0_0_ODFEATURE_CAP -{ +enum SMU_13_0_0_ODFEATURE_CAP { SMU_13_0_0_ODCAP_GFXCLK_LIMITS = 0, SMU_13_0_0_ODCAP_UCLK_LIMITS, SMU_13_0_0_ODCAP_POWER_LIMIT, @@ -62,8 +61,7 @@ enum SMU_13_0_0_ODFEATURE_CAP SMU_13_0_0_ODCAP_COUNT, }; -enum SMU_13_0_0_ODFEATURE_ID -{ +enum SMU_13_0_0_ODFEATURE_ID { SMU_13_0_0_ODFEATURE_GFXCLK_LIMITS = 1 << SMU_13_0_0_ODCAP_GFXCLK_LIMITS, //GFXCLK Limit feature SMU_13_0_0_ODFEATURE_UCLK_LIMITS = 1 << SMU_13_0_0_ODCAP_UCLK_LIMITS, //UCLK Limit feature SMU_13_0_0_ODFEATURE_POWER_LIMIT = 1 << SMU_13_0_0_ODCAP_POWER_LIMIT, //Power Limit feature @@ -85,8 +83,7 @@ enum SMU_13_0_0_ODFEATURE_ID #define SMU_13_0_0_MAX_ODFEATURE 32 //Maximum Number of OD Features -enum SMU_13_0_0_ODSETTING_ID -{ +enum SMU_13_0_0_ODSETTING_ID { SMU_13_0_0_ODSETTING_GFXCLKFMAX = 0, SMU_13_0_0_ODSETTING_GFXCLKFMIN, SMU_13_0_0_ODSETTING_UCLKFMIN, @@ -123,8 +120,7 @@ enum SMU_13_0_0_ODSETTING_ID }; #define SMU_13_0_0_MAX_ODSETTING 64 //Maximum Number of ODSettings -enum SMU_13_0_0_PWRMODE_SETTING -{ +enum SMU_13_0_0_PWRMODE_SETTING { SMU_13_0_0_PMSETTING_POWER_LIMIT_QUIET = 0, SMU_13_0_0_PMSETTING_POWER_LIMIT_BALANCE, SMU_13_0_0_PMSETTING_POWER_LIMIT_TURBO, @@ -144,8 +140,7 @@ enum SMU_13_0_0_PWRMODE_SETTING }; #define SMU_13_0_0_MAX_PMSETTING 32 //Maximum Number of PowerMode Settings -struct smu_13_0_0_overdrive_table -{ +struct smu_13_0_0_overdrive_table { uint8_t revision; //Revision = SMU_13_0_0_PP_OVERDRIVE_VERSION uint8_t reserve[3]; //Zero filled field reserved for future use uint32_t feature_count; //Total number of supported features @@ -156,8 +151,7 @@ struct smu_13_0_0_overdrive_table int16_t pm_setting[SMU_13_0_0_MAX_PMSETTING]; //Optimized power mode feature settings }; -enum SMU_13_0_0_PPCLOCK_ID -{ +enum SMU_13_0_0_PPCLOCK_ID { SMU_13_0_0_PPCLOCK_GFXCLK = 0, SMU_13_0_0_PPCLOCK_SOCCLK, SMU_13_0_0_PPCLOCK_UCLK, @@ -175,8 +169,7 @@ enum SMU_13_0_0_PPCLOCK_ID }; #define SMU_13_0_0_MAX_PPCLOCK 16 //Maximum Number of PP Clocks -struct smu_13_0_0_powerplay_table -{ +struct smu_13_0_0_powerplay_table { struct atom_common_table_header header; //For SMU13, header.format_revision = 15, header.content_revision = 0 uint8_t table_revision; //For SMU13, table_revision = 2 uint8_t padding; diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c index 36c831b280ed..5d28c951a319 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c @@ -191,8 +191,7 @@ static void sumo_construct_vid_mapping_table(struct amdgpu_device *adev, } #if 0 -static const struct kv_lcac_config_values sx_local_cac_cfg_kv[] = -{ +static const struct kv_lcac_config_values sx_local_cac_cfg_kv[] = { { 0, 4, 1 }, { 1, 4, 1 }, { 2, 5, 1 }, @@ -204,32 +203,27 @@ static const struct kv_lcac_config_values sx_local_cac_cfg_kv[] = { 0xffffffff } }; -static const struct kv_lcac_config_values mc0_local_cac_cfg_kv[] = -{ +static const struct kv_lcac_config_values mc0_local_cac_cfg_kv[] = { { 0, 4, 1 }, { 0xffffffff } }; -static const struct kv_lcac_config_values mc1_local_cac_cfg_kv[] = -{ +static const struct kv_lcac_config_values mc1_local_cac_cfg_kv[] = { { 0, 4, 1 }, { 0xffffffff } }; -static const struct kv_lcac_config_values mc2_local_cac_cfg_kv[] = -{ +static const struct kv_lcac_config_values mc2_local_cac_cfg_kv[] = { { 0, 4, 1 }, { 0xffffffff } }; -static const struct kv_lcac_config_values mc3_local_cac_cfg_kv[] = -{ +static const struct kv_lcac_config_values mc3_local_cac_cfg_kv[] = { { 0, 4, 1 }, { 0xffffffff } }; -static const struct kv_lcac_config_values cpl_local_cac_cfg_kv[] = -{ +static const struct kv_lcac_config_values cpl_local_cac_cfg_kv[] = { { 0, 4, 1 }, { 1, 4, 1 }, { 2, 5, 1 }, @@ -260,39 +254,32 @@ static const struct kv_lcac_config_values cpl_local_cac_cfg_kv[] = { 0xffffffff } }; -static const struct kv_lcac_config_reg sx0_cac_config_reg[] = -{ +static const struct kv_lcac_config_reg sx0_cac_config_reg[] = { { 0xc0400d00, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 } }; -static const struct kv_lcac_config_reg mc0_cac_config_reg[] = -{ +static const struct kv_lcac_config_reg mc0_cac_config_reg[] = { { 0xc0400d30, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 } }; -static const struct kv_lcac_config_reg mc1_cac_config_reg[] = -{ +static const struct kv_lcac_config_reg mc1_cac_config_reg[] = { { 0xc0400d3c, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 } }; -static const struct kv_lcac_config_reg mc2_cac_config_reg[] = -{ +static const struct kv_lcac_config_reg mc2_cac_config_reg[] = { { 0xc0400d48, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 } }; -static const struct kv_lcac_config_reg mc3_cac_config_reg[] = -{ +static const struct kv_lcac_config_reg mc3_cac_config_reg[] = { { 0xc0400d54, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 } }; -static const struct kv_lcac_config_reg cpl_cac_config_reg[] = -{ +static const struct kv_lcac_config_reg cpl_cac_config_reg[] = { { 0xc0400d80, 0x003e0000, 17, 0x3fc00000, 22, 0x0001fffe, 1, 0x00000001, 0 } }; #endif -static const struct kv_pt_config_reg didt_config_kv[] = -{ +static const struct kv_pt_config_reg didt_config_kv[] = { { 0x10, 0x000000ff, 0, 0x0, KV_CONFIGREG_DIDT_IND }, { 0x10, 0x0000ff00, 8, 0x0, KV_CONFIGREG_DIDT_IND }, { 0x10, 0x00ff0000, 16, 0x0, KV_CONFIGREG_DIDT_IND }, @@ -1173,9 +1160,9 @@ static void kv_calculate_dfs_bypass_settings(struct amdgpu_device *adev) pi->graphics_level[i].ClkBypassCntl = 2; else if (kv_get_clock_difference(table->entries[i].clk, 26600) < 200) pi->graphics_level[i].ClkBypassCntl = 7; - else if (kv_get_clock_difference(table->entries[i].clk , 20000) < 200) + else if (kv_get_clock_difference(table->entries[i].clk, 20000) < 200) pi->graphics_level[i].ClkBypassCntl = 6; - else if (kv_get_clock_difference(table->entries[i].clk , 10000) < 200) + else if (kv_get_clock_difference(table->entries[i].clk, 10000) < 200) pi->graphics_level[i].ClkBypassCntl = 8; else pi->graphics_level[i].ClkBypassCntl = 0; @@ -1825,7 +1812,7 @@ static void kv_set_valid_clock_range(struct amdgpu_device *adev, if ((new_ps->levels[0].sclk - table->entries[pi->highest_valid].sclk_frequency) > (table->entries[pi->lowest_valid].sclk_frequency - - new_ps->levels[new_ps->num_levels -1].sclk)) + new_ps->levels[new_ps->num_levels - 1].sclk)) pi->highest_valid = pi->lowest_valid; else pi->lowest_valid = pi->highest_valid; @@ -3333,8 +3320,7 @@ static const struct amd_ip_funcs kv_dpm_ip_funcs = { .set_powergating_state = kv_dpm_set_powergating_state, }; -const struct amdgpu_ip_block_version kv_smu_ip_block = -{ +const struct amdgpu_ip_block_version kv_smu_ip_block = { .type = AMD_IP_BLOCK_TYPE_SMC, .major = 1, .minor = 0, diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/r600_dpm.h b/drivers/gpu/drm/amd/pm/legacy-dpm/r600_dpm.h index 055321f61ca7..3e7caa715533 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/r600_dpm.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/r600_dpm.h @@ -117,8 +117,7 @@ enum r600_display_watermark { R600_DISPLAY_WATERMARK_HIGH = 1, }; -enum r600_display_gap -{ +enum r600_display_gap { R600_PM_DISPLAY_GAP_VBLANK_OR_WM = 0, R600_PM_DISPLAY_GAP_VBLANK = 1, R600_PM_DISPLAY_GAP_WATERMARK = 2, diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/sislands_smc.h b/drivers/gpu/drm/amd/pm/legacy-dpm/sislands_smc.h index c7dc117a688c..90ec411c5029 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/sislands_smc.h +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/sislands_smc.h @@ -29,8 +29,7 @@ #define SISLANDS_MAX_SMC_PERFORMANCE_LEVELS_PER_SWSTATE 16 -struct PP_SIslands_Dpm2PerfLevel -{ +struct PP_SIslands_Dpm2PerfLevel { uint8_t MaxPS; uint8_t TgtAct; uint8_t MaxPS_StepInc; @@ -47,8 +46,7 @@ struct PP_SIslands_Dpm2PerfLevel typedef struct PP_SIslands_Dpm2PerfLevel PP_SIslands_Dpm2PerfLevel; -struct PP_SIslands_DPM2Status -{ +struct PP_SIslands_DPM2Status { uint32_t dpm2Flags; uint8_t CurrPSkip; uint8_t CurrPSkipPowerShift; @@ -68,8 +66,7 @@ struct PP_SIslands_DPM2Status typedef struct PP_SIslands_DPM2Status PP_SIslands_DPM2Status; -struct PP_SIslands_DPM2Parameters -{ +struct PP_SIslands_DPM2Parameters { uint32_t TDPLimit; uint32_t NearTDPLimit; uint32_t SafePowerLimit; @@ -78,8 +75,7 @@ struct PP_SIslands_DPM2Parameters }; typedef struct PP_SIslands_DPM2Parameters PP_SIslands_DPM2Parameters; -struct PP_SIslands_PAPMStatus -{ +struct PP_SIslands_PAPMStatus { uint32_t EstimatedDGPU_T; uint32_t EstimatedDGPU_P; uint32_t EstimatedAPU_T; @@ -89,8 +85,7 @@ struct PP_SIslands_PAPMStatus }; typedef struct PP_SIslands_PAPMStatus PP_SIslands_PAPMStatus; -struct PP_SIslands_PAPMParameters -{ +struct PP_SIslands_PAPMParameters { uint32_t NearTDPLimitTherm; uint32_t NearTDPLimitPAPM; uint32_t PlatformPowerLimit; @@ -100,8 +95,7 @@ struct PP_SIslands_PAPMParameters }; typedef struct PP_SIslands_PAPMParameters PP_SIslands_PAPMParameters; -struct SISLANDS_SMC_SCLK_VALUE -{ +struct SISLANDS_SMC_SCLK_VALUE { uint32_t vCG_SPLL_FUNC_CNTL; uint32_t vCG_SPLL_FUNC_CNTL_2; uint32_t vCG_SPLL_FUNC_CNTL_3; @@ -113,8 +107,7 @@ struct SISLANDS_SMC_SCLK_VALUE typedef struct SISLANDS_SMC_SCLK_VALUE SISLANDS_SMC_SCLK_VALUE; -struct SISLANDS_SMC_MCLK_VALUE -{ +struct SISLANDS_SMC_MCLK_VALUE { uint32_t vMPLL_FUNC_CNTL; uint32_t vMPLL_FUNC_CNTL_1; uint32_t vMPLL_FUNC_CNTL_2; @@ -129,8 +122,7 @@ struct SISLANDS_SMC_MCLK_VALUE typedef struct SISLANDS_SMC_MCLK_VALUE SISLANDS_SMC_MCLK_VALUE; -struct SISLANDS_SMC_VOLTAGE_VALUE -{ +struct SISLANDS_SMC_VOLTAGE_VALUE { uint16_t value; uint8_t index; uint8_t phase_settings; @@ -138,8 +130,7 @@ struct SISLANDS_SMC_VOLTAGE_VALUE typedef struct SISLANDS_SMC_VOLTAGE_VALUE SISLANDS_SMC_VOLTAGE_VALUE; -struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL -{ +struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL { uint8_t ACIndex; uint8_t displayWatermark; uint8_t gen2PCIE; @@ -180,8 +171,7 @@ struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL typedef struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL SISLANDS_SMC_HW_PERFORMANCE_LEVEL; -struct SISLANDS_SMC_SWSTATE -{ +struct SISLANDS_SMC_SWSTATE { uint8_t flags; uint8_t levelCount; uint8_t padding2; @@ -205,8 +195,7 @@ struct SISLANDS_SMC_SWSTATE_SINGLE { #define SISLANDS_SMC_VOLTAGEMASK_VDDC_PHASE_SHEDDING 3 #define SISLANDS_SMC_VOLTAGEMASK_MAX 4 -struct SISLANDS_SMC_VOLTAGEMASKTABLE -{ +struct SISLANDS_SMC_VOLTAGEMASKTABLE { uint32_t lowMask[SISLANDS_SMC_VOLTAGEMASK_MAX]; }; @@ -214,8 +203,7 @@ typedef struct SISLANDS_SMC_VOLTAGEMASKTABLE SISLANDS_SMC_VOLTAGEMASKTABLE; #define SISLANDS_MAX_NO_VREG_STEPS 32 -struct SISLANDS_SMC_STATETABLE -{ +struct SISLANDS_SMC_STATETABLE { uint8_t thermalProtectType; uint8_t systemFlags; uint8_t maxVDDCIndexInPPTable; @@ -254,8 +242,7 @@ typedef struct SISLANDS_SMC_STATETABLE SISLANDS_SMC_STATETABLE; #define SI_SMC_SOFT_REGISTER_svi_rework_gpio_id_svd 0x11c #define SI_SMC_SOFT_REGISTER_svi_rework_gpio_id_svc 0x120 -struct PP_SIslands_FanTable -{ +struct PP_SIslands_FanTable { uint8_t fdo_mode; uint8_t padding; int16_t temp_min; @@ -285,8 +272,7 @@ typedef struct PP_SIslands_FanTable PP_SIslands_FanTable; #define SMC_SISLANDS_SCALE_I 7 #define SMC_SISLANDS_SCALE_R 12 -struct PP_SIslands_CacConfig -{ +struct PP_SIslands_CacConfig { uint16_t cac_lkge_lut[SMC_SISLANDS_LKGE_LUT_NUM_OF_TEMP_ENTRIES][SMC_SISLANDS_LKGE_LUT_NUM_OF_VOLT_ENTRIES]; uint32_t lkge_lut_V0; uint32_t lkge_lut_Vstep; @@ -308,23 +294,20 @@ typedef struct PP_SIslands_CacConfig PP_SIslands_CacConfig; #define SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE 16 #define SMC_SISLANDS_MC_REGISTER_ARRAY_SET_COUNT 20 -struct SMC_SIslands_MCRegisterAddress -{ +struct SMC_SIslands_MCRegisterAddress { uint16_t s0; uint16_t s1; }; typedef struct SMC_SIslands_MCRegisterAddress SMC_SIslands_MCRegisterAddress; -struct SMC_SIslands_MCRegisterSet -{ +struct SMC_SIslands_MCRegisterSet { uint32_t value[SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE]; }; typedef struct SMC_SIslands_MCRegisterSet SMC_SIslands_MCRegisterSet; -struct SMC_SIslands_MCRegisters -{ +struct SMC_SIslands_MCRegisters { uint8_t last; uint8_t reserved[3]; SMC_SIslands_MCRegisterAddress address[SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE]; @@ -333,8 +316,7 @@ struct SMC_SIslands_MCRegisters typedef struct SMC_SIslands_MCRegisters SMC_SIslands_MCRegisters; -struct SMC_SIslands_MCArbDramTimingRegisterSet -{ +struct SMC_SIslands_MCArbDramTimingRegisterSet { uint32_t mc_arb_dram_timing; uint32_t mc_arb_dram_timing2; uint8_t mc_arb_rfsh_rate; @@ -344,8 +326,7 @@ struct SMC_SIslands_MCArbDramTimingRegisterSet typedef struct SMC_SIslands_MCArbDramTimingRegisterSet SMC_SIslands_MCArbDramTimingRegisterSet; -struct SMC_SIslands_MCArbDramTimingRegisters -{ +struct SMC_SIslands_MCArbDramTimingRegisters { uint8_t arb_current; uint8_t reserved[3]; SMC_SIslands_MCArbDramTimingRegisterSet data[16]; @@ -353,8 +334,7 @@ struct SMC_SIslands_MCArbDramTimingRegisters typedef struct SMC_SIslands_MCArbDramTimingRegisters SMC_SIslands_MCArbDramTimingRegisters; -struct SMC_SISLANDS_SPLL_DIV_TABLE -{ +struct SMC_SISLANDS_SPLL_DIV_TABLE { uint32_t freq[256]; uint32_t ss[256]; }; @@ -374,8 +354,7 @@ typedef struct SMC_SISLANDS_SPLL_DIV_TABLE SMC_SISLANDS_SPLL_DIV_TABLE; #define SMC_SISLANDS_DTE_MAX_TEMPERATURE_DEPENDENT_ARRAY_SIZE 16 -struct Smc_SIslands_DTE_Configuration -{ +struct Smc_SIslands_DTE_Configuration { uint32_t tau[SMC_SISLANDS_DTE_MAX_FILTER_STAGES]; uint32_t R[SMC_SISLANDS_DTE_MAX_FILTER_STAGES]; uint32_t K; diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c index ff360c699171..9e4f8a4104a3 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c @@ -612,7 +612,7 @@ static int pp_dpm_get_pp_num_states(void *handle, memset(data, 0, sizeof(*data)); - if (!hwmgr || !hwmgr->pm_en ||!hwmgr->ps) + if (!hwmgr || !hwmgr->pm_en || !hwmgr->ps) return -EINVAL; data->nums = hwmgr->num_ps; @@ -644,7 +644,7 @@ static int pp_dpm_get_pp_table(void *handle, char **table) { struct pp_hwmgr *hwmgr = handle; - if (!hwmgr || !hwmgr->pm_en ||!hwmgr->soft_pp_table) + if (!hwmgr || !hwmgr->pm_en || !hwmgr->soft_pp_table) return -EINVAL; *table = (char *)hwmgr->soft_pp_table; @@ -1002,7 +1002,7 @@ static int pp_get_power_limit(void *handle, uint32_t *limit, struct pp_hwmgr *hwmgr = handle; int ret = 0; - if (!hwmgr || !hwmgr->pm_en ||!limit) + if (!hwmgr || !hwmgr->pm_en || !limit) return -EINVAL; if (power_type != PP_PWR_TYPE_SUSTAINED) @@ -1047,7 +1047,7 @@ static int pp_get_display_power_level(void *handle, { struct pp_hwmgr *hwmgr = handle; - if (!hwmgr || !hwmgr->pm_en ||!output) + if (!hwmgr || !hwmgr->pm_en || !output) return -EINVAL; return phm_get_dal_power_level(hwmgr, output); @@ -1120,7 +1120,7 @@ static int pp_get_clock_by_type_with_latency(void *handle, { struct pp_hwmgr *hwmgr = handle; - if (!hwmgr || !hwmgr->pm_en ||!clocks) + if (!hwmgr || !hwmgr->pm_en || !clocks) return -EINVAL; return phm_get_clock_by_type_with_latency(hwmgr, type, clocks); @@ -1132,7 +1132,7 @@ static int pp_get_clock_by_type_with_voltage(void *handle, { struct pp_hwmgr *hwmgr = handle; - if (!hwmgr || !hwmgr->pm_en ||!clocks) + if (!hwmgr || !hwmgr->pm_en || !clocks) return -EINVAL; return phm_get_clock_by_type_with_voltage(hwmgr, type, clocks); @@ -1155,7 +1155,7 @@ static int pp_display_clock_voltage_request(void *handle, { struct pp_hwmgr *hwmgr = handle; - if (!hwmgr || !hwmgr->pm_en ||!clock) + if (!hwmgr || !hwmgr->pm_en || !clock) return -EINVAL; return phm_display_clock_voltage_request(hwmgr, clock); @@ -1167,7 +1167,7 @@ static int pp_get_display_mode_validation_clocks(void *handle, struct pp_hwmgr *hwmgr = handle; int ret = 0; - if (!hwmgr || !hwmgr->pm_en ||!clocks) + if (!hwmgr || !hwmgr->pm_en || !clocks) return -EINVAL; clocks->level = PP_DAL_POWERLEVEL_7; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.c index 45f608838f6e..65b95d6be5c5 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ci_baco.c @@ -38,8 +38,7 @@ #include "gca/gfx_7_2_d.h" #include "gca/gfx_7_2_sh_mask.h" -static const struct baco_cmd_entry gpio_tbl[] = -{ +static const struct baco_cmd_entry gpio_tbl[] = { { CMD_WRITE, mmGPIOPAD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PU_EN, 0, 0, 0, 0x0 }, @@ -52,15 +51,13 @@ static const struct baco_cmd_entry gpio_tbl[] = { CMD_READMODIFYWRITE, mmDC_GPIO_SYNCA_MASK, 0, 0, 0, 0x00001111 } }; -static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = -{ +static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, 0xC0300024 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, 0x1, 0x0, 0, 0x1 }, { CMD_WRITE, mmBIF_FB_EN, 0, 0, 0, 0x0 } }; -static const struct baco_cmd_entry use_bclk_tbl[] = -{ +static const struct baco_cmd_entry use_bclk_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN__SHIFT, 0, 0x1 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL_2 }, @@ -82,8 +79,7 @@ static const struct baco_cmd_entry use_bclk_tbl[] = { CMD_READMODIFYWRITE, mmMPLL_CNTL_MODE, MPLL_CNTL_MODE__MPLL_MCLK_SEL_MASK, MPLL_CNTL_MODE__MPLL_MCLK_SEL__SHIFT, 0, 0x0 } }; -static const struct baco_cmd_entry turn_off_plls_tbl[] = -{ +static const struct baco_cmd_entry turn_off_plls_tbl[] = { { CMD_READMODIFYWRITE, mmDISPPLL_BG_CNTL, DISPPLL_BG_CNTL__DISPPLL_BG_PDN_MASK, DISPPLL_BG_CNTL__DISPPLL_BG_PDN__SHIFT, 0, 0x1 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_CLKPIN_CNTL_DC }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_CLKPIN_CNTL_DC__OSC_EN_MASK, CG_CLKPIN_CNTL_DC__OSC_EN__SHIFT, 0, 0x0 }, @@ -120,8 +116,7 @@ static const struct baco_cmd_entry turn_off_plls_tbl[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, THM_CLK_CNTL__TMON_CLK_SEL_MASK, THM_CLK_CNTL__TMON_CLK_SEL__SHIFT, 0, 0x2 } }; -static const struct baco_cmd_entry enter_baco_tbl[] = -{ +static const struct baco_cmd_entry enter_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, BACO_CNTL__BACO_BCLK_OFF__SHIFT, 0, 0x01 }, { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, 0, 5, 0x02 }, @@ -136,8 +131,7 @@ static const struct baco_cmd_entry enter_baco_tbl[] = #define BACO_CNTL__PWRGOOD_MASK BACO_CNTL__PWRGOOD_GPIO_MASK+BACO_CNTL__PWRGOOD_MEM_MASK+BACO_CNTL__PWRGOOD_DVO_MASK -static const struct baco_cmd_entry exit_baco_tbl[] = -{ +static const struct baco_cmd_entry exit_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_RESET_EN_MASK, BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, BACO_CNTL__BACO_BCLK_OFF__SHIFT, 0, 0x00 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0x00 }, @@ -152,8 +146,7 @@ static const struct baco_cmd_entry exit_baco_tbl[] = { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0x00 } }; -static const struct baco_cmd_entry clean_baco_tbl[] = -{ +static const struct baco_cmd_entry clean_baco_tbl[] = { { CMD_WRITE, mmBIOS_SCRATCH_6, 0, 0, 0, 0 }, { CMD_WRITE, mmBIOS_SCRATCH_7, 0, 0, 0, 0 }, { CMD_WRITE, mmCP_PFP_UCODE_ADDR, 0, 0, 0, 0 } diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.c index 1c73776bd606..fd79337a3536 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/common_baco.c @@ -42,7 +42,7 @@ static bool baco_wait_register(struct pp_hwmgr *hwmgr, u32 reg, u32 mask, u32 va } static bool baco_cmd_handler(struct pp_hwmgr *hwmgr, u32 command, u32 reg, u32 mask, - u32 shift, u32 value, u32 timeout) + u32 shift, u32 value, u32 timeout) { struct amdgpu_device *adev = (struct amdgpu_device *)(hwmgr->adev); u32 data; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.c index c0368f2dfb21..b3e768fa79f2 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/fiji_baco.c @@ -36,8 +36,7 @@ #include "smu/smu_7_1_3_sh_mask.h" -static const struct baco_cmd_entry gpio_tbl[] = -{ +static const struct baco_cmd_entry gpio_tbl[] = { { CMD_WRITE, mmGPIOPAD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PU_EN, 0, 0, 0, 0x0 }, @@ -50,15 +49,13 @@ static const struct baco_cmd_entry gpio_tbl[] = { CMD_READMODIFYWRITE, mmDC_GPIO_SYNCA_MASK, 0, 0, 0, 0x00001111 } }; -static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = -{ +static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, 0xC0300024 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, 0x1, 0x0, 0, 0x1 }, { CMD_WRITE, mmBIF_FB_EN, 0, 0, 0, 0x0 } }; -static const struct baco_cmd_entry use_bclk_tbl[] = -{ +static const struct baco_cmd_entry use_bclk_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN__SHIFT, 0, 0x1 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL_2 }, @@ -78,8 +75,7 @@ static const struct baco_cmd_entry use_bclk_tbl[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL_MASK, MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL__SHIFT, 0, 0x2 } }; -static const struct baco_cmd_entry turn_off_plls_tbl[] = -{ +static const struct baco_cmd_entry turn_off_plls_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_RESET_MASK, CG_SPLL_FUNC_CNTL__SPLL_RESET__SHIFT, 0, 0x1 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_PWRON_MASK, CG_SPLL_FUNC_CNTL__SPLL_PWRON__SHIFT, 0, 0x0 }, @@ -88,8 +84,7 @@ static const struct baco_cmd_entry turn_off_plls_tbl[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, 0x8000000, 0x1b, 0, 0x0 } }; -static const struct baco_cmd_entry clk_req_b_tbl[] = -{ +static const struct baco_cmd_entry clk_req_b_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_CLKPIN_CNTL_2 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_CLKPIN_CNTL_2__FORCE_BIF_REFCLK_EN_MASK, CG_CLKPIN_CNTL_2__FORCE_BIF_REFCLK_EN__SHIFT, 0, 0x0 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixMPLL_BYPASSCLK_SEL }, @@ -104,8 +99,7 @@ static const struct baco_cmd_entry clk_req_b_tbl[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, THM_CLK_CNTL__TMON_CLK_SEL_MASK, THM_CLK_CNTL__TMON_CLK_SEL__SHIFT, 0, 0x1 } }; -static const struct baco_cmd_entry enter_baco_tbl[] = -{ +static const struct baco_cmd_entry enter_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BIF_SCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_SCLK_SWITCH__SHIFT, 0, 0x01 }, { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_BIF_SCLK_SWITCH_MASK, 0, 5, 0x40000 }, @@ -122,8 +116,7 @@ static const struct baco_cmd_entry enter_baco_tbl[] = #define BACO_CNTL__PWRGOOD_MASK BACO_CNTL__PWRGOOD_GPIO_MASK+BACO_CNTL__PWRGOOD_MEM_MASK+BACO_CNTL__PWRGOOD_DVO_MASK -static const struct baco_cmd_entry exit_baco_tbl[] = -{ +static const struct baco_cmd_entry exit_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_RESET_EN_MASK, BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, BACO_CNTL__BACO_BCLK_OFF__SHIFT, 0, 0x00 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0x00 }, @@ -138,8 +131,7 @@ static const struct baco_cmd_entry exit_baco_tbl[] = { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0x00 } }; -static const struct baco_cmd_entry clean_baco_tbl[] = -{ +static const struct baco_cmd_entry clean_baco_tbl[] = { { CMD_WRITE, mmBIOS_SCRATCH_0, 0, 0, 0, 0 }, { CMD_WRITE, mmBIOS_SCRATCH_1, 0, 0, 0, 0 }, { CMD_WRITE, mmBIOS_SCRATCH_2, 0, 0, 0, 0 }, diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr.c index f2cef0930aa9..2b5ac21fee39 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/hwmgr.c @@ -120,7 +120,7 @@ int hwmgr_early_init(struct pp_hwmgr *hwmgr) case CHIP_TOPAZ: hwmgr->smumgr_funcs = &iceland_smu_funcs; topaz_set_asic_special_caps(hwmgr); - hwmgr->feature_mask &= ~ (PP_VBI_TIME_SUPPORT_MASK | + hwmgr->feature_mask &= ~(PP_VBI_TIME_SUPPORT_MASK | PP_ENABLE_GFX_CG_THRU_SMU); hwmgr->pp_table_version = PP_TABLE_V0; hwmgr->od_enabled = false; @@ -133,7 +133,7 @@ int hwmgr_early_init(struct pp_hwmgr *hwmgr) case CHIP_FIJI: hwmgr->smumgr_funcs = &fiji_smu_funcs; fiji_set_asic_special_caps(hwmgr); - hwmgr->feature_mask &= ~ (PP_VBI_TIME_SUPPORT_MASK | + hwmgr->feature_mask &= ~(PP_VBI_TIME_SUPPORT_MASK | PP_ENABLE_GFX_CG_THRU_SMU); break; case CHIP_POLARIS11: @@ -195,7 +195,7 @@ int hwmgr_early_init(struct pp_hwmgr *hwmgr) int hwmgr_sw_init(struct pp_hwmgr *hwmgr) { - if (!hwmgr|| !hwmgr->smumgr_funcs || !hwmgr->smumgr_funcs->smu_init) + if (!hwmgr || !hwmgr->smumgr_funcs || !hwmgr->smumgr_funcs->smu_init) return -EINVAL; phm_register_irq_handlers(hwmgr); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.c index 8f8e296f2fe9..a6a6d43b09f8 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/polaris_baco.c @@ -35,8 +35,7 @@ #include "smu/smu_7_1_3_d.h" #include "smu/smu_7_1_3_sh_mask.h" -static const struct baco_cmd_entry gpio_tbl[] = -{ +static const struct baco_cmd_entry gpio_tbl[] = { { CMD_WRITE, mmGPIOPAD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PU_EN, 0, 0, 0, 0x0 }, @@ -49,15 +48,13 @@ static const struct baco_cmd_entry gpio_tbl[] = { CMD_READMODIFYWRITE, mmDC_GPIO_SYNCA_MASK, 0, 0, 0, 0x00001111 } }; -static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = -{ +static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, 0xC0300024 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, 0x1, 0x0, 0, 0x1 }, { CMD_WRITE, mmBIF_FB_EN, 0, 0, 0, 0x0 } }; -static const struct baco_cmd_entry use_bclk_tbl[] = -{ +static const struct baco_cmd_entry use_bclk_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN__SHIFT, 0, 0x1 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, 0xC0500170 }, @@ -70,8 +67,7 @@ static const struct baco_cmd_entry use_bclk_tbl[] = { CMD_READMODIFYWRITE, mmMPLL_CNTL_MODE, MPLL_CNTL_MODE__MPLL_MCLK_SEL_MASK, MPLL_CNTL_MODE__MPLL_MCLK_SEL__SHIFT, 0, 0x0 } }; -static const struct baco_cmd_entry turn_off_plls_tbl[] = -{ +static const struct baco_cmd_entry turn_off_plls_tbl[] = { { CMD_READMODIFYWRITE, mmDC_GPIO_PAD_STRENGTH_1, DC_GPIO_PAD_STRENGTH_1__GENLK_STRENGTH_SP_MASK, DC_GPIO_PAD_STRENGTH_1__GENLK_STRENGTH_SP__SHIFT, 0, 0x1 }, { CMD_DELAY_US, 0, 0, 0, 1, 0x0 }, { CMD_READMODIFYWRITE, mmMC_SEQ_DRAM, MC_SEQ_DRAM__RST_CTL_MASK, MC_SEQ_DRAM__RST_CTL__SHIFT, 0, 0x1 }, @@ -92,8 +88,7 @@ static const struct baco_cmd_entry turn_off_plls_tbl[] = { CMD_DELAY_US, 0, 0, 0, 5, 0x0 } }; -static const struct baco_cmd_entry clk_req_b_tbl[] = -{ +static const struct baco_cmd_entry clk_req_b_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixTHM_CLK_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, THM_CLK_CNTL__CMON_CLK_SEL_MASK, THM_CLK_CNTL__CMON_CLK_SEL__SHIFT, 0, 0x1 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, THM_CLK_CNTL__TMON_CLK_SEL_MASK, THM_CLK_CNTL__TMON_CLK_SEL__SHIFT, 0, 0x1 }, @@ -108,8 +103,7 @@ static const struct baco_cmd_entry clk_req_b_tbl[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL_MASK, MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL__SHIFT, 0, 0x4 } }; -static const struct baco_cmd_entry enter_baco_tbl[] = -{ +static const struct baco_cmd_entry enter_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BIF_SCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_SCLK_SWITCH__SHIFT, 0, 0x01 }, { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_BIF_SCLK_SWITCH_MASK, 0, 5, 0x40000 }, @@ -126,8 +120,7 @@ static const struct baco_cmd_entry enter_baco_tbl[] = #define BACO_CNTL__PWRGOOD_MASK BACO_CNTL__PWRGOOD_GPIO_MASK+BACO_CNTL__PWRGOOD_MEM_MASK+BACO_CNTL__PWRGOOD_DVO_MASK -static const struct baco_cmd_entry exit_baco_tbl[] = -{ +static const struct baco_cmd_entry exit_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_RESET_EN_MASK, BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, BACO_CNTL__BACO_BCLK_OFF__SHIFT, 0, 0x00 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0x00 }, @@ -142,14 +135,12 @@ static const struct baco_cmd_entry exit_baco_tbl[] = { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0x00 } }; -static const struct baco_cmd_entry clean_baco_tbl[] = -{ +static const struct baco_cmd_entry clean_baco_tbl[] = { { CMD_WRITE, mmBIOS_SCRATCH_6, 0, 0, 0, 0 }, { CMD_WRITE, mmBIOS_SCRATCH_7, 0, 0, 0, 0 } }; -static const struct baco_cmd_entry use_bclk_tbl_vg[] = -{ +static const struct baco_cmd_entry use_bclk_tbl_vg[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN__SHIFT, 0, 0x1 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, 0xC0500170 }, @@ -160,8 +151,7 @@ static const struct baco_cmd_entry use_bclk_tbl_vg[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL_MASK, MPLL_BYPASSCLK_SEL__MPLL_CLKOUT_SEL__SHIFT, 0, 0x2 } }; -static const struct baco_cmd_entry turn_off_plls_tbl_vg[] = -{ +static const struct baco_cmd_entry turn_off_plls_tbl_vg[] = { { CMD_READMODIFYWRITE, mmDC_GPIO_PAD_STRENGTH_1, DC_GPIO_PAD_STRENGTH_1__GENLK_STRENGTH_SP_MASK, DC_GPIO_PAD_STRENGTH_1__GENLK_STRENGTH_SP__SHIFT, 0, 0x1 }, { CMD_DELAY_US, 0, 0, 0, 1, 0x0 }, { CMD_READMODIFYWRITE, mmMC_SEQ_DRAM, MC_SEQ_DRAM__RST_CTL_MASK, MC_SEQ_DRAM__RST_CTL__SHIFT, 0, 0x1 }, diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h index b3103bd4be42..1f987e846628 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h @@ -278,16 +278,14 @@ struct pp_atom_ctrl__avfs_parameters { uint8_t ucReserved; }; -struct _AtomCtrl_HiLoLeakageOffsetTable -{ +struct _AtomCtrl_HiLoLeakageOffsetTable { USHORT usHiLoLeakageThreshold; USHORT usEdcDidtLoDpm7TableOffset; USHORT usEdcDidtHiDpm7TableOffset; }; typedef struct _AtomCtrl_HiLoLeakageOffsetTable AtomCtrl_HiLoLeakageOffsetTable; -struct _AtomCtrl_EDCLeakgeTable -{ +struct _AtomCtrl_EDCLeakgeTable { ULONG DIDT_REG[24]; }; typedef struct _AtomCtrl_EDCLeakgeTable AtomCtrl_EDCLeakgeTable; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.h index 2fc1733bcdcf..e86e05c786d9 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.h @@ -147,8 +147,7 @@ struct pp_atomfwctrl_bios_boot_up_values { uint8_t ucCoolingID; }; -struct pp_atomfwctrl_smc_dpm_parameters -{ +struct pp_atomfwctrl_smc_dpm_parameters { uint8_t liquid1_i2c_address; uint8_t liquid2_i2c_address; uint8_t vr_i2c_address; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h index dac29fe6cfc6..6f54c410c2f9 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h @@ -166,7 +166,7 @@ static fInt fNaturalLog(fInt value) error_term = fAdd(fNegativeOne, value); - return (fAdd(solution, error_term)); + return fAdd(solution, error_term); } static fInt fDecodeLinearFuse(uint32_t fuse_value, fInt f_min, fInt f_range, uint32_t bitlength) @@ -230,7 +230,7 @@ static fInt ConvertToFraction(int X) /*Add all range checking here. Is it possib static fInt fNegate(fInt X) { fInt CONSTANT_NEGONE = ConvertToFraction(-1); - return (fMultiply(X, CONSTANT_NEGONE)); + return fMultiply(X, CONSTANT_NEGONE); } static fInt Convert_ULONG_ToFraction(uint32_t X) @@ -382,14 +382,14 @@ static int ConvertBackToInteger (fInt A) /*THIS is the function that will be use scaledDecimal.full = uGetScaledDecimal(A); - fullNumber = fAdd(scaledDecimal,scaledReal); + fullNumber = fAdd(scaledDecimal, scaledReal); return fullNumber.full; } static fInt fGetSquare(fInt A) { - return fMultiply(A,A); + return fMultiply(A, A); } /* x_new = x_old - (x_old^2 - C) / (2 * x_old) */ @@ -447,7 +447,7 @@ static fInt fSqrt(fInt num) } while (uAbs(error) > 0); - return (x_new); + return x_new; } static void SolveQuadracticEqn(fInt A, fInt B, fInt C, fInt Roots[]) @@ -459,7 +459,7 @@ static void SolveQuadracticEqn(fInt A, fInt B, fInt C, fInt Roots[]) f_CONSTANT100 = ConvertToFraction(100); f_CONSTANT10 = ConvertToFraction(10); - while(GreaterThan(A, f_CONSTANT100) || GreaterThan(B, f_CONSTANT100) || GreaterThan(C, f_CONSTANT100)) { + while (GreaterThan(A, f_CONSTANT100) || GreaterThan(B, f_CONSTANT100) || GreaterThan(C, f_CONSTANT100)) { A = fDivide(A, f_CONSTANT10); B = fDivide(B, f_CONSTANT10); C = fDivide(C, f_CONSTANT10); @@ -515,7 +515,7 @@ static int uGetScaledDecimal (fInt A) /*Converts the fractional portion to whole dec[i] = tmp / (1 << SHIFT_AMOUNT); tmp = tmp - ((1 << SHIFT_AMOUNT)*dec[i]); tmp *= 10; - scaledDecimal = scaledDecimal + dec[i]*uPow(10, PRECISION - 1 -i); + scaledDecimal = scaledDecimal + dec[i]*uPow(10, PRECISION - 1 - i); } return scaledDecimal; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h index b0ac4d121adc..7a31cfa5e7fb 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/pptable_v1_0.h @@ -419,8 +419,7 @@ typedef struct _ATOM_Fiji_PowerTune_Table { USHORT usReserved; } ATOM_Fiji_PowerTune_Table; -typedef struct _ATOM_Polaris_PowerTune_Table -{ +typedef struct _ATOM_Polaris_PowerTune_Table { UCHAR ucRevId; USHORT usTDP; USHORT usConfigurableTDP; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.c index 182118e3fd5f..5794b64507bf 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/processpptables.c @@ -1237,7 +1237,7 @@ static int get_vce_clock_voltage_limit_table(struct pp_hwmgr *hwmgr, const VCEClockInfoArray *array) { unsigned long i; - struct phm_vce_clock_voltage_dependency_table *vce_table = NULL; + struct phm_vce_clock_voltage_dependency_table *vce_table; vce_table = kzalloc(struct_size(vce_table, entries, table->numEntries), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c index 86d6e88c7386..02ba68d7c654 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c @@ -430,37 +430,37 @@ static int smu10_apply_state_adjust_rules(struct pp_hwmgr *hwmgr, } /* temporary hardcoded clock voltage breakdown tables */ -static const DpmClock_t VddDcfClk[]= { +static const DpmClock_t VddDcfClk[] = { { 300, 2600}, { 600, 3200}, { 600, 3600}, }; -static const DpmClock_t VddSocClk[]= { +static const DpmClock_t VddSocClk[] = { { 478, 2600}, { 722, 3200}, { 722, 3600}, }; -static const DpmClock_t VddFClk[]= { +static const DpmClock_t VddFClk[] = { { 400, 2600}, {1200, 3200}, {1200, 3600}, }; -static const DpmClock_t VddDispClk[]= { +static const DpmClock_t VddDispClk[] = { { 435, 2600}, { 661, 3200}, {1086, 3600}, }; -static const DpmClock_t VddDppClk[]= { +static const DpmClock_t VddDppClk[] = { { 435, 2600}, { 661, 3200}, { 661, 3600}, }; -static const DpmClock_t VddPhyClk[]= { +static const DpmClock_t VddPhyClk[] = { { 540, 2600}, { 810, 3200}, { 810, 3600}, @@ -1358,7 +1358,7 @@ static int smu10_set_watermarks_for_clocks_ranges(struct pp_hwmgr *hwmgr, struct amdgpu_device *adev = hwmgr->adev; int i; - smu_set_watermarks_for_clocks_ranges(table,wm_with_clock_ranges); + smu_set_watermarks_for_clocks_ranges(table, wm_with_clock_ranges); if (adev->apu_flags & AMD_APU_IS_RAVEN2) { for (i = 0; i < NUM_WM_RANGES; i++) @@ -1461,7 +1461,7 @@ static int smu10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) phm_get_sysfs_buf(&buf, &size); - size += sysfs_emit_at(buf, size, "%s %16s %s %s %s %s\n",title[0], + size += sysfs_emit_at(buf, size, "%s %16s %s %s %s %s\n", title[0], title[1], title[2], title[3], title[4], title[5]); for (i = 0; i <= PP_SMC_POWER_PROFILE_COMPUTE; i++) diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c index 6841a4bce186..5a2371484a58 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c @@ -83,15 +83,15 @@ #define PCIE_BUS_CLK 10000 #define TCLK (PCIE_BUS_CLK / 10) -static struct profile_mode_setting smu7_profiling[7] = - {{0, 0, 0, 0, 0, 0, 0, 0}, +static struct profile_mode_setting smu7_profiling[7] = { + {0, 0, 0, 0, 0, 0, 0, 0}, {1, 0, 100, 30, 1, 0, 100, 10}, {1, 10, 0, 30, 0, 0, 0, 0}, {0, 0, 0, 0, 1, 10, 16, 31}, {1, 0, 11, 50, 1, 0, 100, 10}, {1, 0, 5, 30, 0, 0, 0, 0}, {0, 0, 0, 0, 0, 0, 0, 0}, - }; +}; #define PPSMC_MSG_SetVBITimeout_VEGAM ((uint16_t) 0x310) @@ -904,7 +904,7 @@ static int smu7_setup_dpm_tables_v1(struct pp_hwmgr *hwmgr) dep_sclk_table->entries[i].clk; data->dpm_table.sclk_table.dpm_levels[data->dpm_table.sclk_table.count].enabled = - (i == 0) ? true : false; + i == 0; data->dpm_table.sclk_table.count++; } } @@ -919,7 +919,7 @@ static int smu7_setup_dpm_tables_v1(struct pp_hwmgr *hwmgr) data->dpm_table.mclk_table.dpm_levels[data->dpm_table.mclk_table.count].value = dep_mclk_table->entries[i].clk; data->dpm_table.mclk_table.dpm_levels[data->dpm_table.mclk_table.count].enabled = - (i == 0) ? true : false; + i == 0; data->dpm_table.mclk_table.count++; } } @@ -950,7 +950,7 @@ static int smu7_odn_initial_default_setting(struct pp_hwmgr *hwmgr) odn_table->odn_core_clock_dpm_levels.num_of_pl = data->golden_dpm_table.sclk_table.count; entries = odn_table->odn_core_clock_dpm_levels.entries; - for (i=0; i<data->golden_dpm_table.sclk_table.count; i++) { + for (i = 0; i < data->golden_dpm_table.sclk_table.count; i++) { entries[i].clock = data->golden_dpm_table.sclk_table.dpm_levels[i].value; entries[i].enabled = true; entries[i].vddc = dep_sclk_table->entries[i].vddc; @@ -962,7 +962,7 @@ static int smu7_odn_initial_default_setting(struct pp_hwmgr *hwmgr) odn_table->odn_memory_clock_dpm_levels.num_of_pl = data->golden_dpm_table.mclk_table.count; entries = odn_table->odn_memory_clock_dpm_levels.entries; - for (i=0; i<data->golden_dpm_table.mclk_table.count; i++) { + for (i = 0; i < data->golden_dpm_table.mclk_table.count; i++) { entries[i].clock = data->golden_dpm_table.mclk_table.dpm_levels[i].value; entries[i].enabled = true; entries[i].vddc = dep_mclk_table->entries[i].vddc; @@ -1798,17 +1798,6 @@ static int smu7_disable_dpm_tasks(struct pp_hwmgr *hwmgr) return result; } -static bool intel_core_rkl_chk(void) -{ -#if IS_ENABLED(CONFIG_X86_64) - struct cpuinfo_x86 *c = &cpu_data(0); - - return (c->x86 == 6 && c->x86_model == INTEL_FAM6_ROCKETLAKE); -#else - return false; -#endif -} - static void smu7_init_dpm_defaults(struct pp_hwmgr *hwmgr) { struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); @@ -1824,18 +1813,19 @@ static void smu7_init_dpm_defaults(struct pp_hwmgr *hwmgr) data->static_screen_threshold = SMU7_STATICSCREENTHRESHOLD_DFLT; data->static_screen_threshold_unit = SMU7_STATICSCREENTHRESHOLDUNIT_DFLT; data->voting_rights_clients[0] = SMU7_VOTINGRIGHTSCLIENTS_DFLT0; - data->voting_rights_clients[1]= SMU7_VOTINGRIGHTSCLIENTS_DFLT1; + data->voting_rights_clients[1] = SMU7_VOTINGRIGHTSCLIENTS_DFLT1; data->voting_rights_clients[2] = SMU7_VOTINGRIGHTSCLIENTS_DFLT2; - data->voting_rights_clients[3]= SMU7_VOTINGRIGHTSCLIENTS_DFLT3; - data->voting_rights_clients[4]= SMU7_VOTINGRIGHTSCLIENTS_DFLT4; - data->voting_rights_clients[5]= SMU7_VOTINGRIGHTSCLIENTS_DFLT5; - data->voting_rights_clients[6]= SMU7_VOTINGRIGHTSCLIENTS_DFLT6; - data->voting_rights_clients[7]= SMU7_VOTINGRIGHTSCLIENTS_DFLT7; + data->voting_rights_clients[3] = SMU7_VOTINGRIGHTSCLIENTS_DFLT3; + data->voting_rights_clients[4] = SMU7_VOTINGRIGHTSCLIENTS_DFLT4; + data->voting_rights_clients[5] = SMU7_VOTINGRIGHTSCLIENTS_DFLT5; + data->voting_rights_clients[6] = SMU7_VOTINGRIGHTSCLIENTS_DFLT6; + data->voting_rights_clients[7] = SMU7_VOTINGRIGHTSCLIENTS_DFLT7; data->mclk_dpm_key_disabled = hwmgr->feature_mask & PP_MCLK_DPM_MASK ? false : true; data->sclk_dpm_key_disabled = hwmgr->feature_mask & PP_SCLK_DPM_MASK ? false : true; data->pcie_dpm_key_disabled = - intel_core_rkl_chk() || !(hwmgr->feature_mask & PP_PCIE_DPM_MASK); + !amdgpu_device_pcie_dynamic_switching_supported() || + !(hwmgr->feature_mask & PP_PCIE_DPM_MASK); /* need to set voltage control types before EVV patching */ data->voltage_control = SMU7_VOLTAGE_CONTROL_NONE; data->vddci_control = SMU7_VOLTAGE_CONTROL_NONE; @@ -2012,7 +2002,7 @@ static int smu7_calculate_ro_range(struct pp_hwmgr *hwmgr) } else if (ASICID_IS_P21(adev->pdev->device, adev->pdev->revision) || ASICID_IS_P31(adev->pdev->device, adev->pdev->revision)) { min = 900; - max= 2100; + max = 2100; } else if (hwmgr->chip_id == CHIP_POLARIS10) { if (adev->pdev->subsystem_vendor == 0x106B) { min = 1000; @@ -4028,7 +4018,7 @@ static int smu7_read_sensor(struct pp_hwmgr *hwmgr, int idx, offset = data->soft_regs_start + smum_get_offsetof(hwmgr, SMU_SoftRegisters, (idx == AMDGPU_PP_SENSOR_GPU_LOAD) ? - AverageGraphicsActivity: + AverageGraphicsActivity : AverageMemoryActivity); activity_percent = cgs_read_ind_register(hwmgr->device, CGS_IND_REG__SMC, offset); @@ -4049,7 +4039,7 @@ static int smu7_read_sensor(struct pp_hwmgr *hwmgr, int idx, *((uint32_t *)value) = data->vce_power_gated ? 0 : 1; *size = 4; return 0; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: return smu7_get_gpu_power(hwmgr, (uint32_t *)value); case AMDGPU_PP_SENSOR_VDDGFX: if ((data->vr_config & VRCONF_VDDGFX_MASK) == diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c index 21be23ec3c79..65001bed0a9a 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_powertune.c @@ -520,8 +520,7 @@ static const struct gpu_pt_config_reg DIDTConfig_Polaris12[] = { { 0xFFFFFFFF } }; -static const struct gpu_pt_config_reg DIDTConfig_Polaris11_Kicker[] = -{ +static const struct gpu_pt_config_reg DIDTConfig_Polaris11_Kicker[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value Type * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -646,7 +645,7 @@ static const struct gpu_pt_config_reg DIDTConfig_Polaris11_Kicker[] = { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_STALL_CTRL_ENABLE_MASK, DIDT_TCP_STALL_CTRL__DIDT_STALL_CTRL_ENABLE__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_HI_MASK, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_HI__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_LO_MASK, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_LO__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, - { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD_MASK, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD__SHIFT,0x01aa, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD_MASK, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD__SHIFT, 0x01aa, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__UNUSED_0_MASK, DIDT_TCP_STALL_CTRL__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_TUNING_CTRL, DIDT_TCP_TUNING_CTRL__DIDT_TUNING_ENABLE_MASK, DIDT_TCP_TUNING_CTRL__DIDT_TUNING_ENABLE__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, @@ -666,8 +665,7 @@ static const struct gpu_pt_config_reg DIDTConfig_Polaris11_Kicker[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct gpu_pt_config_reg GCCACConfig_VegaM[] = -{ +static const struct gpu_pt_config_reg GCCACConfig_VegaM[] = { // --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- // Offset Mask Shift Value Type // --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -703,8 +701,7 @@ static const struct gpu_pt_config_reg GCCACConfig_VegaM[] = { 0xFFFFFFFF } // End of list }; -static const struct gpu_pt_config_reg DIDTConfig_VegaM[] = -{ +static const struct gpu_pt_config_reg DIDTConfig_VegaM[] = { // --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- // Offset Mask Shift Value Type // --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -831,7 +828,7 @@ static const struct gpu_pt_config_reg DIDTConfig_VegaM[] = { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_STALL_CTRL_ENABLE_MASK, DIDT_TCP_STALL_CTRL__DIDT_STALL_CTRL_ENABLE__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_HI_MASK, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_HI__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_LO_MASK, DIDT_TCP_STALL_CTRL__DIDT_STALL_DELAY_LO__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, - { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD_MASK, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD__SHIFT,0x01aa, GPU_CONFIGREG_DIDT_IND }, + { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD_MASK, DIDT_TCP_STALL_CTRL__DIDT_HI_POWER_THRESHOLD__SHIFT, 0x01aa, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_STALL_CTRL, DIDT_TCP_STALL_CTRL__UNUSED_0_MASK, DIDT_TCP_STALL_CTRL__UNUSED_0__SHIFT, 0x0000, GPU_CONFIGREG_DIDT_IND }, { ixDIDT_TCP_TUNING_CTRL, DIDT_TCP_TUNING_CTRL__DIDT_TUNING_ENABLE_MASK, DIDT_TCP_TUNING_CTRL__DIDT_TUNING_ENABLE__SHIFT, 0x0001, GPU_CONFIGREG_DIDT_IND }, @@ -1103,7 +1100,7 @@ int smu7_enable_smc_cac(struct pp_hwmgr *hwmgr) PP_ASSERT_WITH_CODE((0 == smc_result), "Failed to enable CAC in SMC.", result = -1); - data->cac_enabled = (0 == smc_result) ? true : false; + data->cac_enabled = smc_result == 0; } return result; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c index d0b1ab6c4523..79a566f3564a 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.c @@ -696,7 +696,7 @@ int smu_get_voltage_dependency_table_ppt_v1( return -EINVAL); dep_table->count = allowed_dep_table->count; - for (i=0; i<dep_table->count; i++) { + for (i = 0; i < dep_table->count; i++) { dep_table->entries[i].clk = allowed_dep_table->entries[i].clk; dep_table->entries[i].vddInd = allowed_dep_table->entries[i].vddInd; dep_table->entries[i].vdd_offset = allowed_dep_table->entries[i].vdd_offset; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.h index 2a75da1e9f03..83b3c9315143 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu_helper.h @@ -194,7 +194,7 @@ int smu_set_watermarks_for_clocks_ranges(void *wt_table, #define PHM_WAIT_INDIRECT_FIELD_UNEQUAL(hwmgr, port, reg, field, fieldval) \ PHM_WAIT_INDIRECT_REGISTER_UNEQUAL(hwmgr, port, reg, \ (fieldval) << PHM_FIELD_SHIFT(reg, field), \ - PHM_FIELD_MASK(reg, field) ) + PHM_FIELD_MASK(reg, field)) #define PHM_WAIT_VFPF_INDIRECT_REGISTER_UNEQUAL_GIVEN_INDEX(hwmgr, \ diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.c index ea743bea8e29..432d4fd2a0ba 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/tonga_baco.c @@ -36,8 +36,7 @@ #include "smu/smu_7_1_2_sh_mask.h" -static const struct baco_cmd_entry gpio_tbl[] = -{ +static const struct baco_cmd_entry gpio_tbl[] = { { CMD_WRITE, mmGPIOPAD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PU_EN, 0, 0, 0, 0x0 }, @@ -50,15 +49,13 @@ static const struct baco_cmd_entry gpio_tbl[] = { CMD_READMODIFYWRITE, mmDC_GPIO_SYNCA_MASK, 0, 0, 0, 0x00001111 } }; -static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = -{ +static const struct baco_cmd_entry enable_fb_req_rej_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, 0xC0300024 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, 0x1, 0x0, 0, 0x1 }, { CMD_WRITE, mmBIF_FB_EN, 0, 0, 0, 0x0 } }; -static const struct baco_cmd_entry use_bclk_tbl[] = -{ +static const struct baco_cmd_entry use_bclk_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN_MASK, CG_SPLL_FUNC_CNTL__SPLL_BYPASS_EN__SHIFT, 0, 0x1 }, { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL_2 }, @@ -80,8 +77,7 @@ static const struct baco_cmd_entry use_bclk_tbl[] = { CMD_READMODIFYWRITE, mmMPLL_CNTL_MODE, MPLL_CNTL_MODE__MPLL_MCLK_SEL_MASK, MPLL_CNTL_MODE__MPLL_MCLK_SEL__SHIFT, 0, 0x0 } }; -static const struct baco_cmd_entry turn_off_plls_tbl[] = -{ +static const struct baco_cmd_entry turn_off_plls_tbl[] = { { CMD_WRITE, mmGCK_SMC_IND_INDEX, 0, 0, 0, ixCG_SPLL_FUNC_CNTL }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_RESET_MASK, CG_SPLL_FUNC_CNTL__SPLL_RESET__SHIFT, 0, 0x1 }, { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, CG_SPLL_FUNC_CNTL__SPLL_PWRON_MASK, CG_SPLL_FUNC_CNTL__SPLL_PWRON__SHIFT, 0, 0x0 }, @@ -112,8 +108,7 @@ static const struct baco_cmd_entry turn_off_plls_tbl[] = { CMD_READMODIFYWRITE, mmGCK_SMC_IND_DATA, THM_CLK_CNTL__TMON_CLK_SEL_MASK, THM_CLK_CNTL__TMON_CLK_SEL__SHIFT, 0, 0x1 } }; -static const struct baco_cmd_entry enter_baco_tbl[] = -{ +static const struct baco_cmd_entry enter_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BIF_SCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_SCLK_SWITCH__SHIFT, 0, 0x01 }, { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_BIF_SCLK_SWITCH_MASK, 0, 5, 0x40000 }, @@ -130,8 +125,7 @@ static const struct baco_cmd_entry enter_baco_tbl[] = #define BACO_CNTL__PWRGOOD_MASK BACO_CNTL__PWRGOOD_GPIO_MASK+BACO_CNTL__PWRGOOD_MEM_MASK+BACO_CNTL__PWRGOOD_DVO_MASK -static const struct baco_cmd_entry exit_baco_tbl[] = -{ +static const struct baco_cmd_entry exit_baco_tbl[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_RESET_EN_MASK, BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, BACO_CNTL__BACO_BCLK_OFF__SHIFT, 0, 0x00 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0x00 }, @@ -146,22 +140,19 @@ static const struct baco_cmd_entry exit_baco_tbl[] = { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0x00 } }; -static const struct baco_cmd_entry clean_baco_tbl[] = -{ +static const struct baco_cmd_entry clean_baco_tbl[] = { { CMD_WRITE, mmBIOS_SCRATCH_6, 0, 0, 0, 0 }, { CMD_WRITE, mmBIOS_SCRATCH_7, 0, 0, 0, 0 } }; -static const struct baco_cmd_entry gpio_tbl_iceland[] = -{ +static const struct baco_cmd_entry gpio_tbl_iceland[] = { { CMD_WRITE, mmGPIOPAD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PD_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_PU_EN, 0, 0, 0, 0x0 }, { CMD_WRITE, mmGPIOPAD_MASK, 0, 0, 0, 0xff77ffff } }; -static const struct baco_cmd_entry exit_baco_tbl_iceland[] = -{ +static const struct baco_cmd_entry exit_baco_tbl_iceland[] = { { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_RESET_EN_MASK, BACO_CNTL__BACO_RESET_EN__SHIFT, 0, 0x01 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_BCLK_OFF_MASK, BACO_CNTL__BACO_BCLK_OFF__SHIFT, 0, 0x00 }, { CMD_READMODIFYWRITE, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0x00 }, @@ -177,8 +168,7 @@ static const struct baco_cmd_entry exit_baco_tbl_iceland[] = { CMD_WAITFOR, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0x00 } }; -static const struct baco_cmd_entry clean_baco_tbl_iceland[] = -{ +static const struct baco_cmd_entry clean_baco_tbl_iceland[] = { { CMD_WRITE, mmBIOS_SCRATCH_7, 0, 0, 0, 0 } }; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.c index 46bb16c29cf6..6836e98d37be 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_baco.c @@ -31,24 +31,22 @@ -static const struct soc15_baco_cmd_entry pre_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry pre_baco_tbl[] = { {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIF_DOORBELL_CNTL), BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN_MASK, BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN__SHIFT, 0, 1}, {CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIF_FB_EN), 0, 0, 0, 0}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_DSTATE_BYPASS_MASK, BACO_CNTL__BACO_DSTATE_BYPASS__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_RST_INTR_MASK_MASK, BACO_CNTL__BACO_RST_INTR_MASK__SHIFT, 0, 1} }; -static const struct soc15_baco_cmd_entry enter_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry enter_baco_tbl[] = { {CMD_WAITFOR, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__SOC_DOMAIN_IDLE_MASK, THM_BACO_CNTL__SOC_DOMAIN_IDLE__SHIFT, 0xffffffff, 0x80000000}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_DUMMY_EN_MASK, BACO_CNTL__BACO_DUMMY_EN__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_VDCI_RESET_MASK, THM_BACO_CNTL__BACO_SOC_VDCI_RESET__SHIFT, 0, 1}, - {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT,0, 1}, + {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ISO_EN_MASK, THM_BACO_CNTL__BACO_ISO_EN__SHIFT, 0, 1}, - {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT,0, 1}, + {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ANA_ISO_EN_MASK, THM_BACO_CNTL__BACO_ANA_ISO_EN__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 1}, @@ -58,13 +56,12 @@ static const struct soc15_baco_cmd_entry enter_baco_tbl[] = {CMD_WAITFOR, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_MODE_MASK, BACO_CNTL__BACO_MODE__SHIFT, 0xffffffff, 0x100} }; -static const struct soc15_baco_cmd_entry exit_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry exit_baco_tbl[] = { {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0}, - {CMD_DELAY_MS, 0, 0, 0, 0, 0, 0, 10,0}, - {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0,0}, + {CMD_DELAY_MS, 0, 0, 0, 0, 0, 0, 10, 0}, + {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0, 0}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ANA_ISO_EN_MASK, THM_BACO_CNTL__BACO_ANA_ISO_EN__SHIFT, 0, 0}, - {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT,0, 0}, + {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_AEB_ISO_EN_MASK, THM_BACO_CNTL__BACO_AEB_ISO_EN__SHIFT, 0, 0}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_ISO_EN_MASK, THM_BACO_CNTL__BACO_ISO_EN__SHIFT, 0, 0}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_PWROKRAW_CNTL_MASK, THM_BACO_CNTL__BACO_PWROKRAW_CNTL__SHIFT, 0, 1}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SMNCLK_MUX_MASK, THM_BACO_CNTL__BACO_SMNCLK_MUX__SHIFT, 0, 0}, @@ -74,13 +71,12 @@ static const struct soc15_baco_cmd_entry exit_baco_tbl[] = {CMD_WAITFOR, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_EXIT_MASK, 0, 0xffffffff, 0}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(THM, 0, mmTHM_BACO_CNTL), THM_BACO_CNTL__BACO_SB_AXI_FENCE_MASK, THM_BACO_CNTL__BACO_SB_AXI_FENCE__SHIFT, 0, 0}, {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_DUMMY_EN_MASK, BACO_CNTL__BACO_DUMMY_EN__SHIFT, 0, 0}, - {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK ,BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 0}, - {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_EN_MASK , BACO_CNTL__BACO_EN__SHIFT, 0,0}, + {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 0}, + {CMD_READMODIFYWRITE, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 0}, {CMD_WAITFOR, SOC15_REG_ENTRY(NBIF, 0, mmBACO_CNTL), BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0} }; -static const struct soc15_baco_cmd_entry clean_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry clean_baco_tbl[] = { {CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_6), 0, 0, 0, 0}, {CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_7), 0, 0, 0, 0}, }; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c index c51dd4c74fe9..6d6bc6a380b3 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c @@ -1375,8 +1375,7 @@ static int vega10_setup_default_dpm_tables(struct pp_hwmgr *hwmgr) dep_mm_table->entries[i].eclk) { dpm_table->dpm_levels[dpm_table->count].value = dep_mm_table->entries[i].eclk; - dpm_table->dpm_levels[dpm_table->count].enabled = - (i == 0) ? true : false; + dpm_table->dpm_levels[dpm_table->count].enabled = i == 0; dpm_table->count++; } } @@ -1391,8 +1390,7 @@ static int vega10_setup_default_dpm_tables(struct pp_hwmgr *hwmgr) dep_mm_table->entries[i].vclk) { dpm_table->dpm_levels[dpm_table->count].value = dep_mm_table->entries[i].vclk; - dpm_table->dpm_levels[dpm_table->count].enabled = - (i == 0) ? true : false; + dpm_table->dpm_levels[dpm_table->count].enabled = i == 0; dpm_table->count++; } } @@ -1405,8 +1403,7 @@ static int vega10_setup_default_dpm_tables(struct pp_hwmgr *hwmgr) dep_mm_table->entries[i].dclk) { dpm_table->dpm_levels[dpm_table->count].value = dep_mm_table->entries[i].dclk; - dpm_table->dpm_levels[dpm_table->count].enabled = - (i == 0) ? true : false; + dpm_table->dpm_levels[dpm_table->count].enabled = i == 0; dpm_table->count++; } } @@ -3969,7 +3966,7 @@ static int vega10_read_sensor(struct pp_hwmgr *hwmgr, int idx, *((uint32_t *)value) = data->vce_power_gated ? 0 : 1; *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: ret = vega10_get_gpu_power(hwmgr, (uint32_t *)value); break; case AMDGPU_PP_SENSOR_VDDGFX: diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c index 309a9d3bc1b7..3007b054c873 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_powertune.c @@ -30,8 +30,7 @@ #include "pp_debug.h" #include "soc15_common.h" -static const struct vega10_didt_config_reg SEDiDtTuningCtrlConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtTuningCtrlConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -55,8 +54,7 @@ static const struct vega10_didt_config_reg SEDiDtTuningCtrlConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEDiDtCtrl3Config_vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtCtrl3Config_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -120,8 +118,7 @@ static const struct vega10_didt_config_reg SEDiDtCtrl3Config_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEDiDtCtrl2Config_Vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtCtrl2Config_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -149,8 +146,7 @@ static const struct vega10_didt_config_reg SEDiDtCtrl2Config_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEDiDtCtrl1Config_Vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtCtrl1Config_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -172,8 +168,7 @@ static const struct vega10_didt_config_reg SEDiDtCtrl1Config_Vega10[] = }; -static const struct vega10_didt_config_reg SEDiDtWeightConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtWeightConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -201,8 +196,7 @@ static const struct vega10_didt_config_reg SEDiDtWeightConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEDiDtCtrl0Config_Vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtCtrl0Config_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -260,8 +254,7 @@ static const struct vega10_didt_config_reg SEDiDtCtrl0Config_Vega10[] = }; -static const struct vega10_didt_config_reg SEDiDtStallCtrlConfig_vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtStallCtrlConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -293,8 +286,7 @@ static const struct vega10_didt_config_reg SEDiDtStallCtrlConfig_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEDiDtStallPatternConfig_vega10[] = -{ +static const struct vega10_didt_config_reg SEDiDtStallPatternConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -362,8 +354,7 @@ static const struct vega10_didt_config_reg SEDiDtStallPatternConfig_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SELCacConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SELCacConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -401,8 +392,7 @@ static const struct vega10_didt_config_reg SELCacConfig_Vega10[] = }; -static const struct vega10_didt_config_reg SEEDCStallPatternConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCStallPatternConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -431,8 +421,7 @@ static const struct vega10_didt_config_reg SEEDCStallPatternConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEEDCForceStallPatternConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCForceStallPatternConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -451,8 +440,7 @@ static const struct vega10_didt_config_reg SEEDCForceStallPatternConfig_Vega10[] { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEEDCStallDelayConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCStallDelayConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -478,8 +466,7 @@ static const struct vega10_didt_config_reg SEEDCStallDelayConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEEDCThresholdConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCThresholdConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -492,8 +479,7 @@ static const struct vega10_didt_config_reg SEEDCThresholdConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEEDCCtrlResetConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCCtrlResetConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -514,8 +500,7 @@ static const struct vega10_didt_config_reg SEEDCCtrlResetConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEEDCCtrlConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCCtrlConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -536,8 +521,7 @@ static const struct vega10_didt_config_reg SEEDCCtrlConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg SEEDCCtrlForceStallConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg SEEDCCtrlForceStallConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -571,8 +555,7 @@ static const struct vega10_didt_config_reg SEEDCCtrlForceStallConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg GCDiDtDroopCtrlConfig_vega10[] = -{ +static const struct vega10_didt_config_reg GCDiDtDroopCtrlConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -586,8 +569,7 @@ static const struct vega10_didt_config_reg GCDiDtDroopCtrlConfig_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg GCDiDtCtrl0Config_vega10[] = -{ +static const struct vega10_didt_config_reg GCDiDtCtrl0Config_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -601,8 +583,7 @@ static const struct vega10_didt_config_reg GCDiDtCtrl0Config_vega10[] = }; -static const struct vega10_didt_config_reg PSMSEEDCStallPatternConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg PSMSEEDCStallPatternConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -621,8 +602,7 @@ static const struct vega10_didt_config_reg PSMSEEDCStallPatternConfig_Vega10[] { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMSEEDCStallDelayConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg PSMSEEDCStallDelayConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -651,8 +631,7 @@ static const struct vega10_didt_config_reg PSMSEEDCStallDelayConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMSEEDCCtrlResetConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg PSMSEEDCCtrlResetConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -673,8 +652,7 @@ static const struct vega10_didt_config_reg PSMSEEDCCtrlResetConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMSEEDCCtrlConfig_Vega10[] = -{ +static const struct vega10_didt_config_reg PSMSEEDCCtrlConfig_Vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -695,8 +673,7 @@ static const struct vega10_didt_config_reg PSMSEEDCCtrlConfig_Vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMGCEDCDroopCtrlConfig_vega10[] = -{ +static const struct vega10_didt_config_reg PSMGCEDCDroopCtrlConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -710,8 +687,7 @@ static const struct vega10_didt_config_reg PSMGCEDCDroopCtrlConfig_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMGCEDCCtrlResetConfig_vega10[] = -{ +static const struct vega10_didt_config_reg PSMGCEDCCtrlResetConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -726,8 +702,7 @@ static const struct vega10_didt_config_reg PSMGCEDCCtrlResetConfig_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg PSMGCEDCCtrlConfig_vega10[] = -{ +static const struct vega10_didt_config_reg PSMGCEDCCtrlConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -742,8 +717,7 @@ static const struct vega10_didt_config_reg PSMGCEDCCtrlConfig_vega10[] = { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg AvfsPSMResetConfig_vega10[]= -{ +static const struct vega10_didt_config_reg AvfsPSMResetConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -756,8 +730,7 @@ static const struct vega10_didt_config_reg AvfsPSMResetConfig_vega10[]= { 0xFFFFFFFF } /* End of list */ }; -static const struct vega10_didt_config_reg AvfsPSMInitConfig_vega10[] = -{ +static const struct vega10_didt_config_reg AvfsPSMInitConfig_vega10[] = { /* --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- * Offset Mask Shift Value * --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- @@ -919,7 +892,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr) mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { - data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | ( count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); + data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | (count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, data); result = vega10_program_didt_config_registers(hwmgr, SEDiDtStallCtrlConfig_vega10, VEGA10_CONFIGREG_DIDT); @@ -970,7 +943,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr) mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { - data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | ( count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); + data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | (count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, data); result = vega10_program_didt_config_registers(hwmgr, SEDiDtStallCtrlConfig_vega10, VEGA10_CONFIGREG_DIDT); @@ -1031,7 +1004,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr) mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { - data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | ( count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); + data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | (count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, data); result = vega10_program_didt_config_registers(hwmgr, SEDiDtWeightConfig_Vega10, VEGA10_CONFIGREG_DIDT); result |= vega10_program_didt_config_registers(hwmgr, SEEDCStallPatternConfig_Vega10, VEGA10_CONFIGREG_DIDT); @@ -1081,7 +1054,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr) mutex_lock(&adev->grbm_idx_mutex); for (count = 0; count < num_se; count++) { - data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | ( count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); + data = GRBM_GFX_INDEX__INSTANCE_BROADCAST_WRITES_MASK | GRBM_GFX_INDEX__SH_BROADCAST_WRITES_MASK | (count << GRBM_GFX_INDEX__SE_INDEX__SHIFT); WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, data); result = vega10_program_didt_config_registers(hwmgr, PSMSEEDCStallPatternConfig_Vega10, VEGA10_CONFIGREG_DIDT); result |= vega10_program_didt_config_registers(hwmgr, PSMSEEDCStallDelayConfig_Vega10, VEGA10_CONFIGREG_DIDT); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_pptable.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_pptable.h index 9c479bd9a786..8b0590b834cc 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_pptable.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_pptable.h @@ -317,16 +317,14 @@ typedef struct _ATOM_Vega10_Thermal_Controller { UCHAR ucFlags; /* to be defined */ } ATOM_Vega10_Thermal_Controller; -typedef struct _ATOM_Vega10_VCE_State_Record -{ +typedef struct _ATOM_Vega10_VCE_State_Record { UCHAR ucVCEClockIndex; /*index into usVCEDependencyTableOffset of 'ATOM_Vega10_MM_Dependency_Table' type */ UCHAR ucFlag; /* 2 bits indicates memory p-states */ UCHAR ucSCLKIndex; /* index into ATOM_Vega10_SCLK_Dependency_Table */ UCHAR ucMCLKIndex; /* index into ATOM_Vega10_MCLK_Dependency_Table */ } ATOM_Vega10_VCE_State_Record; -typedef struct _ATOM_Vega10_VCE_State_Table -{ +typedef struct _ATOM_Vega10_VCE_State_Table { UCHAR ucRevId; UCHAR ucNumEntries; ATOM_Vega10_VCE_State_Record entries[1]; @@ -361,8 +359,7 @@ typedef struct _ATOM_Vega10_PowerTune_Table { USHORT usTemperatureLimitTedge; } ATOM_Vega10_PowerTune_Table; -typedef struct _ATOM_Vega10_PowerTune_Table_V2 -{ +typedef struct _ATOM_Vega10_PowerTune_Table_V2 { UCHAR ucRevId; USHORT usSocketPowerLimit; USHORT usBatteryPowerLimit; @@ -388,8 +385,7 @@ typedef struct _ATOM_Vega10_PowerTune_Table_V2 USHORT usTemperatureLimitTedge; } ATOM_Vega10_PowerTune_Table_V2; -typedef struct _ATOM_Vega10_PowerTune_Table_V3 -{ +typedef struct _ATOM_Vega10_PowerTune_Table_V3 { UCHAR ucRevId; USHORT usSocketPowerLimit; USHORT usBatteryPowerLimit; @@ -428,15 +424,13 @@ typedef struct _ATOM_Vega10_Hard_Limit_Record { USHORT usVddMemLimit; } ATOM_Vega10_Hard_Limit_Record; -typedef struct _ATOM_Vega10_Hard_Limit_Table -{ +typedef struct _ATOM_Vega10_Hard_Limit_Table { UCHAR ucRevId; UCHAR ucNumEntries; ATOM_Vega10_Hard_Limit_Record entries[1]; } ATOM_Vega10_Hard_Limit_Table; -typedef struct _Vega10_PPTable_Generic_SubTable_Header -{ +typedef struct _Vega10_PPTable_Generic_SubTable_Header { UCHAR ucRevId; } Vega10_PPTable_Generic_SubTable_Header; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.c index bb90d8abf79b..3be616af327e 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_processpptables.c @@ -372,9 +372,9 @@ static int get_mm_clock_voltage_table( return 0; } -static void get_scl_sda_value(uint8_t line, uint8_t *scl, uint8_t* sda) +static void get_scl_sda_value(uint8_t line, uint8_t *scl, uint8_t *sda) { - switch(line){ + switch (line) { case Vega10_I2CLineID_DDC1: *scl = Vega10_I2C_DDC1CLK; *sda = Vega10_I2C_DDC1DATA; @@ -954,7 +954,7 @@ static int init_powerplay_extended_tables( if (!result && powerplay_table->usPixclkDependencyTableOffset) result = get_pix_clk_voltage_dependency_table(hwmgr, &pp_table_info->vdd_dep_on_pixclk, - (const ATOM_Vega10_PIXCLK_Dependency_Table*) + (const ATOM_Vega10_PIXCLK_Dependency_Table *) pixclk_dep_table); if (!result && powerplay_table->usPhyClkDependencyTableOffset) diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.c index bc53cce4f32d..32cc8de296e4 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_baco.c @@ -29,16 +29,14 @@ #include "vega12_ppsmc.h" #include "vega12_baco.h" -static const struct soc15_baco_cmd_entry pre_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry pre_baco_tbl[] = { { CMD_READMODIFYWRITE, NBIF_HWID, 0, mmBIF_DOORBELL_CNTL_BASE_IDX, mmBIF_DOORBELL_CNTL, BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN_MASK, BIF_DOORBELL_CNTL__DOORBELL_MONITOR_EN__SHIFT, 0, 0 }, { CMD_WRITE, NBIF_HWID, 0, mmBIF_FB_EN_BASE_IDX, mmBIF_FB_EN, 0, 0, 0, 0 }, { CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_DSTATE_BYPASS_MASK, BACO_CNTL__BACO_DSTATE_BYPASS__SHIFT, 0, 1 }, { CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_RST_INTR_MASK_MASK, BACO_CNTL__BACO_RST_INTR_MASK__SHIFT, 0, 1 } }; -static const struct soc15_baco_cmd_entry enter_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry enter_baco_tbl[] = { { CMD_WAITFOR, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__SOC_DOMAIN_IDLE_MASK, THM_BACO_CNTL__SOC_DOMAIN_IDLE__SHIFT, 0xffffffff, 0x80000000 }, { CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_EN_MASK, BACO_CNTL__BACO_EN__SHIFT, 0, 1 }, { CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_BIF_LCLK_SWITCH_MASK, BACO_CNTL__BACO_BIF_LCLK_SWITCH__SHIFT, 0, 1 }, @@ -56,8 +54,7 @@ static const struct soc15_baco_cmd_entry enter_baco_tbl[] = { CMD_WAITFOR, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, BACO_CNTL__BACO_MODE__SHIFT, 0xffffffff, 0x100 } }; -static const struct soc15_baco_cmd_entry exit_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry exit_baco_tbl[] = { { CMD_READMODIFYWRITE, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_POWER_OFF_MASK, BACO_CNTL__BACO_POWER_OFF__SHIFT, 0, 0 }, { CMD_DELAY_MS, 0, 0, 0, 0, 0, 0, 10, 0 }, { CMD_READMODIFYWRITE, THM_HWID, 0, mmTHM_BACO_CNTL_BASE_IDX, mmTHM_BACO_CNTL, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF_MASK, THM_BACO_CNTL__BACO_SOC_REFCLK_OFF__SHIFT, 0, 0 }, @@ -77,8 +74,7 @@ static const struct soc15_baco_cmd_entry exit_baco_tbl[] = { CMD_WAITFOR, NBIF_HWID, 0, mmRCC_BACO_CNTL_MISC_BASE_IDX, mmBACO_CNTL, BACO_CNTL__BACO_MODE_MASK, 0, 0xffffffff, 0 } }; -static const struct soc15_baco_cmd_entry clean_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry clean_baco_tbl[] = { { CMD_WRITE, NBIF_HWID, 0, mmBIOS_SCRATCH_6_BASE_IDX, mmBIOS_SCRATCH_6, 0, 0, 0, 0 }, { CMD_WRITE, NBIF_HWID, 0, mmBIOS_SCRATCH_7_BASE_IDX, mmBIOS_SCRATCH_7, 0, 0, 0, 0 } }; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.c index 1937be1cf5b4..460067933de2 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.c @@ -1529,7 +1529,7 @@ static int vega12_read_sensor(struct pp_hwmgr *hwmgr, int idx, *((uint32_t *)value) = data->vce_power_gated ? 0 : 1; *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: ret = vega12_get_gpu_power(hwmgr, (uint32_t *)value); if (!ret) *size = 4; @@ -1623,13 +1623,13 @@ static int vega12_notify_smc_display_config_after_ps_adjustment( if (data->smu_features[GNLD_DPM_DCEFCLK].supported) { clock_req.clock_type = amd_pp_dcef_clock; - clock_req.clock_freq_in_khz = min_clocks.dcefClock/10; + clock_req.clock_freq_in_khz = min_clocks.dcefClock / 10; if (!vega12_display_clock_voltage_request(hwmgr, &clock_req)) { if (data->smu_features[GNLD_DS_DCEFCLK].supported) PP_ASSERT_WITH_CODE( !smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, - min_clocks.dcefClockInSR /100, + min_clocks.dcefClockInSR / 100, NULL), "Attempt to set divider for DCEFCLK Failed!", return -1); @@ -2354,8 +2354,8 @@ static int vega12_apply_clocks_adjust_rules(struct pp_hwmgr *hwmgr) uint32_t i, latency; disable_mclk_switching = ((1 < hwmgr->display_config->num_display) && - !hwmgr->display_config->multi_monitor_in_sync) || - vblank_too_short; + !hwmgr->display_config->multi_monitor_in_sync) || + vblank_too_short; latency = hwmgr->display_config->dce_tolerable_mclk_in_active_latency; /* gfxclk */ @@ -2522,7 +2522,7 @@ static int vega12_set_uclk_to_highest_dpm_level(struct pp_hwmgr *hwmgr, dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level, + (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level, NULL)), "[SetUclkToHightestDpmLevel] Set hard min uclk failed!", return ret); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.h index aa63ae41942d..9f2ce4308548 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_hwmgr.h @@ -38,8 +38,7 @@ #define VG12_PSUEDO_NUM_DCEFCLK_DPM_LEVELS 8 #define VG12_PSUEDO_NUM_UCLK_DPM_LEVELS 4 -enum -{ +enum { GNLD_DPM_PREFETCHER = 0, GNLD_DPM_GFXCLK, GNLD_DPM_UCLK, diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_pptable.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_pptable.h index bf4f5095b80d..9b8435a4d306 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_pptable.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega12_pptable.h @@ -72,8 +72,7 @@ enum ATOM_VEGA12_PPCLOCK_ID { typedef enum ATOM_VEGA12_PPCLOCK_ID ATOM_VEGA12_PPCLOCK_ID; -typedef struct _ATOM_VEGA12_POWERPLAYTABLE -{ +typedef struct _ATOM_VEGA12_POWERPLAYTABLE { struct atom_common_table_header sHeader; UCHAR ucTableRevision; USHORT usTableSize; @@ -92,11 +91,11 @@ typedef struct _ATOM_VEGA12_POWERPLAYTABLE USHORT usODPowerSavePowerLimit; USHORT usSoftwareShutdownTemp; - ULONG PowerSavingClockMax [ATOM_VEGA12_PPCLOCK_COUNT]; - ULONG PowerSavingClockMin [ATOM_VEGA12_PPCLOCK_COUNT]; + ULONG PowerSavingClockMax[ATOM_VEGA12_PPCLOCK_COUNT]; + ULONG PowerSavingClockMin[ATOM_VEGA12_PPCLOCK_COUNT]; - ULONG ODSettingsMax [ATOM_VEGA12_ODSETTING_COUNT]; - ULONG ODSettingsMin [ATOM_VEGA12_ODSETTING_COUNT]; + ULONG ODSettingsMax[ATOM_VEGA12_ODSETTING_COUNT]; + ULONG ODSettingsMin[ATOM_VEGA12_ODSETTING_COUNT]; USHORT usReserve[5]; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.c index 8d99c7a5abf8..994c0d374bfa 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_baco.c @@ -31,8 +31,7 @@ #include "amdgpu_ras.h" -static const struct soc15_baco_cmd_entry clean_baco_tbl[] = -{ +static const struct soc15_baco_cmd_entry clean_baco_tbl[] = { {CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_6), 0, 0, 0, 0}, {CMD_WRITE, SOC15_REG_ENTRY(NBIF, 0, mmBIOS_SCRATCH_7), 0, 0, 0, 0}, }; @@ -90,11 +89,11 @@ int vega20_baco_set_state(struct pp_hwmgr *hwmgr, enum BACO_STATE state) data |= 0x80000000; WREG32_SOC15(THM, 0, mmTHM_BACO_CNTL, data); - if(smum_send_msg_to_smc_with_parameter(hwmgr, + if (smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 0, NULL)) return -EINVAL; } else { - if(smum_send_msg_to_smc_with_parameter(hwmgr, + if (smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_EnterBaco, 1, NULL)) return -EINVAL; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c index 4e19ccbdb807..3b33af30eb0f 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c @@ -1402,7 +1402,7 @@ static int vega20_od8_set_settings( "Failed to export over drive table!", return ret); - switch(index) { + switch (index) { case OD8_SETTING_GFXCLK_FMIN: od_table.GfxclkFmin = (uint16_t)value; break; @@ -2129,7 +2129,7 @@ static int vega20_get_metrics_table(struct pp_hwmgr *hwmgr, return ret; } -static int vega20_get_gpu_power(struct pp_hwmgr *hwmgr, +static int vega20_get_gpu_power(struct pp_hwmgr *hwmgr, int idx, uint32_t *query) { int ret = 0; @@ -2140,10 +2140,17 @@ static int vega20_get_gpu_power(struct pp_hwmgr *hwmgr, return ret; /* For the 40.46 release, they changed the value name */ - if (hwmgr->smu_version == 0x282e00) - *query = metrics_table.AverageSocketPower << 8; - else + switch (idx) { + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: + if (hwmgr->smu_version == 0x282e00) + *query = metrics_table.AverageSocketPower << 8; + else + ret = -EOPNOTSUPP; + break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: *query = metrics_table.CurrSocketPower << 8; + break; + } return ret; } @@ -2253,9 +2260,10 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx, *((uint32_t *)value) = data->vce_power_gated ? 0 : 1; *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: *size = 16; - ret = vega20_get_gpu_power(hwmgr, (uint32_t *)value); + ret = vega20_get_gpu_power(hwmgr, idx, (uint32_t *)value); break; case AMDGPU_PP_SENSOR_VDDGFX: val_vid = (RREG32_SOC15(SMUIO, 0, mmSMUSVI0_TEL_PLANE0) & @@ -2360,7 +2368,7 @@ static int vega20_notify_smc_display_config_after_ps_adjustment( dpm_table->dpm_state.hard_min_level = min_clocks.memoryClock / 100; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level, + (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level, NULL)), "[SetHardMinFreq] Set hard min uclk failed!", return ret); @@ -3579,7 +3587,7 @@ static int vega20_set_uclk_to_highest_dpm_level(struct pp_hwmgr *hwmgr, dpm_table->dpm_state.hard_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetHardMinByFreq, - (PPCLK_UCLK << 16 ) | dpm_table->dpm_state.hard_min_level, + (PPCLK_UCLK << 16) | dpm_table->dpm_state.hard_min_level, NULL)), "[SetUclkToHightestDpmLevel] Set hard min uclk failed!", return ret); @@ -3605,7 +3613,7 @@ static int vega20_set_fclk_to_highest_dpm_level(struct pp_hwmgr *hwmgr) dpm_table->dpm_state.soft_min_level = dpm_table->dpm_levels[dpm_table->count - 1].value; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetSoftMinByFreq, - (PPCLK_FCLK << 16 ) | dpm_table->dpm_state.soft_min_level, + (PPCLK_FCLK << 16) | dpm_table->dpm_state.soft_min_level, NULL)), "[SetFclkToHightestDpmLevel] Set soft min fclk failed!", return ret); @@ -3727,8 +3735,8 @@ static int vega20_apply_clocks_adjust_rules(struct pp_hwmgr *hwmgr) uint32_t i, latency; disable_mclk_switching = ((1 < hwmgr->display_config->num_display) && - !hwmgr->display_config->multi_monitor_in_sync) || - vblank_too_short; + !hwmgr->display_config->multi_monitor_in_sync) || + vblank_too_short; latency = hwmgr->display_config->dce_tolerable_mclk_in_active_latency; /* gfxclk */ diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.h index 075c0094da9c..1ba9b5fe2a5d 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.h @@ -385,8 +385,7 @@ struct vega20_odn_data { struct vega20_odn_temp_table odn_temp_table; }; -enum OD8_FEATURE_ID -{ +enum OD8_FEATURE_ID { OD8_GFXCLK_LIMITS = 1 << 0, OD8_GFXCLK_CURVE = 1 << 1, OD8_UCLK_MAX = 1 << 2, @@ -399,8 +398,7 @@ enum OD8_FEATURE_ID OD8_FAN_ZERO_RPM_CONTROL = 1 << 9 }; -enum OD8_SETTING_ID -{ +enum OD8_SETTING_ID { OD8_SETTING_GFXCLK_FMIN = 0, OD8_SETTING_GFXCLK_FMAX, OD8_SETTING_GFXCLK_FREQ1, diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_pptable.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_pptable.h index 2222e29405c6..b468dddbefff 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_pptable.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_pptable.h @@ -73,14 +73,13 @@ enum ATOM_VEGA20_ODSETTING_ID { }; typedef enum ATOM_VEGA20_ODSETTING_ID ATOM_VEGA20_ODSETTING_ID; -typedef struct _ATOM_VEGA20_OVERDRIVE8_RECORD -{ +typedef struct _ATOM_VEGA20_OVERDRIVE8_RECORD { UCHAR ucODTableRevision; ULONG ODFeatureCount; - UCHAR ODFeatureCapabilities [ATOM_VEGA20_ODFEATURE_MAX_COUNT]; //OD feature support flags + UCHAR ODFeatureCapabilities[ATOM_VEGA20_ODFEATURE_MAX_COUNT]; //OD feature support flags ULONG ODSettingCount; - ULONG ODSettingsMax [ATOM_VEGA20_ODSETTING_MAX_COUNT]; //Upper Limit for each OD Setting - ULONG ODSettingsMin [ATOM_VEGA20_ODSETTING_MAX_COUNT]; //Lower Limit for each OD Setting + ULONG ODSettingsMax[ATOM_VEGA20_ODSETTING_MAX_COUNT]; //Upper Limit for each OD Setting + ULONG ODSettingsMin[ATOM_VEGA20_ODSETTING_MAX_COUNT]; //Lower Limit for each OD Setting } ATOM_VEGA20_OVERDRIVE8_RECORD; enum ATOM_VEGA20_PPCLOCK_ID { @@ -99,16 +98,14 @@ enum ATOM_VEGA20_PPCLOCK_ID { }; typedef enum ATOM_VEGA20_PPCLOCK_ID ATOM_VEGA20_PPCLOCK_ID; -typedef struct _ATOM_VEGA20_POWER_SAVING_CLOCK_RECORD -{ +typedef struct _ATOM_VEGA20_POWER_SAVING_CLOCK_RECORD { UCHAR ucTableRevision; ULONG PowerSavingClockCount; // Count of PowerSavingClock Mode - ULONG PowerSavingClockMax [ATOM_VEGA20_PPCLOCK_MAX_COUNT]; // PowerSavingClock Mode Clock Maximum array In MHz - ULONG PowerSavingClockMin [ATOM_VEGA20_PPCLOCK_MAX_COUNT]; // PowerSavingClock Mode Clock Minimum array In MHz + ULONG PowerSavingClockMax[ATOM_VEGA20_PPCLOCK_MAX_COUNT]; // PowerSavingClock Mode Clock Maximum array In MHz + ULONG PowerSavingClockMin[ATOM_VEGA20_PPCLOCK_MAX_COUNT]; // PowerSavingClock Mode Clock Minimum array In MHz } ATOM_VEGA20_POWER_SAVING_CLOCK_RECORD; -typedef struct _ATOM_VEGA20_POWERPLAYTABLE -{ +typedef struct _ATOM_VEGA20_POWERPLAYTABLE { struct atom_common_table_header sHeader; UCHAR ucTableRevision; USHORT usTableSize; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/hardwaremanager.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hardwaremanager.h index 01a7d66864f2..f4f9a104d170 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/hardwaremanager.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hardwaremanager.h @@ -44,8 +44,7 @@ struct phm_fan_speed_info { }; /* Automatic Power State Throttling */ -enum PHM_AutoThrottleSource -{ +enum PHM_AutoThrottleSource { PHM_AutoThrottleSource_Thermal, PHM_AutoThrottleSource_External }; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h index 612d66aeaab9..81650727a5de 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h @@ -190,8 +190,7 @@ struct phm_vce_clock_voltage_dependency_table { }; -enum SMU_ASIC_RESET_MODE -{ +enum SMU_ASIC_RESET_MODE { SMU_ASIC_RESET_MODE_0, SMU_ASIC_RESET_MODE_1, SMU_ASIC_RESET_MODE_2, @@ -516,7 +515,7 @@ struct phm_vq_budgeting_record { struct phm_vq_budgeting_table { uint8_t numEntries; - struct phm_vq_budgeting_record entries[1]; + struct phm_vq_budgeting_record entries[0]; }; struct phm_clock_and_voltage_limits { @@ -607,8 +606,7 @@ struct phm_ppt_v2_information { uint8_t uc_dcef_dpm_voltage_mode; }; -struct phm_ppt_v3_information -{ +struct phm_ppt_v3_information { uint8_t uc_thermal_controller_type; uint16_t us_small_power_limit1; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/pp_thermal.h b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_thermal.h index f7c41185097e..2003acc70ca0 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/pp_thermal.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/pp_thermal.h @@ -25,14 +25,12 @@ #include "power_state.h" -static const struct PP_TemperatureRange __maybe_unused SMU7ThermalWithDelayPolicy[] = -{ +static const struct PP_TemperatureRange __maybe_unused SMU7ThermalWithDelayPolicy[] = { {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, }; -static const struct PP_TemperatureRange __maybe_unused SMU7ThermalPolicy[] = -{ +static const struct PP_TemperatureRange __maybe_unused SMU7ThermalPolicy[] = { {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, }; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu7.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7.h index e14072d45918..bfce9087a47f 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu7.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7.h @@ -101,8 +101,7 @@ #define VR_SMIO_PATTERN_2 4 #define VR_STATIC_VOLTAGE 5 -struct SMU7_PIDController -{ +struct SMU7_PIDController { uint32_t Ki; int32_t LFWindupUL; int32_t LFWindupLL; @@ -136,8 +135,7 @@ typedef struct SMU7_PIDController SMU7_PIDController; #define SMU7_VCE_MCLK_HANDSHAKE_DISABLE 0x00010000 #define SMU7_VCE_SCLK_HANDSHAKE_DISABLE 0x00020000 -struct SMU7_Firmware_Header -{ +struct SMU7_Firmware_Header { uint32_t Digest[5]; uint32_t Version; uint32_t HeaderSize; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu71.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu71.h index 71c9b2d28640..b5f177412769 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu71.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu71.h @@ -118,8 +118,7 @@ typedef struct { #endif -struct SMU71_PIDController -{ +struct SMU71_PIDController { uint32_t Ki; int32_t LFWindupUpperLim; int32_t LFWindupLowerLim; @@ -133,8 +132,7 @@ struct SMU71_PIDController typedef struct SMU71_PIDController SMU71_PIDController; -struct SMU7_LocalDpmScoreboard -{ +struct SMU7_LocalDpmScoreboard { uint32_t PercentageBusy; int32_t PIDError; @@ -179,8 +177,8 @@ struct SMU7_LocalDpmScoreboard uint8_t DteClampMode; uint8_t FpsClampMode; - uint16_t LevelResidencyCounters [SMU71_MAX_LEVELS_GRAPHICS]; - uint16_t LevelSwitchCounters [SMU71_MAX_LEVELS_GRAPHICS]; + uint16_t LevelResidencyCounters[SMU71_MAX_LEVELS_GRAPHICS]; + uint16_t LevelSwitchCounters[SMU71_MAX_LEVELS_GRAPHICS]; void (*TargetStateCalculator)(uint8_t); void (*SavedTargetStateCalculator)(uint8_t); @@ -200,8 +198,7 @@ typedef struct SMU7_LocalDpmScoreboard SMU7_LocalDpmScoreboard; #define SMU7_MAX_VOLTAGE_CLIENTS 12 -struct SMU7_VoltageScoreboard -{ +struct SMU7_VoltageScoreboard { uint16_t CurrentVoltage; uint16_t HighestVoltage; uint16_t MaxVid; @@ -325,8 +322,7 @@ typedef struct SMU7_PowerScoreboard SMU7_PowerScoreboard; // -------------------------------------------------------------------------------------------------- -struct SMU7_ThermalScoreboard -{ +struct SMU7_ThermalScoreboard { int16_t GpuLimit; int16_t GpuHyst; uint16_t CurrGnbTemp; @@ -360,8 +356,7 @@ typedef struct SMU7_ThermalScoreboard SMU7_ThermalScoreboard; #define SMU7_VCE_SCLK_HANDSHAKE_DISABLE 0x00020000 // All 'soft registers' should be uint32_t. -struct SMU71_SoftRegisters -{ +struct SMU71_SoftRegisters { uint32_t RefClockFrequency; uint32_t PmTimerPeriod; uint32_t FeatureEnables; @@ -413,8 +408,7 @@ struct SMU71_SoftRegisters typedef struct SMU71_SoftRegisters SMU71_SoftRegisters; -struct SMU71_Firmware_Header -{ +struct SMU71_Firmware_Header { uint32_t Digest[5]; uint32_t Version; uint32_t HeaderSize; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu73.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73.h index c6b12a4c00db..cf4b2c3c65bc 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu73.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73.h @@ -37,8 +37,7 @@ enum Poly3rdOrderCoeff { POLY_3RD_ORDER_COUNT }; -struct SMU7_Poly3rdOrder_Data -{ +struct SMU7_Poly3rdOrder_Data { int32_t a; int32_t b; int32_t c; @@ -51,8 +50,7 @@ struct SMU7_Poly3rdOrder_Data typedef struct SMU7_Poly3rdOrder_Data SMU7_Poly3rdOrder_Data; -struct Power_Calculator_Data -{ +struct Power_Calculator_Data { uint16_t NoLoadVoltage; uint16_t LoadVoltage; uint16_t Resistance; @@ -71,8 +69,7 @@ struct Power_Calculator_Data typedef struct Power_Calculator_Data PowerCalculatorData_t; -struct Gc_Cac_Weight_Data -{ +struct Gc_Cac_Weight_Data { uint8_t index; uint32_t value; }; @@ -187,8 +184,7 @@ typedef struct { #define SMU73_THERMAL_CLAMP_MODE_COUNT 8 -struct SMU7_HystController_Data -{ +struct SMU7_HystController_Data { uint16_t waterfall_up; uint16_t waterfall_down; uint16_t waterfall_limit; @@ -199,8 +195,7 @@ struct SMU7_HystController_Data typedef struct SMU7_HystController_Data SMU7_HystController_Data; -struct SMU73_PIDController -{ +struct SMU73_PIDController { uint32_t Ki; int32_t LFWindupUpperLim; int32_t LFWindupLowerLim; @@ -215,8 +210,7 @@ struct SMU73_PIDController typedef struct SMU73_PIDController SMU73_PIDController; -struct SMU7_LocalDpmScoreboard -{ +struct SMU7_LocalDpmScoreboard { uint32_t PercentageBusy; int32_t PIDError; @@ -261,8 +255,8 @@ struct SMU7_LocalDpmScoreboard uint8_t DteClampMode; uint8_t FpsClampMode; - uint16_t LevelResidencyCounters [SMU73_MAX_LEVELS_GRAPHICS]; - uint16_t LevelSwitchCounters [SMU73_MAX_LEVELS_GRAPHICS]; + uint16_t LevelResidencyCounters[SMU73_MAX_LEVELS_GRAPHICS]; + uint16_t LevelSwitchCounters[SMU73_MAX_LEVELS_GRAPHICS]; void (*TargetStateCalculator)(uint8_t); void (*SavedTargetStateCalculator)(uint8_t); @@ -315,8 +309,7 @@ typedef uint8_t (*VoltageChangeHandler_t)(uint16_t, uint8_t); typedef uint32_t SMU_VoltageLevel; -struct SMU7_VoltageScoreboard -{ +struct SMU7_VoltageScoreboard { SMU_VoltageLevel TargetVoltage; uint16_t MaxVid; uint8_t HighestVidOffset; @@ -354,7 +347,7 @@ struct SMU7_VoltageScoreboard VoltageChangeHandler_t functionLinks[6]; - uint16_t * VddcFollower1; + uint16_t *VddcFollower1; int16_t Driver_OD_RequestedVidOffset1; int16_t Driver_OD_RequestedVidOffset2; @@ -366,8 +359,7 @@ typedef struct SMU7_VoltageScoreboard SMU7_VoltageScoreboard; // ------------------------------------------------------------------------------------------------------------------------- #define SMU7_MAX_PCIE_LINK_SPEEDS 3 /* 0:Gen1 1:Gen2 2:Gen3 */ -struct SMU7_PCIeLinkSpeedScoreboard -{ +struct SMU7_PCIeLinkSpeedScoreboard { uint8_t DpmEnable; uint8_t DpmRunning; uint8_t DpmForce; @@ -396,8 +388,7 @@ typedef struct SMU7_PCIeLinkSpeedScoreboard SMU7_PCIeLinkSpeedScoreboard; #define SMU7_SCALE_I 7 #define SMU7_SCALE_R 12 -struct SMU7_PowerScoreboard -{ +struct SMU7_PowerScoreboard { uint32_t GpuPower; uint32_t VddcPower; @@ -436,8 +427,7 @@ typedef struct SMU7_PowerScoreboard SMU7_PowerScoreboard; #define SMU7_VCE_SCLK_HANDSHAKE_DISABLE 0x00020000 // All 'soft registers' should be uint32_t. -struct SMU73_SoftRegisters -{ +struct SMU73_SoftRegisters { uint32_t RefClockFrequency; uint32_t PmTimerPeriod; uint32_t FeatureEnables; @@ -493,8 +483,7 @@ struct SMU73_SoftRegisters typedef struct SMU73_SoftRegisters SMU73_SoftRegisters; -struct SMU73_Firmware_Header -{ +struct SMU73_Firmware_Header { uint32_t Digest[5]; uint32_t Version; uint32_t HeaderSize; @@ -708,9 +697,9 @@ typedef struct VFT_CELL_t VFT_CELL_t; struct VFT_TABLE_t { VFT_CELL_t Cell[TEMP_RANGE_MAXSTEPS][NUM_VFT_COLUMNS]; - uint16_t AvfsGbv [NUM_VFT_COLUMNS]; - uint16_t BtcGbv [NUM_VFT_COLUMNS]; - uint16_t Temperature [TEMP_RANGE_MAXSTEPS]; + uint16_t AvfsGbv[NUM_VFT_COLUMNS]; + uint16_t BtcGbv[NUM_VFT_COLUMNS]; + uint16_t Temperature[TEMP_RANGE_MAXSTEPS]; uint8_t NumTemperatureSteps; uint8_t padding[3]; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu73_discrete.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73_discrete.h index 5916be08a7fe..fd0964ac465e 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu73_discrete.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu73_discrete.h @@ -27,8 +27,7 @@ #pragma pack(push, 1) -struct SMIO_Pattern -{ +struct SMIO_Pattern { uint16_t Voltage; uint8_t Smio; uint8_t padding; @@ -36,8 +35,7 @@ struct SMIO_Pattern typedef struct SMIO_Pattern SMIO_Pattern; -struct SMIO_Table -{ +struct SMIO_Table { SMIO_Pattern Pattern[SMU_MAX_SMIO_LEVELS]; }; @@ -100,8 +98,7 @@ struct SMU73_Discrete_Ulv { typedef struct SMU73_Discrete_Ulv SMU73_Discrete_Ulv; -struct SMU73_Discrete_MemoryLevel -{ +struct SMU73_Discrete_MemoryLevel { uint32_t MinVoltage; uint32_t MinMvdd; @@ -124,10 +121,9 @@ struct SMU73_Discrete_MemoryLevel typedef struct SMU73_Discrete_MemoryLevel SMU73_Discrete_MemoryLevel; -struct SMU73_Discrete_LinkLevel -{ +struct SMU73_Discrete_LinkLevel { uint8_t PcieGenSpeed; ///< 0:PciE-gen1 1:PciE-gen2 2:PciE-gen3 - uint8_t PcieLaneCount; ///< 1=x1, 2=x2, 3=x4, 4=x8, 5=x12, 6=x16 + uint8_t PcieLaneCount; ///< 1=x1, 2=x2, 3=x4, 4=x8, 5=x12, 6=x16 uint8_t EnabledForActivity; uint8_t SPC; uint32_t DownThreshold; @@ -139,8 +135,7 @@ typedef struct SMU73_Discrete_LinkLevel SMU73_Discrete_LinkLevel; // MC ARB DRAM Timing registers. -struct SMU73_Discrete_MCArbDramTimingTableEntry -{ +struct SMU73_Discrete_MCArbDramTimingTableEntry { uint32_t McArbDramTiming; uint32_t McArbDramTiming2; uint8_t McArbBurstTime; @@ -151,16 +146,14 @@ struct SMU73_Discrete_MCArbDramTimingTableEntry typedef struct SMU73_Discrete_MCArbDramTimingTableEntry SMU73_Discrete_MCArbDramTimingTableEntry; -struct SMU73_Discrete_MCArbDramTimingTable -{ +struct SMU73_Discrete_MCArbDramTimingTable { SMU73_Discrete_MCArbDramTimingTableEntry entries[SMU__NUM_SCLK_DPM_STATE][SMU__NUM_MCLK_DPM_LEVELS]; }; typedef struct SMU73_Discrete_MCArbDramTimingTable SMU73_Discrete_MCArbDramTimingTable; // UVD VCLK/DCLK state (level) definition. -struct SMU73_Discrete_UvdLevel -{ +struct SMU73_Discrete_UvdLevel { uint32_t VclkFrequency; uint32_t DclkFrequency; uint32_t MinVoltage; @@ -172,8 +165,7 @@ struct SMU73_Discrete_UvdLevel typedef struct SMU73_Discrete_UvdLevel SMU73_Discrete_UvdLevel; // Clocks for other external blocks (VCE, ACP, SAMU). -struct SMU73_Discrete_ExtClkLevel -{ +struct SMU73_Discrete_ExtClkLevel { uint32_t Frequency; uint32_t MinVoltage; uint8_t Divider; @@ -182,8 +174,7 @@ struct SMU73_Discrete_ExtClkLevel typedef struct SMU73_Discrete_ExtClkLevel SMU73_Discrete_ExtClkLevel; -struct SMU73_Discrete_StateInfo -{ +struct SMU73_Discrete_StateInfo { uint32_t SclkFrequency; uint32_t MclkFrequency; uint32_t VclkFrequency; @@ -206,8 +197,7 @@ struct SMU73_Discrete_StateInfo typedef struct SMU73_Discrete_StateInfo SMU73_Discrete_StateInfo; -struct SMU73_Discrete_DpmTable -{ +struct SMU73_Discrete_DpmTable { // Multi-DPM controller settings SMU73_PIDController GraphicsPIDController; SMU73_PIDController MemoryPIDController; @@ -225,9 +215,9 @@ struct SMU73_Discrete_DpmTable uint32_t MvddLevelCount; - uint8_t BapmVddcVidHiSidd [SMU73_MAX_LEVELS_VDDC]; - uint8_t BapmVddcVidLoSidd [SMU73_MAX_LEVELS_VDDC]; - uint8_t BapmVddcVidHiSidd2 [SMU73_MAX_LEVELS_VDDC]; + uint8_t BapmVddcVidHiSidd[SMU73_MAX_LEVELS_VDDC]; + uint8_t BapmVddcVidLoSidd[SMU73_MAX_LEVELS_VDDC]; + uint8_t BapmVddcVidHiSidd2[SMU73_MAX_LEVELS_VDDC]; uint8_t GraphicsDpmLevelCount; uint8_t MemoryDpmLevelCount; @@ -246,19 +236,19 @@ struct SMU73_Discrete_DpmTable uint32_t Reserved[4]; // State table entries for each DPM state - SMU73_Discrete_GraphicsLevel GraphicsLevel [SMU73_MAX_LEVELS_GRAPHICS]; + SMU73_Discrete_GraphicsLevel GraphicsLevel[SMU73_MAX_LEVELS_GRAPHICS]; SMU73_Discrete_MemoryLevel MemoryACPILevel; - SMU73_Discrete_MemoryLevel MemoryLevel [SMU73_MAX_LEVELS_MEMORY]; - SMU73_Discrete_LinkLevel LinkLevel [SMU73_MAX_LEVELS_LINK]; + SMU73_Discrete_MemoryLevel MemoryLevel[SMU73_MAX_LEVELS_MEMORY]; + SMU73_Discrete_LinkLevel LinkLevel[SMU73_MAX_LEVELS_LINK]; SMU73_Discrete_ACPILevel ACPILevel; - SMU73_Discrete_UvdLevel UvdLevel [SMU73_MAX_LEVELS_UVD]; - SMU73_Discrete_ExtClkLevel VceLevel [SMU73_MAX_LEVELS_VCE]; - SMU73_Discrete_ExtClkLevel AcpLevel [SMU73_MAX_LEVELS_ACP]; - SMU73_Discrete_ExtClkLevel SamuLevel [SMU73_MAX_LEVELS_SAMU]; + SMU73_Discrete_UvdLevel UvdLevel[SMU73_MAX_LEVELS_UVD]; + SMU73_Discrete_ExtClkLevel VceLevel[SMU73_MAX_LEVELS_VCE]; + SMU73_Discrete_ExtClkLevel AcpLevel[SMU73_MAX_LEVELS_ACP]; + SMU73_Discrete_ExtClkLevel SamuLevel[SMU73_MAX_LEVELS_SAMU]; SMU73_Discrete_Ulv Ulv; uint32_t SclkStepSize; - uint32_t Smio [SMU73_MAX_ENTRIES_SMIO]; + uint32_t Smio[SMU73_MAX_ENTRIES_SMIO]; uint8_t UvdBootLevel; uint8_t VceBootLevel; @@ -368,8 +358,7 @@ typedef struct SMU73_Discrete_DpmTable SMU73_Discrete_DpmTable; // --------------------------------------------------- Fan Table ----------------------------------------------------------- -struct SMU73_Discrete_FanTable -{ +struct SMU73_Discrete_FanTable { uint16_t FdoMode; int16_t TempMin; int16_t TempMed; @@ -397,8 +386,7 @@ typedef struct SMU73_Discrete_FanTable SMU73_Discrete_FanTable; -struct SMU7_MclkDpmScoreboard -{ +struct SMU7_MclkDpmScoreboard { uint32_t PercentageBusy; @@ -448,8 +436,8 @@ struct SMU7_MclkDpmScoreboard uint8_t VbiWaitCounter; uint8_t EnabledLevelsChange; - uint16_t LevelResidencyCounters [SMU73_MAX_LEVELS_MEMORY]; - uint16_t LevelSwitchCounters [SMU73_MAX_LEVELS_MEMORY]; + uint16_t LevelResidencyCounters[SMU73_MAX_LEVELS_MEMORY]; + uint16_t LevelSwitchCounters[SMU73_MAX_LEVELS_MEMORY]; void (*TargetStateCalculator)(uint8_t); void (*SavedTargetStateCalculator)(uint8_t); @@ -469,8 +457,7 @@ struct SMU7_MclkDpmScoreboard typedef struct SMU7_MclkDpmScoreboard SMU7_MclkDpmScoreboard; -struct SMU7_UlvScoreboard -{ +struct SMU7_UlvScoreboard { uint8_t EnterUlv; uint8_t ExitUlv; uint8_t UlvActive; @@ -485,8 +472,7 @@ struct SMU7_UlvScoreboard typedef struct SMU7_UlvScoreboard SMU7_UlvScoreboard; -struct VddgfxSavedRegisters -{ +struct VddgfxSavedRegisters { uint32_t GPU_DBG[3]; uint32_t MEC_BaseAddress_Hi; uint32_t MEC_BaseAddress_Lo; @@ -497,8 +483,7 @@ struct VddgfxSavedRegisters typedef struct VddgfxSavedRegisters VddgfxSavedRegisters; -struct SMU7_VddGfxScoreboard -{ +struct SMU7_VddGfxScoreboard { uint8_t VddGfxEnable; uint8_t VddGfxActive; uint8_t VPUResetOccured; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu75.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu75.h index 771523001533..7d5ed7751976 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu75.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu75.h @@ -224,8 +224,8 @@ struct SMU7_LocalDpmScoreboard { uint8_t DteClampMode; uint8_t FpsClampMode; - uint16_t LevelResidencyCounters [SMU75_MAX_LEVELS_GRAPHICS]; - uint16_t LevelSwitchCounters [SMU75_MAX_LEVELS_GRAPHICS]; + uint16_t LevelResidencyCounters[SMU75_MAX_LEVELS_GRAPHICS]; + uint16_t LevelSwitchCounters[SMU75_MAX_LEVELS_GRAPHICS]; void (*TargetStateCalculator)(uint8_t); void (*SavedTargetStateCalculator)(uint8_t); @@ -316,7 +316,7 @@ struct SMU7_VoltageScoreboard { VoltageChangeHandler_t functionLinks[6]; - uint16_t * VddcFollower1; + uint16_t *VddcFollower1; int16_t Driver_OD_RequestedVidOffset1; int16_t Driver_OD_RequestedVidOffset2; }; @@ -677,9 +677,9 @@ typedef struct SCS_CELL_t SCS_CELL_t; struct VFT_TABLE_t { VFT_CELL_t Cell[TEMP_RANGE_MAXSTEPS][NUM_VFT_COLUMNS]; - uint16_t AvfsGbv [NUM_VFT_COLUMNS]; - uint16_t BtcGbv [NUM_VFT_COLUMNS]; - int16_t Temperature [TEMP_RANGE_MAXSTEPS]; + uint16_t AvfsGbv[NUM_VFT_COLUMNS]; + uint16_t BtcGbv[NUM_VFT_COLUMNS]; + int16_t Temperature[TEMP_RANGE_MAXSTEPS]; #ifdef SMU__FIRMWARE_SCKS_PRESENT__1 SCS_CELL_t ScksCell[TEMP_RANGE_MAXSTEPS][NUM_VFT_COLUMNS]; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_fusion.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_fusion.h index 78ada9ffd508..e130f52fe8d6 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_fusion.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu7_fusion.h @@ -36,8 +36,7 @@ #define SMU7_NUM_NON_TES 2 // All 'soft registers' should be uint32_t. -struct SMU7_SoftRegisters -{ +struct SMU7_SoftRegisters { uint32_t RefClockFrequency; uint32_t PmTimerP; uint32_t FeatureEnables; @@ -80,8 +79,7 @@ struct SMU7_SoftRegisters typedef struct SMU7_SoftRegisters SMU7_SoftRegisters; -struct SMU7_Fusion_GraphicsLevel -{ +struct SMU7_Fusion_GraphicsLevel { uint32_t MinVddNb; uint32_t SclkFrequency; @@ -111,8 +109,7 @@ struct SMU7_Fusion_GraphicsLevel typedef struct SMU7_Fusion_GraphicsLevel SMU7_Fusion_GraphicsLevel; -struct SMU7_Fusion_GIOLevel -{ +struct SMU7_Fusion_GIOLevel { uint8_t EnabledForActivity; uint8_t LclkDid; uint8_t Vid; @@ -137,8 +134,7 @@ struct SMU7_Fusion_GIOLevel typedef struct SMU7_Fusion_GIOLevel SMU7_Fusion_GIOLevel; // UVD VCLK/DCLK state (level) definition. -struct SMU7_Fusion_UvdLevel -{ +struct SMU7_Fusion_UvdLevel { uint32_t VclkFrequency; uint32_t DclkFrequency; uint16_t MinVddNb; @@ -155,8 +151,7 @@ struct SMU7_Fusion_UvdLevel typedef struct SMU7_Fusion_UvdLevel SMU7_Fusion_UvdLevel; // Clocks for other external blocks (VCE, ACP, SAMU). -struct SMU7_Fusion_ExtClkLevel -{ +struct SMU7_Fusion_ExtClkLevel { uint32_t Frequency; uint16_t MinVoltage; uint8_t Divider; @@ -166,8 +161,7 @@ struct SMU7_Fusion_ExtClkLevel }; typedef struct SMU7_Fusion_ExtClkLevel SMU7_Fusion_ExtClkLevel; -struct SMU7_Fusion_ACPILevel -{ +struct SMU7_Fusion_ACPILevel { uint32_t Flags; uint32_t MinVddNb; uint32_t SclkFrequency; @@ -181,8 +175,7 @@ struct SMU7_Fusion_ACPILevel typedef struct SMU7_Fusion_ACPILevel SMU7_Fusion_ACPILevel; -struct SMU7_Fusion_NbDpm -{ +struct SMU7_Fusion_NbDpm { uint8_t DpmXNbPsHi; uint8_t DpmXNbPsLo; uint8_t Dpm0PgNbPsHi; @@ -197,8 +190,7 @@ struct SMU7_Fusion_NbDpm typedef struct SMU7_Fusion_NbDpm SMU7_Fusion_NbDpm; -struct SMU7_Fusion_StateInfo -{ +struct SMU7_Fusion_StateInfo { uint32_t SclkFrequency; uint32_t LclkFrequency; uint32_t VclkFrequency; @@ -214,8 +206,7 @@ struct SMU7_Fusion_StateInfo typedef struct SMU7_Fusion_StateInfo SMU7_Fusion_StateInfo; -struct SMU7_Fusion_DpmTable -{ +struct SMU7_Fusion_DpmTable { uint32_t SystemFlags; SMU7_PIDController GraphicsPIDController; @@ -230,12 +221,12 @@ struct SMU7_Fusion_DpmTable uint8_t SamuLevelCount; uint16_t FpsHighT; - SMU7_Fusion_GraphicsLevel GraphicsLevel [SMU__NUM_SCLK_DPM_STATE]; + SMU7_Fusion_GraphicsLevel GraphicsLevel[SMU__NUM_SCLK_DPM_STATE]; SMU7_Fusion_ACPILevel ACPILevel; - SMU7_Fusion_UvdLevel UvdLevel [SMU7_MAX_LEVELS_UVD]; - SMU7_Fusion_ExtClkLevel VceLevel [SMU7_MAX_LEVELS_VCE]; - SMU7_Fusion_ExtClkLevel AcpLevel [SMU7_MAX_LEVELS_ACP]; - SMU7_Fusion_ExtClkLevel SamuLevel [SMU7_MAX_LEVELS_SAMU]; + SMU7_Fusion_UvdLevel UvdLevel[SMU7_MAX_LEVELS_UVD]; + SMU7_Fusion_ExtClkLevel VceLevel[SMU7_MAX_LEVELS_VCE]; + SMU7_Fusion_ExtClkLevel AcpLevel[SMU7_MAX_LEVELS_ACP]; + SMU7_Fusion_ExtClkLevel SamuLevel[SMU7_MAX_LEVELS_SAMU]; uint8_t UvdBootLevel; uint8_t VceBootLevel; @@ -266,10 +257,9 @@ struct SMU7_Fusion_DpmTable }; -struct SMU7_Fusion_GIODpmTable -{ +struct SMU7_Fusion_GIODpmTable { - SMU7_Fusion_GIOLevel GIOLevel [SMU7_MAX_LEVELS_GIO]; + SMU7_Fusion_GIOLevel GIOLevel[SMU7_MAX_LEVELS_GIO]; SMU7_PIDController GioPIDController; diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/smu9_driver_if.h b/drivers/gpu/drm/amd/pm/powerplay/inc/smu9_driver_if.h index faae4b918d90..2c69a5694f94 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/smu9_driver_if.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/smu9_driver_if.h @@ -178,20 +178,20 @@ typedef struct { uint8_t padding8_2[2]; /* SOC Frequencies */ - PllSetting_t GfxclkLevel [NUM_GFXCLK_DPM_LEVELS]; + PllSetting_t GfxclkLevel[NUM_GFXCLK_DPM_LEVELS]; - uint8_t SocclkDid [NUM_SOCCLK_DPM_LEVELS]; /* DID */ - uint8_t SocDpmVoltageIndex [NUM_SOCCLK_DPM_LEVELS]; + uint8_t SocclkDid[NUM_SOCCLK_DPM_LEVELS]; /* DID */ + uint8_t SocDpmVoltageIndex[NUM_SOCCLK_DPM_LEVELS]; - uint8_t VclkDid [NUM_UVD_DPM_LEVELS]; /* DID */ - uint8_t DclkDid [NUM_UVD_DPM_LEVELS]; /* DID */ - uint8_t UvdDpmVoltageIndex [NUM_UVD_DPM_LEVELS]; + uint8_t VclkDid[NUM_UVD_DPM_LEVELS]; /* DID */ + uint8_t DclkDid[NUM_UVD_DPM_LEVELS]; /* DID */ + uint8_t UvdDpmVoltageIndex[NUM_UVD_DPM_LEVELS]; - uint8_t EclkDid [NUM_VCE_DPM_LEVELS]; /* DID */ - uint8_t VceDpmVoltageIndex [NUM_VCE_DPM_LEVELS]; + uint8_t EclkDid[NUM_VCE_DPM_LEVELS]; /* DID */ + uint8_t VceDpmVoltageIndex[NUM_VCE_DPM_LEVELS]; - uint8_t Mp0clkDid [NUM_MP0CLK_DPM_LEVELS]; /* DID */ - uint8_t Mp0DpmVoltageIndex [NUM_MP0CLK_DPM_LEVELS]; + uint8_t Mp0clkDid[NUM_MP0CLK_DPM_LEVELS]; /* DID */ + uint8_t Mp0DpmVoltageIndex[NUM_MP0CLK_DPM_LEVELS]; DisplayClockTable_t DisplayClockTable[DSPCLK_COUNT][NUM_DSPCLK_LEVELS]; QuadraticInt_t DisplayClock2Gfxclk[DSPCLK_COUNT]; diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c index 4bc8db1be738..9e4228232f02 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/ci_smumgr.c @@ -2732,7 +2732,7 @@ static bool ci_is_dpm_running(struct pp_hwmgr *hwmgr) static int ci_smu_init(struct pp_hwmgr *hwmgr) { - struct ci_smumgr *ci_priv = NULL; + struct ci_smumgr *ci_priv; ci_priv = kzalloc(sizeof(struct ci_smumgr), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c index 02c094a06605..5e43ad2b2956 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c @@ -332,7 +332,7 @@ static bool fiji_is_hw_avfs_present(struct pp_hwmgr *hwmgr) static int fiji_smu_init(struct pp_hwmgr *hwmgr) { - struct fiji_smumgr *fiji_priv = NULL; + struct fiji_smumgr *fiji_priv; fiji_priv = kzalloc(sizeof(struct fiji_smumgr), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c index 060fc140c574..97d9802fe673 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c @@ -259,7 +259,7 @@ static int iceland_start_smu(struct pp_hwmgr *hwmgr) static int iceland_smu_init(struct pp_hwmgr *hwmgr) { - struct iceland_smumgr *iceland_priv = NULL; + struct iceland_smumgr *iceland_priv; iceland_priv = kzalloc(sizeof(struct iceland_smumgr), GFP_KERNEL); diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c index e7ed2a7adf8f..ff6b563ecbf5 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c @@ -1888,7 +1888,7 @@ static int polaris10_populate_avfs_parameters(struct pp_hwmgr *hwmgr) (avfs_params.ucEnableGB_VDROOP_TABLE_CKSOFF << BTCGB1_Vdroop_Enable_SHIFT) | (avfs_params.ucEnableGB_FUSE_TABLE_CKSON << AVFSGB0_Vdroop_Enable_SHIFT) | (avfs_params.ucEnableGB_FUSE_TABLE_CKSOFF << AVFSGB1_Vdroop_Enable_SHIFT); - data->apply_avfs_cks_off_voltage = (avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage == 1) ? true : false; + data->apply_avfs_cks_off_voltage = avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage == 1; } return result; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c index acbe41174d7e..6fe6e6abb5d8 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c @@ -226,7 +226,7 @@ static int tonga_start_smu(struct pp_hwmgr *hwmgr) static int tonga_smu_init(struct pp_hwmgr *hwmgr) { - struct tonga_smumgr *tonga_priv = NULL; + struct tonga_smumgr *tonga_priv; tonga_priv = kzalloc(sizeof(struct tonga_smumgr), GFP_KERNEL); if (tonga_priv == NULL) diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.c index 7d024d3facef..34c9f59b889a 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/vegam_smumgr.c @@ -295,9 +295,8 @@ static int vegam_process_firmware_header(struct pp_hwmgr *hwmgr) static bool vegam_is_dpm_running(struct pp_hwmgr *hwmgr) { - return (1 == PHM_READ_INDIRECT_FIELD(hwmgr->device, - CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON)) - ? true : false; + return 1 == PHM_READ_INDIRECT_FIELD(hwmgr->device, + CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON); } static uint32_t vegam_get_mac_definition(uint32_t value) @@ -1660,7 +1659,7 @@ static int vegam_populate_avfs_parameters(struct pp_hwmgr *hwmgr) (avfs_params.ucEnableGB_FUSE_TABLE_CKSON << AVFSGB0_Vdroop_Enable_SHIFT) | (avfs_params.ucEnableGB_FUSE_TABLE_CKSOFF << AVFSGB1_Vdroop_Enable_SHIFT); data->apply_avfs_cks_off_voltage = - (avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage == 1) ? true : false; + avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage == 1; } return result; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index ce41a8309582..f005a90c35af 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -618,7 +618,7 @@ static int smu_set_funcs(struct amdgpu_device *adev) adev->pm.pp_feature &= ~PP_GFXOFF_MASK; arcturus_set_ppt_funcs(smu); /* OD is not supported on Arcturus */ - smu->od_enabled =false; + smu->od_enabled = false; break; case IP_VERSION(13, 0, 2): aldebaran_set_ppt_funcs(smu); @@ -1581,9 +1581,9 @@ static int smu_disable_dpms(struct smu_context *smu) /* * For SMU 13.0.4/11, PMFW will handle the features disablement properly - * for gpu reset case. Driver involvement is unnecessary. + * for gpu reset and S0i3 cases. Driver involvement is unnecessary. */ - if (amdgpu_in_reset(adev)) { + if (amdgpu_in_reset(adev) || adev->in_s0ix) { switch (adev->ip_versions[MP1_HWIP][0]) { case IP_VERSION(13, 0, 4): case IP_VERSION(13, 0, 11): @@ -1648,7 +1648,7 @@ static int smu_hw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct smu_context *smu = adev->powerplay.pp_handle; - if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) + if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) return 0; smu_dpm_set_vcn_enable(smu, false); @@ -1700,7 +1700,7 @@ static int smu_suspend(void *handle) int ret; uint64_t count; - if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) + if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) return 0; if (!smu->pm_enabled) @@ -2217,8 +2217,7 @@ const struct amd_ip_funcs smu_ip_funcs = { .set_powergating_state = smu_set_powergating_state, }; -const struct amdgpu_ip_block_version smu_v11_0_ip_block = -{ +const struct amdgpu_ip_block_version smu_v11_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_SMC, .major = 11, .minor = 0, @@ -2226,8 +2225,7 @@ const struct amdgpu_ip_block_version smu_v11_0_ip_block = .funcs = &smu_ip_funcs, }; -const struct amdgpu_ip_block_version smu_v12_0_ip_block = -{ +const struct amdgpu_ip_block_version smu_v12_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_SMC, .major = 12, .minor = 0, @@ -2235,8 +2233,7 @@ const struct amdgpu_ip_block_version smu_v12_0_ip_block = .funcs = &smu_ip_funcs, }; -const struct amdgpu_ip_block_version smu_v13_0_ip_block = -{ +const struct amdgpu_ip_block_version smu_v13_0_ip_block = { .type = AMD_IP_BLOCK_TYPE_SMC, .major = 13, .minor = 0, @@ -2337,7 +2334,7 @@ int smu_get_power_limit(void *handle, if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) return -EOPNOTSUPP; - switch(pp_power_type) { + switch (pp_power_type) { case PP_PWR_TYPE_SUSTAINED: limit_type = SMU_DEFAULT_PPT_LIMIT; break; @@ -2349,7 +2346,7 @@ int smu_get_power_limit(void *handle, break; } - switch(pp_limit_level){ + switch (pp_limit_level) { case PP_PWR_LIMIT_CURRENT: limit_level = SMU_PPT_LIMIT_CURRENT; break; @@ -2595,7 +2592,7 @@ static int smu_read_sensor(void *handle, *size = 4; break; case AMDGPU_PP_SENSOR_VCN_POWER_STATE: - *(uint32_t *)data = atomic_read(&smu->smu_power.power_gate.vcn_gated) ? 0: 1; + *(uint32_t *)data = atomic_read(&smu->smu_power.power_gate.vcn_gated) ? 0 : 1; *size = 4; break; case AMDGPU_PP_SENSOR_MIN_FAN_RPM: @@ -2868,7 +2865,7 @@ static int smu_set_xgmi_pstate(void *handle, if (smu->ppt_funcs->set_xgmi_pstate) ret = smu->ppt_funcs->set_xgmi_pstate(smu, pstate); - if(ret) + if (ret) dev_err(smu->adev->dev, "Failed to set XGMI pstate!\n"); return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index 6e2069dcb6b9..5a52098bcf16 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -200,29 +200,25 @@ struct smu_power_state { struct smu_hw_power_state hardware; }; -enum smu_power_src_type -{ +enum smu_power_src_type { SMU_POWER_SOURCE_AC, SMU_POWER_SOURCE_DC, SMU_POWER_SOURCE_COUNT, }; -enum smu_ppt_limit_type -{ +enum smu_ppt_limit_type { SMU_DEFAULT_PPT_LIMIT = 0, SMU_FAST_PPT_LIMIT, }; -enum smu_ppt_limit_level -{ +enum smu_ppt_limit_level { SMU_PPT_LIMIT_MIN = -1, SMU_PPT_LIMIT_CURRENT, SMU_PPT_LIMIT_DEFAULT, SMU_PPT_LIMIT_MAX, }; -enum smu_memory_pool_size -{ +enum smu_memory_pool_size { SMU_MEMORY_POOL_SIZE_ZERO = 0, SMU_MEMORY_POOL_SIZE_256_MB = 0x10000000, SMU_MEMORY_POOL_SIZE_512_MB = 0x20000000, @@ -282,8 +278,7 @@ struct smu_clock_info { uint32_t max_bus_bandwidth; }; -struct smu_bios_boot_up_values -{ +struct smu_bios_boot_up_values { uint32_t revision; uint32_t gfxclk; uint32_t uclk; @@ -305,8 +300,7 @@ struct smu_bios_boot_up_values uint32_t firmware_caps; }; -enum smu_table_id -{ +enum smu_table_id { SMU_TABLE_PPTABLE = 0, SMU_TABLE_WATERMARKS, SMU_TABLE_CUSTOM_DPM, @@ -326,8 +320,7 @@ enum smu_table_id SMU_TABLE_COUNT, }; -struct smu_table_context -{ +struct smu_table_context { void *power_play_table; uint32_t power_play_table_size; void *hardcode_pptable; @@ -390,8 +383,7 @@ struct smu_power_context { }; #define SMU_FEATURE_MAX (64) -struct smu_feature -{ +struct smu_feature { uint32_t feature_num; DECLARE_BITMAP(supported, SMU_FEATURE_MAX); DECLARE_BITMAP(allowed, SMU_FEATURE_MAX); @@ -416,21 +408,18 @@ struct mclock_latency_table { struct mclk_latency_entries entries[MAX_REGULAR_DPM_NUM]; }; -enum smu_reset_mode -{ +enum smu_reset_mode { SMU_RESET_MODE_0, SMU_RESET_MODE_1, SMU_RESET_MODE_2, }; -enum smu_baco_state -{ +enum smu_baco_state { SMU_BACO_STATE_ENTER = 0, SMU_BACO_STATE_EXIT, }; -struct smu_baco_context -{ +struct smu_baco_context { uint32_t state; bool platform_support; bool maco_support; @@ -478,8 +467,7 @@ struct stb_context { #define WORKLOAD_POLICY_MAX 7 -struct smu_context -{ +struct smu_context { struct amdgpu_device *adev; struct amdgpu_irq_src irq_source; @@ -1043,10 +1031,7 @@ struct pptable_funcs { enum smu_feature_mask mask); /** - * @notify_display_change: Enable fast memory clock switching. - * - * Allows for fine grained memory clock switching but has more stringent - * timing requirements. + * @notify_display_change: General interface call to let SMU know about DC change */ int (*notify_display_change)(struct smu_context *smu); @@ -1398,6 +1383,7 @@ typedef enum { METRICS_PCIE_RATE, METRICS_PCIE_WIDTH, METRICS_CURR_FANPWM, + METRICS_CURR_SOCKETPOWER, } MetricsMember_t; enum smu_cmn2asic_mapping_type { diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h index 7589faa0232d..779c2524806c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu13_driver_if_v13_0_5.h @@ -23,7 +23,7 @@ #ifndef __SMU13_DRIVER_IF_V13_0_5_H__ #define __SMU13_DRIVER_IF_V13_0_5_H__ -#define SMU13_0_5_DRIVER_IF_VERSION 4 +#define SMU13_0_5_DRIVER_IF_VERSION 5 // Throttler Status Bitmask #define THROTTLER_STATUS_BIT_SPL 0 @@ -103,7 +103,6 @@ typedef struct { uint16_t ThrottlerStatus; uint16_t CurrentSocketPower; //[mW] - uint16_t spare1; } SmuMetrics_t; //Freq in MHz diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_0_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_0_ppsmc.h index 10cff75b44d5..e2ee855c7748 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_0_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_0_ppsmc.h @@ -138,7 +138,10 @@ #define PPSMC_MSG_SetBadMemoryPagesRetiredFlagsPerChannel 0x4A #define PPSMC_MSG_SetPriorityDeltaGain 0x4B #define PPSMC_MSG_AllowIHHostInterrupt 0x4C -#define PPSMC_Message_Count 0x4D + +#define PPSMC_MSG_DALNotPresent 0x4E + +#define PPSMC_Message_Count 0x4F //Debug Dump Message #define DEBUGSMC_MSG_TestMessage 0x1 diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h index 252aef190c5c..9be4051c0865 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h @@ -123,7 +123,7 @@ typedef enum { VOLTAGE_GUARDBAND_COUNT } GFX_GUARDBAND_e; -#define SMU_METRICS_TABLE_VERSION 0x5 +#define SMU_METRICS_TABLE_VERSION 0x7 typedef struct __attribute__((packed, aligned(4))) { uint32_t AccumulationCounter; @@ -198,7 +198,7 @@ typedef struct __attribute__((packed, aligned(4))) { uint32_t SocketThmResidencyAcc; uint32_t VrThmResidencyAcc; uint32_t HbmThmResidencyAcc; - uint32_t spare; + uint32_t GfxLockXCDMak; // New Items at end to maintain driver compatibility uint32_t GfxclkFrequency[8]; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h index ae4f44c4b877..70a4a717fd3f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h @@ -83,13 +83,27 @@ #define PPSMC_MSG_GetMinGfxDpmFreq 0x32 #define PPSMC_MSG_GetMaxGfxDpmFreq 0x33 #define PPSMC_MSG_PrepareForDriverUnload 0x34 -#define PPSMC_Message_Count 0x35 +#define PPSMC_MSG_ReadThrottlerLimit 0x35 +#define PPSMC_MSG_QueryValidMcaCount 0x36 +#define PPSMC_MSG_McaBankDumpDW 0x37 +#define PPSMC_MSG_GetCTFLimit 0x38 +#define PPSMC_Message_Count 0x39 //PPSMC Reset Types for driver msg argument #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET 0x1 #define PPSMC_RESET_TYPE_DRIVER_MODE_2_RESET 0x2 #define PPSMC_RESET_TYPE_DRIVER_MODE_3_RESET 0x3 +//PPSMC Reset Types for driver msg argument +#define PPSMC_THROTTLING_LIMIT_TYPE_SOCKET 0x1 +#define PPSMC_THROTTLING_LIMIT_TYPE_HBM 0x2 + +//CTF/Throttle Limit types +#define PPSMC_AID_THM_TYPE 0x1 +#define PPSMC_CCD_THM_TYPE 0x2 +#define PPSMC_XCD_THM_TYPE 0x3 +#define PPSMC_HBM_THM_TYPE 0x4 + typedef uint32_t PPSMC_Result; typedef uint32_t PPSMC_MSG; diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_11_0_cdr_table.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_11_0_cdr_table.h index beab6d7b28b7..630132c4a76b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_11_0_cdr_table.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_11_0_cdr_table.h @@ -52,8 +52,7 @@ static unsigned int DbiPrbs7[] = //4096 bytes, 256 byte aligned -static unsigned int NoDbiPrbs7[] = -{ +static unsigned int NoDbiPrbs7[] = { 0x0f0f0f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f00f0f, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f0f0f0, 0x0f0ff0f0, 0x0f0f0f0f, 0x0f0f0f0f, 0xf0f0f0f0, 0xf0f00f0f, 0x0f0f0f0f, 0xf0f00f0f, 0x0f0ff0f0, 0x0f0f0f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f00f0f, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0ff0f0, 0xf0f0f0f0, 0xf0f00f0f, 0xf0f0f0f0, 0x0f0f0f0f, 0x0f0ff0f0, 0xf0f00f0f, @@ -121,8 +120,7 @@ static unsigned int NoDbiPrbs7[] = }; // 4096 bytes, 256 byte aligned -static unsigned int DbiPrbs7[] = -{ +static unsigned int DbiPrbs7[] = { 0xffffffff, 0xffffffff, 0xffffffff, 0x0000ffff, 0xffffffff, 0xffffffff, 0x00000000, 0xffffffff, 0xffffffff, 0x0000ffff, 0x0000ffff, 0xffffffff, 0x00000000, 0x00000000, 0xffffffff, 0x0000ffff, 0xffffffff, 0x0000ffff, 0x00000000, 0xffffffff, 0x00000000, 0x0000ffff, 0x0000ffff, 0x0000ffff, 0xffff0000, 0x00000000, 0x00000000, 0xffff0000, 0xffffffff, 0xffffffff, 0x00000000, 0x0000ffff, 0xffffffff, 0x0000ffff, 0xffff0000, 0xffffffff, 0x00000000, 0xffff0000, 0x0000ffff, 0x0000ffff, 0x00000000, 0xffff0000, 0x00000000, 0x0000ffff, 0x00000000, 0xffffffff, 0xffff0000, 0x0000ffff, diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h index 297b70b9388f..e57265cf637c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h @@ -84,6 +84,7 @@ __SMU_DUMMY_MAP(SetTjMax), \ __SMU_DUMMY_MAP(SetFanTemperatureTarget), \ __SMU_DUMMY_MAP(PrepareMp1ForUnload), \ + __SMU_DUMMY_MAP(GetCTFLimit), \ __SMU_DUMMY_MAP(DramLogSetDramAddrHigh), \ __SMU_DUMMY_MAP(DramLogSetDramAddrLow), \ __SMU_DUMMY_MAP(DramLogSetDramSize), \ @@ -245,7 +246,8 @@ __SMU_DUMMY_MAP(AllowGpo), \ __SMU_DUMMY_MAP(Mode2Reset), \ __SMU_DUMMY_MAP(RequestI2cTransaction), \ - __SMU_DUMMY_MAP(GetMetricsTable), + __SMU_DUMMY_MAP(GetMetricsTable), \ + __SMU_DUMMY_MAP(DALNotPresent), #undef __SMU_DUMMY_MAP #define __SMU_DUMMY_MAP(type) SMU_MSG_##type diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h index d466db6f0ad4..a0e5ad0381d6 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h @@ -64,11 +64,9 @@ #define LINK_SPEED_MAX 3 static const __maybe_unused uint16_t link_width[] = {0, 1, 2, 4, 8, 12, 16}; -static const __maybe_unused uint16_t link_speed[] = {25, 50, 80, 160}; static const -struct smu_temperature_range __maybe_unused smu11_thermal_policy[] = -{ +struct smu_temperature_range __maybe_unused smu11_thermal_policy[] = { {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, }; @@ -96,8 +94,8 @@ struct smu_11_0_dpm_table { }; struct smu_11_0_pcie_table { - uint8_t pcie_gen[MAX_PCIE_CONF]; - uint8_t pcie_lane[MAX_PCIE_CONF]; + uint8_t pcie_gen[MAX_PCIE_CONF]; + uint8_t pcie_lane[MAX_PCIE_CONF]; }; struct smu_11_0_dpm_tables { diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_pptable.h index 0116e3d04fad..df7430876e0c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_pptable.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0_pptable.h @@ -101,8 +101,7 @@ enum SMU_11_0_ODSETTING_ID { }; #define SMU_11_0_MAX_ODSETTING 32 //Maximum Number of ODSettings -struct smu_11_0_overdrive_table -{ +struct smu_11_0_overdrive_table { uint8_t revision; //Revision = SMU_11_0_PP_OVERDRIVE_VERSION uint8_t reserve[3]; //Zero filled field reserved for future use uint32_t feature_count; //Total number of supported features @@ -127,8 +126,7 @@ enum SMU_11_0_PPCLOCK_ID { }; #define SMU_11_0_MAX_PPCLOCK 16 //Maximum Number of PP Clocks -struct smu_11_0_power_saving_clock_table -{ +struct smu_11_0_power_saving_clock_table { uint8_t revision; //Revision = SMU_11_0_PP_POWERSAVINGCLOCK_VERSION uint8_t reserve[3]; //Zero filled field reserved for future use uint32_t count; //power_saving_clock_count = SMU_11_0_PPCLOCK_COUNT @@ -136,8 +134,7 @@ struct smu_11_0_power_saving_clock_table uint32_t min[SMU_11_0_MAX_PPCLOCK]; //PowerSavingClock Mode Clock Minimum array In MHz }; -struct smu_11_0_powerplay_table -{ +struct smu_11_0_powerplay_table { struct atom_common_table_header header; uint8_t table_revision; uint16_t table_size; //Driver portion table size. The offset to smc_pptable including header size @@ -145,14 +142,14 @@ struct smu_11_0_powerplay_table uint32_t golden_revision; uint16_t format_id; uint32_t platform_caps; //POWERPLAYABLE::ulPlatformCaps - + uint8_t thermal_controller_type; //one of SMU_11_0_PP_THERMALCONTROLLER uint16_t small_power_limit1; uint16_t small_power_limit2; uint16_t boost_power_limit; - uint16_t od_turbo_power_limit; //Power limit setting for Turbo mode in Performance UI Tuning. - uint16_t od_power_save_power_limit; //Power limit setting for PowerSave/Optimal mode in Performance UI Tuning. + uint16_t od_turbo_power_limit; //Power limit setting for Turbo mode in Performance UI Tuning. + uint16_t od_power_save_power_limit; //Power limit setting for PowerSave/Optimal mode in Performance UI Tuning. uint16_t software_shutdown_temp; uint16_t reserve[6]; //Zero filled field reserved for future use diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h index eadbe0149cae..eb694f9f556d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_7_pptable.h @@ -41,8 +41,7 @@ #define SMU_13_0_7_PP_OVERDRIVE_VERSION 0x83 // OverDrive 8 Table Version 0.2 #define SMU_13_0_7_PP_POWERSAVINGCLOCK_VERSION 0x01 // Power Saving Clock Table Version 1.00 -enum SMU_13_0_7_ODFEATURE_CAP -{ +enum SMU_13_0_7_ODFEATURE_CAP { SMU_13_0_7_ODCAP_GFXCLK_LIMITS = 0, SMU_13_0_7_ODCAP_UCLK_LIMITS, SMU_13_0_7_ODCAP_POWER_LIMIT, @@ -62,8 +61,7 @@ enum SMU_13_0_7_ODFEATURE_CAP SMU_13_0_7_ODCAP_COUNT, }; -enum SMU_13_0_7_ODFEATURE_ID -{ +enum SMU_13_0_7_ODFEATURE_ID { SMU_13_0_7_ODFEATURE_GFXCLK_LIMITS = 1 << SMU_13_0_7_ODCAP_GFXCLK_LIMITS, //GFXCLK Limit feature SMU_13_0_7_ODFEATURE_UCLK_LIMITS = 1 << SMU_13_0_7_ODCAP_UCLK_LIMITS, //UCLK Limit feature SMU_13_0_7_ODFEATURE_POWER_LIMIT = 1 << SMU_13_0_7_ODCAP_POWER_LIMIT, //Power Limit feature @@ -85,8 +83,7 @@ enum SMU_13_0_7_ODFEATURE_ID #define SMU_13_0_7_MAX_ODFEATURE 32 //Maximum Number of OD Features -enum SMU_13_0_7_ODSETTING_ID -{ +enum SMU_13_0_7_ODSETTING_ID { SMU_13_0_7_ODSETTING_GFXCLKFMAX = 0, SMU_13_0_7_ODSETTING_GFXCLKFMIN, SMU_13_0_7_ODSETTING_UCLKFMIN, @@ -123,8 +120,7 @@ enum SMU_13_0_7_ODSETTING_ID }; #define SMU_13_0_7_MAX_ODSETTING 64 //Maximum Number of ODSettings -enum SMU_13_0_7_PWRMODE_SETTING -{ +enum SMU_13_0_7_PWRMODE_SETTING { SMU_13_0_7_PMSETTING_POWER_LIMIT_QUIET = 0, SMU_13_0_7_PMSETTING_POWER_LIMIT_BALANCE, SMU_13_0_7_PMSETTING_POWER_LIMIT_TURBO, @@ -144,8 +140,7 @@ enum SMU_13_0_7_PWRMODE_SETTING }; #define SMU_13_0_7_MAX_PMSETTING 32 //Maximum Number of PowerMode Settings -struct smu_13_0_7_overdrive_table -{ +struct smu_13_0_7_overdrive_table { uint8_t revision; //Revision = SMU_13_0_7_PP_OVERDRIVE_VERSION uint8_t reserve[3]; //Zero filled field reserved for future use uint32_t feature_count; //Total number of supported features @@ -156,8 +151,7 @@ struct smu_13_0_7_overdrive_table int16_t pm_setting[SMU_13_0_7_MAX_PMSETTING]; //Optimized power mode feature settings }; -enum SMU_13_0_7_PPCLOCK_ID -{ +enum SMU_13_0_7_PPCLOCK_ID { SMU_13_0_7_PPCLOCK_GFXCLK = 0, SMU_13_0_7_PPCLOCK_SOCCLK, SMU_13_0_7_PPCLOCK_UCLK, @@ -175,8 +169,7 @@ enum SMU_13_0_7_PPCLOCK_ID }; #define SMU_13_0_7_MAX_PPCLOCK 16 //Maximum Number of PP Clocks -struct smu_13_0_7_powerplay_table -{ +struct smu_13_0_7_powerplay_table { struct atom_common_table_header header; //For PLUM_BONITO, header.format_revision = 15, header.content_revision = 0 uint8_t table_revision; //For PLUM_BONITO, table_revision = 2 uint8_t padding; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index 3bb18396d2f9..704a2b577a0e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -598,7 +598,7 @@ static int arcturus_get_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; @@ -1130,7 +1130,7 @@ static int arcturus_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = arcturus_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); @@ -1169,6 +1169,7 @@ static int arcturus_read_sensor(struct smu_context *smu, ret = smu_v11_0_get_gfx_vdd(smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: default: ret = -EOPNOTSUPP; break; @@ -1482,7 +1483,7 @@ static int arcturus_set_power_profile_mode(struct smu_context *smu, return ret; if ((profile_mode == PP_SMC_POWER_PROFILE_CUSTOM) && - (smu_version >=0x360d00)) { + (smu_version >= 0x360d00)) { ret = smu_cmn_update_table(smu, SMU_TABLE_ACTIVITY_MONITOR_COEFF, WORKLOAD_PPLIB_CUSTOM_BIT, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c index ca4d97b7f576..9548bd3c624b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/cyan_skillfish_ppt.c @@ -154,10 +154,14 @@ cyan_skillfish_get_smu_metrics_data(struct smu_context *smu, case METRICS_CURR_UCLK: *value = metrics->Current.MemclkFrequency; break; - case METRICS_AVERAGE_SOCKETPOWER: + case METRICS_CURR_SOCKETPOWER: *value = (metrics->Current.CurrentSocketPower << 8) / 1000; break; + case METRICS_AVERAGE_SOCKETPOWER: + *value = (metrics->Average.CurrentSocketPower << 8) / + 1000; + break; case METRICS_TEMPERATURE_EDGE: *value = metrics->Current.GfxTemperature / 100 * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; @@ -208,12 +212,18 @@ static int cyan_skillfish_read_sensor(struct smu_context *smu, *(uint32_t *)data *= 100; *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = cyan_skillfish_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: + ret = cyan_skillfish_get_smu_metrics_data(smu, + METRICS_CURR_SOCKETPOWER, + (uint32_t *)data); + *size = 4; + break; case AMDGPU_PP_SENSOR_HOTSPOT_TEMP: ret = cyan_skillfish_get_smu_metrics_data(smu, METRICS_TEMPERATURE_HOTSPOT, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 95f6d821bacb..18487ae10bcf 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -136,7 +136,7 @@ static struct cmn2asic_msg_mapping navi10_message_map[SMU_MSG_MAX_COUNT] = { MSG_MAP(PowerDownJpeg, PPSMC_MSG_PowerDownJpeg, 0), MSG_MAP(BacoAudioD3PME, PPSMC_MSG_BacoAudioD3PME, 0), MSG_MAP(ArmD3, PPSMC_MSG_ArmD3, 0), - MSG_MAP(DAL_DISABLE_DUMMY_PSTATE_CHANGE,PPSMC_MSG_DALDisableDummyPstateChange, 0), + MSG_MAP(DAL_DISABLE_DUMMY_PSTATE_CHANGE, PPSMC_MSG_DALDisableDummyPstateChange, 0), MSG_MAP(DAL_ENABLE_DUMMY_PSTATE_CHANGE, PPSMC_MSG_DALEnableDummyPstateChange, 0), MSG_MAP(GetVoltageByDpm, PPSMC_MSG_GetVoltageByDpm, 0), MSG_MAP(GetVoltageByDpmOverdrive, PPSMC_MSG_GetVoltageByDpmOverdrive, 0), @@ -556,7 +556,7 @@ static int navi10_get_legacy_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_legacy_t *metrics = (SmuMetrics_legacy_t *)smu_table->metrics_table; int ret = 0; @@ -642,7 +642,7 @@ static int navi10_get_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; @@ -731,7 +731,7 @@ static int navi12_get_legacy_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_NV12_legacy_t *metrics = (SmuMetrics_NV12_legacy_t *)smu_table->metrics_table; int ret = 0; @@ -817,7 +817,7 @@ static int navi12_get_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_NV12_t *metrics = (SmuMetrics_NV12_t *)smu_table->metrics_table; int ret = 0; @@ -1686,7 +1686,7 @@ static int navi10_force_clk_levels(struct smu_context *smu, return 0; break; case SMU_DCEFCLK: - dev_info(smu->adev->dev,"Setting DCEFCLK min/max dpm level is not supported!\n"); + dev_info(smu->adev->dev, "Setting DCEFCLK min/max dpm level is not supported!\n"); break; default: @@ -2182,7 +2182,7 @@ static int navi10_read_sensor(struct smu_context *smu, struct smu_table_context *table_context = &smu->smu_table; PPTable_t *pptable = table_context->driver_pptable; - if(!data || !size) + if (!data || !size) return -EINVAL; switch (sensor) { @@ -2202,7 +2202,7 @@ static int navi10_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = navi1x_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); @@ -2240,6 +2240,7 @@ static int navi10_read_sensor(struct smu_context *smu, ret = smu_v11_0_get_gfx_vdd(smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: default: ret = -EOPNOTSUPP; break; @@ -2317,15 +2318,15 @@ static int navi10_display_disable_memory_clock_switch(struct smu_context *smu, uint32_t min_memory_clock = smu->hard_min_uclk_req_from_dal; uint32_t max_memory_clock = max_sustainable_clocks->uclock; - if(smu->disable_uclk_switch == disable_memory_clock_switch) + if (smu->disable_uclk_switch == disable_memory_clock_switch) return 0; - if(disable_memory_clock_switch) + if (disable_memory_clock_switch) ret = smu_v11_0_set_hard_freq_limited_range(smu, SMU_UCLK, max_memory_clock, 0); else ret = smu_v11_0_set_hard_freq_limited_range(smu, SMU_UCLK, min_memory_clock, 0); - if(!ret) + if (!ret) smu->disable_uclk_switch = disable_memory_clock_switch; return ret; @@ -2559,7 +2560,8 @@ static int navi10_set_default_od_settings(struct smu_context *smu) return 0; } -static int navi10_od_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_TABLE_COMMAND type, long input[], uint32_t size) { +static int navi10_od_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_TABLE_COMMAND type, long input[], uint32_t size) +{ int i; int ret = 0; struct smu_table_context *table_context = &smu->smu_table; @@ -3368,7 +3370,7 @@ static ssize_t navi1x_get_gpu_metrics(struct smu_context *smu, ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 0)) && smu_version > 0x002A3B00)) ret = navi10_get_gpu_metrics(smu, table); else - ret =navi10_get_legacy_gpu_metrics(smu, table); + ret = navi10_get_legacy_gpu_metrics(smu, table); break; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index f6599c00a6fd..4bb289f9b4b8 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -588,7 +588,9 @@ err0_out: return -ENOMEM; } -static uint32_t sienna_cichlid_get_throttler_status_locked(struct smu_context *smu) +static uint32_t sienna_cichlid_get_throttler_status_locked(struct smu_context *smu, + bool use_metrics_v3, + bool use_metrics_v2) { struct smu_table_context *smu_table= &smu->smu_table; SmuMetricsExternal_t *metrics_ext = @@ -596,13 +598,11 @@ static uint32_t sienna_cichlid_get_throttler_status_locked(struct smu_context *s uint32_t throttler_status = 0; int i; - if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && - (smu->smc_fw_version >= 0x3A4900)) { + if (use_metrics_v3) { for (i = 0; i < THROTTLER_COUNT; i++) throttler_status |= (metrics_ext->SmuMetrics_V3.ThrottlingPercentage[i] ? 1U << i : 0); - } else if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(11, 0, 7)) && - (smu->smc_fw_version >= 0x3A4300)) { + } else if (use_metrics_v2) { for (i = 0; i < THROTTLER_COUNT; i++) throttler_status |= (metrics_ext->SmuMetrics_V2.ThrottlingPercentage[i] ? 1U << i : 0); @@ -864,7 +864,7 @@ static int sienna_cichlid_get_smu_metrics_data(struct smu_context *smu, metrics->TemperatureVrSoc) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_THROTTLER_STATUS: - *value = sienna_cichlid_get_throttler_status_locked(smu); + *value = sienna_cichlid_get_throttler_status_locked(smu, use_metrics_v3, use_metrics_v2); break; case METRICS_CURR_FANSPEED: *value = use_metrics_v3 ? metrics_v3->CurrFanSpeed : @@ -1902,7 +1902,7 @@ static int sienna_cichlid_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = sienna_cichlid_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); @@ -1927,12 +1927,16 @@ static int sienna_cichlid_read_sensor(struct smu_context *smu, *size = 4; break; case AMDGPU_PP_SENSOR_GFX_MCLK: - ret = sienna_cichlid_get_current_clk_freq_by_table(smu, SMU_UCLK, (uint32_t *)data); + ret = sienna_cichlid_get_smu_metrics_data(smu, + METRICS_CURR_UCLK, + (uint32_t *)data); *(uint32_t *)data *= 100; *size = 4; break; case AMDGPU_PP_SENSOR_GFX_SCLK: - ret = sienna_cichlid_get_current_clk_freq_by_table(smu, SMU_GFXCLK, (uint32_t *)data); + ret = sienna_cichlid_get_smu_metrics_data(smu, + METRICS_AVERAGE_GFXCLK, + (uint32_t *)data); *(uint32_t *)data *= 100; *size = 4; break; @@ -1958,6 +1962,7 @@ static int sienna_cichlid_read_sensor(struct smu_context *smu, ret = -EOPNOTSUPP; } break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: default: ret = -EOPNOTSUPP; break; @@ -4013,7 +4018,7 @@ static ssize_t sienna_cichlid_get_gpu_metrics(struct smu_context *smu, gpu_metrics->current_dclk1 = use_metrics_v3 ? metrics_v3->CurrClock[PPCLK_DCLK_1] : use_metrics_v2 ? metrics_v2->CurrClock[PPCLK_DCLK_1] : metrics->CurrClock[PPCLK_DCLK_1]; - gpu_metrics->throttle_status = sienna_cichlid_get_throttler_status_locked(smu); + gpu_metrics->throttle_status = sienna_cichlid_get_throttler_status_locked(smu, use_metrics_v3, use_metrics_v2); gpu_metrics->indep_throttle_status = smu_cmn_get_indep_throttler_status(gpu_metrics->throttle_status, sienna_cichlid_throttler_map); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index 067b4e0b026c..201cec599842 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -390,6 +390,10 @@ static int vangogh_get_smu_metrics_data(struct smu_context *smu, *value = metrics->Current.UvdActivity; break; case METRICS_AVERAGE_SOCKETPOWER: + *value = (metrics->Average.CurrentSocketPower << 8) / + 1000; + break; + case METRICS_CURR_SOCKETPOWER: *value = (metrics->Current.CurrentSocketPower << 8) / 1000; break; @@ -1536,12 +1540,18 @@ static int vangogh_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = vangogh_common_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: + ret = vangogh_common_get_smu_metrics_data(smu, + METRICS_CURR_SOCKETPOWER, + (uint32_t *)data); + *size = 4; + break; case AMDGPU_PP_SENSOR_EDGE_TEMP: ret = vangogh_common_get_smu_metrics_data(smu, METRICS_TEMPERATURE_EDGE, @@ -1854,6 +1864,86 @@ static ssize_t vangogh_get_gpu_metrics_v2_3(struct smu_context *smu, return sizeof(struct gpu_metrics_v2_3); } +static ssize_t vangogh_get_gpu_metrics_v2_4(struct smu_context *smu, + void **table) +{ + SmuMetrics_t metrics; + struct smu_table_context *smu_table = &smu->smu_table; + struct gpu_metrics_v2_4 *gpu_metrics = + (struct gpu_metrics_v2_4 *)smu_table->gpu_metrics_table; + int ret = 0; + + ret = smu_cmn_get_metrics_table(smu, &metrics, true); + if (ret) + return ret; + + smu_cmn_init_soft_gpu_metrics(gpu_metrics, 2, 4); + + gpu_metrics->temperature_gfx = metrics.Current.GfxTemperature; + gpu_metrics->temperature_soc = metrics.Current.SocTemperature; + memcpy(&gpu_metrics->temperature_core[0], + &metrics.Current.CoreTemperature[0], + sizeof(uint16_t) * 4); + gpu_metrics->temperature_l3[0] = metrics.Current.L3Temperature[0]; + + gpu_metrics->average_temperature_gfx = metrics.Average.GfxTemperature; + gpu_metrics->average_temperature_soc = metrics.Average.SocTemperature; + memcpy(&gpu_metrics->average_temperature_core[0], + &metrics.Average.CoreTemperature[0], + sizeof(uint16_t) * 4); + gpu_metrics->average_temperature_l3[0] = metrics.Average.L3Temperature[0]; + + gpu_metrics->average_gfx_activity = metrics.Current.GfxActivity; + gpu_metrics->average_mm_activity = metrics.Current.UvdActivity; + + gpu_metrics->average_socket_power = metrics.Current.CurrentSocketPower; + gpu_metrics->average_cpu_power = metrics.Current.Power[0]; + gpu_metrics->average_soc_power = metrics.Current.Power[1]; + gpu_metrics->average_gfx_power = metrics.Current.Power[2]; + + gpu_metrics->average_cpu_voltage = metrics.Current.Voltage[0]; + gpu_metrics->average_soc_voltage = metrics.Current.Voltage[1]; + gpu_metrics->average_gfx_voltage = metrics.Current.Voltage[2]; + + gpu_metrics->average_cpu_current = metrics.Current.Current[0]; + gpu_metrics->average_soc_current = metrics.Current.Current[1]; + gpu_metrics->average_gfx_current = metrics.Current.Current[2]; + + memcpy(&gpu_metrics->average_core_power[0], + &metrics.Average.CorePower[0], + sizeof(uint16_t) * 4); + + gpu_metrics->average_gfxclk_frequency = metrics.Average.GfxclkFrequency; + gpu_metrics->average_socclk_frequency = metrics.Average.SocclkFrequency; + gpu_metrics->average_uclk_frequency = metrics.Average.MemclkFrequency; + gpu_metrics->average_fclk_frequency = metrics.Average.MemclkFrequency; + gpu_metrics->average_vclk_frequency = metrics.Average.VclkFrequency; + gpu_metrics->average_dclk_frequency = metrics.Average.DclkFrequency; + + gpu_metrics->current_gfxclk = metrics.Current.GfxclkFrequency; + gpu_metrics->current_socclk = metrics.Current.SocclkFrequency; + gpu_metrics->current_uclk = metrics.Current.MemclkFrequency; + gpu_metrics->current_fclk = metrics.Current.MemclkFrequency; + gpu_metrics->current_vclk = metrics.Current.VclkFrequency; + gpu_metrics->current_dclk = metrics.Current.DclkFrequency; + + memcpy(&gpu_metrics->current_coreclk[0], + &metrics.Current.CoreFrequency[0], + sizeof(uint16_t) * 4); + gpu_metrics->current_l3clk[0] = metrics.Current.L3Frequency[0]; + + gpu_metrics->throttle_status = metrics.Current.ThrottlerStatus; + gpu_metrics->indep_throttle_status = + smu_cmn_get_indep_throttler_status(metrics.Current.ThrottlerStatus, + vangogh_throttler_map); + + gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); + + *table = (void *)gpu_metrics; + + return sizeof(struct gpu_metrics_v2_4); +} + static ssize_t vangogh_get_gpu_metrics(struct smu_context *smu, void **table) { @@ -1923,23 +2013,34 @@ static ssize_t vangogh_common_get_gpu_metrics(struct smu_context *smu, { uint32_t if_version; uint32_t smu_version; + uint32_t smu_program; + uint32_t fw_version; int ret = 0; ret = smu_cmn_get_smc_version(smu, &if_version, &smu_version); - if (ret) { + if (ret) return ret; - } - if (smu_version >= 0x043F3E00) { - if (if_version < 0x3) - ret = vangogh_get_legacy_gpu_metrics_v2_3(smu, table); + smu_program = (smu_version >> 24) & 0xff; + fw_version = smu_version & 0xffffff; + if (smu_program == 6) { + if (fw_version >= 0x3F0800) + ret = vangogh_get_gpu_metrics_v2_4(smu, table); else ret = vangogh_get_gpu_metrics_v2_3(smu, table); + } else { - if (if_version < 0x3) - ret = vangogh_get_legacy_gpu_metrics(smu, table); - else - ret = vangogh_get_gpu_metrics(smu, table); + if (smu_version >= 0x043F3E00) { + if (if_version < 0x3) + ret = vangogh_get_legacy_gpu_metrics_v2_3(smu, table); + else + ret = vangogh_get_gpu_metrics_v2_3(smu, table); + } else { + if (if_version < 0x3) + ret = vangogh_get_legacy_gpu_metrics(smu, table); + else + ret = vangogh_get_gpu_metrics(smu, table); + } } return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index 8a8ba25c9ad7..c8119491c516 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -262,15 +262,15 @@ static int renoir_get_profiling_clk_mask(struct smu_context *smu, /* mclk levels are in reverse order */ *mclk_mask = NUM_MEMCLK_DPM_LEVELS - 1; } else if (level == AMD_DPM_FORCED_LEVEL_PROFILE_PEAK) { - if(sclk_mask) + if (sclk_mask) /* The sclk as gfxclk and has three level about max/min/current */ *sclk_mask = 3 - 1; - if(mclk_mask) + if (mclk_mask) /* mclk levels are in reverse order */ *mclk_mask = 0; - if(soc_mask) + if (soc_mask) *soc_mask = NUM_SOCCLK_DPM_LEVELS - 1; } @@ -1197,7 +1197,7 @@ static int renoir_get_smu_metrics_data(struct smu_context *smu, case METRICS_AVERAGE_VCNACTIVITY: *value = metrics->AverageUvdActivity / 100; break; - case METRICS_AVERAGE_SOCKETPOWER: + case METRICS_CURR_SOCKETPOWER: if (((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 1)) && (adev->pm.fw_version >= 0x40000f)) || ((adev->ip_versions[MP1_HWIP][0] == IP_VERSION(12, 0, 0)) && (adev->pm.fw_version >= 0x373200))) *value = metrics->CurrentSocketPower << 8; @@ -1297,9 +1297,9 @@ static int renoir_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: ret = renoir_get_smu_metrics_data(smu, - METRICS_AVERAGE_SOCKETPOWER, + METRICS_CURR_SOCKETPOWER, (uint32_t *)data); *size = 4; break; @@ -1315,6 +1315,7 @@ static int renoir_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: default: ret = -EOPNOTSUPP; break; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c index c788aa7a99a9..5e408a195860 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c @@ -205,7 +205,8 @@ int smu_v12_0_set_default_dpm_tables(struct smu_context *smu) return smu_cmn_update_table(smu, SMU_TABLE_DPMCLOCKS, 0, smu_table->clocks_table, false); } -int smu_v12_0_mode2_reset(struct smu_context *smu){ +int smu_v12_0_mode2_reset(struct smu_context *smu) +{ return smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GfxDeviceDriverReset, SMU_RESET_MODE_2, NULL); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index ce50ef46e73f..cc3169400c9b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -94,8 +94,7 @@ */ #define SUPPORT_BAD_CHANNEL_INFO_MSG_VERSION 0x00443300 -static const struct smu_temperature_range smu13_thermal_policy[] = -{ +static const struct smu_temperature_range smu13_thermal_policy[] = { {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, }; @@ -196,7 +195,7 @@ static const struct cmn2asic_mapping aldebaran_feature_mask_map[SMU_FEATURE_COUN ALDEBARAN_FEA_MAP(SMU_FEATURE_FW_CTF_BIT, FEATURE_FW_CTF_BIT), ALDEBARAN_FEA_MAP(SMU_FEATURE_THERMAL_BIT, FEATURE_THERMAL_BIT), ALDEBARAN_FEA_MAP(SMU_FEATURE_OUT_OF_BAND_MONITOR_BIT, FEATURE_OUT_OF_BAND_MONITOR_BIT), - ALDEBARAN_FEA_MAP(SMU_FEATURE_XGMI_PER_LINK_PWR_DWN_BIT,FEATURE_XGMI_PER_LINK_PWR_DWN), + ALDEBARAN_FEA_MAP(SMU_FEATURE_XGMI_PER_LINK_PWR_DWN_BIT, FEATURE_XGMI_PER_LINK_PWR_DWN), ALDEBARAN_FEA_MAP(SMU_FEATURE_DF_CSTATE_BIT, FEATURE_DF_CSTATE), }; @@ -580,7 +579,7 @@ static int aldebaran_get_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_t *metrics = (SmuMetrics_t *)smu_table->metrics_table; int ret = 0; @@ -626,9 +625,10 @@ static int aldebaran_get_smu_metrics_data(struct smu_context *smu, break; case METRICS_AVERAGE_SOCKETPOWER: /* Valid power data is available only from primary die */ - *value = aldebaran_is_primary(smu) ? - metrics->AverageSocketPower << 8 : - 0; + if (aldebaran_is_primary(smu)) + *value = metrics->AverageSocketPower << 8; + else + ret = -EOPNOTSUPP; break; case METRICS_TEMPERATURE_EDGE: *value = metrics->TemperatureEdge * @@ -1095,16 +1095,6 @@ static int aldebaran_get_current_activity_percent(struct smu_context *smu, return ret; } -static int aldebaran_get_gpu_power(struct smu_context *smu, uint32_t *value) -{ - if (!value) - return -EINVAL; - - return aldebaran_get_smu_metrics_data(smu, - METRICS_AVERAGE_SOCKETPOWER, - value); -} - static int aldebaran_thermal_get_temperature(struct smu_context *smu, enum amd_pp_sensors sensor, uint32_t *value) @@ -1158,8 +1148,10 @@ static int aldebaran_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: - ret = aldebaran_get_gpu_power(smu, (uint32_t *)data); + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: + ret = aldebaran_get_smu_metrics_data(smu, + METRICS_AVERAGE_SOCKETPOWER, + (uint32_t *)data); *size = 4; break; case AMDGPU_PP_SENSOR_HOTSPOT_TEMP: @@ -1184,6 +1176,7 @@ static int aldebaran_read_sensor(struct smu_context *smu, ret = smu_v13_0_get_gfx_vdd(smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: default: ret = -EOPNOTSUPP; break; @@ -1906,8 +1899,7 @@ static int aldebaran_mode1_reset(struct smu_context *smu) smu_cmn_get_smc_version(smu, NULL, &smu_version); if (smu_version < 0x00440700) { ret = smu_cmn_send_smc_msg(smu, SMU_MSG_Mode1Reset, NULL); - } - else { + } else { /* fatal error triggered by ras, PMFW supports the flag from 68.44.0 */ if ((smu_version >= 0x00442c00) && ras && @@ -2116,7 +2108,7 @@ static const struct pptable_funcs aldebaran_ppt_funcs = { .register_irq_handler = smu_v13_0_register_irq_handler, .set_azalia_d3_pme = smu_v13_0_set_azalia_d3_pme, .get_max_sustainable_clocks_by_dc = smu_v13_0_get_max_sustainable_clocks_by_dc, - .baco_is_support= aldebaran_is_baco_supported, + .baco_is_support = aldebaran_is_baco_supported, .get_dpm_ultimate_freq = smu_v13_0_get_dpm_ultimate_freq, .set_soft_freq_limited_range = aldebaran_set_soft_freq_limited_range, .od_edit_dpm_table = aldebaran_usr_edit_dpm_table, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 9b62b45ebb7f..0232adb95df3 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -83,7 +83,6 @@ MODULE_FIRMWARE("amdgpu/smu_13_0_10.bin"); #define PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT 0xE static const int link_width[] = {0, 1, 2, 4, 8, 12, 16}; -static const int link_speed[] = {25, 50, 80, 160}; const int pmfw_decoded_link_speed[5] = {1, 2, 3, 4, 5}; const int pmfw_decoded_link_width[7] = {0, 1, 2, 4, 8, 12, 16}; @@ -838,12 +837,8 @@ int smu_v13_0_notify_display_change(struct smu_context *smu) { int ret = 0; - if (!smu->pm_enabled) - return ret; - - if (smu_cmn_feature_is_enabled(smu, SMU_FEATURE_DPM_UCLK_BIT) && - smu->adev->gmc.vram_type == AMDGPU_VRAM_TYPE_HBM) - ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetUclkFastSwitch, 1, NULL); + if (!amdgpu_device_has_dc_support(smu->adev)) + ret = smu_cmn_send_smc_msg(smu, SMU_MSG_DALNotPresent, NULL); return ret; } @@ -1121,7 +1116,7 @@ smu_v13_0_display_clock_voltage_request(struct smu_context *smu, ret = smu_v13_0_set_hard_freq_limited_range(smu, clk_select, clk_freq, 0); - if(clk_select == SMU_UCLK) + if (clk_select == SMU_UCLK) smu->hard_min_uclk_req_from_dal = clk_freq; } @@ -1437,8 +1432,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, return 0; } -static const struct amdgpu_irq_src_funcs smu_v13_0_irq_funcs = -{ +static const struct amdgpu_irq_src_funcs smu_v13_0_irq_funcs = { .set = smu_v13_0_set_irq_state, .process = smu_v13_0_irq_process, }; @@ -1933,7 +1927,7 @@ static int smu_v13_0_get_dpm_level_count(struct smu_context *smu, ret = smu_v13_0_get_dpm_freq_by_index(smu, clk_type, 0xff, value); /* SMU v13.0.2 FW returns 0 based max level, increment by one for it */ - if((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2)) && (!ret && value)) + if ((smu->adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2)) && (!ret && value)) ++(*value); return ret; @@ -2264,7 +2258,7 @@ int smu_v13_0_baco_set_state(struct smu_context *smu, if (state == SMU_BACO_STATE_ENTER) { ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_EnterBaco, - smu_baco->maco_support ? + (smu_baco->maco_support && amdgpu_runtime_pm != 1) ? BACO_SEQ_BAMACO : BACO_SEQ_BACO, NULL); } else { diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index b9bde5fa8f8f..3903a47669e4 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -96,6 +96,14 @@ */ #define SUPPORT_ECCTABLE_SMU_13_0_10_VERSION 0x00502200 +#define PP_OD_FEATURE_GFXCLK_FMIN 0 +#define PP_OD_FEATURE_GFXCLK_FMAX 1 +#define PP_OD_FEATURE_UCLK_FMIN 2 +#define PP_OD_FEATURE_UCLK_FMAX 3 +#define PP_OD_FEATURE_GFX_VF_CURVE 4 + +#define LINK_SPEED_MAX 3 + static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] = { MSG_MAP(TestMessage, PPSMC_MSG_TestMessage, 1), MSG_MAP(GetSmuVersion, PPSMC_MSG_GetSmuVersion, 1), @@ -154,6 +162,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_0_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(AllowGpo, PPSMC_MSG_SetGpoAllow, 0), MSG_MAP(AllowIHHostInterrupt, PPSMC_MSG_AllowIHHostInterrupt, 0), MSG_MAP(ReenableAcDcInterrupt, PPSMC_MSG_ReenableAcDcInterrupt, 0), + MSG_MAP(DALNotPresent, PPSMC_MSG_DALNotPresent, 0), }; static struct cmn2asic_mapping smu_v13_0_0_clk_map[SMU_CLK_COUNT] = { @@ -332,10 +341,13 @@ static int smu_v13_0_0_check_powerplay_table(struct smu_context *smu) table_context->power_play_table; struct smu_baco_context *smu_baco = &smu->smu_baco; PPTable_t *pptable = smu->smu_table.driver_pptable; +#if 0 + PPTable_t *pptable = smu->smu_table.driver_pptable; const OverDriveLimits_t * const overdrive_upperlimits = &pptable->SkuTable.OverDriveLimitsBasicMax; const OverDriveLimits_t * const overdrive_lowerlimits = &pptable->SkuTable.OverDriveLimitsMin; +#endif if (powerplay_table->platform_caps & SMU_13_0_0_PP_PLATFORM_CAP_HARDWAREDC) smu->dc_controlled_by_gpio = true; @@ -347,18 +359,30 @@ static int smu_v13_0_0_check_powerplay_table(struct smu_context *smu) if (powerplay_table->platform_caps & SMU_13_0_0_PP_PLATFORM_CAP_MACO) smu_baco->maco_support = true; + /* + * We are in the transition to a new OD mechanism. + * Disable the OD feature support for SMU13 temporarily. + * TODO: get this reverted when new OD mechanism online + */ +#if 0 if (!overdrive_lowerlimits->FeatureCtrlMask || !overdrive_upperlimits->FeatureCtrlMask) smu->od_enabled = false; - table_context->thermal_controller_type = - powerplay_table->thermal_controller_type; - /* * Instead of having its own buffer space and get overdrive_table copied, * smu->od_settings just points to the actual overdrive_table */ smu->od_settings = &powerplay_table->overdrive_table; +#else + smu->od_enabled = false; +#endif + + table_context->thermal_controller_type = + powerplay_table->thermal_controller_type; + + smu->adev->pm.no_fan = + !(pptable->SkuTable.FeaturesToRun[0] & (1 << FEATURE_FAN_CONTROL_BIT)); return 0; } @@ -932,7 +956,7 @@ static int smu_v13_0_0_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = smu_v13_0_0_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); @@ -976,6 +1000,7 @@ static int smu_v13_0_0_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: default: ret = -EOPNOTSUPP; break; @@ -1043,7 +1068,6 @@ static bool smu_v13_0_0_is_od_feature_supported(struct smu_context *smu, static void smu_v13_0_0_get_od_setting_limits(struct smu_context *smu, int od_feature_bit, - bool lower_boundary, int32_t *min, int32_t *max) { @@ -1055,29 +1079,28 @@ static void smu_v13_0_0_get_od_setting_limits(struct smu_context *smu, int32_t od_min_setting, od_max_setting; switch (od_feature_bit) { - case PP_OD_FEATURE_GFXCLK_BIT: - if (lower_boundary) { - od_min_setting = overdrive_lowerlimits->GfxclkFmin; - od_max_setting = overdrive_upperlimits->GfxclkFmin; - } else { - od_min_setting = overdrive_lowerlimits->GfxclkFmax; - od_max_setting = overdrive_upperlimits->GfxclkFmax; - } + case PP_OD_FEATURE_GFXCLK_FMIN: + od_min_setting = overdrive_lowerlimits->GfxclkFmin; + od_max_setting = overdrive_upperlimits->GfxclkFmin; break; - case PP_OD_FEATURE_UCLK_BIT: - if (lower_boundary) { - od_min_setting = overdrive_lowerlimits->UclkFmin; - od_max_setting = overdrive_upperlimits->UclkFmin; - } else { - od_min_setting = overdrive_lowerlimits->UclkFmax; - od_max_setting = overdrive_upperlimits->UclkFmax; - } + case PP_OD_FEATURE_GFXCLK_FMAX: + od_min_setting = overdrive_lowerlimits->GfxclkFmax; + od_max_setting = overdrive_upperlimits->GfxclkFmax; + break; + case PP_OD_FEATURE_UCLK_FMIN: + od_min_setting = overdrive_lowerlimits->UclkFmin; + od_max_setting = overdrive_upperlimits->UclkFmin; break; - case PP_OD_FEATURE_GFX_VF_CURVE_BIT: + case PP_OD_FEATURE_UCLK_FMAX: + od_min_setting = overdrive_lowerlimits->UclkFmax; + od_max_setting = overdrive_upperlimits->UclkFmax; + break; + case PP_OD_FEATURE_GFX_VF_CURVE: od_min_setting = overdrive_lowerlimits->VoltageOffsetPerZoneBoundary; od_max_setting = overdrive_upperlimits->VoltageOffsetPerZoneBoundary; break; default: + od_min_setting = od_max_setting = INT_MAX; break; } @@ -1140,7 +1163,6 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, (OverDriveTableExternal_t *)smu->smu_table.overdrive_table; struct smu_13_0_dpm_table *single_dpm_table; struct smu_13_0_pcie_table *pcie_table; - const int link_width[] = {0, 1, 2, 4, 8, 12, 16}; uint32_t gen_speed, lane_width; int i, curr_freq, size = 0; int32_t min_value, max_value; @@ -1256,7 +1278,7 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, (pcie_table->pcie_lane[i] == 6) ? "x16" : "", pcie_table->clk_freq[i], (gen_speed == DECODE_GEN_SPEED(pcie_table->pcie_gen[i])) && - (lane_width == DECODE_LANE_WIDTH(link_width[pcie_table->pcie_lane[i]])) ? + (lane_width == DECODE_LANE_WIDTH(pcie_table->pcie_lane[i])) ? "*" : ""); break; @@ -1304,13 +1326,11 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) { smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - true, + PP_OD_FEATURE_GFXCLK_FMIN, &min_value, NULL); smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - false, + PP_OD_FEATURE_GFXCLK_FMAX, NULL, &max_value); size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", @@ -1319,13 +1339,11 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) { smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - true, + PP_OD_FEATURE_UCLK_FMIN, &min_value, NULL); smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - false, + PP_OD_FEATURE_UCLK_FMAX, NULL, &max_value); size += sysfs_emit_at(buf, size, "MCLK: %7uMhz %10uMhz\n", @@ -1334,8 +1352,7 @@ static int smu_v13_0_0_print_clk_levels(struct smu_context *smu, if (smu_v13_0_0_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) { smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_GFX_VF_CURVE_BIT, - true, + PP_OD_FEATURE_GFX_VF_CURVE, &min_value, &max_value); size += sysfs_emit_at(buf, size, "VDDC_CURVE: %7dmv %10dmv\n", @@ -1359,7 +1376,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, OverDriveTableExternal_t *od_table = (OverDriveTableExternal_t *)table_context->overdrive_table; struct amdgpu_device *adev = smu->adev; - uint32_t offset_of_featurectrlmask; + uint32_t offset_of_voltageoffset; int32_t minimum, maximum; uint32_t feature_ctrlmask; int i, ret = 0; @@ -1380,8 +1397,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, switch (input[i]) { case 0: smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - true, + PP_OD_FEATURE_GFXCLK_FMIN, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1397,8 +1413,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, case 1: smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - false, + PP_OD_FEATURE_GFXCLK_FMAX, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1443,8 +1458,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, switch (input[i]) { case 0: smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - true, + PP_OD_FEATURE_UCLK_FMIN, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1460,8 +1474,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, case 1: smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - false, + PP_OD_FEATURE_UCLK_FMAX, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1502,8 +1515,7 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, return -EINVAL; smu_v13_0_0_get_od_setting_limits(smu, - PP_OD_FEATURE_GFX_VF_CURVE_BIT, - true, + PP_OD_FEATURE_GFX_VF_CURVE, &minimum, &maximum); if (input[1] < minimum || @@ -1533,10 +1545,10 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, * It does not contain actual informations about user's custom * settings. Thus we do not cache it. */ - offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask); - if (memcmp((u8 *)od_table + offset_of_featurectrlmask, - table_context->user_overdrive_table + offset_of_featurectrlmask, - sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) { + offset_of_voltageoffset = offsetof(OverDriveTable_t, VoltageOffsetPerZoneBoundary); + if (memcmp((u8 *)od_table + offset_of_voltageoffset, + table_context->user_overdrive_table + offset_of_voltageoffset, + sizeof(OverDriveTableExternal_t) - offset_of_voltageoffset)) { smu_v13_0_0_dump_od_table(smu, od_table); ret = smu_v13_0_0_upload_overdrive_table(smu, od_table); @@ -1546,9 +1558,9 @@ static int smu_v13_0_0_od_edit_dpm_table(struct smu_context *smu, } od_table->OverDriveTable.FeatureCtrlMask = 0; - memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask, - (u8 *)od_table + offset_of_featurectrlmask, - sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask); + memcpy(table_context->user_overdrive_table + offset_of_voltageoffset, + (u8 *)od_table + offset_of_voltageoffset, + sizeof(OverDriveTableExternal_t) - offset_of_voltageoffset); if (!memcmp(table_context->user_overdrive_table, table_context->boot_overdrive_table, @@ -1734,7 +1746,7 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu, gpu_metrics->average_vclk1_frequency = metrics->AverageVclk1Frequency; gpu_metrics->average_dclk1_frequency = metrics->AverageDclk1Frequency; - gpu_metrics->current_gfxclk = metrics->CurrClock[PPCLK_GFXCLK]; + gpu_metrics->current_gfxclk = gpu_metrics->average_gfxclk_frequency; gpu_metrics->current_socclk = metrics->CurrClock[PPCLK_SOCCLK]; gpu_metrics->current_uclk = metrics->CurrClock[PPCLK_UCLK]; gpu_metrics->current_vclk0 = metrics->CurrClock[PPCLK_VCLK_0]; @@ -1751,7 +1763,10 @@ static ssize_t smu_v13_0_0_get_gpu_metrics(struct smu_context *smu, gpu_metrics->current_fan_speed = metrics->AvgFanRpm; gpu_metrics->pcie_link_width = metrics->PcieWidth; - gpu_metrics->pcie_link_speed = metrics->PcieRate; + if ((metrics->PcieRate - 1) > LINK_SPEED_MAX) + gpu_metrics->pcie_link_speed = pcie_gen_to_speed(1); + else + gpu_metrics->pcie_link_speed = pcie_gen_to_speed(metrics->PcieRate); gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); @@ -2197,7 +2212,8 @@ static int smu_v13_0_0_baco_enter(struct smu_context *smu) if (adev->in_runpm && smu_cmn_is_audio_func_enabled(adev)) return smu_v13_0_baco_set_armd3_sequence(smu, - smu_baco->maco_support ? BACO_SEQ_BAMACO : BACO_SEQ_BACO); + (smu_baco->maco_support && amdgpu_runtime_pm != 1) ? + BACO_SEQ_BAMACO : BACO_SEQ_BACO); else return smu_v13_0_baco_enter(smu); } @@ -2672,6 +2688,7 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .send_hbm_bad_channel_flag = smu_v13_0_0_send_bad_mem_channel_flag, .gpo_control = smu_v13_0_gpo_control, .get_ecc_info = smu_v13_0_0_get_ecc_info, + .notify_display_change = smu_v13_0_notify_display_change, }; void smu_v13_0_0_set_ppt_funcs(struct smu_context *smu) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c index ef37dda9908f..626591f54bc4 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_4_ppt.c @@ -257,7 +257,7 @@ static ssize_t smu_v13_0_4_get_gpu_metrics(struct smu_context *smu, gpu_metrics->average_gfx_activity = metrics.GfxActivity; gpu_metrics->average_mm_activity = metrics.UvdActivity; - gpu_metrics->average_socket_power = metrics.CurrentSocketPower; + gpu_metrics->average_socket_power = metrics.AverageSocketPower; gpu_metrics->average_gfx_power = metrics.Power[0]; gpu_metrics->average_soc_power = metrics.Power[1]; memcpy(&gpu_metrics->average_core_power[0], @@ -321,6 +321,9 @@ static int smu_v13_0_4_get_smu_metrics_data(struct smu_context *smu, *value = metrics->UvdActivity; break; case METRICS_AVERAGE_SOCKETPOWER: + *value = (metrics->AverageSocketPower << 8) / 1000; + break; + case METRICS_CURR_SOCKETPOWER: *value = (metrics->CurrentSocketPower << 8) / 1000; break; case METRICS_TEMPERATURE_EDGE: @@ -569,12 +572,18 @@ static int smu_v13_0_4_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = smu_v13_0_4_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: + ret = smu_v13_0_4_get_smu_metrics_data(smu, + METRICS_CURR_SOCKETPOWER, + (uint32_t *)data); + *size = 4; + break; case AMDGPU_PP_SENSOR_EDGE_TEMP: ret = smu_v13_0_4_get_smu_metrics_data(smu, METRICS_TEMPERATURE_EDGE, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c index 42f110602eb1..c6e7c2115a26 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c @@ -75,7 +75,7 @@ static struct cmn2asic_msg_mapping smu_v13_0_5_message_map[SMU_MSG_MAX_COUNT] = MSG_MAP(SetDriverDramAddrHigh, PPSMC_MSG_SetDriverDramAddrHigh, 1), MSG_MAP(SetDriverDramAddrLow, PPSMC_MSG_SetDriverDramAddrLow, 1), MSG_MAP(TransferTableSmu2Dram, PPSMC_MSG_TransferTableSmu2Dram, 1), - MSG_MAP(TransferTableDram2Smu, PPSMC_MSG_TransferTableDram2Smu , 1), + MSG_MAP(TransferTableDram2Smu, PPSMC_MSG_TransferTableDram2Smu, 1), MSG_MAP(GetGfxclkFrequency, PPSMC_MSG_GetGfxclkFrequency, 1), MSG_MAP(GetEnabledSmuFeatures, PPSMC_MSG_GetEnabledSmuFeatures, 1), MSG_MAP(SetSoftMaxVcn, PPSMC_MSG_SetSoftMaxVcn, 1), @@ -288,7 +288,7 @@ static int smu_v13_0_5_get_smu_metrics_data(struct smu_context *smu, case METRICS_AVERAGE_VCNACTIVITY: *value = metrics->UvdActivity; break; - case METRICS_AVERAGE_SOCKETPOWER: + case METRICS_CURR_SOCKETPOWER: *value = (metrics->CurrentSocketPower << 8) / 1000; break; case METRICS_TEMPERATURE_EDGE: @@ -332,9 +332,9 @@ static int smu_v13_0_5_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: ret = smu_v13_0_5_get_smu_metrics_data(smu, - METRICS_AVERAGE_SOCKETPOWER, + METRICS_CURR_SOCKETPOWER, (uint32_t *)data); *size = 4; break; @@ -388,6 +388,7 @@ static int smu_v13_0_5_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: default: ret = -EOPNOTSUPP; break; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index 1ac552142763..de80e191a92c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -80,17 +80,23 @@ /* possible frequency drift (1Mhz) */ #define EPSILON 1 -#define smnPCIE_ESM_CTRL 0x193D0 -#define smnPCIE_LC_LINK_WIDTH_CNTL 0x1ab40288 +#define smnPCIE_ESM_CTRL 0x93D0 +#define smnPCIE_LC_LINK_WIDTH_CNTL 0x1a340288 #define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK 0x00000070L #define PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT 0x4 +#define MAX_LINK_WIDTH 6 + +#define smnPCIE_LC_SPEED_CNTL 0x1a340290 +#define PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK 0xE0 +#define PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT 0x5 +#define LINK_SPEED_MAX 4 static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COUNT] = { MSG_MAP(TestMessage, PPSMC_MSG_TestMessage, 0), MSG_MAP(GetSmuVersion, PPSMC_MSG_GetSmuVersion, 1), MSG_MAP(GetDriverIfVersion, PPSMC_MSG_GetDriverIfVersion, 1), - MSG_MAP(EnableAllSmuFeatures, PPSMC_MSG_EnableAllSmuFeatures, 1), - MSG_MAP(DisableAllSmuFeatures, PPSMC_MSG_DisableAllSmuFeatures, 1), + MSG_MAP(EnableAllSmuFeatures, PPSMC_MSG_EnableAllSmuFeatures, 0), + MSG_MAP(DisableAllSmuFeatures, PPSMC_MSG_DisableAllSmuFeatures, 0), MSG_MAP(RequestI2cTransaction, PPSMC_MSG_RequestI2cTransaction, 0), MSG_MAP(GetMetricsTable, PPSMC_MSG_GetMetricsTable, 1), MSG_MAP(GetEnabledSmuFeaturesHigh, PPSMC_MSG_GetEnabledSmuFeaturesHigh, 1), @@ -101,8 +107,8 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU MSG_MAP(SetToolsDramAddrLow, PPSMC_MSG_SetToolsDramAddrLow, 0), MSG_MAP(SetSoftMinByFreq, PPSMC_MSG_SetSoftMinByFreq, 0), MSG_MAP(SetSoftMaxByFreq, PPSMC_MSG_SetSoftMaxByFreq, 0), - MSG_MAP(GetMinDpmFreq, PPSMC_MSG_GetMinDpmFreq, 0), - MSG_MAP(GetMaxDpmFreq, PPSMC_MSG_GetMaxDpmFreq, 0), + MSG_MAP(GetMinDpmFreq, PPSMC_MSG_GetMinDpmFreq, 1), + MSG_MAP(GetMaxDpmFreq, PPSMC_MSG_GetMaxDpmFreq, 1), MSG_MAP(GetDpmFreqByIndex, PPSMC_MSG_GetDpmFreqByIndex, 1), MSG_MAP(SetPptLimit, PPSMC_MSG_SetPptLimit, 0), MSG_MAP(GetPptLimit, PPSMC_MSG_GetPptLimit, 1), @@ -121,11 +127,12 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU MSG_MAP(EnableDeterminism, PPSMC_MSG_EnableDeterminism, 0), MSG_MAP(DisableDeterminism, PPSMC_MSG_DisableDeterminism, 0), MSG_MAP(GfxDriverResetRecovery, PPSMC_MSG_GfxDriverResetRecovery, 0), - MSG_MAP(GetMinGfxclkFrequency, PPSMC_MSG_GetMinGfxDpmFreq, 0), - MSG_MAP(GetMaxGfxclkFrequency, PPSMC_MSG_GetMaxGfxDpmFreq, 0), + MSG_MAP(GetMinGfxclkFrequency, PPSMC_MSG_GetMinGfxDpmFreq, 1), + MSG_MAP(GetMaxGfxclkFrequency, PPSMC_MSG_GetMaxGfxDpmFreq, 1), MSG_MAP(SetSoftMinGfxclk, PPSMC_MSG_SetSoftMinGfxClk, 0), MSG_MAP(SetSoftMaxGfxClk, PPSMC_MSG_SetSoftMaxGfxClk, 0), MSG_MAP(PrepareMp1ForUnload, PPSMC_MSG_PrepareForDriverUnload, 0), + MSG_MAP(GetCTFLimit, PPSMC_MSG_GetCTFLimit, 0), }; static const struct cmn2asic_mapping smu_v13_0_6_clk_map[SMU_CLK_COUNT] = { @@ -325,14 +332,24 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) MetricsTable_t *metrics = (MetricsTable_t *)smu_table->metrics_table; struct PPTable_t *pptable = (struct PPTable_t *)smu_table->driver_pptable; - int ret; - int i; + int ret, i, retry = 100; /* Store one-time values in driver PPTable */ if (!pptable->Init) { - ret = smu_v13_0_6_get_metrics_table(smu, NULL, false); - if (ret) - return ret; + while (--retry) { + ret = smu_v13_0_6_get_metrics_table(smu, NULL, true); + if (ret) + return ret; + + /* Ensure that metrics have been updated */ + if (metrics->AccumulationCounter) + break; + + usleep_range(1000, 1100); + } + + if (!retry) + return -ETIME; pptable->MaxSocketPowerLimit = SMUQ10_TO_UINT(metrics->MaxSocketPowerLimit); @@ -704,20 +721,23 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, case METRICS_AVERAGE_MEMACTIVITY: *value = SMUQ10_TO_UINT(metrics->DramBandwidthUtilization); break; - case METRICS_AVERAGE_SOCKETPOWER: + case METRICS_CURR_SOCKETPOWER: *value = SMUQ10_TO_UINT(metrics->SocketPower) << 8; break; case METRICS_TEMPERATURE_HOTSPOT: - *value = SMUQ10_TO_UINT(metrics->MaxSocketTemperature); + *value = SMUQ10_TO_UINT(metrics->MaxSocketTemperature) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_MEM: - *value = SMUQ10_TO_UINT(metrics->MaxHbmTemperature); + *value = SMUQ10_TO_UINT(metrics->MaxHbmTemperature) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; /* This is the max of all VRs and not just SOC VR. * No need to define another data type for the same. */ case METRICS_TEMPERATURE_VRSOC: - *value = SMUQ10_TO_UINT(metrics->MaxVrTemperature); + *value = SMUQ10_TO_UINT(metrics->MaxVrTemperature) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; default: *value = UINT_MAX; @@ -772,8 +792,6 @@ static int smu_v13_0_6_print_clk_levels(struct smu_context *smu, struct smu_13_0_dpm_table *single_dpm_table; struct smu_dpm_context *smu_dpm = &smu->smu_dpm; struct smu_13_0_dpm_context *dpm_context = NULL; - uint32_t display_levels; - uint32_t freq_values[3] = { 0 }; uint32_t min_clk, max_clk; smu_cmn_get_sysfs_buf(&buf, &size); @@ -798,50 +816,24 @@ static int smu_v13_0_6_print_clk_levels(struct smu_context *smu, return ret; } - single_dpm_table = &(dpm_context->dpm_tables.gfx_table); - ret = smu_v13_0_6_get_clk_table(smu, &clocks, single_dpm_table); - if (ret) { - dev_err(smu->adev->dev, - "Attempt to get gfx clk levels Failed!"); - return ret; - } - - display_levels = clocks.num_levels; - min_clk = pstate_table->gfxclk_pstate.curr.min; max_clk = pstate_table->gfxclk_pstate.curr.max; - freq_values[0] = min_clk; - freq_values[1] = max_clk; - - /* fine-grained dpm has only 2 levels */ - if (now > min_clk && now < max_clk) { - display_levels = clocks.num_levels + 1; - freq_values[2] = max_clk; - freq_values[1] = now; - } - - /* - * For DPM disabled case, there will be only one clock level. - * And it's safe to assume that is always the current clock. - */ - if (display_levels == clocks.num_levels) { - for (i = 0; i < clocks.num_levels; i++) - size += sysfs_emit_at( - buf, size, "%d: %uMhz %s\n", i, - freq_values[i], - (clocks.num_levels == 1) ? - "*" : - (smu_v13_0_6_freqs_in_same_level( - freq_values[i], now) ? - "*" : - "")); + if (!smu_v13_0_6_freqs_in_same_level(now, min_clk) && + !smu_v13_0_6_freqs_in_same_level(now, max_clk)) { + size += sysfs_emit_at(buf, size, "0: %uMhz\n", + min_clk); + size += sysfs_emit_at(buf, size, "1: %uMhz *\n", + now); + size += sysfs_emit_at(buf, size, "2: %uMhz\n", + max_clk); } else { - for (i = 0; i < display_levels; i++) - size += sysfs_emit_at(buf, size, - "%d: %uMhz %s\n", i, - freq_values[i], - i == 1 ? "*" : ""); + size += sysfs_emit_at(buf, size, "0: %uMhz %s\n", + min_clk, + smu_v13_0_6_freqs_in_same_level(now, min_clk) ? "*" : ""); + size += sysfs_emit_at(buf, size, "1: %uMhz %s\n", + max_clk, + smu_v13_0_6_freqs_in_same_level(now, max_clk) ? "*" : ""); } break; @@ -1154,15 +1146,6 @@ static int smu_v13_0_6_get_current_activity_percent(struct smu_context *smu, return ret; } -static int smu_v13_0_6_get_gpu_power(struct smu_context *smu, uint32_t *value) -{ - if (!value) - return -EINVAL; - - return smu_v13_0_6_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, - value); -} - static int smu_v13_0_6_thermal_get_temperature(struct smu_context *smu, enum amd_pp_sensors sensor, uint32_t *value) @@ -1208,8 +1191,10 @@ static int smu_v13_0_6_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: - ret = smu_v13_0_6_get_gpu_power(smu, (uint32_t *)data); + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: + ret = smu_v13_0_6_get_smu_metrics_data(smu, + METRICS_CURR_SOCKETPOWER, + (uint32_t *)data); *size = 4; break; case AMDGPU_PP_SENSOR_HOTSPOT_TEMP: @@ -1235,6 +1220,7 @@ static int smu_v13_0_6_read_sensor(struct smu_context *smu, ret = smu_v13_0_get_gfx_vdd(smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: default: ret = -EOPNOTSUPP; break; @@ -1248,9 +1234,9 @@ static int smu_v13_0_6_get_power_limit(struct smu_context *smu, uint32_t *default_power_limit, uint32_t *max_power_limit) { - struct smu_table_context *smu_table = &smu->smu_table; - struct PPTable_t *pptable = - (struct PPTable_t *)smu_table->driver_pptable; + struct smu_table_context *smu_table = &smu->smu_table; + struct PPTable_t *pptable = + (struct PPTable_t *)smu_table->driver_pptable; uint32_t power_limit = 0; int ret; @@ -1366,8 +1352,7 @@ static int smu_v13_0_6_set_irq_state(struct amdgpu_device *adev, return 0; } -static const struct amdgpu_irq_src_funcs smu_v13_0_6_irq_funcs = -{ +static const struct amdgpu_irq_src_funcs smu_v13_0_6_irq_funcs = { .set = smu_v13_0_6_set_irq_state, .process = smu_v13_0_6_irq_process, }; @@ -1414,6 +1399,9 @@ static int smu_v13_0_6_system_features_control(struct smu_context *smu, struct amdgpu_device *adev = smu->adev; int ret = 0; + if (amdgpu_sriov_vf(adev)) + return 0; + if (enable) { if (!(adev->flags & AMD_IS_APU)) ret = smu_v13_0_system_features_control(smu, enable); @@ -1948,6 +1936,7 @@ smu_v13_0_6_get_current_pcie_link_width_level(struct smu_context *smu) static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; + uint32_t speed_level; uint32_t esm_ctrl; /* TODO: confirm this on real target */ @@ -1955,7 +1944,13 @@ static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu) if ((esm_ctrl >> 15) & 0x1FFFF) return (((esm_ctrl >> 8) & 0x3F) + 128); - return smu_v13_0_get_current_pcie_link_speed(smu); + speed_level = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) & + PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK) + >> PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT; + if (speed_level > LINK_SPEED_MAX) + speed_level = 0; + + return pcie_gen_to_speed(speed_level + 1); } static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table) @@ -1966,6 +1961,7 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table struct amdgpu_device *adev = smu->adev; int ret = 0, inst0, xcc0; MetricsTable_t *metrics; + u16 link_width_level; inst0 = adev->sdma.instance[0].aid_id; xcc0 = GET_INST(GC, 0); @@ -1993,9 +1989,8 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table gpu_metrics->average_socket_power = SMUQ10_TO_UINT(metrics->SocketPower); - /* Energy is reported in 15.625mJ units */ - gpu_metrics->energy_accumulator = - SMUQ10_TO_UINT(metrics->SocketEnergyAcc); + /* Energy counter reported in 15.259uJ (2^-16) units */ + gpu_metrics->energy_accumulator = metrics->SocketEnergyAcc; gpu_metrics->current_gfxclk = SMUQ10_TO_UINT(metrics->GfxclkFrequency[xcc0]); @@ -2017,8 +2012,12 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table gpu_metrics->throttle_status = 0; if (!(adev->flags & AMD_IS_APU)) { + link_width_level = smu_v13_0_6_get_current_pcie_link_width_level(smu); + if (link_width_level > MAX_LINK_WIDTH) + link_width_level = 0; + gpu_metrics->pcie_link_width = - smu_v13_0_6_get_current_pcie_link_width_level(smu); + DECODE_LANE_WIDTH(link_width_level); gpu_metrics->pcie_link_speed = smu_v13_0_6_get_current_pcie_link_speed(smu); } @@ -2083,6 +2082,55 @@ out: return ret; } +static int smu_v13_0_6_get_thermal_temperature_range(struct smu_context *smu, + struct smu_temperature_range *range) +{ + struct amdgpu_device *adev = smu->adev; + u32 aid_temp, xcd_temp, mem_temp; + uint32_t smu_version; + u32 ccd_temp = 0; + int ret; + + if (amdgpu_sriov_vf(smu->adev)) + return 0; + + if (!range) + return -EINVAL; + + /*Check smu version, GetCtfLimit message only supported for smu version 85.69 or higher */ + smu_cmn_get_smc_version(smu, NULL, &smu_version); + if (smu_version < 0x554500) + return 0; + + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GetCTFLimit, + PPSMC_AID_THM_TYPE, &aid_temp); + if (ret) + goto failed; + + if (adev->flags & AMD_IS_APU) { + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GetCTFLimit, + PPSMC_CCD_THM_TYPE, &ccd_temp); + if (ret) + goto failed; + } + + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GetCTFLimit, + PPSMC_XCD_THM_TYPE, &xcd_temp); + if (ret) + goto failed; + + range->hotspot_crit_max = max3(aid_temp, xcd_temp, ccd_temp) * + SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_GetCTFLimit, + PPSMC_HBM_THM_TYPE, &mem_temp); + if (ret) + goto failed; + + range->mem_crit_max = mem_temp * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; +failed: + return ret; +} + static int smu_v13_0_6_mode1_reset(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; @@ -2110,8 +2158,7 @@ static int smu_v13_0_6_mode1_reset(struct smu_context *smu) static bool smu_v13_0_6_is_mode1_reset_supported(struct smu_context *smu) { - /* TODO: Enable this when FW support is added */ - return false; + return true; } static bool smu_v13_0_6_is_mode2_reset_supported(struct smu_context *smu) @@ -2179,6 +2226,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { .get_pp_feature_mask = smu_cmn_get_pp_feature_mask, .set_pp_feature_mask = smu_cmn_set_pp_feature_mask, .get_gpu_metrics = smu_v13_0_6_get_gpu_metrics, + .get_thermal_temperature_range = smu_v13_0_6_get_thermal_temperature_range, .mode1_reset_is_support = smu_v13_0_6_is_mode1_reset_supported, .mode2_reset_is_support = smu_v13_0_6_is_mode2_reset_supported, .mode1_reset = smu_v13_0_6_mode1_reset, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index 3ba02131e682..94ef5b4d116d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -72,6 +72,14 @@ #define MP0_MP1_DATA_REGION_SIZE_COMBOPPTABLE 0x4000 +#define PP_OD_FEATURE_GFXCLK_FMIN 0 +#define PP_OD_FEATURE_GFXCLK_FMAX 1 +#define PP_OD_FEATURE_UCLK_FMIN 2 +#define PP_OD_FEATURE_UCLK_FMAX 3 +#define PP_OD_FEATURE_GFX_VF_CURVE 4 + +#define LINK_SPEED_MAX 3 + static struct cmn2asic_msg_mapping smu_v13_0_7_message_map[SMU_MSG_MAX_COUNT] = { MSG_MAP(TestMessage, PPSMC_MSG_TestMessage, 1), MSG_MAP(GetSmuVersion, PPSMC_MSG_GetSmuVersion, 1), @@ -323,10 +331,12 @@ static int smu_v13_0_7_check_powerplay_table(struct smu_context *smu) struct smu_baco_context *smu_baco = &smu->smu_baco; PPTable_t *smc_pptable = table_context->driver_pptable; BoardTable_t *BoardTable = &smc_pptable->BoardTable; +#if 0 const OverDriveLimits_t * const overdrive_upperlimits = &smc_pptable->SkuTable.OverDriveLimitsBasicMax; const OverDriveLimits_t * const overdrive_lowerlimits = &smc_pptable->SkuTable.OverDriveLimitsMin; +#endif if (powerplay_table->platform_caps & SMU_13_0_7_PP_PLATFORM_CAP_HARDWAREDC) smu->dc_controlled_by_gpio = true; @@ -338,18 +348,22 @@ static int smu_v13_0_7_check_powerplay_table(struct smu_context *smu) if (smu_baco->platform_support && (BoardTable->HsrEnabled || BoardTable->VddqOffEnabled)) smu_baco->maco_support = true; +#if 0 if (!overdrive_lowerlimits->FeatureCtrlMask || !overdrive_upperlimits->FeatureCtrlMask) smu->od_enabled = false; - table_context->thermal_controller_type = - powerplay_table->thermal_controller_type; - /* * Instead of having its own buffer space and get overdrive_table copied, * smu->od_settings just points to the actual overdrive_table */ smu->od_settings = &powerplay_table->overdrive_table; +#else + smu->od_enabled = false; +#endif + + table_context->thermal_controller_type = + powerplay_table->thermal_controller_type; return 0; } @@ -386,8 +400,7 @@ static int smu_v13_0_7_check_fw_status(struct smu_context *smu) } #ifndef atom_smc_dpm_info_table_13_0_7 -struct atom_smc_dpm_info_table_13_0_7 -{ +struct atom_smc_dpm_info_table_13_0_7 { struct atom_common_table_header table_header; BoardTable_t BoardTable; }; @@ -494,7 +507,7 @@ static int smu_v13_0_7_tables_init(struct smu_context *smu) PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_ACTIVITY_MONITOR_COEFF, sizeof(DpmActivityMonitorCoeffIntExternal_t), PAGE_SIZE, - AMDGPU_GEM_DOMAIN_VRAM); + AMDGPU_GEM_DOMAIN_VRAM); SMU_TABLE_INIT(tables, SMU_TABLE_COMBO_PPTABLE, MP0_MP1_DATA_REGION_SIZE_COMBOPPTABLE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM); @@ -728,7 +741,7 @@ static int smu_v13_0_7_get_smu_metrics_data(struct smu_context *smu, MetricsMember_t member, uint32_t *value) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; SmuMetrics_t *metrics = &(((SmuMetricsExternal_t *)(smu_table->metrics_table))->SmuMetrics); int ret = 0; @@ -923,7 +936,7 @@ static int smu_v13_0_7_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: ret = smu_v13_0_7_get_smu_metrics_data(smu, METRICS_AVERAGE_SOCKETPOWER, (uint32_t *)data); @@ -949,7 +962,7 @@ static int smu_v13_0_7_read_sensor(struct smu_context *smu, break; case AMDGPU_PP_SENSOR_GFX_MCLK: ret = smu_v13_0_7_get_smu_metrics_data(smu, - METRICS_AVERAGE_UCLK, + METRICS_CURR_UCLK, (uint32_t *)data); *(uint32_t *)data *= 100; *size = 4; @@ -967,6 +980,7 @@ static int smu_v13_0_7_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: default: ret = -EOPNOTSUPP; break; @@ -1034,7 +1048,6 @@ static bool smu_v13_0_7_is_od_feature_supported(struct smu_context *smu, static void smu_v13_0_7_get_od_setting_limits(struct smu_context *smu, int od_feature_bit, - bool lower_boundary, int32_t *min, int32_t *max) { @@ -1046,29 +1059,28 @@ static void smu_v13_0_7_get_od_setting_limits(struct smu_context *smu, int32_t od_min_setting, od_max_setting; switch (od_feature_bit) { - case PP_OD_FEATURE_GFXCLK_BIT: - if (lower_boundary) { - od_min_setting = overdrive_lowerlimits->GfxclkFmin; - od_max_setting = overdrive_upperlimits->GfxclkFmin; - } else { - od_min_setting = overdrive_lowerlimits->GfxclkFmax; - od_max_setting = overdrive_upperlimits->GfxclkFmax; - } + case PP_OD_FEATURE_GFXCLK_FMIN: + od_min_setting = overdrive_lowerlimits->GfxclkFmin; + od_max_setting = overdrive_upperlimits->GfxclkFmin; break; - case PP_OD_FEATURE_UCLK_BIT: - if (lower_boundary) { - od_min_setting = overdrive_lowerlimits->UclkFmin; - od_max_setting = overdrive_upperlimits->UclkFmin; - } else { - od_min_setting = overdrive_lowerlimits->UclkFmax; - od_max_setting = overdrive_upperlimits->UclkFmax; - } + case PP_OD_FEATURE_GFXCLK_FMAX: + od_min_setting = overdrive_lowerlimits->GfxclkFmax; + od_max_setting = overdrive_upperlimits->GfxclkFmax; + break; + case PP_OD_FEATURE_UCLK_FMIN: + od_min_setting = overdrive_lowerlimits->UclkFmin; + od_max_setting = overdrive_upperlimits->UclkFmin; + break; + case PP_OD_FEATURE_UCLK_FMAX: + od_min_setting = overdrive_lowerlimits->UclkFmax; + od_max_setting = overdrive_upperlimits->UclkFmax; break; - case PP_OD_FEATURE_GFX_VF_CURVE_BIT: + case PP_OD_FEATURE_GFX_VF_CURVE: od_min_setting = overdrive_lowerlimits->VoltageOffsetPerZoneBoundary; od_max_setting = overdrive_upperlimits->VoltageOffsetPerZoneBoundary; break; default: + od_min_setting = od_max_setting = INT_MAX; break; } @@ -1294,13 +1306,11 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu, if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFXCLK_BIT)) { smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - true, + PP_OD_FEATURE_GFXCLK_FMIN, &min_value, NULL); smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - false, + PP_OD_FEATURE_GFXCLK_FMAX, NULL, &max_value); size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", @@ -1309,13 +1319,11 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu, if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_UCLK_BIT)) { smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - true, + PP_OD_FEATURE_UCLK_FMIN, &min_value, NULL); smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - false, + PP_OD_FEATURE_UCLK_FMAX, NULL, &max_value); size += sysfs_emit_at(buf, size, "MCLK: %7uMhz %10uMhz\n", @@ -1324,8 +1332,7 @@ static int smu_v13_0_7_print_clk_levels(struct smu_context *smu, if (smu_v13_0_7_is_od_feature_supported(smu, PP_OD_FEATURE_GFX_VF_CURVE_BIT)) { smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_GFX_VF_CURVE_BIT, - true, + PP_OD_FEATURE_GFX_VF_CURVE, &min_value, &max_value); size += sysfs_emit_at(buf, size, "VDDC_CURVE: %7dmv %10dmv\n", @@ -1349,7 +1356,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, OverDriveTableExternal_t *od_table = (OverDriveTableExternal_t *)table_context->overdrive_table; struct amdgpu_device *adev = smu->adev; - uint32_t offset_of_featurectrlmask; + uint32_t offset_of_voltageoffset; int32_t minimum, maximum; uint32_t feature_ctrlmask; int i, ret = 0; @@ -1370,8 +1377,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, switch (input[i]) { case 0: smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - true, + PP_OD_FEATURE_GFXCLK_FMIN, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1387,8 +1393,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, case 1: smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_BIT, - false, + PP_OD_FEATURE_GFXCLK_FMAX, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1433,8 +1438,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, switch (input[i]) { case 0: smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - true, + PP_OD_FEATURE_UCLK_FMIN, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1450,8 +1454,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, case 1: smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_UCLK_BIT, - false, + PP_OD_FEATURE_UCLK_FMAX, &minimum, &maximum); if (input[i + 1] < minimum || @@ -1492,8 +1495,7 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, return -EINVAL; smu_v13_0_7_get_od_setting_limits(smu, - PP_OD_FEATURE_GFX_VF_CURVE_BIT, - true, + PP_OD_FEATURE_GFX_VF_CURVE, &minimum, &maximum); if (input[1] < minimum || @@ -1523,10 +1525,10 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, * It does not contain actual informations about user's custom * settings. Thus we do not cache it. */ - offset_of_featurectrlmask = offsetof(OverDriveTable_t, FeatureCtrlMask); - if (memcmp((u8 *)od_table + offset_of_featurectrlmask, - table_context->user_overdrive_table + offset_of_featurectrlmask, - sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask)) { + offset_of_voltageoffset = offsetof(OverDriveTable_t, VoltageOffsetPerZoneBoundary); + if (memcmp((u8 *)od_table + offset_of_voltageoffset, + table_context->user_overdrive_table + offset_of_voltageoffset, + sizeof(OverDriveTableExternal_t) - offset_of_voltageoffset)) { smu_v13_0_7_dump_od_table(smu, od_table); ret = smu_v13_0_7_upload_overdrive_table(smu, od_table); @@ -1536,9 +1538,9 @@ static int smu_v13_0_7_od_edit_dpm_table(struct smu_context *smu, } od_table->OverDriveTable.FeatureCtrlMask = 0; - memcpy(table_context->user_overdrive_table + offset_of_featurectrlmask, - (u8 *)od_table + offset_of_featurectrlmask, - sizeof(OverDriveTableExternal_t) - offset_of_featurectrlmask); + memcpy(table_context->user_overdrive_table + offset_of_voltageoffset, + (u8 *)od_table + offset_of_voltageoffset, + sizeof(OverDriveTableExternal_t) - offset_of_voltageoffset); if (!memcmp(table_context->user_overdrive_table, table_context->boot_overdrive_table, @@ -1635,8 +1637,7 @@ static int smu_v13_0_7_force_clk_levels(struct smu_context *smu, return ret; } -static const struct smu_temperature_range smu13_thermal_policy[] = -{ +static const struct smu_temperature_range smu13_thermal_policy[] = { {-273150, 99000, 99000, -273150, 99000, 99000, -273150, 99000, 99000}, { 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000, 120000}, }; @@ -1737,7 +1738,10 @@ static ssize_t smu_v13_0_7_get_gpu_metrics(struct smu_context *smu, gpu_metrics->current_fan_speed = metrics->AvgFanRpm; gpu_metrics->pcie_link_width = metrics->PcieWidth; - gpu_metrics->pcie_link_speed = metrics->PcieRate; + if ((metrics->PcieRate - 1) > LINK_SPEED_MAX) + gpu_metrics->pcie_link_speed = pcie_gen_to_speed(1); + else + gpu_metrics->pcie_link_speed = pcie_gen_to_speed(metrics->PcieRate); gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); @@ -2135,7 +2139,8 @@ static int smu_v13_0_7_baco_enter(struct smu_context *smu) if (adev->in_runpm && smu_cmn_is_audio_func_enabled(adev)) return smu_v13_0_baco_set_armd3_sequence(smu, - smu_baco->maco_support ? BACO_SEQ_BAMACO : BACO_SEQ_BACO); + (smu_baco->maco_support && amdgpu_runtime_pm != 1) ? + BACO_SEQ_BAMACO : BACO_SEQ_BACO); else return smu_v13_0_baco_enter(smu); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c index a1be2029ba4a..2e74d749efdd 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c @@ -365,7 +365,7 @@ static int yellow_carp_get_smu_metrics_data(struct smu_context *smu, case METRICS_AVERAGE_VCNACTIVITY: *value = metrics->UvdActivity; break; - case METRICS_AVERAGE_SOCKETPOWER: + case METRICS_CURR_SOCKETPOWER: *value = (metrics->CurrentSocketPower << 8) / 1000; break; case METRICS_TEMPERATURE_EDGE: @@ -423,9 +423,9 @@ static int yellow_carp_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; - case AMDGPU_PP_SENSOR_GPU_POWER: + case AMDGPU_PP_SENSOR_GPU_INPUT_POWER: ret = yellow_carp_get_smu_metrics_data(smu, - METRICS_AVERAGE_SOCKETPOWER, + METRICS_CURR_SOCKETPOWER, (uint32_t *)data); *size = 4; break; @@ -479,6 +479,7 @@ static int yellow_carp_read_sensor(struct smu_context *smu, (uint32_t *)data); *size = 4; break; + case AMDGPU_PP_SENSOR_GPU_AVG_POWER: default: ret = -EOPNOTSUPP; break; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 3ecb900e6ecd..12618a583e97 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -39,6 +39,8 @@ #define MP1_C2PMSG_90__CONTENT_MASK 0xFFFFFFFFL +const int link_speed[] = {25, 50, 80, 160, 320, 640}; + #undef __SMU_DUMMY_MAP #define __SMU_DUMMY_MAP(type) #type static const char * const __smu_message_names[] = { @@ -691,7 +693,7 @@ int smu_cmn_feature_set_enabled(struct smu_context *smu, #undef __SMU_DUMMY_MAP #define __SMU_DUMMY_MAP(fea) #fea -static const char* __smu_feature_names[] = { +static const char *__smu_feature_names[] = { SMU_FEATURE_MASKS }; @@ -927,7 +929,7 @@ int smu_cmn_get_metrics_table(struct smu_context *smu, void *metrics_table, bool bypass_cache) { - struct smu_table_context *smu_table= &smu->smu_table; + struct smu_table_context *smu_table = &smu->smu_table; uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size; int ret = 0; @@ -969,7 +971,7 @@ void smu_cmn_init_soft_gpu_metrics(void *table, uint8_t frev, uint8_t crev) struct metrics_table_header *header = (struct metrics_table_header *)table; uint16_t structure_size; -#define METRICS_VERSION(a, b) ((a << 16) | b ) +#define METRICS_VERSION(a, b) ((a << 16) | b) switch (METRICS_VERSION(frev, crev)) { case METRICS_VERSION(1, 0): @@ -996,6 +998,9 @@ void smu_cmn_init_soft_gpu_metrics(void *table, uint8_t frev, uint8_t crev) case METRICS_VERSION(2, 3): structure_size = sizeof(struct gpu_metrics_v2_3); break; + case METRICS_VERSION(2, 4): + structure_size = sizeof(struct gpu_metrics_v2_4); + break; default: return; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h index d7cd358a53bd..cc590e27d88a 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h @@ -30,6 +30,14 @@ #define FDO_PWM_MODE_STATIC 1 #define FDO_PWM_MODE_STATIC_RPM 5 +extern const int link_speed[]; + +/* Helper to Convert from PCIE Gen 1/2/3/4/5/6 to 0.1 GT/s speed units */ +static inline int pcie_gen_to_speed(uint32_t gen) +{ + return ((gen == 0) ? link_speed[0] : link_speed[gen - 1]); +} + int smu_cmn_send_msg_without_waiting(struct smu_context *smu, uint16_t msg_index, uint32_t param); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h index ceb13c838067..bcc42abfc768 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h @@ -61,14 +61,14 @@ #define smu_feature_get_enabled_mask(smu, mask) smu_ppt_funcs(get_enabled_mask, -EOPNOTSUPP, smu, mask) #define smu_feature_is_enabled(smu, mask) smu_ppt_funcs(feature_is_enabled, 0, smu, mask) #define smu_disable_all_features_with_exception(smu, mask) smu_ppt_funcs(disable_all_features_with_exception, 0, smu, mask) -#define smu_is_dpm_running(smu) smu_ppt_funcs(is_dpm_running, 0 , smu) +#define smu_is_dpm_running(smu) smu_ppt_funcs(is_dpm_running, 0, smu) #define smu_notify_display_change(smu) smu_ppt_funcs(notify_display_change, 0, smu) #define smu_populate_umd_state_clk(smu) smu_ppt_funcs(populate_umd_state_clk, 0, smu) #define smu_enable_thermal_alert(smu) smu_ppt_funcs(enable_thermal_alert, 0, smu) #define smu_disable_thermal_alert(smu) smu_ppt_funcs(disable_thermal_alert, 0, smu) #define smu_smc_read_sensor(smu, sensor, data, size) smu_ppt_funcs(read_sensor, -EINVAL, smu, sensor, data, size) #define smu_pre_display_config_changed(smu) smu_ppt_funcs(pre_display_config_changed, 0, smu) -#define smu_display_config_changed(smu) smu_ppt_funcs(display_config_changed, 0 , smu) +#define smu_display_config_changed(smu) smu_ppt_funcs(display_config_changed, 0, smu) #define smu_apply_clocks_adjust_rules(smu) smu_ppt_funcs(apply_clocks_adjust_rules, 0, smu) #define smu_notify_smc_display_config(smu) smu_ppt_funcs(notify_smc_display_config, 0, smu) #define smu_run_btc(smu) smu_ppt_funcs(run_btc, 0, smu) diff --git a/drivers/gpu/drm/arm/display/include/malidp_utils.h b/drivers/gpu/drm/arm/display/include/malidp_utils.h index 49a1d7f3539c..9f83baac6ed8 100644 --- a/drivers/gpu/drm/arm/display/include/malidp_utils.h +++ b/drivers/gpu/drm/arm/display/include/malidp_utils.h @@ -35,7 +35,7 @@ static inline void set_range(struct malidp_range *rg, u32 start, u32 end) rg->end = end; } -static inline bool in_range(struct malidp_range *rg, u32 v) +static inline bool malidp_in_range(struct malidp_range *rg, u32 v) { return (v >= rg->start) && (v <= rg->end); } diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c index 916f2c36bf2f..f3e744172673 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_pipeline_state.c @@ -305,12 +305,12 @@ komeda_layer_check_cfg(struct komeda_layer *layer, if (komeda_fb_check_src_coords(kfb, src_x, src_y, src_w, src_h)) return -EINVAL; - if (!in_range(&layer->hsize_in, src_w)) { + if (!malidp_in_range(&layer->hsize_in, src_w)) { DRM_DEBUG_ATOMIC("invalidate src_w %d.\n", src_w); return -EINVAL; } - if (!in_range(&layer->vsize_in, src_h)) { + if (!malidp_in_range(&layer->vsize_in, src_h)) { DRM_DEBUG_ATOMIC("invalidate src_h %d.\n", src_h); return -EINVAL; } @@ -452,14 +452,14 @@ komeda_scaler_check_cfg(struct komeda_scaler *scaler, hsize_out = dflow->out_w; vsize_out = dflow->out_h; - if (!in_range(&scaler->hsize, hsize_in) || - !in_range(&scaler->hsize, hsize_out)) { + if (!malidp_in_range(&scaler->hsize, hsize_in) || + !malidp_in_range(&scaler->hsize, hsize_out)) { DRM_DEBUG_ATOMIC("Invalid horizontal sizes"); return -EINVAL; } - if (!in_range(&scaler->vsize, vsize_in) || - !in_range(&scaler->vsize, vsize_out)) { + if (!malidp_in_range(&scaler->vsize, vsize_in) || + !malidp_in_range(&scaler->vsize, vsize_out)) { DRM_DEBUG_ATOMIC("Invalid vertical sizes"); return -EINVAL; } @@ -574,13 +574,13 @@ komeda_splitter_validate(struct komeda_splitter *splitter, return -EINVAL; } - if (!in_range(&splitter->hsize, dflow->in_w)) { + if (!malidp_in_range(&splitter->hsize, dflow->in_w)) { DRM_DEBUG_ATOMIC("split in_w:%d is out of the acceptable range.\n", dflow->in_w); return -EINVAL; } - if (!in_range(&splitter->vsize, dflow->in_h)) { + if (!malidp_in_range(&splitter->vsize, dflow->in_h)) { DRM_DEBUG_ATOMIC("split in_h: %d exceeds the acceptable range.\n", dflow->in_h); return -EINVAL; @@ -624,13 +624,13 @@ komeda_merger_validate(struct komeda_merger *merger, return -EINVAL; } - if (!in_range(&merger->hsize_merged, output->out_w)) { + if (!malidp_in_range(&merger->hsize_merged, output->out_w)) { DRM_DEBUG_ATOMIC("merged_w: %d is out of the accepted range.\n", output->out_w); return -EINVAL; } - if (!in_range(&merger->vsize_merged, output->out_h)) { + if (!malidp_in_range(&merger->vsize_merged, output->out_h)) { DRM_DEBUG_ATOMIC("merged_h: %d is out of the accepted range.\n", output->out_h); return -EINVAL; @@ -866,8 +866,8 @@ void komeda_complete_data_flow_cfg(struct komeda_layer *layer, * input/output range. */ if (dflow->en_scaling && scaler) - dflow->en_split = !in_range(&scaler->hsize, dflow->in_w) || - !in_range(&scaler->hsize, dflow->out_w); + dflow->en_split = !malidp_in_range(&scaler->hsize, dflow->in_w) || + !malidp_in_range(&scaler->hsize, dflow->out_w); } static bool merger_is_available(struct komeda_pipeline *pipe, diff --git a/drivers/gpu/drm/bridge/ite-it6505.c b/drivers/gpu/drm/bridge/ite-it6505.c index 6c2fcd8b8780..2f300f5ca051 100644 --- a/drivers/gpu/drm/bridge/ite-it6505.c +++ b/drivers/gpu/drm/bridge/ite-it6505.c @@ -2517,9 +2517,11 @@ static irqreturn_t it6505_int_threaded_handler(int unused, void *data) }; int int_status[3], i; - if (it6505->enable_drv_hold || pm_runtime_get_if_in_use(dev) <= 0) + if (it6505->enable_drv_hold || !it6505->powered) return IRQ_HANDLED; + pm_runtime_get_sync(dev); + int_status[0] = it6505_read(it6505, INT_STATUS_01); int_status[1] = it6505_read(it6505, INT_STATUS_02); int_status[2] = it6505_read(it6505, INT_STATUS_03); diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c b/drivers/gpu/drm/bridge/lontium-lt9611.c index 5163e5224aad..9663601ce098 100644 --- a/drivers/gpu/drm/bridge/lontium-lt9611.c +++ b/drivers/gpu/drm/bridge/lontium-lt9611.c @@ -774,9 +774,7 @@ static struct mipi_dsi_device *lt9611_attach_dsi(struct lt9611 *lt9611, dsi->lanes = 4; dsi->format = MIPI_DSI_FMT_RGB888; dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_SYNC_PULSE | - MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_VIDEO_NO_HSA | - MIPI_DSI_MODE_VIDEO_NO_HFP | MIPI_DSI_MODE_VIDEO_NO_HBP | - MIPI_DSI_MODE_NO_EOT_PACKET; + MIPI_DSI_MODE_VIDEO_HSE; ret = devm_mipi_dsi_attach(dev, dsi); if (ret < 0) { diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c index b1df91e37b1b..19bdb32dbc9a 100644 --- a/drivers/gpu/drm/bridge/samsung-dsim.c +++ b/drivers/gpu/drm/bridge/samsung-dsim.c @@ -1401,6 +1401,18 @@ static void samsung_dsim_disable_irq(struct samsung_dsim *dsi) disable_irq(dsi->irq); } +static void samsung_dsim_set_stop_state(struct samsung_dsim *dsi, bool enable) +{ + u32 reg = samsung_dsim_read(dsi, DSIM_ESCMODE_REG); + + if (enable) + reg |= DSIM_FORCE_STOP_STATE; + else + reg &= ~DSIM_FORCE_STOP_STATE; + + samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg); +} + static int samsung_dsim_init(struct samsung_dsim *dsi) { const struct samsung_dsim_driver_data *driver_data = dsi->driver_data; @@ -1460,15 +1472,12 @@ static void samsung_dsim_atomic_enable(struct drm_bridge *bridge, struct drm_bridge_state *old_bridge_state) { struct samsung_dsim *dsi = bridge_to_dsi(bridge); - u32 reg; if (samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) { samsung_dsim_set_display_mode(dsi); samsung_dsim_set_display_enable(dsi, true); } else { - reg = samsung_dsim_read(dsi, DSIM_ESCMODE_REG); - reg &= ~DSIM_FORCE_STOP_STATE; - samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg); + samsung_dsim_set_stop_state(dsi, false); } dsi->state |= DSIM_STATE_VIDOUT_AVAILABLE; @@ -1478,16 +1487,12 @@ static void samsung_dsim_atomic_disable(struct drm_bridge *bridge, struct drm_bridge_state *old_bridge_state) { struct samsung_dsim *dsi = bridge_to_dsi(bridge); - u32 reg; if (!(dsi->state & DSIM_STATE_ENABLED)) return; - if (!samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) { - reg = samsung_dsim_read(dsi, DSIM_ESCMODE_REG); - reg |= DSIM_FORCE_STOP_STATE; - samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg); - } + if (!samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) + samsung_dsim_set_stop_state(dsi, true); dsi->state &= ~DSIM_STATE_VIDOUT_AVAILABLE; } @@ -1790,6 +1795,8 @@ static ssize_t samsung_dsim_host_transfer(struct mipi_dsi_host *host, if (ret) return ret; + samsung_dsim_set_stop_state(dsi, false); + ret = mipi_dsi_create_packet(&xfer.packet, msg); if (ret < 0) return ret; diff --git a/drivers/gpu/drm/ci/arm.config b/drivers/gpu/drm/ci/arm.config new file mode 100644 index 000000000000..871f4de063ad --- /dev/null +++ b/drivers/gpu/drm/ci/arm.config @@ -0,0 +1,69 @@ +CONFIG_LOCALVERSION_AUTO=y +CONFIG_DEBUG_KERNEL=y + +CONFIG_CRYPTO_ZSTD=y +CONFIG_ZRAM_MEMORY_TRACKING=y +CONFIG_ZRAM_WRITEBACK=y +CONFIG_ZRAM=y +CONFIG_ZSMALLOC_STAT=y + +# abootimg with a 'dummy' rootfs fails with root=/dev/nfs +CONFIG_BLK_DEV_INITRD=n + +CONFIG_DEVFREQ_GOV_PERFORMANCE=y +CONFIG_DEVFREQ_GOV_POWERSAVE=y +CONFIG_DEVFREQ_GOV_USERSPACE=y +CONFIG_DEVFREQ_GOV_PASSIVE=y +CONFIG_DEVFREQ_GOV_SIMPLE_ONDEMAND=y + +CONFIG_DRM=y +CONFIG_DRM_ETNAVIV=y +CONFIG_DRM_ROCKCHIP=y +CONFIG_DRM_PANFROST=y +CONFIG_DRM_LIMA=y +CONFIG_DRM_PANEL_SIMPLE=y +CONFIG_PWM_CROS_EC=y +CONFIG_BACKLIGHT_PWM=y + +CONFIG_ROCKCHIP_CDN_DP=n + +CONFIG_SPI_ROCKCHIP=y +CONFIG_PWM_ROCKCHIP=y +CONFIG_PHY_ROCKCHIP_DP=y +CONFIG_DWMAC_ROCKCHIP=y + +CONFIG_MFD_RK808=y +CONFIG_REGULATOR_RK808=y +CONFIG_RTC_DRV_RK808=y +CONFIG_COMMON_CLK_RK808=y + +CONFIG_REGULATOR_FAN53555=y +CONFIG_REGULATOR=y + +CONFIG_REGULATOR_VCTRL=y + +CONFIG_KASAN=n +CONFIG_KASAN_INLINE=n +CONFIG_STACKTRACE=n + +CONFIG_TMPFS=y + +CONFIG_PROVE_LOCKING=n +CONFIG_DEBUG_LOCKDEP=n +CONFIG_SOFTLOCKUP_DETECTOR=n +CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC=n + +CONFIG_FW_LOADER_COMPRESS=y + +CONFIG_USB_USBNET=y +CONFIG_NETDEVICES=y +CONFIG_USB_NET_DRIVERS=y +CONFIG_USB_RTL8152=y +CONFIG_USB_NET_AX8817X=y +CONFIG_USB_NET_SMSC95XX=y + +# TK1 +CONFIG_ARM_TEGRA_DEVFREQ=y + +# 32-bit build failure +CONFIG_DRM_MSM=n diff --git a/drivers/gpu/drm/ci/arm64.config b/drivers/gpu/drm/ci/arm64.config new file mode 100644 index 000000000000..817e18ddfd4f --- /dev/null +++ b/drivers/gpu/drm/ci/arm64.config @@ -0,0 +1,199 @@ +CONFIG_LOCALVERSION_AUTO=y +CONFIG_DEBUG_KERNEL=y + +CONFIG_CRYPTO_ZSTD=y +CONFIG_ZRAM_MEMORY_TRACKING=y +CONFIG_ZRAM_WRITEBACK=y +CONFIG_ZRAM=y +CONFIG_ZSMALLOC_STAT=y + +# abootimg with a 'dummy' rootfs fails with root=/dev/nfs +CONFIG_BLK_DEV_INITRD=n + +CONFIG_DEVFREQ_GOV_PERFORMANCE=y +CONFIG_DEVFREQ_GOV_POWERSAVE=y +CONFIG_DEVFREQ_GOV_USERSPACE=y +CONFIG_DEVFREQ_GOV_PASSIVE=y + +CONFIG_DRM=y +CONFIG_DRM_ROCKCHIP=y +CONFIG_DRM_PANFROST=y +CONFIG_DRM_LIMA=y +CONFIG_DRM_PANEL_SIMPLE=y +CONFIG_DRM_PANEL_EDP=y +CONFIG_DRM_MSM=y +CONFIG_DRM_ETNAVIV=y +CONFIG_DRM_I2C_ADV7511=y +CONFIG_PWM_CROS_EC=y +CONFIG_BACKLIGHT_PWM=y + +CONFIG_ROCKCHIP_CDN_DP=n + +CONFIG_SPI_ROCKCHIP=y +CONFIG_PWM_ROCKCHIP=y +CONFIG_PHY_ROCKCHIP_DP=y +CONFIG_DWMAC_ROCKCHIP=y +CONFIG_STMMAC_ETH=y +CONFIG_TYPEC_FUSB302=y +CONFIG_TYPEC=y +CONFIG_TYPEC_TCPM=y + +# MSM platform bits + +# For CONFIG_QCOM_LMH +CONFIG_OF=y + +CONFIG_ARM_SMMU_QCOM=y +CONFIG_QCOM_COMMAND_DB=y +CONFIG_QCOM_RPMHPD=y +CONFIG_QCOM_RPMPD=y +CONFIG_QCOM_OCMEM=y +CONFIG_SDM_GPUCC_845=y +CONFIG_SDM_VIDEOCC_845=y +CONFIG_SDM_DISPCC_845=y +CONFIG_SDM_LPASSCC_845=y +CONFIG_SDM_CAMCC_845=y +CONFIG_RESET_QCOM_PDC=y +CONFIG_DRM_TI_SN65DSI86=y +CONFIG_I2C_QCOM_GENI=y +CONFIG_SPI_QCOM_GENI=y +CONFIG_PHY_QCOM_QUSB2=y +CONFIG_PHY_QCOM_QMP=y +CONFIG_MSM_GCC_8996=y +CONFIG_QCOM_CLK_APCC_MSM8996=y +CONFIG_QCOM_LLCC=y +CONFIG_QCOM_LMH=y +CONFIG_QCOM_SPMI_TEMP_ALARM=y +CONFIG_QCOM_WDT=y +CONFIG_POWER_RESET_QCOM_PON=y +CONFIG_RTC_DRV_PM8XXX=y +CONFIG_INTERCONNECT=y +CONFIG_INTERCONNECT_QCOM=y +CONFIG_INTERCONNECT_QCOM_MSM8996=y +CONFIG_INTERCONNECT_QCOM_SDM845=y +CONFIG_INTERCONNECT_QCOM_MSM8916=y +CONFIG_INTERCONNECT_QCOM_MSM8996=y +CONFIG_INTERCONNECT_QCOM_OSM_L3=y +CONFIG_INTERCONNECT_QCOM_SC7180=y +CONFIG_INTERCONNECT_QCOM_SM8350=y +CONFIG_CRYPTO_DEV_QCOM_RNG=y +CONFIG_SC_DISPCC_7180=y +CONFIG_SC_GPUCC_7180=y +CONFIG_SM_GPUCC_8350=y +CONFIG_QCOM_SPMI_ADC5=y +CONFIG_DRM_PARADE_PS8640=y +CONFIG_DRM_LONTIUM_LT9611UXC=y +CONFIG_PHY_QCOM_USB_HS=y +CONFIG_QCOM_GPI_DMA=y +CONFIG_USB_ONBOARD_HUB=y +CONFIG_NVMEM_QCOM_QFPROM=y +CONFIG_PHY_QCOM_USB_SNPS_FEMTO_V2=y + + +# db410c ethernet +CONFIG_USB_RTL8152=y +# db820c ethernet +CONFIG_ATL1C=y +# Chromebooks ethernet +CONFIG_USB_ONBOARD_HUB=y +# 888 HDK ethernet +CONFIG_USB_LAN78XX=y + +CONFIG_ARCH_ALPINE=n +CONFIG_ARCH_BCM2835=y +CONFIG_ARCH_BCM_IPROC=n +CONFIG_ARCH_BERLIN=n +CONFIG_ARCH_BRCMSTB=n +CONFIG_ARCH_EXYNOS=n +CONFIG_ARCH_K3=n +CONFIG_ARCH_LAYERSCAPE=n +CONFIG_ARCH_LG1K=n +CONFIG_ARCH_HISI=n +CONFIG_ARCH_MVEBU=n +CONFIG_ARCH_SEATTLE=n +CONFIG_ARCH_SYNQUACER=n +CONFIG_ARCH_RENESAS=n +CONFIG_ARCH_R8A774A1=n +CONFIG_ARCH_R8A774C0=n +CONFIG_ARCH_R8A7795=n +CONFIG_ARCH_R8A7796=n +CONFIG_ARCH_R8A77965=n +CONFIG_ARCH_R8A77970=n +CONFIG_ARCH_R8A77980=n +CONFIG_ARCH_R8A77990=n +CONFIG_ARCH_R8A77995=n +CONFIG_ARCH_STRATIX10=n +CONFIG_ARCH_TEGRA=n +CONFIG_ARCH_SPRD=n +CONFIG_ARCH_THUNDER=n +CONFIG_ARCH_THUNDER2=n +CONFIG_ARCH_UNIPHIER=n +CONFIG_ARCH_VEXPRESS=n +CONFIG_ARCH_XGENE=n +CONFIG_ARCH_ZX=n +CONFIG_ARCH_ZYNQMP=n + +# Strip out some stuff we don't need for graphics testing, to reduce +# the build. +CONFIG_CAN=n +CONFIG_WIRELESS=n +CONFIG_RFKILL=n +CONFIG_WLAN=n + +CONFIG_REGULATOR_FAN53555=y +CONFIG_REGULATOR=y + +CONFIG_REGULATOR_VCTRL=y + +CONFIG_KASAN=n +CONFIG_KASAN_INLINE=n +CONFIG_STACKTRACE=n + +CONFIG_TMPFS=y + +CONFIG_PROVE_LOCKING=n +CONFIG_DEBUG_LOCKDEP=n +CONFIG_SOFTLOCKUP_DETECTOR=y +CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC=y + +CONFIG_DETECT_HUNG_TASK=y + +CONFIG_FW_LOADER_COMPRESS=y +CONFIG_FW_LOADER_USER_HELPER=n + +CONFIG_USB_USBNET=y +CONFIG_NETDEVICES=y +CONFIG_USB_NET_DRIVERS=y +CONFIG_USB_RTL8152=y +CONFIG_USB_NET_AX8817X=y +CONFIG_USB_NET_SMSC95XX=y + +# For amlogic +CONFIG_MESON_GXL_PHY=y +CONFIG_MDIO_BUS_MUX_MESON_G12A=y +CONFIG_DRM_MESON=y + +# For Mediatek +CONFIG_DRM_MEDIATEK=y +CONFIG_PWM_MEDIATEK=y +CONFIG_DRM_MEDIATEK_HDMI=y +CONFIG_GNSS=y +CONFIG_GNSS_MTK_SERIAL=y +CONFIG_HW_RANDOM=y +CONFIG_HW_RANDOM_MTK=y +CONFIG_MTK_DEVAPC=y +CONFIG_PWM_MTK_DISP=y +CONFIG_MTK_CMDQ=y + +# For nouveau. Note that DRM must be a module so that it's loaded after NFS is up to provide the firmware. +CONFIG_ARCH_TEGRA=y +CONFIG_DRM_NOUVEAU=m +CONFIG_DRM_TEGRA=m +CONFIG_R8169=y +CONFIG_STAGING=y +CONFIG_DRM_TEGRA_STAGING=y +CONFIG_TEGRA_HOST1X=y +CONFIG_ARM_TEGRA_DEVFREQ=y +CONFIG_TEGRA_SOCTHERM=y +CONFIG_DRM_TEGRA_DEBUG=y +CONFIG_PWM_TEGRA=y diff --git a/drivers/gpu/drm/ci/build-igt.sh b/drivers/gpu/drm/ci/build-igt.sh new file mode 100644 index 000000000000..500fa4f5c30a --- /dev/null +++ b/drivers/gpu/drm/ci/build-igt.sh @@ -0,0 +1,35 @@ +#!/bin/bash +# SPDX-License-Identifier: MIT + +set -ex + +git clone https://gitlab.freedesktop.org/drm/igt-gpu-tools.git --single-branch --no-checkout +cd igt-gpu-tools +git checkout $IGT_VERSION + +if [[ "$KERNEL_ARCH" = "arm" ]]; then + . ../.gitlab-ci/container/create-cross-file.sh armhf + EXTRA_MESON_ARGS="--cross-file /cross_file-armhf.txt" +fi + +MESON_OPTIONS="-Doverlay=disabled \ + -Dchamelium=disabled \ + -Dvalgrind=disabled \ + -Dman=enabled \ + -Dtests=enabled \ + -Drunner=enabled \ + -Dlibunwind=enabled \ + -Dprefix=/igt" + +mkdir -p /igt +meson build $MESON_OPTIONS $EXTRA_MESON_ARGS +ninja -C build -j${FDO_CI_CONCURRENT:-4} || ninja -C build -j 1 +ninja -C build install + +mkdir -p artifacts/ +tar -cf artifacts/igt.tar /igt + +# Pass needed files to the test stage +S3_ARTIFACT_NAME="igt.tar.gz" +gzip -c artifacts/igt.tar > ${S3_ARTIFACT_NAME} +ci-fairy s3cp --token-file "${CI_JOB_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${KERNEL_ARCH}/${S3_ARTIFACT_NAME} diff --git a/drivers/gpu/drm/ci/build.sh b/drivers/gpu/drm/ci/build.sh new file mode 100644 index 000000000000..7b014287a041 --- /dev/null +++ b/drivers/gpu/drm/ci/build.sh @@ -0,0 +1,157 @@ +#!/bin/bash +# SPDX-License-Identifier: MIT + +set -ex + +# Clean up stale rebases that GitLab might not have removed when reusing a checkout dir +rm -rf .git/rebase-apply + +. .gitlab-ci/container/container_pre_build.sh + +# libssl-dev was uninstalled because it was considered an ephemeral package +apt-get update +apt-get install -y libssl-dev + +if [[ "$KERNEL_ARCH" = "arm64" ]]; then + GCC_ARCH="aarch64-linux-gnu" + DEBIAN_ARCH="arm64" + DEVICE_TREES="arch/arm64/boot/dts/rockchip/rk3399-gru-kevin.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/amlogic/meson-gxl-s805x-libretech-ac.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/allwinner/sun50i-h6-pine-h64.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/amlogic/meson-gxm-khadas-vim2.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/qcom/apq8016-sbc.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/qcom/apq8096-db820c.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/amlogic/meson-g12b-a311d-khadas-vim3.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/mediatek/mt8173-elm-hana.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/mediatek/mt8183-kukui-jacuzzi-juniper-sku16.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/mediatek/mt8192-asurada-spherion-r0.dtb" + DEVICE_TREES+=" arch/arm64/boot/dts/qcom/sc7180-trogdor-lazor-limozeen-nots-r5.dtb" +elif [[ "$KERNEL_ARCH" = "arm" ]]; then + GCC_ARCH="arm-linux-gnueabihf" + DEBIAN_ARCH="armhf" + DEVICE_TREES="arch/arm/boot/dts/rockchip/rk3288-veyron-jaq.dtb" + DEVICE_TREES+=" arch/arm/boot/dts/allwinner/sun8i-h3-libretech-all-h3-cc.dtb" + DEVICE_TREES+=" arch/arm/boot/dts/nxp/imx/imx6q-cubox-i.dtb" + apt-get install -y libssl-dev:armhf +else + GCC_ARCH="x86_64-linux-gnu" + DEBIAN_ARCH="x86_64" + DEVICE_TREES="" +fi + +export ARCH=${KERNEL_ARCH} +export CROSS_COMPILE="${GCC_ARCH}-" + +# The kernel doesn't like the gold linker (or the old lld in our debians). +# Sneak in some override symlinks during kernel build until we can update +# debian. +mkdir -p ld-links +for i in /usr/bin/*-ld /usr/bin/ld; do + i=$(basename $i) + ln -sf /usr/bin/$i.bfd ld-links/$i +done + +NEWPATH=$(pwd)/ld-links +export PATH=$NEWPATH:$PATH + +git config --global user.email "fdo@example.com" +git config --global user.name "freedesktop.org CI" +git config --global pull.rebase true + +# Try to merge fixes from target repo +if [ "$(git ls-remote --exit-code --heads ${UPSTREAM_REPO} ${TARGET_BRANCH}-external-fixes)" ]; then + git pull ${UPSTREAM_REPO} ${TARGET_BRANCH}-external-fixes +fi + +# Try to merge fixes from local repo if this isn't a merge request +if [ -z "$CI_MERGE_REQUEST_PROJECT_PATH" ]; then + if [ "$(git ls-remote --exit-code --heads origin ${TARGET_BRANCH}-external-fixes)" ]; then + git pull origin ${TARGET_BRANCH}-external-fixes + fi +fi + +for opt in $ENABLE_KCONFIGS; do + echo CONFIG_$opt=y >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config +done +for opt in $DISABLE_KCONFIGS; do + echo CONFIG_$opt=n >> drivers/gpu/drm/ci/${KERNEL_ARCH}.config +done + +if [[ -n "${MERGE_FRAGMENT}" ]]; then + ./scripts/kconfig/merge_config.sh ${DEFCONFIG} drivers/gpu/drm/ci/${MERGE_FRAGMENT} +else + make `basename ${DEFCONFIG}` +fi + +make ${KERNEL_IMAGE_NAME} + +mkdir -p /lava-files/ +for image in ${KERNEL_IMAGE_NAME}; do + cp arch/${KERNEL_ARCH}/boot/${image} /lava-files/. +done + +if [[ -n ${DEVICE_TREES} ]]; then + make dtbs + cp ${DEVICE_TREES} /lava-files/. +fi + +make modules +mkdir -p install/modules/ +INSTALL_MOD_PATH=install/modules/ make modules_install + +if [[ ${DEBIAN_ARCH} = "arm64" ]]; then + make Image.lzma + mkimage \ + -f auto \ + -A arm \ + -O linux \ + -d arch/arm64/boot/Image.lzma \ + -C lzma\ + -b arch/arm64/boot/dts/qcom/sdm845-cheza-r3.dtb \ + /lava-files/cheza-kernel + KERNEL_IMAGE_NAME+=" cheza-kernel" + + # Make a gzipped copy of the Image for db410c. + gzip -k /lava-files/Image + KERNEL_IMAGE_NAME+=" Image.gz" +fi + +# Pass needed files to the test stage +mkdir -p install +cp -rfv .gitlab-ci/* install/. +cp -rfv install/common install/ci-common +cp -rfv drivers/gpu/drm/ci/* install/. + +. .gitlab-ci/container/container_post_build.sh + +if [[ "$UPLOAD_TO_MINIO" = "1" ]]; then + xz -7 -c -T${FDO_CI_CONCURRENT:-4} vmlinux > /lava-files/vmlinux.xz + FILES_TO_UPLOAD="$KERNEL_IMAGE_NAME vmlinux.xz" + + if [[ -n $DEVICE_TREES ]]; then + FILES_TO_UPLOAD="$FILES_TO_UPLOAD $(basename -a $DEVICE_TREES)" + fi + + for f in $FILES_TO_UPLOAD; do + ci-fairy s3cp --token-file "${CI_JOB_JWT_FILE}" /lava-files/$f \ + https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/$f + done + + S3_ARTIFACT_NAME="kernel-files.tar.zst" + tar --zstd -cf $S3_ARTIFACT_NAME install + ci-fairy s3cp --token-file "${CI_JOB_JWT_FILE}" ${S3_ARTIFACT_NAME} https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/${S3_ARTIFACT_NAME} + + echo "Download vmlinux.xz from https://${PIPELINE_ARTIFACTS_BASE}/${DEBIAN_ARCH}/vmlinux.xz" +fi + +mkdir -p artifacts/install/lib +mv install/* artifacts/install/. +rm -rf artifacts/install/modules +ln -s common artifacts/install/ci-common + +for image in ${KERNEL_IMAGE_NAME}; do + cp /lava-files/$image artifacts/install/. +done + +tar -C artifacts -cf artifacts/install.tar install +rm -rf artifacts/install diff --git a/drivers/gpu/drm/ci/build.yml b/drivers/gpu/drm/ci/build.yml new file mode 100644 index 000000000000..e6503f1c5927 --- /dev/null +++ b/drivers/gpu/drm/ci/build.yml @@ -0,0 +1,110 @@ +.build: + extends: + - .build-rules + stage: build + artifacts: + paths: + - artifacts + script: + - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build.sh + +.build:arm32: + extends: + - .build + - .use-debian/arm64_build + tags: + - aarch64 + variables: + DEFCONFIG: "arch/arm/configs/multi_v7_defconfig" + KERNEL_IMAGE_NAME: "zImage" + KERNEL_ARCH: "arm" + +.build:arm64: + extends: + - .build + - .use-debian/arm64_build + tags: + - aarch64 + variables: + DEFCONFIG: "arch/arm64/configs/defconfig" + KERNEL_IMAGE_NAME: "Image" + KERNEL_ARCH: "arm64" + +.build:x86_64: + extends: + - .build + - .use-debian/x86_64_build + variables: + DEFCONFIG: "arch/x86/configs/x86_64_defconfig" + KERNEL_IMAGE_NAME: "bzImage" + KERNEL_ARCH: "x86_64" + + +# Build IGT for testing on devices + +igt:arm32: + extends: .build:arm32 + script: + - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh + +igt:arm64: + extends: .build:arm64 + script: + - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh + +igt:x86_64: + extends: .build:x86_64 + script: + - FDO_CI_CONCURRENT=${FDO_CI_CONCURRENT} bash drivers/gpu/drm/ci/build-igt.sh + +# Build kernels for testing on devices + +testing:arm32: + extends: .build:arm32 + variables: + # Would be good to have DEBUG_KMEMLEAK, but it doesn't work well with any of + # PROVE_LOCKING and KASAN as of 5.17. + # + # db410c and db820c don't boot with KASAN_INLINE, probably due to the kernel + # becoming too big for their bootloaders. + ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT" + UPLOAD_TO_MINIO: 1 + MERGE_FRAGMENT: arm.config + +testing:arm64: + extends: .build:arm64 + variables: + # Would be good to have DEBUG_KMEMLEAK, but it doesn't work well with any of + # PROVE_LOCKING and KASAN as of 5.17. + # + # db410c and db820c don't boot with KASAN_INLINE, probably due to the kernel + # becoming too big for their bootloaders. + ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT" + UPLOAD_TO_MINIO: 1 + MERGE_FRAGMENT: arm64.config + +testing:x86_64: + extends: .build:x86_64 + variables: + # Would be good to have DEBUG_KMEMLEAK, but it doesn't work well with any of + # PROVE_LOCKING and KASAN as of 5.17. + # + # db410c and db820c don't boot with KASAN_INLINE, probably due to the kernel + # becoming too big for their bootloaders. + ENABLE_KCONFIGS: "PROVE_LOCKING DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT" + UPLOAD_TO_MINIO: 1 + MERGE_FRAGMENT: x86_64.config + + +# Jobs for build-testing different configurations + +build:arm32: + extends: .build:arm32 + +build-nodebugfs:arm64: + extends: .build:arm64 + variables: + DISABLE_KCONFIGS: "DEBUG_FS" + +build:x86_64: + extends: .build:x86_64 diff --git a/drivers/gpu/drm/ci/check-patch.py b/drivers/gpu/drm/ci/check-patch.py new file mode 100755 index 000000000000..a5f399a20e25 --- /dev/null +++ b/drivers/gpu/drm/ci/check-patch.py @@ -0,0 +1,57 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0-or-later +# +# check-patch.py: run checkpatch.pl across all commits in a branch +# +# Based on qemu/.gitlab-ci.d/check-patch.py +# +# Copyright (C) 2020 Red Hat, Inc. +# Copyright (C) 2022 Collabora Ltd. + +import os +import os.path +import sys +import subprocess + +repourl = "https://gitlab.freedesktop.org/%s.git" % os.environ["CI_MERGE_REQUEST_PROJECT_PATH"] + +# GitLab CI environment does not give us any direct info about the +# base for the user's branch. We thus need to figure out a common +# ancestor between the user's branch and current git master. +os.environ["GIT_DEPTH"] = "1000" +subprocess.call(["git", "remote", "remove", "check-patch"], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL) +subprocess.check_call(["git", "remote", "add", "check-patch", repourl]) +subprocess.check_call(["git", "fetch", "check-patch", os.environ["CI_MERGE_REQUEST_TARGET_BRANCH_NAME"]], + stdout=subprocess.DEVNULL, + stderr=subprocess.DEVNULL) + +ancestor = subprocess.check_output(["git", "merge-base", + "check-patch/%s" % os.environ["CI_MERGE_REQUEST_TARGET_BRANCH_NAME"], "HEAD"], + universal_newlines=True) + +ancestor = ancestor.strip() + +log = subprocess.check_output(["git", "log", "--format=%H %s", + ancestor + "..."], + universal_newlines=True) + +subprocess.check_call(["git", "remote", "rm", "check-patch"]) + +if log == "": + print("\nNo commits since %s, skipping checks\n" % ancestor) + sys.exit(0) + +errors = False + +print("\nChecking all commits since %s...\n" % ancestor, flush=True) + +ret = subprocess.run(["scripts/checkpatch.pl", + "--terse", + "--types", os.environ["CHECKPATCH_TYPES"], + "--git", ancestor + "..."]) + +if ret.returncode != 0: + print(" ❌ FAIL one or more commits failed scripts/checkpatch.pl") + sys.exit(1) + +sys.exit(0) diff --git a/drivers/gpu/drm/ci/container.yml b/drivers/gpu/drm/ci/container.yml new file mode 100644 index 000000000000..9764e7921a4f --- /dev/null +++ b/drivers/gpu/drm/ci/container.yml @@ -0,0 +1,65 @@ +.container: + variables: + CI_REPOSITORY_URL: ${DRM_CI_PROJECT_URL}.git # So ci-templates clones drm-ci instead of the repo to test + CI_COMMIT_SHA: ${DRM_CI_COMMIT_SHA} + +debian/x86_64_build-base: + variables: + EXTRA_LOCAL_PACKAGES: "libcairo-dev libdw-dev libjson-c-dev libkmod2 libkmod-dev libpciaccess-dev libproc2-dev libudev-dev libunwind-dev python3-docutils bc python3-ply libssl-dev bc" + +debian/x86_64_test-gl: + variables: + EXTRA_LOCAL_PACKAGES: "jq libasound2 libcairo2 libdw1 libglib2.0-0 libjson-c5 libkmod-dev libkmod2 libgles2 libproc2-dev" + +debian/arm64_build: + variables: + EXTRA_LOCAL_PACKAGES: "libcairo-dev libdw-dev libjson-c-dev libproc2-dev libkmod2 libkmod-dev libpciaccess-dev libudev-dev libunwind-dev python3-docutils libssl-dev crossbuild-essential-armhf libkmod-dev:armhf libproc2-dev:armhf libunwind-dev:armhf libdw-dev:armhf libpixman-1-dev:armhf libcairo-dev:armhf libudev-dev:armhf libjson-c-dev:armhf" + +.kernel+rootfs: + variables: + EXTRA_LOCAL_PACKAGES: "jq libasound2 libcairo2 libdw1 libglib2.0-0 libjson-c5" + +# Disable container jobs that we won't use +alpine/x86_64_build: + rules: + - when: never + +debian/x86_64_test-vk: + rules: + - when: never + +fedora/x86_64_build: + rules: + - when: never + +debian/android_build: + rules: + - when: never + +debian/x86_64_test-android: + rules: + - when: never + +windows_build_vs2019: + rules: + - when: never + +windows_test_vs2019: + rules: + - when: never + +.debian/x86_64_build-mingw: + rules: + - when: never + +rustfmt: + rules: + - when: never + +windows_vs2019: + rules: + - when: never + +clang-format: + rules: + - when: never
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/gitlab-ci.yml b/drivers/gpu/drm/ci/gitlab-ci.yml new file mode 100644 index 000000000000..2c4df53f5dfe --- /dev/null +++ b/drivers/gpu/drm/ci/gitlab-ci.yml @@ -0,0 +1,251 @@ +variables: + DRM_CI_PROJECT_PATH: &drm-ci-project-path mesa/mesa + DRM_CI_COMMIT_SHA: &drm-ci-commit-sha 0dc961645c4f0241f8512cb0ec3ad59635842072 + + UPSTREAM_REPO: git://anongit.freedesktop.org/drm/drm + TARGET_BRANCH: drm-next + + IGT_VERSION: 471bfababd070e1dac0ebb87470ac4f2ae85e663 + + DEQP_RUNNER_GIT_URL: https://gitlab.freedesktop.org/anholt/deqp-runner.git + DEQP_RUNNER_GIT_TAG: v0.15.0 + + FDO_UPSTREAM_REPO: helen.fornazier/linux # The repo where the git-archive daily runs + MESA_TEMPLATES_COMMIT: &ci-templates-commit d5aa3941aa03c2f716595116354fb81eb8012acb + DRM_CI_PROJECT_URL: https://gitlab.freedesktop.org/${DRM_CI_PROJECT_PATH} + CI_PRE_CLONE_SCRIPT: |- + set -o xtrace + curl -L --retry 4 -f --retry-all-errors --retry-delay 60 -s ${DRM_CI_PROJECT_URL}/-/raw/${DRM_CI_COMMIT_SHA}/.gitlab-ci/download-git-cache.sh -o download-git-cache.sh + bash download-git-cache.sh + rm download-git-cache.sh + set +o xtrace + S3_HOST: s3.freedesktop.org + # per-pipeline artifact storage on MinIO + PIPELINE_ARTIFACTS_BASE: ${S3_HOST}/artifacts/${CI_PROJECT_PATH}/${CI_PIPELINE_ID} + # per-job artifact storage on MinIO + JOB_ARTIFACTS_BASE: ${PIPELINE_ARTIFACTS_BASE}/${CI_JOB_ID} + + LAVA_JOB_PRIORITY: 30 + +default: + before_script: + - export SCRIPTS_DIR=$(mktemp -d) + - curl -L -s --retry 4 -f --retry-all-errors --retry-delay 60 -O --output-dir "${SCRIPTS_DIR}" "${DRM_CI_PROJECT_URL}/-/raw/${DRM_CI_COMMIT_SHA}/.gitlab-ci/setup-test-env.sh" + - source ${SCRIPTS_DIR}/setup-test-env.sh + - echo -e "\e[0Ksection_start:$(date +%s):unset_env_vars_section[collapsed=true]\r\e[0KUnsetting vulnerable environment variables" + - export CI_JOB_JWT_FILE="${CI_JOB_JWT_FILE:-$(mktemp)}" + - echo -n "${CI_JOB_JWT}" > "${CI_JOB_JWT_FILE}" + - unset CI_JOB_JWT + - echo -e "\e[0Ksection_end:$(date +%s):unset_env_vars_section\r\e[0K" + + - echo -e "\e[0Ksection_start:$(date +%s):drm_ci_download_section[collapsed=true]\r\e[0KDownloading mesa from $DRM_CI_PROJECT_URL/-/archive/$DRM_CI_COMMIT_SHA/mesa-$DRM_CI_COMMIT_SHA.tar.gz" + - cd $CI_PROJECT_DIR + - curl --output - $DRM_CI_PROJECT_URL/-/archive/$DRM_CI_COMMIT_SHA/mesa-$DRM_CI_COMMIT_SHA.tar.gz | tar -xz + - mv mesa-$DRM_CI_COMMIT_SHA/.gitlab-ci* . + - rm -rf mesa-$DRM_CI_COMMIT_SHA/ + - echo -e "\e[0Ksection_end:$(date +%s):drm_ci_download_section\r\e[0K" + + after_script: + - > + set +x + + test -e "${CI_JOB_JWT_FILE}" && + export CI_JOB_JWT="$(<${CI_JOB_JWT_FILE})" && + rm "${CI_JOB_JWT_FILE}" + + # Retry when job fails. + retry: + max: 1 + # Ignore runner_unsupported, stale_schedule, archived_failure, or + # unmet_prerequisites + when: + - api_failure + - runner_system_failure + - script_failure + - job_execution_timeout + - scheduler_failure + - data_integrity_failure + - unknown_failure + +include: + - project: 'freedesktop/ci-templates' + ref: 16bc29078de5e0a067ff84a1a199a3760d3b3811 + file: + - '/templates/ci-fairy.yml' + - project: 'freedesktop/ci-templates' + ref: *ci-templates-commit + file: + - '/templates/alpine.yml' + - '/templates/debian.yml' + - '/templates/fedora.yml' + - project: *drm-ci-project-path + ref: *drm-ci-commit-sha + file: + - '/.gitlab-ci/farm-rules.yml' + - '/.gitlab-ci/test-source-dep.yml' + - '/.gitlab-ci/container/gitlab-ci.yml' + - '/.gitlab-ci/test/gitlab-ci.yml' + - '/.gitlab-ci/lava/lava-gitlab-ci.yml' + - drivers/gpu/drm/ci/image-tags.yml + - drivers/gpu/drm/ci/container.yml + - drivers/gpu/drm/ci/static-checks.yml + - drivers/gpu/drm/ci/build.yml + - drivers/gpu/drm/ci/test.yml + - 'https://gitlab.freedesktop.org/gfx-ci/lab-status/-/raw/main/lab-status.yml' + + +stages: + - sanity + - container + - git-archive + - build + - amdgpu + - i915 + - mediatek + - meson + - msm + - rockchip + - virtio-gpu + - lint + +# YAML anchors for rule conditions +# -------------------------------- +.rules-anchors: + rules: + # Pipeline for forked project branch + - if: &is-forked-branch '$CI_COMMIT_BRANCH && $CI_PROJECT_NAMESPACE != "mesa"' + when: manual + # Forked project branch / pre-merge pipeline not for Marge bot + - if: &is-forked-branch-or-pre-merge-not-for-marge '$CI_PROJECT_NAMESPACE != "mesa" || ($GITLAB_USER_LOGIN != "marge-bot" && $CI_PIPELINE_SOURCE == "merge_request_event")' + when: manual + # Pipeline runs for the main branch of the upstream Mesa project + - if: &is-mesa-main '$CI_PROJECT_NAMESPACE == "mesa" && $CI_COMMIT_REF_NAME == $CI_DEFAULT_BRANCH && $CI_COMMIT_BRANCH' + when: always + # Post-merge pipeline + - if: &is-post-merge '$CI_PROJECT_NAMESPACE == "mesa" && $CI_COMMIT_BRANCH' + when: on_success + # Post-merge pipeline, not for Marge Bot + - if: &is-post-merge-not-for-marge '$CI_PROJECT_NAMESPACE == "mesa" && $GITLAB_USER_LOGIN != "marge-bot" && $CI_COMMIT_BRANCH' + when: on_success + # Pre-merge pipeline + - if: &is-pre-merge '$CI_PIPELINE_SOURCE == "merge_request_event"' + when: on_success + # Pre-merge pipeline for Marge Bot + - if: &is-pre-merge-for-marge '$GITLAB_USER_LOGIN == "marge-bot" && $CI_PIPELINE_SOURCE == "merge_request_event"' + when: on_success + +# Rule to filter for only scheduled pipelines. +.scheduled_pipeline-rules: + rules: + - if: &is-scheduled-pipeline '$CI_PIPELINE_SOURCE == "schedule"' + when: on_success + +# Generic rule to not run the job during scheduled pipelines. Jobs that aren't +# something like a nightly run should include this rule. +.no_scheduled_pipelines-rules: + rules: + - if: *is-scheduled-pipeline + when: never + +# When to automatically run the CI for build jobs +.build-rules: + rules: + - !reference [.no_scheduled_pipelines-rules, rules] + # Run automatically once all dependency jobs have passed + - when: on_success + + +.ci-deqp-artifacts: + artifacts: + name: "mesa_${CI_JOB_NAME}" + when: always + untracked: false + paths: + # Watch out! Artifacts are relative to the build dir. + # https://gitlab.com/gitlab-org/gitlab-ce/commit/8788fb925706cad594adf6917a6c5f6587dd1521 + - artifacts + - _build/meson-logs/*.txt + - _build/meson-logs/strace + + +.container-rules: + rules: + - !reference [.no_scheduled_pipelines-rules, rules] + # Run pipeline by default in the main project if any CI pipeline + # configuration files were changed, to ensure docker images are up to date + - if: *is-post-merge + changes: + - drivers/gpu/drm/ci/**/* + when: on_success + # Run pipeline by default if it was triggered by Marge Bot, is for a + # merge request, and any files affecting the pipeline were changed + - if: *is-pre-merge-for-marge + when: on_success + # Run pipeline by default in the main project if it was not triggered by + # Marge Bot, and any files affecting the pipeline were changed + - if: *is-post-merge-not-for-marge + when: on_success + # Allow triggering jobs manually in other cases + - when: manual + + + +# Git archive + +make git archive: + extends: + - .fdo.ci-fairy + stage: git-archive + rules: + - !reference [.scheduled_pipeline-rules, rules] + # ensure we are running on packet + tags: + - packet.net + script: + # Remove drm-ci files we just added + - rm -rf .gitlab-ci.* + + # Compactify the .git directory + - git gc --aggressive + # compress the current folder + - tar -cvzf ../$CI_PROJECT_NAME.tar.gz . + + # login with the JWT token file + - ci-fairy s3cp --token-file "${CI_JOB_JWT_FILE}" ../$CI_PROJECT_NAME.tar.gz https://$S3_HOST/git-cache/$CI_PROJECT_NAMESPACE/$CI_PROJECT_NAME/$CI_PROJECT_NAME.tar.gz + + +# Sanity checks of MR settings and commit logs +sanity: + extends: + - .fdo.ci-fairy + stage: sanity + rules: + - if: *is-pre-merge + when: on_success + # Other cases default to never + variables: + GIT_STRATEGY: none + script: + # ci-fairy check-commits --junit-xml=check-commits.xml + - ci-fairy check-merge-request --require-allow-collaboration --junit-xml=check-merge-request.xml + artifacts: + when: on_failure + reports: + junit: check-*.xml + +# Rules for tests that should not block merging, but should be available to +# optionally run with the "play" button in the UI in pre-merge non-marge +# pipelines. This should appear in "extends:" after any includes of +# test-source-dep.yml rules, so that these rules replace those. +.test-manual-mr: + rules: + - !reference [.no_scheduled_pipelines-rules, rules] + - if: *is-forked-branch-or-pre-merge-not-for-marge + when: manual + variables: + JOB_TIMEOUT: 80 + + +# Jobs that need to pass before spending hardware resources on further testing +.required-for-hardware-jobs: + needs: []
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/igt_runner.sh b/drivers/gpu/drm/ci/igt_runner.sh new file mode 100755 index 000000000000..2bb759165063 --- /dev/null +++ b/drivers/gpu/drm/ci/igt_runner.sh @@ -0,0 +1,77 @@ +#!/bin/sh +# SPDX-License-Identifier: MIT + +set -ex + +export IGT_FORCE_DRIVER=${DRIVER_NAME} +export PATH=$PATH:/igt/bin/ +export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/igt/lib/aarch64-linux-gnu/:/igt/lib/x86_64-linux-gnu:/igt/lib:/igt/lib64 + +# Uncomment the below to debug problems with driver probing +: ' +ls -l /dev/dri/ +cat /sys/kernel/debug/devices_deferred +cat /sys/kernel/debug/device_component/* +' + +# Dump drm state to confirm that kernel was able to find a connected display: +# TODO this path might not exist for all drivers.. maybe run modetest instead? +set +e +cat /sys/kernel/debug/dri/*/state +set -e + +# Cannot use HWCI_KERNEL_MODULES as at that point we don't have the module in /lib +if [ "$IGT_FORCE_DRIVER" = "amdgpu" ]; then + mv /install/modules/lib/modules/* /lib/modules/. + modprobe amdgpu +fi + +if [ -e "/install/xfails/$DRIVER_NAME-$GPU_VERSION-skips.txt" ]; then + IGT_SKIPS="--skips /install/xfails/$DRIVER_NAME-$GPU_VERSION-skips.txt" +fi + +if [ -e "/install/xfails/$DRIVER_NAME-$GPU_VERSION-flakes.txt" ]; then + IGT_FLAKES="--flakes /install/xfails/$DRIVER_NAME-$GPU_VERSION-flakes.txt" +fi + +if [ -e "/install/xfails/$DRIVER_NAME-$GPU_VERSION-fails.txt" ]; then + IGT_FAILS="--baseline /install/xfails/$DRIVER_NAME-$GPU_VERSION-fails.txt" +fi + +if [ "`uname -m`" = "aarch64" ]; then + ARCH="arm64" +elif [ "`uname -m`" = "armv7l" ]; then + ARCH="arm" +else + ARCH="x86_64" +fi + +curl -L --retry 4 -f --retry-all-errors --retry-delay 60 -s ${FDO_HTTP_CACHE_URI:-}$PIPELINE_ARTIFACTS_BASE/$ARCH/igt.tar.gz | tar --zstd -v -x -C / + +set +e +igt-runner \ + run \ + --igt-folder /igt/libexec/igt-gpu-tools \ + --caselist /install/testlist.txt \ + --output /results \ + $IGT_SKIPS \ + $IGT_FLAKES \ + $IGT_FAILS \ + --fraction-start $CI_NODE_INDEX \ + --fraction $CI_NODE_TOTAL \ + --jobs 1 +ret=$? +set -e + +deqp-runner junit \ + --testsuite IGT \ + --results /results/failures.csv \ + --output /results/junit.xml \ + --limit 50 \ + --template "See https://$CI_PROJECT_ROOT_NAMESPACE.pages.freedesktop.org/-/$CI_PROJECT_NAME/-/jobs/$CI_JOB_ID/artifacts/results/{{testcase}}.xml" + +# Store the results also in the simpler format used by the runner in ChromeOS CI +#sed -r 's/(dmesg-warn|pass)/success/g' /results/results.txt > /results/results_simple.txt + +cd $oldpath +exit $ret diff --git a/drivers/gpu/drm/ci/image-tags.yml b/drivers/gpu/drm/ci/image-tags.yml new file mode 100644 index 000000000000..f051b6c547c5 --- /dev/null +++ b/drivers/gpu/drm/ci/image-tags.yml @@ -0,0 +1,15 @@ +variables: + CONTAINER_TAG: "2023-08-10-mesa-uprev" + DEBIAN_X86_64_BUILD_BASE_IMAGE: "debian/x86_64_build-base" + DEBIAN_BASE_TAG: "${CONTAINER_TAG}" + + DEBIAN_X86_64_BUILD_IMAGE_PATH: "debian/x86_64_build" + DEBIAN_BUILD_TAG: "${CONTAINER_TAG}" + + KERNEL_ROOTFS_TAG: "${CONTAINER_TAG}" + + DEBIAN_X86_64_TEST_BASE_IMAGE: "debian/x86_64_test-base" + DEBIAN_X86_64_TEST_IMAGE_GL_PATH: "debian/x86_64_test-gl" + DEBIAN_X86_64_TEST_GL_TAG: "${CONTAINER_TAG}" + + ALPINE_X86_64_LAVA_SSH_TAG: "${CONTAINER_TAG}"
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/lava-submit.sh b/drivers/gpu/drm/ci/lava-submit.sh new file mode 100755 index 000000000000..0c4456b21b0f --- /dev/null +++ b/drivers/gpu/drm/ci/lava-submit.sh @@ -0,0 +1,57 @@ +#!/bin/bash +# SPDX-License-Identifier: MIT + +set -e +set -x + +# Try to use the kernel and rootfs built in mainline first, so we're more +# likely to hit cache +if curl -L --retry 4 -f --retry-all-errors --retry-delay 60 -s "https://${BASE_SYSTEM_MAINLINE_HOST_PATH}/done"; then + BASE_SYSTEM_HOST_PATH="${BASE_SYSTEM_MAINLINE_HOST_PATH}" +else + BASE_SYSTEM_HOST_PATH="${BASE_SYSTEM_FORK_HOST_PATH}" +fi + +rm -rf results +mkdir -p results/job-rootfs-overlay/ + +cp artifacts/ci-common/capture-devcoredump.sh results/job-rootfs-overlay/ +cp artifacts/ci-common/init-*.sh results/job-rootfs-overlay/ +cp artifacts/ci-common/intel-gpu-freq.sh results/job-rootfs-overlay/ +cp "$SCRIPTS_DIR"/setup-test-env.sh results/job-rootfs-overlay/ + +# Prepare env vars for upload. +section_start variables "Variables passed through:" +KERNEL_IMAGE_BASE_URL="https://${BASE_SYSTEM_HOST_PATH}" \ + artifacts/ci-common/generate-env.sh | tee results/job-rootfs-overlay/set-job-env-vars.sh +section_end variables + +tar zcf job-rootfs-overlay.tar.gz -C results/job-rootfs-overlay/ . +ci-fairy s3cp --token-file "${CI_JOB_JWT_FILE}" job-rootfs-overlay.tar.gz "https://${JOB_ROOTFS_OVERLAY_PATH}" + +touch results/lava.log +tail -f results/lava.log & + +PYTHONPATH=artifacts/ artifacts/lava/lava_job_submitter.py \ + submit \ + --dump-yaml \ + --pipeline-info "$CI_JOB_NAME: $CI_PIPELINE_URL on $CI_COMMIT_REF_NAME ${CI_NODE_INDEX}/${CI_NODE_TOTAL}" \ + --rootfs-url-prefix "https://${BASE_SYSTEM_HOST_PATH}" \ + --kernel-url-prefix "https://${PIPELINE_ARTIFACTS_BASE}/${ARCH}" \ + --build-url "${FDO_HTTP_CACHE_URI:-}https://${PIPELINE_ARTIFACTS_BASE}/${ARCH}/kernel-files.tar.zst" \ + --job-rootfs-overlay-url "${FDO_HTTP_CACHE_URI:-}https://${JOB_ROOTFS_OVERLAY_PATH}" \ + --job-timeout-min ${JOB_TIMEOUT:-80} \ + --first-stage-init artifacts/ci-common/init-stage1.sh \ + --ci-project-dir "${CI_PROJECT_DIR}" \ + --device-type "${DEVICE_TYPE}" \ + --dtb-filename "${DTB}" \ + --jwt-file "${CI_JOB_JWT_FILE}" \ + --kernel-image-name "${KERNEL_IMAGE_NAME}" \ + --kernel-image-type "${KERNEL_IMAGE_TYPE}" \ + --boot-method "${BOOT_METHOD}" \ + --visibility-group "${VISIBILITY_GROUP}" \ + --lava-tags "${LAVA_TAGS}" \ + --mesa-job-name "$CI_JOB_NAME" \ + --structured-log-file "results/lava_job_detail.json" \ + --ssh-client-image "${LAVA_SSH_CLIENT_IMAGE}" \ + >> results/lava.log diff --git a/drivers/gpu/drm/ci/static-checks.yml b/drivers/gpu/drm/ci/static-checks.yml new file mode 100644 index 000000000000..13ffa827b7fa --- /dev/null +++ b/drivers/gpu/drm/ci/static-checks.yml @@ -0,0 +1,12 @@ +check-patch: + extends: + - .build + - .use-debian/x86_64_build + script: + - drivers/gpu/drm/ci/check-patch.py + variables: + CHECKPATCH_TYPES: "BAD_SIGN_OFF,BAD_STABLE_ADDRESS_STYLE,COMMIT_COMMENT_SYMBOL,COMMIT_MESSAGE,EMAIL_SUBJECT,FROM_SIGN_OFF_MISMATCH,MISSING_SIGN_OFF,NO_AUTHOR_SIGN_OFF,DIFF_IN_COMMIT_MSG,GERRIT_CHANGE_ID,GIT_COMMIT_ID,UNKNOWN_COMMIT_ID,CODE_INDENT,BIT_MACRO,DOS_LINE_ENDINGS" + rules: + - if: '$CI_PIPELINE_SOURCE == "merge_request_event"' + when: on_success + # Other cases default to never diff --git a/drivers/gpu/drm/ci/test.yml b/drivers/gpu/drm/ci/test.yml new file mode 100644 index 000000000000..6473cddaa7a9 --- /dev/null +++ b/drivers/gpu/drm/ci/test.yml @@ -0,0 +1,335 @@ +.test-rules: + rules: + - if: '$FD_FARM == "offline" && $RUNNER_TAG =~ /^google-freedreno-/' + when: never + - if: '$COLLABORA_FARM == "offline" && $RUNNER_TAG =~ /^mesa-ci-x86-64-lava-/' + when: never + - !reference [.no_scheduled_pipelines-rules, rules] + - when: on_success + +.lava-test: + extends: + - .test-rules + script: + # Note: Build dir (and thus install) may be dirty due to GIT_STRATEGY + - rm -rf install + - tar -xf artifacts/install.tar + - mv install/* artifacts/. + # Override it with our lava-submit.sh script + - ./artifacts/lava-submit.sh + +.lava-igt:arm32: + extends: + - .lava-test:arm32 + variables: + HWCI_TEST_SCRIPT: "/install/igt_runner.sh" + ARCH: "armhf" + dependencies: + - testing:arm32 + needs: + - alpine/x86_64_lava_ssh_client + - kernel+rootfs_arm32 + - debian/x86_64_build + - testing:arm32 + - igt:arm32 + +.lava-igt:arm64: + extends: + - .lava-test:arm64 + variables: + HWCI_TEST_SCRIPT: "/install/igt_runner.sh" + ARCH: "arm64" + dependencies: + - testing:arm64 + needs: + - alpine/x86_64_lava_ssh_client + - kernel+rootfs_arm64 + - debian/x86_64_build + - testing:arm64 + - igt:arm64 + +.lava-igt:x86_64: + extends: + - .lava-test:x86_64 + variables: + HWCI_TEST_SCRIPT: "/install/igt_runner.sh" + ARCH: "x86_64" + dependencies: + - testing:x86_64 + needs: + - alpine/x86_64_lava_ssh_client + - kernel+rootfs_x86_64 + - debian/x86_64_build + - testing:x86_64 + - igt:x86_64 + +.baremetal-igt-arm64: + extends: + - .baremetal-test-arm64 + - .use-debian/arm64_test + - .test-rules + variables: + FDO_CI_CONCURRENT: 10 + HWCI_TEST_SCRIPT: "/install/igt_runner.sh" + S3_ARTIFACT_NAME: "arm64/kernel-files" + BM_KERNEL: https://${PIPELINE_ARTIFACTS_BASE}/arm64/Image.gz + BM_CMDLINE: "ip=dhcp console=ttyMSM0,115200n8 $BM_KERNEL_EXTRA_ARGS root=/dev/nfs rw nfsrootdebug nfsroot=,tcp,nfsvers=4.2 init=/init $BM_KERNELARGS" + needs: + - debian/arm64_test + - job: testing:arm64 + artifacts: false + - igt:arm64 + tags: + - $RUNNER_TAG + +msm:sc7180: + extends: + - .lava-igt:arm64 + stage: msm + parallel: 2 + variables: + DRIVER_NAME: msm + DEVICE_TYPE: sc7180-trogdor-lazor-limozeen + DTB: sc7180-trogdor-lazor-limozeen-nots-r5 + BOOT_METHOD: depthcharge + KERNEL_IMAGE_TYPE: "" + GPU_VERSION: sc7180 + RUNNER_TAG: mesa-ci-x86-64-lava-sc7180-trogdor-lazor-limozeen + +msm:apq8016: + extends: + - .baremetal-igt-arm64 + stage: msm + variables: + DRIVER_NAME: msm + BM_DTB: https://${PIPELINE_ARTIFACTS_BASE}/arm64/apq8016-sbc.dtb + GPU_VERSION: apq8016 + BM_CMDLINE: "ip=dhcp console=ttyMSM0,115200n8 $BM_KERNEL_EXTRA_ARGS root=/dev/nfs rw nfsrootdebug nfsroot=,tcp,nfsvers=4.2 init=/init $BM_KERNELARGS" + RUNNER_TAG: google-freedreno-db410c + script: + - ./install/bare-metal/fastboot.sh + rules: + # TODO: current issue: it is not fiding the NFS root. Fix and remove this rule. + - when: never + +msm:apq8096: + extends: + - .baremetal-igt-arm64 + stage: msm + variables: + DRIVER_NAME: msm + BM_KERNEL_EXTRA_ARGS: maxcpus=2 + BM_DTB: https://${PIPELINE_ARTIFACTS_BASE}/arm64/apq8096-db820c.dtb + GPU_VERSION: apq8096 + RUNNER_TAG: google-freedreno-db820c + script: + - ./install/bare-metal/fastboot.sh + +msm:sdm845: + extends: + - .baremetal-igt-arm64 + stage: msm + parallel: 6 + variables: + DRIVER_NAME: msm + BM_KERNEL: https://${PIPELINE_ARTIFACTS_BASE}/arm64/cheza-kernel + GPU_VERSION: sdm845 + RUNNER_TAG: google-freedreno-cheza + script: + - ./install/bare-metal/cros-servo.sh + +rockchip:rk3288: + extends: + - .lava-igt:arm32 + stage: rockchip + variables: + DRIVER_NAME: rockchip + DEVICE_TYPE: rk3288-veyron-jaq + DTB: ${DEVICE_TYPE} + BOOT_METHOD: depthcharge + KERNEL_IMAGE_TYPE: "zimage" + GPU_VERSION: rk3288 + RUNNER_TAG: mesa-ci-x86-64-lava-rk3288-veyron-jaq + +rockchip:rk3399: + extends: + - .lava-igt:arm64 + stage: rockchip + parallel: 3 + variables: + DRIVER_NAME: rockchip + DEVICE_TYPE: rk3399-gru-kevin + DTB: ${DEVICE_TYPE} + BOOT_METHOD: depthcharge + KERNEL_IMAGE_TYPE: "" + GPU_VERSION: rk3399 + RUNNER_TAG: mesa-ci-x86-64-lava-rk3399-gru-kevin + +.i915: + extends: + - .lava-igt:x86_64 + stage: i915 + variables: + DRIVER_NAME: i915 + DTB: "" + BOOT_METHOD: depthcharge + KERNEL_IMAGE_TYPE: "" + +i915:apl: + extends: + - .i915 + parallel: 12 + variables: + DEVICE_TYPE: asus-C523NA-A20057-coral + GPU_VERSION: apl + RUNNER_TAG: mesa-ci-x86-64-lava-asus-C523NA-A20057-coral + +i915:glk: + extends: + - .i915 + parallel: 5 + variables: + DEVICE_TYPE: hp-x360-12b-ca0010nr-n4020-octopus + GPU_VERSION: glk + RUNNER_TAG: mesa-ci-x86-64-lava-hp-x360-12b-ca0010nr-n4020-octopus + +i915:amly: + extends: + - .i915 + parallel: 8 + variables: + DEVICE_TYPE: asus-C433TA-AJ0005-rammus + GPU_VERSION: amly + RUNNER_TAG: mesa-ci-x86-64-lava-asus-C433TA-AJ0005-rammus + +i915:kbl: + extends: + - .i915 + parallel: 5 + variables: + DEVICE_TYPE: hp-x360-14-G1-sona + GPU_VERSION: kbl + RUNNER_TAG: mesa-ci-x86-64-lava-hp-x360-14-G1-sona + +i915:whl: + extends: + - .i915 + parallel: 8 + variables: + DEVICE_TYPE: dell-latitude-5400-8665U-sarien + GPU_VERSION: whl + RUNNER_TAG: mesa-ci-x86-64-lava-dell-latitude-5400-8665U-sarien + +i915:cml: + extends: + - .i915 + parallel: 6 + variables: + DEVICE_TYPE: asus-C436FA-Flip-hatch + GPU_VERSION: cml + RUNNER_TAG: mesa-ci-x86-64-lava-asus-C436FA-flip-hatch + +i915:tgl: + extends: + - .i915 + parallel: 6 + variables: + DEVICE_TYPE: asus-cx9400-volteer + GPU_VERSION: tgl + RUNNER_TAG: mesa-ci-x86-64-lava-asus-cx9400-volteer + +.amdgpu: + extends: + - .lava-igt:x86_64 + stage: amdgpu + variables: + DRIVER_NAME: amdgpu + DTB: "" + BOOT_METHOD: depthcharge + KERNEL_IMAGE_TYPE: "" + +amdgpu:stoney: + extends: + - .amdgpu + variables: + DEVICE_TYPE: hp-11A-G6-EE-grunt + GPU_VERSION: stoney + RUNNER_TAG: mesa-ci-x86-64-lava-hp-11A-G6-EE-grunt + +.mediatek: + extends: + - .lava-igt:arm64 + stage: mediatek + variables: + DRIVER_NAME: mediatek + DTB: ${DEVICE_TYPE} + BOOT_METHOD: depthcharge + KERNEL_IMAGE_TYPE: "" + +mediatek:mt8173: + extends: + - .mediatek + variables: + DEVICE_TYPE: mt8173-elm-hana + GPU_VERSION: mt8173 + RUNNER_TAG: mesa-ci-x86-64-lava-mt8173-elm-hana + rules: + # TODO: current issue: device is hanging. Fix and remove this rule. + - when: never + +mediatek:mt8183: + extends: + - .mediatek + variables: + DEVICE_TYPE: mt8183-kukui-jacuzzi-juniper-sku16 + GPU_VERSION: mt8183 + RUNNER_TAG: mesa-ci-x86-64-lava-mt8183-kukui-jacuzzi-juniper-sku16 + +# drm-mtk doesn't even probe yet in mainline for mt8192 +.mediatek:mt8192: + extends: + - .mediatek + variables: + DEVICE_TYPE: mt8192-asurada-spherion-r0 + GPU_VERSION: mt8192 + RUNNER_TAG: mesa-ci-x86-64-lava-mt8192-asurada-spherion-r0 + +.meson: + extends: + - .lava-igt:arm64 + stage: meson + variables: + DRIVER_NAME: meson + DTB: ${DEVICE_TYPE} + BOOT_METHOD: u-boot + KERNEL_IMAGE_TYPE: "image" + +meson:g12b: + extends: + - .meson + variables: + DEVICE_TYPE: meson-g12b-a311d-khadas-vim3 + GPU_VERSION: g12b + RUNNER_TAG: mesa-ci-x86-64-lava-meson-g12b-a311d-khadas-vim3 + +virtio_gpu:none: + stage: virtio-gpu + variables: + CROSVM_GALLIUM_DRIVER: llvmpipe + DRIVER_NAME: virtio_gpu + GPU_VERSION: none + extends: + - .test-gl + tags: + - kvm + script: + - ln -sf $CI_PROJECT_DIR/install /install + - mv install/bzImage /lava-files/bzImage + - install/crosvm-runner.sh install/igt_runner.sh + needs: + - debian/x86_64_test-gl + - testing:x86_64 + - igt:x86_64 + rules: + # TODO: current issue: malloc(): corrupted top size. Fix and remove this rule. + - when: never
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/testlist.txt b/drivers/gpu/drm/ci/testlist.txt new file mode 100644 index 000000000000..f82cd90372f4 --- /dev/null +++ b/drivers/gpu/drm/ci/testlist.txt @@ -0,0 +1,2912 @@ +core_auth@getclient-simple +core_auth@getclient-master-drop +core_auth@basic-auth +core_auth@many-magics +core_getclient +core_getstats +core_getversion +core_setmaster_vs_auth +drm_read@invalid-buffer +drm_read@fault-buffer +drm_read@empty-block +drm_read@empty-nonblock +drm_read@short-buffer-block +drm_read@short-buffer-nonblock +drm_read@short-buffer-wakeup +gem_eio@throttle +gem_eio@create +gem_eio@create-ext +gem_eio@context-create +gem_eio@execbuf +gem_eio@banned +gem_eio@suspend +gem_eio@hibernate +gem_eio@in-flight-external +gem_eio@in-flight-suspend +gem_eio@reset-stress +gem_eio@unwedge-stress +gem_eio@wait-immediate +gem_eio@wait-wedge-immediate +gem_eio@in-flight-immediate +gem_eio@in-flight-contexts-immediate +gem_eio@in-flight-internal-immediate +gem_eio@wait-1us +gem_eio@wait-wedge-1us +gem_eio@in-flight-1us +gem_eio@in-flight-contexts-1us +gem_eio@in-flight-internal-1us +gem_eio@wait-10ms +gem_eio@wait-wedge-10ms +gem_eio@in-flight-10ms +gem_eio@in-flight-contexts-10ms +gem_eio@in-flight-internal-10ms +gem_eio@kms +kms_3d +kms_addfb_basic@unused-handle +kms_addfb_basic@unused-pitches +kms_addfb_basic@unused-offsets +kms_addfb_basic@unused-modifier +kms_addfb_basic@clobberred-modifier +kms_addfb_basic@invalid-smem-bo-on-discrete +kms_addfb_basic@legacy-format +kms_addfb_basic@no-handle +kms_addfb_basic@basic +kms_addfb_basic@bad-pitch-0 +kms_addfb_basic@bad-pitch-32 +kms_addfb_basic@bad-pitch-63 +kms_addfb_basic@bad-pitch-128 +kms_addfb_basic@bad-pitch-256 +kms_addfb_basic@bad-pitch-1024 +kms_addfb_basic@bad-pitch-999 +kms_addfb_basic@bad-pitch-65536 +kms_addfb_basic@invalid-get-prop-any +kms_addfb_basic@invalid-get-prop +kms_addfb_basic@invalid-set-prop-any +kms_addfb_basic@invalid-set-prop +kms_addfb_basic@master-rmfb +kms_addfb_basic@addfb25-modifier-no-flag +kms_addfb_basic@addfb25-bad-modifier +kms_addfb_basic@addfb25-x-tiled-mismatch-legacy +kms_addfb_basic@addfb25-x-tiled-legacy +kms_addfb_basic@addfb25-framebuffer-vs-set-tiling +kms_addfb_basic@basic-x-tiled-legacy +kms_addfb_basic@framebuffer-vs-set-tiling +kms_addfb_basic@tile-pitch-mismatch +kms_addfb_basic@basic-y-tiled-legacy +kms_addfb_basic@size-max +kms_addfb_basic@too-wide +kms_addfb_basic@too-high +kms_addfb_basic@bo-too-small +kms_addfb_basic@small-bo +kms_addfb_basic@bo-too-small-due-to-tiling +kms_addfb_basic@addfb25-y-tiled-legacy +kms_addfb_basic@addfb25-yf-tiled-legacy +kms_addfb_basic@addfb25-y-tiled-small-legacy +kms_addfb_basic@addfb25-4-tiled +kms_async_flips@async-flip-with-page-flip-events +kms_async_flips@alternate-sync-async-flip +kms_async_flips@test-time-stamp +kms_async_flips@test-cursor +kms_async_flips@invalid-async-flip +kms_async_flips@crc +kms_atomic@plane-overlay-legacy +kms_atomic@plane-primary-legacy +kms_atomic@plane-primary-overlay-mutable-zpos +kms_atomic@plane-immutable-zpos +kms_atomic@test-only +kms_atomic@plane-cursor-legacy +kms_atomic@plane-invalid-params +kms_atomic@plane-invalid-params-fence +kms_atomic@crtc-invalid-params +kms_atomic@crtc-invalid-params-fence +kms_atomic@atomic-invalid-params +kms_atomic@atomic_plane_damage +kms_atomic_interruptible@legacy-setmode +kms_atomic_interruptible@atomic-setmode +kms_atomic_interruptible@legacy-dpms +kms_atomic_interruptible@legacy-pageflip +kms_atomic_interruptible@legacy-cursor +kms_atomic_interruptible@universal-setplane-primary +kms_atomic_interruptible@universal-setplane-cursor +kms_atomic_transition@plane-primary-toggle-with-vblank-wait +kms_atomic_transition@plane-all-transition +kms_atomic_transition@plane-all-transition-fencing +kms_atomic_transition@plane-all-transition-nonblocking +kms_atomic_transition@plane-all-transition-nonblocking-fencing +kms_atomic_transition@plane-use-after-nonblocking-unbind +kms_atomic_transition@plane-use-after-nonblocking-unbind-fencing +kms_atomic_transition@plane-all-modeset-transition +kms_atomic_transition@plane-all-modeset-transition-fencing +kms_atomic_transition@plane-all-modeset-transition-internal-panels +kms_atomic_transition@plane-all-modeset-transition-fencing-internal-panels +kms_atomic_transition@plane-toggle-modeset-transition +kms_atomic_transition@modeset-transition +kms_atomic_transition@modeset-transition-fencing +kms_atomic_transition@modeset-transition-nonblocking +kms_atomic_transition@modeset-transition-nonblocking-fencing +kms_big_fb@x-tiled-addfb-size-overflow +kms_big_fb@y-tiled-addfb-size-overflow +kms_big_fb@yf-tiled-addfb-size-overflow +kms_big_fb@4-tiled-addfb-size-overflow +kms_big_fb@x-tiled-addfb-size-offset-overflow +kms_big_fb@y-tiled-addfb-size-offset-overflow +kms_big_fb@yf-tiled-addfb-size-offset-overflow +kms_big_fb@4-tiled-addfb-size-offset-overflow +kms_big_fb@linear-addfb +kms_big_fb@x-tiled-addfb +kms_big_fb@y-tiled-addfb +kms_big_fb@yf-tiled-addfb +kms_big_fb@4-tiled-addfb +kms_big_fb@linear-8bpp-rotate-0 +kms_big_fb@linear-8bpp-rotate-90 +kms_big_fb@linear-8bpp-rotate-180 +kms_big_fb@linear-8bpp-rotate-270 +kms_big_fb@linear-16bpp-rotate-0 +kms_big_fb@linear-16bpp-rotate-90 +kms_big_fb@linear-16bpp-rotate-180 +kms_big_fb@linear-16bpp-rotate-270 +kms_big_fb@linear-32bpp-rotate-0 +kms_big_fb@linear-32bpp-rotate-90 +kms_big_fb@linear-32bpp-rotate-180 +kms_big_fb@linear-32bpp-rotate-270 +kms_big_fb@linear-64bpp-rotate-0 +kms_big_fb@linear-64bpp-rotate-90 +kms_big_fb@linear-64bpp-rotate-180 +kms_big_fb@linear-64bpp-rotate-270 +kms_big_fb@x-tiled-8bpp-rotate-0 +kms_big_fb@x-tiled-8bpp-rotate-90 +kms_big_fb@x-tiled-8bpp-rotate-180 +kms_big_fb@x-tiled-8bpp-rotate-270 +kms_big_fb@x-tiled-16bpp-rotate-0 +kms_big_fb@x-tiled-16bpp-rotate-90 +kms_big_fb@x-tiled-16bpp-rotate-180 +kms_big_fb@x-tiled-16bpp-rotate-270 +kms_big_fb@x-tiled-32bpp-rotate-0 +kms_big_fb@x-tiled-32bpp-rotate-90 +kms_big_fb@x-tiled-32bpp-rotate-180 +kms_big_fb@x-tiled-32bpp-rotate-270 +kms_big_fb@x-tiled-64bpp-rotate-0 +kms_big_fb@x-tiled-64bpp-rotate-90 +kms_big_fb@x-tiled-64bpp-rotate-180 +kms_big_fb@x-tiled-64bpp-rotate-270 +kms_big_fb@y-tiled-8bpp-rotate-0 +kms_big_fb@y-tiled-8bpp-rotate-90 +kms_big_fb@y-tiled-8bpp-rotate-180 +kms_big_fb@y-tiled-8bpp-rotate-270 +kms_big_fb@y-tiled-16bpp-rotate-0 +kms_big_fb@y-tiled-16bpp-rotate-90 +kms_big_fb@y-tiled-16bpp-rotate-180 +kms_big_fb@y-tiled-16bpp-rotate-270 +kms_big_fb@y-tiled-32bpp-rotate-0 +kms_big_fb@y-tiled-32bpp-rotate-90 +kms_big_fb@y-tiled-32bpp-rotate-180 +kms_big_fb@y-tiled-32bpp-rotate-270 +kms_big_fb@y-tiled-64bpp-rotate-0 +kms_big_fb@y-tiled-64bpp-rotate-90 +kms_big_fb@y-tiled-64bpp-rotate-180 +kms_big_fb@y-tiled-64bpp-rotate-270 +kms_big_fb@yf-tiled-8bpp-rotate-0 +kms_big_fb@yf-tiled-8bpp-rotate-90 +kms_big_fb@yf-tiled-8bpp-rotate-180 +kms_big_fb@yf-tiled-8bpp-rotate-270 +kms_big_fb@yf-tiled-16bpp-rotate-0 +kms_big_fb@yf-tiled-16bpp-rotate-90 +kms_big_fb@yf-tiled-16bpp-rotate-180 +kms_big_fb@yf-tiled-16bpp-rotate-270 +kms_big_fb@yf-tiled-32bpp-rotate-0 +kms_big_fb@yf-tiled-32bpp-rotate-90 +kms_big_fb@yf-tiled-32bpp-rotate-180 +kms_big_fb@yf-tiled-32bpp-rotate-270 +kms_big_fb@yf-tiled-64bpp-rotate-0 +kms_big_fb@yf-tiled-64bpp-rotate-90 +kms_big_fb@yf-tiled-64bpp-rotate-180 +kms_big_fb@yf-tiled-64bpp-rotate-270 +kms_big_fb@4-tiled-8bpp-rotate-0 +kms_big_fb@4-tiled-8bpp-rotate-90 +kms_big_fb@4-tiled-8bpp-rotate-180 +kms_big_fb@4-tiled-8bpp-rotate-270 +kms_big_fb@4-tiled-16bpp-rotate-0 +kms_big_fb@4-tiled-16bpp-rotate-90 +kms_big_fb@4-tiled-16bpp-rotate-180 +kms_big_fb@4-tiled-16bpp-rotate-270 +kms_big_fb@4-tiled-32bpp-rotate-0 +kms_big_fb@4-tiled-32bpp-rotate-90 +kms_big_fb@4-tiled-32bpp-rotate-180 +kms_big_fb@4-tiled-32bpp-rotate-270 +kms_big_fb@4-tiled-64bpp-rotate-0 +kms_big_fb@4-tiled-64bpp-rotate-90 +kms_big_fb@4-tiled-64bpp-rotate-180 +kms_big_fb@4-tiled-64bpp-rotate-270 +kms_big_fb@linear-max-hw-stride-32bpp-rotate-0 +kms_big_fb@linear-max-hw-stride-32bpp-rotate-180 +kms_big_fb@linear-max-hw-stride-64bpp-rotate-0 +kms_big_fb@linear-max-hw-stride-64bpp-rotate-180 +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-0 +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-0-async-flip +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180 +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0 +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-async-flip +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180 +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-async-flip +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-0-hflip +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip +kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip +kms_big_fb@x-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0 +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-async-flip +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180 +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-async-flip +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0 +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-async-flip +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180 +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-async-flip +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip +kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-hflip +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip +kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0 +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-async-flip +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180 +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-async-flip +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0 +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0-async-flip +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-180 +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-180-async-flip +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-hflip +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-hflip +kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0-hflip +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-180-hflip +kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-0 +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-0-async-flip +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180 +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-async-flip +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0 +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-async-flip +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180 +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180-async-flip +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-0-hflip +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-0-hflip-async-flip +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip +kms_big_fb@4-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180-hflip +kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip +kms_big_joiner@basic +kms_big_joiner@invalid-modeset +kms_big_joiner@2x-modeset +kms_busy@basic +kms_busy@basic-hang +kms_busy@extended-pageflip-modeset-hang-oldfb +kms_busy@extended-pageflip-hang-oldfb +kms_busy@extended-pageflip-hang-newfb +kms_busy@extended-modeset-hang-oldfb +kms_busy@extended-modeset-hang-newfb +kms_busy@extended-modeset-hang-oldfb-with-reset +kms_busy@extended-modeset-hang-newfb-with-reset +kms_bw@linear-tiling-1-displays-1920x1080p +kms_bw@linear-tiling-1-displays-2560x1440p +kms_bw@linear-tiling-1-displays-3840x2160p +kms_bw@linear-tiling-2-displays-1920x1080p +kms_bw@linear-tiling-2-displays-2560x1440p +kms_bw@linear-tiling-2-displays-3840x2160p +kms_bw@linear-tiling-3-displays-1920x1080p +kms_bw@linear-tiling-3-displays-2560x1440p +kms_bw@linear-tiling-3-displays-3840x2160p +kms_bw@linear-tiling-4-displays-1920x1080p +kms_bw@linear-tiling-4-displays-2560x1440p +kms_bw@linear-tiling-4-displays-3840x2160p +kms_bw@linear-tiling-5-displays-1920x1080p +kms_bw@linear-tiling-5-displays-2560x1440p +kms_bw@linear-tiling-5-displays-3840x2160p +kms_bw@linear-tiling-6-displays-1920x1080p +kms_bw@linear-tiling-6-displays-2560x1440p +kms_bw@linear-tiling-6-displays-3840x2160p +kms_bw@linear-tiling-7-displays-1920x1080p +kms_bw@linear-tiling-7-displays-2560x1440p +kms_bw@linear-tiling-7-displays-3840x2160p +kms_bw@linear-tiling-8-displays-1920x1080p +kms_bw@linear-tiling-8-displays-2560x1440p +kms_bw@linear-tiling-8-displays-3840x2160p +kms_ccs@pipe-A-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-A-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-A-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-A-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-A-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-A-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-A-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-A-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-A-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-A-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-A-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-A-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-A-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-A-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-A-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-A-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-A-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-A-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-A-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-A-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-A-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-A-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-A-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-A-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-A-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-A-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-A-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-A-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-A-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-A-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-A-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-A-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-B-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-B-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-B-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-B-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-B-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-B-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-B-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-B-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-B-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-B-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-B-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-B-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-B-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-B-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-B-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-B-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-B-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-B-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-B-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-B-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-B-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-B-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-B-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-B-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-B-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-B-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-B-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-B-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-B-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-B-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-B-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-C-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-C-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-C-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-C-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-C-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-C-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-C-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-C-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-C-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-C-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-C-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-C-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-C-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-C-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-C-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-C-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-C-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-C-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-C-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-C-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-C-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-C-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-C-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-C-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-C-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-C-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-C-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-C-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-C-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-C-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-C-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-D-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-D-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-D-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-D-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-D-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-D-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-D-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-D-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-D-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-D-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-D-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-D-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-D-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-D-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-D-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-D-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-D-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-D-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-D-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-D-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-D-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-D-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-D-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-D-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-D-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-D-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-D-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-D-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-D-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-D-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-D-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-E-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-E-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-E-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-E-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-E-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-E-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-E-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-E-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-E-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-E-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-E-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-E-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-E-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-E-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-E-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-E-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-E-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-E-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-E-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-E-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-E-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-E-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-E-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-E-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-E-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-E-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-E-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-E-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-E-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-E-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-E-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-F-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-F-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-F-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-F-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-F-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-F-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-F-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-F-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-F-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-F-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-F-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-F-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-F-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-F-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-F-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-F-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-F-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-F-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-F-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-F-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-F-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-F-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-F-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-F-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-F-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-F-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-F-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-F-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-F-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-F-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-F-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-G-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-G-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-G-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-G-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-G-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-G-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-G-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-G-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-G-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-G-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-G-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-G-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-G-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-G-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-G-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-G-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-G-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-G-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-G-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-G-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-G-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-G-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-G-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-G-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-G-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-G-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-G-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-G-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-G-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-G-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-G-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-bad-pixel-format-y_tiled_ccs +kms_ccs@pipe-H-bad-pixel-format-yf_tiled_ccs +kms_ccs@pipe-H-bad-pixel-format-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-bad-pixel-format-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-bad-pixel-format-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-bad-pixel-format-4_tiled_dg2_rc_ccs +kms_ccs@pipe-H-bad-pixel-format-4_tiled_dg2_mc_ccs +kms_ccs@pipe-H-bad-pixel-format-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-H-bad-pixel-format-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-bad-pixel-format-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-bad-pixel-format-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-bad-rotation-90-y_tiled_ccs +kms_ccs@pipe-H-bad-rotation-90-yf_tiled_ccs +kms_ccs@pipe-H-bad-rotation-90-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-bad-rotation-90-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-bad-rotation-90-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-bad-rotation-90-4_tiled_dg2_rc_ccs +kms_ccs@pipe-H-bad-rotation-90-4_tiled_dg2_mc_ccs +kms_ccs@pipe-H-bad-rotation-90-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-H-bad-rotation-90-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-bad-rotation-90-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-bad-rotation-90-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-crc-primary-basic-y_tiled_ccs +kms_ccs@pipe-H-crc-primary-basic-yf_tiled_ccs +kms_ccs@pipe-H-crc-primary-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-crc-primary-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-crc-primary-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-crc-primary-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-H-crc-primary-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-H-crc-primary-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-H-crc-primary-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-crc-primary-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-crc-primary-basic-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-crc-primary-rotation-180-y_tiled_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-yf_tiled_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-crc-primary-rotation-180-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-4_tiled_dg2_rc_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-4_tiled_dg2_mc_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-H-crc-primary-rotation-180-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-crc-primary-rotation-180-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-random-ccs-data-y_tiled_ccs +kms_ccs@pipe-H-random-ccs-data-yf_tiled_ccs +kms_ccs@pipe-H-random-ccs-data-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-random-ccs-data-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-random-ccs-data-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-random-ccs-data-4_tiled_dg2_rc_ccs +kms_ccs@pipe-H-random-ccs-data-4_tiled_dg2_mc_ccs +kms_ccs@pipe-H-random-ccs-data-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-H-random-ccs-data-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-random-ccs-data-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-random-ccs-data-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-missing-ccs-buffer-y_tiled_ccs +kms_ccs@pipe-H-missing-ccs-buffer-yf_tiled_ccs +kms_ccs@pipe-H-missing-ccs-buffer-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-missing-ccs-buffer-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-missing-ccs-buffer-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-missing-ccs-buffer-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-missing-ccs-buffer-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-missing-ccs-buffer-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-ccs-on-another-bo-y_tiled_ccs +kms_ccs@pipe-H-ccs-on-another-bo-yf_tiled_ccs +kms_ccs@pipe-H-ccs-on-another-bo-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-ccs-on-another-bo-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-ccs-on-another-bo-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-ccs-on-another-bo-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-ccs-on-another-bo-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-ccs-on-another-bo-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-bad-aux-stride-y_tiled_ccs +kms_ccs@pipe-H-bad-aux-stride-yf_tiled_ccs +kms_ccs@pipe-H-bad-aux-stride-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-bad-aux-stride-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-bad-aux-stride-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-bad-aux-stride-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-bad-aux-stride-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-bad-aux-stride-4_tiled_mtl_rc_ccs_cc +kms_ccs@pipe-H-crc-sprite-planes-basic-y_tiled_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-yf_tiled_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-y_tiled_gen12_rc_ccs_cc +kms_ccs@pipe-H-crc-sprite-planes-basic-y_tiled_gen12_mc_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-4_tiled_dg2_mc_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-4_tiled_dg2_rc_ccs_cc +kms_ccs@pipe-H-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-4_tiled_mtl_mc_ccs +kms_ccs@pipe-H-crc-sprite-planes-basic-4_tiled_mtl_rc_ccs_cc +kms_cdclk@plane-scaling +kms_cdclk@mode-transition +kms_cdclk@mode-transition-all-outputs +kms_color@degamma +kms_color@gamma +kms_color@legacy-gamma +kms_color@legacy-gamma-reset +kms_color@ctm-red-to-blue +kms_color@ctm-green-to-red +kms_color@ctm-blue-to-red +kms_color@ctm-max +kms_color@ctm-negative +kms_color@ctm-0-25 +kms_color@ctm-0-50 +kms_color@ctm-0-75 +kms_color@ctm-signed +kms_color@deep-color +kms_color@invalid-gamma-lut-sizes +kms_color@invalid-degamma-lut-sizes +kms_color@invalid-ctm-matrix-sizes +kms_concurrent@pipe-A +kms_concurrent@pipe-B +kms_concurrent@pipe-C +kms_concurrent@pipe-D +kms_concurrent@pipe-E +kms_concurrent@pipe-F +kms_concurrent@pipe-G +kms_concurrent@pipe-H +kms_content_protection@legacy +kms_content_protection@atomic +kms_content_protection@atomic-dpms +kms_content_protection@LIC +kms_content_protection@type1 +kms_content_protection@mei_interface +kms_content_protection@content_type_change +kms_content_protection@uevent +kms_content_protection@srm +kms_content_protection@dp-mst-type-0 +kms_content_protection@dp-mst-lic-type-0 +kms_content_protection@dp-mst-type-1 +kms_content_protection@dp-mst-lic-type-1 +kms_cursor_crc@cursor-size-change +kms_cursor_crc@cursor-alpha-opaque +kms_cursor_crc@cursor-alpha-transparent +kms_cursor_crc@cursor-dpms +kms_cursor_crc@cursor-suspend +kms_cursor_crc@cursor-onscreen-32x32 +kms_cursor_crc@cursor-offscreen-32x32 +kms_cursor_crc@cursor-sliding-32x32 +kms_cursor_crc@cursor-random-32x32 +kms_cursor_crc@cursor-rapid-movement-32x32 +kms_cursor_crc@cursor-onscreen-32x10 +kms_cursor_crc@cursor-offscreen-32x10 +kms_cursor_crc@cursor-sliding-32x10 +kms_cursor_crc@cursor-random-32x10 +kms_cursor_crc@cursor-rapid-movement-32x10 +kms_cursor_crc@cursor-onscreen-64x64 +kms_cursor_crc@cursor-offscreen-64x64 +kms_cursor_crc@cursor-sliding-64x64 +kms_cursor_crc@cursor-random-64x64 +kms_cursor_crc@cursor-rapid-movement-64x64 +kms_cursor_crc@cursor-onscreen-64x21 +kms_cursor_crc@cursor-offscreen-64x21 +kms_cursor_crc@cursor-sliding-64x21 +kms_cursor_crc@cursor-random-64x21 +kms_cursor_crc@cursor-rapid-movement-64x21 +kms_cursor_crc@cursor-onscreen-128x128 +kms_cursor_crc@cursor-offscreen-128x128 +kms_cursor_crc@cursor-sliding-128x128 +kms_cursor_crc@cursor-random-128x128 +kms_cursor_crc@cursor-rapid-movement-128x128 +kms_cursor_crc@cursor-onscreen-128x42 +kms_cursor_crc@cursor-offscreen-128x42 +kms_cursor_crc@cursor-sliding-128x42 +kms_cursor_crc@cursor-random-128x42 +kms_cursor_crc@cursor-rapid-movement-128x42 +kms_cursor_crc@cursor-onscreen-256x256 +kms_cursor_crc@cursor-offscreen-256x256 +kms_cursor_crc@cursor-sliding-256x256 +kms_cursor_crc@cursor-random-256x256 +kms_cursor_crc@cursor-rapid-movement-256x256 +kms_cursor_crc@cursor-onscreen-256x85 +kms_cursor_crc@cursor-offscreen-256x85 +kms_cursor_crc@cursor-sliding-256x85 +kms_cursor_crc@cursor-random-256x85 +kms_cursor_crc@cursor-rapid-movement-256x85 +kms_cursor_crc@cursor-onscreen-512x512 +kms_cursor_crc@cursor-offscreen-512x512 +kms_cursor_crc@cursor-sliding-512x512 +kms_cursor_crc@cursor-random-512x512 +kms_cursor_crc@cursor-rapid-movement-512x512 +kms_cursor_crc@cursor-onscreen-512x170 +kms_cursor_crc@cursor-offscreen-512x170 +kms_cursor_crc@cursor-sliding-512x170 +kms_cursor_crc@cursor-random-512x170 +kms_cursor_crc@cursor-rapid-movement-512x170 +kms_cursor_crc@cursor-onscreen-max-size +kms_cursor_crc@cursor-offscreen-max-size +kms_cursor_crc@cursor-sliding-max-size +kms_cursor_crc@cursor-random-max-size +kms_cursor_crc@cursor-rapid-movement-max-size +kms_cursor_legacy@single-bo +kms_cursor_legacy@single-move +kms_cursor_legacy@forked-bo +kms_cursor_legacy@forked-move +kms_cursor_legacy@torture-bo +kms_cursor_legacy@torture-move +kms_cursor_legacy@nonblocking-modeset-vs-cursor-atomic +kms_cursor_legacy@long-nonblocking-modeset-vs-cursor-atomic +kms_cursor_legacy@2x-flip-vs-cursor-legacy +kms_cursor_legacy@2x-flip-vs-cursor-atomic +kms_cursor_legacy@2x-long-flip-vs-cursor-legacy +kms_cursor_legacy@2x-long-flip-vs-cursor-atomic +kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic +kms_cursor_legacy@2x-long-nonblocking-modeset-vs-cursor-atomic +kms_cursor_legacy@2x-cursor-vs-flip-legacy +kms_cursor_legacy@2x-long-cursor-vs-flip-legacy +kms_cursor_legacy@2x-cursor-vs-flip-atomic +kms_cursor_legacy@2x-long-cursor-vs-flip-atomic +kms_cursor_legacy@flip-vs-cursor-crc-legacy +kms_cursor_legacy@flip-vs-cursor-crc-atomic +kms_cursor_legacy@flip-vs-cursor-busy-crc-legacy +kms_cursor_legacy@flip-vs-cursor-busy-crc-atomic +kms_cursor_legacy@basic-flip-before-cursor-legacy +kms_cursor_legacy@basic-busy-flip-before-cursor-legacy +kms_cursor_legacy@basic-flip-after-cursor-legacy +kms_cursor_legacy@basic-flip-before-cursor-varying-size +kms_cursor_legacy@basic-busy-flip-before-cursor-varying-size +kms_cursor_legacy@basic-flip-after-cursor-varying-size +kms_cursor_legacy@short-flip-before-cursor-toggle +kms_cursor_legacy@short-busy-flip-before-cursor-toggle +kms_cursor_legacy@short-flip-after-cursor-toggle +kms_cursor_legacy@basic-flip-before-cursor-atomic +kms_cursor_legacy@basic-busy-flip-before-cursor-atomic +kms_cursor_legacy@basic-flip-after-cursor-atomic +kms_cursor_legacy@short-flip-before-cursor-atomic-transitions +kms_cursor_legacy@short-busy-flip-before-cursor-atomic-transitions +kms_cursor_legacy@short-flip-after-cursor-atomic-transitions +kms_cursor_legacy@short-flip-before-cursor-atomic-transitions-varying-size +kms_cursor_legacy@short-busy-flip-before-cursor-atomic-transitions-varying-size +kms_cursor_legacy@short-flip-after-cursor-atomic-transitions-varying-size +kms_cursor_legacy@cursor-vs-flip-legacy +kms_cursor_legacy@flip-vs-cursor-legacy +kms_cursor_legacy@cursorA-vs-flipA-legacy +kms_cursor_legacy@cursorA-vs-flipB-legacy +kms_cursor_legacy@cursorB-vs-flipA-legacy +kms_cursor_legacy@cursorB-vs-flipB-legacy +kms_cursor_legacy@cursor-vs-flip-varying-size +kms_cursor_legacy@flip-vs-cursor-varying-size +kms_cursor_legacy@cursorA-vs-flipA-varying-size +kms_cursor_legacy@cursorA-vs-flipB-varying-size +kms_cursor_legacy@cursorB-vs-flipA-varying-size +kms_cursor_legacy@cursorB-vs-flipB-varying-size +kms_cursor_legacy@cursor-vs-flip-toggle +kms_cursor_legacy@flip-vs-cursor-toggle +kms_cursor_legacy@cursorA-vs-flipA-toggle +kms_cursor_legacy@cursorA-vs-flipB-toggle +kms_cursor_legacy@cursorB-vs-flipA-toggle +kms_cursor_legacy@cursorB-vs-flipB-toggle +kms_cursor_legacy@cursor-vs-flip-atomic +kms_cursor_legacy@flip-vs-cursor-atomic +kms_cursor_legacy@cursorA-vs-flipA-atomic +kms_cursor_legacy@cursorA-vs-flipB-atomic +kms_cursor_legacy@cursorB-vs-flipA-atomic +kms_cursor_legacy@cursorB-vs-flipB-atomic +kms_cursor_legacy@cursor-vs-flip-atomic-transitions +kms_cursor_legacy@flip-vs-cursor-atomic-transitions +kms_cursor_legacy@cursorA-vs-flipA-atomic-transitions +kms_cursor_legacy@cursorA-vs-flipB-atomic-transitions +kms_cursor_legacy@cursorB-vs-flipA-atomic-transitions +kms_cursor_legacy@cursorB-vs-flipB-atomic-transitions +kms_cursor_legacy@cursor-vs-flip-atomic-transitions-varying-size +kms_cursor_legacy@flip-vs-cursor-atomic-transitions-varying-size +kms_cursor_legacy@cursorA-vs-flipA-atomic-transitions-varying-size +kms_cursor_legacy@cursorA-vs-flipB-atomic-transitions-varying-size +kms_cursor_legacy@cursorB-vs-flipA-atomic-transitions-varying-size +kms_cursor_legacy@cursorB-vs-flipB-atomic-transitions-varying-size +kms_dither@FB-8BPC-Vs-Panel-6BPC +kms_dither@FB-8BPC-Vs-Panel-8BPC +kms_dp_aux_dev +kms_tiled_display@basic-test-pattern +kms_tiled_display@basic-test-pattern-with-chamelium +kms_draw_crc@draw-method-mmap-cpu +kms_draw_crc@draw-method-mmap-gtt +kms_draw_crc@draw-method-mmap-wc +kms_draw_crc@draw-method-pwrite +kms_draw_crc@draw-method-blt +kms_draw_crc@draw-method-render +kms_draw_crc@fill-fb +kms_dsc@dsc-basic +kms_dsc@dsc-with-formats +kms_dsc@dsc-with-bpc +kms_dsc@dsc-with-bpc-formats +kms_dsc@dsc-with-output-formats +kms_fbcon_fbt@fbc +kms_fbcon_fbt@psr +kms_fbcon_fbt@fbc-suspend +kms_fbcon_fbt@psr-suspend +kms_fence_pin_leak +kms_flip@nonblocking-read +kms_flip@wf_vblank-ts-check +kms_flip@2x-wf_vblank-ts-check +kms_flip@blocking-wf_vblank +kms_flip@2x-blocking-wf_vblank +kms_flip@absolute-wf_vblank +kms_flip@2x-absolute-wf_vblank +kms_flip@blocking-absolute-wf_vblank +kms_flip@2x-blocking-absolute-wf_vblank +kms_flip@basic-plain-flip +kms_flip@2x-plain-flip +kms_flip@busy-flip +kms_flip@2x-busy-flip +kms_flip@flip-vs-fences +kms_flip@2x-flip-vs-fences +kms_flip@plain-flip-ts-check +kms_flip@2x-plain-flip-ts-check +kms_flip@plain-flip-fb-recreate +kms_flip@2x-plain-flip-fb-recreate +kms_flip@flip-vs-rmfb +kms_flip@2x-flip-vs-rmfb +kms_flip@basic-flip-vs-dpms +kms_flip@2x-flip-vs-dpms +kms_flip@flip-vs-panning +kms_flip@2x-flip-vs-panning +kms_flip@basic-flip-vs-modeset +kms_flip@2x-flip-vs-modeset +kms_flip@flip-vs-expired-vblank +kms_flip@2x-flip-vs-expired-vblank +kms_flip@flip-vs-absolute-wf_vblank +kms_flip@2x-flip-vs-absolute-wf_vblank +kms_flip@basic-flip-vs-wf_vblank +kms_flip@2x-flip-vs-wf_vblank +kms_flip@flip-vs-blocking-wf-vblank +kms_flip@2x-flip-vs-blocking-wf-vblank +kms_flip@flip-vs-modeset-vs-hang +kms_flip@2x-flip-vs-modeset-vs-hang +kms_flip@flip-vs-panning-vs-hang +kms_flip@2x-flip-vs-panning-vs-hang +kms_flip@flip-vs-dpms-off-vs-modeset +kms_flip@2x-flip-vs-dpms-off-vs-modeset +kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset +kms_flip@2x-single-buffer-flip-vs-dpms-off-vs-modeset +kms_flip@dpms-off-confusion +kms_flip@nonexisting-fb +kms_flip@2x-nonexisting-fb +kms_flip@dpms-vs-vblank-race +kms_flip@2x-dpms-vs-vblank-race +kms_flip@modeset-vs-vblank-race +kms_flip@2x-modeset-vs-vblank-race +kms_flip@bo-too-big +kms_flip@flip-vs-suspend +kms_flip@2x-flip-vs-suspend +kms_flip@wf_vblank-ts-check-interruptible +kms_flip@2x-wf_vblank-ts-check-interruptible +kms_flip@absolute-wf_vblank-interruptible +kms_flip@2x-absolute-wf_vblank-interruptible +kms_flip@blocking-absolute-wf_vblank-interruptible +kms_flip@2x-blocking-absolute-wf_vblank-interruptible +kms_flip@plain-flip-interruptible +kms_flip@2x-plain-flip-interruptible +kms_flip@flip-vs-fences-interruptible +kms_flip@2x-flip-vs-fences-interruptible +kms_flip@plain-flip-ts-check-interruptible +kms_flip@2x-plain-flip-ts-check-interruptible +kms_flip@plain-flip-fb-recreate-interruptible +kms_flip@2x-plain-flip-fb-recreate-interruptible +kms_flip@flip-vs-rmfb-interruptible +kms_flip@2x-flip-vs-rmfb-interruptible +kms_flip@flip-vs-panning-interruptible +kms_flip@2x-flip-vs-panning-interruptible +kms_flip@flip-vs-expired-vblank-interruptible +kms_flip@2x-flip-vs-expired-vblank-interruptible +kms_flip@flip-vs-absolute-wf_vblank-interruptible +kms_flip@2x-flip-vs-absolute-wf_vblank-interruptible +kms_flip@flip-vs-wf_vblank-interruptible +kms_flip@2x-flip-vs-wf_vblank-interruptible +kms_flip@flip-vs-dpms-off-vs-modeset-interruptible +kms_flip@2x-flip-vs-dpms-off-vs-modeset-interruptible +kms_flip@single-buffer-flip-vs-dpms-off-vs-modeset-interruptible +kms_flip@2x-single-buffer-flip-vs-dpms-off-vs-modeset-interruptible +kms_flip@dpms-off-confusion-interruptible +kms_flip@nonexisting-fb-interruptible +kms_flip@2x-nonexisting-fb-interruptible +kms_flip@dpms-vs-vblank-race-interruptible +kms_flip@2x-dpms-vs-vblank-race-interruptible +kms_flip@modeset-vs-vblank-race-interruptible +kms_flip@2x-modeset-vs-vblank-race-interruptible +kms_flip@bo-too-big-interruptible +kms_flip@flip-vs-suspend-interruptible +kms_flip@2x-flip-vs-suspend-interruptible +kms_flip_event_leak@basic +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-yftile-to-64bpp-yftile-downscaling +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling +kms_flip_scaled_crc@flip-32bpp-4tile-to-64bpp-4tile-downscaling +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-yftile-to-32bpp-yftile-downscaling +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling +kms_flip_scaled_crc@flip-64bpp-4tile-to-32bpp-4tile-downscaling +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-downscaling +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling +kms_flip_scaled_crc@flip-64bpp-4tile-to-16bpp-4tile-downscaling +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-downscaling +kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytileccs-downscaling +kms_flip_scaled_crc@flip-32bpp-yftile-to-32bpp-yftileccs-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling +kms_flip_scaled_crc@flip-64bpp-4tile-to-32bpp-4tiledg2rcccs-downscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-32bpp-yftile-to-64bpp-yftile-upscaling +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling +kms_flip_scaled_crc@flip-32bpp-4tile-to-64bpp-4tile-upscaling +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-yftile-to-32bpp-yftile-upscaling +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling +kms_flip_scaled_crc@flip-64bpp-4tile-to-32bpp-4tile-upscaling +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-upscaling +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling +kms_flip_scaled_crc@flip-64bpp-4tile-to-16bpp-4tile-upscaling +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-upscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling +kms_flip_scaled_crc@flip-32bpp-4tile-to-32bpp-4tiledg2rcccs-upscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-32bpp-ytileccs-upscaling +kms_flip_scaled_crc@flip-32bpp-yftile-to-32bpp-yftileccs-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling +kms_flip_scaled_crc@flip-64bpp-4tile-to-32bpp-4tiledg2rcccs-upscaling +kms_force_connector_basic@force-load-detect +kms_force_connector_basic@force-connector-state +kms_force_connector_basic@force-edid +kms_force_connector_basic@prune-stale-modes +kms_frontbuffer_tracking@fbc-1p-rte +kms_frontbuffer_tracking@fbc-2p-rte +kms_frontbuffer_tracking@psr-1p-rte +kms_frontbuffer_tracking@psr-2p-rte +kms_frontbuffer_tracking@fbcpsr-1p-rte +kms_frontbuffer_tracking@fbcpsr-2p-rte +kms_frontbuffer_tracking@drrs-1p-rte +kms_frontbuffer_tracking@drrs-2p-rte +kms_frontbuffer_tracking@fbcdrrs-1p-rte +kms_frontbuffer_tracking@fbcdrrs-2p-rte +kms_frontbuffer_tracking@psrdrrs-1p-rte +kms_frontbuffer_tracking@psrdrrs-2p-rte +kms_frontbuffer_tracking@fbcpsrdrrs-1p-rte +kms_frontbuffer_tracking@fbcpsrdrrs-2p-rte +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbc-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbc-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@psr-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@psr-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@psr-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psr-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@psr-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psr-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@drrs-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@drrs-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@drrs-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@drrs-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-offscren-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-shrfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-shrfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-shrfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-shrfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-shrfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-pri-shrfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-draw-render +kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbc-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbc-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbc-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@psr-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@psr-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@psr-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@psr-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@psr-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@psr-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@psr-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@psr-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@psr-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@psr-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@psr-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@psr-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@psr-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@drrs-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@drrs-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@drrs-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-indfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-indfb-msflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-indfb-plflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-shrfb-pgflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-shrfb-msflip-blt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-shrfb-plflip-blt +kms_frontbuffer_tracking@fbc-1p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbc-2p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-1p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-2p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-1p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-2p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-1p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-indfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-2p-shrfb-fliptrack-mmap-gtt +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbc-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@fbc-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@psr-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@drrs-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-move +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-cur-indfb-onoff +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-move +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-onoff +kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbc-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcpsr-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@drrs-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@drrs-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcdrrs-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@psrdrrs-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@psrdrrs-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@psrdrrs-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcpsrdrrs-1p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcpsrdrrs-2p-primscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbcpsrdrrs-2p-scndscrn-spr-indfb-fullscreen +kms_frontbuffer_tracking@fbc-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbc-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@psr-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@psr-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbcpsr-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbcpsr-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@drrs-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@drrs-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbcdrrs-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbcdrrs-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@psrdrrs-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@psrdrrs-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbcpsrdrrs-1p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbcpsrdrrs-2p-pri-indfb-multidraw +kms_frontbuffer_tracking@fbc-farfromfence-mmap-gtt +kms_frontbuffer_tracking@psr-farfromfence-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-farfromfence-mmap-gtt +kms_frontbuffer_tracking@drrs-farfromfence-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-farfromfence-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-farfromfence-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-farfromfence-mmap-gtt +kms_frontbuffer_tracking@fbc-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@fbc-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@fbc-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@fbc-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@fbc-rgb565-draw-pwrite +kms_frontbuffer_tracking@fbc-rgb101010-draw-pwrite +kms_frontbuffer_tracking@fbc-rgb565-draw-blt +kms_frontbuffer_tracking@fbc-rgb101010-draw-blt +kms_frontbuffer_tracking@fbc-rgb565-draw-render +kms_frontbuffer_tracking@fbc-rgb101010-draw-render +kms_frontbuffer_tracking@psr-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@psr-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@psr-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@psr-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@psr-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@psr-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@psr-rgb565-draw-pwrite +kms_frontbuffer_tracking@psr-rgb101010-draw-pwrite +kms_frontbuffer_tracking@psr-rgb565-draw-blt +kms_frontbuffer_tracking@psr-rgb101010-draw-blt +kms_frontbuffer_tracking@psr-rgb565-draw-render +kms_frontbuffer_tracking@psr-rgb101010-draw-render +kms_frontbuffer_tracking@fbcpsr-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsr-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsr-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsr-rgb565-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-rgb101010-draw-pwrite +kms_frontbuffer_tracking@fbcpsr-rgb565-draw-blt +kms_frontbuffer_tracking@fbcpsr-rgb101010-draw-blt +kms_frontbuffer_tracking@fbcpsr-rgb565-draw-render +kms_frontbuffer_tracking@fbcpsr-rgb101010-draw-render +kms_frontbuffer_tracking@drrs-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@drrs-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@drrs-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@drrs-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@drrs-rgb565-draw-pwrite +kms_frontbuffer_tracking@drrs-rgb101010-draw-pwrite +kms_frontbuffer_tracking@drrs-rgb565-draw-blt +kms_frontbuffer_tracking@drrs-rgb101010-draw-blt +kms_frontbuffer_tracking@drrs-rgb565-draw-render +kms_frontbuffer_tracking@drrs-rgb101010-draw-render +kms_frontbuffer_tracking@fbcdrrs-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@fbcdrrs-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@fbcdrrs-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@fbcdrrs-rgb565-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-rgb101010-draw-pwrite +kms_frontbuffer_tracking@fbcdrrs-rgb565-draw-blt +kms_frontbuffer_tracking@fbcdrrs-rgb101010-draw-blt +kms_frontbuffer_tracking@fbcdrrs-rgb565-draw-render +kms_frontbuffer_tracking@fbcdrrs-rgb101010-draw-render +kms_frontbuffer_tracking@psrdrrs-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@psrdrrs-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@psrdrrs-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@psrdrrs-rgb565-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-rgb101010-draw-pwrite +kms_frontbuffer_tracking@psrdrrs-rgb565-draw-blt +kms_frontbuffer_tracking@psrdrrs-rgb101010-draw-blt +kms_frontbuffer_tracking@psrdrrs-rgb565-draw-render +kms_frontbuffer_tracking@psrdrrs-rgb101010-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-rgb565-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-rgb101010-draw-mmap-cpu +kms_frontbuffer_tracking@fbcpsrdrrs-rgb565-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-rgb101010-draw-mmap-gtt +kms_frontbuffer_tracking@fbcpsrdrrs-rgb565-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-rgb101010-draw-mmap-wc +kms_frontbuffer_tracking@fbcpsrdrrs-rgb565-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-rgb101010-draw-pwrite +kms_frontbuffer_tracking@fbcpsrdrrs-rgb565-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-rgb101010-draw-blt +kms_frontbuffer_tracking@fbcpsrdrrs-rgb565-draw-render +kms_frontbuffer_tracking@fbcpsrdrrs-rgb101010-draw-render +kms_frontbuffer_tracking@fbc-indfb-scaledprimary +kms_frontbuffer_tracking@fbc-shrfb-scaledprimary +kms_frontbuffer_tracking@psr-indfb-scaledprimary +kms_frontbuffer_tracking@psr-shrfb-scaledprimary +kms_frontbuffer_tracking@fbcpsr-indfb-scaledprimary +kms_frontbuffer_tracking@fbcpsr-shrfb-scaledprimary +kms_frontbuffer_tracking@drrs-indfb-scaledprimary +kms_frontbuffer_tracking@drrs-shrfb-scaledprimary +kms_frontbuffer_tracking@fbcdrrs-indfb-scaledprimary +kms_frontbuffer_tracking@fbcdrrs-shrfb-scaledprimary +kms_frontbuffer_tracking@psrdrrs-indfb-scaledprimary +kms_frontbuffer_tracking@psrdrrs-shrfb-scaledprimary +kms_frontbuffer_tracking@fbcpsrdrrs-indfb-scaledprimary +kms_frontbuffer_tracking@fbcpsrdrrs-shrfb-scaledprimary +kms_frontbuffer_tracking@fbc-modesetfrombusy +kms_frontbuffer_tracking@fbc-badstride +kms_frontbuffer_tracking@fbc-stridechange +kms_frontbuffer_tracking@fbc-tiling-linear +kms_frontbuffer_tracking@fbc-tiling-y +kms_frontbuffer_tracking@fbc-tiling-4 +kms_frontbuffer_tracking@fbc-suspend +kms_frontbuffer_tracking@psr-modesetfrombusy +kms_frontbuffer_tracking@psr-slowdraw +kms_frontbuffer_tracking@psr-suspend +kms_frontbuffer_tracking@fbcpsr-modesetfrombusy +kms_frontbuffer_tracking@fbcpsr-badstride +kms_frontbuffer_tracking@fbcpsr-stridechange +kms_frontbuffer_tracking@fbcpsr-tiling-linear +kms_frontbuffer_tracking@fbcpsr-tiling-y +kms_frontbuffer_tracking@fbcpsr-tiling-4 +kms_frontbuffer_tracking@fbcpsr-slowdraw +kms_frontbuffer_tracking@fbcpsr-suspend +kms_frontbuffer_tracking@drrs-modesetfrombusy +kms_frontbuffer_tracking@drrs-slowdraw +kms_frontbuffer_tracking@drrs-suspend +kms_frontbuffer_tracking@fbcdrrs-modesetfrombusy +kms_frontbuffer_tracking@fbcdrrs-badstride +kms_frontbuffer_tracking@fbcdrrs-stridechange +kms_frontbuffer_tracking@fbcdrrs-tiling-linear +kms_frontbuffer_tracking@fbcdrrs-tiling-y +kms_frontbuffer_tracking@fbcdrrs-tiling-4 +kms_frontbuffer_tracking@fbcdrrs-slowdraw +kms_frontbuffer_tracking@fbcdrrs-suspend +kms_frontbuffer_tracking@psrdrrs-modesetfrombusy +kms_frontbuffer_tracking@psrdrrs-slowdraw +kms_frontbuffer_tracking@psrdrrs-suspend +kms_frontbuffer_tracking@fbcpsrdrrs-modesetfrombusy +kms_frontbuffer_tracking@fbcpsrdrrs-badstride +kms_frontbuffer_tracking@fbcpsrdrrs-stridechange +kms_frontbuffer_tracking@fbcpsrdrrs-tiling-linear +kms_frontbuffer_tracking@fbcpsrdrrs-tiling-y +kms_frontbuffer_tracking@fbcpsrdrrs-tiling-4 +kms_frontbuffer_tracking@fbcpsrdrrs-slowdraw +kms_frontbuffer_tracking@fbcpsrdrrs-suspend +kms_frontbuffer_tracking@basic +kms_getfb@getfb-handle-zero +kms_getfb@getfb-handle-valid +kms_getfb@getfb-handle-closed +kms_getfb@getfb-handle-not-fb +kms_getfb@getfb-addfb-different-handles +kms_getfb@getfb-repeated-different-handles +kms_getfb@getfb-reject-ccs +kms_getfb@getfb2-handle-zero +kms_getfb@getfb2-handle-closed +kms_getfb@getfb2-handle-not-fb +kms_getfb@getfb2-accept-ccs +kms_getfb@getfb2-into-addfb2 +kms_getfb@getfb-handle-protection +kms_getfb@getfb2-handle-protection +kms_hdmi_inject@inject-4k +kms_hdmi_inject@inject-audio +kms_hdr@bpc-switch +kms_hdr@bpc-switch-dpms +kms_hdr@bpc-switch-suspend +kms_hdr@static-toggle +kms_hdr@static-toggle-dpms +kms_hdr@static-toggle-suspend +kms_hdr@static-swap +kms_hdr@invalid-metadata-sizes +kms_hdr@invalid-hdr +kms_invalid_mode@clock-too-high +kms_invalid_mode@zero-clock +kms_invalid_mode@int-max-clock +kms_invalid_mode@uint-max-clock +kms_invalid_mode@zero-hdisplay +kms_invalid_mode@zero-vdisplay +kms_invalid_mode@bad-hsync-start +kms_invalid_mode@bad-vsync-start +kms_invalid_mode@bad-hsync-end +kms_invalid_mode@bad-vsync-end +kms_invalid_mode@bad-htotal +kms_invalid_mode@bad-vtotal +kms_legacy_colorkey@basic +kms_legacy_colorkey@invalid-plane +kms_multipipe_modeset@basic-max-pipe-crc-check +kms_panel_fitting@legacy +kms_panel_fitting@atomic-fastset +kms_pipe_b_c_ivb@pipe-B-dpms-off-modeset-pipe-C +kms_pipe_b_c_ivb@pipe-B-double-modeset-then-modeset-pipe-C +kms_pipe_b_c_ivb@disable-pipe-B-enable-pipe-C +kms_pipe_b_c_ivb@from-pipe-C-to-B-with-3-lanes +kms_pipe_b_c_ivb@enable-pipe-C-while-B-has-3-lanes +kms_pipe_crc_basic@bad-source +kms_pipe_crc_basic@read-crc +kms_pipe_crc_basic@read-crc-frame-sequence +kms_pipe_crc_basic@nonblocking-crc +kms_pipe_crc_basic@nonblocking-crc-frame-sequence +kms_pipe_crc_basic@suspend-read-crc +kms_pipe_crc_basic@hang-read-crc +kms_pipe_crc_basic@disable-crc-after-crtc +kms_pipe_crc_basic@compare-crc-sanitycheck-xr24 +kms_pipe_crc_basic@compare-crc-sanitycheck-nv12 +kms_plane@pixel-format +kms_plane@pixel-format-source-clamping +kms_plane@plane-position-covered +kms_plane@plane-position-hole +kms_plane@plane-position-hole-dpms +kms_plane@plane-panning-top-left +kms_plane@plane-panning-bottom-right +kms_plane@plane-panning-bottom-right-suspend +kms_plane@invalid-pixel-format-settings +kms_plane_alpha_blend@alpha-basic +kms_plane_alpha_blend@alpha-7efc +kms_plane_alpha_blend@coverage-7efc +kms_plane_alpha_blend@coverage-vs-premult-vs-constant +kms_plane_alpha_blend@alpha-transparent-fb +kms_plane_alpha_blend@alpha-opaque-fb +kms_plane_alpha_blend@constant-alpha-min +kms_plane_alpha_blend@constant-alpha-mid +kms_plane_alpha_blend@constant-alpha-max +kms_plane_cursor@primary +kms_plane_cursor@overlay +kms_plane_cursor@viewport +kms_plane_lowres@tiling-none +kms_plane_lowres@tiling-x +kms_plane_lowres@tiling-y +kms_plane_lowres@tiling-yf +kms_plane_lowres@tiling-4 +kms_plane_multiple@tiling-none +kms_plane_multiple@tiling-x +kms_plane_multiple@tiling-y +kms_plane_multiple@tiling-yf +kms_plane_multiple@tiling-4 +kms_plane_scaling@plane-upscale-with-pixel-format-20x20 +kms_plane_scaling@plane-upscale-with-pixel-format-factor-0-25 +kms_plane_scaling@plane-downscale-with-pixel-format-factor-0-25 +kms_plane_scaling@plane-downscale-with-pixel-format-factor-0-5 +kms_plane_scaling@plane-downscale-with-pixel-format-factor-0-75 +kms_plane_scaling@plane-scaler-with-pixel-format-unity-scaling +kms_plane_scaling@plane-upscale-with-rotation-20x20 +kms_plane_scaling@plane-upscale-with-rotation-factor-0-25 +kms_plane_scaling@plane-downscale-with-rotation-factor-0-25 +kms_plane_scaling@plane-downscale-with-rotation-factor-0-5 +kms_plane_scaling@plane-downscale-with-rotation-factor-0-75 +kms_plane_scaling@plane-scaler-with-rotation-unity-scaling +kms_plane_scaling@plane-upscale-with-modifiers-20x20 +kms_plane_scaling@plane-upscale-with-modifiers-factor-0-25 +kms_plane_scaling@plane-downscale-with-modifiers-factor-0-25 +kms_plane_scaling@plane-downscale-with-modifiers-factor-0-5 +kms_plane_scaling@plane-downscale-with-modifiers-factor-0-75 +kms_plane_scaling@plane-scaler-with-modifiers-unity-scaling +kms_plane_scaling@plane-scaler-with-clipping-clamping-pixel-formats +kms_plane_scaling@plane-scaler-with-clipping-clamping-rotation +kms_plane_scaling@plane-scaler-with-clipping-clamping-modifiers +kms_plane_scaling@planes-upscale-20x20 +kms_plane_scaling@planes-upscale-factor-0-25 +kms_plane_scaling@planes-scaler-unity-scaling +kms_plane_scaling@planes-downscale-factor-0-25 +kms_plane_scaling@planes-downscale-factor-0-5 +kms_plane_scaling@planes-downscale-factor-0-75 +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25 +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-5 +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-75 +kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-25 +kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-5 +kms_plane_scaling@planes-upscale-factor-0-25-downscale-factor-0-75 +kms_plane_scaling@planes-unity-scaling-downscale-factor-0-25 +kms_plane_scaling@planes-unity-scaling-downscale-factor-0-5 +kms_plane_scaling@planes-unity-scaling-downscale-factor-0-75 +kms_plane_scaling@planes-downscale-factor-0-25-upscale-20x20 +kms_plane_scaling@planes-downscale-factor-0-25-upscale-factor-0-25 +kms_plane_scaling@planes-downscale-factor-0-25-unity-scaling +kms_plane_scaling@planes-downscale-factor-0-5-upscale-20x20 +kms_plane_scaling@planes-downscale-factor-0-5-upscale-factor-0-25 +kms_plane_scaling@planes-downscale-factor-0-5-unity-scaling +kms_plane_scaling@planes-downscale-factor-0-75-upscale-20x20 +kms_plane_scaling@planes-downscale-factor-0-75-upscale-factor-0-25 +kms_plane_scaling@planes-downscale-factor-0-75-unity-scaling +kms_plane_scaling@intel-max-src-size +kms_plane_scaling@invalid-num-scalers +kms_plane_scaling@invalid-parameters +kms_plane_scaling@2x-scaler-multi-pipe +kms_prime@basic-crc-hybrid +kms_prime@basic-modeset-hybrid +kms_prime@D3hot +kms_prime@basic-crc-vgem +kms_prop_blob@basic +kms_prop_blob@blob-prop-core +kms_prop_blob@blob-prop-validate +kms_prop_blob@blob-prop-lifetime +kms_prop_blob@blob-multiple +kms_prop_blob@invalid-get-prop-any +kms_prop_blob@invalid-get-prop +kms_prop_blob@invalid-set-prop-any +kms_prop_blob@invalid-set-prop +kms_properties@plane-properties-legacy +kms_properties@plane-properties-atomic +kms_properties@crtc-properties-legacy +kms_properties@crtc-properties-atomic +kms_properties@connector-properties-legacy +kms_properties@connector-properties-atomic +kms_properties@invalid-properties-legacy +kms_properties@invalid-properties-atomic +kms_properties@get_properties-sanity-atomic +kms_properties@get_properties-sanity-non-atomic +kms_psr@basic +kms_psr@no_drrs +kms_psr@primary_page_flip +kms_psr@primary_mmap_gtt +kms_psr@primary_mmap_cpu +kms_psr@primary_blt +kms_psr@primary_render +kms_psr@sprite_mmap_gtt +kms_psr@cursor_mmap_gtt +kms_psr@sprite_mmap_cpu +kms_psr@cursor_mmap_cpu +kms_psr@sprite_blt +kms_psr@cursor_blt +kms_psr@sprite_render +kms_psr@cursor_render +kms_psr@sprite_plane_move +kms_psr@cursor_plane_move +kms_psr@sprite_plane_onoff +kms_psr@cursor_plane_onoff +kms_psr@dpms +kms_psr@suspend +kms_psr@psr2_basic +kms_psr@psr2_no_drrs +kms_psr@psr2_primary_page_flip +kms_psr@psr2_primary_mmap_gtt +kms_psr@psr2_primary_mmap_cpu +kms_psr@psr2_primary_blt +kms_psr@psr2_primary_render +kms_psr@psr2_sprite_mmap_gtt +kms_psr@psr2_cursor_mmap_gtt +kms_psr@psr2_sprite_mmap_cpu +kms_psr@psr2_cursor_mmap_cpu +kms_psr@psr2_sprite_blt +kms_psr@psr2_cursor_blt +kms_psr@psr2_sprite_render +kms_psr@psr2_cursor_render +kms_psr@psr2_sprite_plane_move +kms_psr@psr2_cursor_plane_move +kms_psr@psr2_sprite_plane_onoff +kms_psr@psr2_cursor_plane_onoff +kms_psr@psr2_dpms +kms_psr@psr2_suspend +kms_psr2_sf@primary-plane-update-sf-dmg-area +kms_psr2_sf@primary-plane-update-sf-dmg-area-big-fb +kms_psr2_sf@overlay-plane-update-sf-dmg-area +kms_psr2_sf@cursor-plane-update-sf +kms_psr2_sf@cursor-plane-move-continuous-sf +kms_psr2_sf@cursor-plane-move-continuous-exceed-sf +kms_psr2_sf@cursor-plane-move-continuous-exceed-fully-sf +kms_psr2_sf@plane-move-sf-dmg-area +kms_psr2_sf@overlay-plane-move-continuous-sf +kms_psr2_sf@overlay-plane-move-continuous-exceed-sf +kms_psr2_sf@overlay-plane-move-continuous-exceed-fully-sf +kms_psr2_sf@overlay-primary-update-sf-dmg-area +kms_psr2_sf@overlay-plane-update-continuous-sf +kms_psr2_su@page_flip-XRGB8888 +kms_psr2_su@page_flip-NV12 +kms_psr2_su@page_flip-P010 +kms_psr2_su@frontbuffer-XRGB8888 +kms_pwrite_crc +kms_rmfb@rmfb-ioctl +kms_rmfb@close-fd +kms_rotation_crc@primary-rotation-90 +kms_rotation_crc@primary-rotation-180 +kms_rotation_crc@primary-rotation-270 +kms_rotation_crc@sprite-rotation-90 +kms_rotation_crc@sprite-rotation-180 +kms_rotation_crc@sprite-rotation-270 +kms_rotation_crc@cursor-rotation-180 +kms_rotation_crc@sprite-rotation-90-pos-100-0 +kms_rotation_crc@bad-pixel-format +kms_rotation_crc@bad-tiling +kms_rotation_crc@primary-x-tiled-reflect-x-0 +kms_rotation_crc@primary-x-tiled-reflect-x-180 +kms_rotation_crc@primary-y-tiled-reflect-x-0 +kms_rotation_crc@primary-y-tiled-reflect-x-90 +kms_rotation_crc@primary-y-tiled-reflect-x-180 +kms_rotation_crc@primary-y-tiled-reflect-x-270 +kms_rotation_crc@primary-yf-tiled-reflect-x-0 +kms_rotation_crc@primary-yf-tiled-reflect-x-90 +kms_rotation_crc@primary-yf-tiled-reflect-x-180 +kms_rotation_crc@primary-yf-tiled-reflect-x-270 +kms_rotation_crc@primary-4-tiled-reflect-x-0 +kms_rotation_crc@primary-4-tiled-reflect-x-180 +kms_rotation_crc@multiplane-rotation +kms_rotation_crc@multiplane-rotation-cropping-top +kms_rotation_crc@multiplane-rotation-cropping-bottom +kms_rotation_crc@exhaust-fences +kms_scaling_modes@scaling-mode-full +kms_scaling_modes@scaling-mode-center +kms_scaling_modes@scaling-mode-full-aspect +kms_scaling_modes@scaling-mode-none +kms_selftest@drm_cmdline +kms_selftest@drm_damage +kms_selftest@drm_dp_mst +kms_selftest@drm_format_helper +kms_selftest@drm_format +kms_selftest@framebuffer +kms_selftest@drm_plane +kms_setmode@basic +kms_setmode@basic-clone-single-crtc +kms_setmode@invalid-clone-single-crtc +kms_setmode@invalid-clone-exclusive-crtc +kms_setmode@clone-exclusive-crtc +kms_setmode@invalid-clone-single-crtc-stealing +kms_sysfs_edid_timing +kms_tv_load_detect@load-detect +kms_universal_plane@universal-plane-pipe-A-functional +kms_universal_plane@universal-plane-pipe-A-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-A +kms_universal_plane@cursor-fb-leak-pipe-A +kms_universal_plane@universal-plane-pageflip-windowed-pipe-A +kms_universal_plane@universal-plane-pipe-B-functional +kms_universal_plane@universal-plane-pipe-B-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-B +kms_universal_plane@cursor-fb-leak-pipe-B +kms_universal_plane@universal-plane-pageflip-windowed-pipe-B +kms_universal_plane@universal-plane-pipe-C-functional +kms_universal_plane@universal-plane-pipe-C-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-C +kms_universal_plane@cursor-fb-leak-pipe-C +kms_universal_plane@universal-plane-pageflip-windowed-pipe-C +kms_universal_plane@universal-plane-pipe-D-functional +kms_universal_plane@universal-plane-pipe-D-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-D +kms_universal_plane@cursor-fb-leak-pipe-D +kms_universal_plane@universal-plane-pageflip-windowed-pipe-D +kms_universal_plane@universal-plane-pipe-E-functional +kms_universal_plane@universal-plane-pipe-E-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-E +kms_universal_plane@cursor-fb-leak-pipe-E +kms_universal_plane@universal-plane-pageflip-windowed-pipe-E +kms_universal_plane@universal-plane-pipe-F-functional +kms_universal_plane@universal-plane-pipe-F-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-F +kms_universal_plane@cursor-fb-leak-pipe-F +kms_universal_plane@universal-plane-pageflip-windowed-pipe-F +kms_universal_plane@universal-plane-pipe-G-functional +kms_universal_plane@universal-plane-pipe-G-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-G +kms_universal_plane@cursor-fb-leak-pipe-G +kms_universal_plane@universal-plane-pageflip-windowed-pipe-G +kms_universal_plane@universal-plane-pipe-H-functional +kms_universal_plane@universal-plane-pipe-H-sanity +kms_universal_plane@disable-primary-vs-flip-pipe-H +kms_universal_plane@cursor-fb-leak-pipe-H +kms_universal_plane@universal-plane-pageflip-windowed-pipe-H +kms_vblank@invalid +kms_vblank@crtc-id +kms_vblank@pipe-A-accuracy-idle +kms_vblank@pipe-A-query-idle +kms_vblank@pipe-A-query-idle-hang +kms_vblank@pipe-A-query-forked +kms_vblank@pipe-A-query-forked-hang +kms_vblank@pipe-A-query-busy +kms_vblank@pipe-A-query-busy-hang +kms_vblank@pipe-A-query-forked-busy +kms_vblank@pipe-A-query-forked-busy-hang +kms_vblank@pipe-A-wait-idle +kms_vblank@pipe-A-wait-idle-hang +kms_vblank@pipe-A-wait-forked +kms_vblank@pipe-A-wait-forked-hang +kms_vblank@pipe-A-wait-busy +kms_vblank@pipe-A-wait-busy-hang +kms_vblank@pipe-A-wait-forked-busy +kms_vblank@pipe-A-wait-forked-busy-hang +kms_vblank@pipe-A-ts-continuation-idle +kms_vblank@pipe-A-ts-continuation-idle-hang +kms_vblank@pipe-A-ts-continuation-dpms-rpm +kms_vblank@pipe-A-ts-continuation-dpms-suspend +kms_vblank@pipe-A-ts-continuation-suspend +kms_vblank@pipe-A-ts-continuation-modeset +kms_vblank@pipe-A-ts-continuation-modeset-hang +kms_vblank@pipe-A-ts-continuation-modeset-rpm +kms_vblank@pipe-B-accuracy-idle +kms_vblank@pipe-B-query-idle +kms_vblank@pipe-B-query-idle-hang +kms_vblank@pipe-B-query-forked +kms_vblank@pipe-B-query-forked-hang +kms_vblank@pipe-B-query-busy +kms_vblank@pipe-B-query-busy-hang +kms_vblank@pipe-B-query-forked-busy +kms_vblank@pipe-B-query-forked-busy-hang +kms_vblank@pipe-B-wait-idle +kms_vblank@pipe-B-wait-idle-hang +kms_vblank@pipe-B-wait-forked +kms_vblank@pipe-B-wait-forked-hang +kms_vblank@pipe-B-wait-busy +kms_vblank@pipe-B-wait-busy-hang +kms_vblank@pipe-B-wait-forked-busy +kms_vblank@pipe-B-wait-forked-busy-hang +kms_vblank@pipe-B-ts-continuation-idle +kms_vblank@pipe-B-ts-continuation-idle-hang +kms_vblank@pipe-B-ts-continuation-dpms-rpm +kms_vblank@pipe-B-ts-continuation-dpms-suspend +kms_vblank@pipe-B-ts-continuation-suspend +kms_vblank@pipe-B-ts-continuation-modeset +kms_vblank@pipe-B-ts-continuation-modeset-hang +kms_vblank@pipe-B-ts-continuation-modeset-rpm +kms_vblank@pipe-C-accuracy-idle +kms_vblank@pipe-C-query-idle +kms_vblank@pipe-C-query-idle-hang +kms_vblank@pipe-C-query-forked +kms_vblank@pipe-C-query-forked-hang +kms_vblank@pipe-C-query-busy +kms_vblank@pipe-C-query-busy-hang +kms_vblank@pipe-C-query-forked-busy +kms_vblank@pipe-C-query-forked-busy-hang +kms_vblank@pipe-C-wait-idle +kms_vblank@pipe-C-wait-idle-hang +kms_vblank@pipe-C-wait-forked +kms_vblank@pipe-C-wait-forked-hang +kms_vblank@pipe-C-wait-busy +kms_vblank@pipe-C-wait-busy-hang +kms_vblank@pipe-C-wait-forked-busy +kms_vblank@pipe-C-wait-forked-busy-hang +kms_vblank@pipe-C-ts-continuation-idle +kms_vblank@pipe-C-ts-continuation-idle-hang +kms_vblank@pipe-C-ts-continuation-dpms-rpm +kms_vblank@pipe-C-ts-continuation-dpms-suspend +kms_vblank@pipe-C-ts-continuation-suspend +kms_vblank@pipe-C-ts-continuation-modeset +kms_vblank@pipe-C-ts-continuation-modeset-hang +kms_vblank@pipe-C-ts-continuation-modeset-rpm +kms_vblank@pipe-D-accuracy-idle +kms_vblank@pipe-D-query-idle +kms_vblank@pipe-D-query-idle-hang +kms_vblank@pipe-D-query-forked +kms_vblank@pipe-D-query-forked-hang +kms_vblank@pipe-D-query-busy +kms_vblank@pipe-D-query-busy-hang +kms_vblank@pipe-D-query-forked-busy +kms_vblank@pipe-D-query-forked-busy-hang +kms_vblank@pipe-D-wait-idle +kms_vblank@pipe-D-wait-idle-hang +kms_vblank@pipe-D-wait-forked +kms_vblank@pipe-D-wait-forked-hang +kms_vblank@pipe-D-wait-busy +kms_vblank@pipe-D-wait-busy-hang +kms_vblank@pipe-D-wait-forked-busy +kms_vblank@pipe-D-wait-forked-busy-hang +kms_vblank@pipe-D-ts-continuation-idle +kms_vblank@pipe-D-ts-continuation-idle-hang +kms_vblank@pipe-D-ts-continuation-dpms-rpm +kms_vblank@pipe-D-ts-continuation-dpms-suspend +kms_vblank@pipe-D-ts-continuation-suspend +kms_vblank@pipe-D-ts-continuation-modeset +kms_vblank@pipe-D-ts-continuation-modeset-hang +kms_vblank@pipe-D-ts-continuation-modeset-rpm +kms_vblank@pipe-E-accuracy-idle +kms_vblank@pipe-E-query-idle +kms_vblank@pipe-E-query-idle-hang +kms_vblank@pipe-E-query-forked +kms_vblank@pipe-E-query-forked-hang +kms_vblank@pipe-E-query-busy +kms_vblank@pipe-E-query-busy-hang +kms_vblank@pipe-E-query-forked-busy +kms_vblank@pipe-E-query-forked-busy-hang +kms_vblank@pipe-E-wait-idle +kms_vblank@pipe-E-wait-idle-hang +kms_vblank@pipe-E-wait-forked +kms_vblank@pipe-E-wait-forked-hang +kms_vblank@pipe-E-wait-busy +kms_vblank@pipe-E-wait-busy-hang +kms_vblank@pipe-E-wait-forked-busy +kms_vblank@pipe-E-wait-forked-busy-hang +kms_vblank@pipe-E-ts-continuation-idle +kms_vblank@pipe-E-ts-continuation-idle-hang +kms_vblank@pipe-E-ts-continuation-dpms-rpm +kms_vblank@pipe-E-ts-continuation-dpms-suspend +kms_vblank@pipe-E-ts-continuation-suspend +kms_vblank@pipe-E-ts-continuation-modeset +kms_vblank@pipe-E-ts-continuation-modeset-hang +kms_vblank@pipe-E-ts-continuation-modeset-rpm +kms_vblank@pipe-F-accuracy-idle +kms_vblank@pipe-F-query-idle +kms_vblank@pipe-F-query-idle-hang +kms_vblank@pipe-F-query-forked +kms_vblank@pipe-F-query-forked-hang +kms_vblank@pipe-F-query-busy +kms_vblank@pipe-F-query-busy-hang +kms_vblank@pipe-F-query-forked-busy +kms_vblank@pipe-F-query-forked-busy-hang +kms_vblank@pipe-F-wait-idle +kms_vblank@pipe-F-wait-idle-hang +kms_vblank@pipe-F-wait-forked +kms_vblank@pipe-F-wait-forked-hang +kms_vblank@pipe-F-wait-busy +kms_vblank@pipe-F-wait-busy-hang +kms_vblank@pipe-F-wait-forked-busy +kms_vblank@pipe-F-wait-forked-busy-hang +kms_vblank@pipe-F-ts-continuation-idle +kms_vblank@pipe-F-ts-continuation-idle-hang +kms_vblank@pipe-F-ts-continuation-dpms-rpm +kms_vblank@pipe-F-ts-continuation-dpms-suspend +kms_vblank@pipe-F-ts-continuation-suspend +kms_vblank@pipe-F-ts-continuation-modeset +kms_vblank@pipe-F-ts-continuation-modeset-hang +kms_vblank@pipe-F-ts-continuation-modeset-rpm +kms_vblank@pipe-G-accuracy-idle +kms_vblank@pipe-G-query-idle +kms_vblank@pipe-G-query-idle-hang +kms_vblank@pipe-G-query-forked +kms_vblank@pipe-G-query-forked-hang +kms_vblank@pipe-G-query-busy +kms_vblank@pipe-G-query-busy-hang +kms_vblank@pipe-G-query-forked-busy +kms_vblank@pipe-G-query-forked-busy-hang +kms_vblank@pipe-G-wait-idle +kms_vblank@pipe-G-wait-idle-hang +kms_vblank@pipe-G-wait-forked +kms_vblank@pipe-G-wait-forked-hang +kms_vblank@pipe-G-wait-busy +kms_vblank@pipe-G-wait-busy-hang +kms_vblank@pipe-G-wait-forked-busy +kms_vblank@pipe-G-wait-forked-busy-hang +kms_vblank@pipe-G-ts-continuation-idle +kms_vblank@pipe-G-ts-continuation-idle-hang +kms_vblank@pipe-G-ts-continuation-dpms-rpm +kms_vblank@pipe-G-ts-continuation-dpms-suspend +kms_vblank@pipe-G-ts-continuation-suspend +kms_vblank@pipe-G-ts-continuation-modeset +kms_vblank@pipe-G-ts-continuation-modeset-hang +kms_vblank@pipe-G-ts-continuation-modeset-rpm +kms_vblank@pipe-H-accuracy-idle +kms_vblank@pipe-H-query-idle +kms_vblank@pipe-H-query-idle-hang +kms_vblank@pipe-H-query-forked +kms_vblank@pipe-H-query-forked-hang +kms_vblank@pipe-H-query-busy +kms_vblank@pipe-H-query-busy-hang +kms_vblank@pipe-H-query-forked-busy +kms_vblank@pipe-H-query-forked-busy-hang +kms_vblank@pipe-H-wait-idle +kms_vblank@pipe-H-wait-idle-hang +kms_vblank@pipe-H-wait-forked +kms_vblank@pipe-H-wait-forked-hang +kms_vblank@pipe-H-wait-busy +kms_vblank@pipe-H-wait-busy-hang +kms_vblank@pipe-H-wait-forked-busy +kms_vblank@pipe-H-wait-forked-busy-hang +kms_vblank@pipe-H-ts-continuation-idle +kms_vblank@pipe-H-ts-continuation-idle-hang +kms_vblank@pipe-H-ts-continuation-dpms-rpm +kms_vblank@pipe-H-ts-continuation-dpms-suspend +kms_vblank@pipe-H-ts-continuation-suspend +kms_vblank@pipe-H-ts-continuation-modeset +kms_vblank@pipe-H-ts-continuation-modeset-hang +kms_vblank@pipe-H-ts-continuation-modeset-rpm +kms_vrr@flip-basic +kms_vrr@flip-dpms +kms_vrr@flip-suspend +kms_vrr@flipline +kms_vrr@negative-basic +kms_writeback@writeback-pixel-formats +kms_writeback@writeback-invalid-parameters +kms_writeback@writeback-fb-id +kms_writeback@writeback-check-output +prime_mmap_kms@buffer-sharing diff --git a/drivers/gpu/drm/ci/x86_64.config b/drivers/gpu/drm/ci/x86_64.config new file mode 100644 index 000000000000..1cbd49a5b23a --- /dev/null +++ b/drivers/gpu/drm/ci/x86_64.config @@ -0,0 +1,111 @@ +CONFIG_LOCALVERSION_AUTO=y +CONFIG_DEBUG_KERNEL=y + +CONFIG_CRYPTO_ZSTD=y +CONFIG_ZRAM_MEMORY_TRACKING=y +CONFIG_ZRAM_WRITEBACK=y +CONFIG_ZRAM=y +CONFIG_ZSMALLOC_STAT=y + +CONFIG_PWM=y +CONFIG_PM_DEVFREQ=y +CONFIG_OF=y +CONFIG_CROS_EC=y + +# abootimg with a 'dummy' rootfs fails with root=/dev/nfs +CONFIG_BLK_DEV_INITRD=n + +CONFIG_DEVFREQ_GOV_PERFORMANCE=y +CONFIG_DEVFREQ_GOV_POWERSAVE=y +CONFIG_DEVFREQ_GOV_USERSPACE=y +CONFIG_DEVFREQ_GOV_PASSIVE=y + +CONFIG_DRM=y +CONFIG_DRM_PANEL_SIMPLE=y +CONFIG_PWM_CROS_EC=y +CONFIG_BACKLIGHT_PWM=y + +# Strip out some stuff we don't need for graphics testing, to reduce +# the build. +CONFIG_CAN=n +CONFIG_WIRELESS=n +CONFIG_RFKILL=n +CONFIG_WLAN=n + +CONFIG_REGULATOR_FAN53555=y +CONFIG_REGULATOR=y + +CONFIG_REGULATOR_VCTRL=y + +CONFIG_KASAN=n +CONFIG_KASAN_INLINE=n +CONFIG_STACKTRACE=n + +CONFIG_TMPFS=y + +CONFIG_PROVE_LOCKING=n +CONFIG_DEBUG_LOCKDEP=n +CONFIG_SOFTLOCKUP_DETECTOR=y +CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC=y + +CONFIG_DETECT_HUNG_TASK=y + +CONFIG_USB_USBNET=y +CONFIG_NETDEVICES=y +CONFIG_USB_NET_DRIVERS=y +CONFIG_USB_RTL8152=y +CONFIG_USB_NET_AX8817X=y +CONFIG_USB_NET_SMSC95XX=y +CONFIG_USB_GADGET=y +CONFIG_USB_ETH=y + +CONFIG_FW_LOADER_COMPRESS=y + +# options for AMD devices +CONFIG_X86_AMD_PLATFORM_DEVICE=y +CONFIG_ACPI_VIDEO=y +CONFIG_X86_AMD_FREQ_SENSITIVITY=y +CONFIG_PINCTRL=y +CONFIG_PINCTRL_AMD=y +CONFIG_DRM_AMDGPU=m +CONFIG_DRM_AMDGPU_SI=y +CONFIG_DRM_AMDGPU_USERPTR=y +CONFIG_DRM_AMD_ACP=n +CONFIG_ACPI_WMI=y +CONFIG_MXM_WMI=y +CONFIG_PARPORT=y +CONFIG_PARPORT_PC=y +CONFIG_PARPORT_SERIAL=y +CONFIG_SERIAL_8250_DW=y +CONFIG_CHROME_PLATFORMS=y +CONFIG_KVM_AMD=m + +#options for Intel devices +CONFIG_MFD_INTEL_LPSS_PCI=y +CONFIG_KVM_INTEL=m + +#options for KVM guests +CONFIG_FUSE_FS=y +CONFIG_HYPERVISOR_GUEST=y +CONFIG_KVM=y +CONFIG_KVM_GUEST=y +CONFIG_VIRT_DRIVERS=y +CONFIG_VIRTIO_FS=y +CONFIG_DRM_VIRTIO_GPU=y +CONFIG_SERIAL_8250_CONSOLE=y +CONFIG_VIRTIO_NET=y +CONFIG_VIRTIO_CONSOLE=y +CONFIG_PARAVIRT=y +CONFIG_VIRTIO_BLK=y +CONFIG_VIRTUALIZATION=y +CONFIG_VIRTIO=y +CONFIG_VIRTIO_PCI=y +CONFIG_VIRTIO_MMIO=y +CONFIG_VIRTIO_MMIO_CMDLINE_DEVICES=y +CONFIG_CRYPTO_DEV_VIRTIO=y +CONFIG_HW_RANDOM_VIRTIO=y +CONFIG_BLK_MQ_VIRTIO=y +CONFIG_TUN=y +CONFIG_VSOCKETS=y +CONFIG_VIRTIO_VSOCKETS=y +CONFIG_VHOST_VSOCK=m diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt new file mode 100644 index 000000000000..bd9392536e7c --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-fails.txt @@ -0,0 +1,19 @@ +kms_addfb_basic@bad-pitch-65536,Fail +kms_addfb_basic@bo-too-small,Fail +kms_async_flips@invalid-async-flip,Fail +kms_atomic@plane-immutable-zpos,Fail +kms_atomic_transition@plane-toggle-modeset-transition,Fail +kms_bw@linear-tiling-1-displays-2560x1440p,Fail +kms_bw@linear-tiling-1-displays-3840x2160p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-1920x1080p,Fail +kms_color@degamma,Fail +kms_cursor_crc@cursor-size-change,Fail +kms_cursor_crc@pipe-A-cursor-size-change,Fail +kms_cursor_crc@pipe-B-cursor-size-change,Fail +kms_cursor_legacy@forked-move,Fail +kms_hdr@bpc-switch,Fail +kms_hdr@bpc-switch-dpms,Fail +kms_plane_multiple@atomic-pipe-A-tiling-none,Fail +kms_rmfb@close-fd,Fail +kms_rotation_crc@primary-rotation-180,Fail diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt new file mode 100644 index 000000000000..f8defa0f9e67 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-flakes.txt @@ -0,0 +1,21 @@ +kms_addfb_basic@too-high +kms_async_flips@alternate-sync-async-flip +kms_async_flips@async-flip-with-page-flip-events +kms_async_flips@crc +kms_async_flips@test-cursor +kms_async_flips@test-time-stamp +kms_atomic_transition@plane-all-modeset-transition-internal-panels +kms_atomic_transition@plane-all-transition +kms_atomic_transition@plane-use-after-nonblocking-unbind +kms_bw@linear-tiling-1-displays-1920x1080p +kms_bw@linear-tiling-2-displays-1920x1080p +kms_bw@linear-tiling-2-displays-2560x1440p +kms_bw@linear-tiling-3-displays-2560x1440p +kms_bw@linear-tiling-3-displays-3840x2160p +kms_cursor_crc@pipe-A-cursor-alpha-opaque +kms_cursor_crc@pipe-B-cursor-alpha-opaque +kms_plane@pixel-format +kms_plane_multiple@atomic-pipe-B-tiling-none +kms_plane_scaling@downscale-with-rotation-factor-0-5 +kms_universal_plane@disable-primary-vs-flip-pipe-A +kms_universal_plane@disable-primary-vs-flip-pipe-B diff --git a/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt new file mode 100644 index 000000000000..e2c538a0f954 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/amdgpu-stoney-skips.txt @@ -0,0 +1,2 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.*
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt b/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt new file mode 100644 index 000000000000..5f513c638beb --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-amly-fails.txt @@ -0,0 +1,17 @@ +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_plane_alpha_blend@alpha-basic,Fail +kms_plane_alpha_blend@alpha-opaque-fb,Fail +kms_plane_alpha_blend@alpha-transparent-fb,Fail +kms_plane_alpha_blend@constant-alpha-max,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt new file mode 100644 index 000000000000..d5000515a315 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-amly-flakes.txt @@ -0,0 +1,32 @@ +kms_bw@linear-tiling-2-displays-1920x1080p +kms_bw@linear-tiling-2-displays-2560x1440p +kms_bw@linear-tiling-2-displays-3840x2160p +kms_bw@linear-tiling-3-displays-1920x1080p +kms_bw@linear-tiling-3-displays-2560x1440p +kms_bw@linear-tiling-3-displays-3840x2160p +kms_bw@linear-tiling-4-displays-1920x1080p +kms_bw@linear-tiling-4-displays-2560x1440p +kms_bw@linear-tiling-4-displays-3840x2160p +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling +kms_plane_alpha_blend@pipe-A-alpha-basic +kms_plane_alpha_blend@pipe-A-alpha-opaque-fb +kms_plane_alpha_blend@pipe-A-alpha-transparent-fb +kms_plane_alpha_blend@pipe-A-constant-alpha-max +kms_plane_alpha_blend@pipe-B-alpha-basic +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb +kms_plane_alpha_blend@pipe-B-alpha-transparent-fb +kms_plane_alpha_blend@pipe-B-constant-alpha-max +kms_plane_alpha_blend@pipe-C-alpha-basic +kms_plane_alpha_blend@pipe-C-alpha-opaque-fb +kms_plane_alpha_blend@pipe-C-alpha-transparent-fb +kms_plane_alpha_blend@pipe-C-constant-alpha-max +kms_sysfs_edid_timing diff --git a/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt b/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt new file mode 100644 index 000000000000..fe55540a3f9a --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-amly-skips.txt @@ -0,0 +1,4 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt new file mode 100644 index 000000000000..46397ce38d5a --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-apl-fails.txt @@ -0,0 +1,58 @@ +kms_3d,Timeout +kms_bw@linear-tiling-2-displays-1920x1080p,Fail +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-1920x1080p,Fail +kms_bw@linear-tiling-3-displays-2560x1440p,Fail +kms_bw@linear-tiling-3-displays-3840x2160p,Fail +kms_bw@linear-tiling-4-displays-1920x1080p,Fail +kms_bw@linear-tiling-4-displays-2560x1440p,Fail +kms_bw@linear-tiling-4-displays-3840x2160p,Fail +kms_color@ctm-0-25,Fail +kms_color@ctm-0-50,Fail +kms_color@ctm-0-75,Fail +kms_color@ctm-max,Fail +kms_color@ctm-negative,Fail +kms_color@ctm-red-to-blue,Fail +kms_color@ctm-signed,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling,Fail +kms_hdmi_inject@inject-4k,Timeout +kms_plane@plane-position-hole,Timeout +kms_plane_alpha_blend@alpha-basic,Fail +kms_plane_alpha_blend@alpha-opaque-fb,Fail +kms_plane_alpha_blend@alpha-transparent-fb,Fail +kms_plane_alpha_blend@constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-A-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-A-alpha-transparent-fb,Fail +kms_plane_alpha_blend@pipe-A-constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-B-alpha-transparent-fb,Fail +kms_plane_alpha_blend@pipe-B-constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-C-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-C-alpha-transparent-fb,Fail +kms_plane_alpha_blend@pipe-C-constant-alpha-max,Fail +kms_plane_multiple@tiling-y,Timeout +kms_pwrite_crc,Timeout +kms_sysfs_edid_timing,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-apl-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-apl-flakes.txt new file mode 100644 index 000000000000..331c5841bb41 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-apl-flakes.txt @@ -0,0 +1 @@ +kms_frontbuffer_tracking@fbc-tiling-linear diff --git a/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt new file mode 100644 index 000000000000..3430b215c06e --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-apl-skips.txt @@ -0,0 +1,6 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters +# This is cascading issues +kms_3d
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt new file mode 100644 index 000000000000..6139b410e767 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-cml-fails.txt @@ -0,0 +1,18 @@ +kms_color@ctm-0-25,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_plane_alpha_blend@alpha-basic,Fail +kms_plane_alpha_blend@alpha-opaque-fb,Fail +kms_plane_alpha_blend@alpha-transparent-fb,Fail +kms_plane_alpha_blend@constant-alpha-max,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt new file mode 100644 index 000000000000..0514a7b3fdb0 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-cml-flakes.txt @@ -0,0 +1,38 @@ +kms_bw@linear-tiling-2-displays-1920x1080p +kms_bw@linear-tiling-2-displays-2560x1440p +kms_bw@linear-tiling-2-displays-3840x2160p +kms_bw@linear-tiling-3-displays-1920x1080p +kms_bw@linear-tiling-3-displays-2560x1440p +kms_bw@linear-tiling-3-displays-3840x2160p +kms_bw@linear-tiling-4-displays-1920x1080p +kms_bw@linear-tiling-4-displays-2560x1440p +kms_bw@linear-tiling-4-displays-3840x2160p +kms_draw_crc@draw-method-xrgb8888-render-xtiled +kms_flip@flip-vs-suspend +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling +kms_hdr@bpc-switch-suspend +kms_plane_alpha_blend@constant-alpha-min +kms_plane_alpha_blend@pipe-A-alpha-basic +kms_plane_alpha_blend@pipe-A-alpha-opaque-fb +kms_plane_alpha_blend@pipe-A-alpha-transparent-fb +kms_plane_alpha_blend@pipe-A-constant-alpha-max +kms_plane_alpha_blend@pipe-B-alpha-basic +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb +kms_plane_alpha_blend@pipe-B-alpha-transparent-fb +kms_plane_alpha_blend@pipe-B-constant-alpha-max +kms_plane_alpha_blend@pipe-C-alpha-basic +kms_plane_alpha_blend@pipe-C-alpha-opaque-fb +kms_plane_alpha_blend@pipe-C-alpha-transparent-fb +kms_plane_alpha_blend@pipe-C-constant-alpha-max +kms_psr2_su@page_flip-NV12 +kms_psr2_su@page_flip-P010 +kms_setmode@basic diff --git a/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt b/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt new file mode 100644 index 000000000000..6d3d7ddc377f --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-cml-skips.txt @@ -0,0 +1,2 @@ +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt b/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt new file mode 100644 index 000000000000..5bd432e78129 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-glk-fails.txt @@ -0,0 +1,19 @@ +kms_fbcon_fbt@fbc,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_frontbuffer_tracking@fbcdrrs-tiling-linear,Fail +kms_plane_alpha_blend@alpha-basic,Fail +kms_plane_alpha_blend@alpha-opaque-fb,Fail +kms_plane_alpha_blend@alpha-transparent-fb,Fail +kms_plane_alpha_blend@constant-alpha-max,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt new file mode 100644 index 000000000000..fc41d13a2d56 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-glk-flakes.txt @@ -0,0 +1,41 @@ +kms_bw@linear-tiling-1-displays-3840x2160p +kms_bw@linear-tiling-2-displays-1920x1080p +kms_bw@linear-tiling-2-displays-2560x1440p +kms_bw@linear-tiling-2-displays-3840x2160p +kms_bw@linear-tiling-3-displays-1920x1080p +kms_bw@linear-tiling-3-displays-2560x1440p +kms_bw@linear-tiling-3-displays-3840x2160p +kms_bw@linear-tiling-4-displays-1920x1080p +kms_bw@linear-tiling-4-displays-2560x1440p +kms_bw@linear-tiling-4-displays-3840x2160p +kms_flip@blocking-wf_vblank +kms_flip@wf_vblank-ts-check +kms_flip@wf_vblank-ts-check-interruptible +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling +kms_frontbuffer_tracking@fbc-tiling-linear +kms_plane_alpha_blend@pipe-A-alpha-basic +kms_plane_alpha_blend@pipe-A-alpha-opaque-fb +kms_plane_alpha_blend@pipe-A-alpha-transparent-fb +kms_plane_alpha_blend@pipe-A-constant-alpha-max +kms_plane_alpha_blend@pipe-B-alpha-basic +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb +kms_plane_alpha_blend@pipe-B-alpha-transparent-fb +kms_plane_alpha_blend@pipe-B-constant-alpha-max +kms_plane_alpha_blend@pipe-C-alpha-basic +kms_plane_alpha_blend@pipe-C-alpha-opaque-fb +kms_plane_alpha_blend@pipe-C-alpha-transparent-fb +kms_plane_alpha_blend@pipe-C-constant-alpha-max +kms_prop_blob@invalid-set-prop-any +kms_rotation_crc@multiplane-rotation +kms_rotation_crc@multiplane-rotation-cropping-bottom +kms_rotation_crc@multiplane-rotation-cropping-top +kms_setmode@basic diff --git a/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt b/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt new file mode 100644 index 000000000000..4c7d00ce14bc --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-glk-skips.txt @@ -0,0 +1,5 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* + +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt new file mode 100644 index 000000000000..56ec021a7679 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-kbl-fails.txt @@ -0,0 +1,25 @@ +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-4-displays-2560x1440p,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling,Fail +kms_plane_alpha_blend@alpha-basic,Fail +kms_plane_alpha_blend@alpha-opaque-fb,Fail +kms_plane_alpha_blend@alpha-transparent-fb,Fail +kms_plane_alpha_blend@constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-A-constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-C-constant-alpha-max,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-kbl-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-kbl-flakes.txt new file mode 100644 index 000000000000..f3ba1c4c5d46 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-kbl-flakes.txt @@ -0,0 +1,26 @@ +kms_async_flips@crc +kms_bw@linear-tiling-2-displays-1920x1080p +kms_bw@linear-tiling-2-displays-3840x2160p +kms_bw@linear-tiling-3-displays-1920x1080p +kms_bw@linear-tiling-3-displays-2560x1440p +kms_bw@linear-tiling-3-displays-3840x2160p +kms_bw@linear-tiling-4-displays-1920x1080p +kms_bw@linear-tiling-4-displays-3840x2160p +kms_color@ctm-0-25 +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling +kms_plane_alpha_blend@pipe-A-alpha-basic +kms_plane_alpha_blend@pipe-A-alpha-opaque-fb +kms_plane_alpha_blend@pipe-A-alpha-transparent-fb +kms_plane_alpha_blend@pipe-B-alpha-basic +kms_plane_alpha_blend@pipe-B-alpha-transparent-fb +kms_plane_alpha_blend@pipe-B-constant-alpha-max +kms_plane_alpha_blend@pipe-C-alpha-basic +kms_plane_alpha_blend@pipe-C-alpha-opaque-fb +kms_plane_alpha_blend@pipe-C-alpha-transparent-fb +kms_sysfs_edid_timing diff --git a/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt new file mode 100644 index 000000000000..4c7d00ce14bc --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-kbl-skips.txt @@ -0,0 +1,5 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* + +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt new file mode 100644 index 000000000000..a6da5544e198 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-tgl-fails.txt @@ -0,0 +1,37 @@ +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-1920x1080p,Fail +kms_bw@linear-tiling-3-displays-2560x1440p,Fail +kms_bw@linear-tiling-3-displays-3840x2160p,Fail +kms_bw@linear-tiling-4-displays-1920x1080p,Fail +kms_bw@linear-tiling-4-displays-2560x1440p,Fail +kms_bw@linear-tiling-4-displays-3840x2160p,Fail +kms_bw@linear-tiling-5-displays-1920x1080p,Fail +kms_bw@linear-tiling-5-displays-2560x1440p,Fail +kms_bw@linear-tiling-5-displays-3840x2160p,Fail +kms_color@ctm-0-25,Fail +kms_flip@flip-vs-panning-vs-hang,Timeout +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling,Fail +kms_rotation_crc@bad-pixel-format,Fail +kms_rotation_crc@multiplane-rotation,Fail +kms_rotation_crc@multiplane-rotation-cropping-bottom,Fail +kms_rotation_crc@multiplane-rotation-cropping-top,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-flakes.txt new file mode 100644 index 000000000000..1cd910ee06df --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-tgl-flakes.txt @@ -0,0 +1,5 @@ +kms_draw_crc@.* +kms_flip@blocking-absolute-wf_vblank +kms_flip@bo-too-big-interruptible +kms_flip@busy-flip +kms_flip@flip-vs-rmfb-interruptible diff --git a/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt new file mode 100644 index 000000000000..1d0621750b14 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-tgl-skips.txt @@ -0,0 +1,11 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* + +# GPU hangs, then the whole machine +gem_eio.* + +# Whole machine hangs +kms_flip@absolute-wf_vblank@a-edp1 + +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt b/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt new file mode 100644 index 000000000000..967327ddc1ac --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-whl-fails.txt @@ -0,0 +1,48 @@ +kms_bw@linear-tiling-2-displays-1920x1080p,Fail +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-1920x1080p,Fail +kms_bw@linear-tiling-3-displays-2560x1440p,Fail +kms_bw@linear-tiling-3-displays-3840x2160p,Fail +kms_bw@linear-tiling-4-displays-1920x1080p,Fail +kms_bw@linear-tiling-4-displays-2560x1440p,Fail +kms_bw@linear-tiling-4-displays-3840x2160p,Fail +kms_fbcon_fbt@fbc,Fail +kms_fbcon_fbt@fbc-suspend,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-linear-to-64bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-xtile-to-64bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytile-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-16bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-linear-to-32bpp-linear-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-16bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-xtile-to-32bpp-xtile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytile-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilegen12rcccs-upscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-downscaling,Fail +kms_flip_scaled_crc@flip-64bpp-ytile-to-32bpp-ytilercccs-upscaling,Fail +kms_frontbuffer_tracking@fbc-tiling-linear,Fail +kms_plane_alpha_blend@alpha-basic,Fail +kms_plane_alpha_blend@alpha-opaque-fb,Fail +kms_plane_alpha_blend@alpha-transparent-fb,Fail +kms_plane_alpha_blend@constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-A-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-A-alpha-transparent-fb,Fail +kms_plane_alpha_blend@pipe-A-constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-B-alpha-transparent-fb,Fail +kms_plane_alpha_blend@pipe-B-constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-C-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-C-alpha-transparent-fb,Fail +kms_plane_alpha_blend@pipe-C-constant-alpha-max,Fail diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt b/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt new file mode 100644 index 000000000000..c33202e7e2a1 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-whl-flakes.txt @@ -0,0 +1 @@ +kms_flip@flip-vs-suspend diff --git a/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt b/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt new file mode 100644 index 000000000000..f3be0888a214 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/i915-whl-skips.txt @@ -0,0 +1,2 @@ +# This is generating kernel oops with divide error +kms_plane_scaling@invalid-parameters
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt new file mode 100644 index 000000000000..671916067dba --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-fails.txt @@ -0,0 +1,29 @@ +kms_3d,Fail +kms_addfb_basic@addfb25-bad-modifier,Fail +kms_bw@linear-tiling-1-displays-1920x1080p,Fail +kms_bw@linear-tiling-1-displays-2560x1440p,Fail +kms_bw@linear-tiling-1-displays-3840x2160p,Fail +kms_bw@linear-tiling-2-displays-1920x1080p,Fail +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-1920x1080p,Fail +kms_bw@linear-tiling-3-displays-2560x1440p,Fail +kms_bw@linear-tiling-3-displays-3840x2160p,Fail +kms_color@pipe-A-invalid-gamma-lut-sizes,Fail +kms_color@pipe-B-invalid-gamma-lut-sizes,Fail +kms_force_connector_basic@force-connector-state,Fail +kms_force_connector_basic@force-edid,Fail +kms_force_connector_basic@force-load-detect,Fail +kms_force_connector_basic@prune-stale-modes,Fail +kms_invalid_mode@int-max-clock,Fail +kms_plane_scaling@planes-upscale-20x20,Fail +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25,Fail +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-5,Fail +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-75,Fail +kms_plane_scaling@upscale-with-modifier-20x20,Fail +kms_plane_scaling@upscale-with-pixel-format-20x20,Fail +kms_plane_scaling@upscale-with-rotation-20x20,Fail +kms_properties@get_properties-sanity-atomic,Fail +kms_properties@plane-properties-atomic,Fail +kms_properties@plane-properties-legacy,Fail +kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt new file mode 100644 index 000000000000..e69de29bb2d1 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8173-flakes.txt diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt new file mode 100644 index 000000000000..6ff81d00e84e --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-fails.txt @@ -0,0 +1,10 @@ +kms_addfb_basic@addfb25-bad-modifier,Fail +kms_bw@linear-tiling-1-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-1920x1080p,Fail +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-2560x1440p,Fail +kms_bw@linear-tiling-3-displays-3840x2160p,Fail +kms_color@pipe-A-invalid-gamma-lut-sizes,Fail +kms_plane_scaling@upscale-with-rotation-20x20,Fail +kms_rmfb@close-fd,Fail
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt new file mode 100644 index 000000000000..208890b79eb0 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/mediatek-mt8183-flakes.txt @@ -0,0 +1,14 @@ +core_setmaster_vs_auth +kms_bw@linear-tiling-1-displays-1920x1080p +kms_bw@linear-tiling-1-displays-3840x2160p +kms_bw@linear-tiling-3-displays-1920x1080p +kms_cursor_legacy@cursor-vs-flip-atomic +kms_plane_scaling@invalid-num-scalers +kms_plane_scaling@planes-upscale-20x20 +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-5 +kms_plane_scaling@upscale-with-modifier-20x20 +kms_plane_scaling@upscale-with-pixel-format-20x20 +kms_prop_blob@invalid-set-prop-any +kms_properties@get_properties-sanity-atomic +kms_properties@plane-properties-atomic +kms_properties@plane-properties-legacy
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/meson-g12b-fails.txt b/drivers/gpu/drm/ci/xfails/meson-g12b-fails.txt new file mode 100644 index 000000000000..860e702091e2 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/meson-g12b-fails.txt @@ -0,0 +1,12 @@ +kms_3d,Fail +kms_properties@connector-properties-atomic,Fail +kms_properties@get_properties-sanity-atomic,Fail +kms_properties@get_properties-sanity-non-atomic,Fail +kms_properties@connector-properties-legacy,Fail +kms_cursor_legacy@forked-bo,Fail +kms_cursor_legacy@forked-move,Fail +kms_cursor_legacy@single-bo,Fail +kms_cursor_legacy@single-move,Fail +kms_cursor_legacy@torture-bo,Fail +kms_cursor_legacy@torture-move,Fail +kms_hdmi_inject@inject-4k,Fail
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/meson-g12b-flakes.txt b/drivers/gpu/drm/ci/xfails/meson-g12b-flakes.txt new file mode 100644 index 000000000000..b63329d06767 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/meson-g12b-flakes.txt @@ -0,0 +1,4 @@ +kms_force_connector_basic@force-connector-state +kms_force_connector_basic@force-edid +kms_force_connector_basic@force-load-detect +kms_force_connector_basic@prune-stale-modes
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt b/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt new file mode 100644 index 000000000000..9981682feab2 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-apq8016-fails.txt @@ -0,0 +1,15 @@ +kms_3d,Fail +kms_addfb_basic@addfb25-bad-modifier,Fail +kms_cursor_legacy@all-pipes-forked-bo,Fail +kms_cursor_legacy@all-pipes-forked-move,Fail +kms_cursor_legacy@all-pipes-single-bo,Fail +kms_cursor_legacy@all-pipes-single-move,Fail +kms_cursor_legacy@all-pipes-torture-bo,Fail +kms_cursor_legacy@all-pipes-torture-move,Fail +kms_cursor_legacy@pipe-A-forked-bo,Fail +kms_cursor_legacy@pipe-A-forked-move,Fail +kms_cursor_legacy@pipe-A-single-bo,Fail +kms_cursor_legacy@pipe-A-single-move,Fail +kms_cursor_legacy@pipe-A-torture-bo,Fail +kms_cursor_legacy@pipe-A-torture-move,Fail +kms_hdmi_inject@inject-4k,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt new file mode 100644 index 000000000000..0e3b60d3fade --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-apq8016-flakes.txt @@ -0,0 +1,4 @@ +kms_force_connector_basic@force-connector-state +kms_force_connector_basic@force-edid +kms_force_connector_basic@force-load-detect +kms_force_connector_basic@prune-stale-modes diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8096-fails.txt b/drivers/gpu/drm/ci/xfails/msm-apq8096-fails.txt new file mode 100644 index 000000000000..88a1fc0a3b0d --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-apq8096-fails.txt @@ -0,0 +1,2 @@ +kms_3d,Fail +kms_addfb_basic@addfb25-bad-modifier,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8096-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-apq8096-flakes.txt new file mode 100644 index 000000000000..0e3b60d3fade --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-apq8096-flakes.txt @@ -0,0 +1,4 @@ +kms_force_connector_basic@force-connector-state +kms_force_connector_basic@force-edid +kms_force_connector_basic@force-load-detect +kms_force_connector_basic@prune-stale-modes diff --git a/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt b/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt new file mode 100644 index 000000000000..cd49c8ce2059 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-apq8096-skips.txt @@ -0,0 +1,2 @@ +# Whole machine hangs +kms_cursor_legacy@all-pipes-torture-move
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-fails.txt new file mode 100644 index 000000000000..14adeba3b62d --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-fails.txt @@ -0,0 +1,25 @@ +kms_cursor_legacy@cursor-vs-flip-toggle,Fail +kms_cursor_legacy@cursor-vs-flip-varying-size,Fail +kms_cursor_legacy@cursorA-vs-flipA-atomic-transitions,Crash +kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail +kms_plane@pixel-format,Fail +kms_plane@pixel-format-source-clamping,Fail +kms_plane@plane-position-covered,Fail +kms_plane@plane-position-hole,Fail +kms_plane@plane-position-hole-dpms,Fail +kms_plane_alpha_blend@alpha-7efc,Fail +kms_plane_alpha_blend@coverage-7efc,Fail +kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail +kms_plane_alpha_blend@pipe-A-alpha-7efc,Fail +kms_plane_alpha_blend@pipe-A-coverage-7efc,Fail +kms_plane_alpha_blend@pipe-A-coverage-vs-premult-vs-constant,Fail +kms_plane_alpha_blend@pipe-B-alpha-7efc,Fail +kms_plane_alpha_blend@pipe-B-alpha-basic,Fail +kms_plane_alpha_blend@pipe-B-alpha-opaque-fb,Fail +kms_plane_alpha_blend@pipe-B-constant-alpha-max,Fail +kms_plane_alpha_blend@pipe-B-constant-alpha-mid,Fail +kms_plane_alpha_blend@pipe-B-coverage-7efc,Fail +kms_plane_alpha_blend@pipe-B-coverage-vs-premult-vs-constant,Fail +kms_rmfb@close-fd,Fail +kms_universal_plane@disable-primary-vs-flip-pipe-b,Fail +kms_universal_plane@universal-plane-pipe-B-sanity,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-flakes.txt new file mode 100644 index 000000000000..636563d3e59a --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-flakes.txt @@ -0,0 +1,7 @@ + +# Test ends up reading CRC from frame before cursor update +# bug +# sometimes.. tbd if this is a kernel CRC bug or a test +kms_cursor_crc@.* +kms_plane_multiple@atomic-pipe-A-tiling-none +kms_atomic_transition@modeset-transition-nonblocking-fencing,Fail
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/msm-sc7180-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sc7180-skips.txt new file mode 100644 index 000000000000..410e0eeb3161 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-sc7180-skips.txt @@ -0,0 +1,23 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* + +# Test incorrectly assumes that CTM support implies gamma/degamma +# LUT support. None of the subtests handle the case of only having +# CTM support +kms_color.* + +# 4k@60 is not supported on this hw, but driver doesn't handle it +# too gracefully.. https://gitlab.freedesktop.org/drm/msm/-/issues/15 +kms_bw@linear-tiling-.*-displays-3840x2160p + +# Until igt fix lands: https://patchwork.freedesktop.org/patch/493175/ +kms_bw@linear-tiling-2.* +kms_bw@linear-tiling-3.* +kms_bw@linear-tiling-4.* +kms_bw@linear-tiling-5.* +kms_bw@linear-tiling-6.* + +# igt fix posted: https://patchwork.freedesktop.org/patch/499926/ +# failure mode is flakey due to randomization but fails frequently +# enough to be detected as a Crash or occasionally UnexpectedPass. +kms_plane_multiple@atomic-pipe-A-tiling-none diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt new file mode 100644 index 000000000000..09c0c623cd75 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-fails.txt @@ -0,0 +1,68 @@ +kms_color@ctm-0-25,Fail +kms_color@ctm-0-50,Fail +kms_color@ctm-0-75,Fail +kms_color@ctm-blue-to-red,Fail +kms_color@ctm-green-to-red,Fail +kms_color@ctm-negative,Fail +kms_color@ctm-red-to-blue,Fail +kms_color@ctm-signed,Fail +kms_color@pipe-A-ctm-0-25,Fail +kms_color@pipe-A-ctm-0-5,Fail +kms_color@pipe-A-ctm-0-75,Fail +kms_color@pipe-A-ctm-blue-to-red,Fail +kms_color@pipe-A-ctm-green-to-red,Fail +kms_color@pipe-A-ctm-max,Fail +kms_color@pipe-A-ctm-negative,Fail +kms_color@pipe-A-ctm-red-to-blue,Fail +kms_color@pipe-A-legacy-gamma,Fail +kms_cursor_legacy@basic-flip-after-cursor-atomic,Fail +kms_cursor_legacy@basic-flip-after-cursor-legacy,Fail +kms_cursor_legacy@basic-flip-after-cursor-varying-size,Fail +kms_cursor_legacy@basic-flip-before-cursor-atomic,Fail +kms_cursor_legacy@basic-flip-before-cursor-legacy,Fail +kms_cursor_legacy@basic-flip-before-cursor-varying-size,Fail +kms_cursor_legacy@cursor-vs-flip-atomic,Fail +kms_cursor_legacy@cursor-vs-flip-atomic-transitions,Fail +kms_cursor_legacy@cursor-vs-flip-atomic-transitions-varying-size,Fail +kms_cursor_legacy@cursor-vs-flip-legacy,Fail +kms_cursor_legacy@cursor-vs-flip-toggle,Fail +kms_cursor_legacy@cursor-vs-flip-varying-size,Fail +kms_cursor_legacy@cursorA-vs-flipA-toggle,Fail +kms_cursor_legacy@flip-vs-cursor-atomic,Fail +kms_cursor_legacy@flip-vs-cursor-crc-atomic,Fail +kms_cursor_legacy@flip-vs-cursor-crc-legacy,Fail +kms_cursor_legacy@flip-vs-cursor-legacy,Fail +kms_cursor_legacy@short-flip-after-cursor-atomic-transitions,Fail +kms_cursor_legacy@short-flip-after-cursor-atomic-transitions-varying-size,Fail +kms_cursor_legacy@short-flip-after-cursor-toggle,Fail +kms_cursor_legacy@short-flip-before-cursor-atomic-transitions,Fail +kms_cursor_legacy@short-flip-before-cursor-atomic-transitions-varying-size,Fail +kms_pipe_crc_basic@compare-crc-sanitycheck-nv12,Fail +kms_plane@pixel-format,Fail +kms_plane@pixel-format-source-clamping,Fail +kms_plane_alpha_blend@alpha-7efc,Fail +kms_plane_alpha_blend@coverage-7efc,Fail +kms_plane_alpha_blend@coverage-vs-premult-vs-constant,Fail +kms_plane_alpha_blend@pipe-A-alpha-7efc,Fail +kms_plane_alpha_blend@pipe-A-coverage-7efc,Fail +kms_plane_alpha_blend@pipe-A-coverage-vs-premult-vs-constant,Fail +kms_plane_cursor@overlay,Fail +kms_plane_cursor@pipe-A-overlay-size-128,Fail +kms_plane_cursor@pipe-A-overlay-size-256,Fail +kms_plane_cursor@pipe-A-overlay-size-64,Fail +kms_plane_cursor@pipe-A-viewport-size-128,Fail +kms_plane_cursor@pipe-A-viewport-size-256,Fail +kms_plane_cursor@pipe-A-viewport-size-64,Fail +kms_plane_cursor@viewport,Fail +kms_plane_scaling@downscale-with-pixel-format-factor-0-25,Timeout +kms_plane_scaling@downscale-with-pixel-format-factor-0-5,Timeout +kms_plane_scaling@downscale-with-pixel-format-factor-0-75,Timeout +kms_plane_scaling@plane-downscale-with-pixel-format-factor-0-25,Timeout +kms_plane_scaling@plane-downscale-with-pixel-format-factor-0-5,Timeout +kms_plane_scaling@plane-downscale-with-pixel-format-factor-0-75,Timeout +kms_plane_scaling@plane-scaler-with-clipping-clamping-pixel-formats,Timeout +kms_plane_scaling@plane-scaler-with-pixel-format-unity-scaling,Timeout +kms_plane_scaling@planes-downscale-factor-0-25,Fail +kms_plane_scaling@scaler-with-clipping-clamping,Timeout +kms_plane_scaling@scaler-with-pixel-format-unity-scaling,Timeout +kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt new file mode 100644 index 000000000000..5b3aaab7ac3f --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-flakes.txt @@ -0,0 +1,11 @@ + + +# Test ends up reading CRC from frame before cursor update +# bug +# sometimes.. tbd if this is a kernel CRC bug or a test +kms_cursor_crc@.* +kms_cursor_legacy@flip-vs-cursor-toggle +kms_cursor_legacy@pipe-A-forked-bo +kms_cursor_legacy@pipe-A-forked-move +kms_cursor_legacy@short-flip-before-cursor-toggle +kms_flip@dpms-vs-vblank-race-interruptible diff --git a/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt new file mode 100644 index 000000000000..42675f1c6d76 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/msm-sdm845-skips.txt @@ -0,0 +1,2 @@ +# Hangs machine +kms_bw.*
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt new file mode 100644 index 000000000000..2a1baa948e12 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-fails.txt @@ -0,0 +1,48 @@ +kms_3d,Crash +kms_bw@linear-tiling-2-displays-1920x1080p,Fail +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_bw@linear-tiling-3-displays-1920x1080p,Fail +kms_bw@linear-tiling-3-displays-2560x1440p,Fail +kms_bw@linear-tiling-3-displays-3840x2160p,Fail +kms_force_connector_basic@force-load-detect,Fail +kms_invalid_mode@int-max-clock,Crash +kms_plane@pixel-format,Crash +kms_plane@pixel-format-source-clamping,Crash +kms_plane@plane-position-hole,Crash +kms_plane@plane-position-hole-dpms,Crash +kms_plane_cursor@overlay,Crash +kms_plane_cursor@pipe-A-overlay-size-128,Fail +kms_plane_cursor@pipe-A-overlay-size-256,Fail +kms_plane_cursor@pipe-A-overlay-size-64,Fail +kms_plane_cursor@pipe-A-primary-size-128,Fail +kms_plane_cursor@pipe-A-primary-size-256,Fail +kms_plane_cursor@pipe-A-primary-size-64,Fail +kms_plane_cursor@pipe-A-viewport-size-128,Fail +kms_plane_cursor@pipe-A-viewport-size-256,Fail +kms_plane_cursor@pipe-A-viewport-size-64,Fail +kms_plane_cursor@pipe-B-overlay-size-128,Fail +kms_plane_cursor@pipe-B-overlay-size-256,Fail +kms_plane_cursor@pipe-B-overlay-size-64,Fail +kms_plane_cursor@pipe-B-primary-size-128,Fail +kms_plane_cursor@pipe-B-primary-size-256,Fail +kms_plane_cursor@pipe-B-primary-size-64,Fail +kms_plane_cursor@pipe-B-viewport-size-128,Fail +kms_plane_cursor@pipe-B-viewport-size-256,Fail +kms_plane_cursor@pipe-B-viewport-size-64,Fail +kms_plane_cursor@primary,Crash +kms_plane_cursor@viewport,Crash +kms_plane_lowres@tiling-none,Fail +kms_plane_scaling@downscale-with-modifier-factor-0-25,Fail +kms_plane_scaling@downscale-with-rotation-factor-0-25,Fail +kms_plane_scaling@upscale-with-modifier-20x20,Fail +kms_plane_scaling@upscale-with-modifier-factor-0-25,Fail +kms_plane_scaling@upscale-with-pixel-format-20x20,Fail +kms_plane_scaling@upscale-with-pixel-format-factor-0-25,Fail +kms_plane_scaling@upscale-with-rotation-20x20,Fail +kms_prime@basic-crc,Fail +kms_properties@connector-properties-atomic,Crash +kms_properties@connector-properties-legacy,Crash +kms_properties@get_properties-sanity-atomic,Crash +kms_properties@get_properties-sanity-non-atomic,Crash +kms_setmode@invalid-clone-single-crtc,Crash diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt new file mode 100644 index 000000000000..45c54c75c899 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-flakes.txt @@ -0,0 +1,9 @@ +kms_addfb_basic@addfb25-bad-modifier +kms_cursor_crc@.* +kms_flip@basic-flip-vs-wf_vblank +kms_invalid_mode@int-max-clock,Crash +kms_pipe_crc_basic@.* +kms_properties@connector-properties-atomic,Crash +kms_properties@get_properties-sanity-atomic,Crash +kms_properties@get_properties-sanity-non-atomic,Crash +kms_rmfb@close-fd diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt new file mode 100644 index 000000000000..f20c3574b75a --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3288-skips.txt @@ -0,0 +1,52 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* + +# Too unstable, machine ends up hanging after lots of Oopses +kms_cursor_legacy.* + +# Started hanging the machine on Linux 5.19-rc2: +# +# [IGT] kms_plane_lowres: executing +# [IGT] kms_plane_lowres: starting subtest pipe-F-tiling-y +# [IGT] kms_plane_lowres: exiting, ret=77 +# Console: switching to colour frame buffer device 170x48 +# rockchip-drm display-subsystem: [drm] *ERROR* flip_done timed out +# rockchip-drm display-subsystem: [drm] *ERROR* [CRTC:35:crtc-0] commit wait timed out +# BUG: spinlock bad magic on CPU#3, kms_plane_lowre/482 +# 8<--- cut here --- +# Unable to handle kernel paging request at virtual address 7812078e +# [7812078e] *pgd=00000000 +# Internal error: Oops: 5 [#1] SMP ARM +# Modules linked in: +# CPU: 3 PID: 482 Comm: kms_plane_lowre Tainted: G W 5.19.0-rc2-323596-g00535de92171 #1 +# Hardware name: Rockchip (Device Tree) +# Process kms_plane_lowre (pid: 482, stack limit = 0x1193ac2b) +# spin_dump from do_raw_spin_lock+0xa4/0xe8 +# do_raw_spin_lock from wait_for_completion_timeout+0x2c/0x120 +# wait_for_completion_timeout from drm_crtc_commit_wait+0x18/0x7c +# drm_crtc_commit_wait from drm_atomic_helper_wait_for_dependencies+0x44/0x168 +# drm_atomic_helper_wait_for_dependencies from commit_tail+0x34/0x180 +# commit_tail from drm_atomic_helper_commit+0x164/0x18c +# drm_atomic_helper_commit from drm_atomic_commit+0xac/0xe4 +# drm_atomic_commit from drm_client_modeset_commit_atomic+0x23c/0x284 +# drm_client_modeset_commit_atomic from drm_client_modeset_commit_locked+0x60/0x1c8 +# drm_client_modeset_commit_locked from drm_client_modeset_commit+0x24/0x40 +# drm_client_modeset_commit from drm_fbdev_client_restore+0x58/0x94 +# drm_fbdev_client_restore from drm_client_dev_restore+0x70/0xbc +# drm_client_dev_restore from drm_release+0xf4/0x114 +# drm_release from __fput+0x74/0x240 +# __fput from task_work_run+0x84/0xb4 +# task_work_run from do_exit+0x34c/0xa20 +# do_exit from do_group_exit+0x34/0x98 +# do_group_exit from __wake_up_parent+0x0/0x18 +# Code: e595c008 12843d19 03e00000 03093168 (15940508) +# ---[ end trace 0000000000000000 ]--- +# note: kms_plane_lowre[482] exited with preempt_count 1 +# Fixing recursive fault but reboot is needed! +kms_plane_lowres@pipe-F-tiling-y + +# Take too long, we have only two machines, and these are very flaky +kms_cursor_crc.* + +# Machine is hanging in this test, so skip it +kms_pipe_crc_basic@disable-crc-after-crtc
\ No newline at end of file diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt new file mode 100644 index 000000000000..6db08ba6b008 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-fails.txt @@ -0,0 +1,37 @@ +kms_color@legacy-gamma,Fail +kms_color@pipe-A-legacy-gamma,Fail +kms_color@pipe-B-legacy-gamma,Fail +kms_flip@basic-flip-vs-wf_vblank,Fail +kms_flip@blocking-wf_vblank,Fail +kms_flip@dpms-vs-vblank-race,Fail +kms_flip@flip-vs-absolute-wf_vblank,Fail +kms_flip@flip-vs-absolute-wf_vblank-interruptible,Fail +kms_flip@flip-vs-blocking-wf-vblank,Fail +kms_flip@flip-vs-panning,Fail +kms_flip@flip-vs-panning-interruptible,Fail +kms_flip@flip-vs-wf_vblank-interruptible,Fail +kms_flip@plain-flip-fb-recreate,Fail +kms_flip@plain-flip-fb-recreate-interruptible,Fail +kms_flip@plain-flip-ts-check,Fail +kms_flip@plain-flip-ts-check-interruptible,Fail +kms_flip@wf_vblank-ts-check,Fail +kms_flip@wf_vblank-ts-check-interruptible,Fail +kms_invalid_mode@int-max-clock,Fail +kms_plane@pixel-format,Fail +kms_plane@pixel-format-source-clamping,Fail +kms_plane@plane-panning-bottom-right,Fail +kms_plane@plane-panning-top-left,Fail +kms_plane@plane-position-covered,Fail +kms_plane_cursor@pipe-B-overlay-size-128,Fail +kms_plane_cursor@pipe-B-overlay-size-256,Fail +kms_plane_cursor@pipe-B-overlay-size-64,Fail +kms_plane_cursor@pipe-B-primary-size-128,Fail +kms_plane_cursor@pipe-B-primary-size-256,Fail +kms_plane_cursor@pipe-B-primary-size-64,Fail +kms_plane_cursor@pipe-B-viewport-size-128,Fail +kms_plane_cursor@pipe-B-viewport-size-256,Fail +kms_plane_cursor@pipe-B-viewport-size-64,Fail +kms_plane_multiple@atomic-pipe-B-tiling-none,Fail +kms_plane_multiple@tiling-none,Fail +kms_prime@basic-crc,Fail +kms_rmfb@close-fd,Fail diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt new file mode 100644 index 000000000000..4c0539b4beaf --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-flakes.txt @@ -0,0 +1,23 @@ + +kms_cursor_crc@.* +kms_flip@dpms-vs-vblank-race-interruptible +kms_flip@flip-vs-expired-vblank +kms_flip@modeset-vs-vblank-race-interruptible +kms_pipe_crc_basic@.* +kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-A +kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-B +kms_plane@plane-position-hole +kms_plane_multiple@atomic-pipe-A-tiling-none +kms_plane_multiple@atomic-pipe-B-tiling-none +kms_sequence@get-forked +kms_sequence@get-forked-busy +kms_setmode@basic +kms_universal_plane@universal-plane-pipe-B-functional,UnexpectedPass +kms_vblank@pipe-A-accuracy-idle +kms_vblank@pipe-A-query-busy +kms_vblank@pipe-A-query-forked-busy +kms_vblank@pipe-A-wait-idle +kms_vblank@pipe-B-accuracy-idle +kms_vblank@pipe-B-query-busy +kms_vblank@pipe-B-query-forked-busy +kms_vblank@pipe-B-wait-idle diff --git a/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt new file mode 100644 index 000000000000..10c3d81a919a --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/rockchip-rk3399-skips.txt @@ -0,0 +1,5 @@ +# Suspend to RAM seems to be broken on this machine +.*suspend.* + +# Too unstable, machine ends up hanging after lots of Oopses +kms_cursor_legacy.* diff --git a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt new file mode 100644 index 000000000000..9586b2339f6f --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-fails.txt @@ -0,0 +1,38 @@ +kms_addfb_basic@addfb25-bad-modifier,Fail +kms_addfb_basic@bad-pitch-65536,Fail +kms_addfb_basic@bo-too-small,Fail +kms_addfb_basic@size-max,Fail +kms_addfb_basic@too-high,Fail +kms_atomic_transition@plane-primary-toggle-with-vblank-wait,Fail +kms_bw@linear-tiling-1-displays-1920x1080p,Fail +kms_bw@linear-tiling-1-displays-2560x1440p,Fail +kms_bw@linear-tiling-1-displays-3840x2160p,Fail +kms_bw@linear-tiling-2-displays-1920x1080p,Fail +kms_bw@linear-tiling-2-displays-2560x1440p,Fail +kms_bw@linear-tiling-2-displays-3840x2160p,Fail +kms_invalid_mode@int-max-clock,Fail +kms_plane_scaling@downscale-with-modifier-factor-0-25,Fail +kms_plane_scaling@downscale-with-rotation-factor-0-25,Fail +kms_plane_scaling@planes-upscale-20x20,Fail +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-25,Fail +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-5,Fail +kms_plane_scaling@planes-upscale-20x20-downscale-factor-0-75,Fail +kms_plane_scaling@upscale-with-modifier-20x20,Fail +kms_plane_scaling@upscale-with-modifier-factor-0-25,Fail +kms_plane_scaling@upscale-with-pixel-format-20x20,Fail +kms_plane_scaling@upscale-with-pixel-format-factor-0-25,Fail +kms_plane_scaling@upscale-with-rotation-20x20,Fail +kms_vblank@crtc-id,Fail +kms_vblank@invalid,Fail +kms_vblank@pipe-A-accuracy-idle,Fail +kms_vblank@pipe-A-query-busy,Fail +kms_vblank@pipe-A-query-forked,Fail +kms_vblank@pipe-A-query-forked-busy,Fail +kms_vblank@pipe-A-query-idle,Fail +kms_vblank@pipe-A-ts-continuation-idle,Fail +kms_vblank@pipe-A-ts-continuation-modeset,Fail +kms_vblank@pipe-A-ts-continuation-suspend,Fail +kms_vblank@pipe-A-wait-busy,Fail +kms_vblank@pipe-A-wait-forked,Fail +kms_vblank@pipe-A-wait-forked-busy,Fail +kms_vblank@pipe-A-wait-idle,Fail diff --git a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-flakes.txt b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-flakes.txt new file mode 100644 index 000000000000..e69de29bb2d1 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-flakes.txt diff --git a/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt new file mode 100644 index 000000000000..78be18174012 --- /dev/null +++ b/drivers/gpu/drm/ci/xfails/virtio_gpu-none-skips.txt @@ -0,0 +1,6 @@ +# Hits a "refcount_t: underflow; use-after-free" in virtio_gpu_fence_event_process +# When run in a particular order with other tests +kms_cursor_legacy.* + +# Job just hangs without any output +kms_flip@flip-vs-suspend.*
\ No newline at end of file diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index affce6a8851f..f1a503aafe5a 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -140,6 +140,12 @@ drm_atomic_state_init(struct drm_device *dev, struct drm_atomic_state *state) if (!state->planes) goto fail; + /* + * Because drm_atomic_state can be committed asynchronously we need our + * own reference and cannot rely on the on implied by drm_file in the + * ioctl call. + */ + drm_dev_get(dev); state->dev = dev; drm_dbg_atomic(dev, "Allocated atomic state %p\n", state); @@ -299,7 +305,8 @@ EXPORT_SYMBOL(drm_atomic_state_clear); void __drm_atomic_state_free(struct kref *ref) { struct drm_atomic_state *state = container_of(ref, typeof(*state), ref); - struct drm_mode_config *config = &state->dev->mode_config; + struct drm_device *dev = state->dev; + struct drm_mode_config *config = &dev->mode_config; drm_atomic_state_clear(state); @@ -311,6 +318,8 @@ void __drm_atomic_state_free(struct kref *ref) drm_atomic_state_default_release(state); kfree(state); } + + drm_dev_put(dev); } EXPORT_SYMBOL(__drm_atomic_state_free); diff --git a/drivers/gpu/drm/drm_buddy.c b/drivers/gpu/drm/drm_buddy.c index 7098f125b54a..e6f5ba5f4baf 100644 --- a/drivers/gpu/drm/drm_buddy.c +++ b/drivers/gpu/drm/drm_buddy.c @@ -781,15 +781,15 @@ void drm_buddy_print(struct drm_buddy *mm, struct drm_printer *p) count++; } - drm_printf(p, "order-%d ", order); + drm_printf(p, "order-%2d ", order); free = count * (mm->chunk_size << order); if (free < SZ_1M) - drm_printf(p, "free: %lluKiB", free >> 10); + drm_printf(p, "free: %8llu KiB", free >> 10); else - drm_printf(p, "free: %lluMiB", free >> 20); + drm_printf(p, "free: %8llu MiB", free >> 20); - drm_printf(p, ", pages: %llu\n", count); + drm_printf(p, ", blocks: %llu\n", count); } } EXPORT_SYMBOL(drm_buddy_print); diff --git a/drivers/gpu/drm/drm_client_modeset.c b/drivers/gpu/drm/drm_client_modeset.c index 1b12a3c201a3..871e4e2129d6 100644 --- a/drivers/gpu/drm/drm_client_modeset.c +++ b/drivers/gpu/drm/drm_client_modeset.c @@ -311,6 +311,9 @@ static bool drm_client_target_cloned(struct drm_device *dev, can_clone = true; dmt_mode = drm_mode_find_dmt(dev, 1024, 768, 60, false); + if (!dmt_mode) + goto fail; + for (i = 0; i < connector_count; i++) { if (!enabled[i]) continue; @@ -326,11 +329,13 @@ static bool drm_client_target_cloned(struct drm_device *dev, if (!modes[i]) can_clone = false; } + kfree(dmt_mode); if (can_clone) { DRM_DEBUG_KMS("can clone using 1024x768\n"); return true; } +fail: DRM_INFO("kms: can't enable cloning when we probably wanted to.\n"); return false; } @@ -862,6 +867,7 @@ int drm_client_modeset_probe(struct drm_client_dev *client, unsigned int width, break; } + kfree(modeset->mode); modeset->mode = drm_mode_duplicate(dev, mode); drm_connector_get(connector); modeset->connectors[modeset->num_connectors++] = connector; diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index bf8371dc2a61..c44d5bcf1284 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -2203,6 +2203,7 @@ static int drm_mode_create_colorspace_property(struct drm_connector *connector, /** * drm_mode_create_hdmi_colorspace_property - create hdmi colorspace property * @connector: connector to create the Colorspace property on. + * @supported_colorspaces: bitmap of supported color spaces * * Called by a driver the first time it's needed, must be attached to desired * HDMI connectors. @@ -2227,6 +2228,7 @@ EXPORT_SYMBOL(drm_mode_create_hdmi_colorspace_property); /** * drm_mode_create_dp_colorspace_property - create dp colorspace property * @connector: connector to create the Colorspace property on. + * @supported_colorspaces: bitmap of supported color spaces * * Called by a driver the first time it's needed, must be attached to desired * DP connectors. diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index f95152fac427..340da8257b51 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -3457,6 +3457,10 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_connector *connecto connector->base.id, connector->name); return NULL; } + if (!(pt->misc & DRM_EDID_PT_SEPARATE_SYNC)) { + drm_dbg_kms(dev, "[CONNECTOR:%d:%s] Composite sync not supported\n", + connector->base.id, connector->name); + } /* it is incorrect if hsync/vsync width is zero */ if (!hsync_pulse_width || !vsync_pulse_width) { @@ -3503,27 +3507,10 @@ static struct drm_display_mode *drm_mode_detailed(struct drm_connector *connecto if (info->quirks & EDID_QUIRK_DETAILED_SYNC_PP) { mode->flags |= DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC; } else { - switch (pt->misc & DRM_EDID_PT_SYNC_MASK) { - case DRM_EDID_PT_ANALOG_CSYNC: - case DRM_EDID_PT_BIPOLAR_ANALOG_CSYNC: - drm_dbg_kms(dev, "[CONNECTOR:%d:%s] Analog composite sync!\n", - connector->base.id, connector->name); - mode->flags |= DRM_MODE_FLAG_CSYNC | DRM_MODE_FLAG_NCSYNC; - break; - case DRM_EDID_PT_DIGITAL_CSYNC: - drm_dbg_kms(dev, "[CONNECTOR:%d:%s] Digital composite sync!\n", - connector->base.id, connector->name); - mode->flags |= DRM_MODE_FLAG_CSYNC; - mode->flags |= (pt->misc & DRM_EDID_PT_HSYNC_POSITIVE) ? - DRM_MODE_FLAG_PCSYNC : DRM_MODE_FLAG_NCSYNC; - break; - case DRM_EDID_PT_DIGITAL_SEPARATE_SYNC: - mode->flags |= (pt->misc & DRM_EDID_PT_HSYNC_POSITIVE) ? - DRM_MODE_FLAG_PHSYNC : DRM_MODE_FLAG_NHSYNC; - mode->flags |= (pt->misc & DRM_EDID_PT_VSYNC_POSITIVE) ? - DRM_MODE_FLAG_PVSYNC : DRM_MODE_FLAG_NVSYNC; - break; - } + mode->flags |= (pt->misc & DRM_EDID_PT_HSYNC_POSITIVE) ? + DRM_MODE_FLAG_PHSYNC : DRM_MODE_FLAG_NHSYNC; + mode->flags |= (pt->misc & DRM_EDID_PT_VSYNC_POSITIVE) ? + DRM_MODE_FLAG_PVSYNC : DRM_MODE_FLAG_NVSYNC; } set_size: diff --git a/drivers/gpu/drm/drm_exec.c b/drivers/gpu/drm/drm_exec.c index ff69cf0fb42a..5d2809de4517 100644 --- a/drivers/gpu/drm/drm_exec.c +++ b/drivers/gpu/drm/drm_exec.c @@ -56,7 +56,7 @@ static void drm_exec_unlock_all(struct drm_exec *exec) struct drm_gem_object *obj; unsigned long index; - drm_exec_for_each_locked_object(exec, index, obj) { + drm_exec_for_each_locked_object_reverse(exec, index, obj) { dma_resv_unlock(obj->resv); drm_gem_object_put(obj); } diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 61a5d450cc20..d612133e2cf7 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -301,7 +301,7 @@ static void drm_fb_helper_restore_work_fn(struct work_struct *ignored) static DECLARE_WORK(drm_fb_helper_restore_work, drm_fb_helper_restore_work_fn); -static void drm_fb_helper_sysrq(int dummy1) +static void drm_fb_helper_sysrq(u8 dummy1) { schedule_work(&drm_fb_helper_restore_work); } diff --git a/drivers/gpu/drm/drm_gem_shmem_helper.c b/drivers/gpu/drm/drm_gem_shmem_helper.c index a783d2245599..e435f986cd13 100644 --- a/drivers/gpu/drm/drm_gem_shmem_helper.c +++ b/drivers/gpu/drm/drm_gem_shmem_helper.c @@ -594,7 +594,13 @@ int drm_gem_shmem_mmap(struct drm_gem_shmem_object *shmem, struct vm_area_struct int ret; if (obj->import_attach) { + /* Reset both vm_ops and vm_private_data, so we don't end up with + * vm_ops pointing to our implementation if the dma-buf backend + * doesn't set those fields. + */ vma->vm_private_data = NULL; + vma->vm_ops = NULL; + ret = dma_buf_mmap(obj->dma_buf, vma, 0); /* Drop the reference drm_gem_mmap_obj() acquired.*/ diff --git a/drivers/gpu/drm/drm_gpuva_mgr.c b/drivers/gpu/drm/drm_gpuva_mgr.c index f86bfad74ff8..ad99c9cfedac 100644 --- a/drivers/gpu/drm/drm_gpuva_mgr.c +++ b/drivers/gpu/drm/drm_gpuva_mgr.c @@ -1076,7 +1076,7 @@ __drm_gpuva_sm_map(struct drm_gpuva_manager *mgr, u64 req_addr, u64 req_range, struct drm_gem_object *req_obj, u64 req_offset) { - struct drm_gpuva *va, *next, *prev = NULL; + struct drm_gpuva *va, *next; u64 req_end = req_addr + req_range; int ret; @@ -1106,7 +1106,7 @@ __drm_gpuva_sm_map(struct drm_gpuva_manager *mgr, ret = op_unmap_cb(ops, priv, va, merge); if (ret) return ret; - goto next; + continue; } if (end > req_end) { @@ -1151,7 +1151,7 @@ __drm_gpuva_sm_map(struct drm_gpuva_manager *mgr, ret = op_remap_cb(ops, priv, &p, NULL, &u); if (ret) return ret; - goto next; + continue; } if (end > req_end) { @@ -1184,7 +1184,7 @@ __drm_gpuva_sm_map(struct drm_gpuva_manager *mgr, ret = op_unmap_cb(ops, priv, va, merge); if (ret) return ret; - goto next; + continue; } if (end > req_end) { @@ -1205,8 +1205,6 @@ __drm_gpuva_sm_map(struct drm_gpuva_manager *mgr, break; } } -next: - prev = va; } return op_map_cb(ops, priv, diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index 2fb9bf901a2c..3f479483d7d8 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -262,6 +262,26 @@ static bool drm_kms_helper_enable_hpd(struct drm_device *dev) } #define DRM_OUTPUT_POLL_PERIOD (10*HZ) +static void reschedule_output_poll_work(struct drm_device *dev) +{ + unsigned long delay = DRM_OUTPUT_POLL_PERIOD; + + if (dev->mode_config.delayed_event) + /* + * FIXME: + * + * Use short (1s) delay to handle the initial delayed event. + * This delay should not be needed, but Optimus/nouveau will + * fail in a mysterious way if the delayed event is handled as + * soon as possible like it is done in + * drm_helper_probe_single_connector_modes() in case the poll + * was enabled before. + */ + delay = HZ; + + schedule_delayed_work(&dev->mode_config.output_poll_work, delay); +} + /** * drm_kms_helper_poll_enable - re-enable output polling. * @dev: drm_device @@ -279,37 +299,41 @@ static bool drm_kms_helper_enable_hpd(struct drm_device *dev) */ void drm_kms_helper_poll_enable(struct drm_device *dev) { - bool poll = false; - unsigned long delay = DRM_OUTPUT_POLL_PERIOD; - if (!dev->mode_config.poll_enabled || !drm_kms_helper_poll || dev->mode_config.poll_running) return; - poll = drm_kms_helper_enable_hpd(dev); - - if (dev->mode_config.delayed_event) { - /* - * FIXME: - * - * Use short (1s) delay to handle the initial delayed event. - * This delay should not be needed, but Optimus/nouveau will - * fail in a mysterious way if the delayed event is handled as - * soon as possible like it is done in - * drm_helper_probe_single_connector_modes() in case the poll - * was enabled before. - */ - poll = true; - delay = HZ; - } - - if (poll) - schedule_delayed_work(&dev->mode_config.output_poll_work, delay); + if (drm_kms_helper_enable_hpd(dev) || + dev->mode_config.delayed_event) + reschedule_output_poll_work(dev); dev->mode_config.poll_running = true; } EXPORT_SYMBOL(drm_kms_helper_poll_enable); +/** + * drm_kms_helper_poll_reschedule - reschedule the output polling work + * @dev: drm_device + * + * This function reschedules the output polling work, after polling for a + * connector has been enabled. + * + * Drivers must call this helper after enabling polling for a connector by + * setting %DRM_CONNECTOR_POLL_CONNECT / %DRM_CONNECTOR_POLL_DISCONNECT flags + * in drm_connector::polled. Note that after disabling polling by clearing these + * flags for a connector will stop the output polling work automatically if + * the polling is disabled for all other connectors as well. + * + * The function can be called only after polling has been enabled by calling + * drm_kms_helper_poll_init() / drm_kms_helper_poll_enable(). + */ +void drm_kms_helper_poll_reschedule(struct drm_device *dev) +{ + if (dev->mode_config.poll_running) + reschedule_output_poll_work(dev); +} +EXPORT_SYMBOL(drm_kms_helper_poll_reschedule); + static enum drm_connector_status drm_helper_probe_detect_ctx(struct drm_connector *connector, bool force) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c index cf741c5c82d2..384df1659be6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c @@ -53,11 +53,12 @@ static inline void CMD_END(struct etnaviv_cmdbuf *buffer) OUT(buffer, VIV_FE_END_HEADER_OP_END); } -static inline void CMD_WAIT(struct etnaviv_cmdbuf *buffer) +static inline void CMD_WAIT(struct etnaviv_cmdbuf *buffer, + unsigned int waitcycles) { buffer->user_size = ALIGN(buffer->user_size, 8); - OUT(buffer, VIV_FE_WAIT_HEADER_OP_WAIT | 200); + OUT(buffer, VIV_FE_WAIT_HEADER_OP_WAIT | waitcycles); } static inline void CMD_LINK(struct etnaviv_cmdbuf *buffer, @@ -168,7 +169,7 @@ u16 etnaviv_buffer_init(struct etnaviv_gpu *gpu) /* initialize buffer */ buffer->user_size = 0; - CMD_WAIT(buffer); + CMD_WAIT(buffer, gpu->fe_waitcycles); CMD_LINK(buffer, 2, etnaviv_cmdbuf_get_va(buffer, &gpu->mmu_context->cmdbuf_mapping) + buffer->user_size - 4); @@ -320,7 +321,7 @@ void etnaviv_sync_point_queue(struct etnaviv_gpu *gpu, unsigned int event) CMD_END(buffer); /* Append waitlink */ - CMD_WAIT(buffer); + CMD_WAIT(buffer, gpu->fe_waitcycles); CMD_LINK(buffer, 2, etnaviv_cmdbuf_get_va(buffer, &gpu->mmu_context->cmdbuf_mapping) + buffer->user_size - 4); @@ -503,7 +504,7 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, CMD_LOAD_STATE(buffer, VIVS_GL_EVENT, VIVS_GL_EVENT_EVENT_ID(event) | VIVS_GL_EVENT_FROM_PE); - CMD_WAIT(buffer); + CMD_WAIT(buffer, gpu->fe_waitcycles); CMD_LINK(buffer, 2, etnaviv_cmdbuf_get_va(buffer, &gpu->mmu_context->cmdbuf_mapping) + buffer->user_size - 4); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c b/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c index 9dc20d892c15..721d633aece9 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c @@ -121,6 +121,9 @@ void etnaviv_cmdbuf_free(struct etnaviv_cmdbuf *cmdbuf) int order = order_base_2(ALIGN(cmdbuf->size, SUBALLOC_GRANULE) / SUBALLOC_GRANULE); + if (!suballoc) + return; + mutex_lock(&suballoc->lock); bitmap_release_region(suballoc->granule_map, cmdbuf->suballoc_offset / SUBALLOC_GRANULE, diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index ea55f6b7b744..a8d3fa81e4ec 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -6,7 +6,9 @@ #include <linux/component.h> #include <linux/dma-mapping.h> #include <linux/module.h> -#include <linux/of_platform.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> #include <linux/uaccess.h> #include <drm/drm_debugfs.h> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_dump.c b/drivers/gpu/drm/etnaviv/etnaviv_dump.c index 44b5f3c35aab..898f84a0fc30 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_dump.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_dump.c @@ -130,9 +130,9 @@ void etnaviv_core_dump(struct etnaviv_gem_submit *submit) return; etnaviv_dump_core = false; - mutex_lock(&gpu->mmu_context->lock); + mutex_lock(&submit->mmu_context->lock); - mmu_size = etnaviv_iommu_dump_size(gpu->mmu_context); + mmu_size = etnaviv_iommu_dump_size(submit->mmu_context); /* We always dump registers, mmu, ring, hanging cmdbuf and end marker */ n_obj = 5; @@ -162,7 +162,7 @@ void etnaviv_core_dump(struct etnaviv_gem_submit *submit) iter.start = __vmalloc(file_size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); if (!iter.start) { - mutex_unlock(&gpu->mmu_context->lock); + mutex_unlock(&submit->mmu_context->lock); dev_warn(gpu->dev, "failed to allocate devcoredump file\n"); return; } @@ -174,18 +174,18 @@ void etnaviv_core_dump(struct etnaviv_gem_submit *submit) memset(iter.hdr, 0, iter.data - iter.start); etnaviv_core_dump_registers(&iter, gpu); - etnaviv_core_dump_mmu(&iter, gpu->mmu_context, mmu_size); + etnaviv_core_dump_mmu(&iter, submit->mmu_context, mmu_size); etnaviv_core_dump_mem(&iter, ETDUMP_BUF_RING, gpu->buffer.vaddr, gpu->buffer.size, etnaviv_cmdbuf_get_va(&gpu->buffer, - &gpu->mmu_context->cmdbuf_mapping)); + &submit->mmu_context->cmdbuf_mapping)); etnaviv_core_dump_mem(&iter, ETDUMP_BUF_CMD, submit->cmdbuf.vaddr, submit->cmdbuf.size, etnaviv_cmdbuf_get_va(&submit->cmdbuf, - &gpu->mmu_context->cmdbuf_mapping)); + &submit->mmu_context->cmdbuf_mapping)); - mutex_unlock(&gpu->mmu_context->lock); + mutex_unlock(&submit->mmu_context->lock); /* Reserve space for the bomap */ if (n_bomap_pages) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.h b/drivers/gpu/drm/etnaviv/etnaviv_gem.h index baa81cbf701a..a42d260cac2c 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.h @@ -97,7 +97,6 @@ struct etnaviv_gem_submit { struct list_head node; /* GPU active submit list */ struct etnaviv_cmdbuf cmdbuf; struct pid *pid; /* submitting process */ - bool runtime_resumed; u32 exec_state; u32 flags; unsigned int nr_pmrs; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c index 45403ea38906..2416c526f9b0 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c @@ -362,9 +362,6 @@ static void submit_cleanup(struct kref *kref) container_of(kref, struct etnaviv_gem_submit, refcount); unsigned i; - if (submit->runtime_resumed) - pm_runtime_put_autosuspend(submit->gpu->dev); - if (submit->cmdbuf.suballoc) etnaviv_cmdbuf_free(&submit->cmdbuf); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index bbc9c54871f4..9276756e1397 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -493,6 +493,14 @@ static void etnaviv_gpu_update_clock(struct etnaviv_gpu *gpu) clock |= VIVS_HI_CLOCK_CONTROL_FSCALE_VAL(fscale); etnaviv_gpu_load_clock(gpu, clock); } + + /* + * Choose number of wait cycles to target a ~30us (1/32768) max latency + * until new work is picked up by the FE when it polls in the idle loop. + * If the GPU base frequency is unknown use 200 wait cycles. + */ + gpu->fe_waitcycles = clamp(gpu->base_rate_core >> (15 - gpu->freq_scale), + 200UL, 0xffffUL); } static int etnaviv_hw_reset(struct etnaviv_gpu *gpu) @@ -576,7 +584,7 @@ static int etnaviv_hw_reset(struct etnaviv_gpu *gpu) /* We rely on the GPU running, so program the clock */ etnaviv_gpu_update_clock(gpu); - gpu->fe_running = false; + gpu->state = ETNA_GPU_STATE_RESET; gpu->exec_state = -1; if (gpu->mmu_context) etnaviv_iommu_context_put(gpu->mmu_context); @@ -651,8 +659,6 @@ void etnaviv_gpu_start_fe(struct etnaviv_gpu *gpu, u32 address, u16 prefetch) VIVS_MMUv2_SEC_COMMAND_CONTROL_ENABLE | VIVS_MMUv2_SEC_COMMAND_CONTROL_PREFETCH(prefetch)); } - - gpu->fe_running = true; } static void etnaviv_gpu_start_fe_idleloop(struct etnaviv_gpu *gpu, @@ -661,6 +667,8 @@ static void etnaviv_gpu_start_fe_idleloop(struct etnaviv_gpu *gpu, u16 prefetch; u32 address; + WARN_ON(gpu->state != ETNA_GPU_STATE_INITIALIZED); + /* setup the MMU */ etnaviv_iommu_restore(gpu, context); @@ -670,6 +678,8 @@ static void etnaviv_gpu_start_fe_idleloop(struct etnaviv_gpu *gpu, &gpu->mmu_context->cmdbuf_mapping); etnaviv_gpu_start_fe(gpu, address, prefetch); + + gpu->state = ETNA_GPU_STATE_RUNNING; } static void etnaviv_gpu_setup_pulse_eater(struct etnaviv_gpu *gpu) @@ -705,6 +715,9 @@ static void etnaviv_gpu_setup_pulse_eater(struct etnaviv_gpu *gpu) static void etnaviv_gpu_hw_init(struct etnaviv_gpu *gpu) { + WARN_ON(!(gpu->state == ETNA_GPU_STATE_IDENTIFIED || + gpu->state == ETNA_GPU_STATE_RESET)); + if ((etnaviv_is_model_rev(gpu, GC320, 0x5007) || etnaviv_is_model_rev(gpu, GC320, 0x5220)) && gpu_read(gpu, VIVS_HI_CHIP_TIME) != 0x2062400) { @@ -751,6 +764,8 @@ static void etnaviv_gpu_hw_init(struct etnaviv_gpu *gpu) etnaviv_gpu_setup_pulse_eater(gpu); gpu_write(gpu, VIVS_HI_INTR_ENBL, ~0U); + + gpu->state = ETNA_GPU_STATE_INITIALIZED; } int etnaviv_gpu_init(struct etnaviv_gpu *gpu) @@ -793,6 +808,8 @@ int etnaviv_gpu_init(struct etnaviv_gpu *gpu) (gpu->identity.minor_features10 & chipMinorFeatures10_SECURITY_AHB)) gpu->sec_mode = ETNA_SEC_KERNEL; + gpu->state = ETNA_GPU_STATE_IDENTIFIED; + ret = etnaviv_hw_reset(gpu); if (ret) { dev_err(gpu->dev, "GPU reset failed\n"); @@ -859,8 +876,6 @@ int etnaviv_gpu_init(struct etnaviv_gpu *gpu) pm_runtime_mark_last_busy(gpu->dev); pm_runtime_put_autosuspend(gpu->dev); - gpu->initialized = true; - return 0; fail: @@ -1059,50 +1074,6 @@ pm_put: } #endif -void etnaviv_gpu_recover_hang(struct etnaviv_gem_submit *submit) -{ - struct etnaviv_gpu *gpu = submit->gpu; - char *comm = NULL, *cmd = NULL; - struct task_struct *task; - unsigned int i; - - dev_err(gpu->dev, "recover hung GPU!\n"); - - task = get_pid_task(submit->pid, PIDTYPE_PID); - if (task) { - comm = kstrdup(task->comm, GFP_KERNEL); - cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); - put_task_struct(task); - } - - if (comm && cmd) - dev_err(gpu->dev, "offending task: %s (%s)\n", comm, cmd); - - kfree(cmd); - kfree(comm); - - if (pm_runtime_get_sync(gpu->dev) < 0) - goto pm_put; - - mutex_lock(&gpu->lock); - - etnaviv_hw_reset(gpu); - - /* complete all events, the GPU won't do it after the reset */ - spin_lock(&gpu->event_spinlock); - for_each_set_bit(i, gpu->event_bitmap, ETNA_NR_EVENTS) - complete(&gpu->event_free); - bitmap_zero(gpu->event_bitmap, ETNA_NR_EVENTS); - spin_unlock(&gpu->event_spinlock); - - etnaviv_gpu_hw_init(gpu); - - mutex_unlock(&gpu->lock); - pm_runtime_mark_last_busy(gpu->dev); -pm_put: - pm_runtime_put_autosuspend(gpu->dev); -} - /* fence object management */ struct etnaviv_fence { struct etnaviv_gpu *gpu; @@ -1183,20 +1154,22 @@ static int event_alloc(struct etnaviv_gpu *gpu, unsigned nr_events, unsigned int *events) { unsigned long timeout = msecs_to_jiffies(10 * 10000); - unsigned i, acquired = 0; + unsigned i, acquired = 0, rpm_count = 0; + int ret; for (i = 0; i < nr_events; i++) { - unsigned long ret; + unsigned long remaining; - ret = wait_for_completion_timeout(&gpu->event_free, timeout); + remaining = wait_for_completion_timeout(&gpu->event_free, timeout); - if (!ret) { + if (!remaining) { dev_err(gpu->dev, "wait_for_completion_timeout failed"); + ret = -EBUSY; goto out; } acquired++; - timeout = ret; + timeout = remaining; } spin_lock(&gpu->event_spinlock); @@ -1211,13 +1184,23 @@ static int event_alloc(struct etnaviv_gpu *gpu, unsigned nr_events, spin_unlock(&gpu->event_spinlock); + for (i = 0; i < nr_events; i++) { + ret = pm_runtime_resume_and_get(gpu->dev); + if (ret) + goto out_rpm; + rpm_count++; + } + return 0; +out_rpm: + for (i = 0; i < rpm_count; i++) + pm_runtime_put_autosuspend(gpu->dev); out: for (i = 0; i < acquired; i++) complete(&gpu->event_free); - return -EBUSY; + return ret; } static void event_free(struct etnaviv_gpu *gpu, unsigned int event) @@ -1229,6 +1212,8 @@ static void event_free(struct etnaviv_gpu *gpu, unsigned int event) clear_bit(event, gpu->event_bitmap); complete(&gpu->event_free); } + + pm_runtime_put_autosuspend(gpu->dev); } /* @@ -1371,15 +1356,6 @@ struct dma_fence *etnaviv_gpu_submit(struct etnaviv_gem_submit *submit) unsigned int i, nr_events = 1, event[3]; int ret; - if (!submit->runtime_resumed) { - ret = pm_runtime_get_sync(gpu->dev); - if (ret < 0) { - pm_runtime_put_noidle(gpu->dev); - return NULL; - } - submit->runtime_resumed = true; - } - /* * if there are performance monitor requests we need to have * - a sync point to re-configure gpu and process ETNA_PM_PROCESS_PRE @@ -1407,7 +1383,7 @@ struct dma_fence *etnaviv_gpu_submit(struct etnaviv_gem_submit *submit) goto out_unlock; } - if (!gpu->fe_running) + if (gpu->state == ETNA_GPU_STATE_INITIALIZED) etnaviv_gpu_start_fe_idleloop(gpu, submit->mmu_context); if (submit->prev_mmu_context) @@ -1454,6 +1430,49 @@ static void sync_point_worker(struct work_struct *work) etnaviv_gpu_start_fe(gpu, addr + 2, 2); } +void etnaviv_gpu_recover_hang(struct etnaviv_gem_submit *submit) +{ + struct etnaviv_gpu *gpu = submit->gpu; + char *comm = NULL, *cmd = NULL; + struct task_struct *task; + unsigned int i; + + dev_err(gpu->dev, "recover hung GPU!\n"); + + task = get_pid_task(submit->pid, PIDTYPE_PID); + if (task) { + comm = kstrdup(task->comm, GFP_KERNEL); + cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); + put_task_struct(task); + } + + if (comm && cmd) + dev_err(gpu->dev, "offending task: %s (%s)\n", comm, cmd); + + kfree(cmd); + kfree(comm); + + if (pm_runtime_get_sync(gpu->dev) < 0) + goto pm_put; + + mutex_lock(&gpu->lock); + + etnaviv_hw_reset(gpu); + + /* complete all events, the GPU won't do it after the reset */ + spin_lock(&gpu->event_spinlock); + for_each_set_bit(i, gpu->event_bitmap, ETNA_NR_EVENTS) + event_free(gpu, i); + spin_unlock(&gpu->event_spinlock); + + etnaviv_gpu_hw_init(gpu); + + mutex_unlock(&gpu->lock); + pm_runtime_mark_last_busy(gpu->dev); +pm_put: + pm_runtime_put_autosuspend(gpu->dev); +} + static void dump_mmu_fault(struct etnaviv_gpu *gpu) { static const char *fault_reasons[] = { @@ -1520,6 +1539,8 @@ static irqreturn_t irq_handler(int irq, void *data) if (intr & VIVS_HI_INTR_ACKNOWLEDGE_MMU_EXCEPTION) { dump_mmu_fault(gpu); + gpu->state = ETNA_GPU_STATE_FAULT; + drm_sched_fault(&gpu->sched); intr &= ~VIVS_HI_INTR_ACKNOWLEDGE_MMU_EXCEPTION; } @@ -1628,9 +1649,9 @@ int etnaviv_gpu_wait_idle(struct etnaviv_gpu *gpu, unsigned int timeout_ms) } while (1); } -static int etnaviv_gpu_hw_suspend(struct etnaviv_gpu *gpu) +static void etnaviv_gpu_hw_suspend(struct etnaviv_gpu *gpu) { - if (gpu->initialized && gpu->fe_running) { + if (gpu->state == ETNA_GPU_STATE_RUNNING) { /* Replace the last WAIT with END */ mutex_lock(&gpu->lock); etnaviv_buffer_end(gpu); @@ -1643,12 +1664,10 @@ static int etnaviv_gpu_hw_suspend(struct etnaviv_gpu *gpu) */ etnaviv_gpu_wait_idle(gpu, 100); - gpu->fe_running = false; + gpu->state = ETNA_GPU_STATE_INITIALIZED; } gpu->exec_state = -1; - - return etnaviv_gpu_clk_disable(gpu); } static int etnaviv_gpu_hw_resume(struct etnaviv_gpu *gpu) @@ -1733,13 +1752,11 @@ static int etnaviv_gpu_bind(struct device *dev, struct device *master, if (ret) goto out_workqueue; - if (IS_ENABLED(CONFIG_PM)) - ret = pm_runtime_get_sync(gpu->dev); - else + if (!IS_ENABLED(CONFIG_PM)) { ret = etnaviv_gpu_clk_enable(gpu); - if (ret < 0) - goto out_sched; - + if (ret < 0) + goto out_sched; + } gpu->drm = drm; gpu->fence_context = dma_fence_context_alloc(1); @@ -1751,9 +1768,6 @@ static int etnaviv_gpu_bind(struct device *dev, struct device *master, priv->gpu[priv->num_gpus++] = gpu; - pm_runtime_mark_last_busy(gpu->dev); - pm_runtime_put_autosuspend(gpu->dev); - return 0; out_sched: @@ -1785,16 +1799,14 @@ static void etnaviv_gpu_unbind(struct device *dev, struct device *master, pm_runtime_put_sync_suspend(gpu->dev); } else { etnaviv_gpu_hw_suspend(gpu); + etnaviv_gpu_clk_disable(gpu); } if (gpu->mmu_context) etnaviv_iommu_context_put(gpu->mmu_context); - if (gpu->initialized) { - etnaviv_cmdbuf_free(&gpu->buffer); - etnaviv_iommu_global_fini(gpu); - gpu->initialized = false; - } + etnaviv_cmdbuf_free(&gpu->buffer); + etnaviv_iommu_global_fini(gpu); gpu->drm = NULL; xa_destroy(&gpu->user_fences); @@ -1918,7 +1930,11 @@ static int etnaviv_gpu_rpm_suspend(struct device *dev) return -EBUSY; } - return etnaviv_gpu_hw_suspend(gpu); + etnaviv_gpu_hw_suspend(gpu); + + gpu->state = ETNA_GPU_STATE_IDENTIFIED; + + return etnaviv_gpu_clk_disable(gpu); } static int etnaviv_gpu_rpm_resume(struct device *dev) @@ -1931,7 +1947,7 @@ static int etnaviv_gpu_rpm_resume(struct device *dev) return ret; /* Re-initialise the basic hardware state */ - if (gpu->drm && gpu->initialized) { + if (gpu->state == ETNA_GPU_STATE_IDENTIFIED) { ret = etnaviv_gpu_hw_resume(gpu); if (ret) { etnaviv_gpu_clk_disable(gpu); diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h index 98c6f9c320fc..197e0037732e 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h @@ -95,6 +95,15 @@ struct clk; #define ETNA_NR_EVENTS 30 +enum etnaviv_gpu_state { + ETNA_GPU_STATE_UNKNOWN = 0, + ETNA_GPU_STATE_IDENTIFIED, + ETNA_GPU_STATE_RESET, + ETNA_GPU_STATE_INITIALIZED, + ETNA_GPU_STATE_RUNNING, + ETNA_GPU_STATE_FAULT, +}; + struct etnaviv_gpu { struct drm_device *drm; struct thermal_cooling_device *cooling; @@ -105,8 +114,7 @@ struct etnaviv_gpu { struct workqueue_struct *wq; struct mutex sched_lock; struct drm_gpu_scheduler sched; - bool initialized; - bool fe_running; + enum etnaviv_gpu_state state; /* 'ring'-buffer: */ struct etnaviv_cmdbuf buffer; @@ -150,6 +158,7 @@ struct etnaviv_gpu { struct clk *clk_shader; unsigned int freq_scale; + unsigned int fe_waitcycles; unsigned long base_rate_core; unsigned long base_rate_shader; }; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c index 2e63afa6c798..67201242438b 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_hwdb.c @@ -39,6 +39,37 @@ static const struct etnaviv_chip_identity etnaviv_chip_identities[] = { .minor_features11 = 0x0, }, { + .model = 0x520, + .revision = 0x5341, + .product_id = 0x5202, + .customer_id = 0x204, + .eco_id = 0, + .stream_count = 1, + .register_max = 64, + .thread_count = 256, + .shader_core_count = 1, + .vertex_cache_size = 8, + .vertex_output_buffer_size = 512, + .pixel_pipes = 1, + .instruction_count = 256, + .num_constants = 168, + .buffer_size = 0, + .varyings_count = 8, + .features = 0xe02c7eca, + .minor_features0 = 0xe9399eff, + .minor_features1 = 0xfe1fb2db, + .minor_features2 = 0xcedf0080, + .minor_features3 = 0x10800005, + .minor_features4 = 0x20000000, + .minor_features5 = 0x00020880, + .minor_features6 = 0x00000000, + .minor_features7 = 0x00001000, + .minor_features8 = 0x00000000, + .minor_features9 = 0x00000000, + .minor_features10 = 0x00000000, + .minor_features11 = 0x00000000, + }, + { .model = 0x7000, .revision = 0x6202, .product_id = 0x70003, @@ -197,6 +228,38 @@ static const struct etnaviv_chip_identity etnaviv_chip_identities[] = { .minor_features10 = 0x108048c0, .minor_features11 = 0x00000010, }, + { + .model = 0x8000, + .revision = 0x8002, + .product_id = 0x5080009, + .customer_id = 0x9f, + .eco_id = 0x6000000, + .stream_count = 8, + .register_max = 64, + .thread_count = 256, + .shader_core_count = 1, + .nn_core_count = 6, + .vertex_cache_size = 16, + .vertex_output_buffer_size = 1024, + .pixel_pipes = 1, + .instruction_count = 512, + .num_constants = 320, + .buffer_size = 0, + .varyings_count = 16, + .features = 0xe0287cac, + .minor_features0 = 0xc1799eff, + .minor_features1 = 0xfefbfadb, + .minor_features2 = 0xeb9d6fbf, + .minor_features3 = 0xedfffced, + .minor_features4 = 0xd30dafc7, + .minor_features5 = 0x7b5ac333, + .minor_features6 = 0xfc8ee200, + .minor_features7 = 0x03fffa6f, + .minor_features8 = 0x00fe0ef0, + .minor_features9 = 0x0088003c, + .minor_features10 = 0x108048c0, + .minor_features11 = 0x00000010, + }, }; bool etnaviv_fill_identity_from_hwdb(struct etnaviv_gpu *gpu) diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c index 67bdce5326c6..4fa72567183a 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c @@ -553,6 +553,9 @@ void etnaviv_iommu_global_fini(struct etnaviv_gpu *gpu) struct etnaviv_drm_private *priv = gpu->drm->dev_private; struct etnaviv_iommu_global *global = priv->mmu_global; + if (!global) + return; + if (--global->use > 0) return; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c b/drivers/gpu/drm/etnaviv/etnaviv_sched.c index 1ae87dfd19c4..345fec6cb1a4 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c @@ -55,8 +55,9 @@ static enum drm_gpu_sched_stat etnaviv_sched_timedout_job(struct drm_sched_job */ dma_addr = gpu_read(gpu, VIVS_FE_DMA_ADDRESS); change = dma_addr - gpu->hangcheck_dma_addr; - if (gpu->completed_fence != gpu->hangcheck_fence || - change < 0 || change > 16) { + if (gpu->state == ETNA_GPU_STATE_RUNNING && + (gpu->completed_fence != gpu->hangcheck_fence || + change < 0 || change > 16)) { gpu->hangcheck_dma_addr = dma_addr; gpu->hangcheck_fence = gpu->completed_fence; goto out_no_timeout; diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index 4153f302de7c..d19e796c2061 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -39,13 +39,12 @@ static void exynos_drm_crtc_atomic_disable(struct drm_crtc *crtc, if (exynos_crtc->ops->atomic_disable) exynos_crtc->ops->atomic_disable(exynos_crtc); + spin_lock_irq(&crtc->dev->event_lock); if (crtc->state->event && !crtc->state->active) { - spin_lock_irq(&crtc->dev->event_lock); drm_crtc_send_vblank_event(crtc, crtc->state->event); - spin_unlock_irq(&crtc->dev->event_lock); - crtc->state->event = NULL; } + spin_unlock_irq(&crtc->dev->event_lock); } static int exynos_crtc_atomic_check(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index 964dceb28c1e..34cdabc30b4f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -1426,6 +1426,6 @@ struct platform_driver gsc_driver = { .name = "exynos-drm-gsc", .owner = THIS_MODULE, .pm = &gsc_pm_ops, - .of_match_table = of_match_ptr(exynos_drm_gsc_of_match), + .of_match_table = exynos_drm_gsc_of_match, }, }; diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 789dce9e2608..79f65eff6bb2 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -23,6 +23,11 @@ subdir-ccflags-y += $(call cc-option, -Wunused-but-set-variable) subdir-ccflags-y += $(call cc-disable-warning, frame-address) subdir-ccflags-$(CONFIG_DRM_I915_WERROR) += -Werror +# Fine grained warnings disable +CFLAGS_i915_pci.o = $(call cc-disable-warning, override-init) +CFLAGS_display/intel_display_device.o = $(call cc-disable-warning, override-init) +CFLAGS_display/intel_fbdev.o = $(call cc-disable-warning, override-init) + subdir-ccflags-y += -I$(srctree)/$(src) # Please keep these build lists sorted! @@ -127,6 +132,7 @@ gt-y += \ gt/intel_sseu.o \ gt/intel_sseu_debugfs.o \ gt/intel_timeline.o \ + gt/intel_tlb.o \ gt/intel_wopcm.o \ gt/intel_workarounds.o \ gt/shmem_utils.o \ @@ -192,7 +198,8 @@ i915-y += \ gt/uc/intel_gsc_fw.o \ gt/uc/intel_gsc_proxy.o \ gt/uc/intel_gsc_uc.o \ - gt/uc/intel_gsc_uc_heci_cmd_submit.o\ + gt/uc/intel_gsc_uc_debugfs.o \ + gt/uc/intel_gsc_uc_heci_cmd_submit.o \ gt/uc/intel_guc.o \ gt/uc/intel_guc_ads.o \ gt/uc/intel_guc_capture.o \ diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c index 112d91d81fdc..4c7187f7913e 100644 --- a/drivers/gpu/drm/i915/display/g4x_dp.c +++ b/drivers/gpu/drm/i915/display/g4x_dp.c @@ -1259,6 +1259,9 @@ bool g4x_dp_init(struct drm_i915_private *dev_priv, struct drm_encoder *encoder; struct intel_connector *intel_connector; + if (!assert_port_valid(dev_priv, port)) + return false; + devdata = intel_bios_encoder_data_lookup(dev_priv, port); /* FIXME bail? */ @@ -1270,6 +1273,8 @@ bool g4x_dp_init(struct drm_i915_private *dev_priv, if (!dig_port) return false; + dig_port->aux_ch = AUX_CH_NONE; + intel_connector = intel_connector_alloc(); if (!intel_connector) goto err_connector_alloc; @@ -1373,6 +1378,9 @@ bool g4x_dp_init(struct drm_i915_private *dev_priv, intel_infoframe_init(dig_port); dig_port->aux_ch = intel_dp_aux_ch(intel_encoder); + if (dig_port->aux_ch == AUX_CH_NONE) + goto err_init_connector; + if (!intel_dp_init_connector(dig_port, intel_connector)) goto err_init_connector; diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c index 5c187e6e0472..634b14116d9d 100644 --- a/drivers/gpu/drm/i915/display/g4x_hdmi.c +++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c @@ -659,6 +659,20 @@ int g4x_hdmi_connector_atomic_check(struct drm_connector *connector, return ret; } +static bool is_hdmi_port_valid(struct drm_i915_private *i915, enum port port) +{ + if (IS_G4X(i915) || IS_VALLEYVIEW(i915)) + return port == PORT_B || port == PORT_C; + else + return port == PORT_B || port == PORT_C || port == PORT_D; +} + +static bool assert_hdmi_port_valid(struct drm_i915_private *i915, enum port port) +{ + return !drm_WARN(&i915->drm, !is_hdmi_port_valid(i915, port), + "Platform does not support HDMI %c\n", port_name(port)); +} + void g4x_hdmi_init(struct drm_i915_private *dev_priv, i915_reg_t hdmi_reg, enum port port) { @@ -667,6 +681,12 @@ void g4x_hdmi_init(struct drm_i915_private *dev_priv, struct intel_encoder *intel_encoder; struct intel_connector *intel_connector; + if (!assert_port_valid(dev_priv, port)) + return; + + if (!assert_hdmi_port_valid(dev_priv, port)) + return; + devdata = intel_bios_encoder_data_lookup(dev_priv, port); /* FIXME bail? */ @@ -678,6 +698,8 @@ void g4x_hdmi_init(struct drm_i915_private *dev_priv, if (!dig_port) return; + dig_port->aux_ch = AUX_CH_NONE; + intel_connector = intel_connector_alloc(); if (!intel_connector) { kfree(dig_port); @@ -753,6 +775,5 @@ void g4x_hdmi_init(struct drm_i915_private *dev_priv, intel_infoframe_init(dig_port); - dig_port->aux_ch = intel_dp_aux_ch(intel_encoder); intel_hdmi_init_connector(dig_port, intel_connector); } diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c index c133928a0655..ad6488e9c2b2 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.c +++ b/drivers/gpu/drm/i915/display/icl_dsi.c @@ -444,7 +444,8 @@ static void gen11_dsi_config_phy_lanes_sequence(struct intel_encoder *encoder) intel_de_write(dev_priv, ICL_PORT_TX_DW2_GRP(phy), tmp); /* For EHL, TGL, set latency optimization for PCS_DW1 lanes */ - if (IS_JSL_EHL(dev_priv) || (DISPLAY_VER(dev_priv) >= 12)) { + if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv) || + (DISPLAY_VER(dev_priv) >= 12)) { intel_de_rmw(dev_priv, ICL_PORT_PCS_DW1_AUX(phy), LATENCY_OPTIM_MASK, LATENCY_OPTIM_VAL(0)); @@ -528,31 +529,16 @@ gen11_dsi_setup_dphy_timings(struct intel_encoder *encoder, enum port port; enum phy phy; - /* Program T-INIT master registers */ - for_each_dsi_port(port, intel_dsi->ports) - intel_de_rmw(dev_priv, ICL_DSI_T_INIT_MASTER(port), - DSI_T_INIT_MASTER_MASK, intel_dsi->init_count); - /* Program DPHY clock lanes timings */ - for_each_dsi_port(port, intel_dsi->ports) { + for_each_dsi_port(port, intel_dsi->ports) intel_de_write(dev_priv, DPHY_CLK_TIMING_PARAM(port), intel_dsi->dphy_reg); - /* shadow register inside display core */ - intel_de_write(dev_priv, DSI_CLK_TIMING_PARAM(port), - intel_dsi->dphy_reg); - } - /* Program DPHY data lanes timings */ - for_each_dsi_port(port, intel_dsi->ports) { + for_each_dsi_port(port, intel_dsi->ports) intel_de_write(dev_priv, DPHY_DATA_TIMING_PARAM(port), intel_dsi->dphy_data_lane_reg); - /* shadow register inside display core */ - intel_de_write(dev_priv, DSI_DATA_TIMING_PARAM(port), - intel_dsi->dphy_data_lane_reg); - } - /* * If DSI link operating at or below an 800 MHz, * TA_SURE should be override and programmed to @@ -561,26 +547,55 @@ gen11_dsi_setup_dphy_timings(struct intel_encoder *encoder, */ if (DISPLAY_VER(dev_priv) == 11) { if (afe_clk(encoder, crtc_state) <= 800000) { - for_each_dsi_port(port, intel_dsi->ports) { + for_each_dsi_port(port, intel_dsi->ports) intel_de_rmw(dev_priv, DPHY_TA_TIMING_PARAM(port), TA_SURE_MASK, TA_SURE_OVERRIDE | TA_SURE(0)); - - /* shadow register inside display core */ - intel_de_rmw(dev_priv, DSI_TA_TIMING_PARAM(port), - TA_SURE_MASK, - TA_SURE_OVERRIDE | TA_SURE(0)); - } } } - if (IS_JSL_EHL(dev_priv)) { + if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) { for_each_dsi_phy(phy, intel_dsi->phys) intel_de_rmw(dev_priv, ICL_DPHY_CHKN(phy), 0, ICL_DPHY_CHKN_AFE_OVER_PPI_STRAP); } } +static void +gen11_dsi_setup_timings(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); + enum port port; + + /* Program T-INIT master registers */ + for_each_dsi_port(port, intel_dsi->ports) + intel_de_rmw(dev_priv, ICL_DSI_T_INIT_MASTER(port), + DSI_T_INIT_MASTER_MASK, intel_dsi->init_count); + + /* shadow register inside display core */ + for_each_dsi_port(port, intel_dsi->ports) + intel_de_write(dev_priv, DSI_CLK_TIMING_PARAM(port), + intel_dsi->dphy_reg); + + /* shadow register inside display core */ + for_each_dsi_port(port, intel_dsi->ports) + intel_de_write(dev_priv, DSI_DATA_TIMING_PARAM(port), + intel_dsi->dphy_data_lane_reg); + + /* shadow register inside display core */ + if (DISPLAY_VER(dev_priv) == 11) { + if (afe_clk(encoder, crtc_state) <= 800000) { + for_each_dsi_port(port, intel_dsi->ports) { + intel_de_rmw(dev_priv, DSI_TA_TIMING_PARAM(port), + TA_SURE_MASK, + TA_SURE_OVERRIDE | TA_SURE(0)); + } + } + } +} + static void gen11_dsi_gate_clocks(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); @@ -1090,11 +1105,15 @@ gen11_dsi_enable_port_and_phy(struct intel_encoder *encoder, /* step 4c: configure voltage swing and skew */ gen11_dsi_voltage_swing_program_seq(encoder); + /* setup D-PHY timings */ + gen11_dsi_setup_dphy_timings(encoder, crtc_state); + /* enable DDI buffer */ gen11_dsi_enable_ddi_buffer(encoder); - /* setup D-PHY timings */ - gen11_dsi_setup_dphy_timings(encoder, crtc_state); + gen11_dsi_gate_clocks(encoder); + + gen11_dsi_setup_timings(encoder, crtc_state); /* Since transcoder is configured to take events from GPIO */ gen11_dsi_config_util_pin(encoder, true); @@ -1104,9 +1123,6 @@ gen11_dsi_enable_port_and_phy(struct intel_encoder *encoder, /* Step (4h, 4i, 4j, 4k): Configure transcoder */ gen11_dsi_configure_transcoder(encoder, crtc_state); - - /* Step 4l: Gate DDI clocks */ - gen11_dsi_gate_clocks(encoder); } static void gen11_dsi_powerup_panel(struct intel_encoder *encoder) @@ -1138,12 +1154,7 @@ static void gen11_dsi_powerup_panel(struct intel_encoder *encoder) "error setting max return pkt size%d\n", tmp); } - /* panel power on related mipi dsi vbt sequences */ - intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_ON); - msleep(intel_dsi->panel_on_delay); - intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DEASSERT_RESET); intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_INIT_OTP); - intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DISPLAY_ON); /* ensure all panel commands dispatched before enabling transcoder */ wait_for_cmds_dispatched_to_panel(encoder); @@ -1154,6 +1165,14 @@ static void gen11_dsi_pre_pll_enable(struct intel_atomic_state *state, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { + struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); + + intel_dsi_wait_panel_power_cycle(intel_dsi); + + intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_ON); + msleep(intel_dsi->panel_on_delay); + intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DEASSERT_RESET); + /* step2: enable IO power */ gen11_dsi_enable_io_power(encoder); @@ -1225,9 +1244,7 @@ static void gen11_dsi_enable(struct intel_atomic_state *state, const struct drm_connector_state *conn_state) { struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); - struct intel_crtc *crtc = to_intel_crtc(conn_state->crtc); - - drm_WARN_ON(state->base.dev, crtc_state->has_pch_encoder); + struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); /* Wa_1409054076:icl,jsl,ehl */ icl_apply_kvmr_pipe_a_wa(encoder, crtc->pipe, true); @@ -1238,6 +1255,8 @@ static void gen11_dsi_enable(struct intel_atomic_state *state, /* step6d: enable dsi transcoder */ gen11_dsi_enable_transcoder(encoder); + intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DISPLAY_ON); + /* step7: enable backlight */ intel_backlight_enable(crtc_state, conn_state); intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_BACKLIGHT_ON); @@ -1271,8 +1290,6 @@ static void gen11_dsi_powerdown_panel(struct intel_encoder *encoder) struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_DISPLAY_OFF); - intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_ASSERT_RESET); - intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_OFF); /* ensure cmds dispatched to panel */ wait_for_cmds_dispatched_to_panel(encoder); @@ -1373,11 +1390,21 @@ static void gen11_dsi_disable(struct intel_atomic_state *state, const struct drm_connector_state *old_conn_state) { struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); - struct intel_crtc *crtc = to_intel_crtc(old_conn_state->crtc); /* step1: turn off backlight */ intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_BACKLIGHT_OFF); intel_backlight_disable(old_conn_state); +} + +static void gen11_dsi_post_disable(struct intel_atomic_state *state, + struct intel_encoder *encoder, + const struct intel_crtc_state *old_crtc_state, + const struct drm_connector_state *old_conn_state) +{ + struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); + struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->uapi.crtc); + + intel_crtc_vblank_off(old_crtc_state); /* step2d,e: disable transcoder and wait */ gen11_dsi_disable_transcoder(encoder); @@ -1391,6 +1418,9 @@ static void gen11_dsi_disable(struct intel_atomic_state *state, /* step2h,i,j: deconfig trancoder */ gen11_dsi_deconfigure_trancoder(encoder); + intel_dsc_disable(old_crtc_state); + skl_scaler_disable(old_crtc_state); + /* step3: disable port */ gen11_dsi_disable_port(encoder); @@ -1398,18 +1428,13 @@ static void gen11_dsi_disable(struct intel_atomic_state *state, /* step4: disable IO power */ gen11_dsi_disable_io_power(encoder); -} -static void gen11_dsi_post_disable(struct intel_atomic_state *state, - struct intel_encoder *encoder, - const struct intel_crtc_state *old_crtc_state, - const struct drm_connector_state *old_conn_state) -{ - intel_crtc_vblank_off(old_crtc_state); + intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_ASSERT_RESET); - intel_dsc_disable(old_crtc_state); + msleep(intel_dsi->panel_off_delay); + intel_dsi_vbt_exec_sequence(intel_dsi, MIPI_SEQ_POWER_OFF); - skl_scaler_disable(old_crtc_state); + intel_dsi->panel_power_off_time = ktime_get_boottime(); } static enum drm_mode_status gen11_dsi_mode_valid(struct drm_connector *connector, @@ -1909,7 +1934,8 @@ static void icl_dsi_add_properties(struct intel_connector *connector) fixed_mode->vdisplay); } -void icl_dsi_init(struct drm_i915_private *dev_priv) +void icl_dsi_init(struct drm_i915_private *dev_priv, + const struct intel_bios_encoder_data *devdata) { struct intel_dsi *intel_dsi; struct intel_encoder *encoder; @@ -1917,7 +1943,8 @@ void icl_dsi_init(struct drm_i915_private *dev_priv) struct drm_connector *connector; enum port port; - if (!intel_bios_is_dsi_present(dev_priv, &port)) + port = intel_bios_encoder_port(devdata); + if (port == PORT_NONE) return; intel_dsi = kzalloc(sizeof(*intel_dsi), GFP_KERNEL); @@ -1934,6 +1961,8 @@ void icl_dsi_init(struct drm_i915_private *dev_priv) intel_dsi->attached_connector = intel_connector; connector = &intel_connector->base; + encoder->devdata = devdata; + /* register DSI encoder with DRM subsystem */ drm_encoder_init(&dev_priv->drm, &encoder->base, &gen11_dsi_encoder_funcs, DRM_MODE_ENCODER_DSI, "DSI %c", port_name(port)); @@ -1957,6 +1986,7 @@ void icl_dsi_init(struct drm_i915_private *dev_priv) encoder->get_power_domains = gen11_dsi_get_power_domains; encoder->disable_clock = gen11_dsi_gate_clocks; encoder->is_clock_enabled = gen11_dsi_is_clock_enabled; + encoder->shutdown = intel_dsi_shutdown; /* register DSI connector with DRM subsystem */ drm_connector_init(&dev_priv->drm, connector, &gen11_dsi_connector_funcs, @@ -1968,7 +1998,8 @@ void icl_dsi_init(struct drm_i915_private *dev_priv) /* attach connector to encoder */ intel_connector_attach_encoder(intel_connector, encoder); - encoder->devdata = intel_bios_encoder_data_lookup(dev_priv, port); + intel_dsi->panel_power_off_time = ktime_get_boottime(); + intel_bios_init_panel_late(dev_priv, &intel_connector->panel, encoder->devdata, NULL); mutex_lock(&dev_priv->drm.mode_config.mutex); diff --git a/drivers/gpu/drm/i915/display/icl_dsi.h b/drivers/gpu/drm/i915/display/icl_dsi.h index b4861b56b5b2..43fa7d72eeb1 100644 --- a/drivers/gpu/drm/i915/display/icl_dsi.h +++ b/drivers/gpu/drm/i915/display/icl_dsi.h @@ -7,9 +7,11 @@ #define __ICL_DSI_H__ struct drm_i915_private; +struct intel_bios_encoder_data; struct intel_crtc_state; -void icl_dsi_init(struct drm_i915_private *i915); +void icl_dsi_init(struct drm_i915_private *dev_priv, + const struct intel_bios_encoder_data *devdata); void icl_dsi_frame_update(struct intel_crtc_state *crtc_state); #endif /* __ICL_DSI_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c index 7d9578ebae55..60a492e186ab 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c @@ -212,6 +212,7 @@ intel_plane_relative_data_rate(const struct intel_crtc_state *crtc_state, struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane); const struct drm_framebuffer *fb = plane_state->hw.fb; int width, height; + unsigned int rel_data_rate; if (plane->id == PLANE_CURSOR) return 0; @@ -241,7 +242,11 @@ intel_plane_relative_data_rate(const struct intel_crtc_state *crtc_state, height /= 2; } - return width * height * fb->format->cpp[color_plane]; + rel_data_rate = width * height * fb->format->cpp[color_plane]; + + return intel_adjusted_rate(&plane_state->uapi.src, + &plane_state->uapi.dst, + rel_data_rate); } int intel_plane_calc_min_cdclk(struct intel_atomic_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index 34a397adbd6b..f735b035436c 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -2230,122 +2230,6 @@ static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin) return 0; } -static enum port get_port_by_ddc_pin(struct drm_i915_private *i915, u8 ddc_pin) -{ - enum port port; - - if (!ddc_pin) - return PORT_NONE; - - for_each_port(port) { - const struct intel_bios_encoder_data *devdata = - i915->display.vbt.ports[port]; - - if (devdata && ddc_pin == devdata->child.ddc_pin) - return port; - } - - return PORT_NONE; -} - -static void sanitize_ddc_pin(struct intel_bios_encoder_data *devdata, - enum port port) -{ - struct drm_i915_private *i915 = devdata->i915; - struct child_device_config *child; - u8 mapped_ddc_pin; - enum port p; - - if (!devdata->child.ddc_pin) - return; - - mapped_ddc_pin = map_ddc_pin(i915, devdata->child.ddc_pin); - if (!intel_gmbus_is_valid_pin(i915, mapped_ddc_pin)) { - drm_dbg_kms(&i915->drm, - "Port %c has invalid DDC pin %d, " - "sticking to defaults\n", - port_name(port), mapped_ddc_pin); - devdata->child.ddc_pin = 0; - return; - } - - p = get_port_by_ddc_pin(i915, devdata->child.ddc_pin); - if (p == PORT_NONE) - return; - - drm_dbg_kms(&i915->drm, - "port %c trying to use the same DDC pin (0x%x) as port %c, " - "disabling port %c DVI/HDMI support\n", - port_name(port), mapped_ddc_pin, - port_name(p), port_name(p)); - - /* - * If we have multiple ports supposedly sharing the pin, then dvi/hdmi - * couldn't exist on the shared port. Otherwise they share the same ddc - * pin and system couldn't communicate with them separately. - * - * Give inverse child device order the priority, last one wins. Yes, - * there are real machines (eg. Asrock B250M-HDV) where VBT has both - * port A and port E with the same AUX ch and we must pick port E :( - */ - child = &i915->display.vbt.ports[p]->child; - - child->device_type &= ~DEVICE_TYPE_TMDS_DVI_SIGNALING; - child->device_type |= DEVICE_TYPE_NOT_HDMI_OUTPUT; - - child->ddc_pin = 0; -} - -static enum port get_port_by_aux_ch(struct drm_i915_private *i915, u8 aux_ch) -{ - enum port port; - - if (!aux_ch) - return PORT_NONE; - - for_each_port(port) { - const struct intel_bios_encoder_data *devdata = - i915->display.vbt.ports[port]; - - if (devdata && aux_ch == devdata->child.aux_channel) - return port; - } - - return PORT_NONE; -} - -static void sanitize_aux_ch(struct intel_bios_encoder_data *devdata, - enum port port) -{ - struct drm_i915_private *i915 = devdata->i915; - struct child_device_config *child; - enum port p; - - p = get_port_by_aux_ch(i915, devdata->child.aux_channel); - if (p == PORT_NONE) - return; - - drm_dbg_kms(&i915->drm, - "port %c trying to use the same AUX CH (0x%x) as port %c, " - "disabling port %c DP support\n", - port_name(port), devdata->child.aux_channel, - port_name(p), port_name(p)); - - /* - * If we have multiple ports supposedly sharing the aux channel, then DP - * couldn't exist on the shared port. Otherwise they share the same aux - * channel and system couldn't communicate with them separately. - * - * Give inverse child device order the priority, last one wins. Yes, - * there are real machines (eg. Asrock B250M-HDV) where VBT has both - * port A and port E with the same AUX ch and we must pick port E :( - */ - child = &i915->display.vbt.ports[p]->child; - - child->device_type &= ~DEVICE_TYPE_DISPLAYPORT_OUTPUT; - child->aux_channel = 0; -} - static u8 dvo_port_type(u8 dvo_port) { switch (dvo_port) { @@ -2490,6 +2374,19 @@ dsi_dvo_port_to_port(struct drm_i915_private *i915, u8 dvo_port) } } +enum port intel_bios_encoder_port(const struct intel_bios_encoder_data *devdata) +{ + struct drm_i915_private *i915 = devdata->i915; + const struct child_device_config *child = &devdata->child; + enum port port; + + port = dvo_port_to_port(i915, child->dvo_port); + if (port == PORT_NONE && DISPLAY_VER(i915) >= 11) + port = dsi_dvo_port_to_port(i915, child->dvo_port); + + return port; +} + static int parse_bdb_230_dp_max_link_rate(const int vbt_max_link_rate) { switch (vbt_max_link_rate) { @@ -2600,7 +2497,7 @@ intel_bios_encoder_supports_edp(const struct intel_bios_encoder_data *devdata) devdata->child.device_type & DEVICE_TYPE_INTERNAL_CONNECTOR; } -static bool +bool intel_bios_encoder_supports_dsi(const struct intel_bios_encoder_data *devdata) { return devdata->child.device_type & DEVICE_TYPE_MIPI_OUTPUT; @@ -2615,7 +2512,8 @@ intel_bios_encoder_is_lspcon(const struct intel_bios_encoder_data *devdata) /* This is an index in the HDMI/DVI DDI buffer translation table, or -1 */ int intel_bios_hdmi_level_shift(const struct intel_bios_encoder_data *devdata) { - if (!devdata || devdata->i915->display.vbt.version < 158) + if (!devdata || devdata->i915->display.vbt.version < 158 || + DISPLAY_VER(devdata->i915) >= 14) return -1; return devdata->child.hdmi_level_shifter_value; @@ -2658,13 +2556,17 @@ static bool is_port_valid(struct drm_i915_private *i915, enum port port) return true; } -static void print_ddi_port(const struct intel_bios_encoder_data *devdata, - enum port port) +static void print_ddi_port(const struct intel_bios_encoder_data *devdata) { struct drm_i915_private *i915 = devdata->i915; const struct child_device_config *child = &devdata->child; bool is_dvi, is_hdmi, is_dp, is_edp, is_dsi, is_crt, supports_typec_usb, supports_tbt; int dp_boost_level, dp_max_link_rate, hdmi_boost_level, hdmi_level_shift, max_tmds_clock; + enum port port; + + port = intel_bios_encoder_port(devdata); + if (port == PORT_NONE) + return; is_dvi = intel_bios_encoder_supports_dvi(devdata); is_dp = intel_bios_encoder_supports_dp(devdata); @@ -2728,12 +2630,9 @@ static void print_ddi_port(const struct intel_bios_encoder_data *devdata, static void parse_ddi_port(struct intel_bios_encoder_data *devdata) { struct drm_i915_private *i915 = devdata->i915; - const struct child_device_config *child = &devdata->child; enum port port; - port = dvo_port_to_port(i915, child->dvo_port); - if (port == PORT_NONE && DISPLAY_VER(i915) >= 11) - port = dsi_dvo_port_to_port(i915, child->dvo_port); + port = intel_bios_encoder_port(devdata); if (port == PORT_NONE) return; @@ -2744,22 +2643,7 @@ static void parse_ddi_port(struct intel_bios_encoder_data *devdata) return; } - if (i915->display.vbt.ports[port]) { - drm_dbg_kms(&i915->drm, - "More than one child device for port %c in VBT, using the first.\n", - port_name(port)); - return; - } - sanitize_device_type(devdata, port); - - if (intel_bios_encoder_supports_dvi(devdata)) - sanitize_ddc_pin(devdata, port); - - if (intel_bios_encoder_supports_dp(devdata)) - sanitize_aux_ch(devdata, port); - - i915->display.vbt.ports[port] = devdata; } static bool has_ddi_port_info(struct drm_i915_private *i915) @@ -2770,7 +2654,6 @@ static bool has_ddi_port_info(struct drm_i915_private *i915) static void parse_ddi_ports(struct drm_i915_private *i915) { struct intel_bios_encoder_data *devdata; - enum port port; if (!has_ddi_port_info(i915)) return; @@ -2778,10 +2661,8 @@ static void parse_ddi_ports(struct drm_i915_private *i915) list_for_each_entry(devdata, &i915->display.vbt.display_devices, node) parse_ddi_port(devdata); - for_each_port(port) { - if (i915->display.vbt.ports[port]) - print_ddi_port(i915->display.vbt.ports[port], port); - } + list_for_each_entry(devdata, &i915->display.vbt.display_devices, node) + print_ddi_port(devdata); } static void @@ -3659,6 +3540,27 @@ enum aux_ch intel_bios_dp_aux_ch(const struct intel_bios_encoder_data *devdata) return map_aux_ch(devdata->i915, devdata->child.aux_channel); } +bool intel_bios_dp_has_shared_aux_ch(const struct intel_bios_encoder_data *devdata) +{ + struct drm_i915_private *i915; + u8 aux_channel; + int count = 0; + + if (!devdata || !devdata->child.aux_channel) + return false; + + i915 = devdata->i915; + aux_channel = devdata->child.aux_channel; + + list_for_each_entry(devdata, &i915->display.vbt.display_devices, node) { + if (intel_bios_encoder_supports_dp(devdata) && + aux_channel == devdata->child.aux_channel) + count++; + } + + return count > 1; +} + int intel_bios_dp_boost_level(const struct intel_bios_encoder_data *devdata) { if (!devdata || devdata->i915->display.vbt.version < 196 || !devdata->child.iboost) @@ -3706,5 +3608,22 @@ bool intel_bios_encoder_hpd_invert(const struct intel_bios_encoder_data *devdata const struct intel_bios_encoder_data * intel_bios_encoder_data_lookup(struct drm_i915_private *i915, enum port port) { - return i915->display.vbt.ports[port]; + struct intel_bios_encoder_data *devdata; + + list_for_each_entry(devdata, &i915->display.vbt.display_devices, node) { + if (intel_bios_encoder_port(devdata) == port) + return devdata; + } + + return NULL; +} + +void intel_bios_for_each_encoder(struct drm_i915_private *i915, + void (*func)(struct drm_i915_private *i915, + const struct intel_bios_encoder_data *devdata)) +{ + struct intel_bios_encoder_data *devdata; + + list_for_each_entry(devdata, &i915->display.vbt.display_devices, node) + func(i915, devdata); } diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h index 45fae97d9719..49e24b7cf675 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.h +++ b/drivers/gpu/drm/i915/display/intel_bios.h @@ -263,17 +263,24 @@ bool intel_bios_encoder_supports_dp(const struct intel_bios_encoder_data *devdat bool intel_bios_encoder_supports_edp(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_supports_typec_usb(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_supports_tbt(const struct intel_bios_encoder_data *devdata); +bool intel_bios_encoder_supports_dsi(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_supports_dp_dual_mode(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_is_lspcon(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_lane_reversal(const struct intel_bios_encoder_data *devdata); bool intel_bios_encoder_hpd_invert(const struct intel_bios_encoder_data *devdata); +enum port intel_bios_encoder_port(const struct intel_bios_encoder_data *devdata); enum aux_ch intel_bios_dp_aux_ch(const struct intel_bios_encoder_data *devdata); int intel_bios_dp_boost_level(const struct intel_bios_encoder_data *devdata); int intel_bios_dp_max_lane_count(const struct intel_bios_encoder_data *devdata); int intel_bios_dp_max_link_rate(const struct intel_bios_encoder_data *devdata); +bool intel_bios_dp_has_shared_aux_ch(const struct intel_bios_encoder_data *devdata); int intel_bios_hdmi_boost_level(const struct intel_bios_encoder_data *devdata); int intel_bios_hdmi_ddc_pin(const struct intel_bios_encoder_data *devdata); int intel_bios_hdmi_level_shift(const struct intel_bios_encoder_data *devdata); int intel_bios_hdmi_max_tmds_clock(const struct intel_bios_encoder_data *devdata); +void intel_bios_for_each_encoder(struct drm_i915_private *i915, + void (*func)(struct drm_i915_private *i915, + const struct intel_bios_encoder_data *devdata)); + #endif /* _INTEL_BIOS_H_ */ diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c index 4207863b7b2a..2fb030b1ff1d 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -37,6 +37,7 @@ #include "intel_pci_config.h" #include "intel_pcode.h" #include "intel_psr.h" +#include "intel_vdsc.h" #include "vlv_sideband.h" /** @@ -469,7 +470,7 @@ static void hsw_get_cdclk(struct drm_i915_private *dev_priv, cdclk_config->cdclk = 450000; else if (freq == LCPLL_CLK_FREQ_450) cdclk_config->cdclk = 450000; - else if (IS_HSW_ULT(dev_priv)) + else if (IS_HASWELL_ULT(dev_priv)) cdclk_config->cdclk = 337500; else cdclk_config->cdclk = 540000; @@ -2607,9 +2608,16 @@ int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state) * When we decide to use only one VDSC engine, since * each VDSC operates with 1 ppc throughput, pixel clock * cannot be higher than the VDSC clock (cdclk) + * If there 2 VDSC engines, then pixel clock can't be higher than + * VDSC clock(cdclk) * 2 and so on. */ - if (crtc_state->dsc.compression_enable && !crtc_state->dsc.dsc_split) - min_cdclk = max(min_cdclk, (int)crtc_state->pixel_rate); + if (crtc_state->dsc.compression_enable) { + int num_vdsc_instances = intel_dsc_get_num_vdsc_instances(crtc_state); + + min_cdclk = max_t(int, min_cdclk, + DIV_ROUND_UP(crtc_state->pixel_rate, + num_vdsc_instances)); + } /* * HACK. Currently for TGL/DG2 platforms we calculate @@ -3147,7 +3155,7 @@ static int intel_compute_max_dotclk(struct drm_i915_private *dev_priv) */ void intel_update_max_cdclk(struct drm_i915_private *dev_priv) { - if (IS_JSL_EHL(dev_priv)) { + if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) { if (dev_priv->display.cdclk.hw.ref == 24000) dev_priv->display.cdclk.max_cdclk_freq = 552000; else @@ -3192,9 +3200,9 @@ void intel_update_max_cdclk(struct drm_i915_private *dev_priv) */ if (intel_de_read(dev_priv, FUSE_STRAP) & HSW_CDCLK_LIMIT) dev_priv->display.cdclk.max_cdclk_freq = 450000; - else if (IS_BDW_ULX(dev_priv)) + else if (IS_BROADWELL_ULX(dev_priv)) dev_priv->display.cdclk.max_cdclk_freq = 450000; - else if (IS_BDW_ULT(dev_priv)) + else if (IS_BROADWELL_ULT(dev_priv)) dev_priv->display.cdclk.max_cdclk_freq = 540000; else dev_priv->display.cdclk.max_cdclk_freq = 675000; @@ -3559,10 +3567,10 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv) dev_priv->display.cdclk.table = dg2_cdclk_table; } else if (IS_ALDERLAKE_P(dev_priv)) { /* Wa_22011320316:adl-p[a0] */ - if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { + if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { dev_priv->display.cdclk.table = adlp_a_step_cdclk_table; dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; - } else if (IS_ADLP_RPLU(dev_priv)) { + } else if (IS_RAPTORLAKE_U(dev_priv)) { dev_priv->display.cdclk.table = rplu_cdclk_table; dev_priv->display.funcs.cdclk = &rplu_cdclk_funcs; } else { @@ -3575,7 +3583,7 @@ void intel_init_cdclk_hooks(struct drm_i915_private *dev_priv) } else if (DISPLAY_VER(dev_priv) >= 12) { dev_priv->display.funcs.cdclk = &tgl_cdclk_funcs; dev_priv->display.cdclk.table = icl_cdclk_table; - } else if (IS_JSL_EHL(dev_priv)) { + } else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) { dev_priv->display.funcs.cdclk = &ehl_cdclk_funcs; dev_priv->display.cdclk.table = icl_cdclk_table; } else if (DISPLAY_VER(dev_priv) >= 11) { diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c index 8966e6560516..454607b4a02a 100644 --- a/drivers/gpu/drm/i915/display/intel_color.c +++ b/drivers/gpu/drm/i915/display/intel_color.c @@ -1453,6 +1453,16 @@ static int glk_degamma_lut_size(struct drm_i915_private *i915) return 35; } +/* + * change_lut_val_precision: helper function to upscale or downscale lut values. + * Parameters 'to' and 'from' needs to be less than 32. This should be sufficient + * as currently there are no lut values exceeding 32 bit. + */ +static u32 change_lut_val_precision(u32 lut_val, int to, int from) +{ + return mul_u32_u32(lut_val, (1 << to)) / (1 << from); +} + static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state, const struct drm_property_blob *blob) { @@ -1487,8 +1497,15 @@ static void glk_load_degamma_lut(const struct intel_crtc_state *crtc_state, * ToDo: Extend to max 7.0. Enable 32 bit input value * as compared to just 16 to achieve this. */ + u32 lut_val; + + if (DISPLAY_VER(i915) >= 14) + lut_val = change_lut_val_precision(lut[i].green, 24, 16); + else + lut_val = lut[i].green; + ilk_lut_write(crtc_state, PRE_CSC_GAMC_DATA(pipe), - lut[i].green); + lut_val); } /* Clamp values > 1.0. */ @@ -3439,6 +3456,14 @@ static struct drm_property_blob *glk_read_degamma_lut(struct intel_crtc *crtc) for (i = 0; i < lut_size; i++) { u32 val = intel_de_read_fw(dev_priv, PRE_CSC_GAMC_DATA(pipe)); + /* + * For MTL and beyond, convert back the 24 bit lut values + * read from HW to 16 bit values to maintain parity with + * userspace values + */ + if (DISPLAY_VER(dev_priv) >= 14) + val = change_lut_val_precision(val, 16, 24); + lut[i].red = val; lut[i].green = val; lut[i].blue = val; diff --git a/drivers/gpu/drm/i915/display/intel_combo_phy.c b/drivers/gpu/drm/i915/display/intel_combo_phy.c index 922a6d87b553..e2a220cf2e57 100644 --- a/drivers/gpu/drm/i915/display/intel_combo_phy.c +++ b/drivers/gpu/drm/i915/display/intel_combo_phy.c @@ -141,7 +141,7 @@ static bool has_phy_misc(struct drm_i915_private *i915, enum phy phy) if (IS_ALDERLAKE_S(i915)) return phy == PHY_A; - else if (IS_JSL_EHL(i915) || + else if ((IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) || IS_ROCKETLAKE(i915) || IS_DG1(i915)) return phy < PHY_C; @@ -242,7 +242,7 @@ static bool icl_combo_phy_verify_state(struct drm_i915_private *dev_priv, ret &= check_phy_reg(dev_priv, phy, ICL_PORT_COMP_DW8(phy), IREFGEN, IREFGEN); - if (IS_JSL_EHL(dev_priv)) { + if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) { if (ehl_vbt_ddi_d_present(dev_priv)) expected_val = ICL_PHY_MISC_MUX_DDID; @@ -333,7 +333,8 @@ static void icl_combo_phys_init(struct drm_i915_private *dev_priv) * "internal" child devices. */ val = intel_de_read(dev_priv, ICL_PHY_MISC(phy)); - if (IS_JSL_EHL(dev_priv) && phy == PHY_A) { + if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && + phy == PHY_A) { val &= ~ICL_PHY_MISC_MUX_DDID; if (ehl_vbt_ddi_d_present(dev_priv)) diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index ab7cd5e60a0a..809074758687 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -1064,6 +1064,8 @@ void intel_crt_init(struct drm_i915_private *dev_priv) } if (HAS_DDI(dev_priv)) { + assert_port_valid(dev_priv, PORT_E); + crt->base.port = PORT_E; crt->base.get_config = hsw_crt_get_config; crt->base.get_hw_state = intel_ddi_get_hw_state; diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c index 719447ce86e7..1b00ef2c6185 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c @@ -116,6 +116,7 @@ static int intel_cx0_wait_for_ack(struct drm_i915_private *i915, enum port port, XELPDP_MSGBUS_TIMEOUT_SLOW, val)) { drm_dbg_kms(&i915->drm, "PHY %c Timeout waiting for message ACK. Status: 0x%x\n", phy_name(phy), *val); + intel_cx0_bus_reset(i915, port, lane); return -ETIMEDOUT; } @@ -158,10 +159,8 @@ static int __intel_cx0_read_once(struct drm_i915_private *i915, enum port port, XELPDP_PORT_M2P_ADDRESS(addr)); ack = intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_READ_ACK, lane, &val); - if (ack < 0) { - intel_cx0_bus_reset(i915, port, lane); + if (ack < 0) return ack; - } intel_clear_response_ready_flag(i915, port, lane); @@ -202,6 +201,7 @@ static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port, int lane, u16 addr, u8 data, bool committed) { enum phy phy = intel_port_to_phy(i915, port); + int ack; u32 val; if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), @@ -230,10 +230,9 @@ static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port, } if (committed) { - if (intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_WRITE_ACK, lane, &val) < 0) { - intel_cx0_bus_reset(i915, port, lane); - return -EINVAL; - } + ack = intel_cx0_wait_for_ack(i915, port, XELPDP_PORT_P2M_COMMAND_WRITE_ACK, lane, &val); + if (ack < 0) + return ack; } else if ((intel_de_read(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane)) & XELPDP_PORT_P2M_ERROR_SET)) { drm_dbg_kms(&i915->drm, diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.h b/drivers/gpu/drm/i915/display/intel_cx0_phy.h index f99809af257d..4c4db5cdcbd0 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.h +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.h @@ -43,8 +43,5 @@ int intel_c20pll_calc_port_clock(struct intel_encoder *encoder, void intel_cx0_phy_set_signal_levels(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); int intel_cx0_phy_check_hdmi_link_rate(struct intel_hdmi *hdmi, int clock); -void intel_cx0_phy_ddi_vswing_sequence(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - u32 level); int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder); #endif /* __INTEL_CX0_PHY_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index 090f242e610c..84bbf854337a 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -32,6 +32,7 @@ #include "i915_drv.h" #include "i915_reg.h" +#include "icl_dsi.h" #include "intel_audio.h" #include "intel_audio_regs.h" #include "intel_backlight.h" @@ -3582,7 +3583,8 @@ void intel_ddi_compute_min_voltage_level(struct drm_i915_private *dev_priv, { if (DISPLAY_VER(dev_priv) >= 12 && crtc_state->port_clock > 594000) crtc_state->min_voltage_level = 2; - else if (IS_JSL_EHL(dev_priv) && crtc_state->port_clock > 594000) + else if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && + crtc_state->port_clock > 594000) crtc_state->min_voltage_level = 3; else if (DISPLAY_VER(dev_priv) >= 11 && crtc_state->port_clock > 594000) crtc_state->min_voltage_level = 1; @@ -4653,13 +4655,95 @@ static void intel_ddi_tc_encoder_shutdown_complete(struct intel_encoder *encoder #define port_tc_name(port) ((port) - PORT_TC1 + '1') #define tc_port_name(tc_port) ((tc_port) - TC_PORT_1 + '1') -void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) +static bool port_strap_detected(struct drm_i915_private *i915, enum port port) +{ + /* straps not used on skl+ */ + if (DISPLAY_VER(i915) >= 9) + return true; + + switch (port) { + case PORT_A: + return intel_de_read(i915, DDI_BUF_CTL(PORT_A)) & DDI_INIT_DISPLAY_DETECTED; + case PORT_B: + return intel_de_read(i915, SFUSE_STRAP) & SFUSE_STRAP_DDIB_DETECTED; + case PORT_C: + return intel_de_read(i915, SFUSE_STRAP) & SFUSE_STRAP_DDIC_DETECTED; + case PORT_D: + return intel_de_read(i915, SFUSE_STRAP) & SFUSE_STRAP_DDID_DETECTED; + case PORT_E: + return true; /* no strap for DDI-E */ + default: + MISSING_CASE(port); + return false; + } +} + +static bool need_aux_ch(struct intel_encoder *encoder, bool init_dp) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + enum phy phy = intel_port_to_phy(i915, encoder->port); + + return init_dp || intel_phy_is_tc(i915, phy); +} + +static bool assert_has_icl_dsi(struct drm_i915_private *i915) +{ + return !drm_WARN(&i915->drm, !IS_ALDERLAKE_P(i915) && + !IS_TIGERLAKE(i915) && DISPLAY_VER(i915) != 11, + "Platform does not support DSI\n"); +} + +static bool port_in_use(struct drm_i915_private *i915, enum port port) +{ + struct intel_encoder *encoder; + + for_each_intel_encoder(&i915->drm, encoder) { + /* FIXME what about second port for dual link DSI? */ + if (encoder->port == port) + return true; + } + + return false; +} + +void intel_ddi_init(struct drm_i915_private *dev_priv, + const struct intel_bios_encoder_data *devdata) { struct intel_digital_port *dig_port; struct intel_encoder *encoder; - const struct intel_bios_encoder_data *devdata; bool init_hdmi, init_dp; - enum phy phy = intel_port_to_phy(dev_priv, port); + enum port port; + enum phy phy; + + port = intel_bios_encoder_port(devdata); + if (port == PORT_NONE) + return; + + if (!port_strap_detected(dev_priv, port)) { + drm_dbg_kms(&dev_priv->drm, + "Port %c strap not detected\n", port_name(port)); + return; + } + + if (!assert_port_valid(dev_priv, port)) + return; + + if (port_in_use(dev_priv, port)) { + drm_dbg_kms(&dev_priv->drm, + "Port %c already claimed\n", port_name(port)); + return; + } + + if (intel_bios_encoder_supports_dsi(devdata)) { + /* BXT/GLK handled elsewhere, for now at least */ + if (!assert_has_icl_dsi(dev_priv)) + return; + + icl_dsi_init(dev_priv, devdata); + return; + } + + phy = intel_port_to_phy(dev_priv, port); /* * On platforms with HTI (aka HDPORT), if it's enabled at boot it may @@ -4673,14 +4757,6 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) return; } - devdata = intel_bios_encoder_data_lookup(dev_priv, port); - if (!devdata) { - drm_dbg_kms(&dev_priv->drm, - "VBT says port %c is not present\n", - port_name(port)); - return; - } - init_hdmi = intel_bios_encoder_supports_dvi(devdata) || intel_bios_encoder_supports_hdmi(devdata); init_dp = intel_bios_encoder_supports_dp(devdata); @@ -4715,6 +4791,8 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) if (!dig_port) return; + dig_port->aux_ch = AUX_CH_NONE; + encoder = &dig_port->base; encoder->devdata = devdata; @@ -4801,7 +4879,7 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) encoder->disable_clock = dg1_ddi_disable_clock; encoder->is_clock_enabled = dg1_ddi_is_clock_enabled; encoder->get_config = dg1_ddi_get_config; - } else if (IS_JSL_EHL(dev_priv)) { + } else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) { if (intel_ddi_is_tc(dev_priv, port)) { encoder->enable_clock = jsl_ddi_tc_enable_clock; encoder->disable_clock = jsl_ddi_tc_disable_clock; @@ -4872,7 +4950,7 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) encoder->hpd_pin = rkl_hpd_pin(dev_priv, port); else if (DISPLAY_VER(dev_priv) >= 12) encoder->hpd_pin = tgl_hpd_pin(dev_priv, port); - else if (IS_JSL_EHL(dev_priv)) + else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) encoder->hpd_pin = ehl_hpd_pin(dev_priv, port); else if (DISPLAY_VER(dev_priv) == 11) encoder->hpd_pin = icl_hpd_pin(dev_priv, port); @@ -4895,7 +4973,12 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) dig_port->dp.output_reg = INVALID_MMIO_REG; dig_port->max_lanes = intel_ddi_max_lanes(dig_port); - dig_port->aux_ch = intel_dp_aux_ch(encoder); + + if (need_aux_ch(encoder, init_dp)) { + dig_port->aux_ch = intel_dp_aux_ch(encoder); + if (dig_port->aux_ch == AUX_CH_NONE) + goto err; + } if (intel_phy_is_tc(dev_priv, phy)) { bool is_legacy = diff --git a/drivers/gpu/drm/i915/display/intel_ddi.h b/drivers/gpu/drm/i915/display/intel_ddi.h index 2bc034042a93..4999c0ee229b 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.h +++ b/drivers/gpu/drm/i915/display/intel_ddi.h @@ -11,6 +11,7 @@ struct drm_connector_state; struct drm_i915_private; struct intel_atomic_state; +struct intel_bios_encoder_data; struct intel_connector; struct intel_crtc; struct intel_crtc_state; @@ -50,7 +51,8 @@ void hsw_prepare_dp_ddi_buffers(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); void intel_wait_ddi_buf_idle(struct drm_i915_private *dev_priv, enum port port); -void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port); +void intel_ddi_init(struct drm_i915_private *dev_priv, + const struct intel_bios_encoder_data *devdata); bool intel_ddi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe); void intel_ddi_enable_transcoder_func(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); diff --git a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c index b7d20485bde5..de809e2d9cac 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c +++ b/drivers/gpu/drm/i915/display/intel_ddi_buf_trans.c @@ -1049,12 +1049,26 @@ static const union intel_ddi_buf_trans_entry _mtl_c10_trans_dp14[] = { { .snps = { 62, 0, 0 } }, /* preset 9 */ }; -static const struct intel_ddi_buf_trans mtl_cx0_trans = { +static const struct intel_ddi_buf_trans mtl_c10_trans_dp14 = { .entries = _mtl_c10_trans_dp14, .num_entries = ARRAY_SIZE(_mtl_c10_trans_dp14), .hdmi_default_entry = ARRAY_SIZE(_mtl_c10_trans_dp14) - 1, }; +/* DP1.4 */ +static const union intel_ddi_buf_trans_entry _mtl_c20_trans_dp14[] = { + { .snps = { 20, 0, 0 } }, /* preset 0 */ + { .snps = { 24, 0, 4 } }, /* preset 1 */ + { .snps = { 30, 0, 9 } }, /* preset 2 */ + { .snps = { 34, 0, 14 } }, /* preset 3 */ + { .snps = { 29, 0, 0 } }, /* preset 4 */ + { .snps = { 34, 0, 5 } }, /* preset 5 */ + { .snps = { 38, 0, 10 } }, /* preset 6 */ + { .snps = { 36, 0, 0 } }, /* preset 7 */ + { .snps = { 40, 0, 6 } }, /* preset 8 */ + { .snps = { 48, 0, 0 } }, /* preset 9 */ +}; + /* DP2.0 */ static const union intel_ddi_buf_trans_entry _mtl_c20_trans_uhbr[] = { { .snps = { 48, 0, 0 } }, /* preset 0 */ @@ -1072,7 +1086,7 @@ static const union intel_ddi_buf_trans_entry _mtl_c20_trans_uhbr[] = { { .snps = { 37, 4, 7 } }, /* preset 12 */ { .snps = { 33, 4, 11 } }, /* preset 13 */ { .snps = { 40, 8, 0 } }, /* preset 14 */ - { .snps = { 28, 2, 2 } }, /* preset 15 */ + { .snps = { 30, 2, 2 } }, /* preset 15 */ }; /* HDMI2.0 */ @@ -1090,6 +1104,12 @@ static const struct intel_ddi_buf_trans mtl_c20_trans_hdmi = { .hdmi_default_entry = 0, }; +static const struct intel_ddi_buf_trans mtl_c20_trans_dp14 = { + .entries = _mtl_c20_trans_dp14, + .num_entries = ARRAY_SIZE(_mtl_c20_trans_dp14), + .hdmi_default_entry = ARRAY_SIZE(_mtl_c20_trans_dp14) - 1, +}; + static const struct intel_ddi_buf_trans mtl_c20_trans_uhbr = { .entries = _mtl_c20_trans_uhbr, .num_entries = ARRAY_SIZE(_mtl_c20_trans_uhbr), @@ -1390,7 +1410,7 @@ tgl_get_combo_buf_trans_dp(struct intel_encoder *encoder, struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); if (crtc_state->port_clock > 270000) { - if (IS_TGL_UY(dev_priv)) { + if (IS_TIGERLAKE_UY(dev_priv)) { return intel_get_buf_trans(&tgl_uy_combo_phy_trans_dp_hbr2, n_entries); } else { @@ -1678,8 +1698,10 @@ mtl_get_cx0_buf_trans(struct intel_encoder *encoder, return intel_get_buf_trans(&mtl_c20_trans_uhbr, n_entries); else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) && !(intel_is_c10phy(i915, phy))) return intel_get_buf_trans(&mtl_c20_trans_hdmi, n_entries); + else if (!intel_is_c10phy(i915, phy)) + return intel_get_buf_trans(&mtl_c20_trans_dp14, n_entries); else - return intel_get_buf_trans(&mtl_cx0_trans, n_entries); + return intel_get_buf_trans(&mtl_c10_trans_dp14, n_entries); } void intel_ddi_buf_trans_init(struct intel_encoder *encoder) @@ -1718,15 +1740,15 @@ void intel_ddi_buf_trans_init(struct intel_encoder *encoder) encoder->get_buf_trans = icl_get_mg_buf_trans; } else if (IS_GEMINILAKE(i915) || IS_BROXTON(i915)) { encoder->get_buf_trans = bxt_get_buf_trans; - } else if (IS_CML_ULX(i915) || IS_CFL_ULX(i915) || IS_KBL_ULX(i915)) { + } else if (IS_COMETLAKE_ULX(i915) || IS_COFFEELAKE_ULX(i915) || IS_KABYLAKE_ULX(i915)) { encoder->get_buf_trans = kbl_y_get_buf_trans; - } else if (IS_CML_ULT(i915) || IS_CFL_ULT(i915) || IS_KBL_ULT(i915)) { + } else if (IS_COMETLAKE_ULT(i915) || IS_COFFEELAKE_ULT(i915) || IS_KABYLAKE_ULT(i915)) { encoder->get_buf_trans = kbl_u_get_buf_trans; } else if (IS_COMETLAKE(i915) || IS_COFFEELAKE(i915) || IS_KABYLAKE(i915)) { encoder->get_buf_trans = kbl_get_buf_trans; - } else if (IS_SKL_ULX(i915)) { + } else if (IS_SKYLAKE_ULX(i915)) { encoder->get_buf_trans = skl_y_get_buf_trans; - } else if (IS_SKL_ULT(i915)) { + } else if (IS_SKYLAKE_ULT(i915)) { encoder->get_buf_trans = skl_u_get_buf_trans; } else if (IS_SKYLAKE(i915)) { encoder->get_buf_trans = skl_get_buf_trans; diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 16603d591f56..763ab569d8f3 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -53,7 +53,6 @@ #include "i915_utils.h" #include "i9xx_plane.h" #include "i9xx_wm.h" -#include "icl_dsi.h" #include "intel_atomic.h" #include "intel_atomic_plane.h" #include "intel_audio.h" @@ -1750,7 +1749,7 @@ bool intel_phy_is_combo(struct drm_i915_private *dev_priv, enum phy phy) return phy <= PHY_E; else if (IS_DG1(dev_priv) || IS_ROCKETLAKE(dev_priv)) return phy <= PHY_D; - else if (IS_JSL_EHL(dev_priv)) + else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) return phy <= PHY_C; else if (IS_ALDERLAKE_P(dev_priv) || IS_DISPLAY_VER(dev_priv, 11, 12)) return phy <= PHY_B; @@ -1802,7 +1801,8 @@ enum phy intel_port_to_phy(struct drm_i915_private *i915, enum port port) return PHY_B + port - PORT_TC1; else if ((IS_DG1(i915) || IS_ROCKETLAKE(i915)) && port >= PORT_TC1) return PHY_C + port - PORT_TC1; - else if (IS_JSL_EHL(i915) && port == PORT_D) + else if ((IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) && + port == PORT_D) return PHY_A; return PHY_A + port - PORT_A; @@ -3153,6 +3153,10 @@ static void bdw_set_pipe_misc(const struct intel_crtc_state *crtc_state) if (DISPLAY_VER(dev_priv) >= 12) val |= PIPE_MISC_PIXEL_ROUNDING_TRUNC; + /* allow PSR with sprite enabled */ + if (IS_BROADWELL(dev_priv)) + val |= PIPE_MISC_PSR_MASK_SPRITE_ENABLE; + intel_de_write(dev_priv, PIPE_MISC(crtc->pipe), val); } @@ -7143,7 +7147,11 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) */ intel_uncore_arm_unclaimed_mmio_detection(&dev_priv->uncore); } - intel_display_power_put(dev_priv, POWER_DOMAIN_DC_OFF, wakeref); + /* + * Delay re-enabling DC states by 17 ms to avoid the off->on->off + * toggling overhead at and above 60 FPS. + */ + intel_display_power_put_async_delay(dev_priv, POWER_DOMAIN_DC_OFF, wakeref, 17); intel_runtime_pm_put(&dev_priv->runtime_pm, state->wakeref); /* @@ -7370,7 +7378,7 @@ static bool intel_ddi_crt_present(struct drm_i915_private *dev_priv) if (DISPLAY_VER(dev_priv) >= 9) return false; - if (IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv)) + if (IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv)) return false; if (HAS_PCH_LPT_H(dev_priv) && @@ -7387,6 +7395,12 @@ static bool intel_ddi_crt_present(struct drm_i915_private *dev_priv) return true; } +bool assert_port_valid(struct drm_i915_private *i915, enum port port) +{ + return !drm_WARN(&i915->drm, !(DISPLAY_RUNTIME_INFO(i915)->port_mask & BIT(port)), + "Platform does not support port %c\n", port_name(port)); +} + void intel_setup_outputs(struct drm_i915_private *dev_priv) { struct intel_encoder *encoder; @@ -7397,93 +7411,14 @@ void intel_setup_outputs(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) return; - if (IS_METEORLAKE(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_TC1); - intel_ddi_init(dev_priv, PORT_TC2); - intel_ddi_init(dev_priv, PORT_TC3); - intel_ddi_init(dev_priv, PORT_TC4); - } else if (IS_DG2(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_C); - intel_ddi_init(dev_priv, PORT_D_XELPD); - intel_ddi_init(dev_priv, PORT_TC1); - } else if (IS_ALDERLAKE_P(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_TC1); - intel_ddi_init(dev_priv, PORT_TC2); - intel_ddi_init(dev_priv, PORT_TC3); - intel_ddi_init(dev_priv, PORT_TC4); - icl_dsi_init(dev_priv); - } else if (IS_ALDERLAKE_S(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_TC1); - intel_ddi_init(dev_priv, PORT_TC2); - intel_ddi_init(dev_priv, PORT_TC3); - intel_ddi_init(dev_priv, PORT_TC4); - } else if (IS_DG1(dev_priv) || IS_ROCKETLAKE(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_TC1); - intel_ddi_init(dev_priv, PORT_TC2); - } else if (DISPLAY_VER(dev_priv) >= 12) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_TC1); - intel_ddi_init(dev_priv, PORT_TC2); - intel_ddi_init(dev_priv, PORT_TC3); - intel_ddi_init(dev_priv, PORT_TC4); - intel_ddi_init(dev_priv, PORT_TC5); - intel_ddi_init(dev_priv, PORT_TC6); - icl_dsi_init(dev_priv); - } else if (IS_JSL_EHL(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_C); - intel_ddi_init(dev_priv, PORT_D); - icl_dsi_init(dev_priv); - } else if (DISPLAY_VER(dev_priv) == 11) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_C); - intel_ddi_init(dev_priv, PORT_D); - intel_ddi_init(dev_priv, PORT_E); - intel_ddi_init(dev_priv, PORT_F); - icl_dsi_init(dev_priv); - } else if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_C); - vlv_dsi_init(dev_priv); - } else if (DISPLAY_VER(dev_priv) >= 9) { - intel_ddi_init(dev_priv, PORT_A); - intel_ddi_init(dev_priv, PORT_B); - intel_ddi_init(dev_priv, PORT_C); - intel_ddi_init(dev_priv, PORT_D); - intel_ddi_init(dev_priv, PORT_E); - } else if (HAS_DDI(dev_priv)) { - u32 found; - + if (HAS_DDI(dev_priv)) { if (intel_ddi_crt_present(dev_priv)) intel_crt_init(dev_priv); - /* Haswell uses DDI functions to detect digital outputs. */ - found = intel_de_read(dev_priv, DDI_BUF_CTL(PORT_A)) & DDI_INIT_DISPLAY_DETECTED; - if (found) - intel_ddi_init(dev_priv, PORT_A); - - found = intel_de_read(dev_priv, SFUSE_STRAP); - if (found & SFUSE_STRAP_DDIB_DETECTED) - intel_ddi_init(dev_priv, PORT_B); - if (found & SFUSE_STRAP_DDIC_DETECTED) - intel_ddi_init(dev_priv, PORT_C); - if (found & SFUSE_STRAP_DDID_DETECTED) - intel_ddi_init(dev_priv, PORT_D); - if (found & SFUSE_STRAP_DDIF_DETECTED) - intel_ddi_init(dev_priv, PORT_F); + intel_bios_for_each_encoder(dev_priv, intel_ddi_init); + + if (IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) + vlv_dsi_init(dev_priv); } else if (HAS_PCH_SPLIT(dev_priv)) { int found; diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h index c744c021af23..49ac8473b988 100644 --- a/drivers/gpu/drm/i915/display/intel_display.h +++ b/drivers/gpu/drm/i915/display/intel_display.h @@ -113,7 +113,7 @@ enum i9xx_plane_id { #define for_each_dbuf_slice(__dev_priv, __slice) \ for ((__slice) = DBUF_S1; (__slice) < I915_MAX_DBUF_SLICES; (__slice)++) \ - for_each_if(INTEL_INFO(__dev_priv)->display->dbuf.slice_mask & BIT(__slice)) + for_each_if(DISPLAY_INFO(__dev_priv)->dbuf.slice_mask & BIT(__slice)) #define for_each_dbuf_slice_in_mask(__dev_priv, __slice, __mask) \ for_each_dbuf_slice((__dev_priv), (__slice)) \ @@ -539,6 +539,8 @@ void assert_transcoder(struct drm_i915_private *dev_priv, #define assert_transcoder_enabled(d, t) assert_transcoder(d, t, true) #define assert_transcoder_disabled(d, t) assert_transcoder(d, t, false) +bool assert_port_valid(struct drm_i915_private *i915, enum port port); + /* * Use I915_STATE_WARN(x) (rather than WARN() and WARN_ON()) for hw state sanity * checks to check for unexpected conditions which may not necessarily be a user diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h index 8d2243c71dd8..53e5c33e08c3 100644 --- a/drivers/gpu/drm/i915/display/intel_display_core.h +++ b/drivers/gpu/drm/i915/display/intel_display_core.h @@ -17,6 +17,7 @@ #include <drm/drm_modeset_lock.h> #include "intel_cdclk.h" +#include "intel_display_device.h" #include "intel_display_limits.h" #include "intel_display_power.h" #include "intel_dpll_mgr.h" @@ -33,7 +34,6 @@ struct i915_audio_component; struct i915_hdcp_arbiter; struct intel_atomic_state; struct intel_audio_funcs; -struct intel_bios_encoder_data; struct intel_cdclk_funcs; struct intel_cdclk_vals; struct intel_color_funcs; @@ -218,7 +218,6 @@ struct intel_vbt_data { struct list_head display_devices; struct list_head bdb_blocks; - struct intel_bios_encoder_data *ports[I915_MAX_PORTS]; /* Non-NULL if port present. */ struct sdvo_device_mapping { u8 initialized; u8 dvo_port; @@ -430,6 +429,14 @@ struct intel_display { } hti; struct { + /* Access with DISPLAY_INFO() */ + const struct intel_display_device_info *__device_info; + + /* Access with DISPLAY_RUNTIME_INFO() */ + struct intel_display_runtime_info __runtime_info; + } info; + + struct { bool false_color; } ips; diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index 165e2c7e3126..63c1fb9e479f 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -819,8 +819,7 @@ static ssize_t i915_displayport_test_active_write(struct file *file, if (IS_ERR(input_buffer)) return PTR_ERR(input_buffer); - drm_dbg(&to_i915(dev)->drm, - "Copied %d bytes from user\n", (unsigned int)len); + drm_dbg(dev, "Copied %d bytes from user\n", (unsigned int)len); drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(connector, &conn_iter) { @@ -839,8 +838,7 @@ static ssize_t i915_displayport_test_active_write(struct file *file, status = kstrtoint(input_buffer, 10, &val); if (status < 0) break; - drm_dbg(&to_i915(dev)->drm, - "Got %d for test active\n", val); + drm_dbg(dev, "Got %d for test active\n", val); /* To prevent erroneous activation of the compliance * testing code, only accept an actual value of 1 here */ diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c index 3fd30e7f0062..c39f8a15d8aa 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.c +++ b/drivers/gpu/drm/i915/display/intel_display_device.c @@ -16,9 +16,6 @@ #include "intel_display_reg_defs.h" #include "intel_fbc.h" -__diag_push(); -__diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); - static const struct intel_display_device_info no_display = {}; #define PIPE_A_OFFSET 0x70000 @@ -187,10 +184,6 @@ static const struct intel_display_device_info no_display = {}; .__runtime_defaults.cpu_transcoder_mask = \ BIT(TRANSCODER_A) | BIT(TRANSCODER_B) -static const struct intel_display_device_info i830_display = { - I830_DISPLAY, -}; - #define I845_DISPLAY \ .has_overlay = 1, \ .overlay_needs_physical = 1, \ @@ -203,19 +196,29 @@ static const struct intel_display_device_info i830_display = { .__runtime_defaults.pipe_mask = BIT(PIPE_A), \ .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) +static const struct intel_display_device_info i830_display = { + I830_DISPLAY, + + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C), /* DVO A/B/C */ +}; + static const struct intel_display_device_info i845_display = { I845_DISPLAY, + + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C), /* DVO B/C */ }; static const struct intel_display_device_info i85x_display = { I830_DISPLAY, + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C), /* DVO B/C */ .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; static const struct intel_display_device_info i865g_display = { I845_DISPLAY, + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C), /* DVO B/C */ .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -228,7 +231,8 @@ static const struct intel_display_device_info i865g_display = { .__runtime_defaults.ip.ver = 3, \ .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ .__runtime_defaults.cpu_transcoder_mask = \ - BIT(TRANSCODER_A) | BIT(TRANSCODER_B) + BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C) /* SDVO B/C */ static const struct intel_display_device_info i915g_display = { GEN3_DISPLAY, @@ -293,6 +297,8 @@ static const struct intel_display_device_info pnv_display = { static const struct intel_display_device_info i965g_display = { GEN4_DISPLAY, .has_overlay = 1, + + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C), /* SDVO B/C */ }; static const struct intel_display_device_info i965gm_display = { @@ -300,17 +306,21 @@ static const struct intel_display_device_info i965gm_display = { .has_overlay = 1, .supports_tv = 1, + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C), /* SDVO B/C */ .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; static const struct intel_display_device_info g45_display = { GEN4_DISPLAY, + + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D), /* SDVO/HDMI/DP B/C, DP D */ }; static const struct intel_display_device_info gm45_display = { GEN4_DISPLAY, .supports_tv = 1, + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D), /* SDVO/HDMI/DP B/C, DP D */ .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -323,7 +333,8 @@ static const struct intel_display_device_info gm45_display = { .__runtime_defaults.ip.ver = 5, \ .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), \ .__runtime_defaults.cpu_transcoder_mask = \ - BIT(TRANSCODER_A) | BIT(TRANSCODER_B) + BIT(TRANSCODER_A) | BIT(TRANSCODER_B), \ + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D) /* DP A, SDVO/HDMI/DP B, HDMI/DP C/D */ static const struct intel_display_device_info ilk_d_display = { ILK_DISPLAY, @@ -345,6 +356,7 @@ static const struct intel_display_device_info snb_display = { .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D), /* DP A, SDVO/HDMI/DP B, HDMI/DP C/D */ .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -358,6 +370,7 @@ static const struct intel_display_device_info ivb_display = { .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D), /* DP A, SDVO/HDMI/DP B, HDMI/DP C/D */ .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -373,6 +386,7 @@ static const struct intel_display_device_info vlv_display = { .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B), .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B), + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C), /* HDMI/DP B/C */ }; static const struct intel_display_device_info hsw_display = { @@ -380,6 +394,8 @@ static const struct intel_display_device_info hsw_display = { .has_dp_mst = 1, .has_fpga_dbg = 1, .has_hotplug = 1, + .has_psr = 1, + .has_psr_hw_tracking = 1, HSW_PIPE_OFFSETS, IVB_CURSOR_OFFSETS, IVB_COLORS, @@ -389,6 +405,7 @@ static const struct intel_display_device_info hsw_display = { .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D) | BIT(PORT_E), .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -397,6 +414,8 @@ static const struct intel_display_device_info bdw_display = { .has_dp_mst = 1, .has_fpga_dbg = 1, .has_hotplug = 1, + .has_psr = 1, + .has_psr_hw_tracking = 1, HSW_PIPE_OFFSETS, IVB_CURSOR_OFFSETS, IVB_COLORS, @@ -406,6 +425,7 @@ static const struct intel_display_device_info bdw_display = { .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D) | BIT(PORT_E), .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -421,6 +441,7 @@ static const struct intel_display_device_info chv_display = { .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), + .__runtime_defaults.port_mask = BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D), /* HDMI/DP B/C/D */ }; static const struct intel_display_device_info skl_display = { @@ -444,6 +465,7 @@ static const struct intel_display_device_info skl_display = { .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D) | BIT(PORT_E), .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), }; @@ -467,7 +489,8 @@ static const struct intel_display_device_info skl_display = { .__runtime_defaults.cpu_transcoder_mask = \ BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \ - BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C) + BIT(TRANSCODER_DSI_A) | BIT(TRANSCODER_DSI_C), \ + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) static const struct intel_display_device_info bxt_display = { GEN9_LP_DISPLAY, @@ -484,46 +507,57 @@ static const struct intel_display_device_info glk_display = { .__runtime_defaults.ip.ver = 10, }; -static const struct intel_display_device_info gen11_display = { - .abox_mask = BIT(0), - .dbuf.size = 2048, - .dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), - .has_ddi = 1, - .has_dp_mst = 1, - .has_fpga_dbg = 1, - .has_hotplug = 1, - .has_ipc = 1, - .has_psr = 1, - .has_psr_hw_tracking = 1, - .pipe_offsets = { - [TRANSCODER_A] = PIPE_A_OFFSET, - [TRANSCODER_B] = PIPE_B_OFFSET, - [TRANSCODER_C] = PIPE_C_OFFSET, - [TRANSCODER_EDP] = PIPE_EDP_OFFSET, - [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, - [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, - }, - .trans_offsets = { - [TRANSCODER_A] = TRANSCODER_A_OFFSET, - [TRANSCODER_B] = TRANSCODER_B_OFFSET, - [TRANSCODER_C] = TRANSCODER_C_OFFSET, - [TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, - [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, - [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, - }, - IVB_CURSOR_OFFSETS, - ICL_COLORS, +#define ICL_DISPLAY \ + .abox_mask = BIT(0), \ + .dbuf.size = 2048, \ + .dbuf.slice_mask = BIT(DBUF_S1) | BIT(DBUF_S2), \ + .has_ddi = 1, \ + .has_dp_mst = 1, \ + .has_fpga_dbg = 1, \ + .has_hotplug = 1, \ + .has_ipc = 1, \ + .has_psr = 1, \ + .has_psr_hw_tracking = 1, \ + .pipe_offsets = { \ + [TRANSCODER_A] = PIPE_A_OFFSET, \ + [TRANSCODER_B] = PIPE_B_OFFSET, \ + [TRANSCODER_C] = PIPE_C_OFFSET, \ + [TRANSCODER_EDP] = PIPE_EDP_OFFSET, \ + [TRANSCODER_DSI_0] = PIPE_DSI0_OFFSET, \ + [TRANSCODER_DSI_1] = PIPE_DSI1_OFFSET, \ + }, \ + .trans_offsets = { \ + [TRANSCODER_A] = TRANSCODER_A_OFFSET, \ + [TRANSCODER_B] = TRANSCODER_B_OFFSET, \ + [TRANSCODER_C] = TRANSCODER_C_OFFSET, \ + [TRANSCODER_EDP] = TRANSCODER_EDP_OFFSET, \ + [TRANSCODER_DSI_0] = TRANSCODER_DSI0_OFFSET, \ + [TRANSCODER_DSI_1] = TRANSCODER_DSI1_OFFSET, \ + }, \ + IVB_CURSOR_OFFSETS, \ + ICL_COLORS, \ + \ + .__runtime_defaults.ip.ver = 11, \ + .__runtime_defaults.has_dmc = 1, \ + .__runtime_defaults.has_dsc = 1, \ + .__runtime_defaults.has_hdcp = 1, \ + .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), \ + .__runtime_defaults.cpu_transcoder_mask = \ + BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | \ + BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | \ + BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), \ + .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A) - .__runtime_defaults.ip.ver = 11, - .__runtime_defaults.has_dmc = 1, - .__runtime_defaults.has_dsc = 1, - .__runtime_defaults.has_hdcp = 1, - .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), - .__runtime_defaults.cpu_transcoder_mask = - BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | - BIT(TRANSCODER_C) | BIT(TRANSCODER_EDP) | - BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), - .__runtime_defaults.fbc_mask = BIT(INTEL_FBC_A), +static const struct intel_display_device_info icl_display = { + ICL_DISPLAY, + + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D) | BIT(PORT_E), +}; + +static const struct intel_display_device_info jsl_ehl_display = { + ICL_DISPLAY, + + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D), }; #define XE_D_DISPLAY \ @@ -571,6 +605,20 @@ static const struct intel_display_device_info gen11_display = { static const struct intel_display_device_info tgl_display = { XE_D_DISPLAY, + + /* + * FIXME DDI C/combo PHY C missing due to combo PHY + * code making a mess on SKUs where the PHY is missing. + */ + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | + BIT(PORT_TC1) | BIT(PORT_TC2) | BIT(PORT_TC3) | BIT(PORT_TC4) | BIT(PORT_TC5) | BIT(PORT_TC6), +}; + +static const struct intel_display_device_info dg1_display = { + XE_D_DISPLAY, + + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | + BIT(PORT_TC1) | BIT(PORT_TC2), }; static const struct intel_display_device_info rkl_display = { @@ -582,12 +630,17 @@ static const struct intel_display_device_info rkl_display = { .__runtime_defaults.pipe_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C), .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | + BIT(PORT_TC1) | BIT(PORT_TC2), }; static const struct intel_display_device_info adl_s_display = { XE_D_DISPLAY, .has_hti = 1, .has_psr_hw_tracking = 0, + + .__runtime_defaults.port_mask = BIT(PORT_A) | + BIT(PORT_TC1) | BIT(PORT_TC2) | BIT(PORT_TC3) | BIT(PORT_TC4), }; #define XE_LPD_FEATURES \ @@ -642,6 +695,8 @@ static const struct intel_display_device_info xe_lpd_display = { BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C) | BIT(TRANSCODER_D) | BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | + BIT(PORT_TC1) | BIT(PORT_TC2) | BIT(PORT_TC3) | BIT(PORT_TC4), }; static const struct intel_display_device_info xe_hpd_display = { @@ -651,6 +706,8 @@ static const struct intel_display_device_info xe_hpd_display = { .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C) | BIT(TRANSCODER_D), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_C) | BIT(PORT_D_XELPD) | + BIT(PORT_TC1), }; static const struct intel_display_device_info xe_lpdp_display = { @@ -663,14 +720,28 @@ static const struct intel_display_device_info xe_lpdp_display = { .__runtime_defaults.cpu_transcoder_mask = BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C) | BIT(TRANSCODER_D), + .__runtime_defaults.port_mask = BIT(PORT_A) | BIT(PORT_B) | + BIT(PORT_TC1) | BIT(PORT_TC2) | BIT(PORT_TC3) | BIT(PORT_TC4), }; -__diag_pop(); +/* + * Separate detection for no display cases to keep the display id array simple. + * + * IVB Q requires subvendor and subdevice matching to differentiate from IVB D + * GT2 server. + */ +static bool has_no_display(struct pci_dev *pdev) +{ + static const struct pci_device_id ids[] = { + INTEL_IVB_Q_IDS(0), + {} + }; + + return pci_match_id(ids, pdev); +} #undef INTEL_VGA_DEVICE -#undef INTEL_QUANTA_VGA_DEVICE #define INTEL_VGA_DEVICE(id, info) { id, info } -#define INTEL_QUANTA_VGA_DEVICE(info) { 0x16a, info } static const struct { u32 devid; @@ -695,7 +766,6 @@ static const struct { INTEL_IRONLAKE_M_IDS(&ilk_m_display), INTEL_SNB_D_IDS(&snb_display), INTEL_SNB_M_IDS(&snb_display), - INTEL_IVB_Q_IDS(NULL), /* must be first IVB in list */ INTEL_IVB_M_IDS(&ivb_display), INTEL_IVB_D_IDS(&ivb_display), INTEL_HSW_IDS(&hsw_display), @@ -707,11 +777,11 @@ static const struct { INTEL_GLK_IDS(&glk_display), INTEL_KBL_IDS(&skl_display), INTEL_CFL_IDS(&skl_display), - INTEL_ICL_11_IDS(&gen11_display), - INTEL_EHL_IDS(&gen11_display), - INTEL_JSL_IDS(&gen11_display), + INTEL_ICL_11_IDS(&icl_display), + INTEL_EHL_IDS(&jsl_ehl_display), + INTEL_JSL_IDS(&jsl_ehl_display), INTEL_TGL_12_IDS(&tgl_display), - INTEL_DG1_IDS(&tgl_display), + INTEL_DG1_IDS(&dg1_display), INTEL_RKL_IDS(&rkl_display), INTEL_ADLS_IDS(&adl_s_display), INTEL_RPLS_IDS(&adl_s_display), @@ -743,6 +813,15 @@ probe_gmdid_display(struct drm_i915_private *i915, u16 *ver, u16 *rel, u16 *step u32 val; int i; + /* The caller expects to ver, rel and step to be initialized + * here, and there's no good way to check when there was a + * failure and no_display was returned. So initialize all these + * values here zero, to be sure. + */ + *ver = 0; + *rel = 0; + *step = 0; + addr = pci_iomap_range(pdev, 0, i915_mmio_reg_offset(GMD_ID_DISPLAY), sizeof(u32)); if (!addr) { drm_err(&i915->drm, "Cannot map MMIO BAR to read display GMD_ID\n"); @@ -752,9 +831,10 @@ probe_gmdid_display(struct drm_i915_private *i915, u16 *ver, u16 *rel, u16 *step val = ioread32(addr); pci_iounmap(pdev, addr); - if (val == 0) - /* Platform doesn't have display */ + if (val == 0) { + drm_dbg_kms(&i915->drm, "Device doesn't have display\n"); return &no_display; + } *ver = REG_FIELD_GET(GMD_ID_ARCH_MASK, val); *rel = REG_FIELD_GET(GMD_ID_RELEASE_MASK, val); @@ -780,6 +860,11 @@ intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid, if (has_gmdid) return probe_gmdid_display(i915, gmdid_ver, gmdid_rel, gmdid_step); + if (has_no_display(pdev)) { + drm_dbg_kms(&i915->drm, "Device doesn't have display\n"); + return &no_display; + } + for (i = 0; i < ARRAY_SIZE(intel_display_ids); i++) { if (intel_display_ids[i].devid == pdev->device) return intel_display_ids[i].info; @@ -796,8 +881,12 @@ void intel_display_device_info_runtime_init(struct drm_i915_private *i915) struct intel_display_runtime_info *display_runtime = DISPLAY_RUNTIME_INFO(i915); enum pipe pipe; + BUILD_BUG_ON(BITS_PER_TYPE(display_runtime->pipe_mask) < I915_MAX_PIPES); + BUILD_BUG_ON(BITS_PER_TYPE(display_runtime->cpu_transcoder_mask) < I915_MAX_TRANSCODERS); + BUILD_BUG_ON(BITS_PER_TYPE(display_runtime->port_mask) < I915_MAX_PORTS); + /* Wa_14011765242: adl-s A0,A1 */ - if (IS_ADLS_DISPLAY_STEP(i915, STEP_A0, STEP_A2)) + if (IS_ALDERLAKE_S(i915) && IS_DISPLAY_STEP(i915, STEP_A0, STEP_A2)) for_each_pipe(i915, pipe) display_runtime->num_scalers[pipe] = 0; else if (DISPLAY_VER(i915) >= 11) { @@ -915,3 +1004,24 @@ void intel_display_device_info_runtime_init(struct drm_i915_private *i915) display_fused_off: memset(display_runtime, 0, sizeof(*display_runtime)); } + +void intel_display_device_info_print(const struct intel_display_device_info *info, + const struct intel_display_runtime_info *runtime, + struct drm_printer *p) +{ + if (runtime->ip.rel) + drm_printf(p, "display version: %u.%02u\n", + runtime->ip.ver, + runtime->ip.rel); + else + drm_printf(p, "display version: %u\n", + runtime->ip.ver); + +#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, str_yes_no(info->name)) + DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG); +#undef PRINT_FLAG + + drm_printf(p, "has_hdcp: %s\n", str_yes_no(runtime->has_hdcp)); + drm_printf(p, "has_dmc: %s\n", str_yes_no(runtime->has_dmc)); + drm_printf(p, "has_dsc: %s\n", str_yes_no(runtime->has_dsc)); +} diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h index 706ff2aa1f55..215e682bd8b7 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.h +++ b/drivers/gpu/drm/i915/display/intel_display_device.h @@ -11,6 +11,7 @@ #include "intel_display_limits.h" struct drm_i915_private; +struct drm_printer; #define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \ /* Keep in alphabetical order */ \ @@ -53,7 +54,7 @@ struct drm_i915_private; #define HAS_GMCH(i915) (DISPLAY_INFO(i915)->has_gmch) #define HAS_HW_SAGV_WM(i915) (DISPLAY_VER(i915) >= 13 && !IS_DGFX(i915)) #define HAS_IPC(i915) (DISPLAY_INFO(i915)->has_ipc) -#define HAS_IPS(i915) (IS_HSW_ULT(i915) || IS_BROADWELL(i915)) +#define HAS_IPS(i915) (IS_HASWELL_ULT(i915) || IS_BROADWELL(i915)) #define HAS_LSPCON(i915) (IS_DISPLAY_VER(i915, 9, 10)) #define HAS_MBUS_JOINING(i915) (IS_ALDERLAKE_P(i915) || DISPLAY_VER(i915) >= 14) #define HAS_MSO(i915) (DISPLAY_VER(i915) >= 12) @@ -79,6 +80,7 @@ struct intel_display_runtime_info { u8 pipe_mask; u8 cpu_transcoder_mask; + u16 port_mask; u8 num_sprites[I915_MAX_PIPES]; u8 num_scalers[I915_MAX_PIPES]; @@ -126,4 +128,8 @@ intel_display_device_probe(struct drm_i915_private *i915, bool has_gmdid, u16 *ver, u16 *rel, u16 *step); void intel_display_device_info_runtime_init(struct drm_i915_private *i915); +void intel_display_device_info_print(const struct intel_display_device_info *info, + const struct intel_display_runtime_info *runtime, + struct drm_printer *p); + #endif diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c index b909814ae02b..8f144d4d3c39 100644 --- a/drivers/gpu/drm/i915/display/intel_display_driver.c +++ b/drivers/gpu/drm/i915/display/intel_display_driver.c @@ -28,6 +28,7 @@ #include "intel_crtc.h" #include "intel_display_debugfs.h" #include "intel_display_driver.h" +#include "intel_display_irq.h" #include "intel_display_power.h" #include "intel_display_types.h" #include "intel_dkl_phy.h" @@ -177,6 +178,7 @@ void intel_display_driver_early_probe(struct drm_i915_private *i915) if (!HAS_DISPLAY(i915)) return; + intel_display_irq_init(i915); intel_dkl_phy_init(i915); intel_color_init_hooks(i915); intel_init_cdclk_hooks(i915); diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c index ae2578741dfe..62ce55475554 100644 --- a/drivers/gpu/drm/i915/display/intel_display_irq.c +++ b/drivers/gpu/drm/i915/display/intel_display_irq.c @@ -749,6 +749,20 @@ void ivb_display_irq_handler(struct drm_i915_private *dev_priv, u32 de_iir) if (de_iir & DE_ERR_INT_IVB) ivb_err_int_handler(dev_priv); + if (de_iir & DE_EDP_PSR_INT_HSW) { + struct intel_encoder *encoder; + + for_each_intel_encoder_with_psr(&dev_priv->drm, encoder) { + struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + u32 psr_iir; + + psr_iir = intel_uncore_rmw(&dev_priv->uncore, + EDP_PSR_IIR, 0, 0); + intel_psr_irq_handler(intel_dp, psr_iir); + break; + } + } + if (de_iir & DE_AUX_CHANNEL_A_IVB) intel_dp_aux_irq_handler(dev_priv); @@ -1135,7 +1149,7 @@ void gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl) u32 gen11_gu_misc_irq_ack(struct drm_i915_private *i915, const u32 master_ctl) { - void __iomem * const regs = i915->uncore.regs; + void __iomem * const regs = intel_uncore_regs(&i915->uncore); u32 iir; if (!(master_ctl & GEN11_GU_MISC_IRQ)) @@ -1156,7 +1170,7 @@ void gen11_gu_misc_irq_handler(struct drm_i915_private *i915, const u32 iir) void gen11_display_irq_handler(struct drm_i915_private *i915) { - void __iomem * const regs = i915->uncore.regs; + void __iomem * const regs = intel_uncore_regs(&i915->uncore); const u32 disp_ctl = raw_reg_read(regs, GEN11_DISPLAY_INT_CTL); disable_rpm_wakeref_asserts(&i915->runtime_pm); @@ -1523,7 +1537,7 @@ void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv, * to avoid races with the irq handler, assuming we have MSI. Shared legacy * interrupts could still race. */ -void ibx_irq_postinstall(struct drm_i915_private *dev_priv) +static void ibx_irq_postinstall(struct drm_i915_private *dev_priv) { struct intel_uncore *uncore = &dev_priv->uncore; u32 mask; @@ -1569,6 +1583,50 @@ void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv) vlv_display_irq_reset(dev_priv); } +void ilk_de_irq_postinstall(struct drm_i915_private *i915) +{ + struct intel_uncore *uncore = &i915->uncore; + u32 display_mask, extra_mask; + + if (GRAPHICS_VER(i915) >= 7) { + display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE_IVB | + DE_PCH_EVENT_IVB | DE_AUX_CHANNEL_A_IVB); + extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB | + DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB | + DE_PLANE_FLIP_DONE_IVB(PLANE_C) | + DE_PLANE_FLIP_DONE_IVB(PLANE_B) | + DE_PLANE_FLIP_DONE_IVB(PLANE_A) | + DE_DP_A_HOTPLUG_IVB); + } else { + display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT | + DE_AUX_CHANNEL_A | DE_PIPEB_CRC_DONE | + DE_PIPEA_CRC_DONE | DE_POISON); + extra_mask = (DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | + DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN | + DE_PLANE_FLIP_DONE(PLANE_A) | + DE_PLANE_FLIP_DONE(PLANE_B) | + DE_DP_A_HOTPLUG); + } + + if (IS_HASWELL(i915)) { + gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR); + display_mask |= DE_EDP_PSR_INT_HSW; + } + + if (IS_IRONLAKE_M(i915)) + extra_mask |= DE_PCU_EVENT; + + i915->irq_mask = ~display_mask; + + ibx_irq_postinstall(i915); + + GEN3_IRQ_INIT(uncore, DE, i915->irq_mask, + display_mask | extra_mask); +} + +static void mtp_irq_postinstall(struct drm_i915_private *i915); +static void icp_irq_postinstall(struct drm_i915_private *i915); + void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) { struct intel_uncore *uncore = &dev_priv->uncore; @@ -1586,6 +1644,13 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) return; + if (DISPLAY_VER(dev_priv) >= 14) + mtp_irq_postinstall(dev_priv); + else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) + icp_irq_postinstall(dev_priv); + else if (HAS_PCH_SPLIT(dev_priv)) + ibx_irq_postinstall(dev_priv); + if (DISPLAY_VER(dev_priv) <= 10) de_misc_masked |= GEN8_DE_MISC_GSE; @@ -1652,7 +1717,7 @@ void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) } } -void mtp_irq_postinstall(struct drm_i915_private *i915) +static void mtp_irq_postinstall(struct drm_i915_private *i915) { struct intel_uncore *uncore = &i915->uncore; u32 sde_mask = SDE_GMBUS_ICP | SDE_PICAINTERRUPT; @@ -1666,7 +1731,7 @@ void mtp_irq_postinstall(struct drm_i915_private *i915) GEN3_IRQ_INIT(uncore, SDE, ~sde_mask, 0xffffffff); } -void icp_irq_postinstall(struct drm_i915_private *dev_priv) +static void icp_irq_postinstall(struct drm_i915_private *dev_priv) { struct intel_uncore *uncore = &dev_priv->uncore; u32 mask = SDE_GMBUS_ICP; @@ -1685,3 +1750,30 @@ void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv) GEN11_DISPLAY_IRQ_ENABLE); } +void dg1_de_irq_postinstall(struct drm_i915_private *i915) +{ + if (!HAS_DISPLAY(i915)) + return; + + gen8_de_irq_postinstall(i915); + intel_uncore_write(&i915->uncore, GEN11_DISPLAY_INT_CTL, + GEN11_DISPLAY_IRQ_ENABLE); +} + +void intel_display_irq_init(struct drm_i915_private *i915) +{ + i915->drm.vblank_disable_immediate = true; + + /* + * Most platforms treat the display irq block as an always-on power + * domain. vlv/chv can disable it at runtime and need special care to + * avoid writing any of the display block registers outside of the power + * domain. We defer setting up the display irqs in this case to the + * runtime pm. + */ + i915->display_irqs_enabled = true; + if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) + i915->display_irqs_enabled = false; + + intel_hotplug_irq_init(i915); +} diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.h b/drivers/gpu/drm/i915/display/intel_display_irq.h index 874893f4f16d..2a090dd6abd7 100644 --- a/drivers/gpu/drm/i915/display/intel_display_irq.h +++ b/drivers/gpu/drm/i915/display/intel_display_irq.h @@ -58,12 +58,11 @@ void vlv_display_irq_reset(struct drm_i915_private *i915); void gen8_display_irq_reset(struct drm_i915_private *i915); void gen11_display_irq_reset(struct drm_i915_private *i915); -void ibx_irq_postinstall(struct drm_i915_private *i915); void vlv_display_irq_postinstall(struct drm_i915_private *i915); -void icp_irq_postinstall(struct drm_i915_private *i915); +void ilk_de_irq_postinstall(struct drm_i915_private *i915); void gen8_de_irq_postinstall(struct drm_i915_private *i915); -void mtp_irq_postinstall(struct drm_i915_private *i915); void gen11_de_irq_postinstall(struct drm_i915_private *i915); +void dg1_de_irq_postinstall(struct drm_i915_private *i915); u32 i915_pipestat_enable_mask(struct drm_i915_private *i915, enum pipe pipe); void i915_enable_pipestat(struct drm_i915_private *i915, enum pipe pipe, u32 status_mask); @@ -78,4 +77,6 @@ void i965_pipestat_irq_handler(struct drm_i915_private *i915, u32 iir, u32 pipe_ void valleyview_pipestat_irq_handler(struct drm_i915_private *i915, u32 pipe_stats[I915_MAX_PIPES]); void i8xx_pipestat_irq_handler(struct drm_i915_private *i915, u16 iir, u32 pipe_stats[I915_MAX_PIPES]); +void intel_display_irq_init(struct drm_i915_private *i915); + #endif /* __INTEL_DISPLAY_IRQ_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c index db5437043904..9e01054c2430 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.c +++ b/drivers/gpu/drm/i915/display/intel_display_power.c @@ -10,6 +10,7 @@ #include "i915_reg.h" #include "intel_backlight_regs.h" #include "intel_cdclk.h" +#include "intel_clock_gating.h" #include "intel_combo_phy.h" #include "intel_de.h" #include "intel_display_power.h" @@ -457,6 +458,17 @@ async_put_domains_clear_domain(struct i915_power_domains *power_domains, clear_bit(domain, power_domains->async_put_domains[1].bits); } +static void +cancel_async_put_work(struct i915_power_domains *power_domains, bool sync) +{ + if (sync) + cancel_delayed_work_sync(&power_domains->async_put_work); + else + cancel_delayed_work(&power_domains->async_put_work); + + power_domains->async_put_next_delay = 0; +} + static bool intel_display_power_grab_async_put_ref(struct drm_i915_private *dev_priv, enum intel_display_power_domain domain) @@ -477,7 +489,7 @@ intel_display_power_grab_async_put_ref(struct drm_i915_private *dev_priv, if (!bitmap_empty(async_put_mask.bits, POWER_DOMAIN_NUM)) goto out_verify; - cancel_delayed_work(&power_domains->async_put_work); + cancel_async_put_work(power_domains, false); intel_runtime_pm_put_raw(&dev_priv->runtime_pm, fetch_and_zero(&power_domains->async_put_wakeref)); out_verify: @@ -608,7 +620,8 @@ static void __intel_display_power_put(struct drm_i915_private *dev_priv, static void queue_async_put_domains_work(struct i915_power_domains *power_domains, - intel_wakeref_t wakeref) + intel_wakeref_t wakeref, + int delay_ms) { struct drm_i915_private *i915 = container_of(power_domains, struct drm_i915_private, @@ -617,7 +630,7 @@ queue_async_put_domains_work(struct i915_power_domains *power_domains, power_domains->async_put_wakeref = wakeref; drm_WARN_ON(&i915->drm, !queue_delayed_work(system_unbound_wq, &power_domains->async_put_work, - msecs_to_jiffies(100))); + msecs_to_jiffies(delay_ms))); } static void @@ -680,13 +693,15 @@ intel_display_power_put_async_work(struct work_struct *work) bitmap_zero(power_domains->async_put_domains[1].bits, POWER_DOMAIN_NUM); queue_async_put_domains_work(power_domains, - fetch_and_zero(&new_work_wakeref)); + fetch_and_zero(&new_work_wakeref), + power_domains->async_put_next_delay); + power_domains->async_put_next_delay = 0; } else { /* * Cancel the work that got queued after this one got dequeued, * since here we released the corresponding async-put reference. */ - cancel_delayed_work(&power_domains->async_put_work); + cancel_async_put_work(power_domains, false); } out_verify: @@ -705,19 +720,25 @@ out_verify: * @i915: i915 device instance * @domain: power domain to reference * @wakeref: wakeref acquired for the reference that is being released + * @delay_ms: delay of powering down the power domain * * This function drops the power domain reference obtained by * intel_display_power_get*() and schedules a work to power down the * corresponding hardware block if this is the last reference. + * The power down is delayed by @delay_ms if this is >= 0, or by a default + * 100 ms otherwise. */ void __intel_display_power_put_async(struct drm_i915_private *i915, enum intel_display_power_domain domain, - intel_wakeref_t wakeref) + intel_wakeref_t wakeref, + int delay_ms) { struct i915_power_domains *power_domains = &i915->display.power.domains; struct intel_runtime_pm *rpm = &i915->runtime_pm; intel_wakeref_t work_wakeref = intel_runtime_pm_get_raw(rpm); + delay_ms = delay_ms >= 0 ? delay_ms : 100; + mutex_lock(&power_domains->lock); if (power_domains->domain_use_count[domain] > 1) { @@ -731,10 +752,13 @@ void __intel_display_power_put_async(struct drm_i915_private *i915, /* Let a pending work requeue itself or queue a new one. */ if (power_domains->async_put_wakeref) { set_bit(domain, power_domains->async_put_domains[1].bits); + power_domains->async_put_next_delay = max(power_domains->async_put_next_delay, + delay_ms); } else { set_bit(domain, power_domains->async_put_domains[0].bits); queue_async_put_domains_work(power_domains, - fetch_and_zero(&work_wakeref)); + fetch_and_zero(&work_wakeref), + delay_ms); } out_verify: @@ -774,7 +798,7 @@ void intel_display_power_flush_work(struct drm_i915_private *i915) async_put_domains_mask(power_domains, &async_put_mask); release_async_put_domains(power_domains, &async_put_mask); - cancel_delayed_work(&power_domains->async_put_work); + cancel_async_put_work(power_domains, false); out_verify: verify_async_put_domains_state(power_domains); @@ -798,7 +822,7 @@ intel_display_power_flush_work_sync(struct drm_i915_private *i915) struct i915_power_domains *power_domains = &i915->display.power.domains; intel_display_power_flush_work(i915); - cancel_delayed_work_sync(&power_domains->async_put_work); + cancel_async_put_work(power_domains, true); verify_async_put_domains_state(power_domains); @@ -1385,9 +1409,8 @@ static void hsw_disable_pc8(struct drm_i915_private *dev_priv) hsw_restore_lcpll(dev_priv); intel_init_pch_refclk(dev_priv); - if (HAS_PCH_LPT_LP(dev_priv)) - intel_de_rmw(dev_priv, SOUTH_DSPCLK_GATE_D, - 0, PCH_LP_PARTITION_LEVEL_DISABLE); + /* Many display registers don't survive PC8+ */ + intel_clock_gating_init(dev_priv); } static void intel_pch_reset_handshake(struct drm_i915_private *dev_priv, @@ -1586,7 +1609,7 @@ static void tgl_bw_buddy_init(struct drm_i915_private *dev_priv) return; if (IS_ALDERLAKE_S(dev_priv) || - IS_RKL_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) + (IS_ROCKETLAKE(dev_priv) && IS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0))) /* Wa_1409767108 */ table = wa_1409767108_buddy_page_masks; else diff --git a/drivers/gpu/drm/i915/display/intel_display_power.h b/drivers/gpu/drm/i915/display/intel_display_power.h index df38632c6237..d3b5d04b7b07 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power.h +++ b/drivers/gpu/drm/i915/display/intel_display_power.h @@ -12,9 +12,6 @@ #include "intel_wakeref.h" enum aux_ch; -enum dpio_channel; -enum dpio_phy; -enum i915_drm_suspend_mode; enum port; struct drm_i915_private; struct i915_power_well; @@ -154,6 +151,7 @@ struct i915_power_domains { struct delayed_work async_put_work; intel_wakeref_t async_put_wakeref; struct intel_power_domain_mask async_put_domains[2]; + int async_put_next_delay; struct i915_power_well *power_wells; }; @@ -200,7 +198,8 @@ intel_display_power_get_if_enabled(struct drm_i915_private *dev_priv, enum intel_display_power_domain domain); void __intel_display_power_put_async(struct drm_i915_private *i915, enum intel_display_power_domain domain, - intel_wakeref_t wakeref); + intel_wakeref_t wakeref, + int delay_ms); void intel_display_power_flush_work(struct drm_i915_private *i915); #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_RUNTIME_PM) void intel_display_power_put(struct drm_i915_private *dev_priv, @@ -211,7 +210,16 @@ intel_display_power_put_async(struct drm_i915_private *i915, enum intel_display_power_domain domain, intel_wakeref_t wakeref) { - __intel_display_power_put_async(i915, domain, wakeref); + __intel_display_power_put_async(i915, domain, wakeref, -1); +} + +static inline void +intel_display_power_put_async_delay(struct drm_i915_private *i915, + enum intel_display_power_domain domain, + intel_wakeref_t wakeref, + int delay_ms) +{ + __intel_display_power_put_async(i915, domain, wakeref, delay_ms); } #else void intel_display_power_put_unchecked(struct drm_i915_private *dev_priv, @@ -230,7 +238,16 @@ intel_display_power_put_async(struct drm_i915_private *i915, enum intel_display_power_domain domain, intel_wakeref_t wakeref) { - __intel_display_power_put_async(i915, domain, -1); + __intel_display_power_put_async(i915, domain, -1, -1); +} + +static inline void +intel_display_power_put_async_delay(struct drm_i915_private *i915, + enum intel_display_power_domain domain, + intel_wakeref_t wakeref, + int delay_ms) +{ + __intel_display_power_put_async(i915, domain, -1, delay_ms); } #endif diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.h b/drivers/gpu/drm/i915/display/intel_display_power_well.h index 1015bba4af01..a8736588314d 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_well.h +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.h @@ -11,7 +11,6 @@ #include "intel_dpio_phy.h" struct drm_i915_private; -struct i915_power_well; struct i915_power_well_ops; struct intel_encoder; diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index 9f40da20e88d..e0e4cb529284 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -500,7 +500,7 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp) else if (IS_ALDERLAKE_P(dev_priv) || IS_ALDERLAKE_S(dev_priv) || IS_DG1(dev_priv) || IS_ROCKETLAKE(dev_priv)) max_rate = 810000; - else if (IS_JSL_EHL(dev_priv)) + else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) max_rate = ehl_max_source_rate(intel_dp); else max_rate = icl_max_source_rate(intel_dp); @@ -510,7 +510,7 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp) } else if (DISPLAY_VER(dev_priv) == 9) { source_rates = skl_rates; size = ARRAY_SIZE(skl_rates); - } else if ((IS_HASWELL(dev_priv) && !IS_HSW_ULX(dev_priv)) || + } else if ((IS_HASWELL(dev_priv) && !IS_HASWELL_ULX(dev_priv)) || IS_BROADWELL(dev_priv)) { source_rates = hsw_rates; size = ARRAY_SIZE(hsw_rates); @@ -713,9 +713,18 @@ u32 intel_dp_dsc_nearest_valid_bpp(struct drm_i915_private *i915, u32 bpp, u32 p /* * According to BSpec, 27 is the max DSC output bpp, - * 8 is the min DSC output bpp + * 8 is the min DSC output bpp. + * While we can still clamp higher bpp values to 27, saving bandwidth, + * if it is required to oompress up to bpp < 8, means we can't do + * that and probably means we can't fit the required mode, even with + * DSC enabled. */ - bits_per_pixel = clamp_t(u32, bits_per_pixel, 8, 27); + if (bits_per_pixel < 8) { + drm_dbg_kms(&i915->drm, "Unsupported BPP %u, min 8\n", + bits_per_pixel); + return 0; + } + bits_per_pixel = min_t(u32, bits_per_pixel, 27); } else { /* Find the nearest match in the array of known BPPs from VESA */ for (i = 0; i < ARRAY_SIZE(valid_dsc_bpp) - 1; i++) { @@ -5503,8 +5512,13 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp, /* * VBT and straps are liars. Also check HPD as that seems * to be the most reliable piece of information available. + * + * ... expect on devices that forgot to hook HPD up for eDP + * (eg. Acer Chromebook C710), so we'll check it only if multiple + * ports are attempting to use the same AUX CH, according to VBT. */ - if (!intel_digital_port_connected(encoder)) { + if (intel_bios_dp_has_shared_aux_ch(encoder->devdata) && + !intel_digital_port_connected(encoder)) { /* * If this fails, presume the DPCD answer came * from some other port using the same AUX CH. diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c index 197c6e81db14..2d173bd495a3 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c @@ -14,7 +14,7 @@ #include "intel_pps.h" #include "intel_tc.h" -static u32 intel_dp_aux_pack(const u8 *src, int src_bytes) +u32 intel_dp_aux_pack(const u8 *src, int src_bytes) { int i; u32 v = 0; @@ -792,25 +792,60 @@ static enum aux_ch default_aux_ch(struct intel_encoder *encoder) return (enum aux_ch)encoder->port; } +static struct intel_encoder * +get_encoder_by_aux_ch(struct intel_encoder *encoder, + enum aux_ch aux_ch) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_encoder *other; + + for_each_intel_encoder(&i915->drm, other) { + if (other == encoder) + continue; + + if (!intel_encoder_is_dig_port(other)) + continue; + + if (enc_to_dig_port(other)->aux_ch == aux_ch) + return other; + } + + return NULL; +} + enum aux_ch intel_dp_aux_ch(struct intel_encoder *encoder) { struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_encoder *other; + const char *source; enum aux_ch aux_ch; aux_ch = intel_bios_dp_aux_ch(encoder->devdata); - if (aux_ch != AUX_CH_NONE) { - drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] using AUX %c (VBT)\n", - encoder->base.base.id, encoder->base.name, - aux_ch_name(aux_ch)); - return aux_ch; + source = "VBT"; + + if (aux_ch == AUX_CH_NONE) { + aux_ch = default_aux_ch(encoder); + source = "platform default"; } - aux_ch = default_aux_ch(encoder); + if (aux_ch == AUX_CH_NONE) + return AUX_CH_NONE; + + /* FIXME validate aux_ch against platform caps */ + + other = get_encoder_by_aux_ch(encoder, aux_ch); + if (other) { + drm_dbg_kms(&i915->drm, + "[ENCODER:%d:%s] AUX CH %c already claimed by [ENCODER:%d:%s]\n", + encoder->base.base.id, encoder->base.name, aux_ch_name(aux_ch), + other->base.base.id, other->base.name); + return AUX_CH_NONE; + } drm_dbg_kms(&i915->drm, - "[ENCODER:%d:%s] using AUX %c (platform default)\n", + "[ENCODER:%d:%s] Using AUX CH %c (%s)\n", encoder->base.base.id, encoder->base.name, - aux_ch_name(aux_ch)); + aux_ch_name(aux_ch), source); return aux_ch; } diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.h b/drivers/gpu/drm/i915/display/intel_dp_aux.h index 5b608f9d3499..8447f3e601fe 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.h +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.h @@ -6,6 +6,8 @@ #ifndef __INTEL_DP_AUX_H__ #define __INTEL_DP_AUX_H__ +#include <linux/types.h> + enum aux_ch; struct drm_i915_private; struct intel_dp; @@ -17,5 +19,6 @@ void intel_dp_aux_init(struct intel_dp *intel_dp); enum aux_ch intel_dp_aux_ch(struct intel_encoder *encoder); void intel_dp_aux_irq_handler(struct drm_i915_private *i915); +u32 intel_dp_aux_pack(const u8 *src, int src_bytes); #endif /* __INTEL_DP_AUX_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index 6b2d8a1e2aa9..6d68b36292d3 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -21,6 +21,7 @@ * DEALINGS IN THE SOFTWARE. */ +#include <linux/math.h> #include <linux/string_helpers.h> #include "i915_reg.h" @@ -191,7 +192,8 @@ intel_combo_pll_enable_reg(struct drm_i915_private *i915, { if (IS_DG1(i915)) return DG1_DPLL_ENABLE(pll->info->id); - else if (IS_JSL_EHL(i915) && (pll->info->id == DPLL_ID_EHL_DPLL4)) + else if ((IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) && + (pll->info->id == DPLL_ID_EHL_DPLL4)) return MG_PLL_ENABLE(0); return ICL_DPLL_ENABLE(pll->info->id); @@ -927,7 +929,7 @@ static int hsw_ddi_wrpll_get_freq(struct drm_i915_private *dev_priv, switch (wrpll & WRPLL_REF_MASK) { case WRPLL_REF_SPECIAL_HSW: /* Muxed-SSC for BDW, non-SSC for non-ULT HSW. */ - if (IS_HASWELL(dev_priv) && !IS_HSW_ULT(dev_priv)) { + if (IS_HASWELL(dev_priv) && !IS_HASWELL_ULT(dev_priv)) { refclk = dev_priv->display.dpll.ref_clks.nssc; break; } @@ -2460,8 +2462,8 @@ static void icl_wrpll_params_populate(struct skl_wrpll_params *params, static bool ehl_combo_pll_div_frac_wa_needed(struct drm_i915_private *i915) { - return ((IS_PLATFORM(i915, INTEL_ELKHARTLAKE) && - IS_JSL_EHL_DISPLAY_STEP(i915, STEP_B0, STEP_FOREVER)) || + return (((IS_ELKHARTLAKE(i915) || IS_JASPERLAKE(i915)) && + IS_DISPLAY_STEP(i915, STEP_B0, STEP_FOREVER)) || IS_TIGERLAKE(i915) || IS_ALDERLAKE_S(i915) || IS_ALDERLAKE_P(i915)) && i915->display.dpll.ref_clks.nssc == 38400; } @@ -3226,7 +3228,8 @@ static int icl_get_combo_phy_dpll(struct intel_atomic_state *state, BIT(DPLL_ID_EHL_DPLL4) | BIT(DPLL_ID_ICL_DPLL1) | BIT(DPLL_ID_ICL_DPLL0); - } else if (IS_JSL_EHL(dev_priv) && port != PORT_A) { + } else if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && + port != PORT_A) { dpll_mask = BIT(DPLL_ID_EHL_DPLL4) | BIT(DPLL_ID_ICL_DPLL1) | @@ -3567,7 +3570,8 @@ static bool icl_pll_get_hw_state(struct drm_i915_private *dev_priv, hw_state->div0 &= TGL_DPLL0_DIV0_AFC_STARTUP_MASK; } } else { - if (IS_JSL_EHL(dev_priv) && id == DPLL_ID_EHL_DPLL4) { + if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && + id == DPLL_ID_EHL_DPLL4) { hw_state->cfgcr0 = intel_de_read(dev_priv, ICL_DPLL_CFGCR0(4)); hw_state->cfgcr1 = intel_de_read(dev_priv, @@ -3623,7 +3627,8 @@ static void icl_dpll_write(struct drm_i915_private *dev_priv, cfgcr1_reg = TGL_DPLL_CFGCR1(id); div0_reg = TGL_DPLL0_DIV0(id); } else { - if (IS_JSL_EHL(dev_priv) && id == DPLL_ID_EHL_DPLL4) { + if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && + id == DPLL_ID_EHL_DPLL4) { cfgcr0_reg = ICL_DPLL_CFGCR0(4); cfgcr1_reg = ICL_DPLL_CFGCR1(4); } else { @@ -3781,7 +3786,7 @@ static void adlp_cmtg_clock_gating_wa(struct drm_i915_private *i915, struct inte { u32 val; - if (!IS_ADLP_DISPLAY_STEP(i915, STEP_A0, STEP_B0) || + if (!(IS_ALDERLAKE_P(i915) && IS_DISPLAY_STEP(i915, STEP_A0, STEP_B0)) || pll->info->id != DPLL_ID_ICL_DPLL0) return; /* @@ -3806,7 +3811,7 @@ static void combo_pll_enable(struct drm_i915_private *dev_priv, { i915_reg_t enable_reg = intel_combo_pll_enable_reg(dev_priv, pll); - if (IS_JSL_EHL(dev_priv) && + if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && pll->info->id == DPLL_ID_EHL_DPLL4) { /* @@ -3914,7 +3919,7 @@ static void combo_pll_disable(struct drm_i915_private *dev_priv, icl_pll_disable(dev_priv, pll, enable_reg); - if (IS_JSL_EHL(dev_priv) && + if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && pll->info->id == DPLL_ID_EHL_DPLL4) intel_display_power_put(dev_priv, POWER_DOMAIN_DC_OFF, pll->wakeref); @@ -4150,7 +4155,7 @@ void intel_shared_dpll_init(struct drm_i915_private *dev_priv) dpll_mgr = &rkl_pll_mgr; else if (DISPLAY_VER(dev_priv) >= 12) dpll_mgr = &tgl_pll_mgr; - else if (IS_JSL_EHL(dev_priv)) + else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) dpll_mgr = &ehl_pll_mgr; else if (DISPLAY_VER(dev_priv) >= 11) dpll_mgr = &icl_pll_mgr; @@ -4335,7 +4340,8 @@ static void readout_dpll_hw_state(struct drm_i915_private *i915, pll->on = intel_dpll_get_hw_state(i915, pll, &pll->state.hw_state); - if (IS_JSL_EHL(i915) && pll->on && + if ((IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) && + pll->on && pll->info->id == DPLL_ID_EHL_DPLL4) { pll->wakeref = intel_display_power_get(i915, POWER_DOMAIN_DC_OFF); diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h index ba62eb5d7c51..04e6810954b2 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h @@ -29,13 +29,6 @@ #include "intel_wakeref.h" -/*FIXME: Move this to a more appropriate place. */ -#define abs_diff(a, b) ({ \ - typeof(a) __a = (a); \ - typeof(b) __b = (b); \ - (void) (&__a == &__b); \ - __a > __b ? (__a - __b) : (__b - __a); }) - enum tc_port; struct drm_i915_private; struct intel_atomic_state; diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c index 7c5fddb203ba..fbfd8f959f17 100644 --- a/drivers/gpu/drm/i915/display/intel_dpt.c +++ b/drivers/gpu/drm/i915/display/intel_dpt.c @@ -166,6 +166,8 @@ struct i915_vma *intel_dpt_pin(struct i915_address_space *vm) i915_vma_get(vma); } + dpt->obj->mm.dirty = true; + atomic_dec(&i915->gpu_error.pending_fb_pin); intel_runtime_pm_put(&i915->runtime_pm, wakeref); @@ -261,7 +263,7 @@ intel_dpt_create(struct intel_framebuffer *fb) dpt_obj = i915_gem_object_create_stolen(i915, size); if (IS_ERR(dpt_obj) && !HAS_LMEM(i915)) { drm_dbg_kms(&i915->drm, "Allocating dpt from smem\n"); - dpt_obj = i915_gem_object_create_internal(i915, size); + dpt_obj = i915_gem_object_create_shmem(i915, size); } if (IS_ERR(dpt_obj)) return ERR_CAST(dpt_obj); diff --git a/drivers/gpu/drm/i915/display/intel_dsi.c b/drivers/gpu/drm/i915/display/intel_dsi.c index 5efdd471ac2b..d3cf6a652221 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi.c +++ b/drivers/gpu/drm/i915/display/intel_dsi.c @@ -9,6 +9,26 @@ #include "intel_dsi.h" #include "intel_panel.h" +void intel_dsi_wait_panel_power_cycle(struct intel_dsi *intel_dsi) +{ + ktime_t panel_power_on_time; + s64 panel_power_off_duration; + + panel_power_on_time = ktime_get_boottime(); + panel_power_off_duration = ktime_ms_delta(panel_power_on_time, + intel_dsi->panel_power_off_time); + + if (panel_power_off_duration < (s64)intel_dsi->panel_pwr_cycle_delay) + msleep(intel_dsi->panel_pwr_cycle_delay - panel_power_off_duration); +} + +void intel_dsi_shutdown(struct intel_encoder *encoder) +{ + struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); + + intel_dsi_wait_panel_power_cycle(intel_dsi); +} + int intel_dsi_bitrate(const struct intel_dsi *intel_dsi) { int bpp = mipi_dsi_pixel_format_to_bpp(intel_dsi->pixel_format); diff --git a/drivers/gpu/drm/i915/display/intel_dsi.h b/drivers/gpu/drm/i915/display/intel_dsi.h index ce80bd8be519..083390e5e442 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi.h +++ b/drivers/gpu/drm/i915/display/intel_dsi.h @@ -173,5 +173,7 @@ enum drm_mode_status intel_dsi_mode_valid(struct drm_connector *connector, struct intel_dsi_host *intel_dsi_host_init(struct intel_dsi *intel_dsi, const struct mipi_dsi_host_ops *funcs, enum port port); +void intel_dsi_wait_panel_power_cycle(struct intel_dsi *intel_dsi); +void intel_dsi_shutdown(struct intel_encoder *encoder); #endif /* _INTEL_DSI_H */ diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c index c7935ea498c4..e56ec3f2d84a 100644 --- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c @@ -235,7 +235,7 @@ static const u8 *mipi_exec_delay(struct intel_dsi *intel_dsi, const u8 *data) struct drm_i915_private *i915 = to_i915(intel_dsi->base.base.dev); u32 delay = *((const u32 *) data); - drm_dbg_kms(&i915->drm, "\n"); + drm_dbg_kms(&i915->drm, "%d usecs\n", delay); usleep_range(delay, delay + 10); data += 4; diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c index 9884678743b6..b386894c3a6d 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo.c +++ b/drivers/gpu/drm/i915/display/intel_dvo.c @@ -509,6 +509,8 @@ void intel_dvo_init(struct drm_i915_private *i915) return; } + assert_port_valid(i915, intel_dvo->dev.port); + encoder->type = INTEL_OUTPUT_DVO; encoder->power_domain = POWER_DOMAIN_PORT_OTHER; encoder->port = intel_dvo->dev.port; diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index 7f8b2d7713c7..25382022cd27 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -47,6 +47,7 @@ #include "i915_reg.h" #include "i915_utils.h" #include "i915_vgpu.h" +#include "i915_vma.h" #include "intel_cdclk.h" #include "intel_de.h" #include "intel_display_trace.h" @@ -94,8 +95,7 @@ struct intel_fbc { struct mutex lock; unsigned int busy_bits; - struct drm_mm_node compressed_fb; - struct drm_mm_node compressed_llb; + struct i915_stolen_fb compressed_fb, compressed_llb; enum intel_fbc_id id; @@ -332,15 +332,16 @@ static void i8xx_fbc_program_cfb(struct intel_fbc *fbc) { struct drm_i915_private *i915 = fbc->i915; - GEM_BUG_ON(range_overflows_end_t(u64, i915->dsm.stolen.start, - fbc->compressed_fb.start, U32_MAX)); - GEM_BUG_ON(range_overflows_end_t(u64, i915->dsm.stolen.start, - fbc->compressed_llb.start, U32_MAX)); - + GEM_BUG_ON(range_overflows_end_t(u64, i915_gem_stolen_area_address(i915), + i915_gem_stolen_node_offset(&fbc->compressed_fb), + U32_MAX)); + GEM_BUG_ON(range_overflows_end_t(u64, i915_gem_stolen_area_address(i915), + i915_gem_stolen_node_offset(&fbc->compressed_llb), + U32_MAX)); intel_de_write(i915, FBC_CFB_BASE, - i915->dsm.stolen.start + fbc->compressed_fb.start); + i915_gem_stolen_node_address(i915, &fbc->compressed_fb)); intel_de_write(i915, FBC_LL_BASE, - i915->dsm.stolen.start + fbc->compressed_llb.start); + i915_gem_stolen_node_address(i915, &fbc->compressed_llb)); } static const struct intel_fbc_funcs i8xx_fbc_funcs = { @@ -447,7 +448,8 @@ static void g4x_fbc_program_cfb(struct intel_fbc *fbc) { struct drm_i915_private *i915 = fbc->i915; - intel_de_write(i915, DPFC_CB_BASE, fbc->compressed_fb.start); + intel_de_write(i915, DPFC_CB_BASE, + i915_gem_stolen_node_offset(&fbc->compressed_fb)); } static const struct intel_fbc_funcs g4x_fbc_funcs = { @@ -498,7 +500,8 @@ static void ilk_fbc_program_cfb(struct intel_fbc *fbc) { struct drm_i915_private *i915 = fbc->i915; - intel_de_write(i915, ILK_DPFC_CB_BASE(fbc->id), fbc->compressed_fb.start); + intel_de_write(i915, ILK_DPFC_CB_BASE(fbc->id), + i915_gem_stolen_node_offset(&fbc->compressed_fb)); } static const struct intel_fbc_funcs ilk_fbc_funcs = { @@ -605,7 +608,7 @@ static void ivb_fbc_activate(struct intel_fbc *fbc) else if (DISPLAY_VER(i915) == 9) skl_fbc_program_cfb_stride(fbc); - if (to_gt(i915)->ggtt->num_fences) + if (intel_gt_support_legacy_fencing(to_gt(i915))) snb_fbc_program_fence(fbc); intel_de_write(i915, ILK_DPFC_CONTROL(fbc->id), @@ -713,7 +716,7 @@ static u64 intel_fbc_stolen_end(struct drm_i915_private *i915) * underruns, even if that range is not reserved by the BIOS. */ if (IS_BROADWELL(i915) || (DISPLAY_VER(i915) == 9 && !IS_BROXTON(i915))) - end = resource_size(&i915->dsm.stolen) - 8 * 1024 * 1024; + end = i915_gem_stolen_area_size(i915) - 8 * 1024 * 1024; else end = U64_MAX; @@ -770,9 +773,9 @@ static int intel_fbc_alloc_cfb(struct intel_fbc *fbc, int ret; drm_WARN_ON(&i915->drm, - drm_mm_node_allocated(&fbc->compressed_fb)); + i915_gem_stolen_node_allocated(&fbc->compressed_fb)); drm_WARN_ON(&i915->drm, - drm_mm_node_allocated(&fbc->compressed_llb)); + i915_gem_stolen_node_allocated(&fbc->compressed_llb)); if (DISPLAY_VER(i915) < 5 && !IS_G4X(i915)) { ret = i915_gem_stolen_insert_node(i915, &fbc->compressed_llb, @@ -792,15 +795,14 @@ static int intel_fbc_alloc_cfb(struct intel_fbc *fbc, drm_dbg_kms(&i915->drm, "reserved %llu bytes of contiguous stolen space for FBC, limit: %d\n", - fbc->compressed_fb.size, fbc->limit); - + i915_gem_stolen_node_size(&fbc->compressed_fb), fbc->limit); return 0; err_llb: - if (drm_mm_node_allocated(&fbc->compressed_llb)) + if (i915_gem_stolen_node_allocated(&fbc->compressed_llb)) i915_gem_stolen_remove_node(i915, &fbc->compressed_llb); err: - if (drm_mm_initialized(&i915->mm.stolen)) + if (i915_gem_stolen_initialized(i915)) drm_info_once(&i915->drm, "not enough stolen space for compressed buffer (need %d more bytes), disabling. Hint: you may be able to increase stolen memory size in the BIOS to avoid this.\n", size); return -ENOSPC; } @@ -825,9 +827,9 @@ static void __intel_fbc_cleanup_cfb(struct intel_fbc *fbc) if (WARN_ON(intel_fbc_hw_is_active(fbc))) return; - if (drm_mm_node_allocated(&fbc->compressed_llb)) + if (i915_gem_stolen_node_allocated(&fbc->compressed_llb)) i915_gem_stolen_remove_node(i915, &fbc->compressed_llb); - if (drm_mm_node_allocated(&fbc->compressed_fb)) + if (i915_gem_stolen_node_allocated(&fbc->compressed_fb)) i915_gem_stolen_remove_node(i915, &fbc->compressed_fb); } @@ -990,11 +992,10 @@ static void intel_fbc_update_state(struct intel_atomic_state *state, fbc_state->fence_y_offset = intel_plane_fence_y_offset(plane_state); drm_WARN_ON(&i915->drm, plane_state->flags & PLANE_HAS_FENCE && - !plane_state->ggtt_vma->fence); + !intel_gt_support_legacy_fencing(to_gt(i915))); - if (plane_state->flags & PLANE_HAS_FENCE && - plane_state->ggtt_vma->fence) - fbc_state->fence_id = plane_state->ggtt_vma->fence->id; + if (plane_state->flags & PLANE_HAS_FENCE) + fbc_state->fence_id = i915_vma_fence_id(plane_state->ggtt_vma); else fbc_state->fence_id = -1; @@ -1021,7 +1022,7 @@ static bool intel_fbc_is_fence_ok(const struct intel_plane_state *plane_state) */ return DISPLAY_VER(i915) >= 9 || (plane_state->flags & PLANE_HAS_FENCE && - plane_state->ggtt_vma->fence); + i915_vma_fence_id(plane_state->ggtt_vma) != -1); } static bool intel_fbc_is_cfb_ok(const struct intel_plane_state *plane_state) @@ -1030,7 +1031,8 @@ static bool intel_fbc_is_cfb_ok(const struct intel_plane_state *plane_state) struct intel_fbc *fbc = plane->fbc; return intel_fbc_min_limit(plane_state) <= fbc->limit && - intel_fbc_cfb_size(plane_state) <= fbc->compressed_fb.size * fbc->limit; + intel_fbc_cfb_size(plane_state) <= fbc->limit * + i915_gem_stolen_node_size(&fbc->compressed_fb); } static bool intel_fbc_is_ok(const struct intel_plane_state *plane_state) @@ -1054,6 +1056,11 @@ static int intel_fbc_check_plane(struct intel_atomic_state *state, if (!fbc) return 0; + if (!i915_gem_stolen_initialized(i915)) { + plane_state->no_fbc_reason = "stolen memory not initialised"; + return 0; + } + if (intel_vgpu_active(i915)) { plane_state->no_fbc_reason = "VGPU active"; return 0; @@ -1707,9 +1714,6 @@ void intel_fbc_init(struct drm_i915_private *i915) { enum intel_fbc_id fbc_id; - if (!drm_mm_initialized(&i915->mm.stolen)) - DISPLAY_RUNTIME_INFO(i915)->fbc_mask = 0; - if (need_fbc_vtd_wa(i915)) DISPLAY_RUNTIME_INFO(i915)->fbc_mask = 0; diff --git a/drivers/gpu/drm/i915/display/intel_fbdev.c b/drivers/gpu/drm/i915/display/intel_fbdev.c index f1439827bc80..31d0d695d567 100644 --- a/drivers/gpu/drm/i915/display/intel_fbdev.c +++ b/drivers/gpu/drm/i915/display/intel_fbdev.c @@ -135,9 +135,6 @@ static int intel_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma) return i915_gem_fb_mmap(obj, vma); } -__diag_push(); -__diag_ignore_all("-Woverride-init", "Allow overriding the default ops"); - static const struct fb_ops intelfb_ops = { .owner = THIS_MODULE, __FB_DEFAULT_DEFERRED_OPS_RDWR(intel_fbdev), @@ -149,8 +146,6 @@ static const struct fb_ops intelfb_ops = { .fb_mmap = intel_fbdev_mmap, }; -__diag_pop(); - static int intelfb_alloc(struct drm_fb_helper *helper, struct drm_fb_helper_surface_size *sizes) { @@ -187,8 +182,10 @@ static int intelfb_alloc(struct drm_fb_helper *helper, * If the FB is too big, just don't use it since fbdev is not very * important and we should probably use that space with FBC or other * features. + * + * Also skip stolen on MTL as Wa_22018444074 mitigation. */ - if (size * 2 < dev_priv->dsm.usable_size) + if (!(IS_METEORLAKE(dev_priv)) && size * 2 < dev_priv->dsm.usable_size) obj = i915_gem_object_create_stolen(dev_priv, size); if (IS_ERR(obj)) obj = i915_gem_object_create_shmem(dev_priv, size); diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.c b/drivers/gpu/drm/i915/display/intel_frontbuffer.c index 17a7aa8b28c2..22392f94b626 100644 --- a/drivers/gpu/drm/i915/display/intel_frontbuffer.c +++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.c @@ -167,7 +167,7 @@ void __intel_fb_invalidate(struct intel_frontbuffer *front, enum fb_op_origin origin, unsigned int frontbuffer_bits) { - struct drm_i915_private *i915 = to_i915(front->obj->base.dev); + struct drm_i915_private *i915 = intel_bo_to_i915(front->obj); if (origin == ORIGIN_CS) { spin_lock(&i915->display.fb_tracking.lock); @@ -188,7 +188,7 @@ void __intel_fb_flush(struct intel_frontbuffer *front, enum fb_op_origin origin, unsigned int frontbuffer_bits) { - struct drm_i915_private *i915 = to_i915(front->obj->base.dev); + struct drm_i915_private *i915 = intel_bo_to_i915(front->obj); if (origin == ORIGIN_CS) { spin_lock(&i915->display.fb_tracking.lock); @@ -221,24 +221,18 @@ static void frontbuffer_retire(struct i915_active *ref) } static void frontbuffer_release(struct kref *ref) - __releases(&to_i915(front->obj->base.dev)->display.fb_tracking.lock) + __releases(&intel_bo_to_i915(front->obj)->display.fb_tracking.lock) { struct intel_frontbuffer *front = container_of(ref, typeof(*front), ref); struct drm_i915_gem_object *obj = front->obj; - struct i915_vma *vma; - drm_WARN_ON(obj->base.dev, atomic_read(&front->bits)); + drm_WARN_ON(&intel_bo_to_i915(obj)->drm, atomic_read(&front->bits)); - spin_lock(&obj->vma.lock); - for_each_ggtt_vma(vma, obj) { - i915_vma_clear_scanout(vma); - vma->display_alignment = I915_GTT_MIN_ALIGNMENT; - } - spin_unlock(&obj->vma.lock); + i915_ggtt_clear_scanout(obj); - RCU_INIT_POINTER(obj->frontbuffer, NULL); - spin_unlock(&to_i915(obj->base.dev)->display.fb_tracking.lock); + i915_gem_object_set_frontbuffer(obj, NULL); + spin_unlock(&intel_bo_to_i915(obj)->display.fb_tracking.lock); i915_active_fini(&front->write); @@ -249,10 +243,10 @@ static void frontbuffer_release(struct kref *ref) struct intel_frontbuffer * intel_frontbuffer_get(struct drm_i915_gem_object *obj) { - struct drm_i915_private *i915 = to_i915(obj->base.dev); - struct intel_frontbuffer *front; + struct drm_i915_private *i915 = intel_bo_to_i915(obj); + struct intel_frontbuffer *front, *cur; - front = __intel_frontbuffer_get(obj); + front = i915_gem_object_get_frontbuffer(obj); if (front) return front; @@ -269,24 +263,18 @@ intel_frontbuffer_get(struct drm_i915_gem_object *obj) I915_ACTIVE_RETIRE_SLEEPS); spin_lock(&i915->display.fb_tracking.lock); - if (rcu_access_pointer(obj->frontbuffer)) { - kfree(front); - front = rcu_dereference_protected(obj->frontbuffer, true); - kref_get(&front->ref); - } else { - i915_gem_object_get(obj); - rcu_assign_pointer(obj->frontbuffer, front); - } + cur = i915_gem_object_set_frontbuffer(obj, front); spin_unlock(&i915->display.fb_tracking.lock); - - return front; + if (cur != front) + kfree(front); + return cur; } void intel_frontbuffer_put(struct intel_frontbuffer *front) { kref_put_lock(&front->ref, frontbuffer_release, - &to_i915(front->obj->base.dev)->display.fb_tracking.lock); + &intel_bo_to_i915(front->obj)->display.fb_tracking.lock); } /** @@ -315,13 +303,13 @@ void intel_frontbuffer_track(struct intel_frontbuffer *old, BUILD_BUG_ON(I915_MAX_PLANES > INTEL_FRONTBUFFER_BITS_PER_PIPE); if (old) { - drm_WARN_ON(old->obj->base.dev, + drm_WARN_ON(&intel_bo_to_i915(old->obj)->drm, !(atomic_read(&old->bits) & frontbuffer_bits)); atomic_andnot(frontbuffer_bits, &old->bits); } if (new) { - drm_WARN_ON(new->obj->base.dev, + drm_WARN_ON(&intel_bo_to_i915(new->obj)->drm, atomic_read(&new->bits) & frontbuffer_bits); atomic_or(frontbuffer_bits, &new->bits); } diff --git a/drivers/gpu/drm/i915/display/intel_frontbuffer.h b/drivers/gpu/drm/i915/display/intel_frontbuffer.h index 3c474ed937fb..72d89be3284b 100644 --- a/drivers/gpu/drm/i915/display/intel_frontbuffer.h +++ b/drivers/gpu/drm/i915/display/intel_frontbuffer.h @@ -28,7 +28,6 @@ #include <linux/bits.h> #include <linux/kref.h> -#include "gem/i915_gem_object_types.h" #include "i915_active_types.h" struct drm_i915_private; @@ -75,33 +74,6 @@ void intel_frontbuffer_flip(struct drm_i915_private *i915, void intel_frontbuffer_put(struct intel_frontbuffer *front); -static inline struct intel_frontbuffer * -__intel_frontbuffer_get(const struct drm_i915_gem_object *obj) -{ - struct intel_frontbuffer *front; - - if (likely(!rcu_access_pointer(obj->frontbuffer))) - return NULL; - - rcu_read_lock(); - do { - front = rcu_dereference(obj->frontbuffer); - if (!front) - break; - - if (unlikely(!kref_get_unless_zero(&front->ref))) - continue; - - if (likely(front == rcu_access_pointer(obj->frontbuffer))) - break; - - intel_frontbuffer_put(front); - } while (1); - rcu_read_unlock(); - - return front; -} - struct intel_frontbuffer * intel_frontbuffer_get(struct drm_i915_gem_object *obj); diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 34fabadefaf6..a42549fa9691 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -177,8 +177,11 @@ bool intel_hdcp2_capable(struct intel_connector *connector) struct intel_gt *gt = i915->media_gt; struct intel_gsc_uc *gsc = gt ? >->uc.gsc : NULL; - if (!gsc || !intel_uc_fw_is_running(&gsc->fw)) + if (!gsc || !intel_uc_fw_is_running(&gsc->fw)) { + drm_dbg_kms(&i915->drm, + "GSC components required for HDCP2.2 are not ready\n"); return false; + } } /* MEI/GSC interface is solid depending on which is used */ diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c index 72573ce1d0e9..d753db3eef15 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c @@ -6,6 +6,7 @@ #include <drm/i915_hdcp_interface.h> #include "gem/i915_gem_region.h" +#include "gt/intel_gt.h" #include "gt/uc/intel_gsc_uc_heci_cmd_submit.h" #include "i915_drv.h" #include "i915_utils.h" @@ -621,24 +622,26 @@ static int intel_hdcp_gsc_initialize_message(struct drm_i915_private *i915, struct intel_gt *gt = i915->media_gt; struct drm_i915_gem_object *obj = NULL; struct i915_vma *vma = NULL; - void *cmd; + void *cmd_in, *cmd_out; int err; - /* allocate object of one page for HDCP command memory and store it */ - obj = i915_gem_object_create_shmem(i915, PAGE_SIZE); + /* allocate object of two page for HDCP command memory and store it */ + obj = i915_gem_object_create_shmem(i915, 2 * PAGE_SIZE); if (IS_ERR(obj)) { drm_err(&i915->drm, "Failed to allocate HDCP streaming command!\n"); return PTR_ERR(obj); } - cmd = i915_gem_object_pin_map_unlocked(obj, i915_coherent_map_type(i915, obj, true)); - if (IS_ERR(cmd)) { + cmd_in = i915_gem_object_pin_map_unlocked(obj, intel_gt_coherent_map_type(gt, obj, true)); + if (IS_ERR(cmd_in)) { drm_err(&i915->drm, "Failed to map gsc message page!\n"); - err = PTR_ERR(cmd); + err = PTR_ERR(cmd_in); goto out_unpin; } + cmd_out = cmd_in + PAGE_SIZE; + vma = i915_vma_instance(obj, >->ggtt->vm, NULL); if (IS_ERR(vma)) { err = PTR_ERR(vma); @@ -649,9 +652,10 @@ static int intel_hdcp_gsc_initialize_message(struct drm_i915_private *i915, if (err) goto out_unmap; - memset(cmd, 0, obj->base.size); + memset(cmd_in, 0, obj->base.size); - hdcp_message->hdcp_cmd = cmd; + hdcp_message->hdcp_cmd_in = cmd_in; + hdcp_message->hdcp_cmd_out = cmd_out; hdcp_message->vma = vma; return 0; @@ -691,6 +695,8 @@ static void intel_hdcp_gsc_free_message(struct drm_i915_private *i915) struct intel_hdcp_gsc_message *hdcp_message = i915->display.hdcp.hdcp_message; + hdcp_message->hdcp_cmd_in = NULL; + hdcp_message->hdcp_cmd_out = NULL; i915_vma_unpin_and_release(&hdcp_message->vma, I915_VMA_RELEASE_MAP); kfree(hdcp_message); } @@ -721,38 +727,42 @@ void intel_hdcp_gsc_fini(struct drm_i915_private *i915) } static int intel_gsc_send_sync(struct drm_i915_private *i915, - struct intel_gsc_mtl_header *header, u64 addr, + struct intel_gsc_mtl_header *header_in, + struct intel_gsc_mtl_header *header_out, + u64 addr_in, u64 addr_out, size_t msg_out_len) { struct intel_gt *gt = i915->media_gt; int ret; - header->flags = 0; - ret = intel_gsc_uc_heci_cmd_submit_packet(>->uc.gsc, addr, - header->message_size, - addr, - msg_out_len + sizeof(*header)); + ret = intel_gsc_uc_heci_cmd_submit_packet(>->uc.gsc, addr_in, + header_in->message_size, + addr_out, + msg_out_len + sizeof(*header_out)); if (ret) { drm_err(&i915->drm, "failed to send gsc HDCP msg (%d)\n", ret); return ret; } /* - * Checking validity marker for memory sanity + * Checking validity marker and header status to see if some error has + * blocked us from sending message to gsc cs */ - if (header->validity_marker != GSC_HECI_VALIDITY_MARKER) { + if (header_out->validity_marker != GSC_HECI_VALIDITY_MARKER) { drm_err(&i915->drm, "invalid validity marker\n"); return -EINVAL; } - if (header->status != 0) { + if (header_out->status != 0) { drm_err(&i915->drm, "header status indicates error %d\n", - header->status); + header_out->status); return -EINVAL; } - if (header->flags & GSC_OUTFLAG_MSG_PENDING) + if (header_out->flags & GSC_OUTFLAG_MSG_PENDING) { + header_in->gsc_message_handle = header_out->gsc_message_handle; return -EAGAIN; + } return 0; } @@ -769,11 +779,11 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in, size_t msg_out_len) { struct intel_gt *gt = i915->media_gt; - struct intel_gsc_mtl_header *header; - const size_t max_msg_size = PAGE_SIZE - sizeof(*header); + struct intel_gsc_mtl_header *header_in, *header_out; + const size_t max_msg_size = PAGE_SIZE - sizeof(*header_in); struct intel_hdcp_gsc_message *hdcp_message; - u64 addr, host_session_id; - u32 reply_size, msg_size; + u64 addr_in, addr_out, host_session_id; + u32 reply_size, msg_size_in, msg_size_out; int ret, tries = 0; if (!intel_uc_uses_gsc_uc(>->uc)) @@ -782,16 +792,20 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in, if (msg_in_len > max_msg_size || msg_out_len > max_msg_size) return -ENOSPC; + msg_size_in = msg_in_len + sizeof(*header_in); + msg_size_out = msg_out_len + sizeof(*header_out); hdcp_message = i915->display.hdcp.hdcp_message; - header = hdcp_message->hdcp_cmd; - addr = i915_ggtt_offset(hdcp_message->vma); + header_in = hdcp_message->hdcp_cmd_in; + header_out = hdcp_message->hdcp_cmd_out; + addr_in = i915_ggtt_offset(hdcp_message->vma); + addr_out = addr_in + PAGE_SIZE; - msg_size = msg_in_len + sizeof(*header); - memset(header, 0, msg_size); + memset(header_in, 0, msg_size_in); + memset(header_out, 0, msg_size_out); get_random_bytes(&host_session_id, sizeof(u64)); - intel_gsc_uc_heci_cmd_emit_mtl_header(header, HECI_MEADDRESS_HDCP, - msg_size, host_session_id); - memcpy(hdcp_message->hdcp_cmd + sizeof(*header), msg_in, msg_in_len); + intel_gsc_uc_heci_cmd_emit_mtl_header(header_in, HECI_MEADDRESS_HDCP, + msg_size_in, host_session_id); + memcpy(hdcp_message->hdcp_cmd_in + sizeof(*header_in), msg_in, msg_in_len); /* * Keep sending request in case the pending bit is set no need to add @@ -800,7 +814,8 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in, * 20 times each message 50 ms apart */ do { - ret = intel_gsc_send_sync(i915, header, addr, msg_out_len); + ret = intel_gsc_send_sync(i915, header_in, header_out, addr_in, + addr_out, msg_out_len); /* Only try again if gsc says so */ if (ret != -EAGAIN) @@ -814,7 +829,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in, goto err; /* we use the same mem for the reply, so header is in the same loc */ - reply_size = header->message_size - sizeof(*header); + reply_size = header_out->message_size - sizeof(*header_out); if (reply_size > msg_out_len) { drm_warn(&i915->drm, "caller with insufficient HDCP reply size %u (%d)\n", reply_size, (u32)msg_out_len); @@ -824,7 +839,7 @@ ssize_t intel_hdcp_gsc_msg_send(struct drm_i915_private *i915, u8 *msg_in, reply_size, (u32)msg_out_len); } - memcpy(msg_out, hdcp_message->hdcp_cmd + sizeof(*header), msg_out_len); + memcpy(msg_out, hdcp_message->hdcp_cmd_out + sizeof(*header_out), msg_out_len); err: return ret; diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h index 5cc9fd2e88f6..cbf96551e534 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h +++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.h @@ -13,7 +13,8 @@ struct drm_i915_private; struct intel_hdcp_gsc_message { struct i915_vma *vma; - void *hdcp_cmd; + void *hdcp_cmd_in; + void *hdcp_cmd_out; }; bool intel_hdcp_gsc_cs_required(struct drm_i915_private *i915); diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index 7ac5e6c5e00d..94a7e1537f42 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -2880,21 +2880,12 @@ static u8 g4x_port_to_ddc_pin(struct drm_i915_private *dev_priv, return ddc_pin; } -static u8 intel_hdmi_ddc_pin(struct intel_encoder *encoder) +static u8 intel_hdmi_default_ddc_pin(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum port port = encoder->port; u8 ddc_pin; - ddc_pin = intel_bios_hdmi_ddc_pin(encoder->devdata); - if (ddc_pin) { - drm_dbg_kms(&dev_priv->drm, - "[ENCODER:%d:%s] Using DDC pin 0x%x (VBT)\n", - encoder->base.base.id, encoder->base.name, - ddc_pin); - return ddc_pin; - } - if (IS_ALDERLAKE_S(dev_priv)) ddc_pin = adls_port_to_ddc_pin(dev_priv, port); else if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1) @@ -2903,7 +2894,8 @@ static u8 intel_hdmi_ddc_pin(struct intel_encoder *encoder) ddc_pin = rkl_port_to_ddc_pin(dev_priv, port); else if (DISPLAY_VER(dev_priv) == 9 && HAS_PCH_TGP(dev_priv)) ddc_pin = gen9bc_tgp_port_to_ddc_pin(dev_priv, port); - else if (IS_JSL_EHL(dev_priv) && HAS_PCH_TGP(dev_priv)) + else if ((IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) && + HAS_PCH_TGP(dev_priv)) ddc_pin = mcc_port_to_ddc_pin(dev_priv, port); else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) ddc_pin = icl_port_to_ddc_pin(dev_priv, port); @@ -2916,10 +2908,62 @@ static u8 intel_hdmi_ddc_pin(struct intel_encoder *encoder) else ddc_pin = g4x_port_to_ddc_pin(dev_priv, port); - drm_dbg_kms(&dev_priv->drm, - "[ENCODER:%d:%s] Using DDC pin 0x%x (platform default)\n", + return ddc_pin; +} + +static struct intel_encoder * +get_encoder_by_ddc_pin(struct intel_encoder *encoder, u8 ddc_pin) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_encoder *other; + + for_each_intel_encoder(&i915->drm, other) { + if (other == encoder) + continue; + + if (!intel_encoder_is_dig_port(other)) + continue; + + if (enc_to_dig_port(other)->hdmi.ddc_bus == ddc_pin) + return other; + } + + return NULL; +} + +static u8 intel_hdmi_ddc_pin(struct intel_encoder *encoder) +{ + struct drm_i915_private *i915 = to_i915(encoder->base.dev); + struct intel_encoder *other; + const char *source; + u8 ddc_pin; + + ddc_pin = intel_bios_hdmi_ddc_pin(encoder->devdata); + source = "VBT"; + + if (!ddc_pin) { + ddc_pin = intel_hdmi_default_ddc_pin(encoder); + source = "platform default"; + } + + if (!intel_gmbus_is_valid_pin(i915, ddc_pin)) { + drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] Invalid DDC pin %d\n", + encoder->base.base.id, encoder->base.name, ddc_pin); + return 0; + } + + other = get_encoder_by_ddc_pin(encoder, ddc_pin); + if (other) { + drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] DDC pin %d already claimed by [ENCODER:%d:%s]\n", + encoder->base.base.id, encoder->base.name, ddc_pin, + other->base.base.id, other->base.name); + return 0; + } + + drm_dbg_kms(&i915->drm, + "[ENCODER:%d:%s] Using DDC pin 0x%x (%s)\n", encoder->base.base.id, encoder->base.name, - ddc_pin); + ddc_pin, source); return ddc_pin; } @@ -2990,6 +3034,9 @@ void intel_hdmi_init_connector(struct intel_digital_port *dig_port, return; intel_hdmi->ddc_bus = intel_hdmi_ddc_pin(intel_encoder); + if (!intel_hdmi->ddc_bus) + return; + ddc = intel_gmbus_get_adapter(dev_priv, intel_hdmi->ddc_bus); drm_connector_init_with_ddc(dev, connector, diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c index 1160fa20433b..69a83dcc1996 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug.c @@ -211,7 +211,7 @@ intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv) /* Enable polling and queue hotplug re-enabling. */ if (hpd_disabled) { - drm_kms_helper_poll_enable(&dev_priv->drm); + drm_kms_helper_poll_reschedule(&dev_priv->drm); mod_delayed_work(dev_priv->unordered_wq, &dev_priv->display.hotplug.reenable_work, msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); @@ -376,6 +376,8 @@ static void i915_hotplug_work_func(struct work_struct *work) u32 changed = 0, retry = 0; u32 hpd_event_bits; u32 hpd_retry_bits; + struct drm_connector *first_changed_connector = NULL; + int changed_connectors = 0; mutex_lock(&dev_priv->drm.mode_config.mutex); drm_dbg_kms(&dev_priv->drm, "running encoder hotplug functions\n"); @@ -428,6 +430,11 @@ static void i915_hotplug_work_func(struct work_struct *work) break; case INTEL_HOTPLUG_CHANGED: changed |= hpd_bit; + changed_connectors++; + if (!first_changed_connector) { + drm_connector_get(&connector->base); + first_changed_connector = &connector->base; + } break; case INTEL_HOTPLUG_RETRY: retry |= hpd_bit; @@ -438,9 +445,14 @@ static void i915_hotplug_work_func(struct work_struct *work) drm_connector_list_iter_end(&conn_iter); mutex_unlock(&dev_priv->drm.mode_config.mutex); - if (changed) + if (changed_connectors == 1) + drm_kms_helper_connector_hotplug_event(first_changed_connector); + else if (changed_connectors > 0) drm_kms_helper_hotplug_event(&dev_priv->drm); + if (first_changed_connector) + drm_connector_put(first_changed_connector); + /* Remove shared HPD pins that have changed */ retry &= ~changed; if (retry) { @@ -649,7 +661,7 @@ static void i915_hpd_poll_init_work(struct work_struct *work) drm_connector_list_iter_end(&conn_iter); if (enabled) - drm_kms_helper_poll_enable(&dev_priv->drm); + drm_kms_helper_poll_reschedule(&dev_priv->drm); mutex_unlock(&dev_priv->drm.mode_config.mutex); diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c index f95fa793fabb..95a7ea94f417 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c @@ -842,6 +842,8 @@ static void icp_hpd_irq_setup(struct drm_i915_private *dev_priv) if (INTEL_PCH_TYPE(dev_priv) <= PCH_TGP) intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + else + intel_uncore_write(&dev_priv->uncore, SHPD_FILTER_CNT, SHPD_FILTER_CNT_250); ibx_display_interrupt_update(dev_priv, hotplug_irqs, enabled_irqs); @@ -1049,7 +1051,7 @@ static void mtp_hpd_irq_setup(struct drm_i915_private *i915) enabled_irqs = intel_hpd_enabled_irqs(i915, i915->display.hotplug.pch_hpd); hotplug_irqs = intel_hpd_hotplug_irqs(i915, i915->display.hotplug.pch_hpd); - intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_500_ADJ); + intel_de_write(i915, SHPD_FILTER_CNT, SHPD_FILTER_CNT_250); mtp_hpd_invert(i915); ibx_display_interrupt_update(i915, hotplug_irqs, enabled_irqs); diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index d6fe2bbabe55..09c1aa1427ad 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -1348,11 +1348,12 @@ out_unlock: static int get_registers(struct intel_overlay *overlay, bool use_phys) { struct drm_i915_private *i915 = overlay->i915; - struct drm_i915_gem_object *obj; + struct drm_i915_gem_object *obj = ERR_PTR(-ENODEV); struct i915_vma *vma; int err; - obj = i915_gem_object_create_stolen(i915, PAGE_SIZE); + if (!IS_METEORLAKE(i915)) /* Wa_22018444074 */ + obj = i915_gem_object_create_stolen(i915, PAGE_SIZE); if (IS_ERR(obj)) obj = i915_gem_object_create_internal(i915, PAGE_SIZE); if (IS_ERR(obj)) diff --git a/drivers/gpu/drm/i915/display/intel_pch_refclk.c b/drivers/gpu/drm/i915/display/intel_pch_refclk.c index f4c09cc37a5e..9583e86b602a 100644 --- a/drivers/gpu/drm/i915/display/intel_pch_refclk.c +++ b/drivers/gpu/drm/i915/display/intel_pch_refclk.c @@ -423,7 +423,7 @@ static bool wrpll_uses_pch_ssc(struct drm_i915_private *dev_priv, if ((ctl & WRPLL_REF_MASK) == WRPLL_REF_PCH_SSC) return true; - if ((IS_BROADWELL(dev_priv) || IS_HSW_ULT(dev_priv)) && + if ((IS_BROADWELL(dev_priv) || IS_HASWELL_ULT(dev_priv)) && (ctl & WRPLL_REF_MASK) == WRPLL_REF_MUXED_SSC_BDW && (fuse_strap & HSW_CPU_SSC_ENABLE) == 0) return true; diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 56c17283ba2d..97d5eef10130 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -234,23 +234,91 @@ static u32 psr_irq_mask_get(struct intel_dp *intel_dp) EDP_PSR_MASK(intel_dp->psr.transcoder); } -static void psr_irq_control(struct intel_dp *intel_dp) +static i915_reg_t psr_ctl_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) { - struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); - i915_reg_t imr_reg; - u32 mask; + if (DISPLAY_VER(dev_priv) >= 8) + return EDP_PSR_CTL(cpu_transcoder); + else + return HSW_SRD_CTL; +} + +static i915_reg_t psr_debug_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) +{ + if (DISPLAY_VER(dev_priv) >= 8) + return EDP_PSR_DEBUG(cpu_transcoder); + else + return HSW_SRD_DEBUG; +} + +static i915_reg_t psr_perf_cnt_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) +{ + if (DISPLAY_VER(dev_priv) >= 8) + return EDP_PSR_PERF_CNT(cpu_transcoder); + else + return HSW_SRD_PERF_CNT; +} + +static i915_reg_t psr_status_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) +{ + if (DISPLAY_VER(dev_priv) >= 8) + return EDP_PSR_STATUS(cpu_transcoder); + else + return HSW_SRD_STATUS; +} + +static i915_reg_t psr_imr_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) +{ + if (DISPLAY_VER(dev_priv) >= 12) + return TRANS_PSR_IMR(cpu_transcoder); + else + return EDP_PSR_IMR; +} +static i915_reg_t psr_iir_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) +{ if (DISPLAY_VER(dev_priv) >= 12) - imr_reg = TRANS_PSR_IMR(intel_dp->psr.transcoder); + return TRANS_PSR_IIR(cpu_transcoder); + else + return EDP_PSR_IIR; +} + +static i915_reg_t psr_aux_ctl_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder) +{ + if (DISPLAY_VER(dev_priv) >= 8) + return EDP_PSR_AUX_CTL(cpu_transcoder); + else + return HSW_SRD_AUX_CTL; +} + +static i915_reg_t psr_aux_data_reg(struct drm_i915_private *dev_priv, + enum transcoder cpu_transcoder, int i) +{ + if (DISPLAY_VER(dev_priv) >= 8) + return EDP_PSR_AUX_DATA(cpu_transcoder, i); else - imr_reg = EDP_PSR_IMR; + return HSW_SRD_AUX_DATA(i); +} + +static void psr_irq_control(struct intel_dp *intel_dp) +{ + struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; + u32 mask; mask = psr_irq_psr_error_bit_get(intel_dp); if (intel_dp->psr.debug & I915_PSR_DEBUG_IRQ) mask |= psr_irq_post_exit_bit_get(intel_dp) | psr_irq_pre_entry_bit_get(intel_dp); - intel_de_rmw(dev_priv, imr_reg, psr_irq_mask_get(intel_dp), ~mask); + intel_de_rmw(dev_priv, psr_imr_reg(dev_priv, cpu_transcoder), + psr_irq_mask_get(intel_dp), ~mask); } static void psr_event_print(struct drm_i915_private *i915, @@ -296,12 +364,6 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir) struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); enum transcoder cpu_transcoder = intel_dp->psr.transcoder; ktime_t time_ns = ktime_get(); - i915_reg_t imr_reg; - - if (DISPLAY_VER(dev_priv) >= 12) - imr_reg = TRANS_PSR_IMR(cpu_transcoder); - else - imr_reg = EDP_PSR_IMR; if (psr_iir & psr_irq_pre_entry_bit_get(intel_dp)) { intel_dp->psr.last_entry_attempt = time_ns; @@ -339,7 +401,8 @@ void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir) * again so we don't care about unmask the interruption * or unset irq_aux_error. */ - intel_de_rmw(dev_priv, imr_reg, 0, psr_irq_psr_error_bit_get(intel_dp)); + intel_de_rmw(dev_priv, psr_imr_reg(dev_priv, cpu_transcoder), + 0, psr_irq_psr_error_bit_get(intel_dp)); queue_work(dev_priv->unordered_wq, &intel_dp->psr.work); } @@ -467,6 +530,43 @@ void intel_psr_init_dpcd(struct intel_dp *intel_dp) } } +static void hsw_psr_setup_aux(struct intel_dp *intel_dp) +{ + struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; + u32 aux_clock_divider, aux_ctl; + /* write DP_SET_POWER=D0 */ + static const u8 aux_msg[] = { + [0] = (DP_AUX_NATIVE_WRITE << 4) | ((DP_SET_POWER >> 16) & 0xf), + [1] = (DP_SET_POWER >> 8) & 0xff, + [2] = DP_SET_POWER & 0xff, + [3] = 1 - 1, + [4] = DP_SET_POWER_D0, + }; + int i; + + BUILD_BUG_ON(sizeof(aux_msg) > 20); + for (i = 0; i < sizeof(aux_msg); i += 4) + intel_de_write(dev_priv, + psr_aux_data_reg(dev_priv, cpu_transcoder, i >> 2), + intel_dp_aux_pack(&aux_msg[i], sizeof(aux_msg) - i)); + + aux_clock_divider = intel_dp->get_aux_clock_divider(intel_dp, 0); + + /* Start with bits set for DDI_AUX_CTL register */ + aux_ctl = intel_dp->get_aux_send_ctl(intel_dp, sizeof(aux_msg), + aux_clock_divider); + + /* Select only valid bits for SRD_AUX_CTL */ + aux_ctl &= EDP_PSR_AUX_CTL_TIME_OUT_MASK | + EDP_PSR_AUX_CTL_MESSAGE_SIZE_MASK | + EDP_PSR_AUX_CTL_PRECHARGE_2US_MASK | + EDP_PSR_AUX_CTL_BIT_CLOCK_2X_MASK; + + intel_de_write(dev_priv, psr_aux_ctl_reg(dev_priv, cpu_transcoder), + aux_ctl); +} + static void intel_psr_enable_sink(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); @@ -528,6 +628,15 @@ static u32 intel_psr1_get_tp_time(struct intel_dp *intel_dp) else val |= EDP_PSR_TP2_TP3_TIME_2500us; + /* + * WA 0479: hsw,bdw + * "Do not skip both TP1 and TP2/TP3" + */ + if (DISPLAY_VER(dev_priv) < 9 && + connector->panel.vbt.psr.tp1_wakeup_time_us == 0 && + connector->panel.vbt.psr.tp2_tp3_wakeup_time_us == 0) + val |= EDP_PSR_TP2_TP3_TIME_100us; + check_tp3_sel: if (intel_dp_source_supports_tps3(dev_priv) && drm_dp_tps3_supported(intel_dp->dpcd)) @@ -577,7 +686,7 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp) if (DISPLAY_VER(dev_priv) >= 8) val |= EDP_PSR_CRC_ENABLE; - intel_de_rmw(dev_priv, EDP_PSR_CTL(cpu_transcoder), + intel_de_rmw(dev_priv, psr_ctl_reg(dev_priv, cpu_transcoder), ~EDP_PSR_RESTORE_PSR_ACTIVE_CTX_MASK, val); } @@ -639,7 +748,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) } /* Wa_22012278275:adl-p */ - if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_E0)) { + if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_E0)) { static const u8 map[] = { 2, /* 5 lines */ 1, /* 6 lines */ @@ -685,7 +794,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is * recommending keep this bit unset while PSR2 is enabled. */ - intel_de_write(dev_priv, EDP_PSR_CTL(cpu_transcoder), 0); + intel_de_write(dev_priv, psr_ctl_reg(dev_priv, cpu_transcoder), 0); intel_de_write(dev_priv, EDP_PSR2_CTL(cpu_transcoder), val); } @@ -697,8 +806,10 @@ transcoder_has_psr2(struct drm_i915_private *dev_priv, enum transcoder cpu_trans return cpu_transcoder == TRANSCODER_A || cpu_transcoder == TRANSCODER_B; else if (DISPLAY_VER(dev_priv) >= 12) return cpu_transcoder == TRANSCODER_A; - else + else if (DISPLAY_VER(dev_priv) >= 9) return cpu_transcoder == TRANSCODER_EDP; + else + return false; } static u32 intel_get_frame_time_us(const struct intel_crtc_state *cstate) @@ -807,7 +918,7 @@ tgl_dc3co_exitline_compute_config(struct intel_dp *intel_dp, return; /* Wa_16011303918:adl-p */ - if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) + if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) return; /* @@ -963,7 +1074,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp, return false; /* JSL and EHL only supports eDP 1.3 */ - if (IS_JSL_EHL(dev_priv)) { + if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) { drm_dbg_kms(&dev_priv->drm, "PSR2 not supported by phy\n"); return false; } @@ -975,7 +1086,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp, return false; } - if (IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { + if (IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { drm_dbg_kms(&dev_priv->drm, "PSR2 not completely functional in this stepping\n"); return false; } @@ -1033,7 +1144,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp, /* Wa_16011303918:adl-p */ if (crtc_state->vrr.enable && - IS_ADLP_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { + IS_ALDERLAKE_P(dev_priv) && IS_DISPLAY_STEP(dev_priv, STEP_A0, STEP_B0)) { drm_dbg_kms(&dev_priv->drm, "PSR2 not enabled, not compatible with HW stepping + VRR\n"); return false; @@ -1201,13 +1312,15 @@ static void intel_psr_activate(struct intel_dp *intel_dp) struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); enum transcoder cpu_transcoder = intel_dp->psr.transcoder; - if (transcoder_has_psr2(dev_priv, cpu_transcoder)) - drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)) & EDP_PSR2_ENABLE); + drm_WARN_ON(&dev_priv->drm, + transcoder_has_psr2(dev_priv, cpu_transcoder) && + intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)) & EDP_PSR2_ENABLE); drm_WARN_ON(&dev_priv->drm, - intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)) & EDP_PSR_ENABLE); + intel_de_read(dev_priv, psr_ctl_reg(dev_priv, cpu_transcoder)) & EDP_PSR_ENABLE); + drm_WARN_ON(&dev_priv->drm, intel_dp->psr.active); + lockdep_assert_held(&intel_dp->psr.lock); /* psr1 and psr2 are mutually exclusive.*/ @@ -1272,6 +1385,13 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp, u32 mask; /* + * Only HSW and BDW have PSR AUX registers that need to be setup. + * SKL+ use hardcoded values PSR AUX transactions + */ + if (DISPLAY_VER(dev_priv) < 9) + hsw_psr_setup_aux(intel_dp); + + /* * Per Spec: Avoid continuous PSR exit by masking MEMUP and HPD also * mask LPSP to avoid dependency on other drivers that might block * runtime_pm besides preventing other hw tracking issues now we @@ -1282,11 +1402,18 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp, EDP_PSR_DEBUG_MASK_LPSP | EDP_PSR_DEBUG_MASK_MAX_SLEEP; - if (DISPLAY_VER(dev_priv) < 11) + /* + * No separate pipe reg write mask on hsw/bdw, so have to unmask all + * registers in order to keep the CURSURFLIVE tricks working :( + */ + if (IS_DISPLAY_VER(dev_priv, 9, 10)) mask |= EDP_PSR_DEBUG_MASK_DISP_REG_WRITE; - intel_de_write(dev_priv, EDP_PSR_DEBUG(cpu_transcoder), - mask); + /* allow PSR with sprite enabled */ + if (IS_HASWELL(dev_priv)) + mask |= EDP_PSR_DEBUG_MASK_SPRITE_ENABLE; + + intel_de_write(dev_priv, psr_debug_reg(dev_priv, cpu_transcoder), mask); psr_irq_control(intel_dp); @@ -1352,10 +1479,7 @@ static bool psr_interrupt_error_check(struct intel_dp *intel_dp) * first time that PSR HW tries to activate so lets keep PSR disabled * to avoid any rendering problems. */ - if (DISPLAY_VER(dev_priv) >= 12) - val = intel_de_read(dev_priv, TRANS_PSR_IIR(cpu_transcoder)); - else - val = intel_de_read(dev_priv, EDP_PSR_IIR); + val = intel_de_read(dev_priv, psr_iir_reg(dev_priv, cpu_transcoder)); val &= psr_irq_psr_error_bit_get(intel_dp); if (val) { intel_dp->psr.sink_not_reliable = true; @@ -1418,7 +1542,7 @@ static void intel_psr_exit(struct intel_dp *intel_dp) drm_WARN_ON(&dev_priv->drm, val & EDP_PSR2_ENABLE); } - val = intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)); + val = intel_de_read(dev_priv, psr_ctl_reg(dev_priv, cpu_transcoder)); drm_WARN_ON(&dev_priv->drm, val & EDP_PSR_ENABLE); return; @@ -1432,7 +1556,7 @@ static void intel_psr_exit(struct intel_dp *intel_dp) drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR2_ENABLE)); } else { - val = intel_de_rmw(dev_priv, EDP_PSR_CTL(cpu_transcoder), + val = intel_de_rmw(dev_priv, psr_ctl_reg(dev_priv, cpu_transcoder), EDP_PSR_ENABLE, 0); drm_WARN_ON(&dev_priv->drm, !(val & EDP_PSR_ENABLE)); @@ -1451,7 +1575,7 @@ static void intel_psr_wait_exit_locked(struct intel_dp *intel_dp) psr_status = EDP_PSR2_STATUS(cpu_transcoder); psr_status_mask = EDP_PSR2_STATUS_STATE_MASK; } else { - psr_status = EDP_PSR_STATUS(cpu_transcoder); + psr_status = psr_status_reg(dev_priv, cpu_transcoder); psr_status_mask = EDP_PSR_STATUS_STATE_MASK; } @@ -2151,7 +2275,7 @@ static int _psr1_ready_for_pipe_update_locked(struct intel_dp *intel_dp) * defensive enough to cover everything. */ return intel_de_wait_for_clear(dev_priv, - EDP_PSR_STATUS(cpu_transcoder), + psr_status_reg(dev_priv, cpu_transcoder), EDP_PSR_STATUS_STATE_MASK, 50); } @@ -2205,7 +2329,7 @@ static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp) reg = EDP_PSR2_STATUS(cpu_transcoder); mask = EDP_PSR2_STATUS_STATE_MASK; } else { - reg = EDP_PSR_STATUS(cpu_transcoder); + reg = psr_status_reg(dev_priv, cpu_transcoder); mask = EDP_PSR_STATUS_STATE_MASK; } @@ -2825,7 +2949,7 @@ psr_source_status(struct intel_dp *intel_dp, struct seq_file *m) "SRDOFFACK", "SRDENT_ON", }; - val = intel_de_read(dev_priv, EDP_PSR_STATUS(cpu_transcoder)); + val = intel_de_read(dev_priv, psr_status_reg(dev_priv, cpu_transcoder)); status_val = REG_FIELD_GET(EDP_PSR_STATUS_STATE_MASK, val); if (status_val < ARRAY_SIZE(live_status)) status = live_status[status_val]; @@ -2872,7 +2996,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp) val = intel_de_read(dev_priv, EDP_PSR2_CTL(cpu_transcoder)); enabled = val & EDP_PSR2_ENABLE; } else { - val = intel_de_read(dev_priv, EDP_PSR_CTL(cpu_transcoder)); + val = intel_de_read(dev_priv, psr_ctl_reg(dev_priv, cpu_transcoder)); enabled = val & EDP_PSR_ENABLE; } seq_printf(m, "Source PSR ctl: %s [0x%08x]\n", @@ -2884,7 +3008,7 @@ static int intel_psr_status(struct seq_file *m, struct intel_dp *intel_dp) /* * SKL+ Perf counter is reset to 0 everytime DC state is entered */ - val = intel_de_read(dev_priv, EDP_PSR_PERF_CNT(cpu_transcoder)); + val = intel_de_read(dev_priv, psr_perf_cnt_reg(dev_priv, cpu_transcoder)); seq_printf(m, "Performance counter: %u\n", REG_FIELD_GET(EDP_PSR_PERF_CNT_MASK, val)); diff --git a/drivers/gpu/drm/i915/display/intel_psr_regs.h b/drivers/gpu/drm/i915/display/intel_psr_regs.h index 8750cb0d8d9d..d39951383c92 100644 --- a/drivers/gpu/drm/i915/display/intel_psr_regs.h +++ b/drivers/gpu/drm/i915/display/intel_psr_regs.h @@ -7,6 +7,7 @@ #define __INTEL_PSR_REGS_H__ #include "intel_display_reg_defs.h" +#include "intel_dp_aux_regs.h" #define TRANS_EXITLINE(trans) _MMIO_TRANS2((trans), _TRANS_EXITLINE_A) #define EXITLINE_ENABLE REG_BIT(31) @@ -19,6 +20,7 @@ * HSW PSR registers are relative to DDIA(_DDI_BUF_CTL_A + 0x800) with just one * instance of it */ +#define HSW_SRD_CTL _MMIO(0x64800) #define _SRD_CTL_A 0x60800 #define _SRD_CTL_EDP 0x6f800 #define EDP_PSR_CTL(tran) _MMIO_TRANS2(tran, _SRD_CTL_A) @@ -79,10 +81,22 @@ #define EDP_PSR_PRE_ENTRY(trans) (TGL_PSR_PRE_ENTRY << \ _EDP_PSR_TRANS_SHIFT(trans)) +#define HSW_SRD_AUX_CTL _MMIO(0x64810) +#define _SRD_AUX_CTL_A 0x60810 +#define _SRD_AUX_CTL_EDP 0x6f810 +#define EDP_PSR_AUX_CTL(tran) _MMIO_TRANS2(tran, _SRD_AUX_CTL_A) +#define EDP_PSR_AUX_CTL_TIME_OUT_MASK DP_AUX_CH_CTL_TIME_OUT_MASK +#define EDP_PSR_AUX_CTL_MESSAGE_SIZE_MASK DP_AUX_CH_CTL_MESSAGE_SIZE_MASK +#define EDP_PSR_AUX_CTL_PRECHARGE_2US_MASK DP_AUX_CH_CTL_PRECHARGE_2US_MASK +#define EDP_PSR_AUX_CTL_ERROR_INTERRUPT REG_BIT(11) +#define EDP_PSR_AUX_CTL_BIT_CLOCK_2X_MASK DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK + +#define HSW_SRD_AUX_DATA(i) _MMIO(0x64814 + (i) * 4) /* 5 registers */ #define _SRD_AUX_DATA_A 0x60814 #define _SRD_AUX_DATA_EDP 0x6f814 #define EDP_PSR_AUX_DATA(tran, i) _MMIO_TRANS2(tran, _SRD_AUX_DATA_A + (i) * 4) /* 5 registers */ +#define HSW_SRD_STATUS _MMIO(0x64840) #define _SRD_STATUS_A 0x60840 #define _SRD_STATUS_EDP 0x6f840 #define EDP_PSR_STATUS(tran) _MMIO_TRANS2(tran, _SRD_STATUS_A) @@ -107,12 +121,14 @@ #define EDP_PSR_STATUS_SENDING_TP1 REG_BIT(4) #define EDP_PSR_STATUS_IDLE_MASK REG_GENMASK(3, 0) +#define HSW_SRD_PERF_CNT _MMIO(0x64844) #define _SRD_PERF_CNT_A 0x60844 #define _SRD_PERF_CNT_EDP 0x6f844 #define EDP_PSR_PERF_CNT(tran) _MMIO_TRANS2(tran, _SRD_PERF_CNT_A) #define EDP_PSR_PERF_CNT_MASK REG_GENMASK(23, 0) /* PSR_MASK on SKL+ */ +#define HSW_SRD_DEBUG _MMIO(0x64860) #define _SRD_DEBUG_A 0x60860 #define _SRD_DEBUG_EDP 0x6f860 #define EDP_PSR_DEBUG(tran) _MMIO_TRANS2(tran, _SRD_DEBUG_A) diff --git a/drivers/gpu/drm/i915/display/intel_qp_tables.c b/drivers/gpu/drm/i915/display/intel_qp_tables.c index 6e86c0971d24..543cdc46aa1d 100644 --- a/drivers/gpu/drm/i915/display/intel_qp_tables.c +++ b/drivers/gpu/drm/i915/display/intel_qp_tables.c @@ -17,13 +17,17 @@ /* from BPP 6 to 36 in steps of 0.5 */ #define RC_RANGE_QP444_12BPC_MAX_NUM_BPP 61 -/* from BPP 6 to 24 in steps of 0.5 */ +/* For YCbCr420 the bits_per_pixel sent in PPS params + * is double the target bpp. The below values represent + * the target bpp. + */ +/* from BPP 4 to 12 in steps of 0.5 */ #define RC_RANGE_QP420_8BPC_MAX_NUM_BPP 17 -/* from BPP 6 to 30 in steps of 0.5 */ +/* from BPP 4 to 15 in steps of 0.5 */ #define RC_RANGE_QP420_10BPC_MAX_NUM_BPP 23 -/* from BPP 6 to 36 in steps of 0.5 */ +/* from BPP 4 to 18 in steps of 0.5 */ #define RC_RANGE_QP420_12BPC_MAX_NUM_BPP 29 /* diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index 21f92123c844..7d25a64698e2 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -2097,7 +2097,7 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector) const struct edid *edid = drm_edid_raw(drm_edid); /* DDC bus is shared, match EDID to connector type */ - if (edid->input & DRM_EDID_INPUT_DIGITAL) + if (edid && edid->input & DRM_EDID_INPUT_DIGITAL) status = connector_status_connected; else status = connector_status_disconnected; @@ -2752,7 +2752,7 @@ static struct intel_sdvo_connector *intel_sdvo_connector_alloc(void) __drm_atomic_helper_connector_reset(&sdvo_connector->base.base, &conn_state->base.base); - INIT_LIST_HEAD(&sdvo_connector->base.panel.fixed_modes); + intel_panel_init_alloc(&sdvo_connector->base); return sdvo_connector; } @@ -3313,13 +3313,19 @@ intel_sdvo_init_ddc_proxy(struct intel_sdvo *sdvo, return i2c_add_adapter(&sdvo->ddc) == 0; } -static void assert_sdvo_port_valid(const struct drm_i915_private *dev_priv, - enum port port) +static bool is_sdvo_port_valid(struct drm_i915_private *dev_priv, enum port port) { if (HAS_PCH_SPLIT(dev_priv)) - drm_WARN_ON(&dev_priv->drm, port != PORT_B); + return port == PORT_B; else - drm_WARN_ON(&dev_priv->drm, port != PORT_B && port != PORT_C); + return port == PORT_B || port == PORT_C; +} + +static bool assert_sdvo_port_valid(struct drm_i915_private *dev_priv, + enum port port) +{ + return !drm_WARN(&dev_priv->drm, !is_sdvo_port_valid(dev_priv, port), + "Platform does not support SDVO %c\n", port_name(port)); } bool intel_sdvo_init(struct drm_i915_private *dev_priv, @@ -3329,7 +3335,11 @@ bool intel_sdvo_init(struct drm_i915_private *dev_priv, struct intel_sdvo *intel_sdvo; int i; - assert_sdvo_port_valid(dev_priv, port); + if (!assert_port_valid(dev_priv, port)) + return false; + + if (!assert_sdvo_port_valid(dev_priv, port)) + return false; intel_sdvo = kzalloc(sizeof(*intel_sdvo), GFP_KERNEL); if (!intel_sdvo) diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.c b/drivers/gpu/drm/i915/display/intel_vdsc.c index bd9116d2cd76..9d76c2756784 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.c +++ b/drivers/gpu/drm/i915/display/intel_vdsc.c @@ -53,22 +53,32 @@ static bool is_pipe_dsc(struct intel_crtc *crtc, enum transcoder cpu_transcoder) } static void +intel_vdsc_set_min_max_qp(struct drm_dsc_config *vdsc_cfg, int buf, + int bpp) +{ + int bpc = vdsc_cfg->bits_per_component; + + /* Read range_minqp and range_max_qp from qp tables */ + vdsc_cfg->rc_range_params[buf].range_min_qp = + intel_lookup_range_min_qp(bpc, buf, bpp, vdsc_cfg->native_420); + vdsc_cfg->rc_range_params[buf].range_max_qp = + intel_lookup_range_max_qp(bpc, buf, bpp, vdsc_cfg->native_420); +} + +/* + * We are using the method provided in DSC 1.2a C-Model in codec_main.c + * Above method use a common formula to derive values for any combination of DSC + * variables. The formula approach may yield slight differences in the derived PPS + * parameters from the original parameter sets. These differences are not consequential + * to the coding performance because all parameter sets have been shown to produce + * visually lossless quality (provides the same PPS values as + * DSCParameterValuesVESA V1-2 spreadsheet). + */ +static void calculate_rc_params(struct drm_dsc_config *vdsc_cfg) { int bpc = vdsc_cfg->bits_per_component; int bpp = vdsc_cfg->bits_per_pixel >> 4; - static const s8 ofs_und6[] = { - 0, -2, -2, -4, -6, -6, -8, -8, -8, -10, -10, -12, -12, -12, -12 - }; - static const s8 ofs_und8[] = { - 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -10, -12, -12, -12 - }; - static const s8 ofs_und12[] = { - 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -10, -12, -12, -12 - }; - static const s8 ofs_und15[] = { - 10, 8, 6, 4, 2, 0, -2, -4, -6, -8, -10, -10, -12, -12, -12 - }; int qp_bpc_modifier = (bpc - 8) * 2; u32 res, buf_i, bpp_i; @@ -78,6 +88,28 @@ calculate_rc_params(struct drm_dsc_config *vdsc_cfg) else vdsc_cfg->first_line_bpg_offset = 2 * (vdsc_cfg->slice_height - 1); + /* + * According to DSC 1.2 spec in Section 4.1 if native_420 is set: + * -second_line_bpg_offset is 12 in general and equal to 2*(slice_height-1) if slice + * height < 8. + * -second_line_offset_adj is 512 as shown by emperical values to yield best chroma + * preservation in second line. + * -nsl_bpg_offset is calculated as second_line_offset/slice_height -1 then rounded + * up to 16 fractional bits, we left shift second line offset by 11 to preserve 11 + * fractional bits. + */ + if (vdsc_cfg->native_420) { + if (vdsc_cfg->slice_height >= 8) + vdsc_cfg->second_line_bpg_offset = 12; + else + vdsc_cfg->second_line_bpg_offset = + 2 * (vdsc_cfg->slice_height - 1); + + vdsc_cfg->second_line_offset_adj = 512; + vdsc_cfg->nsl_bpg_offset = DIV_ROUND_UP(vdsc_cfg->second_line_bpg_offset << 11, + vdsc_cfg->slice_height - 1); + } + /* Our hw supports only 444 modes as of today */ if (bpp >= 12) vdsc_cfg->initial_offset = 2048; @@ -97,33 +129,88 @@ calculate_rc_params(struct drm_dsc_config *vdsc_cfg) vdsc_cfg->rc_quant_incr_limit0 = 11 + qp_bpc_modifier; vdsc_cfg->rc_quant_incr_limit1 = 11 + qp_bpc_modifier; - bpp_i = (2 * (bpp - 6)); - for (buf_i = 0; buf_i < DSC_NUM_BUF_RANGES; buf_i++) { - u8 range_bpg_offset; - - /* Read range_minqp and range_max_qp from qp tables */ - vdsc_cfg->rc_range_params[buf_i].range_min_qp = - intel_lookup_range_min_qp(bpc, buf_i, bpp_i, vdsc_cfg->native_420); - vdsc_cfg->rc_range_params[buf_i].range_max_qp = - intel_lookup_range_max_qp(bpc, buf_i, bpp_i, vdsc_cfg->native_420); - - /* Calculate range_bpg_offset */ - if (bpp <= 6) { - range_bpg_offset = ofs_und6[buf_i]; - } else if (bpp <= 8) { - res = DIV_ROUND_UP(((bpp - 6) * (ofs_und8[buf_i] - ofs_und6[buf_i])), 2); - range_bpg_offset = ofs_und6[buf_i] + res; - } else if (bpp <= 12) { - range_bpg_offset = ofs_und8[buf_i]; - } else if (bpp <= 15) { - res = DIV_ROUND_UP(((bpp - 12) * (ofs_und15[buf_i] - ofs_und12[buf_i])), 3); - range_bpg_offset = ofs_und12[buf_i] + res; - } else { - range_bpg_offset = ofs_und15[buf_i]; + if (vdsc_cfg->native_420) { + static const s8 ofs_und4[] = { + 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -12, -12, -12, -12 + }; + static const s8 ofs_und5[] = { + 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -10, -12, -12, -12 + }; + static const s8 ofs_und6[] = { + 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -10, -12, -12, -12 + }; + static const s8 ofs_und8[] = { + 10, 8, 6, 4, 2, 0, -2, -4, -6, -8, -10, -10, -12, -12, -12 + }; + + bpp_i = bpp - 8; + for (buf_i = 0; buf_i < DSC_NUM_BUF_RANGES; buf_i++) { + u8 range_bpg_offset; + + intel_vdsc_set_min_max_qp(vdsc_cfg, buf_i, bpp_i); + + /* Calculate range_bpg_offset */ + if (bpp <= 8) { + range_bpg_offset = ofs_und4[buf_i]; + } else if (bpp <= 10) { + res = DIV_ROUND_UP(((bpp - 8) * + (ofs_und5[buf_i] - ofs_und4[buf_i])), 2); + range_bpg_offset = ofs_und4[buf_i] + res; + } else if (bpp <= 12) { + res = DIV_ROUND_UP(((bpp - 10) * + (ofs_und6[buf_i] - ofs_und5[buf_i])), 2); + range_bpg_offset = ofs_und5[buf_i] + res; + } else if (bpp <= 16) { + res = DIV_ROUND_UP(((bpp - 12) * + (ofs_und8[buf_i] - ofs_und6[buf_i])), 4); + range_bpg_offset = ofs_und6[buf_i] + res; + } else { + range_bpg_offset = ofs_und8[buf_i]; + } + + vdsc_cfg->rc_range_params[buf_i].range_bpg_offset = + range_bpg_offset & DSC_RANGE_BPG_OFFSET_MASK; + } + } else { + static const s8 ofs_und6[] = { + 0, -2, -2, -4, -6, -6, -8, -8, -8, -10, -10, -12, -12, -12, -12 + }; + static const s8 ofs_und8[] = { + 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -10, -12, -12, -12 + }; + static const s8 ofs_und12[] = { + 2, 0, 0, -2, -4, -6, -8, -8, -8, -10, -10, -10, -12, -12, -12 + }; + static const s8 ofs_und15[] = { + 10, 8, 6, 4, 2, 0, -2, -4, -6, -8, -10, -10, -12, -12, -12 + }; + + bpp_i = (2 * (bpp - 6)); + for (buf_i = 0; buf_i < DSC_NUM_BUF_RANGES; buf_i++) { + u8 range_bpg_offset; + + intel_vdsc_set_min_max_qp(vdsc_cfg, buf_i, bpp_i); + + /* Calculate range_bpg_offset */ + if (bpp <= 6) { + range_bpg_offset = ofs_und6[buf_i]; + } else if (bpp <= 8) { + res = DIV_ROUND_UP(((bpp - 6) * + (ofs_und8[buf_i] - ofs_und6[buf_i])), 2); + range_bpg_offset = ofs_und6[buf_i] + res; + } else if (bpp <= 12) { + range_bpg_offset = ofs_und8[buf_i]; + } else if (bpp <= 15) { + res = DIV_ROUND_UP(((bpp - 12) * + (ofs_und15[buf_i] - ofs_und12[buf_i])), 3); + range_bpg_offset = ofs_und12[buf_i] + res; + } else { + range_bpg_offset = ofs_und15[buf_i]; + } + + vdsc_cfg->rc_range_params[buf_i].range_bpg_offset = + range_bpg_offset & DSC_RANGE_BPG_OFFSET_MASK; } - - vdsc_cfg->rc_range_params[buf_i].range_bpg_offset = - range_bpg_offset & DSC_RANGE_BPG_OFFSET_MASK; } } @@ -190,30 +277,12 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config) vdsc_cfg->bits_per_pixel = compressed_bpp << 4; /* - * According to DSC 1.2 specs in Section 4.1 if native_420 is set: - * -We need to double the current bpp. - * -second_line_bpg_offset is 12 in general and equal to 2*(slice_height-1) if slice - * height < 8. - * -second_line_offset_adj is 512 as shown by emperical values to yeild best chroma - * preservation in second line. - * -nsl_bpg_offset is calculated as second_line_offset/slice_height -1 then rounded - * up to 16 fractional bits, we left shift second line offset by 11 to preserve 11 - * fractional bits. + * According to DSC 1.2 specs in Section 4.1 if native_420 is set + * we need to double the current bpp. */ - if (vdsc_cfg->native_420) { + if (vdsc_cfg->native_420) vdsc_cfg->bits_per_pixel <<= 1; - if (vdsc_cfg->slice_height >= 8) - vdsc_cfg->second_line_bpg_offset = 12; - else - vdsc_cfg->second_line_bpg_offset = - 2 * (vdsc_cfg->slice_height - 1); - - vdsc_cfg->second_line_offset_adj = 512; - vdsc_cfg->nsl_bpg_offset = DIV_ROUND_UP(vdsc_cfg->second_line_bpg_offset << 11, - vdsc_cfg->slice_height - 1); - } - vdsc_cfg->bits_per_component = pipe_config->pipe_bpp / 3; drm_dsc_set_rc_buf_thresh(vdsc_cfg); @@ -237,18 +306,6 @@ int intel_dsc_compute_params(struct intel_crtc_state *pipe_config) if (ret) return ret; - - /* - * FIXME: verify that the hardware actually needs these - * modifications rather than them being simple typos. - */ - if (compressed_bpp == 6 && - vdsc_cfg->bits_per_component == 8) - vdsc_cfg->rc_quant_incr_limit1 = 23; - - if (compressed_bpp == 8 && - vdsc_cfg->bits_per_component == 14) - vdsc_cfg->rc_range_params[0].range_bpg_offset = 0; } /* @@ -293,6 +350,16 @@ intel_dsc_power_domain(struct intel_crtc *crtc, enum transcoder cpu_transcoder) return POWER_DOMAIN_TRANSCODER_VDSC_PW2; } +int intel_dsc_get_num_vdsc_instances(const struct intel_crtc_state *crtc_state) +{ + int num_vdsc_instances = (crtc_state->dsc.dsc_split) ? 2 : 1; + + if (crtc_state->bigjoiner_pipes) + num_vdsc_instances *= 2; + + return num_vdsc_instances; +} + static void intel_dsc_pps_configure(const struct intel_crtc_state *crtc_state) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); @@ -303,11 +370,8 @@ static void intel_dsc_pps_configure(const struct intel_crtc_state *crtc_state) u32 pps_val = 0; u32 rc_buf_thresh_dword[4]; u32 rc_range_params_dword[8]; - u8 num_vdsc_instances = (crtc_state->dsc.dsc_split) ? 2 : 1; int i = 0; - - if (crtc_state->bigjoiner_pipes) - num_vdsc_instances *= 2; + int num_vdsc_instances = intel_dsc_get_num_vdsc_instances(crtc_state); /* Populate PICTURE_PARAMETER_SET_0 registers */ pps_val = DSC_VER_MAJ | vdsc_cfg->dsc_version_minor << diff --git a/drivers/gpu/drm/i915/display/intel_vdsc.h b/drivers/gpu/drm/i915/display/intel_vdsc.h index 8763f00fa7e2..2cc41ff08909 100644 --- a/drivers/gpu/drm/i915/display/intel_vdsc.h +++ b/drivers/gpu/drm/i915/display/intel_vdsc.h @@ -22,6 +22,7 @@ void intel_dsc_get_config(struct intel_crtc_state *crtc_state); enum intel_display_power_domain intel_dsc_power_domain(struct intel_crtc *crtc, enum transcoder cpu_transcoder); struct intel_crtc *intel_dsc_get_bigjoiner_secondary(const struct intel_crtc *primary_crtc); +int intel_dsc_get_num_vdsc_instances(const struct intel_crtc_state *crtc_state); void intel_dsc_dsi_pps_write(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state); void intel_dsc_dp_pps_write(struct intel_encoder *encoder, diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index 6b01a0b68b97..ffc15d278a39 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -2174,7 +2174,7 @@ static bool skl_plane_has_rc_ccs(struct drm_i915_private *i915, return false; /* Wa_22011186057 */ - if (IS_ADLP_DISPLAY_STEP(i915, STEP_A0, STEP_B0)) + if (IS_ALDERLAKE_P(i915) && IS_DISPLAY_STEP(i915, STEP_A0, STEP_B0)) return false; if (DISPLAY_VER(i915) >= 11) @@ -2196,11 +2196,11 @@ static bool gen12_plane_has_mc_ccs(struct drm_i915_private *i915, /* Wa_14010477008 */ if (IS_DG1(i915) || IS_ROCKETLAKE(i915) || - IS_TGL_DISPLAY_STEP(i915, STEP_A0, STEP_D0)) + (IS_TIGERLAKE(i915) && IS_DISPLAY_STEP(i915, STEP_A0, STEP_D0))) return false; /* Wa_22011186057 */ - if (IS_ADLP_DISPLAY_STEP(i915, STEP_A0, STEP_B0)) + if (IS_ALDERLAKE_P(i915) && IS_DISPLAY_STEP(i915, STEP_A0, STEP_B0)) return false; /* Wa_14013215631 */ diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c index ae2f3ab3e73d..a96e7d028c5c 100644 --- a/drivers/gpu/drm/i915/display/vlv_dsi.c +++ b/drivers/gpu/drm/i915/display/vlv_dsi.c @@ -671,20 +671,6 @@ static void intel_dsi_port_disable(struct intel_encoder *encoder) intel_de_posting_read(dev_priv, port_ctrl); } } - -static void intel_dsi_wait_panel_power_cycle(struct intel_dsi *intel_dsi) -{ - ktime_t panel_power_on_time; - s64 panel_power_off_duration; - - panel_power_on_time = ktime_get_boottime(); - panel_power_off_duration = ktime_ms_delta(panel_power_on_time, - intel_dsi->panel_power_off_time); - - if (panel_power_off_duration < (s64)intel_dsi->panel_pwr_cycle_delay) - msleep(intel_dsi->panel_pwr_cycle_delay - panel_power_off_duration); -} - static void intel_dsi_prepare(struct intel_encoder *intel_encoder, const struct intel_crtc_state *pipe_config); static void intel_dsi_unprepare(struct intel_encoder *encoder); @@ -831,8 +817,6 @@ static void bxt_dsi_enable(struct intel_atomic_state *state, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { - drm_WARN_ON(state->base.dev, crtc_state->has_pch_encoder); - intel_crtc_vblank_on(crtc_state); } @@ -943,13 +927,6 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state, intel_dsi->panel_power_off_time = ktime_get_boottime(); } -static void intel_dsi_shutdown(struct intel_encoder *encoder) -{ - struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder); - - intel_dsi_wait_panel_power_cycle(intel_dsi); -} - static bool intel_dsi_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe) { diff --git a/drivers/gpu/drm/i915/gem/i915_gem_domain.c b/drivers/gpu/drm/i915/gem/i915_gem_domain.c index dfaaa8b66ac3..ffddec1d2a76 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_domain.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_domain.c @@ -68,10 +68,8 @@ flush_write_domain(struct drm_i915_gem_object *obj, unsigned int flush_domains) switch (obj->write_domain) { case I915_GEM_DOMAIN_GTT: spin_lock(&obj->vma.lock); - for_each_ggtt_vma(vma, obj) { - if (i915_vma_unset_ggtt_write(vma)) - intel_gt_flush_ggtt_writes(vma->vm->gt); - } + for_each_ggtt_vma(vma, obj) + i915_vma_flush_writes(vma); spin_unlock(&obj->vma.lock); i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index cfd7929587d8..5a687a3686bd 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -2229,8 +2229,8 @@ static int i915_reset_gen7_sol_offsets(struct i915_request *rq) u32 *cs; int i; - if (GRAPHICS_VER(rq->engine->i915) != 7 || rq->engine->id != RCS0) { - drm_dbg(&rq->engine->i915->drm, "sol reset is gen7/rcs only\n"); + if (GRAPHICS_VER(rq->i915) != 7 || rq->engine->id != RCS0) { + drm_dbg(&rq->i915->drm, "sol reset is gen7/rcs only\n"); return -EINVAL; } @@ -2691,6 +2691,7 @@ static int eb_select_engine(struct i915_execbuffer *eb) { struct intel_context *ce, *child; + struct intel_gt *gt; unsigned int idx; int err; @@ -2714,10 +2715,17 @@ eb_select_engine(struct i915_execbuffer *eb) } } eb->num_batches = ce->parallel.number_children + 1; + gt = ce->engine->gt; for_each_child(ce, child) intel_context_get(child); - intel_gt_pm_get(ce->engine->gt); + intel_gt_pm_get(gt); + /* + * Keep GT0 active on MTL so that i915_vma_parked() doesn't + * free VMAs while execbuf ioctl is validating VMAs. + */ + if (gt->info.id) + intel_gt_pm_get(to_gt(gt->i915)); if (!test_bit(CONTEXT_ALLOC_BIT, &ce->flags)) { err = intel_context_alloc_state(ce); @@ -2756,7 +2764,10 @@ eb_select_engine(struct i915_execbuffer *eb) return err; err: - intel_gt_pm_put(ce->engine->gt); + if (gt->info.id) + intel_gt_pm_put(to_gt(gt->i915)); + + intel_gt_pm_put(gt); for_each_child(ce, child) intel_context_put(child); intel_context_put(ce); @@ -2769,6 +2780,12 @@ eb_put_engine(struct i915_execbuffer *eb) struct intel_context *child; i915_vm_put(eb->context->vm); + /* + * This works in conjunction with eb_select_engine() to prevent + * i915_vma_parked() from interfering while execbuf validates vmas. + */ + if (eb->gt->info.id) + intel_gt_pm_put(to_gt(eb->gt->i915)); intel_gt_pm_put(eb->gt); for_each_child(eb->context, child) intel_context_put(child); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 97ac6fb37958..ef9346ed6d0f 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -226,7 +226,7 @@ bool i915_gem_object_can_bypass_llc(struct drm_i915_gem_object *obj) * it, but since i915 takes the stance of always zeroing memory before * handing it to userspace, we need to prevent this. */ - return IS_JSL_EHL(i915); + return (IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)); } static void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file) @@ -469,7 +469,7 @@ void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, { struct intel_frontbuffer *front; - front = __intel_frontbuffer_get(obj); + front = i915_gem_object_get_frontbuffer(obj); if (front) { intel_frontbuffer_flush(front, origin); intel_frontbuffer_put(front); @@ -481,7 +481,7 @@ void __i915_gem_object_invalidate_frontbuffer(struct drm_i915_gem_object *obj, { struct intel_frontbuffer *front; - front = __intel_frontbuffer_get(obj); + front = i915_gem_object_get_frontbuffer(obj); if (front) { intel_frontbuffer_invalidate(front, origin); intel_frontbuffer_put(front); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 884a17275b3a..f607b87890dd 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -716,10 +716,6 @@ void *__must_check i915_gem_object_pin_map(struct drm_i915_gem_object *obj, void *__must_check i915_gem_object_pin_map_unlocked(struct drm_i915_gem_object *obj, enum i915_map_type type); -enum i915_map_type i915_coherent_map_type(struct drm_i915_private *i915, - struct drm_i915_gem_object *obj, - bool always_coherent); - void __i915_gem_object_flush_map(struct drm_i915_gem_object *obj, unsigned long offset, unsigned long size); @@ -891,4 +887,71 @@ static inline int i915_gem_object_userptr_validate(struct drm_i915_gem_object *o #endif +/** + * i915_gem_object_get_frontbuffer - Get the object's frontbuffer + * @obj: The object whose frontbuffer to get. + * + * Get pointer to object's frontbuffer if such exists. Please note that RCU + * mechanism is used to handle e.g. ongoing removal of frontbuffer pointer. + * + * Return: pointer to object's frontbuffer is such exists or NULL + */ +static inline struct intel_frontbuffer * +i915_gem_object_get_frontbuffer(const struct drm_i915_gem_object *obj) +{ + struct intel_frontbuffer *front; + + if (likely(!rcu_access_pointer(obj->frontbuffer))) + return NULL; + + rcu_read_lock(); + do { + front = rcu_dereference(obj->frontbuffer); + if (!front) + break; + + if (unlikely(!kref_get_unless_zero(&front->ref))) + continue; + + if (likely(front == rcu_access_pointer(obj->frontbuffer))) + break; + + intel_frontbuffer_put(front); + } while (1); + rcu_read_unlock(); + + return front; +} + +/** + * i915_gem_object_set_frontbuffer - Set the object's frontbuffer + * @obj: The object whose frontbuffer to set. + * @front: The frontbuffer to set + * + * Set object's frontbuffer pointer. If frontbuffer is already set for the + * object keep it and return it's pointer to the caller. Please note that RCU + * mechanism is used to handle e.g. ongoing removal of frontbuffer pointer. This + * function is protected by i915->display.fb_tracking.lock + * + * Return: pointer to frontbuffer which was set. + */ +static inline struct intel_frontbuffer * +i915_gem_object_set_frontbuffer(struct drm_i915_gem_object *obj, + struct intel_frontbuffer *front) +{ + struct intel_frontbuffer *cur = front; + + if (!front) { + RCU_INIT_POINTER(obj->frontbuffer, NULL); + } else if (rcu_access_pointer(obj->frontbuffer)) { + cur = rcu_dereference_protected(obj->frontbuffer, true); + kref_get(&cur->ref); + } else { + drm_gem_object_get(intel_bo_to_drm_bo(obj)); + rcu_assign_pointer(obj->frontbuffer, front); + } + + return cur; +} + #endif diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index e72c57716bee..2292404007c8 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -17,6 +17,8 @@ #include "i915_selftest.h" #include "i915_vma_resource.h" +#include "gt/intel_gt_defines.h" + struct drm_i915_gem_object; struct intel_fronbuffer; struct intel_memory_region; @@ -675,7 +677,7 @@ struct drm_i915_gem_object { */ bool dirty:1; - u32 tlb; + u32 tlb[I915_MAX_GT]; } mm; struct { @@ -718,6 +720,9 @@ struct drm_i915_gem_object { }; }; +#define intel_bo_to_drm_bo(bo) (&(bo)->base) +#define intel_bo_to_i915(bo) to_i915(intel_bo_to_drm_bo(bo)->dev) + static inline struct drm_i915_gem_object * to_intel_bo(struct drm_gem_object *gem) { diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c index 89fc8ea6bcfc..6b6d22c19411 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c @@ -7,7 +7,7 @@ #include <drm/drm_cache.h> #include "gt/intel_gt.h" -#include "gt/intel_gt_pm.h" +#include "gt/intel_tlb.h" #include "i915_drv.h" #include "i915_gem_object.h" @@ -193,13 +193,16 @@ static void unmap_object(struct drm_i915_gem_object *obj, void *ptr) static void flush_tlb_invalidate(struct drm_i915_gem_object *obj) { struct drm_i915_private *i915 = to_i915(obj->base.dev); - struct intel_gt *gt = to_gt(i915); + struct intel_gt *gt; + int id; - if (!obj->mm.tlb) - return; + for_each_gt(gt, i915, id) { + if (!obj->mm.tlb[id]) + return; - intel_gt_invalidate_tlb(gt, obj->mm.tlb); - obj->mm.tlb = 0; + intel_gt_invalidate_tlb_full(gt, obj->mm.tlb[id]); + obj->mm.tlb[id] = 0; + } } struct sg_table * @@ -465,21 +468,6 @@ void *i915_gem_object_pin_map_unlocked(struct drm_i915_gem_object *obj, return ret; } -enum i915_map_type i915_coherent_map_type(struct drm_i915_private *i915, - struct drm_i915_gem_object *obj, - bool always_coherent) -{ - /* - * Wa_22016122933: always return I915_MAP_WC for MTL - */ - if (i915_gem_object_is_lmem(obj) || IS_METEORLAKE(i915)) - return I915_MAP_WC; - if (HAS_LLC(i915) || always_coherent) - return I915_MAP_WB; - else - return I915_MAP_WC; -} - void __i915_gem_object_flush_map(struct drm_i915_gem_object *obj, unsigned long offset, unsigned long size) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c index 3b094d36a0b0..1a766d8e7cce 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c @@ -892,7 +892,7 @@ i915_gem_stolen_lmem_setup(struct drm_i915_private *i915, u16 type, } else { resource_size_t lmem_range; - lmem_range = intel_gt_mcr_read_any(&i915->gt0, XEHP_TILE0_ADDR_RANGE) & 0xFFFF; + lmem_range = intel_gt_mcr_read_any(to_gt(i915), XEHP_TILE0_ADDR_RANGE) & 0xFFFF; lmem_size = lmem_range >> XEHP_TILE_LMEM_RANGE_SHIFT; lmem_size *= SZ_1G; } @@ -974,3 +974,39 @@ bool i915_gem_object_is_stolen(const struct drm_i915_gem_object *obj) { return obj->ops == &i915_gem_object_stolen_ops; } + +bool i915_gem_stolen_initialized(const struct drm_i915_private *i915) +{ + return drm_mm_initialized(&i915->mm.stolen); +} + +u64 i915_gem_stolen_area_address(const struct drm_i915_private *i915) +{ + return i915->dsm.stolen.start; +} + +u64 i915_gem_stolen_area_size(const struct drm_i915_private *i915) +{ + return resource_size(&i915->dsm.stolen); +} + +u64 i915_gem_stolen_node_address(const struct drm_i915_private *i915, + const struct drm_mm_node *node) +{ + return i915->dsm.stolen.start + i915_gem_stolen_node_offset(node); +} + +bool i915_gem_stolen_node_allocated(const struct drm_mm_node *node) +{ + return drm_mm_node_allocated(node); +} + +u64 i915_gem_stolen_node_offset(const struct drm_mm_node *node) +{ + return node->start; +} + +u64 i915_gem_stolen_node_size(const struct drm_mm_node *node) +{ + return node->size; +} diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.h b/drivers/gpu/drm/i915/gem/i915_gem_stolen.h index d5005a39d130..258381d1c054 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.h @@ -12,6 +12,8 @@ struct drm_i915_private; struct drm_mm_node; struct drm_i915_gem_object; +#define i915_stolen_fb drm_mm_node + int i915_gem_stolen_insert_node(struct drm_i915_private *dev_priv, struct drm_mm_node *node, u64 size, unsigned alignment); @@ -36,4 +38,15 @@ bool i915_gem_object_is_stolen(const struct drm_i915_gem_object *obj); #define I915_GEM_STOLEN_BIAS SZ_128K +bool i915_gem_stolen_initialized(const struct drm_i915_private *i915); +u64 i915_gem_stolen_area_address(const struct drm_i915_private *i915); +u64 i915_gem_stolen_area_size(const struct drm_i915_private *i915); + +u64 i915_gem_stolen_node_address(const struct drm_i915_private *i915, + const struct drm_mm_node *node); + +bool i915_gem_stolen_node_allocated(const struct drm_mm_node *node); +u64 i915_gem_stolen_node_offset(const struct drm_mm_node *node); +u64 i915_gem_stolen_node_size(const struct drm_mm_node *node); + #endif /* __I915_GEM_STOLEN_H__ */ diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c index df6c9a84252c..6b9f6cf50bf6 100644 --- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c @@ -1246,8 +1246,10 @@ static int igt_write_huge(struct drm_i915_private *i915, * times in succession a possibility by enlarging the permutation array. */ order = i915_random_order(count * count, &prng); - if (!order) - return -ENOMEM; + if (!order) { + err = -ENOMEM; + goto out; + } max_page_size = rounddown_pow_of_two(obj->mm.page_sizes.sg); max = div_u64(max - size, max_page_size); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c index a93a90b15907..d8f4a10d71de 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c @@ -13,12 +13,12 @@ #include "selftests/igt_spinner.h" static int igt_fill_check_buffer(struct drm_i915_gem_object *obj, + struct intel_gt *gt, bool fill) { - struct drm_i915_private *i915 = to_i915(obj->base.dev); unsigned int i, count = obj->base.size / sizeof(u32); enum i915_map_type map_type = - i915_coherent_map_type(i915, obj, false); + intel_gt_coherent_map_type(gt, obj, false); u32 *cur; int err = 0; @@ -66,7 +66,7 @@ static int igt_create_migrate(struct intel_gt *gt, enum intel_region_id src, if (err) continue; - err = igt_fill_check_buffer(obj, true); + err = igt_fill_check_buffer(obj, gt, true); if (err) continue; @@ -86,7 +86,7 @@ static int igt_create_migrate(struct intel_gt *gt, enum intel_region_id src, if (err) continue; - err = igt_fill_check_buffer(obj, false); + err = igt_fill_check_buffer(obj, gt, false); } i915_gem_object_put(obj); @@ -233,7 +233,7 @@ static int __igt_lmem_pages_migrate(struct intel_gt *gt, continue; if (!vma) { - err = igt_fill_check_buffer(obj, true); + err = igt_fill_check_buffer(obj, gt, true); if (err) continue; } @@ -276,7 +276,7 @@ static int __igt_lmem_pages_migrate(struct intel_gt *gt, if (err) goto out_unlock; } else { - err = igt_fill_check_buffer(obj, false); + err = igt_fill_check_buffer(obj, gt, false); } out_unlock: diff --git a/drivers/gpu/drm/i915/gt/gen2_engine_cs.c b/drivers/gpu/drm/i915/gt/gen2_engine_cs.c index 1c82caf525c3..8fe0499308ff 100644 --- a/drivers/gpu/drm/i915/gt/gen2_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen2_engine_cs.c @@ -76,7 +76,7 @@ int gen4_emit_flush_rcs(struct i915_request *rq, u32 mode) cmd = MI_FLUSH; if (mode & EMIT_INVALIDATE) { cmd |= MI_EXE_FLUSH; - if (IS_G4X(rq->engine->i915) || GRAPHICS_VER(rq->engine->i915) == 5) + if (IS_G4X(rq->i915) || GRAPHICS_VER(rq->i915) == 5) cmd |= MI_INVALIDATE_ISP; } diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c index 23857cc08eca..a4ff55aa5e55 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c @@ -39,11 +39,11 @@ int gen8_emit_flush_rcs(struct i915_request *rq, u32 mode) * On GEN9: before VF_CACHE_INVALIDATE we need to emit a NULL * pipe control. */ - if (GRAPHICS_VER(rq->engine->i915) == 9) + if (GRAPHICS_VER(rq->i915) == 9) vf_flush_wa = true; /* WaForGAMHang:kbl */ - if (IS_KBL_GRAPHICS_STEP(rq->engine->i915, 0, STEP_C0)) + if (IS_KABYLAKE(rq->i915) && IS_GRAPHICS_STEP(rq->i915, 0, STEP_C0)) dc_flush_wa = true; } @@ -165,14 +165,60 @@ static u32 preparser_disable(bool state) return MI_ARB_CHECK | 1 << 8 | state; } -u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv_reg) +static i915_reg_t gen12_get_aux_inv_reg(struct intel_engine_cs *engine) { - u32 gsi_offset = gt->uncore->gsi_offset; + switch (engine->id) { + case RCS0: + return GEN12_CCS_AUX_INV; + case BCS0: + return GEN12_BCS0_AUX_INV; + case VCS0: + return GEN12_VD0_AUX_INV; + case VCS2: + return GEN12_VD2_AUX_INV; + case VECS0: + return GEN12_VE0_AUX_INV; + case CCS0: + return GEN12_CCS0_AUX_INV; + default: + return INVALID_MMIO_REG; + } +} + +static bool gen12_needs_ccs_aux_inv(struct intel_engine_cs *engine) +{ + i915_reg_t reg = gen12_get_aux_inv_reg(engine); + + if (IS_PONTEVECCHIO(engine->i915)) + return false; + + /* + * So far platforms supported by i915 having flat ccs do not require + * AUX invalidation. Check also whether the engine requires it. + */ + return i915_mmio_reg_valid(reg) && !HAS_FLAT_CCS(engine->i915); +} + +u32 *gen12_emit_aux_table_inv(struct intel_engine_cs *engine, u32 *cs) +{ + i915_reg_t inv_reg = gen12_get_aux_inv_reg(engine); + u32 gsi_offset = engine->gt->uncore->gsi_offset; + + if (!gen12_needs_ccs_aux_inv(engine)) + return cs; *cs++ = MI_LOAD_REGISTER_IMM(1) | MI_LRI_MMIO_REMAP_EN; *cs++ = i915_mmio_reg_offset(inv_reg) + gsi_offset; *cs++ = AUX_INV; - *cs++ = MI_NOOP; + + *cs++ = MI_SEMAPHORE_WAIT_TOKEN | + MI_SEMAPHORE_REGISTER_POLL | + MI_SEMAPHORE_POLL | + MI_SEMAPHORE_SAD_EQ_SDD; + *cs++ = 0; + *cs++ = i915_mmio_reg_offset(inv_reg) + gsi_offset; + *cs++ = 0; + *cs++ = 0; return cs; } @@ -180,8 +226,8 @@ u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv static int mtl_dummy_pipe_control(struct i915_request *rq) { /* Wa_14016712196 */ - if (IS_MTL_GRAPHICS_STEP(rq->engine->i915, M, STEP_A0, STEP_B0) || - IS_MTL_GRAPHICS_STEP(rq->engine->i915, P, STEP_A0, STEP_B0)) { + if (IS_MTL_GRAPHICS_STEP(rq->i915, M, STEP_A0, STEP_B0) || + IS_MTL_GRAPHICS_STEP(rq->i915, P, STEP_A0, STEP_B0)) { u32 *cs; /* dummy PIPE_CONTROL + depth flush */ @@ -202,8 +248,13 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) { struct intel_engine_cs *engine = rq->engine; - if (mode & EMIT_FLUSH) { - u32 flags = 0; + /* + * On Aux CCS platforms the invalidation of the Aux + * table requires quiescing memory traffic beforehand + */ + if (mode & EMIT_FLUSH || gen12_needs_ccs_aux_inv(engine)) { + u32 bit_group_0 = 0; + u32 bit_group_1 = 0; int err; u32 *cs; @@ -211,32 +262,40 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) if (err) return err; - flags |= PIPE_CONTROL_TILE_CACHE_FLUSH; - flags |= PIPE_CONTROL_FLUSH_L3; - flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH; - flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH; + bit_group_0 |= PIPE_CONTROL0_HDC_PIPELINE_FLUSH; + + /* + * When required, in MTL and beyond platforms we + * need to set the CCS_FLUSH bit in the pipe control + */ + if (GRAPHICS_VER_FULL(rq->i915) >= IP_VER(12, 70)) + bit_group_0 |= PIPE_CONTROL_CCS_FLUSH; + + bit_group_1 |= PIPE_CONTROL_TILE_CACHE_FLUSH; + bit_group_1 |= PIPE_CONTROL_FLUSH_L3; + bit_group_1 |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH; + bit_group_1 |= PIPE_CONTROL_DEPTH_CACHE_FLUSH; /* Wa_1409600907:tgl,adl-p */ - flags |= PIPE_CONTROL_DEPTH_STALL; - flags |= PIPE_CONTROL_DC_FLUSH_ENABLE; - flags |= PIPE_CONTROL_FLUSH_ENABLE; + bit_group_1 |= PIPE_CONTROL_DEPTH_STALL; + bit_group_1 |= PIPE_CONTROL_DC_FLUSH_ENABLE; + bit_group_1 |= PIPE_CONTROL_FLUSH_ENABLE; - flags |= PIPE_CONTROL_STORE_DATA_INDEX; - flags |= PIPE_CONTROL_QW_WRITE; + bit_group_1 |= PIPE_CONTROL_STORE_DATA_INDEX; + bit_group_1 |= PIPE_CONTROL_QW_WRITE; - flags |= PIPE_CONTROL_CS_STALL; + bit_group_1 |= PIPE_CONTROL_CS_STALL; if (!HAS_3D_PIPELINE(engine->i915)) - flags &= ~PIPE_CONTROL_3D_ARCH_FLAGS; + bit_group_1 &= ~PIPE_CONTROL_3D_ARCH_FLAGS; else if (engine->class == COMPUTE_CLASS) - flags &= ~PIPE_CONTROL_3D_ENGINE_FLAGS; + bit_group_1 &= ~PIPE_CONTROL_3D_ENGINE_FLAGS; cs = intel_ring_begin(rq, 6); if (IS_ERR(cs)) return PTR_ERR(cs); - cs = gen12_emit_pipe_control(cs, - PIPE_CONTROL0_HDC_PIPELINE_FLUSH, - flags, LRC_PPHWSP_SCRATCH_ADDR); + cs = gen12_emit_pipe_control(cs, bit_group_0, bit_group_1, + LRC_PPHWSP_SCRATCH_ADDR); intel_ring_advance(rq, cs); } @@ -267,10 +326,9 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) else if (engine->class == COMPUTE_CLASS) flags &= ~PIPE_CONTROL_3D_ENGINE_FLAGS; - if (!HAS_FLAT_CCS(rq->engine->i915)) - count = 8 + 4; - else - count = 8; + count = 8; + if (gen12_needs_ccs_aux_inv(rq->engine)) + count += 8; cs = intel_ring_begin(rq, count); if (IS_ERR(cs)) @@ -285,11 +343,7 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) cs = gen8_emit_pipe_control(cs, flags, LRC_PPHWSP_SCRATCH_ADDR); - if (!HAS_FLAT_CCS(rq->engine->i915)) { - /* hsdes: 1809175790 */ - cs = gen12_emit_aux_table_inv(rq->engine->gt, - cs, GEN12_GFX_CCS_AUX_NV); - } + cs = gen12_emit_aux_table_inv(engine, cs); *cs++ = preparser_disable(false); intel_ring_advance(rq, cs); @@ -300,21 +354,14 @@ int gen12_emit_flush_rcs(struct i915_request *rq, u32 mode) int gen12_emit_flush_xcs(struct i915_request *rq, u32 mode) { - intel_engine_mask_t aux_inv = 0; - u32 cmd, *cs; + u32 cmd = 4; + u32 *cs; - cmd = 4; if (mode & EMIT_INVALIDATE) { cmd += 2; - if (!HAS_FLAT_CCS(rq->engine->i915) && - (rq->engine->class == VIDEO_DECODE_CLASS || - rq->engine->class == VIDEO_ENHANCEMENT_CLASS)) { - aux_inv = rq->engine->mask & - ~GENMASK(_BCS(I915_MAX_BCS - 1), BCS0); - if (aux_inv) - cmd += 4; - } + if (gen12_needs_ccs_aux_inv(rq->engine)) + cmd += 8; } cs = intel_ring_begin(rq, cmd); @@ -338,6 +385,10 @@ int gen12_emit_flush_xcs(struct i915_request *rq, u32 mode) cmd |= MI_INVALIDATE_TLB; if (rq->engine->class == VIDEO_DECODE_CLASS) cmd |= MI_INVALIDATE_BSD; + + if (gen12_needs_ccs_aux_inv(rq->engine) && + rq->engine->class == COPY_ENGINE_CLASS) + cmd |= MI_FLUSH_DW_CCS; } *cs++ = cmd; @@ -345,14 +396,7 @@ int gen12_emit_flush_xcs(struct i915_request *rq, u32 mode) *cs++ = 0; /* upper addr */ *cs++ = 0; /* value */ - if (aux_inv) { /* hsdes: 1809175790 */ - if (rq->engine->class == VIDEO_DECODE_CLASS) - cs = gen12_emit_aux_table_inv(rq->engine->gt, - cs, GEN12_VD0_AUX_NV); - else - cs = gen12_emit_aux_table_inv(rq->engine->gt, - cs, GEN12_VE0_AUX_NV); - } + cs = gen12_emit_aux_table_inv(rq->engine, cs); if (mode & EMIT_INVALIDATE) *cs++ = preparser_disable(false); @@ -754,7 +798,7 @@ u32 *gen12_emit_fini_breadcrumb_xcs(struct i915_request *rq, u32 *cs) u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs) { - struct drm_i915_private *i915 = rq->engine->i915; + struct drm_i915_private *i915 = rq->i915; u32 flags = (PIPE_CONTROL_CS_STALL | PIPE_CONTROL_TLB_INVALIDATE | PIPE_CONTROL_TILE_CACHE_FLUSH | @@ -775,7 +819,7 @@ u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs) /* Wa_1409600907 */ flags |= PIPE_CONTROL_DEPTH_STALL; - if (!HAS_3D_PIPELINE(rq->engine->i915)) + if (!HAS_3D_PIPELINE(rq->i915)) flags &= ~PIPE_CONTROL_3D_ARCH_FLAGS; else if (rq->engine->class == COMPUTE_CLASS) flags &= ~PIPE_CONTROL_3D_ENGINE_FLAGS; diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.h b/drivers/gpu/drm/i915/gt/gen8_engine_cs.h index 655e5c00ddc2..867ba697aceb 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.h +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.h @@ -13,6 +13,7 @@ #include "intel_gt_regs.h" #include "intel_gpu_commands.h" +struct intel_engine_cs; struct intel_gt; struct i915_request; @@ -46,28 +47,32 @@ u32 *gen8_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs); u32 *gen11_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs); u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs); -u32 *gen12_emit_aux_table_inv(struct intel_gt *gt, u32 *cs, const i915_reg_t inv_reg); +u32 *gen12_emit_aux_table_inv(struct intel_engine_cs *engine, u32 *cs); static inline u32 * -__gen8_emit_pipe_control(u32 *batch, u32 flags0, u32 flags1, u32 offset) +__gen8_emit_pipe_control(u32 *batch, u32 bit_group_0, + u32 bit_group_1, u32 offset) { memset(batch, 0, 6 * sizeof(u32)); - batch[0] = GFX_OP_PIPE_CONTROL(6) | flags0; - batch[1] = flags1; + batch[0] = GFX_OP_PIPE_CONTROL(6) | bit_group_0; + batch[1] = bit_group_1; batch[2] = offset; return batch + 6; } -static inline u32 *gen8_emit_pipe_control(u32 *batch, u32 flags, u32 offset) +static inline u32 *gen8_emit_pipe_control(u32 *batch, + u32 bit_group_1, u32 offset) { - return __gen8_emit_pipe_control(batch, 0, flags, offset); + return __gen8_emit_pipe_control(batch, 0, bit_group_1, offset); } -static inline u32 *gen12_emit_pipe_control(u32 *batch, u32 flags0, u32 flags1, u32 offset) +static inline u32 *gen12_emit_pipe_control(u32 *batch, u32 bit_group_0, + u32 bit_group_1, u32 offset) { - return __gen8_emit_pipe_control(batch, flags0, flags1, offset); + return __gen8_emit_pipe_control(batch, bit_group_0, + bit_group_1, offset); } static inline u32 * diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index 0aff5bb13c53..ee15486fed0d 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -1333,6 +1333,7 @@ static int measure_breadcrumb_dw(struct intel_context *ce) if (!frame) return -ENOMEM; + frame->rq.i915 = engine->i915; frame->rq.engine = engine; frame->rq.context = ce; rcu_assign_pointer(frame->rq.timeline, ce->timeline); diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c index 21af0ec52223..b538b5c04948 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c @@ -39,7 +39,7 @@ static void dbg_poison_ce(struct intel_context *ce) if (ce->state) { struct drm_i915_gem_object *obj = ce->state->obj; - int type = i915_coherent_map_type(ce->engine->i915, obj, true); + int type = intel_gt_coherent_map_type(ce->engine->gt, obj, true); void *map; if (!i915_gem_object_trylock(obj, NULL)) diff --git a/drivers/gpu/drm/i915/gt/intel_engine_types.h b/drivers/gpu/drm/i915/gt/intel_engine_types.h index e99a6fa03d45..a7e677598004 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_types.h +++ b/drivers/gpu/drm/i915/gt/intel_engine_types.h @@ -58,6 +58,7 @@ struct i915_perf_group; typedef u32 intel_engine_mask_t; #define ALL_ENGINES ((intel_engine_mask_t)~0ul) +#define VIRTUAL_ENGINES BIT(BITS_PER_TYPE(intel_engine_mask_t) - 1) struct intel_hw_status_page { struct list_head timelines; diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c index 2ebd937f3b4c..8a641bcf777c 100644 --- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c @@ -2718,7 +2718,7 @@ static int emit_pdps(struct i915_request *rq) int err, i; u32 *cs; - GEM_BUG_ON(intel_vgpu_active(rq->engine->i915)); + GEM_BUG_ON(intel_vgpu_active(rq->i915)); /* * Beware ye of the dragons, this sequence is magic! @@ -3556,16 +3556,16 @@ int intel_execlists_submission_setup(struct intel_engine_cs *engine) lrc_init_wa_ctx(engine); if (HAS_LOGICAL_RING_ELSQ(i915)) { - execlists->submit_reg = uncore->regs + + execlists->submit_reg = intel_uncore_regs(uncore) + i915_mmio_reg_offset(RING_EXECLIST_SQ_CONTENTS(base)); - execlists->ctrl_reg = uncore->regs + + execlists->ctrl_reg = intel_uncore_regs(uncore) + i915_mmio_reg_offset(RING_EXECLIST_CONTROL(base)); engine->fw_domain = intel_uncore_forcewake_for_reg(engine->uncore, RING_EXECLIST_CONTROL(engine->mmio_base), FW_REG_WRITE); } else { - execlists->submit_reg = uncore->regs + + execlists->submit_reg = intel_uncore_regs(uncore) + i915_mmio_reg_offset(RING_ELSP(base)); } diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c index 37d0b0fe791d..40371b8a9bbb 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c @@ -818,7 +818,7 @@ i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj, if (obj->bit_17 == NULL) { obj->bit_17 = bitmap_zalloc(page_count, GFP_KERNEL); if (obj->bit_17 == NULL) { - drm_err(&to_i915(obj->base.dev)->drm, + drm_err(obj->base.dev, "Failed to allocate memory for bit 17 record\n"); return; } diff --git a/drivers/gpu/drm/i915/gt/intel_gpu_commands.h b/drivers/gpu/drm/i915/gt/intel_gpu_commands.h index 5d143e2a8db0..2bd8d98d2110 100644 --- a/drivers/gpu/drm/i915/gt/intel_gpu_commands.h +++ b/drivers/gpu/drm/i915/gt/intel_gpu_commands.h @@ -121,6 +121,7 @@ #define MI_SEMAPHORE_TARGET(engine) ((engine)<<15) #define MI_SEMAPHORE_WAIT MI_INSTR(0x1c, 2) /* GEN8+ */ #define MI_SEMAPHORE_WAIT_TOKEN MI_INSTR(0x1c, 3) /* GEN12+ */ +#define MI_SEMAPHORE_REGISTER_POLL (1 << 16) #define MI_SEMAPHORE_POLL (1 << 15) #define MI_SEMAPHORE_SAD_GT_SDD (0 << 12) #define MI_SEMAPHORE_SAD_GTE_SDD (1 << 12) @@ -299,6 +300,7 @@ #define PIPE_CONTROL_QW_WRITE (1<<14) #define PIPE_CONTROL_POST_SYNC_OP_MASK (3<<14) #define PIPE_CONTROL_DEPTH_STALL (1<<13) +#define PIPE_CONTROL_CCS_FLUSH (1<<13) /* MTL+ */ #define PIPE_CONTROL_WRITE_FLUSH (1<<12) #define PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH (1<<12) /* gen6+ */ #define PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE (1<<11) /* MBZ on ILK */ diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index 7a008e829d4d..449f0b7fc843 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -33,6 +33,7 @@ #include "intel_rps.h" #include "intel_sa_media.h" #include "intel_gt_sysfs.h" +#include "intel_tlb.h" #include "intel_uncore.h" #include "shmem_utils.h" @@ -50,8 +51,7 @@ void intel_gt_common_init_early(struct intel_gt *gt) intel_gt_init_reset(gt); intel_gt_init_requests(gt); intel_gt_init_timelines(gt); - mutex_init(>->tlb.invalidate_lock); - seqcount_mutex_init(>->tlb.seqno, >->tlb.invalidate_lock); + intel_gt_init_tlb(gt); intel_gt_pm_init_early(gt); intel_wopcm_init_early(>->wopcm); @@ -179,7 +179,7 @@ int intel_gt_init_hw(struct intel_gt *gt) if (IS_HASWELL(i915)) intel_uncore_write(uncore, HSW_MI_PREDICATE_RESULT_2, - IS_HSW_GT3(i915) ? + IS_HASWELL_GT3(i915) ? LOWER_SLICE_ENABLED : LOWER_SLICE_DISABLED); /* Apply the GT workarounds... */ @@ -466,7 +466,7 @@ static int intel_gt_init_scratch(struct intel_gt *gt, unsigned int size) obj = i915_gem_object_create_lmem(i915, size, I915_BO_ALLOC_VOLATILE | I915_BO_ALLOC_GPU_ONLY); - if (IS_ERR(obj)) + if (IS_ERR(obj) && !IS_METEORLAKE(i915)) /* Wa_22018444074 */ obj = i915_gem_object_create_stolen(i915, size); if (IS_ERR(obj)) obj = i915_gem_object_create_internal(i915, size); @@ -846,7 +846,7 @@ void intel_gt_driver_late_release_all(struct drm_i915_private *i915) intel_gt_fini_requests(gt); intel_gt_fini_reset(gt); intel_gt_fini_timelines(gt); - mutex_destroy(>->tlb.invalidate_lock); + intel_gt_fini_tlb(gt); intel_engines_free(gt); } } @@ -887,7 +887,7 @@ static int intel_gt_tile_setup(struct intel_gt *gt, phys_addr_t phys_addr) int intel_gt_probe_all(struct drm_i915_private *i915) { struct pci_dev *pdev = to_pci_dev(i915->drm.dev); - struct intel_gt *gt = &i915->gt0; + struct intel_gt *gt = to_gt(i915); const struct intel_gt_definition *gtdef; phys_addr_t phys_addr; unsigned int mmio_bar; @@ -904,7 +904,7 @@ int intel_gt_probe_all(struct drm_i915_private *i915) */ gt->i915 = i915; gt->name = "Primary GT"; - gt->info.engine_mask = RUNTIME_INFO(i915)->platform_engine_mask; + gt->info.engine_mask = INTEL_INFO(i915)->platform_engine_mask; gt_dbg(gt, "Setting up %s\n", gt->name); ret = intel_gt_tile_setup(gt, phys_addr); @@ -1004,136 +1004,18 @@ void intel_gt_info_print(const struct intel_gt_info *info, intel_sseu_dump(&info->sseu, p); } -/* - * HW architecture suggest typical invalidation time at 40us, - * with pessimistic cases up to 100us and a recommendation to - * cap at 1ms. We go a bit higher just in case. - */ -#define TLB_INVAL_TIMEOUT_US 100 -#define TLB_INVAL_TIMEOUT_MS 4 - -/* - * On Xe_HP the TLB invalidation registers are located at the same MMIO offsets - * but are now considered MCR registers. Since they exist within a GAM range, - * the primary instance of the register rolls up the status from each unit. - */ -static int wait_for_invalidate(struct intel_engine_cs *engine) +enum i915_map_type intel_gt_coherent_map_type(struct intel_gt *gt, + struct drm_i915_gem_object *obj, + bool always_coherent) { - if (engine->tlb_inv.mcr) - return intel_gt_mcr_wait_for_reg(engine->gt, - engine->tlb_inv.reg.mcr_reg, - engine->tlb_inv.done, - 0, - TLB_INVAL_TIMEOUT_US, - TLB_INVAL_TIMEOUT_MS); - else - return __intel_wait_for_register_fw(engine->gt->uncore, - engine->tlb_inv.reg.reg, - engine->tlb_inv.done, - 0, - TLB_INVAL_TIMEOUT_US, - TLB_INVAL_TIMEOUT_MS, - NULL); -} - -static void mmio_invalidate_full(struct intel_gt *gt) -{ - struct drm_i915_private *i915 = gt->i915; - struct intel_uncore *uncore = gt->uncore; - struct intel_engine_cs *engine; - intel_engine_mask_t awake, tmp; - enum intel_engine_id id; - unsigned long flags; - - if (GRAPHICS_VER(i915) < 8) - return; - - intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL); - - intel_gt_mcr_lock(gt, &flags); - spin_lock(&uncore->lock); /* serialise invalidate with GT reset */ - - awake = 0; - for_each_engine(engine, gt, id) { - if (!intel_engine_pm_is_awake(engine)) - continue; - - if (engine->tlb_inv.mcr) - intel_gt_mcr_multicast_write_fw(gt, - engine->tlb_inv.reg.mcr_reg, - engine->tlb_inv.request); - else - intel_uncore_write_fw(uncore, - engine->tlb_inv.reg.reg, - engine->tlb_inv.request); - - awake |= engine->mask; - } - - GT_TRACE(gt, "invalidated engines %08x\n", awake); - - /* Wa_2207587034:tgl,dg1,rkl,adl-s,adl-p */ - if (awake && - (IS_TIGERLAKE(i915) || - IS_DG1(i915) || - IS_ROCKETLAKE(i915) || - IS_ALDERLAKE_S(i915) || - IS_ALDERLAKE_P(i915))) - intel_uncore_write_fw(uncore, GEN12_OA_TLB_INV_CR, 1); - - spin_unlock(&uncore->lock); - intel_gt_mcr_unlock(gt, flags); - - for_each_engine_masked(engine, gt, awake, tmp) { - if (wait_for_invalidate(engine)) - gt_err_ratelimited(gt, - "%s TLB invalidation did not complete in %ums!\n", - engine->name, TLB_INVAL_TIMEOUT_MS); - } - /* - * Use delayed put since a) we mostly expect a flurry of TLB - * invalidations so it is good to avoid paying the forcewake cost and - * b) it works around a bug in Icelake which cannot cope with too rapid - * transitions. + * Wa_22016122933: always return I915_MAP_WC for Media + * version 13.0 when the object is on the Media GT */ - intel_uncore_forcewake_put_delayed(uncore, FORCEWAKE_ALL); -} - -static bool tlb_seqno_passed(const struct intel_gt *gt, u32 seqno) -{ - u32 cur = intel_gt_tlb_seqno(gt); - - /* Only skip if a *full* TLB invalidate barrier has passed */ - return (s32)(cur - ALIGN(seqno, 2)) > 0; -} - -void intel_gt_invalidate_tlb(struct intel_gt *gt, u32 seqno) -{ - intel_wakeref_t wakeref; - - if (I915_SELFTEST_ONLY(gt->awake == -ENODEV)) - return; - - if (intel_gt_is_wedged(gt)) - return; - - if (tlb_seqno_passed(gt, seqno)) - return; - - with_intel_gt_pm_if_awake(gt, wakeref) { - mutex_lock(>->tlb.invalidate_lock); - if (tlb_seqno_passed(gt, seqno)) - goto unlock; - - mmio_invalidate_full(gt); - - write_seqcount_invalidate(>->tlb.seqno); -unlock: - mutex_unlock(>->tlb.invalidate_lock); - } + if (i915_gem_object_is_lmem(obj) || intel_gt_needs_wa_22016122933(gt)) + return I915_MAP_WC; + if (HAS_LLC(gt->i915) || always_coherent) + return I915_MAP_WB; + else + return I915_MAP_WC; } - -#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) -#include "selftest_tlb.c" -#endif diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index d2f4fbde5f9f..6c34547b58b5 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -6,6 +6,7 @@ #ifndef __INTEL_GT__ #define __INTEL_GT__ +#include "i915_drv.h" #include "intel_engine_types.h" #include "intel_gt_types.h" #include "intel_reset.h" @@ -24,6 +25,11 @@ static inline bool gt_is_root(struct intel_gt *gt) return !gt->info.id; } +static inline bool intel_gt_needs_wa_22016122933(struct intel_gt *gt) +{ + return MEDIA_VER_FULL(gt->i915) == IP_VER(13, 0) && gt->type == GT_MEDIA; +} + static inline struct intel_gt *uc_to_gt(struct intel_uc *uc) { return container_of(uc, struct intel_gt, uc); @@ -107,16 +113,8 @@ void intel_gt_info_print(const struct intel_gt_info *info, void intel_gt_watchdog_work(struct work_struct *work); -static inline u32 intel_gt_tlb_seqno(const struct intel_gt *gt) -{ - return seqprop_sequence(>->tlb.seqno); -} - -static inline u32 intel_gt_next_invalidate_tlb_full(const struct intel_gt *gt) -{ - return intel_gt_tlb_seqno(gt) | 1; -} - -void intel_gt_invalidate_tlb(struct intel_gt *gt, u32 seqno); +enum i915_map_type intel_gt_coherent_map_type(struct intel_gt *gt, + struct drm_i915_gem_object *obj, + bool always_coherent); #endif /* __INTEL_GT_H__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gt_defines.h b/drivers/gpu/drm/i915/gt/intel_gt_defines.h new file mode 100644 index 000000000000..5017788bac8f --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_gt_defines.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2019 Intel Corporation + */ + +#ifndef __INTEL_GT_DEFINES__ +#define __INTEL_GT_DEFINES__ + +#define I915_MAX_GT 2 + +#endif diff --git a/drivers/gpu/drm/i915/gt/intel_gt_irq.c b/drivers/gpu/drm/i915/gt/intel_gt_irq.c index 62fd00c9e519..77fb57223465 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_irq.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_irq.c @@ -31,7 +31,7 @@ static u32 gen11_gt_engine_identity(struct intel_gt *gt, const unsigned int bank, const unsigned int bit) { - void __iomem * const regs = gt->uncore->regs; + void __iomem * const regs = intel_uncore_regs(gt->uncore); u32 timeout_ts; u32 ident; @@ -148,7 +148,7 @@ gen11_gt_identity_handler(struct intel_gt *gt, const u32 identity) static void gen11_gt_bank_handler(struct intel_gt *gt, const unsigned int bank) { - void __iomem * const regs = gt->uncore->regs; + void __iomem * const regs = intel_uncore_regs(gt->uncore); unsigned long intr_dw; unsigned int bit; @@ -183,7 +183,7 @@ void gen11_gt_irq_handler(struct intel_gt *gt, const u32 master_ctl) bool gen11_gt_reset_one_iir(struct intel_gt *gt, const unsigned int bank, const unsigned int bit) { - void __iomem * const regs = gt->uncore->regs; + void __iomem * const regs = intel_uncore_regs(gt->uncore); u32 dw; lockdep_assert_held(gt->irq_lock); @@ -404,7 +404,7 @@ void gen6_gt_irq_handler(struct intel_gt *gt, u32 gt_iir) void gen8_gt_irq_handler(struct intel_gt *gt, u32 master_ctl) { - void __iomem * const regs = gt->uncore->regs; + void __iomem * const regs = intel_uncore_regs(gt->uncore); u32 iir; if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) { diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h index 718cb2c80f79..2cdfb2f713d0 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h @@ -332,9 +332,11 @@ #define GEN8_PRIVATE_PAT_HI _MMIO(0x40e0 + 4) #define GEN10_PAT_INDEX(index) _MMIO(0x40e0 + (index) * 4) #define BSD_HWS_PGA_GEN7 _MMIO(0x4180) -#define GEN12_GFX_CCS_AUX_NV _MMIO(0x4208) -#define GEN12_VD0_AUX_NV _MMIO(0x4218) -#define GEN12_VD1_AUX_NV _MMIO(0x4228) + +#define GEN12_CCS_AUX_INV _MMIO(0x4208) +#define GEN12_VD0_AUX_INV _MMIO(0x4218) +#define GEN12_VE0_AUX_INV _MMIO(0x4238) +#define GEN12_BCS0_AUX_INV _MMIO(0x4248) #define GEN8_RTCR _MMIO(0x4260) #define GEN8_M1TCR _MMIO(0x4264) @@ -342,14 +344,12 @@ #define GEN8_BTCR _MMIO(0x426c) #define GEN8_VTCR _MMIO(0x4270) -#define GEN12_VD2_AUX_NV _MMIO(0x4298) -#define GEN12_VD3_AUX_NV _MMIO(0x42a8) -#define GEN12_VE0_AUX_NV _MMIO(0x4238) - #define BLT_HWS_PGA_GEN7 _MMIO(0x4280) -#define GEN12_VE1_AUX_NV _MMIO(0x42b8) +#define GEN12_VD2_AUX_INV _MMIO(0x4298) +#define GEN12_CCS0_AUX_INV _MMIO(0x42c8) #define AUX_INV REG_BIT(0) + #define VEBOX_HWS_PGA_GEN7 _MMIO(0x4380) #define GEN12_AUX_ERR_DBG _MMIO(0x43f4) diff --git a/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c b/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c index ee2b44f896a2..f0dea54880af 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c @@ -701,6 +701,80 @@ static const struct attribute *media_perf_power_attrs[] = { }; static ssize_t +rps_up_threshold_pct_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name); + struct intel_rps *rps = >->rps; + + return sysfs_emit(buf, "%u\n", intel_rps_get_up_threshold(rps)); +} + +static ssize_t +rps_up_threshold_pct_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name); + struct intel_rps *rps = >->rps; + int ret; + u8 val; + + ret = kstrtou8(buf, 10, &val); + if (ret) + return ret; + + ret = intel_rps_set_up_threshold(rps, val); + + return ret == 0 ? count : ret; +} + +static struct kobj_attribute rps_up_threshold_pct = + __ATTR(rps_up_threshold_pct, + 0664, + rps_up_threshold_pct_show, + rps_up_threshold_pct_store); + +static ssize_t +rps_down_threshold_pct_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name); + struct intel_rps *rps = >->rps; + + return sysfs_emit(buf, "%u\n", intel_rps_get_down_threshold(rps)); +} + +static ssize_t +rps_down_threshold_pct_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name); + struct intel_rps *rps = >->rps; + int ret; + u8 val; + + ret = kstrtou8(buf, 10, &val); + if (ret) + return ret; + + ret = intel_rps_set_down_threshold(rps, val); + + return ret == 0 ? count : ret; +} + +static struct kobj_attribute rps_down_threshold_pct = + __ATTR(rps_down_threshold_pct, + 0664, + rps_down_threshold_pct_show, + rps_down_threshold_pct_store); + +static const struct attribute * const gen6_gt_rps_attrs[] = { + &rps_up_threshold_pct.attr, + &rps_down_threshold_pct.attr, + NULL +}; + +static ssize_t default_min_freq_mhz_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { struct intel_gt *gt = kobj_to_gt(kobj->parent); @@ -722,9 +796,37 @@ default_max_freq_mhz_show(struct kobject *kobj, struct kobj_attribute *attr, cha static struct kobj_attribute default_max_freq_mhz = __ATTR(rps_max_freq_mhz, 0444, default_max_freq_mhz_show, NULL); +static ssize_t +default_rps_up_threshold_pct_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + struct intel_gt *gt = kobj_to_gt(kobj->parent); + + return sysfs_emit(buf, "%u\n", gt->defaults.rps_up_threshold); +} + +static struct kobj_attribute default_rps_up_threshold_pct = +__ATTR(rps_up_threshold_pct, 0444, default_rps_up_threshold_pct_show, NULL); + +static ssize_t +default_rps_down_threshold_pct_show(struct kobject *kobj, + struct kobj_attribute *attr, + char *buf) +{ + struct intel_gt *gt = kobj_to_gt(kobj->parent); + + return sysfs_emit(buf, "%u\n", gt->defaults.rps_down_threshold); +} + +static struct kobj_attribute default_rps_down_threshold_pct = +__ATTR(rps_down_threshold_pct, 0444, default_rps_down_threshold_pct_show, NULL); + static const struct attribute * const rps_defaults_attrs[] = { &default_min_freq_mhz.attr, &default_max_freq_mhz.attr, + &default_rps_up_threshold_pct.attr, + &default_rps_down_threshold_pct.attr, NULL }; @@ -752,6 +854,12 @@ static int intel_sysfs_rps_init(struct intel_gt *gt, struct kobject *kobj) if (IS_VALLEYVIEW(gt->i915) || IS_CHERRYVIEW(gt->i915)) ret = sysfs_create_file(kobj, vlv_attr); + if (is_object_gt(kobj) && !intel_uc_uses_guc_slpc(>->uc)) { + ret = sysfs_create_files(kobj, gen6_gt_rps_attrs); + if (ret) + return ret; + } + return ret; } diff --git a/drivers/gpu/drm/i915/gt/intel_gt_types.h b/drivers/gpu/drm/i915/gt/intel_gt_types.h index f08c2556aa25..def7dd0eb6f1 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_types.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_types.h @@ -83,6 +83,9 @@ enum intel_submission_method { struct gt_defaults { u32 min_freq; u32 max_freq; + + u8 rps_up_threshold; + u8 rps_down_threshold; }; enum intel_gt_type { @@ -306,4 +309,6 @@ enum intel_gt_scratch_field { INTEL_GT_SCRATCH_FIELD_COHERENTL3_WA = 256, }; +#define intel_gt_support_legacy_fencing(gt) ((gt)->ggtt->num_fences > 0) + #endif /* __INTEL_GT_TYPES_H__ */ diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c index 731d9f2bbc56..13944a14ea2d 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.c +++ b/drivers/gpu/drm/i915/gt/intel_gtt.c @@ -89,7 +89,7 @@ int map_pt_dma(struct i915_address_space *vm, struct drm_i915_gem_object *obj) enum i915_map_type type; void *vaddr; - type = i915_coherent_map_type(vm->i915, obj, true); + type = intel_gt_coherent_map_type(vm->gt, obj, true); vaddr = i915_gem_object_pin_map_unlocked(obj, type); if (IS_ERR(vaddr)) return PTR_ERR(vaddr); @@ -103,7 +103,7 @@ int map_pt_dma_locked(struct i915_address_space *vm, struct drm_i915_gem_object enum i915_map_type type; void *vaddr; - type = i915_coherent_map_type(vm->i915, obj, true); + type = intel_gt_coherent_map_type(vm->gt, obj, true); vaddr = i915_gem_object_pin_map(obj, type); if (IS_ERR(vaddr)) return PTR_ERR(vaddr); diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index a4ec20aaafe2..957d0aeb0c02 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -1092,8 +1092,16 @@ __lrc_alloc_state(struct intel_context *ce, struct intel_engine_cs *engine) obj = i915_gem_object_create_lmem(engine->i915, context_size, I915_BO_ALLOC_PM_VOLATILE); - if (IS_ERR(obj)) + if (IS_ERR(obj)) { obj = i915_gem_object_create_shmem(engine->i915, context_size); + /* + * Wa_22016122933: For Media version 13.0, all Media GT shared + * memory needs to be mapped as WC on CPU side and UC (PAT + * index 2) on GPU side. + */ + if (intel_gt_needs_wa_22016122933(engine->gt)) + i915_gem_object_set_cache_coherency(obj, I915_CACHE_NONE); + } if (IS_ERR(obj)) return ERR_CAST(obj); @@ -1184,9 +1192,9 @@ lrc_pre_pin(struct intel_context *ce, GEM_BUG_ON(!i915_vma_is_pinned(ce->state)); *vaddr = i915_gem_object_pin_map(ce->state->obj, - i915_coherent_map_type(ce->engine->i915, - ce->state->obj, - false) | + intel_gt_coherent_map_type(ce->engine->gt, + ce->state->obj, + false) | I915_MAP_OVERRIDE); return PTR_ERR_OR_ZERO(*vaddr); @@ -1364,10 +1372,7 @@ gen12_emit_indirect_ctx_rcs(const struct intel_context *ce, u32 *cs) IS_DG2_G11(ce->engine->i915)) cs = gen8_emit_pipe_control(cs, PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE, 0); - /* hsdes: 1809175790 */ - if (!HAS_FLAT_CCS(ce->engine->i915)) - cs = gen12_emit_aux_table_inv(ce->engine->gt, - cs, GEN12_GFX_CCS_AUX_NV); + cs = gen12_emit_aux_table_inv(ce->engine, cs); /* Wa_16014892111 */ if (IS_MTL_GRAPHICS_STEP(ce->engine->i915, M, STEP_A0, STEP_B0) || @@ -1392,17 +1397,7 @@ gen12_emit_indirect_ctx_xcs(const struct intel_context *ce, u32 *cs) PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE, 0); - /* hsdes: 1809175790 */ - if (!HAS_FLAT_CCS(ce->engine->i915)) { - if (ce->engine->class == VIDEO_DECODE_CLASS) - cs = gen12_emit_aux_table_inv(ce->engine->gt, - cs, GEN12_VD0_AUX_NV); - else if (ce->engine->class == VIDEO_ENHANCEMENT_CLASS) - cs = gen12_emit_aux_table_inv(ce->engine->gt, - cs, GEN12_VE0_AUX_NV); - } - - return cs; + return gen12_emit_aux_table_inv(ce->engine, cs); } static void diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.c b/drivers/gpu/drm/i915/gt/intel_migrate.c index 6023288b0e2d..576e5ef0289b 100644 --- a/drivers/gpu/drm/i915/gt/intel_migrate.c +++ b/drivers/gpu/drm/i915/gt/intel_migrate.c @@ -366,7 +366,7 @@ static int emit_pte(struct i915_request *rq, u64 offset, int length) { - bool has_64K_pages = HAS_64K_PAGES(rq->engine->i915); + bool has_64K_pages = HAS_64K_PAGES(rq->i915); const u64 encode = rq->context->vm->pte_encode(0, pat_index, is_lmem ? PTE_LM : 0); struct intel_ring *ring = rq->ring; @@ -375,7 +375,7 @@ static int emit_pte(struct i915_request *rq, u32 page_size; u32 *hdr, *cs; - GEM_BUG_ON(GRAPHICS_VER(rq->engine->i915) < 8); + GEM_BUG_ON(GRAPHICS_VER(rq->i915) < 8); page_size = I915_GTT_PAGE_SIZE; dword_length = 0x400; @@ -531,7 +531,7 @@ static int emit_copy_ccs(struct i915_request *rq, u32 dst_offset, u8 dst_access, u32 src_offset, u8 src_access, int size) { - struct drm_i915_private *i915 = rq->engine->i915; + struct drm_i915_private *i915 = rq->i915; int mocs = rq->engine->gt->mocs.uc_index << 1; u32 num_ccs_blks; u32 *cs; @@ -581,7 +581,7 @@ static int emit_copy_ccs(struct i915_request *rq, static int emit_copy(struct i915_request *rq, u32 dst_offset, u32 src_offset, int size) { - const int ver = GRAPHICS_VER(rq->engine->i915); + const int ver = GRAPHICS_VER(rq->i915); u32 instance = rq->engine->instance; u32 *cs; @@ -917,7 +917,7 @@ out_ce: static int emit_clear(struct i915_request *rq, u32 offset, int size, u32 value, bool is_lmem) { - struct drm_i915_private *i915 = rq->engine->i915; + struct drm_i915_private *i915 = rq->i915; int mocs = rq->engine->gt->mocs.uc_index << 1; const int ver = GRAPHICS_VER(i915); int ring_sz; diff --git a/drivers/gpu/drm/i915/gt/intel_ppgtt.c b/drivers/gpu/drm/i915/gt/intel_ppgtt.c index 436756bfbb1a..d07a4f97b943 100644 --- a/drivers/gpu/drm/i915/gt/intel_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ppgtt.c @@ -8,6 +8,7 @@ #include "gem/i915_gem_lmem.h" #include "i915_trace.h" +#include "intel_gt.h" #include "intel_gtt.h" #include "gen6_ppgtt.h" #include "gen8_ppgtt.h" @@ -210,8 +211,7 @@ void ppgtt_unbind_vma(struct i915_address_space *vm, return; vm->clear_range(vm, vma_res->start, vma_res->vma_size); - if (vma_res->tlb) - vma_invalidate_tlb(vm, vma_res->tlb); + vma_invalidate_tlb(vm, vma_res->tlb); } static unsigned long pd_count(u64 size, int shift) diff --git a/drivers/gpu/drm/i915/gt/intel_region_lmem.c b/drivers/gpu/drm/i915/gt/intel_region_lmem.c index 2a3217e2890f..f8512aee58a8 100644 --- a/drivers/gpu/drm/i915/gt/intel_region_lmem.c +++ b/drivers/gpu/drm/i915/gt/intel_region_lmem.c @@ -220,7 +220,7 @@ static struct intel_memory_region *setup_lmem(struct intel_gt *gt) resource_size_t lmem_range; u64 tile_stolen, flat_ccs_base; - lmem_range = intel_gt_mcr_read_any(&i915->gt0, XEHP_TILE0_ADDR_RANGE) & 0xFFFF; + lmem_range = intel_gt_mcr_read_any(to_gt(i915), XEHP_TILE0_ADDR_RANGE) & 0xFFFF; lmem_size = lmem_range >> XEHP_TILE_LMEM_RANGE_SHIFT; lmem_size *= SZ_1G; diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c index e2152f75ba2e..cc6bd21a3e51 100644 --- a/drivers/gpu/drm/i915/gt/intel_reset.c +++ b/drivers/gpu/drm/i915/gt/intel_reset.c @@ -35,9 +35,6 @@ #define RESET_MAX_RETRIES 3 -/* XXX How to handle concurrent GGTT updates using tiling registers? */ -#define RESET_UNDER_STOP_MACHINE 0 - static void client_mark_guilty(struct i915_gem_context *ctx, bool banned) { struct drm_i915_file_private *file_priv = ctx->file_priv; diff --git a/drivers/gpu/drm/i915/gt/intel_ring.c b/drivers/gpu/drm/i915/gt/intel_ring.c index fb99143be98e..59da4b7bd262 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring.c +++ b/drivers/gpu/drm/i915/gt/intel_ring.c @@ -13,6 +13,7 @@ #include "intel_engine_regs.h" #include "intel_gpu_commands.h" #include "intel_ring.h" +#include "intel_gt.h" #include "intel_timeline.h" unsigned int intel_ring_update_space(struct intel_ring *ring) @@ -56,7 +57,7 @@ int intel_ring_pin(struct intel_ring *ring, struct i915_gem_ww_ctx *ww) if (i915_vma_is_map_and_fenceable(vma) && !HAS_LLC(vma->vm->i915)) { addr = (void __force *)i915_vma_pin_iomap(vma); } else { - int type = i915_coherent_map_type(vma->vm->i915, vma->obj, false); + int type = intel_gt_coherent_map_type(vma->vm->gt, vma->obj, false); addr = i915_gem_object_pin_map(vma->obj, type); } diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c index 3fd795c3263f..92085ffd23de 100644 --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c @@ -805,7 +805,7 @@ static int mi_set_context(struct i915_request *rq, static int remap_l3_slice(struct i915_request *rq, int slice) { #define L3LOG_DW (GEN7_L3LOG_SIZE / sizeof(u32)) - u32 *cs, *remap_info = rq->engine->i915->l3_parity.remap_info[slice]; + u32 *cs, *remap_info = rq->i915->l3_parity.remap_info[slice]; int i; if (!remap_info) diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index e92e626d4994..092542f53aad 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -16,7 +16,9 @@ #include "intel_gt.h" #include "intel_gt_clock_utils.h" #include "intel_gt_irq.h" +#include "intel_gt_pm.h" #include "intel_gt_pm_irq.h" +#include "intel_gt_print.h" #include "intel_gt_regs.h" #include "intel_mchbar_regs.h" #include "intel_pcode.h" @@ -672,7 +674,6 @@ static void rps_set_power(struct intel_rps *rps, int new_power) { struct intel_gt *gt = rps_to_gt(rps); struct intel_uncore *uncore = gt->uncore; - u32 threshold_up = 0, threshold_down = 0; /* in % */ u32 ei_up = 0, ei_down = 0; lockdep_assert_held(&rps->power.mutex); @@ -680,9 +681,6 @@ static void rps_set_power(struct intel_rps *rps, int new_power) if (new_power == rps->power.mode) return; - threshold_up = 95; - threshold_down = 85; - /* Note the units here are not exactly 1us, but 1280ns. */ switch (new_power) { case LOW_POWER: @@ -709,17 +707,22 @@ static void rps_set_power(struct intel_rps *rps, int new_power) GT_TRACE(gt, "changing power mode [%d], up %d%% @ %dus, down %d%% @ %dus\n", - new_power, threshold_up, ei_up, threshold_down, ei_down); + new_power, + rps->power.up_threshold, ei_up, + rps->power.down_threshold, ei_down); set(uncore, GEN6_RP_UP_EI, intel_gt_ns_to_pm_interval(gt, ei_up * 1000)); set(uncore, GEN6_RP_UP_THRESHOLD, - intel_gt_ns_to_pm_interval(gt, ei_up * threshold_up * 10)); + intel_gt_ns_to_pm_interval(gt, + ei_up * rps->power.up_threshold * 10)); set(uncore, GEN6_RP_DOWN_EI, intel_gt_ns_to_pm_interval(gt, ei_down * 1000)); set(uncore, GEN6_RP_DOWN_THRESHOLD, - intel_gt_ns_to_pm_interval(gt, ei_down * threshold_down * 10)); + intel_gt_ns_to_pm_interval(gt, + ei_down * + rps->power.down_threshold * 10)); set(uncore, GEN6_RP_CONTROL, (GRAPHICS_VER(gt->i915) > 9 ? 0 : GEN6_RP_MEDIA_TURBO) | @@ -731,8 +734,6 @@ static void rps_set_power(struct intel_rps *rps, int new_power) skip_hw_write: rps->power.mode = new_power; - rps->power.up_threshold = threshold_up; - rps->power.down_threshold = threshold_down; } static void gen6_rps_set_thresholds(struct intel_rps *rps, u8 val) @@ -1559,10 +1560,12 @@ void intel_rps_enable(struct intel_rps *rps) return; GT_TRACE(rps_to_gt(rps), - "min:%x, max:%x, freq:[%d, %d]\n", + "min:%x, max:%x, freq:[%d, %d], thresholds:[%u, %u]\n", rps->min_freq, rps->max_freq, intel_gpu_freq(rps, rps->min_freq), - intel_gpu_freq(rps, rps->max_freq)); + intel_gpu_freq(rps, rps->max_freq), + rps->power.up_threshold, + rps->power.down_threshold); GEM_BUG_ON(rps->max_freq < rps->min_freq); GEM_BUG_ON(rps->idle_freq > rps->max_freq); @@ -2015,6 +2018,12 @@ void intel_rps_init(struct intel_rps *rps) } } + /* Set default thresholds in % */ + rps->power.up_threshold = 95; + rps_to_gt(rps)->defaults.rps_up_threshold = rps->power.up_threshold; + rps->power.down_threshold = 85; + rps_to_gt(rps)->defaults.rps_down_threshold = rps->power.down_threshold; + /* Finally allow us to boost to max by default */ rps->boost_freq = rps->max_freq; rps->idle_freq = rps->min_freq; @@ -2569,6 +2578,58 @@ int intel_rps_set_min_frequency(struct intel_rps *rps, u32 val) return set_min_freq(rps, val); } +u8 intel_rps_get_up_threshold(struct intel_rps *rps) +{ + return rps->power.up_threshold; +} + +static int rps_set_threshold(struct intel_rps *rps, u8 *threshold, u8 val) +{ + int ret; + + if (val > 100) + return -EINVAL; + + ret = mutex_lock_interruptible(&rps->lock); + if (ret) + return ret; + + if (*threshold == val) + goto out_unlock; + + *threshold = val; + + /* Force reset. */ + rps->last_freq = -1; + mutex_lock(&rps->power.mutex); + rps->power.mode = -1; + mutex_unlock(&rps->power.mutex); + + intel_rps_set(rps, clamp(rps->cur_freq, + rps->min_freq_softlimit, + rps->max_freq_softlimit)); + +out_unlock: + mutex_unlock(&rps->lock); + + return ret; +} + +int intel_rps_set_up_threshold(struct intel_rps *rps, u8 threshold) +{ + return rps_set_threshold(rps, &rps->power.up_threshold, threshold); +} + +u8 intel_rps_get_down_threshold(struct intel_rps *rps) +{ + return rps->power.down_threshold; +} + +int intel_rps_set_down_threshold(struct intel_rps *rps, u8 threshold) +{ + return rps_set_threshold(rps, &rps->power.down_threshold, threshold); +} + static void intel_rps_set_manual(struct intel_rps *rps, bool enable) { struct intel_uncore *uncore = rps_to_uncore(rps); diff --git a/drivers/gpu/drm/i915/gt/intel_rps.h b/drivers/gpu/drm/i915/gt/intel_rps.h index a3fa987aa91f..92fb01f5a452 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.h +++ b/drivers/gpu/drm/i915/gt/intel_rps.h @@ -37,6 +37,10 @@ void intel_rps_mark_interactive(struct intel_rps *rps, bool interactive); int intel_gpu_freq(struct intel_rps *rps, int val); int intel_freq_opcode(struct intel_rps *rps, int val); +u8 intel_rps_get_up_threshold(struct intel_rps *rps); +int intel_rps_set_up_threshold(struct intel_rps *rps, u8 threshold); +u8 intel_rps_get_down_threshold(struct intel_rps *rps); +int intel_rps_set_down_threshold(struct intel_rps *rps, u8 threshold); u32 intel_rps_read_actual_frequency(struct intel_rps *rps); u32 intel_rps_read_actual_frequency_fw(struct intel_rps *rps); u32 intel_rps_get_requested_frequency(struct intel_rps *rps); diff --git a/drivers/gpu/drm/i915/gt/intel_sa_media.c b/drivers/gpu/drm/i915/gt/intel_sa_media.c index e8f3d18c12b8..8c1dbcbcbc4f 100644 --- a/drivers/gpu/drm/i915/gt/intel_sa_media.c +++ b/drivers/gpu/drm/i915/gt/intel_sa_media.c @@ -29,7 +29,7 @@ int intel_sa_mediagt_setup(struct intel_gt *gt, phys_addr_t phys_addr, * Standalone media shares the general MMIO space with the primary * GT. We'll re-use the primary GT's mapping. */ - uncore->regs = i915->uncore.regs; + uncore->regs = intel_uncore_regs(&i915->uncore); if (drm_WARN_ON(&i915->drm, uncore->regs == NULL)) return -EIO; diff --git a/drivers/gpu/drm/i915/gt/intel_sseu.c b/drivers/gpu/drm/i915/gt/intel_sseu.c index 1141f875f5bd..f602895f6d0d 100644 --- a/drivers/gpu/drm/i915/gt/intel_sseu.c +++ b/drivers/gpu/drm/i915/gt/intel_sseu.c @@ -302,7 +302,7 @@ static void gen11_sseu_info_init(struct intel_gt *gt) u8 eu_en; u8 s_en; - if (IS_JSL_EHL(gt->i915)) + if (IS_JASPERLAKE(gt->i915) || IS_ELKHARTLAKE(gt->i915)) intel_sseu_set_info(sseu, 1, 4, 8); else intel_sseu_set_info(sseu, 1, 8, 8); diff --git a/drivers/gpu/drm/i915/gt/intel_tlb.c b/drivers/gpu/drm/i915/gt/intel_tlb.c new file mode 100644 index 000000000000..139608c30d97 --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_tlb.c @@ -0,0 +1,159 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include "i915_drv.h" +#include "i915_perf_oa_regs.h" +#include "intel_engine_pm.h" +#include "intel_gt.h" +#include "intel_gt_mcr.h" +#include "intel_gt_pm.h" +#include "intel_gt_print.h" +#include "intel_gt_regs.h" +#include "intel_tlb.h" + +/* + * HW architecture suggest typical invalidation time at 40us, + * with pessimistic cases up to 100us and a recommendation to + * cap at 1ms. We go a bit higher just in case. + */ +#define TLB_INVAL_TIMEOUT_US 100 +#define TLB_INVAL_TIMEOUT_MS 4 + +/* + * On Xe_HP the TLB invalidation registers are located at the same MMIO offsets + * but are now considered MCR registers. Since they exist within a GAM range, + * the primary instance of the register rolls up the status from each unit. + */ +static int wait_for_invalidate(struct intel_engine_cs *engine) +{ + if (engine->tlb_inv.mcr) + return intel_gt_mcr_wait_for_reg(engine->gt, + engine->tlb_inv.reg.mcr_reg, + engine->tlb_inv.done, + 0, + TLB_INVAL_TIMEOUT_US, + TLB_INVAL_TIMEOUT_MS); + else + return __intel_wait_for_register_fw(engine->gt->uncore, + engine->tlb_inv.reg.reg, + engine->tlb_inv.done, + 0, + TLB_INVAL_TIMEOUT_US, + TLB_INVAL_TIMEOUT_MS, + NULL); +} + +static void mmio_invalidate_full(struct intel_gt *gt) +{ + struct drm_i915_private *i915 = gt->i915; + struct intel_uncore *uncore = gt->uncore; + struct intel_engine_cs *engine; + intel_engine_mask_t awake, tmp; + enum intel_engine_id id; + unsigned long flags; + + if (GRAPHICS_VER(i915) < 8) + return; + + intel_uncore_forcewake_get(uncore, FORCEWAKE_ALL); + + intel_gt_mcr_lock(gt, &flags); + spin_lock(&uncore->lock); /* serialise invalidate with GT reset */ + + awake = 0; + for_each_engine(engine, gt, id) { + if (!intel_engine_pm_is_awake(engine)) + continue; + + if (engine->tlb_inv.mcr) + intel_gt_mcr_multicast_write_fw(gt, + engine->tlb_inv.reg.mcr_reg, + engine->tlb_inv.request); + else + intel_uncore_write_fw(uncore, + engine->tlb_inv.reg.reg, + engine->tlb_inv.request); + + awake |= engine->mask; + } + + GT_TRACE(gt, "invalidated engines %08x\n", awake); + + /* Wa_2207587034:tgl,dg1,rkl,adl-s,adl-p */ + if (awake && + (IS_TIGERLAKE(i915) || + IS_DG1(i915) || + IS_ROCKETLAKE(i915) || + IS_ALDERLAKE_S(i915) || + IS_ALDERLAKE_P(i915))) + intel_uncore_write_fw(uncore, GEN12_OA_TLB_INV_CR, 1); + + spin_unlock(&uncore->lock); + intel_gt_mcr_unlock(gt, flags); + + for_each_engine_masked(engine, gt, awake, tmp) { + if (wait_for_invalidate(engine)) + gt_err_ratelimited(gt, + "%s TLB invalidation did not complete in %ums!\n", + engine->name, TLB_INVAL_TIMEOUT_MS); + } + + /* + * Use delayed put since a) we mostly expect a flurry of TLB + * invalidations so it is good to avoid paying the forcewake cost and + * b) it works around a bug in Icelake which cannot cope with too rapid + * transitions. + */ + intel_uncore_forcewake_put_delayed(uncore, FORCEWAKE_ALL); +} + +static bool tlb_seqno_passed(const struct intel_gt *gt, u32 seqno) +{ + u32 cur = intel_gt_tlb_seqno(gt); + + /* Only skip if a *full* TLB invalidate barrier has passed */ + return (s32)(cur - ALIGN(seqno, 2)) > 0; +} + +void intel_gt_invalidate_tlb_full(struct intel_gt *gt, u32 seqno) +{ + intel_wakeref_t wakeref; + + if (I915_SELFTEST_ONLY(gt->awake == -ENODEV)) + return; + + if (intel_gt_is_wedged(gt)) + return; + + if (tlb_seqno_passed(gt, seqno)) + return; + + with_intel_gt_pm_if_awake(gt, wakeref) { + mutex_lock(>->tlb.invalidate_lock); + if (tlb_seqno_passed(gt, seqno)) + goto unlock; + + mmio_invalidate_full(gt); + + write_seqcount_invalidate(>->tlb.seqno); +unlock: + mutex_unlock(>->tlb.invalidate_lock); + } +} + +void intel_gt_init_tlb(struct intel_gt *gt) +{ + mutex_init(>->tlb.invalidate_lock); + seqcount_mutex_init(>->tlb.seqno, >->tlb.invalidate_lock); +} + +void intel_gt_fini_tlb(struct intel_gt *gt) +{ + mutex_destroy(>->tlb.invalidate_lock); +} + +#if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) +#include "selftest_tlb.c" +#endif diff --git a/drivers/gpu/drm/i915/gt/intel_tlb.h b/drivers/gpu/drm/i915/gt/intel_tlb.h new file mode 100644 index 000000000000..337327af92ac --- /dev/null +++ b/drivers/gpu/drm/i915/gt/intel_tlb.h @@ -0,0 +1,29 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef INTEL_TLB_H +#define INTEL_TLB_H + +#include <linux/seqlock.h> +#include <linux/types.h> + +#include "intel_gt_types.h" + +void intel_gt_invalidate_tlb_full(struct intel_gt *gt, u32 seqno); + +void intel_gt_init_tlb(struct intel_gt *gt); +void intel_gt_fini_tlb(struct intel_gt *gt); + +static inline u32 intel_gt_tlb_seqno(const struct intel_gt *gt) +{ + return seqprop_sequence(>->tlb.seqno); +} + +static inline u32 intel_gt_next_invalidate_tlb_full(const struct intel_gt *gt) +{ + return intel_gt_tlb_seqno(gt) | 1; +} + +#endif /* INTEL_TLB_H */ diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 4d2dece96011..3ae0dbd39eaa 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -123,6 +123,22 @@ static void wa_init_finish(struct i915_wa_list *wal) wal->wa_count, wal->name, wal->engine_name); } +static enum forcewake_domains +wal_get_fw_for_rmw(struct intel_uncore *uncore, const struct i915_wa_list *wal) +{ + enum forcewake_domains fw = 0; + struct i915_wa *wa; + unsigned int i; + + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) + fw |= intel_uncore_forcewake_for_reg(uncore, + wa->reg, + FW_REG_READ | + FW_REG_WRITE); + + return fw; +} + static void _wa_add(struct i915_wa_list *wal, const struct i915_wa *wa) { unsigned int addr = i915_mmio_reg_offset(wa->reg); @@ -225,13 +241,13 @@ static void wa_mcr_add(struct i915_wa_list *wal, i915_mcr_reg_t reg, static void wa_write_clr_set(struct i915_wa_list *wal, i915_reg_t reg, u32 clear, u32 set) { - wa_add(wal, reg, clear, set, clear, false); + wa_add(wal, reg, clear, set, clear | set, false); } static void wa_mcr_write_clr_set(struct i915_wa_list *wal, i915_mcr_reg_t reg, u32 clear, u32 set) { - wa_mcr_add(wal, reg, clear, set, clear, false); + wa_mcr_add(wal, reg, clear, set, clear | set, false); } static void @@ -404,7 +420,7 @@ static void bdw_ctx_workarounds_init(struct intel_engine_cs *engine, /* WaForceContextSaveRestoreNonCoherent:bdw */ HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT | /* WaDisableFenceDestinationToSLM:bdw (pre-prod) */ - (IS_BDW_GT3(i915) ? HDC_FENCE_DEST_SLM_DISABLE : 0)); + (IS_BROADWELL_GT3(i915) ? HDC_FENCE_DEST_SLM_DISABLE : 0)); } static void chv_ctx_workarounds_init(struct intel_engine_cs *engine, @@ -584,7 +600,7 @@ static void kbl_ctx_workarounds_init(struct intel_engine_cs *engine, gen9_ctx_workarounds_init(engine, wal); /* WaToEnableHwFixForPushConstHWBug:kbl */ - if (IS_KBL_GRAPHICS_STEP(i915, STEP_C0, STEP_FOREVER)) + if (IS_KABYLAKE(i915) && IS_GRAPHICS_STEP(i915, STEP_C0, STEP_FOREVER)) wa_masked_en(wal, COMMON_SLICE_CHICKEN2, GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION); @@ -621,10 +637,7 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) { /* Wa_1406697149 (WaDisableBankHangMode:icl) */ - wa_write(wal, - GEN8_L3CNTLREG, - intel_uncore_read(engine->uncore, GEN8_L3CNTLREG) | - GEN8_ERRDETBCTRL); + wa_write(wal, GEN8_L3CNTLREG, GEN8_ERRDETBCTRL); /* WaForceEnableNonCoherent:icl * This is not the same workaround as in early Gen9 platforms, where @@ -653,7 +666,7 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine, /* Wa_1604278689:icl,ehl */ wa_write(wal, IVB_FBC_RT_BASE, 0xFFFFFFFF & ~ILK_FBC_RT_VALID); wa_write_clr_set(wal, IVB_FBC_RT_BASE_UPPER, - 0, /* write-only register; skip validation */ + 0, 0xFFFFFFFF); /* Wa_1406306137:icl,ehl */ @@ -670,38 +683,8 @@ static void dg2_ctx_gt_tuning_init(struct intel_engine_cs *engine, wa_mcr_masked_en(wal, CHICKEN_RASTER_2, TBIMR_FAST_CLIP); wa_mcr_write_clr_set(wal, XEHP_L3SQCREG5, L3_PWM_TIMER_INIT_VAL_MASK, REG_FIELD_PREP(L3_PWM_TIMER_INIT_VAL_MASK, 0x7f)); - wa_mcr_add(wal, - XEHP_FF_MODE2, - FF_MODE2_TDS_TIMER_MASK, - FF_MODE2_TDS_TIMER_128, - 0, false); -} - -/* - * These settings aren't actually workarounds, but general tuning settings that - * need to be programmed on several platforms. - */ -static void gen12_ctx_gt_tuning_init(struct intel_engine_cs *engine, - struct i915_wa_list *wal) -{ - /* - * Although some platforms refer to it as Wa_1604555607, we need to - * program it even on those that don't explicitly list that - * workaround. - * - * Note that the programming of this register is further modified - * according to the FF_MODE2 guidance given by Wa_1608008084:gen12. - * Wa_1608008084 tells us the FF_MODE2 register will return the wrong - * value when read. The default value for this register is zero for all - * fields and there are no bit masks. So instead of doing a RMW we - * should just write TDS timer value. For the same reason read - * verification is ignored. - */ - wa_add(wal, - GEN12_FF_MODE2, - FF_MODE2_TDS_TIMER_MASK, - FF_MODE2_TDS_TIMER_128, - 0, false); + wa_mcr_write_clr_set(wal, XEHP_FF_MODE2, FF_MODE2_TDS_TIMER_MASK, + FF_MODE2_TDS_TIMER_128); } static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine, @@ -709,8 +692,6 @@ static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine, { struct drm_i915_private *i915 = engine->i915; - gen12_ctx_gt_tuning_init(engine, wal); - /* * Wa_1409142259:tgl,dg1,adl-p * Wa_1409347922:tgl,dg1,adl-p @@ -732,15 +713,27 @@ static void gen12_ctx_workarounds_init(struct intel_engine_cs *engine, GEN9_PREEMPT_GPGPU_THREAD_GROUP_LEVEL); /* - * Wa_16011163337 + * Wa_16011163337 - GS_TIMER + * + * TDS_TIMER: Although some platforms refer to it as Wa_1604555607, we + * need to program it even on those that don't explicitly list that + * workaround. + * + * Note that the programming of GEN12_FF_MODE2 is further modified + * according to the FF_MODE2 guidance given by Wa_1608008084. + * Wa_1608008084 tells us the FF_MODE2 register will return the wrong + * value when read from the CPU. * - * Like in gen12_ctx_gt_tuning_init(), read verification is ignored due - * to Wa_1608008084. + * The default value for this register is zero for all fields. + * So instead of doing a RMW we should just write the desired values + * for TDS and GS timers. Note that since the readback can't be trusted, + * the clear mask is just set to ~0 to make sure other bits are not + * inadvertently set. For the same reason read verification is ignored. */ wa_add(wal, GEN12_FF_MODE2, - FF_MODE2_GS_TIMER_MASK, - FF_MODE2_GS_TIMER_224, + ~0, + FF_MODE2_TDS_TIMER_128 | FF_MODE2_GS_TIMER_224, 0, false); if (!IS_DG1(i915)) { @@ -987,6 +980,9 @@ void intel_engine_init_ctx_wa(struct intel_engine_cs *engine) int intel_engine_emit_ctx_wa(struct i915_request *rq) { struct i915_wa_list *wal = &rq->engine->ctx_wa_list; + struct intel_uncore *uncore = rq->engine->uncore; + enum forcewake_domains fw; + unsigned long flags; struct i915_wa *wa; unsigned int i; u32 *cs; @@ -1003,13 +999,36 @@ int intel_engine_emit_ctx_wa(struct i915_request *rq) if (IS_ERR(cs)) return PTR_ERR(cs); + fw = wal_get_fw_for_rmw(uncore, wal); + + intel_gt_mcr_lock(wal->gt, &flags); + spin_lock(&uncore->lock); + intel_uncore_forcewake_get__locked(uncore, fw); + *cs++ = MI_LOAD_REGISTER_IMM(wal->count); for (i = 0, wa = wal->list; i < wal->count; i++, wa++) { + u32 val; + + /* Skip reading the register if it's not really needed */ + if (wa->masked_reg || (wa->clr | wa->set) == U32_MAX) { + val = wa->set; + } else { + val = wa->is_mcr ? + intel_gt_mcr_read_any_fw(wal->gt, wa->mcr_reg) : + intel_uncore_read_fw(uncore, wa->reg); + val &= ~wa->clr; + val |= wa->set; + } + *cs++ = i915_mmio_reg_offset(wa->reg); - *cs++ = wa->set; + *cs++ = val; } *cs++ = MI_NOOP; + intel_uncore_forcewake_put__locked(uncore, fw); + spin_unlock(&uncore->lock); + intel_gt_mcr_unlock(wal->gt, flags); + intel_ring_advance(rq, cs); ret = rq->engine->emit_flush(rq, EMIT_BARRIER); @@ -1173,7 +1192,7 @@ skl_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:skl */ - if (IS_SKL_GRAPHICS_STEP(gt->i915, STEP_A0, STEP_H0)) + if (IS_SKYLAKE(gt->i915) && IS_GRAPHICS_STEP(gt->i915, STEP_A0, STEP_H0)) wa_write_or(wal, GEN9_GAMT_ECO_REG_RW_IA, GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); @@ -1185,7 +1204,7 @@ kbl_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) gen9_gt_workarounds_init(gt, wal); /* WaDisableDynamicCreditSharing:kbl */ - if (IS_KBL_GRAPHICS_STEP(gt->i915, 0, STEP_C0)) + if (IS_KABYLAKE(gt->i915) && IS_GRAPHICS_STEP(gt->i915, 0, STEP_C0)) wa_write_or(wal, GAMT_CHKN_BIT_REG, GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); @@ -1441,7 +1460,8 @@ icl_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) /* Wa_1607087056:icl,ehl,jsl */ if (IS_ICELAKE(i915) || - IS_JSL_EHL_GRAPHICS_STEP(i915, STEP_A0, STEP_B0)) + ((IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) && + IS_GRAPHICS_STEP(i915, STEP_A0, STEP_B0))) wa_write_or(wal, GEN11_SLICE_UNIT_LEVEL_CLKGATE, L3_CLKGATE_DIS | L3_CR2X_CLKGATE_DIS); @@ -1485,6 +1505,18 @@ gen12_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) /* Wa_14011059788:tgl,rkl,adl-s,dg1,adl-p */ wa_mcr_write_or(wal, GEN10_DFR_RATIO_EN_AND_CHICKEN, DFR_DISABLE); + + /* + * Wa_14015795083 + * + * Firmware on some gen12 platforms locks the MISCCPCTL register, + * preventing i915 from modifying it for this workaround. Skip the + * readback verification for this workaround on debug builds; if the + * workaround doesn't stick due to firmware behavior, it's not an error + * that we want CI to flag. + */ + wa_add(wal, GEN7_MISCCPCTL, GEN12_DOP_CLOCK_GATE_RENDER_ENABLE, + 0, 0, false); } static void @@ -1710,7 +1742,6 @@ static void xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) { /* Wa_14018778641 / Wa_18018781329 */ - wa_mcr_write_or(wal, RENDER_MOD_CTRL, FORCE_MISS_FTLB); wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB); /* Wa_22016670082 */ @@ -1743,8 +1774,6 @@ xelpmp_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) * GT, the media GT's versions are regular singleton registers. */ wa_write_or(wal, XELPMP_GSC_MOD_CTRL, FORCE_MISS_FTLB); - wa_write_or(wal, XELPMP_VDBX_MOD_CTRL, FORCE_MISS_FTLB); - wa_write_or(wal, XELPMP_VEBX_MOD_CTRL, FORCE_MISS_FTLB); debug_dump_steering(gt); } @@ -1850,22 +1879,6 @@ void intel_gt_init_workarounds(struct intel_gt *gt) wa_init_finish(wal); } -static enum forcewake_domains -wal_get_fw_for_rmw(struct intel_uncore *uncore, const struct i915_wa_list *wal) -{ - enum forcewake_domains fw = 0; - struct i915_wa *wa; - unsigned int i; - - for (i = 0, wa = wal->list; i < wal->count; i++, wa++) - fw |= intel_uncore_forcewake_for_reg(uncore, - wa->reg, - FW_REG_READ | - FW_REG_WRITE); - - return fw; -} - static bool wa_verify(struct intel_gt *gt, const struct i915_wa *wa, u32 cur, const char *name, const char *from) @@ -2933,7 +2946,7 @@ xcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal) struct drm_i915_private *i915 = engine->i915; /* WaKBLVECSSemaphoreWaitPoll:kbl */ - if (IS_KBL_GRAPHICS_STEP(i915, STEP_A0, STEP_F0)) { + if (IS_KABYLAKE(i915) && IS_GRAPHICS_STEP(i915, STEP_A0, STEP_F0)) { wa_write(wal, RING_SEMA_WAIT_POLL(engine->mmio_base), 1); @@ -3237,7 +3250,7 @@ wa_list_srm(struct i915_request *rq, const struct i915_wa_list *wal, struct i915_vma *vma) { - struct drm_i915_private *i915 = rq->engine->i915; + struct drm_i915_private *i915 = rq->i915; unsigned int i, count = 0; const struct i915_wa *wa; u32 srm, *cs; @@ -3336,7 +3349,7 @@ retry: err = 0; for (i = 0, wa = wal->list; i < wal->count; i++, wa++) { - if (mcr_range(rq->engine->i915, i915_mmio_reg_offset(wa->reg))) + if (mcr_range(rq->i915, i915_mmio_reg_offset(wa->reg))) continue; if (!wa_verify(wal->gt, wa, results[i], wal->name, from)) diff --git a/drivers/gpu/drm/i915/gt/selftest_context.c b/drivers/gpu/drm/i915/gt/selftest_context.c index 76fbae358072..47070cba7eb1 100644 --- a/drivers/gpu/drm/i915/gt/selftest_context.c +++ b/drivers/gpu/drm/i915/gt/selftest_context.c @@ -88,8 +88,9 @@ static int __live_context_size(struct intel_engine_cs *engine) goto err; vaddr = i915_gem_object_pin_map_unlocked(ce->state->obj, - i915_coherent_map_type(engine->i915, - ce->state->obj, false)); + intel_gt_coherent_map_type(engine->gt, + ce->state->obj, + false)); if (IS_ERR(vaddr)) { err = PTR_ERR(vaddr); intel_context_unpin(ce); diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c index 78cdfc6f315f..86cecf7a1105 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c @@ -62,7 +62,7 @@ static int write_timestamp(struct i915_request *rq, int slot) return PTR_ERR(cs); cmd = MI_STORE_REGISTER_MEM | MI_USE_GGTT; - if (GRAPHICS_VER(rq->engine->i915) >= 8) + if (GRAPHICS_VER(rq->i915) >= 8) cmd++; *cs++ = cmd; *cs++ = i915_mmio_reg_offset(timestamp_reg(rq->engine)); diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c index 8b0d84f2aad2..0dd4d00ee894 100644 --- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c +++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c @@ -73,7 +73,7 @@ static int hang_init(struct hang *h, struct intel_gt *gt) h->seqno = memset(vaddr, 0xff, PAGE_SIZE); vaddr = i915_gem_object_pin_map_unlocked(h->obj, - i915_coherent_map_type(gt->i915, h->obj, false)); + intel_gt_coherent_map_type(gt, h->obj, false)); if (IS_ERR(vaddr)) { err = PTR_ERR(vaddr); goto err_unpin_hws; @@ -119,7 +119,7 @@ hang_create_request(struct hang *h, struct intel_engine_cs *engine) return ERR_CAST(obj); } - vaddr = i915_gem_object_pin_map_unlocked(obj, i915_coherent_map_type(gt->i915, obj, false)); + vaddr = i915_gem_object_pin_map_unlocked(obj, intel_gt_coherent_map_type(gt, obj, false)); if (IS_ERR(vaddr)) { i915_gem_object_put(obj); i915_vm_put(vm); diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c index a78a3d2c2e16..5f826b6dcf5d 100644 --- a/drivers/gpu/drm/i915/gt/selftest_lrc.c +++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c @@ -1292,9 +1292,9 @@ static int compare_isolation(struct intel_engine_cs *engine, } lrc = i915_gem_object_pin_map_unlocked(ce->state->obj, - i915_coherent_map_type(engine->i915, - ce->state->obj, - false)); + intel_gt_coherent_map_type(engine->gt, + ce->state->obj, + false)); if (IS_ERR(lrc)) { err = PTR_ERR(lrc); goto err_B1; diff --git a/drivers/gpu/drm/i915/gt/selftest_mocs.c b/drivers/gpu/drm/i915/gt/selftest_mocs.c index a8446ab82501..d73e438fb85f 100644 --- a/drivers/gpu/drm/i915/gt/selftest_mocs.c +++ b/drivers/gpu/drm/i915/gt/selftest_mocs.c @@ -137,7 +137,7 @@ static int read_mocs_table(struct i915_request *rq, if (!table) return 0; - if (HAS_GLOBAL_MOCS_REGISTERS(rq->engine->i915)) + if (HAS_GLOBAL_MOCS_REGISTERS(rq->i915)) addr = global_mocs_offset() + gt->uncore->gsi_offset; else addr = mocs_offset(rq->engine); diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c index 2ceeadecc639..a7189c2d660c 100644 --- a/drivers/gpu/drm/i915/gt/selftest_rc6.c +++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c @@ -140,7 +140,7 @@ static const u32 *__live_rc6_ctx(struct intel_context *ce) } cmd = MI_STORE_REGISTER_MEM | MI_USE_GGTT; - if (GRAPHICS_VER(rq->engine->i915) >= 8) + if (GRAPHICS_VER(rq->i915) >= 8) cmd++; *cs++ = cmd; diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c index 39c3ec12df1a..fa36cf920bde 100644 --- a/drivers/gpu/drm/i915/gt/selftest_timeline.c +++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c @@ -459,12 +459,12 @@ static int emit_ggtt_store_dw(struct i915_request *rq, u32 addr, u32 value) if (IS_ERR(cs)) return PTR_ERR(cs); - if (GRAPHICS_VER(rq->engine->i915) >= 8) { + if (GRAPHICS_VER(rq->i915) >= 8) { *cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; *cs++ = addr; *cs++ = 0; *cs++ = value; - } else if (GRAPHICS_VER(rq->engine->i915) >= 4) { + } else if (GRAPHICS_VER(rq->i915) >= 4) { *cs++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; *cs++ = 0; *cs++ = addr; diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c index 3bd6b540257b..7e41f69fc818 100644 --- a/drivers/gpu/drm/i915/gt/selftest_tlb.c +++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c @@ -6,6 +6,7 @@ #include "i915_selftest.h" #include "gem/i915_gem_internal.h" +#include "gem/i915_gem_lmem.h" #include "gem/i915_gem_region.h" #include "gen8_engine_cs.h" @@ -354,7 +355,7 @@ out_a: static void tlbinv_full(struct i915_address_space *vm, u64 addr, u64 length) { - intel_gt_invalidate_tlb(vm->gt, intel_gt_tlb_seqno(vm->gt) | 1); + intel_gt_invalidate_tlb_full(vm->gt, intel_gt_tlb_seqno(vm->gt) | 1); } static int invalidate_full(void *arg) diff --git a/drivers/gpu/drm/i915/gt/shmem_utils.c b/drivers/gpu/drm/i915/gt/shmem_utils.c index 449c9ed44382..bccc3a1200bc 100644 --- a/drivers/gpu/drm/i915/gt/shmem_utils.c +++ b/drivers/gpu/drm/i915/gt/shmem_utils.c @@ -33,7 +33,6 @@ struct file *shmem_create_from_data(const char *name, void *data, size_t len) struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) { - struct drm_i915_private *i915 = to_i915(obj->base.dev); enum i915_map_type map_type; struct file *file; void *ptr; @@ -44,7 +43,7 @@ struct file *shmem_create_from_object(struct drm_i915_gem_object *obj) return file; } - map_type = i915_coherent_map_type(i915, obj, true); + map_type = i915_gem_object_is_lmem(obj) ? I915_MAP_WC : I915_MAP_WB; ptr = i915_gem_object_pin_map_unlocked(obj, map_type); if (IS_ERR(ptr)) return ERR_CAST(ptr); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h index 714f0c256118..6d009a905269 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_binary_headers.h @@ -8,6 +8,74 @@ #include <linux/types.h> +struct intel_gsc_version { + u16 major; + u16 minor; + u16 hotfix; + u16 build; +} __packed; + +struct intel_gsc_partition { + u32 offset; + u32 size; +} __packed; + +struct intel_gsc_layout_pointers { + u8 rom_bypass_vector[16]; + + /* size of pointers layout not including ROM bypass vector */ + u16 size; + + /* + * bit0: Backup copy of layout pointers exist + * bits1-15: reserved + */ + u8 flags; + + u8 reserved; + + u32 crc32; + + struct intel_gsc_partition datap; + struct intel_gsc_partition boot1; + struct intel_gsc_partition boot2; + struct intel_gsc_partition boot3; + struct intel_gsc_partition boot4; + struct intel_gsc_partition boot5; + struct intel_gsc_partition temp_pages; +} __packed; + +/* Boot partition structures */ +struct intel_gsc_bpdt_header { + u32 signature; +#define INTEL_GSC_BPDT_HEADER_SIGNATURE 0x000055AA + + u16 descriptor_count; /* num of entries after the header */ + + u8 version; + u8 configuration; + + u32 crc32; + + u32 build_version; + struct intel_gsc_version tool_version; +} __packed; + +struct intel_gsc_bpdt_entry { + /* + * Bits 0-15: BPDT entry type + * Bits 16-17: reserved + * Bit 18: code sub-partition + * Bits 19-31: reserved + */ + u32 type; +#define INTEL_GSC_BPDT_ENTRY_TYPE_MASK GENMASK(15, 0) +#define INTEL_GSC_BPDT_ENTRY_TYPE_GSC_RBE 0x1 + + u32 sub_partition_offset; /* from the base of the BPDT header */ + u32 sub_partition_size; +} __packed; + /* Code partition directory (CPD) structures */ struct intel_gsc_cpd_header_v2 { u32 header_marker; @@ -44,13 +112,6 @@ struct intel_gsc_cpd_entry { u8 reserved[4]; } __packed; -struct intel_gsc_version { - u16 major; - u16 minor; - u16 hotfix; - u16 build; -} __packed; - struct intel_gsc_manifest_header { u32 header_type; /* 0x4 for manifest type */ u32 header_length; /* in dwords */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c index 60e9c6c9e775..e2e42b3e0d5d 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.c @@ -3,48 +3,216 @@ * Copyright © 2022 Intel Corporation */ +#include "gem/i915_gem_lmem.h" #include "gt/intel_engine_pm.h" #include "gt/intel_gpu_commands.h" #include "gt/intel_gt.h" #include "gt/intel_gt_print.h" #include "gt/intel_ring.h" +#include "intel_gsc_binary_headers.h" #include "intel_gsc_fw.h" - -#define GSC_FW_STATUS_REG _MMIO(0x116C40) -#define GSC_FW_CURRENT_STATE REG_GENMASK(3, 0) -#define GSC_FW_CURRENT_STATE_RESET 0 -#define GSC_FW_PROXY_STATE_NORMAL 5 -#define GSC_FW_INIT_COMPLETE_BIT REG_BIT(9) +#include "intel_gsc_uc_heci_cmd_submit.h" +#include "i915_reg.h" static bool gsc_is_in_reset(struct intel_uncore *uncore) { - u32 fw_status = intel_uncore_read(uncore, GSC_FW_STATUS_REG); + u32 fw_status = intel_uncore_read(uncore, HECI_FWSTS(MTL_GSC_HECI1_BASE, 1)); - return REG_FIELD_GET(GSC_FW_CURRENT_STATE, fw_status) == - GSC_FW_CURRENT_STATE_RESET; + return REG_FIELD_GET(HECI1_FWSTS1_CURRENT_STATE, fw_status) == + HECI1_FWSTS1_CURRENT_STATE_RESET; } -static u32 gsc_uc_get_fw_status(struct intel_uncore *uncore) +static u32 gsc_uc_get_fw_status(struct intel_uncore *uncore, bool needs_wakeref) { intel_wakeref_t wakeref; u32 fw_status = 0; - with_intel_runtime_pm(uncore->rpm, wakeref) - fw_status = intel_uncore_read(uncore, GSC_FW_STATUS_REG); + if (needs_wakeref) + wakeref = intel_runtime_pm_get(uncore->rpm); + + fw_status = intel_uncore_read(uncore, HECI_FWSTS(MTL_GSC_HECI1_BASE, 1)); + if (needs_wakeref) + intel_runtime_pm_put(uncore->rpm, wakeref); return fw_status; } -bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc) +bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc, bool needs_wakeref) +{ + return REG_FIELD_GET(HECI1_FWSTS1_CURRENT_STATE, + gsc_uc_get_fw_status(gsc_uc_to_gt(gsc)->uncore, + needs_wakeref)) == + HECI1_FWSTS1_PROXY_STATE_NORMAL; +} + +int intel_gsc_uc_fw_proxy_get_status(struct intel_gsc_uc *gsc) { - return REG_FIELD_GET(GSC_FW_CURRENT_STATE, - gsc_uc_get_fw_status(gsc_uc_to_gt(gsc)->uncore)) == - GSC_FW_PROXY_STATE_NORMAL; + if (!(IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY))) + return -ENODEV; + if (!intel_uc_fw_is_loadable(&gsc->fw)) + return -ENODEV; + if (__intel_uc_fw_status(&gsc->fw) == INTEL_UC_FIRMWARE_LOAD_FAIL) + return -ENOLINK; + if (!intel_gsc_uc_fw_proxy_init_done(gsc, true)) + return -EAGAIN; + + return 0; } bool intel_gsc_uc_fw_init_done(struct intel_gsc_uc *gsc) { - return gsc_uc_get_fw_status(gsc_uc_to_gt(gsc)->uncore) & GSC_FW_INIT_COMPLETE_BIT; + return gsc_uc_get_fw_status(gsc_uc_to_gt(gsc)->uncore, false) & + HECI1_FWSTS1_INIT_COMPLETE; +} + +static inline u32 cpd_entry_offset(const struct intel_gsc_cpd_entry *entry) +{ + return entry->offset & INTEL_GSC_CPD_ENTRY_OFFSET_MASK; +} + +int intel_gsc_fw_get_binary_info(struct intel_uc_fw *gsc_fw, const void *data, size_t size) +{ + struct intel_gsc_uc *gsc = container_of(gsc_fw, struct intel_gsc_uc, fw); + struct intel_gt *gt = gsc_uc_to_gt(gsc); + const struct intel_gsc_layout_pointers *layout = data; + const struct intel_gsc_bpdt_header *bpdt_header = NULL; + const struct intel_gsc_bpdt_entry *bpdt_entry = NULL; + const struct intel_gsc_cpd_header_v2 *cpd_header = NULL; + const struct intel_gsc_cpd_entry *cpd_entry = NULL; + const struct intel_gsc_manifest_header *manifest; + size_t min_size = sizeof(*layout); + int i; + + if (size < min_size) { + gt_err(gt, "GSC FW too small! %zu < %zu\n", size, min_size); + return -ENODATA; + } + + /* + * The GSC binary starts with the pointer layout, which contains the + * locations of the various partitions of the binary. The one we're + * interested in to get the version is the boot1 partition, where we can + * find a BPDT header followed by entries, one of which points to the + * RBE sub-section of the partition. From here, we can parse the CPD + * header and the following entries to find the manifest location + * (entry identified by the "RBEP.man" name), from which we can finally + * extract the version. + * + * -------------------------------------------------- + * [ intel_gsc_layout_pointers ] + * [ ... ] + * [ boot1.offset >---------------------------]------o + * [ ... ] | + * -------------------------------------------------- | + * | + * -------------------------------------------------- | + * [ intel_gsc_bpdt_header ]<-----o + * -------------------------------------------------- + * [ intel_gsc_bpdt_entry[] ] + * [ entry1 ] + * [ ... ] + * [ entryX ] + * [ type == GSC_RBE ] + * [ offset >-----------------------------]------o + * [ ... ] | + * -------------------------------------------------- | + * | + * -------------------------------------------------- | + * [ intel_gsc_cpd_header_v2 ]<-----o + * -------------------------------------------------- + * [ intel_gsc_cpd_entry[] ] + * [ entry1 ] + * [ ... ] + * [ entryX ] + * [ "RBEP.man" ] + * [ ... ] + * [ offset >----------------------------]------o + * [ ... ] | + * -------------------------------------------------- | + * | + * -------------------------------------------------- | + * [ intel_gsc_manifest_header ]<-----o + * [ ... ] + * [ intel_gsc_version fw_version ] + * [ ... ] + * -------------------------------------------------- + */ + + min_size = layout->boot1.offset + layout->boot1.size; + if (size < min_size) { + gt_err(gt, "GSC FW too small for boot section! %zu < %zu\n", + size, min_size); + return -ENODATA; + } + + min_size = sizeof(*bpdt_header); + if (layout->boot1.size < min_size) { + gt_err(gt, "GSC FW boot section too small for BPDT header: %u < %zu\n", + layout->boot1.size, min_size); + return -ENODATA; + } + + bpdt_header = data + layout->boot1.offset; + if (bpdt_header->signature != INTEL_GSC_BPDT_HEADER_SIGNATURE) { + gt_err(gt, "invalid signature for BPDT header: 0x%08x!\n", + bpdt_header->signature); + return -EINVAL; + } + + min_size += sizeof(*bpdt_entry) * bpdt_header->descriptor_count; + if (layout->boot1.size < min_size) { + gt_err(gt, "GSC FW boot section too small for BPDT entries: %u < %zu\n", + layout->boot1.size, min_size); + return -ENODATA; + } + + bpdt_entry = (void *)bpdt_header + sizeof(*bpdt_header); + for (i = 0; i < bpdt_header->descriptor_count; i++, bpdt_entry++) { + if ((bpdt_entry->type & INTEL_GSC_BPDT_ENTRY_TYPE_MASK) != + INTEL_GSC_BPDT_ENTRY_TYPE_GSC_RBE) + continue; + + cpd_header = (void *)bpdt_header + bpdt_entry->sub_partition_offset; + min_size = bpdt_entry->sub_partition_offset + sizeof(*cpd_header); + break; + } + + if (!cpd_header) { + gt_err(gt, "couldn't find CPD header in GSC binary!\n"); + return -ENODATA; + } + + if (layout->boot1.size < min_size) { + gt_err(gt, "GSC FW boot section too small for CPD header: %u < %zu\n", + layout->boot1.size, min_size); + return -ENODATA; + } + + if (cpd_header->header_marker != INTEL_GSC_CPD_HEADER_MARKER) { + gt_err(gt, "invalid marker for CPD header in GSC bin: 0x%08x!\n", + cpd_header->header_marker); + return -EINVAL; + } + + min_size += sizeof(*cpd_entry) * cpd_header->num_of_entries; + if (layout->boot1.size < min_size) { + gt_err(gt, "GSC FW boot section too small for CPD entries: %u < %zu\n", + layout->boot1.size, min_size); + return -ENODATA; + } + + cpd_entry = (void *)cpd_header + cpd_header->header_length; + for (i = 0; i < cpd_header->num_of_entries; i++, cpd_entry++) { + if (strcmp(cpd_entry->name, "RBEP.man") == 0) { + manifest = (void *)cpd_header + cpd_entry_offset(cpd_entry); + intel_uc_fw_version_from_gsc_manifest(&gsc->release, + manifest); + gsc->security_version = manifest->security_version; + break; + } + } + + return 0; } static int emit_gsc_fw_load(struct i915_request *rq, struct intel_gsc_uc *gsc) @@ -114,48 +282,25 @@ out_rq: static int gsc_fw_load_prepare(struct intel_gsc_uc *gsc) { struct intel_gt *gt = gsc_uc_to_gt(gsc); - struct drm_i915_private *i915 = gt->i915; - struct drm_i915_gem_object *obj; - void *src, *dst; + void *src; if (!gsc->local) return -ENODEV; - obj = gsc->local->obj; - - if (obj->base.size < gsc->fw.size) + if (gsc->local->size < gsc->fw.size) return -ENOSPC; - /* - * Wa_22016122933: For MTL the shared memory needs to be mapped - * as WC on CPU side and UC (PAT index 2) on GPU side - */ - if (IS_METEORLAKE(i915)) - i915_gem_object_set_cache_coherency(obj, I915_CACHE_NONE); - - dst = i915_gem_object_pin_map_unlocked(obj, - i915_coherent_map_type(i915, obj, true)); - if (IS_ERR(dst)) - return PTR_ERR(dst); - src = i915_gem_object_pin_map_unlocked(gsc->fw.obj, - i915_coherent_map_type(i915, gsc->fw.obj, true)); - if (IS_ERR(src)) { - i915_gem_object_unpin_map(obj); + intel_gt_coherent_map_type(gt, gsc->fw.obj, true)); + if (IS_ERR(src)) return PTR_ERR(src); - } - memset(dst, 0, obj->base.size); - memcpy(dst, src, gsc->fw.size); + memcpy_toio(gsc->local_vaddr, src, gsc->fw.size); + memset_io(gsc->local_vaddr + gsc->fw.size, 0, gsc->local->size - gsc->fw.size); - /* - * Wa_22016122933: Making sure the data in dst is - * visible to GSC right away - */ intel_guc_write_barrier(>->uc.guc); i915_gem_object_unpin_map(gsc->fw.obj); - i915_gem_object_unpin_map(obj); return 0; } @@ -163,12 +308,94 @@ static int gsc_fw_load_prepare(struct intel_gsc_uc *gsc) static int gsc_fw_wait(struct intel_gt *gt) { return intel_wait_for_register(gt->uncore, - GSC_FW_STATUS_REG, - GSC_FW_INIT_COMPLETE_BIT, - GSC_FW_INIT_COMPLETE_BIT, + HECI_FWSTS(MTL_GSC_HECI1_BASE, 1), + HECI1_FWSTS1_INIT_COMPLETE, + HECI1_FWSTS1_INIT_COMPLETE, 500); } +struct intel_gsc_mkhi_header { + u8 group_id; +#define MKHI_GROUP_ID_GFX_SRV 0x30 + + u8 command; +#define MKHI_GFX_SRV_GET_HOST_COMPATIBILITY_VERSION (0x42) + + u8 reserved; + u8 result; +} __packed; + +struct mtl_gsc_ver_msg_in { + struct intel_gsc_mtl_header header; + struct intel_gsc_mkhi_header mkhi; +} __packed; + +struct mtl_gsc_ver_msg_out { + struct intel_gsc_mtl_header header; + struct intel_gsc_mkhi_header mkhi; + u16 proj_major; + u16 compat_major; + u16 compat_minor; + u16 reserved[5]; +} __packed; + +#define GSC_VER_PKT_SZ SZ_4K + +static int gsc_fw_query_compatibility_version(struct intel_gsc_uc *gsc) +{ + struct intel_gt *gt = gsc_uc_to_gt(gsc); + struct mtl_gsc_ver_msg_in *msg_in; + struct mtl_gsc_ver_msg_out *msg_out; + struct i915_vma *vma; + u64 offset; + void *vaddr; + int err; + + err = intel_guc_allocate_and_map_vma(>->uc.guc, GSC_VER_PKT_SZ * 2, + &vma, &vaddr); + if (err) { + gt_err(gt, "failed to allocate vma for GSC version query\n"); + return err; + } + + offset = i915_ggtt_offset(vma); + msg_in = vaddr; + msg_out = vaddr + GSC_VER_PKT_SZ; + + intel_gsc_uc_heci_cmd_emit_mtl_header(&msg_in->header, + HECI_MEADDRESS_MKHI, + sizeof(*msg_in), 0); + msg_in->mkhi.group_id = MKHI_GROUP_ID_GFX_SRV; + msg_in->mkhi.command = MKHI_GFX_SRV_GET_HOST_COMPATIBILITY_VERSION; + + err = intel_gsc_uc_heci_cmd_submit_packet(>->uc.gsc, + offset, + sizeof(*msg_in), + offset + GSC_VER_PKT_SZ, + GSC_VER_PKT_SZ); + if (err) { + gt_err(gt, + "failed to submit GSC request for compatibility version: %d\n", + err); + goto out_vma; + } + + if (msg_out->header.message_size != sizeof(*msg_out)) { + gt_err(gt, "invalid GSC reply length %u [expected %zu], s=0x%x, f=0x%x, r=0x%x\n", + msg_out->header.message_size, sizeof(*msg_out), + msg_out->header.status, msg_out->header.flags, msg_out->mkhi.result); + err = -EPROTO; + goto out_vma; + } + + gsc->fw.file_selected.ver.major = msg_out->compat_major; + gsc->fw.file_selected.ver.minor = msg_out->compat_minor; + +out_vma: + i915_vma_unpin_and_release(&vma, I915_VMA_RELEASE_MAP); + return err; +} + int intel_gsc_uc_fw_upload(struct intel_gsc_uc *gsc) { struct intel_gt *gt = gsc_uc_to_gt(gsc); @@ -226,10 +453,24 @@ int intel_gsc_uc_fw_upload(struct intel_gsc_uc *gsc) if (err) goto fail; + err = gsc_fw_query_compatibility_version(gsc); + if (err) + goto fail; + + /* we only support compatibility version 1.0 at the moment */ + err = intel_uc_check_file_version(gsc_fw, NULL); + if (err) + goto fail; + /* FW is not fully operational until we enable SW proxy */ intel_uc_fw_change_status(gsc_fw, INTEL_UC_FIRMWARE_TRANSFERRED); - gt_info(gt, "Loaded GSC firmware %s\n", gsc_fw->file_selected.path); + gt_info(gt, "Loaded GSC firmware %s (cv%u.%u, r%u.%u.%u.%u, svn %u)\n", + gsc_fw->file_selected.path, + gsc_fw->file_selected.ver.major, gsc_fw->file_selected.ver.minor, + gsc->release.major, gsc->release.minor, + gsc->release.patch, gsc->release.build, + gsc->security_version); return 0; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h index fff8928218df..bc9dd0de8aaf 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_fw.h @@ -9,10 +9,13 @@ #include <linux/types.h> struct intel_gsc_uc; +struct intel_uc_fw; struct intel_uncore; +int intel_gsc_fw_get_binary_info(struct intel_uc_fw *gsc_fw, const void *data, size_t size); int intel_gsc_uc_fw_upload(struct intel_gsc_uc *gsc); bool intel_gsc_uc_fw_init_done(struct intel_gsc_uc *gsc); -bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc); +bool intel_gsc_uc_fw_proxy_init_done(struct intel_gsc_uc *gsc, bool needs_wakeref); +int intel_gsc_uc_fw_proxy_get_status(struct intel_gsc_uc *gsc); #endif diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c index c659cc01f32f..0d3b22a74365 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.c @@ -7,10 +7,11 @@ #include "gt/intel_gt.h" #include "gt/intel_gt_print.h" -#include "intel_gsc_uc.h" #include "intel_gsc_fw.h" -#include "i915_drv.h" #include "intel_gsc_proxy.h" +#include "intel_gsc_uc.h" +#include "i915_drv.h" +#include "i915_reg.h" static void gsc_work(struct work_struct *work) { @@ -61,8 +62,18 @@ static void gsc_work(struct work_struct *work) } ret = intel_gsc_proxy_request_handler(gsc); - if (ret) + if (ret) { + if (actions & GSC_ACTION_FW_LOAD) { + /* + * A proxy failure right after firmware load means the proxy-init + * step has failed so mark GSC as not usable after this + */ + drm_err(>->i915->drm, + "GSC proxy handler failed to init\n"); + intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_LOAD_FAIL); + } goto out_put; + } /* mark the GSC FW init as done the first time we run this */ if (actions & GSC_ACTION_FW_LOAD) { @@ -71,12 +82,13 @@ static void gsc_work(struct work_struct *work) * complete the request handling cleanly, so we need to check the * status register to check if the proxy init was actually successful */ - if (intel_gsc_uc_fw_proxy_init_done(gsc)) { + if (intel_gsc_uc_fw_proxy_init_done(gsc, false)) { drm_dbg(>->i915->drm, "GSC Proxy initialized\n"); intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_RUNNING); } else { drm_err(>->i915->drm, "GSC status reports proxy init not complete\n"); + intel_uc_fw_change_status(&gsc->fw, INTEL_UC_FIRMWARE_LOAD_FAIL); } } } @@ -98,7 +110,7 @@ static bool gsc_engine_supported(struct intel_gt *gt) GEM_BUG_ON(!gt_is_root(gt) && !gt->info.engine_mask); if (gt_is_root(gt)) - mask = RUNTIME_INFO(gt->i915)->platform_engine_mask; + mask = INTEL_INFO(gt->i915)->platform_engine_mask; else mask = gt->info.engine_mask; @@ -133,26 +145,85 @@ void intel_gsc_uc_init_early(struct intel_gsc_uc *gsc) } } +static int gsc_allocate_and_map_vma(struct intel_gsc_uc *gsc, u32 size) +{ + struct intel_gt *gt = gsc_uc_to_gt(gsc); + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + void __iomem *vaddr; + int ret = 0; + + /* + * The GSC FW doesn't immediately suspend after becoming idle, so there + * is a chance that it could still be awake after we successfully + * return from the pci suspend function, even if there are no pending + * operations. + * The FW might therefore try to access memory for its suspend operation + * after the kernel has completed the HW suspend flow; this can cause + * issues if the FW is mapped in normal RAM memory, as some of the + * involved HW units might've already lost power. + * The driver must therefore avoid this situation and the recommended + * way to do so is to use stolen memory for the GSC memory allocation, + * because stolen memory takes a different path in HW and it is + * guaranteed to always work as long as the GPU itself is awake (which + * it must be if the GSC is awake). + */ + obj = i915_gem_object_create_stolen(gt->i915, size); + if (IS_ERR(obj)) + return PTR_ERR(obj); + + vma = i915_gem_object_ggtt_pin(obj, NULL, 0, 0, 0); + if (IS_ERR(vma)) { + ret = PTR_ERR(vma); + goto err; + } + + vaddr = i915_vma_pin_iomap(vma); + i915_vma_unpin(vma); + if (IS_ERR(vaddr)) { + ret = PTR_ERR(vaddr); + goto err; + } + + i915_vma_make_unshrinkable(vma); + + gsc->local = vma; + gsc->local_vaddr = vaddr; + + return 0; + +err: + i915_gem_object_put(obj); + return ret; +} + +static void gsc_unmap_and_free_vma(struct intel_gsc_uc *gsc) +{ + struct i915_vma *vma = fetch_and_zero(&gsc->local); + + if (!vma) + return; + + gsc->local_vaddr = NULL; + i915_vma_unpin_iomap(vma); + i915_gem_object_put(vma->obj); +} + int intel_gsc_uc_init(struct intel_gsc_uc *gsc) { static struct lock_class_key gsc_lock; struct intel_gt *gt = gsc_uc_to_gt(gsc); struct intel_engine_cs *engine = gt->engine[GSC0]; struct intel_context *ce; - struct i915_vma *vma; int err; err = intel_uc_fw_init(&gsc->fw); if (err) goto out; - vma = intel_guc_allocate_vma(>->uc.guc, SZ_8M); - if (IS_ERR(vma)) { - err = PTR_ERR(vma); + err = gsc_allocate_and_map_vma(gsc, SZ_4M); + if (err) goto out_fw; - } - - gsc->local = vma; ce = intel_engine_create_pinned_context(engine, engine->gt->vm, SZ_4K, I915_GEM_HWS_GSC_ADDR, @@ -173,7 +244,7 @@ int intel_gsc_uc_init(struct intel_gsc_uc *gsc) return 0; out_vma: - i915_vma_unpin_and_release(&gsc->local, 0); + gsc_unmap_and_free_vma(gsc); out_fw: intel_uc_fw_fini(&gsc->fw); out: @@ -197,7 +268,7 @@ void intel_gsc_uc_fini(struct intel_gsc_uc *gsc) if (gsc->ce) intel_engine_destroy_pinned_context(fetch_and_zero(&gsc->ce)); - i915_vma_unpin_and_release(&gsc->local, 0); + gsc_unmap_and_free_vma(gsc); intel_uc_fw_fini(&gsc->fw); } @@ -245,3 +316,45 @@ void intel_gsc_uc_load_start(struct intel_gsc_uc *gsc) queue_work(gsc->wq, &gsc->work); } + +void intel_gsc_uc_load_status(struct intel_gsc_uc *gsc, struct drm_printer *p) +{ + struct intel_gt *gt = gsc_uc_to_gt(gsc); + struct intel_uncore *uncore = gt->uncore; + intel_wakeref_t wakeref; + + if (!intel_gsc_uc_is_supported(gsc)) { + drm_printf(p, "GSC not supported\n"); + return; + } + + if (!intel_gsc_uc_is_wanted(gsc)) { + drm_printf(p, "GSC disabled\n"); + return; + } + + drm_printf(p, "GSC firmware: %s\n", gsc->fw.file_selected.path); + if (gsc->fw.file_selected.path != gsc->fw.file_wanted.path) + drm_printf(p, "GSC firmware wanted: %s\n", gsc->fw.file_wanted.path); + drm_printf(p, "\tstatus: %s\n", intel_uc_fw_status_repr(gsc->fw.status)); + + drm_printf(p, "Release: %u.%u.%u.%u\n", + gsc->release.major, gsc->release.minor, + gsc->release.patch, gsc->release.build); + + drm_printf(p, "Compatibility Version: %u.%u [min expected %u.%u]\n", + gsc->fw.file_selected.ver.major, gsc->fw.file_selected.ver.minor, + gsc->fw.file_wanted.ver.major, gsc->fw.file_wanted.ver.minor); + + drm_printf(p, "SVN: %u\n", gsc->security_version); + + with_intel_runtime_pm(uncore->rpm, wakeref) { + u32 i; + + for (i = 1; i <= 6; i++) { + u32 status = intel_uncore_read(uncore, + HECI_FWSTS(MTL_GSC_HECI1_BASE, i)); + drm_printf(p, "HECI1 FWSTST%u = 0x%08x\n", i, status); + } + } +} diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h index a2a0813b8a76..c8082cf200fc 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc.h @@ -8,6 +8,7 @@ #include "intel_uc_fw.h" +struct drm_printer; struct i915_vma; struct intel_context; struct i915_gsc_proxy_component; @@ -17,7 +18,26 @@ struct intel_gsc_uc { struct intel_uc_fw fw; /* GSC-specific additions */ + + /* + * The GSC has 3 version numbers: + * - Release version (incremented with each build) + * - Security version (incremented on security fix) + * - Compatibility version (incremented on interface change) + * + * The one we care about to use the binary is the last one, so that's + * the one we save inside the intel_uc_fw structure. The other two + * versions are only used for debug/info purposes, so we save them here. + * + * Note that the release and security versions are available in the + * binary header, while the compatibility version must be queried after + * loading the binary. + */ + struct intel_uc_fw_ver release; + u32 security_version; + struct i915_vma *local; /* private memory for GSC usage */ + void __iomem *local_vaddr; /* pointer to access the private memory */ struct intel_context *ce; /* for submission to GSC FW via GSC engine */ /* for delayed load and proxy handling */ @@ -44,6 +64,7 @@ void intel_gsc_uc_suspend(struct intel_gsc_uc *gsc); void intel_gsc_uc_resume(struct intel_gsc_uc *gsc); void intel_gsc_uc_flush_work(struct intel_gsc_uc *gsc); void intel_gsc_uc_load_start(struct intel_gsc_uc *gsc); +void intel_gsc_uc_load_status(struct intel_gsc_uc *gsc, struct drm_printer *p); static inline bool intel_gsc_uc_is_supported(struct intel_gsc_uc *gsc) { diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_debugfs.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_debugfs.c new file mode 100644 index 000000000000..5baacd822a1c --- /dev/null +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_debugfs.c @@ -0,0 +1,39 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2023 Intel Corporation + */ + +#include <drm/drm_print.h> + +#include "gt/intel_gt.h" +#include "gt/intel_gt_debugfs.h" +#include "gt/intel_gt_print.h" +#include "intel_gsc_uc.h" +#include "intel_gsc_uc_debugfs.h" +#include "i915_drv.h" + +static int gsc_info_show(struct seq_file *m, void *data) +{ + struct drm_printer p = drm_seq_file_printer(m); + struct intel_gsc_uc *gsc = m->private; + + if (!intel_gsc_uc_is_supported(gsc)) + return -ENODEV; + + intel_gsc_uc_load_status(gsc, &p); + + return 0; +} +DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE(gsc_info); + +void intel_gsc_uc_debugfs_register(struct intel_gsc_uc *gsc_uc, struct dentry *root) +{ + static const struct intel_gt_debugfs_file files[] = { + { "gsc_info", &gsc_info_fops, NULL }, + }; + + if (!intel_gsc_uc_is_supported(gsc_uc)) + return; + + intel_gt_debugfs_register_files(root, files, ARRAY_SIZE(files), gsc_uc); +} diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_debugfs.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_debugfs.h new file mode 100644 index 000000000000..3415ad39aabb --- /dev/null +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_debugfs.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2023 Intel Corporation + */ + +#ifndef DEBUGFS_GSC_UC_H +#define DEBUGFS_GSC_UC_H + +struct intel_gsc_uc; +struct dentry; + +void intel_gsc_uc_debugfs_register(struct intel_gsc_uc *gsc, struct dentry *root); + +#endif /* DEBUGFS_GSC_UC_H */ diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h index ef70e304904a..09d3fbdad05a 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.h @@ -17,6 +17,7 @@ struct intel_gsc_mtl_header { #define GSC_HECI_VALIDITY_MARKER 0xA578875A u8 heci_client_id; +#define HECI_MEADDRESS_MKHI 7 #define HECI_MEADDRESS_PROXY 10 #define HECI_MEADDRESS_PXP 17 #define HECI_MEADDRESS_HDCP 18 diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c index 2eb891b270ae..569b5fe94c41 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c @@ -745,10 +745,11 @@ struct i915_vma *intel_guc_allocate_vma(struct intel_guc *guc, u32 size) return ERR_CAST(obj); /* - * Wa_22016122933: For MTL the shared memory needs to be mapped - * as WC on CPU side and UC (PAT index 2) on GPU side + * Wa_22016122933: For Media version 13.0, all Media GT shared + * memory needs to be mapped as WC on CPU side and UC (PAT + * index 2) on GPU side. */ - if (IS_METEORLAKE(gt->i915)) + if (intel_gt_needs_wa_22016122933(gt)) i915_gem_object_set_cache_coherency(obj, I915_CACHE_NONE); vma = i915_vma_instance(obj, >->ggtt->vm, NULL); @@ -792,8 +793,8 @@ int intel_guc_allocate_and_map_vma(struct intel_guc *guc, u32 size, return PTR_ERR(vma); vaddr = i915_gem_object_pin_map_unlocked(vma->obj, - i915_coherent_map_type(guc_to_gt(guc)->i915, - vma->obj, true)); + intel_gt_coherent_map_type(guc_to_gt(guc), + vma->obj, true)); if (IS_ERR(vaddr)) { i915_vma_unpin_and_release(&vma, 0); return PTR_ERR(vaddr); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c index f28a3a83742d..97eadd08181d 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c @@ -960,10 +960,6 @@ static int ct_read(struct intel_guc_ct *ct, struct ct_incoming_msg **msg) /* now update descriptor */ WRITE_ONCE(desc->head, head); - /* - * Wa_22016122933: Making sure the head update is - * visible to GuC right away - */ intel_guc_write_barrier(ct_to_guc(ct)); return available - len; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c index 364d0d546ec8..0f79cb658518 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c @@ -251,9 +251,11 @@ static int guc_wait_ucode(struct intel_guc *guc) if (ret == 0) ret = -ENXIO; } else if (delta_ms > 200) { - guc_warn(guc, "excessive init time: %lldms! [freq = %dMHz, before = %dMHz, status = 0x%08X, count = %d, ret = %d]\n", - delta_ms, intel_rps_read_actual_frequency(&uncore->gt->rps), - before_freq, status, count, ret); + guc_warn(guc, "excessive init time: %lldms! [status = 0x%08X, count = %d, ret = %d]\n", + delta_ms, status, count, ret); + guc_warn(guc, "excessive init time: [freq = %dMHz, before = %dMHz, perf_limit_reasons = 0x%08X]\n", + intel_rps_read_actual_frequency(&uncore->gt->rps), before_freq, + intel_uncore_read(uncore, intel_gt_perf_limit_reasons_reg(gt))); } else { guc_dbg(guc, "init took %lldms, freq = %dMHz, before = %dMHz, status = 0x%08X, count = %d, ret = %d\n", delta_ms, intel_rps_read_actual_frequency(&uncore->gt->rps), diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_hwconfig.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_hwconfig.c index 852bea0208ce..cc9569af7f0c 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_hwconfig.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_hwconfig.c @@ -94,7 +94,7 @@ static int guc_hwconfig_fill_buffer(struct intel_guc *guc, struct intel_hwconfig static bool has_table(struct drm_i915_private *i915) { - if (IS_ALDERLAKE_P(i915) && !IS_ADLP_N(i915)) + if (IS_ALDERLAKE_P(i915) && !IS_ALDERLAKE_P_N(i915)) return true; if (GRAPHICS_VER_FULL(i915) >= IP_VER(12, 55)) return true; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c index ee9f83af7cf6..477df260ae3a 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c @@ -470,12 +470,19 @@ int intel_guc_slpc_set_ignore_eff_freq(struct intel_guc_slpc *slpc, bool val) ret = slpc_set_param(slpc, SLPC_PARAM_IGNORE_EFFICIENT_FREQUENCY, val); - if (ret) + if (ret) { guc_probe_error(slpc_to_guc(slpc), "Failed to set efficient freq(%d): %pe\n", val, ERR_PTR(ret)); - else + } else { slpc->ignore_eff_freq = val; + /* Set min to RPn when we disable efficient freq */ + if (val) + ret = slpc_set_param(slpc, + SLPC_PARAM_GLOBAL_MIN_GT_UNSLICE_FREQ_MHZ, + slpc->min_freq); + } + intel_runtime_pm_put(&i915->runtime_pm, wakeref); mutex_unlock(&slpc->lock); return ret; @@ -602,9 +609,8 @@ static int slpc_set_softlimits(struct intel_guc_slpc *slpc) return ret; if (!slpc->min_freq_softlimit) { - ret = intel_guc_slpc_get_min_freq(slpc, &slpc->min_freq_softlimit); - if (unlikely(ret)) - return ret; + /* Min softlimit is initialized to RPn */ + slpc->min_freq_softlimit = slpc->min_freq; slpc_to_gt(slpc)->defaults.min_freq = slpc->min_freq_softlimit; } else { return intel_guc_slpc_set_min_freq(slpc, @@ -755,6 +761,9 @@ int intel_guc_slpc_enable(struct intel_guc_slpc *slpc) return ret; } + /* Set cached value of ignore efficient freq */ + intel_guc_slpc_set_ignore_eff_freq(slpc, slpc->ignore_eff_freq); + /* Revert SLPC min/max to softlimits if necessary */ ret = slpc_set_softlimits(slpc); if (unlikely(ret)) { @@ -765,9 +774,6 @@ int intel_guc_slpc_enable(struct intel_guc_slpc *slpc) /* Set cached media freq ratio mode */ intel_guc_slpc_set_media_ratio_mode(slpc, slpc->media_ratio_mode); - /* Set cached value of ignore efficient freq */ - intel_guc_slpc_set_ignore_eff_freq(slpc, slpc->ignore_eff_freq); - return 0; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index a0e3ef1c65d2..b5b7f2fe8c78 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -5470,6 +5470,9 @@ guc_create_virtual(struct intel_engine_cs **siblings, unsigned int count, ve->base.flags = I915_ENGINE_IS_VIRTUAL; + BUILD_BUG_ON(ilog2(VIRTUAL_ENGINES) < I915_NUM_ENGINES); + ve->base.mask = VIRTUAL_ENGINES; + intel_context_init(&ve->context, &ve->base); for (n = 0; n < count; n++) { diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c index ddd146265beb..ba9e07fc2b57 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c @@ -26,6 +26,7 @@ * The kernel driver is only responsible for loading the HuC firmware and * triggering its security authentication. This is done differently depending * on the platform: + * * - older platforms (from Gen9 to most Gen12s): the load is performed via DMA * and the authentication via GuC * - DG2: load and authentication are both performed via GSC. @@ -33,6 +34,7 @@ * not-DG2 older platforms), while the authentication is done in 2-steps, * a first auth for clear-media workloads via GuC and a second one for all * workloads via GSC. + * * On platforms where the GuC does the authentication, to correctly do so the * HuC binary must be loaded before the GuC one. * Loading the HuC is optional; however, not using the HuC might negatively @@ -265,7 +267,7 @@ static bool vcs_supported(struct intel_gt *gt) GEM_BUG_ON(!gt_is_root(gt) && !gt->info.engine_mask); if (gt_is_root(gt)) - mask = RUNTIME_INFO(gt->i915)->platform_engine_mask; + mask = INTEL_INFO(gt->i915)->platform_engine_mask; else mask = gt->info.engine_mask; @@ -308,9 +310,9 @@ void intel_huc_init_early(struct intel_huc *huc) huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HUC_LOAD_SUCCESSFUL; huc->status[INTEL_HUC_AUTH_BY_GSC].value = HUC_LOAD_SUCCESSFUL; } else { - huc->status[INTEL_HUC_AUTH_BY_GSC].reg = HECI_FWSTS5(MTL_GSC_HECI1_BASE); - huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HECI_FWSTS5_HUC_AUTH_DONE; - huc->status[INTEL_HUC_AUTH_BY_GSC].value = HECI_FWSTS5_HUC_AUTH_DONE; + huc->status[INTEL_HUC_AUTH_BY_GSC].reg = HECI_FWSTS(MTL_GSC_HECI1_BASE, 5); + huc->status[INTEL_HUC_AUTH_BY_GSC].mask = HECI1_FWSTS5_HUC_AUTH_DONE; + huc->status[INTEL_HUC_AUTH_BY_GSC].value = HECI1_FWSTS5_HUC_AUTH_DONE; } } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c index e608152fecfc..b648238cc675 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc_fw.c @@ -27,7 +27,6 @@ struct mtl_huc_auth_msg_out { int intel_huc_fw_auth_via_gsccs(struct intel_huc *huc) { struct intel_gt *gt = huc_to_gt(huc); - struct drm_i915_private *i915 = gt->i915; struct drm_i915_gem_object *obj; struct mtl_huc_auth_msg_in *msg_in; struct mtl_huc_auth_msg_out *msg_out; @@ -43,7 +42,7 @@ int intel_huc_fw_auth_via_gsccs(struct intel_huc *huc) pkt_offset = i915_ggtt_offset(huc->heci_pkt); pkt_vaddr = i915_gem_object_pin_map_unlocked(obj, - i915_coherent_map_type(i915, obj, true)); + intel_gt_coherent_map_type(gt, obj, true)); if (IS_ERR(pkt_vaddr)) return PTR_ERR(pkt_vaddr); @@ -107,15 +106,6 @@ out_unpin: return err; } -static void get_version_from_gsc_manifest(struct intel_uc_fw_ver *ver, const void *data) -{ - const struct intel_gsc_manifest_header *manifest = data; - - ver->major = manifest->fw_version.major; - ver->minor = manifest->fw_version.minor; - ver->patch = manifest->fw_version.hotfix; -} - static bool css_valid(const void *data, size_t size) { const struct uc_css_header *css = data; @@ -227,8 +217,8 @@ int intel_huc_fw_get_binary_info(struct intel_uc_fw *huc_fw, const void *data, s for (i = 0; i < header->num_of_entries; i++, entry++) { if (strcmp(entry->name, "HUCP.man") == 0) - get_version_from_gsc_manifest(&huc_fw->file_selected.ver, - data + entry_offset(entry)); + intel_uc_fw_version_from_gsc_manifest(&huc_fw->file_selected.ver, + data + entry_offset(entry)); if (strcmp(entry->name, "huc_fw") == 0) { u32 offset = entry_offset(entry); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index 18250fb64bd8..98b103375b7a 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -43,7 +43,7 @@ static void uc_expand_default_options(struct intel_uc *uc) } /* Intermediate platforms are HuC authentication only */ - if (IS_ALDERLAKE_S(i915) && !IS_ADLS_RPLS(i915)) { + if (IS_ALDERLAKE_S(i915) && !IS_RAPTORLAKE_S(i915)) { i915->params.enable_guc = ENABLE_GUC_LOAD_HUC; return; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_debugfs.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_debugfs.c index 2f93cc4e408a..6d541c866edb 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_debugfs.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_debugfs.c @@ -10,6 +10,7 @@ #include "gt/intel_gt_debugfs.h" #include "intel_guc_debugfs.h" +#include "intel_gsc_uc_debugfs.h" #include "intel_huc_debugfs.h" #include "intel_uc.h" #include "intel_uc_debugfs.h" @@ -58,6 +59,7 @@ void intel_uc_debugfs_register(struct intel_uc *uc, struct dentry *gt_root) intel_gt_debugfs_register_files(root, files, ARRAY_SIZE(files), uc); + intel_gsc_uc_debugfs_register(&uc->gsc, root); intel_guc_debugfs_register(&uc->guc, root); intel_huc_debugfs_register(&uc->huc, root); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index 944725e62414..8be005de1d28 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -11,7 +11,10 @@ #include <drm/drm_print.h> #include "gem/i915_gem_lmem.h" +#include "gt/intel_gt.h" #include "gt/intel_gt_print.h" +#include "intel_gsc_binary_headers.h" +#include "intel_gsc_fw.h" #include "intel_uc_fw.h" #include "intel_uc_fw_abi.h" #include "i915_drv.h" @@ -277,7 +280,7 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw) * ADL-S, otherwise the GuC might attempt to fetch a config table that * does not exist. */ - if (IS_ADLP_N(i915)) + if (IS_ALDERLAKE_P_N(i915)) p = INTEL_ALDERLAKE_S; GEM_BUG_ON(uc_fw->type >= ARRAY_SIZE(blobs_all)); @@ -468,6 +471,17 @@ static void __uc_fw_user_override(struct drm_i915_private *i915, struct intel_uc } } +void intel_uc_fw_version_from_gsc_manifest(struct intel_uc_fw_ver *ver, + const void *data) +{ + const struct intel_gsc_manifest_header *manifest = data; + + ver->major = manifest->fw_version.major; + ver->minor = manifest->fw_version.minor; + ver->patch = manifest->fw_version.hotfix; + ver->build = manifest->fw_version.build; +} + /** * intel_uc_fw_init_early - initialize the uC object and select the firmware * @uc_fw: uC firmware @@ -668,13 +682,18 @@ static int check_gsc_manifest(struct intel_gt *gt, const struct firmware *fw, struct intel_uc_fw *uc_fw) { - if (uc_fw->type != INTEL_UC_FW_TYPE_HUC) { - gt_err(gt, "trying to GSC-parse a non-HuC binary"); + switch (uc_fw->type) { + case INTEL_UC_FW_TYPE_HUC: + intel_huc_fw_get_binary_info(uc_fw, fw->data, fw->size); + break; + case INTEL_UC_FW_TYPE_GSC: + intel_gsc_fw_get_binary_info(uc_fw, fw->data, fw->size); + break; + default: + MISSING_CASE(uc_fw->type); return -EINVAL; } - intel_huc_fw_get_binary_info(uc_fw, fw->data, fw->size); - if (uc_fw->dma_start_offset) { u32 delta = uc_fw->dma_start_offset; @@ -734,10 +753,6 @@ static int check_fw_header(struct intel_gt *gt, { int err = 0; - /* GSC FW version is queried after the FW is loaded */ - if (uc_fw->type == INTEL_UC_FW_TYPE_GSC) - return 0; - if (uc_fw->has_gsc_headers) err = check_gsc_manifest(gt, fw, uc_fw); else @@ -773,6 +788,80 @@ static int try_firmware_load(struct intel_uc_fw *uc_fw, const struct firmware ** return 0; } +static int check_mtl_huc_guc_compatibility(struct intel_gt *gt, + struct intel_uc_fw_file *huc_selected) +{ + struct intel_uc_fw_file *guc_selected = >->uc.guc.fw.file_selected; + struct intel_uc_fw_ver *huc_ver = &huc_selected->ver; + struct intel_uc_fw_ver *guc_ver = &guc_selected->ver; + bool new_huc, new_guc; + + /* we can only do this check after having fetched both GuC and HuC */ + GEM_BUG_ON(!huc_selected->path || !guc_selected->path); + + /* + * Due to changes in the authentication flow for MTL, HuC 8.5.1 or newer + * requires GuC 70.7.0 or newer. Older HuC binaries will instead require + * GuC < 70.7.0. + */ + new_huc = huc_ver->major > 8 || + (huc_ver->major == 8 && huc_ver->minor > 5) || + (huc_ver->major == 8 && huc_ver->minor == 5 && huc_ver->patch >= 1); + + new_guc = guc_ver->major > 70 || + (guc_ver->major == 70 && guc_ver->minor >= 7); + + if (new_huc != new_guc) { + UNEXPECTED(gt, "HuC %u.%u.%u is incompatible with GuC %u.%u.%u\n", + huc_ver->major, huc_ver->minor, huc_ver->patch, + guc_ver->major, guc_ver->minor, guc_ver->patch); + gt_info(gt, "MTL GuC 70.7.0+ and HuC 8.5.1+ don't work with older releases\n"); + return -ENOEXEC; + } + + return 0; +} + +int intel_uc_check_file_version(struct intel_uc_fw *uc_fw, bool *old_ver) +{ + struct intel_gt *gt = __uc_fw_to_gt(uc_fw); + struct intel_uc_fw_file *wanted = &uc_fw->file_wanted; + struct intel_uc_fw_file *selected = &uc_fw->file_selected; + int ret; + + /* + * MTL has some compatibility issues with early GuC/HuC binaries + * not working with newer ones. This is specific to MTL and we + * don't expect it to extend to other platforms. + */ + if (IS_METEORLAKE(gt->i915) && uc_fw->type == INTEL_UC_FW_TYPE_HUC) { + ret = check_mtl_huc_guc_compatibility(gt, selected); + if (ret) + return ret; + } + + if (!wanted->ver.major || !selected->ver.major) + return 0; + + /* Check the file's major version was as it claimed */ + if (selected->ver.major != wanted->ver.major) { + UNEXPECTED(gt, "%s firmware %s: unexpected version: %u.%u != %u.%u\n", + intel_uc_fw_type_repr(uc_fw->type), selected->path, + selected->ver.major, selected->ver.minor, + wanted->ver.major, wanted->ver.minor); + if (!intel_uc_fw_is_overridden(uc_fw)) + return -ENOEXEC; + } else if (old_ver) { + if (selected->ver.minor < wanted->ver.minor) + *old_ver = true; + else if ((selected->ver.minor == wanted->ver.minor) && + (selected->ver.patch < wanted->ver.patch)) + *old_ver = true; + } + + return 0; +} + /** * intel_uc_fw_fetch - fetch uC firmware * @uc_fw: uC firmware @@ -840,25 +929,9 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw) goto fail; } - if (uc_fw->file_wanted.ver.major && uc_fw->file_selected.ver.major) { - /* Check the file's major version was as it claimed */ - if (uc_fw->file_selected.ver.major != uc_fw->file_wanted.ver.major) { - UNEXPECTED(gt, "%s firmware %s: unexpected version: %u.%u != %u.%u\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->file_selected.path, - uc_fw->file_selected.ver.major, uc_fw->file_selected.ver.minor, - uc_fw->file_wanted.ver.major, uc_fw->file_wanted.ver.minor); - if (!intel_uc_fw_is_overridden(uc_fw)) { - err = -ENOEXEC; - goto fail; - } - } else { - if (uc_fw->file_selected.ver.minor < uc_fw->file_wanted.ver.minor) - old_ver = true; - else if ((uc_fw->file_selected.ver.minor == uc_fw->file_wanted.ver.minor) && - (uc_fw->file_selected.ver.patch < uc_fw->file_wanted.ver.patch)) - old_ver = true; - } - } + err = intel_uc_check_file_version(uc_fw, &old_ver); + if (err) + goto fail; if (old_ver && uc_fw->file_selected.ver.major) { /* Preserve the version that was really wanted */ @@ -1125,7 +1198,7 @@ static int uc_fw_rsa_data_create(struct intel_uc_fw *uc_fw) return PTR_ERR(vma); vaddr = i915_gem_object_pin_map_unlocked(vma->obj, - i915_coherent_map_type(gt->i915, vma->obj, true)); + intel_gt_coherent_map_type(gt, vma->obj, true)); if (IS_ERR(vaddr)) { i915_vma_unpin_and_release(&vma, 0); err = PTR_ERR(vaddr); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h index 054f02811971..9a431726c8d5 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h @@ -70,6 +70,7 @@ struct intel_uc_fw_ver { u32 major; u32 minor; u32 patch; + u32 build; }; /* @@ -289,6 +290,9 @@ static inline u32 intel_uc_fw_get_upload_size(struct intel_uc_fw *uc_fw) return __intel_uc_fw_get_upload_size(uc_fw); } +void intel_uc_fw_version_from_gsc_manifest(struct intel_uc_fw_ver *ver, + const void *data); +int intel_uc_check_file_version(struct intel_uc_fw *uc_fw, bool *old_ver); void intel_uc_fw_init_early(struct intel_uc_fw *uc_fw, enum intel_uc_fw_type type, bool needs_ggtt_mapping); diff --git a/drivers/gpu/drm/i915/gt/uc/selftest_guc.c b/drivers/gpu/drm/i915/gt/uc/selftest_guc.c index 1fd760539f77..bfb72143566f 100644 --- a/drivers/gpu/drm/i915/gt/uc/selftest_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/selftest_guc.c @@ -204,9 +204,9 @@ static int intel_guc_steal_guc_ids(void *arg) if (IS_ERR(rq)) { ret = PTR_ERR(rq); rq = NULL; - if (ret != -EAGAIN) { - guc_err(guc, "Failed to create request %d: %pe\n", - context_index, ERR_PTR(ret)); + if ((ret != -EAGAIN) || !last) { + guc_err(guc, "Failed to create %srequest %d: %pe\n", + last ? "" : "first ", context_index, ERR_PTR(ret)); goto err_spin_rq; } } else { diff --git a/drivers/gpu/drm/i915/gvt/edid.c b/drivers/gpu/drm/i915/gvt/edid.c index 2a0438f12a14..af9afdb53c7f 100644 --- a/drivers/gpu/drm/i915/gvt/edid.c +++ b/drivers/gpu/drm/i915/gvt/edid.c @@ -491,7 +491,7 @@ void intel_gvt_i2c_handle_aux_ch_write(struct intel_vgpu *vgpu, return; } - msg_length = REG_FIELD_GET(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, reg); + msg_length = REG_FIELD_GET(DP_AUX_CH_CTL_MESSAGE_SIZE_MASK, value); // check the msg in DATA register. msg = vgpu_vreg(vgpu, offset + 4); diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index 4ec85308379a..094fca9b0e73 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -49,22 +49,6 @@ static bool enable_out_of_sync = false; static int preallocated_oos_pages = 8192; -static bool intel_gvt_is_valid_gfn(struct intel_vgpu *vgpu, unsigned long gfn) -{ - struct kvm *kvm = vgpu->vfio_device.kvm; - int idx; - bool ret; - - if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status)) - return false; - - idx = srcu_read_lock(&kvm->srcu); - ret = kvm_is_visible_gfn(kvm, gfn); - srcu_read_unlock(&kvm->srcu, idx); - - return ret; -} - /* * validate a gm address and related range size, * translate it to host gm address @@ -1161,31 +1145,6 @@ static inline void ppgtt_generate_shadow_entry(struct intel_gvt_gtt_entry *se, ops->set_pfn(se, s->shadow_page.mfn); } -/* - * Check if can do 2M page - * @vgpu: target vgpu - * @entry: target pfn's gtt entry - * - * Return 1 if 2MB huge gtt shadowing is possible, 0 if miscondition, - * negative if found err. - */ -static int is_2MB_gtt_possible(struct intel_vgpu *vgpu, - struct intel_gvt_gtt_entry *entry) -{ - const struct intel_gvt_gtt_pte_ops *ops = vgpu->gvt->gtt.pte_ops; - kvm_pfn_t pfn; - - if (!HAS_PAGE_SIZES(vgpu->gvt->gt->i915, I915_GTT_PAGE_SIZE_2M)) - return 0; - - if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status)) - return -EINVAL; - pfn = gfn_to_pfn(vgpu->vfio_device.kvm, ops->get_pfn(entry)); - if (is_error_noslot_pfn(pfn)) - return -EINVAL; - return PageTransHuge(pfn_to_page(pfn)); -} - static int split_2MB_gtt_entry(struct intel_vgpu *vgpu, struct intel_vgpu_ppgtt_spt *spt, unsigned long index, struct intel_gvt_gtt_entry *se) @@ -1279,7 +1238,7 @@ static int ppgtt_populate_shadow_entry(struct intel_vgpu *vgpu, { const struct intel_gvt_gtt_pte_ops *pte_ops = vgpu->gvt->gtt.pte_ops; struct intel_gvt_gtt_entry se = *ge; - unsigned long gfn, page_size = PAGE_SIZE; + unsigned long gfn; dma_addr_t dma_addr; int ret; @@ -1291,6 +1250,9 @@ static int ppgtt_populate_shadow_entry(struct intel_vgpu *vgpu, switch (ge->type) { case GTT_TYPE_PPGTT_PTE_4K_ENTRY: gvt_vdbg_mm("shadow 4K gtt entry\n"); + ret = intel_gvt_dma_map_guest_page(vgpu, gfn, PAGE_SIZE, &dma_addr); + if (ret) + return -ENXIO; break; case GTT_TYPE_PPGTT_PTE_64K_ENTRY: gvt_vdbg_mm("shadow 64K gtt entry\n"); @@ -1302,25 +1264,20 @@ static int ppgtt_populate_shadow_entry(struct intel_vgpu *vgpu, return split_64KB_gtt_entry(vgpu, spt, index, &se); case GTT_TYPE_PPGTT_PTE_2M_ENTRY: gvt_vdbg_mm("shadow 2M gtt entry\n"); - ret = is_2MB_gtt_possible(vgpu, ge); - if (ret == 0) + if (!HAS_PAGE_SIZES(vgpu->gvt->gt->i915, I915_GTT_PAGE_SIZE_2M) || + intel_gvt_dma_map_guest_page(vgpu, gfn, + I915_GTT_PAGE_SIZE_2M, &dma_addr)) return split_2MB_gtt_entry(vgpu, spt, index, &se); - else if (ret < 0) - return ret; - page_size = I915_GTT_PAGE_SIZE_2M; break; case GTT_TYPE_PPGTT_PTE_1G_ENTRY: gvt_vgpu_err("GVT doesn't support 1GB entry\n"); return -EINVAL; default: GEM_BUG_ON(1); + return -EINVAL; } - /* direct shadow */ - ret = intel_gvt_dma_map_guest_page(vgpu, gfn, page_size, &dma_addr); - if (ret) - return -ENXIO; - + /* Successfully shadowed a 4K or 2M page (without splitting). */ pte_ops->set_pfn(&se, dma_addr >> PAGE_SHIFT); ppgtt_set_shadow_entry(spt, &se, index); return 0; @@ -1329,11 +1286,9 @@ static int ppgtt_populate_shadow_entry(struct intel_vgpu *vgpu, static int ppgtt_populate_spt(struct intel_vgpu_ppgtt_spt *spt) { struct intel_vgpu *vgpu = spt->vgpu; - struct intel_gvt *gvt = vgpu->gvt; - const struct intel_gvt_gtt_pte_ops *ops = gvt->gtt.pte_ops; struct intel_vgpu_ppgtt_spt *s; struct intel_gvt_gtt_entry se, ge; - unsigned long gfn, i; + unsigned long i; int ret; trace_spt_change(spt->vgpu->id, "born", spt, @@ -1350,13 +1305,6 @@ static int ppgtt_populate_spt(struct intel_vgpu_ppgtt_spt *spt) ppgtt_generate_shadow_entry(&se, s, &ge); ppgtt_set_shadow_entry(spt, &se, i); } else { - gfn = ops->get_pfn(&ge); - if (!intel_gvt_is_valid_gfn(vgpu, gfn)) { - ops->set_pfn(&se, gvt->gtt.scratch_mfn); - ppgtt_set_shadow_entry(spt, &se, i); - continue; - } - ret = ppgtt_populate_shadow_entry(vgpu, spt, i, &ge); if (ret) goto fail; @@ -1845,6 +1793,9 @@ static int shadow_ppgtt_mm(struct intel_vgpu_mm *mm) if (mm->ppgtt_mm.shadowed) return 0; + if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status)) + return -EINVAL; + mm->ppgtt_mm.shadowed = true; for (index = 0; index < ARRAY_SIZE(mm->ppgtt_mm.guest_pdps); index++) { @@ -2331,14 +2282,6 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, m.val64 = e.val64; m.type = e.type; - /* one PTE update may be issued in multiple writes and the - * first write may not construct a valid gfn - */ - if (!intel_gvt_is_valid_gfn(vgpu, gfn)) { - ops->set_pfn(&m, gvt->gtt.scratch_mfn); - goto out; - } - ret = intel_gvt_dma_map_guest_page(vgpu, gfn, PAGE_SIZE, &dma_addr); if (ret) { @@ -2355,7 +2298,6 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off, ops->clear_present(&m); } -out: ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index); ggtt_get_host_entry(ggtt_mm, &e, g_gtt_index); @@ -2876,24 +2818,6 @@ void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old) } /** - * intel_vgpu_reset_gtt - reset the all GTT related status - * @vgpu: a vGPU - * - * This function is called from vfio core to reset reset all - * GTT related status, including GGTT, PPGTT, scratch page. - * - */ -void intel_vgpu_reset_gtt(struct intel_vgpu *vgpu) -{ - /* Shadow pages are only created when there is no page - * table tracking data, so remove page tracking data after - * removing the shadow pages. - */ - intel_vgpu_destroy_all_ppgtt_mm(vgpu); - intel_vgpu_reset_ggtt(vgpu, true); -} - -/** * intel_gvt_restore_ggtt - restore all vGPU's ggtt entries * @gvt: intel gvt device * diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h index a3b0f59ec8bd..4cb183e06e95 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.h +++ b/drivers/gpu/drm/i915/gvt/gtt.h @@ -224,7 +224,6 @@ void intel_vgpu_reset_ggtt(struct intel_vgpu *vgpu, bool invalidate_old); void intel_vgpu_invalidate_ppgtt(struct intel_vgpu *vgpu); int intel_gvt_init_gtt(struct intel_gvt *gvt); -void intel_vgpu_reset_gtt(struct intel_vgpu *vgpu); void intel_gvt_clean_gtt(struct intel_gvt *gvt); struct intel_vgpu_mm *intel_gvt_find_ppgtt_mm(struct intel_vgpu *vgpu, diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index 2d65800d8e93..53a0a42a50db 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -34,10 +34,11 @@ #define _GVT_H_ #include <uapi/linux/pci_regs.h> -#include <linux/kvm_host.h> #include <linux/vfio.h> #include <linux/mdev.h> +#include <asm/kvm_page_track.h> + #include "i915_drv.h" #include "intel_gvt.h" diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index de675d799c7d..42ce20e72db7 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -106,12 +106,10 @@ struct gvt_dma { #define vfio_dev_to_vgpu(vfio_dev) \ container_of((vfio_dev), struct intel_vgpu, vfio_device) -static void kvmgt_page_track_write(struct kvm_vcpu *vcpu, gpa_t gpa, - const u8 *val, int len, - struct kvm_page_track_notifier_node *node); -static void kvmgt_page_track_flush_slot(struct kvm *kvm, - struct kvm_memory_slot *slot, - struct kvm_page_track_notifier_node *node); +static void kvmgt_page_track_write(gpa_t gpa, const u8 *val, int len, + struct kvm_page_track_notifier_node *node); +static void kvmgt_page_track_remove_region(gfn_t gfn, unsigned long nr_pages, + struct kvm_page_track_notifier_node *node); static ssize_t intel_vgpu_show_description(struct mdev_type *mtype, char *buf) { @@ -161,8 +159,7 @@ static int gvt_pin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn, if (npage == 0) base_page = cur_page; - else if (base_page + npage != cur_page) { - gvt_vgpu_err("The pages are not continuous\n"); + else if (page_to_pfn(base_page) + npage != page_to_pfn(cur_page)) { ret = -EINVAL; npage++; goto err; @@ -172,7 +169,8 @@ static int gvt_pin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn, *page = base_page; return 0; err: - gvt_unpin_guest_page(vgpu, gfn, npage * PAGE_SIZE); + if (npage) + gvt_unpin_guest_page(vgpu, gfn, npage * PAGE_SIZE); return ret; } @@ -352,6 +350,8 @@ __kvmgt_protect_table_find(struct intel_vgpu *info, gfn_t gfn) { struct kvmgt_pgfn *p, *res = NULL; + lockdep_assert_held(&info->vgpu_lock); + hash_for_each_possible(info->ptable, p, hnode, gfn) { if (gfn == p->gfn) { res = p; @@ -654,21 +654,19 @@ out: static int intel_vgpu_open_device(struct vfio_device *vfio_dev) { struct intel_vgpu *vgpu = vfio_dev_to_vgpu(vfio_dev); - - if (!vgpu->vfio_device.kvm || - vgpu->vfio_device.kvm->mm != current->mm) { - gvt_vgpu_err("KVM is required to use Intel vGPU\n"); - return -ESRCH; - } + int ret; if (__kvmgt_vgpu_exist(vgpu)) return -EEXIST; vgpu->track_node.track_write = kvmgt_page_track_write; - vgpu->track_node.track_flush_slot = kvmgt_page_track_flush_slot; - kvm_get_kvm(vgpu->vfio_device.kvm); - kvm_page_track_register_notifier(vgpu->vfio_device.kvm, - &vgpu->track_node); + vgpu->track_node.track_remove_region = kvmgt_page_track_remove_region; + ret = kvm_page_track_register_notifier(vgpu->vfio_device.kvm, + &vgpu->track_node); + if (ret) { + gvt_vgpu_err("KVM is required to use Intel vGPU\n"); + return ret; + } set_bit(INTEL_VGPU_STATUS_ATTACHED, vgpu->status); @@ -703,7 +701,6 @@ static void intel_vgpu_close_device(struct vfio_device *vfio_dev) kvm_page_track_unregister_notifier(vgpu->vfio_device.kvm, &vgpu->track_node); - kvm_put_kvm(vgpu->vfio_device.kvm); kvmgt_protect_table_destroy(vgpu); gvt_cache_destroy(vgpu); @@ -1474,6 +1471,7 @@ static const struct vfio_device_ops intel_vgpu_dev_ops = { .bind_iommufd = vfio_iommufd_emulated_bind, .unbind_iommufd = vfio_iommufd_emulated_unbind, .attach_ioas = vfio_iommufd_emulated_attach_ioas, + .detach_ioas = vfio_iommufd_emulated_detach_ioas, }; static int intel_vgpu_probe(struct mdev_device *mdev) @@ -1546,95 +1544,70 @@ static struct mdev_driver intel_vgpu_mdev_driver = { int intel_gvt_page_track_add(struct intel_vgpu *info, u64 gfn) { - struct kvm *kvm = info->vfio_device.kvm; - struct kvm_memory_slot *slot; - int idx; + int r; if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, info->status)) return -ESRCH; - idx = srcu_read_lock(&kvm->srcu); - slot = gfn_to_memslot(kvm, gfn); - if (!slot) { - srcu_read_unlock(&kvm->srcu, idx); - return -EINVAL; - } - - write_lock(&kvm->mmu_lock); - if (kvmgt_gfn_is_write_protected(info, gfn)) - goto out; + return 0; - kvm_slot_page_track_add_page(kvm, slot, gfn, KVM_PAGE_TRACK_WRITE); - kvmgt_protect_table_add(info, gfn); + r = kvm_write_track_add_gfn(info->vfio_device.kvm, gfn); + if (r) + return r; -out: - write_unlock(&kvm->mmu_lock); - srcu_read_unlock(&kvm->srcu, idx); + kvmgt_protect_table_add(info, gfn); return 0; } int intel_gvt_page_track_remove(struct intel_vgpu *info, u64 gfn) { - struct kvm *kvm = info->vfio_device.kvm; - struct kvm_memory_slot *slot; - int idx; + int r; if (!test_bit(INTEL_VGPU_STATUS_ATTACHED, info->status)) return -ESRCH; - idx = srcu_read_lock(&kvm->srcu); - slot = gfn_to_memslot(kvm, gfn); - if (!slot) { - srcu_read_unlock(&kvm->srcu, idx); - return -EINVAL; - } - - write_lock(&kvm->mmu_lock); - if (!kvmgt_gfn_is_write_protected(info, gfn)) - goto out; + return 0; - kvm_slot_page_track_remove_page(kvm, slot, gfn, KVM_PAGE_TRACK_WRITE); - kvmgt_protect_table_del(info, gfn); + r = kvm_write_track_remove_gfn(info->vfio_device.kvm, gfn); + if (r) + return r; -out: - write_unlock(&kvm->mmu_lock); - srcu_read_unlock(&kvm->srcu, idx); + kvmgt_protect_table_del(info, gfn); return 0; } -static void kvmgt_page_track_write(struct kvm_vcpu *vcpu, gpa_t gpa, - const u8 *val, int len, - struct kvm_page_track_notifier_node *node) +static void kvmgt_page_track_write(gpa_t gpa, const u8 *val, int len, + struct kvm_page_track_notifier_node *node) { struct intel_vgpu *info = container_of(node, struct intel_vgpu, track_node); - if (kvmgt_gfn_is_write_protected(info, gpa_to_gfn(gpa))) + mutex_lock(&info->vgpu_lock); + + if (kvmgt_gfn_is_write_protected(info, gpa >> PAGE_SHIFT)) intel_vgpu_page_track_handler(info, gpa, (void *)val, len); + + mutex_unlock(&info->vgpu_lock); } -static void kvmgt_page_track_flush_slot(struct kvm *kvm, - struct kvm_memory_slot *slot, - struct kvm_page_track_notifier_node *node) +static void kvmgt_page_track_remove_region(gfn_t gfn, unsigned long nr_pages, + struct kvm_page_track_notifier_node *node) { - int i; - gfn_t gfn; + unsigned long i; struct intel_vgpu *info = container_of(node, struct intel_vgpu, track_node); - write_lock(&kvm->mmu_lock); - for (i = 0; i < slot->npages; i++) { - gfn = slot->base_gfn + i; - if (kvmgt_gfn_is_write_protected(info, gfn)) { - kvm_slot_page_track_remove_page(kvm, slot, gfn, - KVM_PAGE_TRACK_WRITE); - kvmgt_protect_table_del(info, gfn); - } + mutex_lock(&info->vgpu_lock); + + for (i = 0; i < nr_pages; i++) { + if (kvmgt_gfn_is_write_protected(info, gfn + i)) + kvmgt_protect_table_del(info, gfn + i); } - write_unlock(&kvm->mmu_lock); + + mutex_unlock(&info->vgpu_lock); } void intel_vgpu_detach_regions(struct intel_vgpu *vgpu) diff --git a/drivers/gpu/drm/i915/gvt/page_track.c b/drivers/gpu/drm/i915/gvt/page_track.c index df34e73cba41..60a65435556d 100644 --- a/drivers/gpu/drm/i915/gvt/page_track.c +++ b/drivers/gpu/drm/i915/gvt/page_track.c @@ -162,13 +162,9 @@ int intel_vgpu_page_track_handler(struct intel_vgpu *vgpu, u64 gpa, struct intel_vgpu_page_track *page_track; int ret = 0; - mutex_lock(&vgpu->vgpu_lock); - page_track = intel_vgpu_find_page_track(vgpu, gpa >> PAGE_SHIFT); - if (!page_track) { - ret = -ENXIO; - goto out; - } + if (!page_track) + return -ENXIO; if (unlikely(vgpu->failsafe)) { /* Remove write protection to prevent furture traps. */ @@ -179,7 +175,5 @@ int intel_vgpu_page_track_handler(struct intel_vgpu *vgpu, u64 gpa, gvt_err("guest page write error, gpa %llx\n", gpa); } -out: - mutex_unlock(&vgpu->vgpu_lock); return ret; } diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index f4055804aad1..a5c8005ec484 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -974,7 +974,7 @@ static void update_guest_context(struct intel_vgpu_workload *workload) context_page_num = rq->engine->context_size; context_page_num = context_page_num >> PAGE_SHIFT; - if (IS_BROADWELL(rq->engine->i915) && rq->engine->id == RCS0) + if (IS_BROADWELL(rq->i915) && rq->engine->id == RCS0) context_page_num = 19; context_base = (void *) ctx->lrc_reg_state - diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c index 8ef93889061a..5ec293011d99 100644 --- a/drivers/gpu/drm/i915/i915_active.c +++ b/drivers/gpu/drm/i915/i915_active.c @@ -449,8 +449,11 @@ int i915_active_add_request(struct i915_active *ref, struct i915_request *rq) } } while (unlikely(is_barrier(active))); - if (!__i915_active_fence_set(active, fence)) + fence = __i915_active_fence_set(active, fence); + if (!fence) __i915_active_acquire(ref); + else + dma_fence_put(fence); out: i915_active_release(ref); @@ -469,13 +472,9 @@ __i915_active_set_fence(struct i915_active *ref, return NULL; } - rcu_read_lock(); prev = __i915_active_fence_set(active, fence); - if (prev) - prev = dma_fence_get_rcu(prev); - else + if (!prev) __i915_active_acquire(ref); - rcu_read_unlock(); return prev; } @@ -1019,10 +1018,11 @@ void i915_request_add_active_barriers(struct i915_request *rq) * * Records the new @fence as the last active fence along its timeline in * this active tracker, moving the tracking callbacks from the previous - * fence onto this one. Returns the previous fence (if not already completed), - * which the caller must ensure is executed before the new fence. To ensure - * that the order of fences within the timeline of the i915_active_fence is - * understood, it should be locked by the caller. + * fence onto this one. Gets and returns a reference to the previous fence + * (if not already completed), which the caller must put after making sure + * that it is executed before the new fence. To ensure that the order of + * fences within the timeline of the i915_active_fence is understood, it + * should be locked by the caller. */ struct dma_fence * __i915_active_fence_set(struct i915_active_fence *active, @@ -1031,7 +1031,23 @@ __i915_active_fence_set(struct i915_active_fence *active, struct dma_fence *prev; unsigned long flags; - if (fence == rcu_access_pointer(active->fence)) + /* + * In case of fences embedded in i915_requests, their memory is + * SLAB_FAILSAFE_BY_RCU, then it can be reused right after release + * by new requests. Then, there is a risk of passing back a pointer + * to a new, completely unrelated fence that reuses the same memory + * while tracked under a different active tracker. Combined with i915 + * perf open/close operations that build await dependencies between + * engine kernel context requests and user requests from different + * timelines, this can lead to dependency loops and infinite waits. + * + * As a countermeasure, we try to get a reference to the active->fence + * first, so if we succeed and pass it back to our user then it is not + * released and potentially reused by an unrelated request before the + * user has a chance to set up an await dependency on it. + */ + prev = i915_active_fence_get(active); + if (fence == prev) return fence; GEM_BUG_ON(test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &fence->flags)); @@ -1040,27 +1056,56 @@ __i915_active_fence_set(struct i915_active_fence *active, * Consider that we have two threads arriving (A and B), with * C already resident as the active->fence. * - * A does the xchg first, and so it sees C or NULL depending - * on the timing of the interrupt handler. If it is NULL, the - * previous fence must have been signaled and we know that - * we are first on the timeline. If it is still present, - * we acquire the lock on that fence and serialise with the interrupt - * handler, in the process removing it from any future interrupt - * callback. A will then wait on C before executing (if present). - * - * As B is second, it sees A as the previous fence and so waits for - * it to complete its transition and takes over the occupancy for - * itself -- remembering that it needs to wait on A before executing. + * Both A and B have got a reference to C or NULL, depending on the + * timing of the interrupt handler. Let's assume that if A has got C + * then it has locked C first (before B). * * Note the strong ordering of the timeline also provides consistent * nesting rules for the fence->lock; the inner lock is always the * older lock. */ spin_lock_irqsave(fence->lock, flags); - prev = xchg(__active_fence_slot(active), fence); - if (prev) { - GEM_BUG_ON(prev == fence); + if (prev) spin_lock_nested(prev->lock, SINGLE_DEPTH_NESTING); + + /* + * A does the cmpxchg first, and so it sees C or NULL, as before, or + * something else, depending on the timing of other threads and/or + * interrupt handler. If not the same as before then A unlocks C if + * applicable and retries, starting from an attempt to get a new + * active->fence. Meanwhile, B follows the same path as A. + * Once A succeeds with cmpxch, B fails again, retires, gets A from + * active->fence, locks it as soon as A completes, and possibly + * succeeds with cmpxchg. + */ + while (cmpxchg(__active_fence_slot(active), prev, fence) != prev) { + if (prev) { + spin_unlock(prev->lock); + dma_fence_put(prev); + } + spin_unlock_irqrestore(fence->lock, flags); + + prev = i915_active_fence_get(active); + GEM_BUG_ON(prev == fence); + + spin_lock_irqsave(fence->lock, flags); + if (prev) + spin_lock_nested(prev->lock, SINGLE_DEPTH_NESTING); + } + + /* + * If prev is NULL then the previous fence must have been signaled + * and we know that we are first on the timeline. If it is still + * present then, having the lock on that fence already acquired, we + * serialise with the interrupt handler, in the process of removing it + * from any future interrupt callback. A will then wait on C before + * executing (if present). + * + * As B is second, it sees A as the previous fence and so waits for + * it to complete its transition and takes over the occupancy for + * itself -- remembering that it needs to wait on A before executing. + */ + if (prev) { __list_del_entry(&active->cb.node); spin_unlock(prev->lock); /* serialise with prev->cb_list */ } @@ -1077,11 +1122,7 @@ int i915_active_fence_set(struct i915_active_fence *active, int err = 0; /* Must maintain timeline ordering wrt previous active requests */ - rcu_read_lock(); fence = __i915_active_fence_set(active, &rq->fence); - if (fence) /* but the previous fence may not belong to that timeline! */ - fence = dma_fence_get_rcu(fence); - rcu_read_unlock(); if (fence) { err = i915_request_await_dma_fence(rq, fence); dma_fence_put(fence); diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 76ccd4e03e31..4de44cf1026d 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -67,6 +67,7 @@ static int i915_capabilities(struct seq_file *m, void *data) seq_printf(m, "pch: %d\n", INTEL_PCH_TYPE(i915)); intel_device_info_print(INTEL_INFO(i915), RUNTIME_INFO(i915), &p); + intel_display_device_info_print(DISPLAY_INFO(i915), DISPLAY_RUNTIME_INFO(i915), &p); i915_print_iommu_status(i915, &p); intel_gt_info_print(&to_gt(i915)->info, &p); intel_driver_caps_print(&i915->caps, &p); diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index 222d0a1f3b55..ec4d26b3c17c 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -175,7 +175,7 @@ static void intel_detect_preproduction_hw(struct drm_i915_private *dev_priv) { bool pre = false; - pre |= IS_HSW_EARLY_SDV(dev_priv); + pre |= IS_HASWELL_EARLY_SDV(dev_priv); pre |= IS_SKYLAKE(dev_priv) && INTEL_REVID(dev_priv) < 0x6; pre |= IS_BROXTON(dev_priv) && INTEL_REVID(dev_priv) < 0xA; pre |= IS_KABYLAKE(dev_priv) && INTEL_REVID(dev_priv) < 0x1; @@ -443,7 +443,6 @@ static int i915_pcode_init(struct drm_i915_private *i915) static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) { struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); - struct pci_dev *root_pdev; int ret; if (i915_inject_probe_failure(dev_priv)) @@ -557,15 +556,6 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) intel_bw_init_hw(dev_priv); - /* - * FIXME: Temporary hammer to avoid freezing the machine on our DGFX - * This should be totally removed when we handle the pci states properly - * on runtime PM and on s2idle cases. - */ - root_pdev = pcie_find_root_port(pdev); - if (root_pdev) - pci_d3cold_disable(root_pdev); - return 0; err_opregion: @@ -591,7 +581,6 @@ err_perf: static void i915_driver_hw_remove(struct drm_i915_private *dev_priv) { struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); - struct pci_dev *root_pdev; i915_perf_fini(dev_priv); @@ -599,10 +588,6 @@ static void i915_driver_hw_remove(struct drm_i915_private *dev_priv) if (pdev->msi_enabled) pci_disable_msi(pdev); - - root_pdev = pcie_find_root_port(pdev); - if (root_pdev) - pci_d3cold_enable(root_pdev); } /** @@ -711,6 +696,8 @@ static void i915_welcome_messages(struct drm_i915_private *dev_priv) intel_device_info_print(INTEL_INFO(dev_priv), RUNTIME_INFO(dev_priv), &p); + intel_display_device_info_print(DISPLAY_INFO(dev_priv), + DISPLAY_RUNTIME_INFO(dev_priv), &p); i915_print_iommu_status(dev_priv, &p); for_each_gt(gt, dev_priv, i) intel_gt_info_print(>->info, &p); @@ -1517,6 +1504,8 @@ static int intel_runtime_suspend(struct device *kdev) { struct drm_i915_private *dev_priv = kdev_to_i915(kdev); struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; + struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); + struct pci_dev *root_pdev; struct intel_gt *gt; int ret, i; @@ -1568,6 +1557,15 @@ static int intel_runtime_suspend(struct device *kdev) drm_err(&dev_priv->drm, "Unclaimed access detected prior to suspending\n"); + /* + * FIXME: Temporary hammer to avoid freezing the machine on our DGFX + * This should be totally removed when we handle the pci states properly + * on runtime PM. + */ + root_pdev = pcie_find_root_port(pdev); + if (root_pdev) + pci_d3cold_disable(root_pdev); + rpm->suspended = true; /* @@ -1606,6 +1604,8 @@ static int intel_runtime_resume(struct device *kdev) { struct drm_i915_private *dev_priv = kdev_to_i915(kdev); struct intel_runtime_pm *rpm = &dev_priv->runtime_pm; + struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); + struct pci_dev *root_pdev; struct intel_gt *gt; int ret, i; @@ -1619,6 +1619,11 @@ static int intel_runtime_resume(struct device *kdev) intel_opregion_notify_adapter(dev_priv, PCI_D0); rpm->suspended = false; + + root_pdev = pcie_find_root_port(pdev); + if (root_pdev) + pci_d3cold_enable(root_pdev); + if (intel_uncore_unclaimed_mmio(&dev_priv->uncore)) drm_dbg(&dev_priv->drm, "Unclaimed access during suspend, bios?\n"); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index b4cf6f0f636d..7a8ce7239bc9 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -203,9 +203,8 @@ struct drm_i915_private { /* i915 device parameters */ struct i915_params params; - const struct intel_device_info __info; /* Use INTEL_INFO() to access. */ + const struct intel_device_info *__info; /* Use INTEL_INFO() to access. */ struct intel_runtime_info __runtime; /* Use RUNTIME_INFO() to access. */ - struct intel_display_runtime_info __display_runtime; /* Access with DISPLAY_RUNTIME_INFO() */ struct intel_driver_caps caps; struct i915_dsm dsm; @@ -324,7 +323,6 @@ struct drm_i915_private { /* * i915->gt[0] == &i915->gt0 */ -#define I915_MAX_GT 2 struct intel_gt *gt[I915_MAX_GT]; struct kobject *sysfs_gt; @@ -416,10 +414,10 @@ static inline struct intel_gt *to_gt(struct drm_i915_private *i915) (engine__) && (engine__)->uabi_class == (class__); \ (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node))) -#define INTEL_INFO(i915) (&(i915)->__info) -#define DISPLAY_INFO(i915) (INTEL_INFO(i915)->display) +#define INTEL_INFO(i915) ((i915)->__info) #define RUNTIME_INFO(i915) (&(i915)->__runtime) -#define DISPLAY_RUNTIME_INFO(i915) (&(i915)->__display_runtime) +#define DISPLAY_INFO(i915) ((i915)->display.info.__device_info) +#define DISPLAY_RUNTIME_INFO(i915) (&(i915)->display.info.__runtime_info) #define DRIVER_CAPS(i915) (&(i915)->caps) #define INTEL_DEVID(i915) (RUNTIME_INFO(i915)->device_id) @@ -563,8 +561,8 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define IS_COFFEELAKE(i915) IS_PLATFORM(i915, INTEL_COFFEELAKE) #define IS_COMETLAKE(i915) IS_PLATFORM(i915, INTEL_COMETLAKE) #define IS_ICELAKE(i915) IS_PLATFORM(i915, INTEL_ICELAKE) -#define IS_JSL_EHL(i915) (IS_PLATFORM(i915, INTEL_JASPERLAKE) || \ - IS_PLATFORM(i915, INTEL_ELKHARTLAKE)) +#define IS_JASPERLAKE(i915) IS_PLATFORM(i915, INTEL_JASPERLAKE) +#define IS_ELKHARTLAKE(i915) IS_PLATFORM(i915, INTEL_ELKHARTLAKE) #define IS_TIGERLAKE(i915) IS_PLATFORM(i915, INTEL_TIGERLAKE) #define IS_ROCKETLAKE(i915) IS_PLATFORM(i915, INTEL_ROCKETLAKE) #define IS_DG1(i915) IS_PLATFORM(i915, INTEL_DG1) @@ -585,105 +583,77 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, IS_SUBPLATFORM(i915, INTEL_DG2, INTEL_SUBPLATFORM_G11) #define IS_DG2_G12(i915) \ IS_SUBPLATFORM(i915, INTEL_DG2, INTEL_SUBPLATFORM_G12) -#define IS_ADLS_RPLS(i915) \ +#define IS_RAPTORLAKE_S(i915) \ IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_S, INTEL_SUBPLATFORM_RPL) -#define IS_ADLP_N(i915) \ +#define IS_ALDERLAKE_P_N(i915) \ IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_N) -#define IS_ADLP_RPLP(i915) \ +#define IS_RAPTORLAKE_P(i915) \ IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_RPL) -#define IS_ADLP_RPLU(i915) \ +#define IS_RAPTORLAKE_U(i915) \ IS_SUBPLATFORM(i915, INTEL_ALDERLAKE_P, INTEL_SUBPLATFORM_RPLU) -#define IS_HSW_EARLY_SDV(i915) (IS_HASWELL(i915) && \ +#define IS_HASWELL_EARLY_SDV(i915) (IS_HASWELL(i915) && \ (INTEL_DEVID(i915) & 0xFF00) == 0x0C00) -#define IS_BDW_ULT(i915) \ +#define IS_BROADWELL_ULT(i915) \ IS_SUBPLATFORM(i915, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULT) -#define IS_BDW_ULX(i915) \ +#define IS_BROADWELL_ULX(i915) \ IS_SUBPLATFORM(i915, INTEL_BROADWELL, INTEL_SUBPLATFORM_ULX) -#define IS_BDW_GT3(i915) (IS_BROADWELL(i915) && \ +#define IS_BROADWELL_GT3(i915) (IS_BROADWELL(i915) && \ INTEL_INFO(i915)->gt == 3) -#define IS_HSW_ULT(i915) \ +#define IS_HASWELL_ULT(i915) \ IS_SUBPLATFORM(i915, INTEL_HASWELL, INTEL_SUBPLATFORM_ULT) -#define IS_HSW_GT3(i915) (IS_HASWELL(i915) && \ +#define IS_HASWELL_GT3(i915) (IS_HASWELL(i915) && \ INTEL_INFO(i915)->gt == 3) -#define IS_HSW_GT1(i915) (IS_HASWELL(i915) && \ +#define IS_HASWELL_GT1(i915) (IS_HASWELL(i915) && \ INTEL_INFO(i915)->gt == 1) /* ULX machines are also considered ULT. */ -#define IS_HSW_ULX(i915) \ +#define IS_HASWELL_ULX(i915) \ IS_SUBPLATFORM(i915, INTEL_HASWELL, INTEL_SUBPLATFORM_ULX) -#define IS_SKL_ULT(i915) \ +#define IS_SKYLAKE_ULT(i915) \ IS_SUBPLATFORM(i915, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULT) -#define IS_SKL_ULX(i915) \ +#define IS_SKYLAKE_ULX(i915) \ IS_SUBPLATFORM(i915, INTEL_SKYLAKE, INTEL_SUBPLATFORM_ULX) -#define IS_KBL_ULT(i915) \ +#define IS_KABYLAKE_ULT(i915) \ IS_SUBPLATFORM(i915, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULT) -#define IS_KBL_ULX(i915) \ +#define IS_KABYLAKE_ULX(i915) \ IS_SUBPLATFORM(i915, INTEL_KABYLAKE, INTEL_SUBPLATFORM_ULX) -#define IS_SKL_GT2(i915) (IS_SKYLAKE(i915) && \ +#define IS_SKYLAKE_GT2(i915) (IS_SKYLAKE(i915) && \ INTEL_INFO(i915)->gt == 2) -#define IS_SKL_GT3(i915) (IS_SKYLAKE(i915) && \ +#define IS_SKYLAKE_GT3(i915) (IS_SKYLAKE(i915) && \ INTEL_INFO(i915)->gt == 3) -#define IS_SKL_GT4(i915) (IS_SKYLAKE(i915) && \ +#define IS_SKYLAKE_GT4(i915) (IS_SKYLAKE(i915) && \ INTEL_INFO(i915)->gt == 4) -#define IS_KBL_GT2(i915) (IS_KABYLAKE(i915) && \ +#define IS_KABYLAKE_GT2(i915) (IS_KABYLAKE(i915) && \ INTEL_INFO(i915)->gt == 2) -#define IS_KBL_GT3(i915) (IS_KABYLAKE(i915) && \ +#define IS_KABYLAKE_GT3(i915) (IS_KABYLAKE(i915) && \ INTEL_INFO(i915)->gt == 3) -#define IS_CFL_ULT(i915) \ +#define IS_COFFEELAKE_ULT(i915) \ IS_SUBPLATFORM(i915, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULT) -#define IS_CFL_ULX(i915) \ +#define IS_COFFEELAKE_ULX(i915) \ IS_SUBPLATFORM(i915, INTEL_COFFEELAKE, INTEL_SUBPLATFORM_ULX) -#define IS_CFL_GT2(i915) (IS_COFFEELAKE(i915) && \ +#define IS_COFFEELAKE_GT2(i915) (IS_COFFEELAKE(i915) && \ INTEL_INFO(i915)->gt == 2) -#define IS_CFL_GT3(i915) (IS_COFFEELAKE(i915) && \ +#define IS_COFFEELAKE_GT3(i915) (IS_COFFEELAKE(i915) && \ INTEL_INFO(i915)->gt == 3) -#define IS_CML_ULT(i915) \ +#define IS_COMETLAKE_ULT(i915) \ IS_SUBPLATFORM(i915, INTEL_COMETLAKE, INTEL_SUBPLATFORM_ULT) -#define IS_CML_ULX(i915) \ +#define IS_COMETLAKE_ULX(i915) \ IS_SUBPLATFORM(i915, INTEL_COMETLAKE, INTEL_SUBPLATFORM_ULX) -#define IS_CML_GT2(i915) (IS_COMETLAKE(i915) && \ +#define IS_COMETLAKE_GT2(i915) (IS_COMETLAKE(i915) && \ INTEL_INFO(i915)->gt == 2) #define IS_ICL_WITH_PORT_F(i915) \ IS_SUBPLATFORM(i915, INTEL_ICELAKE, INTEL_SUBPLATFORM_PORTF) -#define IS_TGL_UY(i915) \ +#define IS_TIGERLAKE_UY(i915) \ IS_SUBPLATFORM(i915, INTEL_TIGERLAKE, INTEL_SUBPLATFORM_UY) -#define IS_SKL_GRAPHICS_STEP(p, since, until) (IS_SKYLAKE(p) && IS_GRAPHICS_STEP(p, since, until)) -#define IS_KBL_GRAPHICS_STEP(i915, since, until) \ - (IS_KABYLAKE(i915) && IS_GRAPHICS_STEP(i915, since, until)) -#define IS_KBL_DISPLAY_STEP(i915, since, until) \ - (IS_KABYLAKE(i915) && IS_DISPLAY_STEP(i915, since, until)) -#define IS_JSL_EHL_GRAPHICS_STEP(p, since, until) \ - (IS_JSL_EHL(p) && IS_GRAPHICS_STEP(p, since, until)) -#define IS_JSL_EHL_DISPLAY_STEP(p, since, until) \ - (IS_JSL_EHL(p) && IS_DISPLAY_STEP(p, since, until)) -#define IS_TGL_DISPLAY_STEP(__i915, since, until) \ - (IS_TIGERLAKE(__i915) && \ - IS_DISPLAY_STEP(__i915, since, until)) - -#define IS_RKL_DISPLAY_STEP(p, since, until) \ - (IS_ROCKETLAKE(p) && IS_DISPLAY_STEP(p, since, until)) -#define IS_ADLS_DISPLAY_STEP(__i915, since, until) \ - (IS_ALDERLAKE_S(__i915) && \ - IS_DISPLAY_STEP(__i915, since, until)) -#define IS_ADLS_GRAPHICS_STEP(__i915, since, until) \ - (IS_ALDERLAKE_S(__i915) && \ - IS_GRAPHICS_STEP(__i915, since, until)) -#define IS_ADLP_DISPLAY_STEP(__i915, since, until) \ - (IS_ALDERLAKE_P(__i915) && \ - IS_DISPLAY_STEP(__i915, since, until)) - -#define IS_ADLP_GRAPHICS_STEP(__i915, since, until) \ - (IS_ALDERLAKE_P(__i915) && \ - IS_GRAPHICS_STEP(__i915, since, until)) #define IS_XEHPSDV_GRAPHICS_STEP(__i915, since, until) \ (IS_XEHPSDV(__i915) && IS_GRAPHICS_STEP(__i915, since, until)) @@ -801,7 +771,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, /* WaRsDisableCoarsePowerGating:skl,cnl */ #define NEEDS_WaRsDisableCoarsePowerGating(i915) \ - (IS_SKL_GT3(i915) || IS_SKL_GT4(i915)) + (IS_SKYLAKE_GT3(i915) || IS_SKYLAKE_GT4(i915)) /* With the 945 and later, Y tiling got adjusted so that it was 32 128-byte * rows, which changed the alignment requirements and fence programming. @@ -839,7 +809,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, */ #define HAS_64K_PAGES(i915) (INTEL_INFO(i915)->has_64k_pages) -#define HAS_REGION(i915, i) (RUNTIME_INFO(i915)->memory_regions & (i)) +#define HAS_REGION(i915, i) (INTEL_INFO(i915)->memory_regions & (i)) #define HAS_LMEM(i915) HAS_REGION(i915, REGION_LMEM) #define HAS_EXTRA_GT_LIST(i915) (INTEL_INFO(i915)->extra_gt_list) @@ -862,7 +832,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, /* DPF == dynamic parity feature */ #define HAS_L3_DPF(i915) (INTEL_INFO(i915)->has_l3_dpf) -#define NUM_L3_SLICES(i915) (IS_HSW_GT3(i915) ? \ +#define NUM_L3_SLICES(i915) (IS_HASWELL_GT3(i915) ? \ 2 : HAS_L3_DPF(i915)) /* Only valid when HAS_DISPLAY() is true */ diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 0c38bfb60c9a..4008bb09fdb5 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -649,6 +649,8 @@ static void err_print_capabilities(struct drm_i915_error_state_buf *m, struct drm_printer p = i915_error_printer(m); intel_device_info_print(&error->device_info, &error->runtime_info, &p); + intel_display_device_info_print(&error->display_device_info, + &error->display_runtime_info, &p); intel_driver_caps_print(&error->driver_caps, &p); } @@ -1173,9 +1175,9 @@ i915_vma_coredump_create(const struct intel_gt *gt, drm_clflush_pages(&page, 1); - s = kmap(page); + s = kmap_local_page(page); ret = compress_page(compress, s, dst, false); - kunmap(page); + kunmap_local(s); drm_clflush_pages(&page, 1); @@ -1983,6 +1985,10 @@ static void capture_gen(struct i915_gpu_coredump *error) memcpy(&error->runtime_info, RUNTIME_INFO(i915), sizeof(error->runtime_info)); + memcpy(&error->display_device_info, DISPLAY_INFO(i915), + sizeof(error->display_device_info)); + memcpy(&error->display_runtime_info, DISPLAY_RUNTIME_INFO(i915), + sizeof(error->display_runtime_info)); error->driver_caps = i915->caps; } diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h index a78c061ce26f..9f5971f5e980 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.h +++ b/drivers/gpu/drm/i915/i915_gpu_error.h @@ -14,6 +14,7 @@ #include <drm/drm_mm.h> +#include "display/intel_display_device.h" #include "gt/intel_engine.h" #include "gt/intel_gt_types.h" #include "gt/uc/intel_uc_fw.h" @@ -209,6 +210,8 @@ struct i915_gpu_coredump { struct intel_device_info device_info; struct intel_runtime_info runtime_info; + struct intel_display_device_info display_device_info; + struct intel_display_runtime_info display_runtime_info; struct intel_driver_caps driver_caps; struct i915_params params; diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 82fbabcdd7a5..1bfcfbe6e30b 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -423,7 +423,7 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg) static irqreturn_t ilk_irq_handler(int irq, void *arg) { struct drm_i915_private *i915 = arg; - void __iomem * const regs = i915->uncore.regs; + void __iomem * const regs = intel_uncore_regs(&i915->uncore); u32 de_iir, gt_iir, de_ier, sde_ier = 0; irqreturn_t ret = IRQ_NONE; @@ -511,7 +511,7 @@ static inline void gen8_master_intr_enable(void __iomem * const regs) static irqreturn_t gen8_irq_handler(int irq, void *arg) { struct drm_i915_private *dev_priv = arg; - void __iomem * const regs = dev_priv->uncore.regs; + void __iomem * const regs = intel_uncore_regs(&dev_priv->uncore); u32 master_ctl; if (!intel_irqs_enabled(dev_priv)) @@ -561,7 +561,7 @@ static inline void gen11_master_intr_enable(void __iomem * const regs) static irqreturn_t gen11_irq_handler(int irq, void *arg) { struct drm_i915_private *i915 = arg; - void __iomem * const regs = i915->uncore.regs; + void __iomem * const regs = intel_uncore_regs(&i915->uncore); struct intel_gt *gt = to_gt(i915); u32 master_ctl; u32 gu_misc_iir; @@ -619,7 +619,7 @@ static irqreturn_t dg1_irq_handler(int irq, void *arg) { struct drm_i915_private * const i915 = arg; struct intel_gt *gt = to_gt(i915); - void __iomem * const regs = gt->uncore->regs; + void __iomem * const regs = intel_uncore_regs(gt->uncore); u32 master_tile_ctl, master_ctl; u32 gu_misc_iir; @@ -711,7 +711,7 @@ static void gen8_irq_reset(struct drm_i915_private *dev_priv) { struct intel_uncore *uncore = &dev_priv->uncore; - gen8_master_intr_disable(uncore->regs); + gen8_master_intr_disable(intel_uncore_regs(uncore)); gen8_gt_irq_reset(to_gt(dev_priv)); gen8_display_irq_reset(dev_priv); @@ -727,7 +727,7 @@ static void gen11_irq_reset(struct drm_i915_private *dev_priv) struct intel_gt *gt = to_gt(dev_priv); struct intel_uncore *uncore = gt->uncore; - gen11_master_intr_disable(dev_priv->uncore.regs); + gen11_master_intr_disable(intel_uncore_regs(&dev_priv->uncore)); gen11_gt_irq_reset(gt); gen11_display_irq_reset(dev_priv); @@ -742,7 +742,7 @@ static void dg1_irq_reset(struct drm_i915_private *dev_priv) struct intel_gt *gt; unsigned int i; - dg1_master_intr_disable(dev_priv->uncore.regs); + dg1_master_intr_disable(intel_uncore_regs(&dev_priv->uncore)); for_each_gt(gt, dev_priv, i) gen11_gt_irq_reset(gt); @@ -772,45 +772,9 @@ static void cherryview_irq_reset(struct drm_i915_private *dev_priv) static void ilk_irq_postinstall(struct drm_i915_private *dev_priv) { - struct intel_uncore *uncore = &dev_priv->uncore; - u32 display_mask, extra_mask; - - if (GRAPHICS_VER(dev_priv) >= 7) { - display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE_IVB | - DE_PCH_EVENT_IVB | DE_AUX_CHANNEL_A_IVB); - extra_mask = (DE_PIPEC_VBLANK_IVB | DE_PIPEB_VBLANK_IVB | - DE_PIPEA_VBLANK_IVB | DE_ERR_INT_IVB | - DE_PLANE_FLIP_DONE_IVB(PLANE_C) | - DE_PLANE_FLIP_DONE_IVB(PLANE_B) | - DE_PLANE_FLIP_DONE_IVB(PLANE_A) | - DE_DP_A_HOTPLUG_IVB); - } else { - display_mask = (DE_MASTER_IRQ_CONTROL | DE_GSE | DE_PCH_EVENT | - DE_AUX_CHANNEL_A | DE_PIPEB_CRC_DONE | - DE_PIPEA_CRC_DONE | DE_POISON); - extra_mask = (DE_PIPEA_VBLANK | DE_PIPEB_VBLANK | - DE_PIPEB_FIFO_UNDERRUN | DE_PIPEA_FIFO_UNDERRUN | - DE_PLANE_FLIP_DONE(PLANE_A) | - DE_PLANE_FLIP_DONE(PLANE_B) | - DE_DP_A_HOTPLUG); - } - - if (IS_HASWELL(dev_priv)) { - gen3_assert_iir_is_zero(uncore, EDP_PSR_IIR); - display_mask |= DE_EDP_PSR_INT_HSW; - } - - if (IS_IRONLAKE_M(dev_priv)) - extra_mask |= DE_PCU_EVENT; - - dev_priv->irq_mask = ~display_mask; - - ibx_irq_postinstall(dev_priv); - gen5_gt_irq_postinstall(to_gt(dev_priv)); - GEN3_IRQ_INIT(uncore, DE, dev_priv->irq_mask, - display_mask | extra_mask); + ilk_de_irq_postinstall(dev_priv); } static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv) @@ -828,15 +792,10 @@ static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv) static void gen8_irq_postinstall(struct drm_i915_private *dev_priv) { - if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - icp_irq_postinstall(dev_priv); - else if (HAS_PCH_SPLIT(dev_priv)) - ibx_irq_postinstall(dev_priv); - gen8_gt_irq_postinstall(to_gt(dev_priv)); gen8_de_irq_postinstall(dev_priv); - gen8_master_intr_enable(dev_priv->uncore.regs); + gen8_master_intr_enable(intel_uncore_regs(&dev_priv->uncore)); } static void gen11_irq_postinstall(struct drm_i915_private *dev_priv) @@ -845,15 +804,12 @@ static void gen11_irq_postinstall(struct drm_i915_private *dev_priv) struct intel_uncore *uncore = gt->uncore; u32 gu_misc_masked = GEN11_GU_MISC_GSE; - if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) - icp_irq_postinstall(dev_priv); - gen11_gt_irq_postinstall(gt); gen11_de_irq_postinstall(dev_priv); GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked); - gen11_master_intr_enable(uncore->regs); + gen11_master_intr_enable(intel_uncore_regs(uncore)); intel_uncore_posting_read(&dev_priv->uncore, GEN11_GFX_MSTR_IRQ); } @@ -869,18 +825,9 @@ static void dg1_irq_postinstall(struct drm_i915_private *dev_priv) GEN3_IRQ_INIT(uncore, GEN11_GU_MISC_, ~gu_misc_masked, gu_misc_masked); - if (HAS_DISPLAY(dev_priv)) { - if (DISPLAY_VER(dev_priv) >= 14) - mtp_irq_postinstall(dev_priv); - else - icp_irq_postinstall(dev_priv); + dg1_de_irq_postinstall(dev_priv); - gen8_de_irq_postinstall(dev_priv); - intel_uncore_write(&dev_priv->uncore, GEN11_DISPLAY_INT_CTL, - GEN11_DISPLAY_IRQ_ENABLE); - } - - dg1_master_intr_enable(uncore->regs); + dg1_master_intr_enable(intel_uncore_regs(uncore)); intel_uncore_posting_read(uncore, DG1_MSTR_TILE_INTR); } @@ -1343,23 +1290,6 @@ void intel_irq_init(struct drm_i915_private *dev_priv) /* pre-gen11 the guc irqs bits are in the upper 16 bits of the pm reg */ if (HAS_GT_UC(dev_priv) && GRAPHICS_VER(dev_priv) < 11) to_gt(dev_priv)->pm_guc_events = GUC_INTR_GUC2HOST << 16; - - if (!HAS_DISPLAY(dev_priv)) - return; - - dev_priv->drm.vblank_disable_immediate = true; - - /* Most platforms treat the display irq block as an always-on - * power domain. vlv/chv can disable it at runtime and need - * special care to avoid writing any of the display block registers - * outside of the power domain. We defer setting up the display irqs - * in this case to the runtime pm. - */ - dev_priv->display_irqs_enabled = true; - if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - dev_priv->display_irqs_enabled = false; - - intel_hotplug_irq_init(dev_priv); } /** diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 3d7a5db9833b..fcacdc21643c 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -38,9 +38,6 @@ #include "i915_reg.h" #include "intel_pci_config.h" -__diag_push(); -__diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); - #define PLATFORM(x) .platform = (x) #define GEN(x) \ .__runtime.graphics.ip.ver = (x), \ @@ -84,7 +81,7 @@ __diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); .__runtime.page_sizes = I915_GTT_PAGE_SIZE_4K #define GEN_DEFAULT_REGIONS \ - .__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_SMEM + .memory_regions = REGION_SMEM | REGION_STOLEN_SMEM #define I830_FEATURES \ GEN(2), \ @@ -93,7 +90,7 @@ __diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); .has_3d_pipeline = 1, \ .hws_needs_physical = 1, \ .unfenced_needs_alignment = 1, \ - .__runtime.platform_engine_mask = BIT(RCS0), \ + .platform_engine_mask = BIT(RCS0), \ .has_snoop = true, \ .has_coherent_ggtt = false, \ .dma_mask_size = 32, \ @@ -108,7 +105,7 @@ __diag_ignore_all("-Woverride-init", "Allow overriding inherited members"); .gpu_reset_clobbers_display = true, \ .hws_needs_physical = 1, \ .unfenced_needs_alignment = 1, \ - .__runtime.platform_engine_mask = BIT(RCS0), \ + .platform_engine_mask = BIT(RCS0), \ .has_snoop = true, \ .has_coherent_ggtt = false, \ .dma_mask_size = 32, \ @@ -140,7 +137,7 @@ static const struct intel_device_info i865g_info = { #define GEN3_FEATURES \ GEN(3), \ .gpu_reset_clobbers_display = true, \ - .__runtime.platform_engine_mask = BIT(RCS0), \ + .platform_engine_mask = BIT(RCS0), \ .has_3d_pipeline = 1, \ .has_snoop = true, \ .has_coherent_ggtt = true, \ @@ -203,7 +200,7 @@ static const struct intel_device_info pnv_m_info = { #define GEN4_FEATURES \ GEN(4), \ .gpu_reset_clobbers_display = true, \ - .__runtime.platform_engine_mask = BIT(RCS0), \ + .platform_engine_mask = BIT(RCS0), \ .has_3d_pipeline = 1, \ .has_snoop = true, \ .has_coherent_ggtt = true, \ @@ -231,7 +228,7 @@ static const struct intel_device_info i965gm_info = { static const struct intel_device_info g45_info = { GEN4_FEATURES, PLATFORM(INTEL_G45), - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), + .platform_engine_mask = BIT(RCS0) | BIT(VCS0), .gpu_reset_clobbers_display = false, }; @@ -239,13 +236,13 @@ static const struct intel_device_info gm45_info = { GEN4_FEATURES, PLATFORM(INTEL_GM45), .is_mobile = 1, - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), + .platform_engine_mask = BIT(RCS0) | BIT(VCS0), .gpu_reset_clobbers_display = false, }; #define GEN5_FEATURES \ GEN(5), \ - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0), \ + .platform_engine_mask = BIT(RCS0) | BIT(VCS0), \ .has_3d_pipeline = 1, \ .has_snoop = true, \ .has_coherent_ggtt = true, \ @@ -271,7 +268,7 @@ static const struct intel_device_info ilk_m_info = { #define GEN6_FEATURES \ GEN(6), \ - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ .has_3d_pipeline = 1, \ .has_coherent_ggtt = true, \ .has_llc = 1, \ @@ -319,7 +316,7 @@ static const struct intel_device_info snb_m_gt2_info = { #define GEN7_FEATURES \ GEN(7), \ - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), \ .has_3d_pipeline = 1, \ .has_coherent_ggtt = true, \ .has_llc = 1, \ @@ -387,7 +384,7 @@ static const struct intel_device_info vlv_info = { .__runtime.ppgtt_size = 31, .has_snoop = true, .has_coherent_ggtt = false, - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0), GEN_DEFAULT_PAGE_SIZES, GEN_DEFAULT_REGIONS, LEGACY_CACHELEVEL, @@ -395,7 +392,7 @@ static const struct intel_device_info vlv_info = { #define G75_FEATURES \ GEN7_FEATURES, \ - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \ + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \ .has_rc6p = 0 /* RC6p removed-by HSW */, \ .has_runtime_pm = 1 @@ -453,7 +450,7 @@ static const struct intel_device_info bdw_rsvd_info = { static const struct intel_device_info bdw_gt3_info = { BDW_PLATFORM, .gt = 3, - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1), }; @@ -461,7 +458,7 @@ static const struct intel_device_info chv_info = { PLATFORM(INTEL_CHERRYVIEW), GEN(8), .is_lp = 1, - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), .has_64bit_reloc = 1, .has_runtime_pm = 1, .has_rc6 = 1, @@ -505,7 +502,7 @@ static const struct intel_device_info skl_gt2_info = { #define SKL_GT3_PLUS_PLATFORM \ SKL_PLATFORM, \ - .__runtime.platform_engine_mask = \ + .platform_engine_mask = \ BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1) @@ -522,7 +519,7 @@ static const struct intel_device_info skl_gt4_info = { #define GEN9_LP_FEATURES \ GEN(9), \ .is_lp = 1, \ - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \ + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0), \ .has_3d_pipeline = 1, \ .has_64bit_reloc = 1, \ .has_runtime_pm = 1, \ @@ -568,7 +565,7 @@ static const struct intel_device_info kbl_gt2_info = { static const struct intel_device_info kbl_gt3_info = { KBL_PLATFORM, .gt = 3, - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1), }; @@ -589,7 +586,7 @@ static const struct intel_device_info cfl_gt2_info = { static const struct intel_device_info cfl_gt3_info = { CFL_PLATFORM, .gt = 3, - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(VCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS1), }; @@ -622,21 +619,21 @@ static const struct intel_device_info cml_gt2_info = { static const struct intel_device_info icl_info = { GEN11_FEATURES, PLATFORM(INTEL_ICELAKE), - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), }; static const struct intel_device_info ehl_info = { GEN11_FEATURES, PLATFORM(INTEL_ELKHARTLAKE), - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0), + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0), .__runtime.ppgtt_size = 36, }; static const struct intel_device_info jsl_info = { GEN11_FEATURES, PLATFORM(INTEL_JASPERLAKE), - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0), + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VCS0) | BIT(VECS0), .__runtime.ppgtt_size = 36, }; @@ -651,19 +648,19 @@ static const struct intel_device_info jsl_info = { static const struct intel_device_info tgl_info = { GEN12_FEATURES, PLATFORM(INTEL_TIGERLAKE), - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), }; static const struct intel_device_info rkl_info = { GEN12_FEATURES, PLATFORM(INTEL_ROCKETLAKE), - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0), }; #define DGFX_FEATURES \ - .__runtime.memory_regions = REGION_SMEM | REGION_LMEM | REGION_STOLEN_LMEM, \ + .memory_regions = REGION_SMEM | REGION_LMEM | REGION_STOLEN_LMEM, \ .has_llc = 0, \ .has_pxp = 0, \ .has_snoop = 1, \ @@ -676,7 +673,7 @@ static const struct intel_device_info dg1_info = { .__runtime.graphics.ip.rel = 10, PLATFORM(INTEL_DG1), .require_force_probe = 1, - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), /* Wa_16011227922 */ @@ -686,7 +683,7 @@ static const struct intel_device_info dg1_info = { static const struct intel_device_info adl_s_info = { GEN12_FEATURES, PLATFORM(INTEL_ALDERLAKE_S), - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), .dma_mask_size = 39, }; @@ -694,7 +691,7 @@ static const struct intel_device_info adl_s_info = { static const struct intel_device_info adl_p_info = { GEN12_FEATURES, PLATFORM(INTEL_ALDERLAKE_P), - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VCS0) | BIT(VCS2), .__runtime.ppgtt_size = 48, .dma_mask_size = 39, @@ -746,7 +743,7 @@ static const struct intel_device_info xehpsdv_info = { PLATFORM(INTEL_XEHPSDV), .has_64k_pages = 1, .has_media_ratio_mode = 1, - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VECS1) | BIT(VECS2) | BIT(VECS3) | BIT(VCS0) | BIT(VCS1) | BIT(VCS2) | BIT(VCS3) | @@ -766,7 +763,7 @@ static const struct intel_device_info xehpsdv_info = { .has_guc_deprivilege = 1, \ .has_heci_pxp = 1, \ .has_media_ratio_mode = 1, \ - .__runtime.platform_engine_mask = \ + .platform_engine_mask = \ BIT(RCS0) | BIT(BCS0) | \ BIT(VECS0) | BIT(VECS1) | \ BIT(VCS0) | BIT(VCS2) | \ @@ -801,7 +798,7 @@ static const struct intel_device_info pvc_info = { PLATFORM(INTEL_PONTEVECCHIO), .has_flat_ccs = 0, .max_pat_index = 7, - .__runtime.platform_engine_mask = + .platform_engine_mask = BIT(BCS0) | BIT(VCS0) | BIT(CCS0) | BIT(CCS1) | BIT(CCS2) | BIT(CCS3), @@ -838,16 +835,14 @@ static const struct intel_device_info mtl_info = { .has_snoop = 1, .max_pat_index = 4, .has_pxp = 1, - .__runtime.memory_regions = REGION_SMEM | REGION_STOLEN_LMEM, - .__runtime.platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(CCS0), + .memory_regions = REGION_SMEM | REGION_STOLEN_LMEM, + .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(CCS0), .require_force_probe = 1, MTL_CACHELEVEL, }; #undef PLATFORM -__diag_pop(); - /* * Make sure any device matches here are from most specific to most * general. For example, since the Quanta match is based on the subsystem diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 7413c11fb562..04bc1f4a1115 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -1319,7 +1319,7 @@ __store_reg_to_mem(struct i915_request *rq, i915_reg_t reg, u32 ggtt_offset) u32 *cs, cmd; cmd = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; - if (GRAPHICS_VER(rq->engine->i915) >= 8) + if (GRAPHICS_VER(rq->i915) >= 8) cmd++; cs = intel_ring_begin(rq, 4); @@ -4431,6 +4431,7 @@ static const struct i915_range mtl_oam_b_counters[] = { static const struct i915_range xehp_oa_b_counters[] = { { .start = 0xdc48, .end = 0xdc48 }, /* OAA_ENABLE_REG */ { .start = 0xdd00, .end = 0xdd48 }, /* OAG_LCE0_0 - OAA_LENABLE_REG */ + {} }; static const struct i915_range gen7_oa_mux_regs[] = { diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 7a4f462e8b70..aefad14ab27a 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -941,8 +941,30 @@ #define HECI_H_GS1(base) _MMIO((base) + 0xc4c) #define HECI_H_GS1_ER_PREP REG_BIT(0) -#define HECI_FWSTS5(base) _MMIO((base) + 0xc68) -#define HECI_FWSTS5_HUC_AUTH_DONE (1 << 19) +/* + * The FWSTS register values are FW defined and can be different between + * HECI1 and HECI2 + */ +#define HECI_FWSTS1 0xc40 +#define HECI1_FWSTS1_CURRENT_STATE REG_GENMASK(3, 0) +#define HECI1_FWSTS1_CURRENT_STATE_RESET 0 +#define HECI1_FWSTS1_PROXY_STATE_NORMAL 5 +#define HECI1_FWSTS1_INIT_COMPLETE REG_BIT(9) +#define HECI_FWSTS2 0xc48 +#define HECI_FWSTS3 0xc60 +#define HECI_FWSTS4 0xc64 +#define HECI_FWSTS5 0xc68 +#define HECI1_FWSTS5_HUC_AUTH_DONE (1 << 19) +#define HECI_FWSTS6 0xc6c + +/* the FWSTS regs are 1-based, so we use -base for index 0 to get an invalid reg */ +#define HECI_FWSTS(base, x) _MMIO((base) + _PICK(x, -(base), \ + HECI_FWSTS1, \ + HECI_FWSTS2, \ + HECI_FWSTS3, \ + HECI_FWSTS4, \ + HECI_FWSTS5, \ + HECI_FWSTS6)) #define HSW_GTT_CACHE_EN _MMIO(0x4024) #define GTT_CACHE_EN_ALL 0xF0007FFF @@ -4917,6 +4939,7 @@ #define SHPD_FILTER_CNT _MMIO(0xc4038) #define SHPD_FILTER_CNT_500_ADJ 0x001D9 +#define SHPD_FILTER_CNT_250 0x000F8 #define _PCH_DPLL_A 0xc6014 #define _PCH_DPLL_B 0xc6018 diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index 32323bb801a1..f59081066a19 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -134,9 +134,7 @@ static void i915_fence_release(struct dma_fence *fence) i915_sw_fence_fini(&rq->semaphore); /* - * Keep one request on each engine for reserved use under mempressure - * do not use with virtual engines as this really is only needed for - * kernel contexts. + * Keep one request on each engine for reserved use under mempressure. * * We do not hold a reference to the engine here and so have to be * very careful in what rq->engine we poke. The virtual engine is @@ -166,8 +164,7 @@ static void i915_fence_release(struct dma_fence *fence) * know that if the rq->execution_mask is a single bit, rq->engine * can be a physical engine with the exact corresponding mask. */ - if (!intel_engine_is_virtual(rq->engine) && - is_power_of_2(rq->execution_mask) && + if (is_power_of_2(rq->execution_mask) && !cmpxchg(&rq->engine->request_pool, NULL, rq)) return; @@ -1353,7 +1350,7 @@ __i915_request_await_external(struct i915_request *rq, struct dma_fence *fence) { mark_external(rq); return i915_sw_fence_await_dma_fence(&rq->submit, fence, - i915_fence_context_timeout(rq->engine->i915, + i915_fence_context_timeout(rq->i915, fence->context), I915_FENCE_GFP); } @@ -1661,6 +1658,11 @@ __i915_request_ensure_parallel_ordering(struct i915_request *rq, request_to_parent(rq)->parallel.last_rq = i915_request_get(rq); + /* + * Users have to put a reference potentially got by + * __i915_active_fence_set() to the returned request + * when no longer needed + */ return to_request(__i915_active_fence_set(&timeline->last_request, &rq->fence)); } @@ -1707,6 +1709,10 @@ __i915_request_ensure_ordering(struct i915_request *rq, 0); } + /* + * Users have to put the reference to prev potentially got + * by __i915_active_fence_set() when no longer needed + */ return prev; } @@ -1760,6 +1766,8 @@ __i915_request_add_to_timeline(struct i915_request *rq) prev = __i915_request_ensure_ordering(rq, timeline); else prev = __i915_request_ensure_parallel_ordering(rq, timeline); + if (prev) + i915_request_put(prev); /* * Make sure that no request gazumped us - if it was allocated after diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h index f6f9228a1351..ce1cbee1b39d 100644 --- a/drivers/gpu/drm/i915/i915_trace.h +++ b/drivers/gpu/drm/i915/i915_trace.h @@ -277,7 +277,7 @@ TRACE_EVENT(i915_request_queue, ), TP_fast_assign( - __entry->dev = rq->engine->i915->drm.primary->index; + __entry->dev = rq->i915->drm.primary->index; __entry->class = rq->engine->uabi_class; __entry->instance = rq->engine->uabi_instance; __entry->ctx = rq->fence.context; @@ -304,7 +304,7 @@ DECLARE_EVENT_CLASS(i915_request, ), TP_fast_assign( - __entry->dev = rq->engine->i915->drm.primary->index; + __entry->dev = rq->i915->drm.primary->index; __entry->class = rq->engine->uabi_class; __entry->instance = rq->engine->uabi_instance; __entry->ctx = rq->fence.context; @@ -353,7 +353,7 @@ TRACE_EVENT(i915_request_in, ), TP_fast_assign( - __entry->dev = rq->engine->i915->drm.primary->index; + __entry->dev = rq->i915->drm.primary->index; __entry->class = rq->engine->uabi_class; __entry->instance = rq->engine->uabi_instance; __entry->ctx = rq->fence.context; @@ -382,7 +382,7 @@ TRACE_EVENT(i915_request_out, ), TP_fast_assign( - __entry->dev = rq->engine->i915->drm.primary->index; + __entry->dev = rq->i915->drm.primary->index; __entry->class = rq->engine->uabi_class; __entry->instance = rq->engine->uabi_instance; __entry->ctx = rq->fence.context; @@ -623,7 +623,7 @@ TRACE_EVENT(i915_request_wait_begin, * less desirable. */ TP_fast_assign( - __entry->dev = rq->engine->i915->drm.primary->index; + __entry->dev = rq->i915->drm.primary->index; __entry->class = rq->engine->uabi_class; __entry->instance = rq->engine->uabi_instance; __entry->ctx = rq->fence.context; diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index ffb425ba591c..6f180ee13853 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -34,6 +34,7 @@ #include "gt/intel_engine_heartbeat.h" #include "gt/intel_gt.h" #include "gt/intel_gt_requests.h" +#include "gt/intel_tlb.h" #include "i915_drv.h" #include "i915_gem_evict.h" @@ -74,14 +75,14 @@ static void vma_print_allocator(struct i915_vma *vma, const char *reason) char buf[512]; if (!vma->node.stack) { - drm_dbg(&to_i915(vma->obj->base.dev)->drm, + drm_dbg(vma->obj->base.dev, "vma.node [%08llx + %08llx] %s: unknown owner\n", vma->node.start, vma->node.size, reason); return; } stack_depot_snprint(vma->node.stack, buf, sizeof(buf), 0); - drm_dbg(&to_i915(vma->obj->base.dev)->drm, + drm_dbg(vma->obj->base.dev, "vma.node [%08llx + %08llx] %s: inserted at %s\n", vma->node.start, vma->node.size, reason, buf); } @@ -805,7 +806,7 @@ i915_vma_insert(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, * attempt to find space. */ if (size > end - 2 * guard) { - drm_dbg(&to_i915(vma->obj->base.dev)->drm, + drm_dbg(vma->obj->base.dev, "Attempting to bind an object larger than the aperture: request=%llu > %s aperture=%llu\n", size, flags & PIN_MAPPABLE ? "mappable" : "total", end); return -ENOSPC; @@ -1339,6 +1340,12 @@ err_unpin: void vma_invalidate_tlb(struct i915_address_space *vm, u32 *tlb) { + struct intel_gt *gt; + int id; + + if (!tlb) + return; + /* * Before we release the pages that were bound by this vma, we * must invalidate all the TLBs that may still have a reference @@ -1347,7 +1354,9 @@ void vma_invalidate_tlb(struct i915_address_space *vm, u32 *tlb) * the most recent TLB invalidation seqno, and if we have not yet * flushed the TLBs upon release, perform a full invalidation. */ - WRITE_ONCE(*tlb, intel_gt_next_invalidate_tlb_full(vm->gt)); + for_each_gt(gt, vm->i915, id) + WRITE_ONCE(tlb[id], + intel_gt_next_invalidate_tlb_full(gt)); } static void __vma_put_pages(struct i915_vma *vma, unsigned int count) @@ -1629,6 +1638,26 @@ int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, return err; } +/** + * i915_ggtt_clear_scanout - Clear scanout flag for all objects ggtt vmas + * @obj: i915 GEM object + * This function clears scanout flags for objects ggtt vmas. These flags are set + * when object is pinned for display use and this function to clear them all is + * targeted to be called by frontbuffer tracking code when the frontbuffer is + * about to be released. + */ +void i915_ggtt_clear_scanout(struct drm_i915_gem_object *obj) +{ + struct i915_vma *vma; + + spin_lock(&obj->vma.lock); + for_each_ggtt_vma(vma, obj) { + i915_vma_clear_scanout(vma); + vma->display_alignment = I915_GTT_MIN_ALIGNMENT; + } + spin_unlock(&obj->vma.lock); +} + static void __vma_close(struct i915_vma *vma, struct intel_gt *gt) { /* @@ -1908,7 +1937,7 @@ int _i915_vma_move_to_active(struct i915_vma *vma, if (flags & EXEC_OBJECT_WRITE) { struct intel_frontbuffer *front; - front = __intel_frontbuffer_get(obj); + front = i915_gem_object_get_frontbuffer(obj); if (unlikely(front)) { if (intel_frontbuffer_invalidate(front, ORIGIN_CS)) i915_active_add_request(&front->write, rq); @@ -1994,7 +2023,7 @@ struct dma_fence *__i915_vma_evict(struct i915_vma *vma, bool async) if (async) unbind_fence = i915_vma_resource_unbind(vma_res, - &vma->obj->mm.tlb); + vma->obj->mm.tlb); else unbind_fence = i915_vma_resource_unbind(vma_res, NULL); @@ -2011,7 +2040,7 @@ struct dma_fence *__i915_vma_evict(struct i915_vma *vma, bool async) dma_fence_put(unbind_fence); unbind_fence = NULL; } - vma_invalidate_tlb(vma->vm, &vma->obj->mm.tlb); + vma_invalidate_tlb(vma->vm, vma->obj->mm.tlb); } /* diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h index 9a9729205d5b..e356dfb883d3 100644 --- a/drivers/gpu/drm/i915/i915_vma.h +++ b/drivers/gpu/drm/i915/i915_vma.h @@ -418,6 +418,11 @@ i915_vma_unpin_fence(struct i915_vma *vma) __i915_vma_unpin_fence(vma); } +static inline int i915_vma_fence_id(const struct i915_vma *vma) +{ + return vma->fence ? vma->fence->id : -1; +} + void i915_vma_parked(struct intel_gt *gt); static inline bool i915_vma_is_scanout(const struct i915_vma *vma) @@ -435,6 +440,8 @@ static inline void i915_vma_clear_scanout(struct i915_vma *vma) clear_bit(I915_VMA_SCANOUT_BIT, __i915_vma_flags(vma)); } +void i915_ggtt_clear_scanout(struct drm_i915_gem_object *obj); + #define for_each_until(cond) if (cond) break; else /** diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c index a27600bc5976..81a4d32734e9 100644 --- a/drivers/gpu/drm/i915/intel_clock_gating.c +++ b/drivers/gpu/drm/i915/intel_clock_gating.c @@ -456,12 +456,12 @@ static void kbl_init_clock_gating(struct drm_i915_private *i915) intel_uncore_rmw(&i915->uncore, FBC_LLC_READ_CTRL, 0, FBC_LLC_FULLY_OPEN); /* WaDisableSDEUnitClockGating:kbl */ - if (IS_KBL_GRAPHICS_STEP(i915, 0, STEP_C0)) + if (IS_KABYLAKE(i915) && IS_GRAPHICS_STEP(i915, 0, STEP_C0)) intel_uncore_rmw(&i915->uncore, GEN8_UCGCTL6, 0, GEN8_SDEUNIT_CLOCK_GATE_DISABLE); /* WaDisableGamClockGating:kbl */ - if (IS_KBL_GRAPHICS_STEP(i915, 0, STEP_C0)) + if (IS_KABYLAKE(i915) && IS_GRAPHICS_STEP(i915, 0, STEP_C0)) intel_uncore_rmw(&i915->uncore, GEN6_UCGCTL1, 0, GEN6_GAMUNIT_CLOCK_GATE_DISABLE); @@ -559,9 +559,20 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915) static void hsw_init_clock_gating(struct drm_i915_private *i915) { + enum pipe pipe; + /* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */ intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(PIPE_A), 0, HSW_FBCQ_DIS); + /* WaPsrDPAMaskVBlankInSRD:hsw */ + intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD); + + for_each_pipe(i915, pipe) { + /* WaPsrDPRSUnmaskVBlankInSRD:hsw */ + intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe), + 0, HSW_UNMASK_VBL_TO_REGS_IN_SRD); + } + /* This is required by WaCatErrorRejectionIssue:hsw */ intel_uncore_rmw(&i915->uncore, GEN7_SQ_CHICKEN_MBCUNIT_CONFIG, 0, GEN7_SQ_CHICKEN_MBCUNIT_SQINTMOB); diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index 6e49caf241a5..ea0ec6174ce5 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -93,9 +93,6 @@ void intel_device_info_print(const struct intel_device_info *info, const struct intel_runtime_info *runtime, struct drm_printer *p) { - const struct intel_display_runtime_info *display_runtime = - &info->display->__runtime_defaults; - if (runtime->graphics.ip.rel) drm_printf(p, "graphics version: %u.%02u\n", runtime->graphics.ip.ver, @@ -112,21 +109,13 @@ void intel_device_info_print(const struct intel_device_info *info, drm_printf(p, "media version: %u\n", runtime->media.ip.ver); - if (display_runtime->ip.rel) - drm_printf(p, "display version: %u.%02u\n", - display_runtime->ip.ver, - display_runtime->ip.rel); - else - drm_printf(p, "display version: %u\n", - display_runtime->ip.ver); - drm_printf(p, "graphics stepping: %s\n", intel_step_name(runtime->step.graphics_step)); drm_printf(p, "media stepping: %s\n", intel_step_name(runtime->step.media_step)); drm_printf(p, "display stepping: %s\n", intel_step_name(runtime->step.display_step)); drm_printf(p, "base die stepping: %s\n", intel_step_name(runtime->step.basedie_step)); drm_printf(p, "gt: %d\n", info->gt); - drm_printf(p, "memory-regions: 0x%x\n", runtime->memory_regions); + drm_printf(p, "memory-regions: 0x%x\n", info->memory_regions); drm_printf(p, "page-sizes: 0x%x\n", runtime->page_sizes); drm_printf(p, "platform: %s\n", intel_platform_name(info->platform)); drm_printf(p, "ppgtt-size: %d\n", runtime->ppgtt_size); @@ -138,15 +127,6 @@ void intel_device_info_print(const struct intel_device_info *info, #undef PRINT_FLAG drm_printf(p, "has_pooled_eu: %s\n", str_yes_no(runtime->has_pooled_eu)); - -#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, str_yes_no(info->display->name)) - DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG); -#undef PRINT_FLAG - - drm_printf(p, "has_hdcp: %s\n", str_yes_no(display_runtime->has_hdcp)); - drm_printf(p, "has_dmc: %s\n", str_yes_no(display_runtime->has_dmc)); - drm_printf(p, "has_dsc: %s\n", str_yes_no(display_runtime->has_dsc)); - drm_printf(p, "rawclk rate: %u kHz\n", runtime->rawclk_freq); } @@ -260,15 +240,19 @@ static void intel_device_info_subplatform_init(struct drm_i915_private *i915) if (find_devid(devid, subplatform_ult_ids, ARRAY_SIZE(subplatform_ult_ids))) { mask = BIT(INTEL_SUBPLATFORM_ULT); + if (IS_HASWELL(i915) || IS_BROADWELL(i915)) + DISPLAY_RUNTIME_INFO(i915)->port_mask &= ~BIT(PORT_D); } else if (find_devid(devid, subplatform_ulx_ids, ARRAY_SIZE(subplatform_ulx_ids))) { mask = BIT(INTEL_SUBPLATFORM_ULX); if (IS_HASWELL(i915) || IS_BROADWELL(i915)) { /* ULX machines are also considered ULT. */ mask |= BIT(INTEL_SUBPLATFORM_ULT); + DISPLAY_RUNTIME_INFO(i915)->port_mask &= ~BIT(PORT_D); } } else if (find_devid(devid, subplatform_portf_ids, ARRAY_SIZE(subplatform_portf_ids))) { + DISPLAY_RUNTIME_INFO(i915)->port_mask |= BIT(PORT_F); mask = BIT(INTEL_SUBPLATFORM_PORTF); } else if (find_devid(devid, subplatform_uy_ids, ARRAY_SIZE(subplatform_uy_ids))) { @@ -380,13 +364,6 @@ void intel_device_info_runtime_init_early(struct drm_i915_private *i915) intel_device_info_subplatform_init(i915); } -/* FIXME: Remove this, and make device info a const pointer to rodata. */ -static struct intel_device_info * -mkwrite_device_info(struct drm_i915_private *i915) -{ - return (struct intel_device_info *)INTEL_INFO(i915); -} - static const struct intel_display_device_info no_display = {}; /** @@ -407,7 +384,6 @@ static const struct intel_display_device_info no_display = {}; */ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) { - struct intel_device_info *info = mkwrite_device_info(dev_priv); struct intel_runtime_info *runtime = RUNTIME_INFO(dev_priv); if (HAS_DISPLAY(dev_priv)) @@ -417,7 +393,7 @@ void intel_device_info_runtime_init(struct drm_i915_private *dev_priv) if (!HAS_DISPLAY(dev_priv)) { dev_priv->drm.driver_features &= ~(DRIVER_MODESET | DRIVER_ATOMIC); - info->display = &no_display; + dev_priv->display.info.__device_info = &no_display; } /* Disable nuclear pageflip by default on pre-g4x */ @@ -447,26 +423,24 @@ void intel_device_info_driver_create(struct drm_i915_private *i915, u16 device_id, const struct intel_device_info *match_info) { - struct intel_device_info *info; struct intel_runtime_info *runtime; u16 ver, rel, step; - /* Setup the write-once "constant" device info */ - info = mkwrite_device_info(i915); - memcpy(info, match_info, sizeof(*info)); + /* Setup INTEL_INFO() */ + i915->__info = match_info; /* Initialize initial runtime info from static const data and pdev. */ runtime = RUNTIME_INFO(i915); memcpy(runtime, &INTEL_INFO(i915)->__runtime, sizeof(*runtime)); /* Probe display support */ - info->display = intel_display_device_probe(i915, info->has_gmd_id, - &ver, &rel, &step); + i915->display.info.__device_info = intel_display_device_probe(i915, HAS_GMD_ID(i915), + &ver, &rel, &step); memcpy(DISPLAY_RUNTIME_INFO(i915), &DISPLAY_INFO(i915)->__runtime_defaults, sizeof(*DISPLAY_RUNTIME_INFO(i915))); - if (info->has_gmd_id) { + if (HAS_GMD_ID(i915)) { DISPLAY_RUNTIME_INFO(i915)->ip.ver = ver; DISPLAY_RUNTIME_INFO(i915)->ip.rel = rel; DISPLAY_RUNTIME_INFO(i915)->ip.step = step; diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index 069291b3bd37..dbfe6443457b 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -29,8 +29,6 @@ #include "intel_step.h" -#include "display/intel_display_device.h" - #include "gt/intel_engine_types.h" #include "gt/intel_context_types.h" #include "gt/intel_sseu.h" @@ -212,8 +210,6 @@ struct intel_runtime_info { u16 device_id; - intel_engine_mask_t platform_engine_mask; /* Engines supported by the HW */ - u32 rawclk_freq; struct intel_step_info step; @@ -223,8 +219,6 @@ struct intel_runtime_info { enum intel_ppgtt_type ppgtt_type; unsigned int ppgtt_size; /* log2, e.g. 31/32/48 bits */ - u32 memory_regions; /* regions supported by the HW */ - bool has_pooled_eu; }; @@ -237,12 +231,13 @@ struct intel_device_info { u8 gt; /* GT number, 0 if undefined */ + intel_engine_mask_t platform_engine_mask; /* Engines supported by the HW */ + u32 memory_regions; /* regions supported by the HW */ + #define DEFINE_FLAG(name) u8 name:1 DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG); #undef DEFINE_FLAG - const struct intel_display_device_info *display; - /* * Initial runtime info. Do not access outside of i915_driver_create(). */ diff --git a/drivers/gpu/drm/i915/intel_step.c b/drivers/gpu/drm/i915/intel_step.c index 8a9ff6227e53..c02a6f156a00 100644 --- a/drivers/gpu/drm/i915/intel_step.c +++ b/drivers/gpu/drm/i915/intel_step.c @@ -192,16 +192,16 @@ void intel_step_init(struct drm_i915_private *i915) } else if (IS_XEHPSDV(i915)) { revids = xehpsdv_revids; size = ARRAY_SIZE(xehpsdv_revids); - } else if (IS_ADLP_N(i915)) { + } else if (IS_ALDERLAKE_P_N(i915)) { revids = adlp_n_revids; size = ARRAY_SIZE(adlp_n_revids); - } else if (IS_ADLP_RPLP(i915)) { + } else if (IS_RAPTORLAKE_P(i915)) { revids = adlp_rplp_revids; size = ARRAY_SIZE(adlp_rplp_revids); } else if (IS_ALDERLAKE_P(i915)) { revids = adlp_revids; size = ARRAY_SIZE(adlp_revids); - } else if (IS_ADLS_RPLS(i915)) { + } else if (IS_RAPTORLAKE_S(i915)) { revids = adls_rpls_revids; size = ARRAY_SIZE(adls_rpls_revids); } else if (IS_ALDERLAKE_S(i915)) { @@ -213,13 +213,13 @@ void intel_step_init(struct drm_i915_private *i915) } else if (IS_ROCKETLAKE(i915)) { revids = rkl_revids; size = ARRAY_SIZE(rkl_revids); - } else if (IS_TGL_UY(i915)) { + } else if (IS_TIGERLAKE_UY(i915)) { revids = tgl_uy_revids; size = ARRAY_SIZE(tgl_uy_revids); } else if (IS_TIGERLAKE(i915)) { revids = tgl_revids; size = ARRAY_SIZE(tgl_revids); - } else if (IS_JSL_EHL(i915)) { + } else if (IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) { revids = jsl_ehl_revids; size = ARRAY_SIZE(jsl_ehl_revids); } else if (IS_ICELAKE(i915)) { diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 796ebfe6c550..dfefad5a5fec 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1925,25 +1925,31 @@ __unclaimed_previous_reg_debug(struct intel_uncore *uncore, i915_mmio_reg_offset(reg)); } -static inline void -unclaimed_reg_debug(struct intel_uncore *uncore, - const i915_reg_t reg, - const bool read, - const bool before) +static inline bool __must_check +unclaimed_reg_debug_header(struct intel_uncore *uncore, + const i915_reg_t reg, const bool read) { if (likely(!uncore->i915->params.mmio_debug) || !uncore->debug) - return; + return false; /* interrupts are disabled and re-enabled around uncore->lock usage */ lockdep_assert_held(&uncore->lock); - if (before) { - spin_lock(&uncore->debug->lock); - __unclaimed_previous_reg_debug(uncore, reg, read); - } else { - __unclaimed_reg_debug(uncore, reg, read); - spin_unlock(&uncore->debug->lock); - } + spin_lock(&uncore->debug->lock); + __unclaimed_previous_reg_debug(uncore, reg, read); + + return true; +} + +static inline void +unclaimed_reg_debug_footer(struct intel_uncore *uncore, + const i915_reg_t reg, const bool read) +{ + /* interrupts are disabled and re-enabled around uncore->lock usage */ + lockdep_assert_held(&uncore->lock); + + __unclaimed_reg_debug(uncore, reg, read); + spin_unlock(&uncore->debug->lock); } #define __vgpu_read(x) \ @@ -2001,13 +2007,15 @@ __gen2_read(64) #define GEN6_READ_HEADER(x) \ u32 offset = i915_mmio_reg_offset(reg); \ unsigned long irqflags; \ + bool unclaimed_reg_debug; \ u##x val = 0; \ assert_rpm_wakelock_held(uncore->rpm); \ spin_lock_irqsave(&uncore->lock, irqflags); \ - unclaimed_reg_debug(uncore, reg, true, true) + unclaimed_reg_debug = unclaimed_reg_debug_header(uncore, reg, true) #define GEN6_READ_FOOTER \ - unclaimed_reg_debug(uncore, reg, true, false); \ + if (unclaimed_reg_debug) \ + unclaimed_reg_debug_footer(uncore, reg, true); \ spin_unlock_irqrestore(&uncore->lock, irqflags); \ trace_i915_reg_rw(false, reg, val, sizeof(val), trace); \ return val @@ -2105,13 +2113,15 @@ __gen2_write(32) #define GEN6_WRITE_HEADER \ u32 offset = i915_mmio_reg_offset(reg); \ unsigned long irqflags; \ + bool unclaimed_reg_debug; \ trace_i915_reg_rw(true, reg, val, sizeof(val), trace); \ assert_rpm_wakelock_held(uncore->rpm); \ spin_lock_irqsave(&uncore->lock, irqflags); \ - unclaimed_reg_debug(uncore, reg, false, true) + unclaimed_reg_debug = unclaimed_reg_debug_header(uncore, reg, false) #define GEN6_WRITE_FOOTER \ - unclaimed_reg_debug(uncore, reg, false, false); \ + if (unclaimed_reg_debug) \ + unclaimed_reg_debug_footer(uncore, reg, false); \ spin_unlock_irqrestore(&uncore->lock, irqflags) #define __gen6_write(x) \ diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h index 9ea1f4864a3a..f419c311a0de 100644 --- a/drivers/gpu/drm/i915/intel_uncore.h +++ b/drivers/gpu/drm/i915/intel_uncore.h @@ -496,6 +496,11 @@ static inline int intel_uncore_write_and_verify(struct intel_uncore *uncore, return (reg_val & mask) != expected_val ? -EINVAL : 0; } +static inline void __iomem *intel_uncore_regs(struct intel_uncore *uncore) +{ + return uncore->regs; +} + /* * The raw_reg_{read,write} macros are intended as a micro-optimization for * interrupt handlers so that the pointer indirection on uncore->regs can diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c index bb2e15329f34..38ec754d0ec8 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c @@ -162,8 +162,8 @@ static struct intel_gt *find_gt_for_required_teelink(struct drm_i915_private *i9 * for HuC authentication. For now, its limited to DG2. */ if (IS_ENABLED(CONFIG_INTEL_MEI_PXP) && IS_ENABLED(CONFIG_INTEL_MEI_GSC) && - intel_huc_is_loaded_by_gsc(&i915->gt0.uc.huc) && intel_uc_uses_huc(&i915->gt0.uc)) - return &i915->gt0; + intel_huc_is_loaded_by_gsc(&to_gt(i915)->uc.huc) && intel_uc_uses_huc(&to_gt(i915)->uc)) + return to_gt(i915); return NULL; } @@ -188,8 +188,8 @@ static struct intel_gt *find_gt_for_required_protected_content(struct drm_i915_p * Else we rely on mei-pxp module but only on legacy platforms * prior to having separate media GTs and has a valid VDBOX. */ - if (IS_ENABLED(CONFIG_INTEL_MEI_PXP) && !i915->media_gt && VDBOX_MASK(&i915->gt0)) - return &i915->gt0; + if (IS_ENABLED(CONFIG_INTEL_MEI_PXP) && !i915->media_gt && VDBOX_MASK(to_gt(i915))) + return to_gt(i915); return NULL; } diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c index f13890ec7db1..2a600184a077 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_gsccs.c @@ -6,6 +6,7 @@ #include "gem/i915_gem_internal.h" #include "gt/intel_context.h" +#include "gt/intel_gt.h" #include "gt/uc/intel_gsc_fw.h" #include "gt/uc/intel_gsc_uc_heci_cmd_submit.h" @@ -197,7 +198,7 @@ bool intel_pxp_gsccs_is_ready_for_sessions(struct intel_pxp *pxp) * are out of order) will suffice. */ if (intel_huc_is_authenticated(&pxp->ctrl_gt->uc.huc, INTEL_HUC_AUTH_BY_GSC) && - intel_gsc_uc_fw_proxy_init_done(&pxp->ctrl_gt->uc.gsc)) + intel_gsc_uc_fw_proxy_init_done(&pxp->ctrl_gt->uc.gsc, true)) return true; return false; @@ -336,7 +337,7 @@ gsccs_create_buffer(struct intel_gt *gt, } /* return a virtual pointer */ - *map = i915_gem_object_pin_map_unlocked(obj, i915_coherent_map_type(i915, obj, true)); + *map = i915_gem_object_pin_map_unlocked(obj, intel_gt_coherent_map_type(gt, obj, true)); if (IS_ERR(*map)) { drm_err(&i915->drm, "Failed to map gsccs backend %s.\n", bufname); err = PTR_ERR(*map); diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c b/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c index 1ce07d7e8769..80bb00189865 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c @@ -11,6 +11,7 @@ #include "gem/i915_gem_lmem.h" #include "i915_drv.h" +#include "gt/intel_gt.h" #include "intel_pxp.h" #include "intel_pxp_cmd_interface_42.h" @@ -245,7 +246,9 @@ static int alloc_streaming_command(struct intel_pxp *pxp) } /* map the lmem into the virtual memory pointer */ - cmd = i915_gem_object_pin_map_unlocked(obj, i915_coherent_map_type(i915, obj, true)); + cmd = i915_gem_object_pin_map_unlocked(obj, + intel_gt_coherent_map_type(pxp->ctrl_gt, + obj, true)); if (IS_ERR(cmd)) { drm_err(&i915->drm, "Failed to map gsc message page!\n"); err = PTR_ERR(cmd); diff --git a/drivers/gpu/drm/i915/selftests/i915_perf.c b/drivers/gpu/drm/i915/selftests/i915_perf.c index d4608b220123..403134a7acec 100644 --- a/drivers/gpu/drm/i915/selftests/i915_perf.c +++ b/drivers/gpu/drm/i915/selftests/i915_perf.c @@ -168,7 +168,7 @@ static int write_timestamp(struct i915_request *rq, int slot) return PTR_ERR(cs); len = 5; - if (GRAPHICS_VER(rq->engine->i915) >= 8) + if (GRAPHICS_VER(rq->i915) >= 8) len++; *cs++ = GFX_OP_PIPE_CONTROL(len); diff --git a/drivers/gpu/drm/i915/selftests/i915_selftest.c b/drivers/gpu/drm/i915/selftests/i915_selftest.c index 39da0fb0d6d2..ee79e0809a6d 100644 --- a/drivers/gpu/drm/i915/selftests/i915_selftest.c +++ b/drivers/gpu/drm/i915/selftests/i915_selftest.c @@ -24,6 +24,8 @@ #include <linux/random.h> #include "gt/intel_gt_pm.h" +#include "gt/uc/intel_gsc_fw.h" + #include "i915_driver.h" #include "i915_drv.h" #include "i915_selftest.h" @@ -127,6 +129,31 @@ static void set_default_test_all(struct selftest *st, unsigned int count) st[i].enabled = true; } +static bool +__gsc_proxy_init_progressing(struct intel_gsc_uc *gsc) +{ + return intel_gsc_uc_fw_proxy_get_status(gsc) == -EAGAIN; +} + +static void +__wait_gsc_proxy_completed(struct drm_i915_private *i915) +{ + bool need_to_wait = (IS_ENABLED(CONFIG_INTEL_MEI_GSC_PROXY) && + i915->media_gt && + HAS_ENGINE(i915->media_gt, GSC0) && + intel_uc_fw_is_loadable(&i915->media_gt->uc.gsc.fw)); + /* + * The gsc proxy component depends on the kernel component driver load ordering + * and in corner cases (the first time after an IFWI flash), init-completion + * firmware flows take longer. + */ + unsigned long timeout_ms = 8000; + + if (need_to_wait && wait_for(!__gsc_proxy_init_progressing(&i915->media_gt->uc.gsc), + timeout_ms)) + pr_warn(DRIVER_NAME "Timed out waiting for gsc_proxy_completion!\n"); +} + static int __run_selftests(const char *name, struct selftest *st, unsigned int count, @@ -206,6 +233,8 @@ int i915_live_selftests(struct pci_dev *pdev) if (!i915_selftest.live) return 0; + __wait_gsc_proxy_completed(pdev_to_i915(pdev)); + err = run_selftests(live, pdev_to_i915(pdev)); if (err) { i915_selftest.live = err; @@ -227,6 +256,8 @@ int i915_perf_selftests(struct pci_dev *pdev) if (!i915_selftest.perf) return 0; + __wait_gsc_proxy_completed(pdev_to_i915(pdev)); + err = run_selftests(perf, pdev_to_i915(pdev)); if (err) { i915_selftest.perf = err; diff --git a/drivers/gpu/drm/i915/selftests/igt_spinner.c b/drivers/gpu/drm/i915/selftests/igt_spinner.c index 618d9386d554..0f064930ef11 100644 --- a/drivers/gpu/drm/i915/selftests/igt_spinner.c +++ b/drivers/gpu/drm/i915/selftests/igt_spinner.c @@ -97,7 +97,7 @@ int igt_spinner_pin(struct igt_spinner *spin, if (!spin->batch) { unsigned int mode; - mode = i915_coherent_map_type(spin->gt->i915, spin->obj, false); + mode = intel_gt_coherent_map_type(spin->gt, spin->obj, false); vaddr = igt_spinner_pin_obj(ce, ww, spin->obj, mode, &spin->batch_vma); if (IS_ERR(vaddr)) return PTR_ERR(vaddr); @@ -159,15 +159,15 @@ igt_spinner_create_request(struct igt_spinner *spin, batch = spin->batch; - if (GRAPHICS_VER(rq->engine->i915) >= 8) { + if (GRAPHICS_VER(rq->i915) >= 8) { *batch++ = MI_STORE_DWORD_IMM_GEN4; *batch++ = lower_32_bits(hws_address(hws, rq)); *batch++ = upper_32_bits(hws_address(hws, rq)); - } else if (GRAPHICS_VER(rq->engine->i915) >= 6) { + } else if (GRAPHICS_VER(rq->i915) >= 6) { *batch++ = MI_STORE_DWORD_IMM_GEN4; *batch++ = 0; *batch++ = hws_address(hws, rq); - } else if (GRAPHICS_VER(rq->engine->i915) >= 4) { + } else if (GRAPHICS_VER(rq->i915) >= 4) { *batch++ = MI_STORE_DWORD_IMM_GEN4 | MI_USE_GGTT; *batch++ = 0; *batch++ = hws_address(hws, rq); @@ -179,11 +179,11 @@ igt_spinner_create_request(struct igt_spinner *spin, *batch++ = arbitration_command; - if (GRAPHICS_VER(rq->engine->i915) >= 8) + if (GRAPHICS_VER(rq->i915) >= 8) *batch++ = MI_BATCH_BUFFER_START | BIT(8) | 1; - else if (IS_HASWELL(rq->engine->i915)) + else if (IS_HASWELL(rq->i915)) *batch++ = MI_BATCH_BUFFER_START | MI_BATCH_PPGTT_HSW; - else if (GRAPHICS_VER(rq->engine->i915) >= 6) + else if (GRAPHICS_VER(rq->i915) >= 6) *batch++ = MI_BATCH_BUFFER_START; else *batch++ = MI_BATCH_BUFFER_START | MI_BATCH_GTT; @@ -201,7 +201,7 @@ igt_spinner_create_request(struct igt_spinner *spin, } flags = 0; - if (GRAPHICS_VER(rq->engine->i915) <= 5) + if (GRAPHICS_VER(rq->i915) <= 5) flags |= I915_DISPATCH_SECURE; err = engine->emit_bb_start(rq, i915_vma_offset(vma), PAGE_SIZE, flags); diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c index e4281508d580..03ea75cd84dd 100644 --- a/drivers/gpu/drm/i915/selftests/intel_uncore.c +++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c @@ -210,7 +210,7 @@ static int live_forcewake_ops(void *arg) for_each_engine(engine, gt, id) { i915_reg_t mmio = _MMIO(engine->mmio_base + r->offset); - u32 __iomem *reg = uncore->regs + engine->mmio_base + r->offset; + u32 __iomem *reg = intel_uncore_regs(uncore) + engine->mmio_base + r->offset; enum forcewake_domains fw_domains; u32 val; diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c index 4de6a4e8280d..da0b269606c5 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c @@ -114,7 +114,7 @@ static struct dev_pm_domain pm_domain = { static void mock_gt_probe(struct drm_i915_private *i915) { - i915->gt[0] = &i915->gt0; + i915->gt[0] = to_gt(i915); i915->gt[0]->name = "Mock GT"; } @@ -123,8 +123,8 @@ static const struct intel_device_info mock_info = { .__runtime.page_sizes = (I915_GTT_PAGE_SIZE_4K | I915_GTT_PAGE_SIZE_64K | I915_GTT_PAGE_SIZE_2M), - .__runtime.memory_regions = REGION_SMEM, - .__runtime.platform_engine_mask = BIT(0), + .memory_regions = REGION_SMEM, + .platform_engine_mask = BIT(0), /* simply use legacy cache level for mock device */ .max_pat_index = 3, diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c index 9f0651d48d41..15492b69f698 100644 --- a/drivers/gpu/drm/i915/soc/intel_dram.c +++ b/drivers/gpu/drm/i915/soc/intel_dram.c @@ -704,7 +704,7 @@ void intel_dram_edram_detect(struct drm_i915_private *i915) if (!(IS_HASWELL(i915) || IS_BROADWELL(i915) || GRAPHICS_VER(i915) >= 9)) return; - edram_cap = __raw_uncore_read32(&i915->uncore, HSW_EDRAM_CAP); + edram_cap = intel_uncore_read_fw(&i915->uncore, HSW_EDRAM_CAP); /* NB: We can't write IDICR yet because we don't have gt funcs set up */ diff --git a/drivers/gpu/drm/i915/soc/intel_gmch.c b/drivers/gpu/drm/i915/soc/intel_gmch.c index 6d0204942f7a..49c7fb16e934 100644 --- a/drivers/gpu/drm/i915/soc/intel_gmch.c +++ b/drivers/gpu/drm/i915/soc/intel_gmch.c @@ -47,11 +47,9 @@ intel_alloc_mchbar_resource(struct drm_i915_private *i915) mchbar_addr = ((u64)temp_hi << 32) | temp_lo; /* If ACPI doesn't have it, assume we need to allocate it ourselves */ -#ifdef CONFIG_PNP - if (mchbar_addr && + if (IS_ENABLED(CONFIG_PNP) && mchbar_addr && pnp_range_reserved(mchbar_addr, mchbar_addr + MCHBAR_SIZE)) return 0; -#endif /* Get some space for it */ i915->gmch.mch_res.name = "i915 MCHBAR"; diff --git a/drivers/gpu/drm/i915/soc/intel_pch.c b/drivers/gpu/drm/i915/soc/intel_pch.c index ba9843cb1b13..19a8f27c404e 100644 --- a/drivers/gpu/drm/i915/soc/intel_pch.c +++ b/drivers/gpu/drm/i915/soc/intel_pch.c @@ -32,21 +32,21 @@ intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id) drm_WARN_ON(&dev_priv->drm, !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv)); drm_WARN_ON(&dev_priv->drm, - IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv)); + IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv)); return PCH_LPT; case INTEL_PCH_LPT_LP_DEVICE_ID_TYPE: drm_dbg_kms(&dev_priv->drm, "Found LynxPoint LP PCH\n"); drm_WARN_ON(&dev_priv->drm, !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv)); drm_WARN_ON(&dev_priv->drm, - !IS_HSW_ULT(dev_priv) && !IS_BDW_ULT(dev_priv)); + !IS_HASWELL_ULT(dev_priv) && !IS_BROADWELL_ULT(dev_priv)); return PCH_LPT; case INTEL_PCH_WPT_DEVICE_ID_TYPE: drm_dbg_kms(&dev_priv->drm, "Found WildcatPoint PCH\n"); drm_WARN_ON(&dev_priv->drm, !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv)); drm_WARN_ON(&dev_priv->drm, - IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv)); + IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv)); /* WPT is LPT compatible */ return PCH_LPT; case INTEL_PCH_WPT_LP_DEVICE_ID_TYPE: @@ -54,7 +54,7 @@ intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id) drm_WARN_ON(&dev_priv->drm, !IS_HASWELL(dev_priv) && !IS_BROADWELL(dev_priv)); drm_WARN_ON(&dev_priv->drm, - !IS_HSW_ULT(dev_priv) && !IS_BDW_ULT(dev_priv)); + !IS_HASWELL_ULT(dev_priv) && !IS_BROADWELL_ULT(dev_priv)); /* WPT is LPT compatible */ return PCH_LPT; case INTEL_PCH_SPT_DEVICE_ID_TYPE: @@ -115,7 +115,8 @@ intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id) return PCH_ICP; case INTEL_PCH_MCC_DEVICE_ID_TYPE: drm_dbg_kms(&dev_priv->drm, "Found Mule Creek Canyon PCH\n"); - drm_WARN_ON(&dev_priv->drm, !IS_JSL_EHL(dev_priv)); + drm_WARN_ON(&dev_priv->drm, !(IS_JASPERLAKE(dev_priv) || + IS_ELKHARTLAKE(dev_priv))); /* MCC is TGP compatible */ return PCH_TGP; case INTEL_PCH_TGP_DEVICE_ID_TYPE: @@ -127,7 +128,8 @@ intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id) return PCH_TGP; case INTEL_PCH_JSP_DEVICE_ID_TYPE: drm_dbg_kms(&dev_priv->drm, "Found Jasper Lake PCH\n"); - drm_WARN_ON(&dev_priv->drm, !IS_JSL_EHL(dev_priv)); + drm_WARN_ON(&dev_priv->drm, !(IS_JASPERLAKE(dev_priv) || + IS_ELKHARTLAKE(dev_priv))); /* JSP is ICP compatible */ return PCH_ICP; case INTEL_PCH_ADP_DEVICE_ID_TYPE: @@ -177,7 +179,7 @@ intel_virt_detect_pch(const struct drm_i915_private *dev_priv, id = INTEL_PCH_ADP_DEVICE_ID_TYPE; else if (IS_TIGERLAKE(dev_priv) || IS_ROCKETLAKE(dev_priv)) id = INTEL_PCH_TGP_DEVICE_ID_TYPE; - else if (IS_JSL_EHL(dev_priv)) + else if (IS_JASPERLAKE(dev_priv) || IS_ELKHARTLAKE(dev_priv)) id = INTEL_PCH_MCC_DEVICE_ID_TYPE; else if (IS_ICELAKE(dev_priv)) id = INTEL_PCH_ICP_DEVICE_ID_TYPE; @@ -186,7 +188,7 @@ intel_virt_detect_pch(const struct drm_i915_private *dev_priv, id = INTEL_PCH_CNP_DEVICE_ID_TYPE; else if (IS_KABYLAKE(dev_priv) || IS_SKYLAKE(dev_priv)) id = INTEL_PCH_SPT_DEVICE_ID_TYPE; - else if (IS_HSW_ULT(dev_priv) || IS_BDW_ULT(dev_priv)) + else if (IS_HASWELL_ULT(dev_priv) || IS_BROADWELL_ULT(dev_priv)) id = INTEL_PCH_LPT_LP_DEVICE_ID_TYPE; else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) id = INTEL_PCH_LPT_DEVICE_ID_TYPE; diff --git a/drivers/gpu/drm/imx/ipuv3/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3/ipuv3-crtc.c index 8819e40d5e6e..ef29c9a61a46 100644 --- a/drivers/gpu/drm/imx/ipuv3/ipuv3-crtc.c +++ b/drivers/gpu/drm/imx/ipuv3/ipuv3-crtc.c @@ -310,7 +310,7 @@ static void ipu_crtc_mode_set_nofb(struct drm_crtc *crtc) dev_warn(ipu_crtc->dev, "8-pixel align hactive %d -> %d\n", sig_cfg.mode.hactive, new_hactive); - sig_cfg.mode.hfront_porch = new_hactive - sig_cfg.mode.hactive; + sig_cfg.mode.hfront_porch -= new_hactive - sig_cfg.mode.hactive; sig_cfg.mode.hactive = new_hactive; } diff --git a/drivers/gpu/drm/mediatek/Kconfig b/drivers/gpu/drm/mediatek/Kconfig index b451dee64d34..76cab28e010c 100644 --- a/drivers/gpu/drm/mediatek/Kconfig +++ b/drivers/gpu/drm/mediatek/Kconfig @@ -26,6 +26,7 @@ config DRM_MEDIATEK_DP select PHY_MTK_DP select DRM_DISPLAY_HELPER select DRM_DISPLAY_DP_HELPER + select DRM_DP_AUX_BUS help DRM/KMS Display Port driver for MediaTek SoCs. diff --git a/drivers/gpu/drm/mediatek/mtk_disp_aal.c b/drivers/gpu/drm/mediatek/mtk_disp_aal.c index 136b1328e459..4da9ac93b29e 100644 --- a/drivers/gpu/drm/mediatek/mtk_disp_aal.c +++ b/drivers/gpu/drm/mediatek/mtk_disp_aal.c @@ -25,11 +25,6 @@ struct mtk_disp_aal_data { bool has_gamma; }; -/** - * struct mtk_disp_aal - DISP_AAL driver structure - * @ddp_comp - structure containing type enum and hardware resources - * @crtc - associated crtc to report irq events to - */ struct mtk_disp_aal { struct clk *clk; void __iomem *regs; diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ccorr.c b/drivers/gpu/drm/mediatek/mtk_disp_ccorr.c index bd859008eff9..4234ff7485e8 100644 --- a/drivers/gpu/drm/mediatek/mtk_disp_ccorr.c +++ b/drivers/gpu/drm/mediatek/mtk_disp_ccorr.c @@ -33,11 +33,6 @@ struct mtk_disp_ccorr_data { u32 matrix_bits; }; -/** - * struct mtk_disp_ccorr - DISP_CCORR driver structure - * @ddp_comp - structure containing type enum and hardware resources - * @crtc - associated crtc to report irq events to - */ struct mtk_disp_ccorr { struct clk *clk; void __iomem *regs; diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c b/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c index 878575bb9671..6bf6367853fb 100644 --- a/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c +++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl_adaptor.c @@ -427,7 +427,7 @@ static int ovl_adaptor_comp_init(struct device *dev, struct component_match **ma continue; } - type = (enum mtk_ovl_adaptor_comp_type)of_id->data; + type = (enum mtk_ovl_adaptor_comp_type)(uintptr_t)of_id->data; id = ovl_adaptor_comp_get_id(dev, node, type); if (id < 0) { dev_warn(dev, "Skipping unknown component %pOF\n", diff --git a/drivers/gpu/drm/mediatek/mtk_dp.c b/drivers/gpu/drm/mediatek/mtk_dp.c index 4dbc5be156f4..2cb47f663756 100644 --- a/drivers/gpu/drm/mediatek/mtk_dp.c +++ b/drivers/gpu/drm/mediatek/mtk_dp.c @@ -4,6 +4,7 @@ * Copyright (c) 2022 BayLibre */ +#include <drm/display/drm_dp_aux_bus.h> #include <drm/display/drm_dp.h> #include <drm/display/drm_dp_helper.h> #include <drm/drm_atomic_helper.h> @@ -100,6 +101,7 @@ struct mtk_dp_efuse_fmt { struct mtk_dp { bool enabled; bool need_debounce; + int irq; u8 max_lanes; u8 max_linkrate; u8 rx_cap[DP_RECEIVER_CAP_SIZE]; @@ -847,7 +849,7 @@ static int mtk_dp_aux_do_transfer(struct mtk_dp *mtk_dp, bool is_read, u8 cmd, u32 phy_status = mtk_dp_read(mtk_dp, MTK_DP_AUX_P0_3628) & AUX_RX_PHY_STATE_AUX_TX_P0_MASK; if (phy_status != AUX_RX_PHY_STATE_AUX_TX_P0_RX_IDLE) { - drm_err(mtk_dp->drm_dev, + dev_err(mtk_dp->dev, "AUX Rx Aux hang, need SW reset\n"); return -EIO; } @@ -1009,6 +1011,11 @@ static void mtk_dp_initialize_aux_settings(struct mtk_dp *mtk_dp) mtk_dp_update_bits(mtk_dp, MTK_DP_AUX_P0_37C8, MTK_ATOP_EN_AUX_TX_P0, MTK_ATOP_EN_AUX_TX_P0); + + /* Set complete reply mode for AUX */ + mtk_dp_update_bits(mtk_dp, MTK_DP_AUX_P0_3690, + RX_REPLY_COMPLETE_MODE_AUX_TX_P0, + RX_REPLY_COMPLETE_MODE_AUX_TX_P0); } static void mtk_dp_initialize_digital_settings(struct mtk_dp *mtk_dp) @@ -1251,6 +1258,29 @@ static void mtk_dp_audio_mute(struct mtk_dp *mtk_dp, bool mute) val[2], AU_TS_CFG_DP_ENC0_P0_MASK); } +static void mtk_dp_aux_panel_poweron(struct mtk_dp *mtk_dp, bool pwron) +{ + if (pwron) { + /* power on aux */ + mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, + DP_PWR_STATE_BANDGAP_TPLL_LANE, + DP_PWR_STATE_MASK); + + /* power on panel */ + drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D0); + usleep_range(2000, 5000); + } else { + /* power off panel */ + drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D3); + usleep_range(2000, 3000); + + /* power off aux */ + mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, + DP_PWR_STATE_BANDGAP_TPLL, + DP_PWR_STATE_MASK); + } +} + static void mtk_dp_power_enable(struct mtk_dp *mtk_dp) { mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_RESET_AND_PROBE, @@ -1284,9 +1314,11 @@ static void mtk_dp_power_disable(struct mtk_dp *mtk_dp) static void mtk_dp_initialize_priv_data(struct mtk_dp *mtk_dp) { + bool plugged_in = (mtk_dp->bridge.type == DRM_MODE_CONNECTOR_eDP); + mtk_dp->train_info.link_rate = DP_LINK_BW_5_4; mtk_dp->train_info.lane_count = mtk_dp->max_lanes; - mtk_dp->train_info.cable_plugged_in = false; + mtk_dp->train_info.cable_plugged_in = plugged_in; mtk_dp->info.format = DP_PIXELFORMAT_RGB; memset(&mtk_dp->info.vm, 0, sizeof(struct videomode)); @@ -1588,7 +1620,19 @@ static int mtk_dp_parse_capabilities(struct mtk_dp *mtk_dp) u8 val; ssize_t ret; - drm_dp_read_dpcd_caps(&mtk_dp->aux, mtk_dp->rx_cap); + /* + * If we're eDP and capabilities were already parsed we can skip + * reading again because eDP panels aren't hotpluggable hence the + * caps and training information won't ever change in a boot life + */ + if (mtk_dp->bridge.type == DRM_MODE_CONNECTOR_eDP && + mtk_dp->rx_cap[DP_MAX_LINK_RATE] && + mtk_dp->train_info.sink_ssc) + return 0; + + ret = drm_dp_read_dpcd_caps(&mtk_dp->aux, mtk_dp->rx_cap); + if (ret < 0) + return ret; if (drm_dp_tps4_supported(mtk_dp->rx_cap)) mtk_dp->train_info.channel_eq_pattern = DP_TRAINING_PATTERN_4; @@ -1615,10 +1659,13 @@ static int mtk_dp_parse_capabilities(struct mtk_dp *mtk_dp) return ret == 0 ? -EIO : ret; } - if (val) - drm_dp_dpcd_writeb(&mtk_dp->aux, - DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0, - val); + if (val) { + ret = drm_dp_dpcd_writeb(&mtk_dp->aux, + DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0, + val); + if (ret < 0) + return ret; + } } return 0; @@ -1798,10 +1845,6 @@ static void mtk_dp_init_port(struct mtk_dp *mtk_dp) mtk_dp_initialize_settings(mtk_dp); mtk_dp_initialize_aux_settings(mtk_dp); mtk_dp_initialize_digital_settings(mtk_dp); - - mtk_dp_update_bits(mtk_dp, MTK_DP_AUX_P0_3690, - RX_REPLY_COMPLETE_MODE_AUX_TX_P0, - RX_REPLY_COMPLETE_MODE_AUX_TX_P0); mtk_dp_initialize_hpd_detect_settings(mtk_dp); mtk_dp_digital_sw_reset(mtk_dp); @@ -1877,6 +1920,31 @@ static irqreturn_t mtk_dp_hpd_event(int hpd, void *dev) return IRQ_WAKE_THREAD; } +static int mtk_dp_wait_hpd_asserted(struct drm_dp_aux *mtk_aux, unsigned long wait_us) +{ + struct mtk_dp *mtk_dp = container_of(mtk_aux, struct mtk_dp, aux); + u32 val; + int ret; + + ret = regmap_read_poll_timeout(mtk_dp->regs, MTK_DP_TRANS_P0_3414, + val, !!(val & HPD_DB_DP_TRANS_P0_MASK), + wait_us / 100, wait_us); + if (ret) { + mtk_dp->train_info.cable_plugged_in = false; + return ret; + } + + mtk_dp->train_info.cable_plugged_in = true; + + ret = mtk_dp_parse_capabilities(mtk_dp); + if (ret) { + drm_err(mtk_dp->drm_dev, "Can't parse capabilities\n"); + return ret; + } + + return 0; +} + static int mtk_dp_dt_parse(struct mtk_dp *mtk_dp, struct platform_device *pdev) { @@ -1918,6 +1986,9 @@ static int mtk_dp_dt_parse(struct mtk_dp *mtk_dp, static void mtk_dp_update_plugged_status(struct mtk_dp *mtk_dp) { + if (!mtk_dp->data->audio_supported || !mtk_dp->audio_enable) + return; + mutex_lock(&mtk_dp->update_plugged_status_lock); if (mtk_dp->plugged_cb && mtk_dp->codec_dev) mtk_dp->plugged_cb(mtk_dp->codec_dev, @@ -1936,16 +2007,9 @@ static enum drm_connector_status mtk_dp_bdg_detect(struct drm_bridge *bridge) if (!mtk_dp->train_info.cable_plugged_in) return ret; - if (!enabled) { - /* power on aux */ - mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, - DP_PWR_STATE_BANDGAP_TPLL_LANE, - DP_PWR_STATE_MASK); + if (!enabled) + mtk_dp_aux_panel_poweron(mtk_dp, true); - /* power on panel */ - drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D0); - usleep_range(2000, 5000); - } /* * Some dongles still source HPD when they do not connect to any * sink device. To avoid this, we need to read the sink count @@ -1957,16 +2021,8 @@ static enum drm_connector_status mtk_dp_bdg_detect(struct drm_bridge *bridge) if (DP_GET_SINK_COUNT(sink_count)) ret = connector_status_connected; - if (!enabled) { - /* power off panel */ - drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D3); - usleep_range(2000, 3000); - - /* power off aux */ - mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, - DP_PWR_STATE_BANDGAP_TPLL, - DP_PWR_STATE_MASK); - } + if (!enabled) + mtk_dp_aux_panel_poweron(mtk_dp, false); return ret; } @@ -1982,15 +2038,7 @@ static struct edid *mtk_dp_get_edid(struct drm_bridge *bridge, if (!enabled) { drm_atomic_bridge_chain_pre_enable(bridge, connector->state->state); - - /* power on aux */ - mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, - DP_PWR_STATE_BANDGAP_TPLL_LANE, - DP_PWR_STATE_MASK); - - /* power on panel */ - drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D0); - usleep_range(2000, 5000); + mtk_dp_aux_panel_poweron(mtk_dp, true); } new_edid = drm_get_edid(connector, &mtk_dp->aux.ddc); @@ -2010,15 +2058,7 @@ static struct edid *mtk_dp_get_edid(struct drm_bridge *bridge, } if (!enabled) { - /* power off panel */ - drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D3); - usleep_range(2000, 3000); - - /* power off aux */ - mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, - DP_PWR_STATE_BANDGAP_TPLL, - DP_PWR_STATE_MASK); - + mtk_dp_aux_panel_poweron(mtk_dp, false); drm_atomic_bridge_chain_post_disable(bridge, connector->state->state); } @@ -2028,15 +2068,14 @@ static struct edid *mtk_dp_get_edid(struct drm_bridge *bridge, static ssize_t mtk_dp_aux_transfer(struct drm_dp_aux *mtk_aux, struct drm_dp_aux_msg *msg) { - struct mtk_dp *mtk_dp; + struct mtk_dp *mtk_dp = container_of(mtk_aux, struct mtk_dp, aux); bool is_read; u8 request; size_t accessed_bytes = 0; int ret; - mtk_dp = container_of(mtk_aux, struct mtk_dp, aux); - - if (!mtk_dp->train_info.cable_plugged_in) { + if (mtk_dp->bridge.type != DRM_MODE_CONNECTOR_eDP && + !mtk_dp->train_info.cable_plugged_in) { ret = -EAGAIN; goto err; } @@ -2057,7 +2096,7 @@ static ssize_t mtk_dp_aux_transfer(struct drm_dp_aux *mtk_aux, is_read = true; break; default: - drm_err(mtk_aux->drm_dev, "invalid aux cmd = %d\n", + dev_err(mtk_dp->dev, "invalid aux cmd = %d\n", msg->request); ret = -EINVAL; goto err; @@ -2073,7 +2112,7 @@ static ssize_t mtk_dp_aux_transfer(struct drm_dp_aux *mtk_aux, to_access, &msg->reply); if (ret) { - drm_info(mtk_dp->drm_dev, + dev_info(mtk_dp->dev, "Failed to do AUX transfer: %d\n", ret); goto err; } @@ -2143,7 +2182,11 @@ static int mtk_dp_bridge_attach(struct drm_bridge *bridge, mtk_dp->drm_dev = bridge->dev; - mtk_dp_hwirq_enable(mtk_dp, true); + if (mtk_dp->bridge.type != DRM_MODE_CONNECTOR_eDP) { + irq_clear_status_flags(mtk_dp->irq, IRQ_NOAUTOEN); + enable_irq(mtk_dp->irq); + mtk_dp_hwirq_enable(mtk_dp, true); + } return 0; @@ -2158,7 +2201,10 @@ static void mtk_dp_bridge_detach(struct drm_bridge *bridge) { struct mtk_dp *mtk_dp = mtk_dp_from_bridge(bridge); - mtk_dp_hwirq_enable(mtk_dp, false); + if (mtk_dp->bridge.type != DRM_MODE_CONNECTOR_eDP) { + mtk_dp_hwirq_enable(mtk_dp, false); + disable_irq(mtk_dp->irq); + } mtk_dp->drm_dev = NULL; mtk_dp_poweroff(mtk_dp); drm_dp_aux_unregister(&mtk_dp->aux); @@ -2178,15 +2224,7 @@ static void mtk_dp_bridge_atomic_enable(struct drm_bridge *bridge, return; } - /* power on aux */ - mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, - DP_PWR_STATE_BANDGAP_TPLL_LANE, - DP_PWR_STATE_MASK); - - if (mtk_dp->train_info.cable_plugged_in) { - drm_dp_dpcd_writeb(&mtk_dp->aux, DP_SET_POWER, DP_SET_POWER_D0); - usleep_range(2000, 5000); - } + mtk_dp_aux_panel_poweron(mtk_dp, true); /* Training */ ret = mtk_dp_training(mtk_dp); @@ -2481,11 +2519,62 @@ static int mtk_dp_register_audio_driver(struct device *dev) return PTR_ERR_OR_ZERO(mtk_dp->audio_pdev); } +static int mtk_dp_register_phy(struct mtk_dp *mtk_dp) +{ + struct device *dev = mtk_dp->dev; + + mtk_dp->phy_dev = platform_device_register_data(dev, "mediatek-dp-phy", + PLATFORM_DEVID_AUTO, + &mtk_dp->regs, + sizeof(struct regmap *)); + if (IS_ERR(mtk_dp->phy_dev)) + return dev_err_probe(dev, PTR_ERR(mtk_dp->phy_dev), + "Failed to create device mediatek-dp-phy\n"); + + mtk_dp_get_calibration_data(mtk_dp); + + mtk_dp->phy = devm_phy_get(&mtk_dp->phy_dev->dev, "dp"); + if (IS_ERR(mtk_dp->phy)) { + platform_device_unregister(mtk_dp->phy_dev); + return dev_err_probe(dev, PTR_ERR(mtk_dp->phy), "Failed to get phy\n"); + } + + return 0; +} + +static int mtk_dp_edp_link_panel(struct drm_dp_aux *mtk_aux) +{ + struct mtk_dp *mtk_dp = container_of(mtk_aux, struct mtk_dp, aux); + struct device *dev = mtk_aux->dev; + int ret; + + mtk_dp->next_bridge = devm_drm_of_get_bridge(dev, dev->of_node, 1, 0); + + /* Power off the DP and AUX: either detection is done, or no panel present */ + mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, + DP_PWR_STATE_BANDGAP_TPLL, + DP_PWR_STATE_MASK); + mtk_dp_power_disable(mtk_dp); + + if (IS_ERR(mtk_dp->next_bridge)) { + ret = PTR_ERR(mtk_dp->next_bridge); + mtk_dp->next_bridge = NULL; + return ret; + } + + /* For eDP, we add the bridge only if the panel was found */ + ret = devm_drm_bridge_add(dev, &mtk_dp->bridge); + if (ret) + return ret; + + return 0; +} + static int mtk_dp_probe(struct platform_device *pdev) { struct mtk_dp *mtk_dp; struct device *dev = &pdev->dev; - int ret, irq_num; + int ret; mtk_dp = devm_kzalloc(dev, sizeof(*mtk_dp), GFP_KERNEL); if (!mtk_dp) @@ -2494,42 +2583,49 @@ static int mtk_dp_probe(struct platform_device *pdev) mtk_dp->dev = dev; mtk_dp->data = (struct mtk_dp_data *)of_device_get_match_data(dev); - irq_num = platform_get_irq(pdev, 0); - if (irq_num < 0) - return dev_err_probe(dev, irq_num, - "failed to request dp irq resource\n"); - - mtk_dp->next_bridge = devm_drm_of_get_bridge(dev, dev->of_node, 1, 0); - if (IS_ERR(mtk_dp->next_bridge) && - PTR_ERR(mtk_dp->next_bridge) == -ENODEV) - mtk_dp->next_bridge = NULL; - else if (IS_ERR(mtk_dp->next_bridge)) - return dev_err_probe(dev, PTR_ERR(mtk_dp->next_bridge), - "Failed to get bridge\n"); - ret = mtk_dp_dt_parse(mtk_dp, pdev); if (ret) return dev_err_probe(dev, ret, "Failed to parse dt\n"); - drm_dp_aux_init(&mtk_dp->aux); - mtk_dp->aux.name = "aux_mtk_dp"; - mtk_dp->aux.transfer = mtk_dp_aux_transfer; - - spin_lock_init(&mtk_dp->irq_thread_lock); + /* + * Request the interrupt and install service routine only if we are + * on full DisplayPort. + * For eDP, polling the HPD instead is more convenient because we + * don't expect any (un)plug events during runtime, hence we can + * avoid some locking. + */ + if (mtk_dp->data->bridge_type != DRM_MODE_CONNECTOR_eDP) { + mtk_dp->irq = platform_get_irq(pdev, 0); + if (mtk_dp->irq < 0) + return dev_err_probe(dev, mtk_dp->irq, + "failed to request dp irq resource\n"); + + spin_lock_init(&mtk_dp->irq_thread_lock); + + irq_set_status_flags(mtk_dp->irq, IRQ_NOAUTOEN); + ret = devm_request_threaded_irq(dev, mtk_dp->irq, mtk_dp_hpd_event, + mtk_dp_hpd_event_thread, + IRQ_TYPE_LEVEL_HIGH, dev_name(dev), + mtk_dp); + if (ret) + return dev_err_probe(dev, ret, + "failed to request mediatek dptx irq\n"); - ret = devm_request_threaded_irq(dev, irq_num, mtk_dp_hpd_event, - mtk_dp_hpd_event_thread, - IRQ_TYPE_LEVEL_HIGH, dev_name(dev), - mtk_dp); - if (ret) - return dev_err_probe(dev, ret, - "failed to request mediatek dptx irq\n"); + mtk_dp->need_debounce = true; + timer_setup(&mtk_dp->debounce_timer, mtk_dp_debounce_timer, 0); + } - mutex_init(&mtk_dp->update_plugged_status_lock); + mtk_dp->aux.name = "aux_mtk_dp"; + mtk_dp->aux.dev = dev; + mtk_dp->aux.transfer = mtk_dp_aux_transfer; + mtk_dp->aux.wait_hpd_asserted = mtk_dp_wait_hpd_asserted; + drm_dp_aux_init(&mtk_dp->aux); platform_set_drvdata(pdev, mtk_dp); if (mtk_dp->data->audio_supported) { + mutex_init(&mtk_dp->update_plugged_status_lock); + ret = mtk_dp_register_audio_driver(dev); if (ret) { dev_err(dev, "Failed to register audio driver: %d\n", @@ -2538,35 +2634,59 @@ static int mtk_dp_probe(struct platform_device *pdev) } } - mtk_dp->phy_dev = platform_device_register_data(dev, "mediatek-dp-phy", - PLATFORM_DEVID_AUTO, - &mtk_dp->regs, - sizeof(struct regmap *)); - if (IS_ERR(mtk_dp->phy_dev)) - return dev_err_probe(dev, PTR_ERR(mtk_dp->phy_dev), - "Failed to create device mediatek-dp-phy\n"); - - mtk_dp_get_calibration_data(mtk_dp); - - mtk_dp->phy = devm_phy_get(&mtk_dp->phy_dev->dev, "dp"); - - if (IS_ERR(mtk_dp->phy)) { - platform_device_unregister(mtk_dp->phy_dev); - return dev_err_probe(dev, PTR_ERR(mtk_dp->phy), - "Failed to get phy\n"); - } + ret = mtk_dp_register_phy(mtk_dp); + if (ret) + return ret; mtk_dp->bridge.funcs = &mtk_dp_bridge_funcs; mtk_dp->bridge.of_node = dev->of_node; - - mtk_dp->bridge.ops = - DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID | DRM_BRIDGE_OP_HPD; mtk_dp->bridge.type = mtk_dp->data->bridge_type; - drm_bridge_add(&mtk_dp->bridge); + if (mtk_dp->bridge.type == DRM_MODE_CONNECTOR_eDP) { + /* + * Set the data lanes to idle in case the bootloader didn't + * properly close the eDP port to avoid stalls and then + * reinitialize, reset and power on the AUX block. + */ + mtk_dp_set_idle_pattern(mtk_dp, true); + mtk_dp_initialize_aux_settings(mtk_dp); + mtk_dp_power_enable(mtk_dp); + + /* Disable HW interrupts: we don't need any for eDP */ + mtk_dp_hwirq_enable(mtk_dp, false); - mtk_dp->need_debounce = true; - timer_setup(&mtk_dp->debounce_timer, mtk_dp_debounce_timer, 0); + /* + * Power on the AUX to allow reading the EDID from aux-bus: + * please note that it is necessary to call power off in the + * .done_probing() callback (mtk_dp_edp_link_panel), as only + * there we can safely assume that we finished reading EDID. + */ + mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, + DP_PWR_STATE_BANDGAP_TPLL_LANE, + DP_PWR_STATE_MASK); + + ret = devm_of_dp_aux_populate_bus(&mtk_dp->aux, mtk_dp_edp_link_panel); + if (ret) { + /* -ENODEV this means that the panel is not on the aux-bus */ + if (ret == -ENODEV) { + ret = mtk_dp_edp_link_panel(&mtk_dp->aux); + if (ret) + return ret; + } else { + mtk_dp_update_bits(mtk_dp, MTK_DP_TOP_PWR_STATE, + DP_PWR_STATE_BANDGAP_TPLL, + DP_PWR_STATE_MASK); + mtk_dp_power_disable(mtk_dp); + return ret; + } + } + } else { + mtk_dp->bridge.ops = DRM_BRIDGE_OP_DETECT | + DRM_BRIDGE_OP_EDID | DRM_BRIDGE_OP_HPD; + ret = devm_drm_bridge_add(dev, &mtk_dp->bridge); + if (ret) + return dev_err_probe(dev, ret, "Failed to add bridge\n"); + } pm_runtime_enable(dev); pm_runtime_get_sync(dev); @@ -2580,8 +2700,8 @@ static void mtk_dp_remove(struct platform_device *pdev) pm_runtime_put(&pdev->dev); pm_runtime_disable(&pdev->dev); - del_timer_sync(&mtk_dp->debounce_timer); - drm_bridge_remove(&mtk_dp->bridge); + if (mtk_dp->data->bridge_type != DRM_MODE_CONNECTOR_eDP) + del_timer_sync(&mtk_dp->debounce_timer); platform_device_unregister(mtk_dp->phy_dev); if (mtk_dp->audio_pdev) platform_device_unregister(mtk_dp->audio_pdev); @@ -2593,7 +2713,8 @@ static int mtk_dp_suspend(struct device *dev) struct mtk_dp *mtk_dp = dev_get_drvdata(dev); mtk_dp_power_disable(mtk_dp); - mtk_dp_hwirq_enable(mtk_dp, false); + if (mtk_dp->bridge.type != DRM_MODE_CONNECTOR_eDP) + mtk_dp_hwirq_enable(mtk_dp, false); pm_runtime_put_sync(dev); return 0; @@ -2605,7 +2726,8 @@ static int mtk_dp_resume(struct device *dev) pm_runtime_get_sync(dev); mtk_dp_init_port(mtk_dp); - mtk_dp_hwirq_enable(mtk_dp, true); + if (mtk_dp->bridge.type != DRM_MODE_CONNECTOR_eDP) + mtk_dp_hwirq_enable(mtk_dp, true); mtk_dp_power_enable(mtk_dp); return 0; diff --git a/drivers/gpu/drm/mediatek/mtk_dpi.c b/drivers/gpu/drm/mediatek/mtk_dpi.c index 0ef722c24150..2f931e4e2b60 100644 --- a/drivers/gpu/drm/mediatek/mtk_dpi.c +++ b/drivers/gpu/drm/mediatek/mtk_dpi.c @@ -1006,7 +1006,6 @@ static int mtk_dpi_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; struct mtk_dpi *dpi; - struct resource *mem; int ret; dpi = devm_kzalloc(dev, sizeof(*dpi), GFP_KERNEL); @@ -1037,49 +1036,34 @@ static int mtk_dpi_probe(struct platform_device *pdev) dev_dbg(&pdev->dev, "Cannot find pinctrl active!\n"); } } - mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); - dpi->regs = devm_ioremap_resource(dev, mem); - if (IS_ERR(dpi->regs)) { - ret = PTR_ERR(dpi->regs); - dev_err(dev, "Failed to ioremap mem resource: %d\n", ret); - return ret; - } + dpi->regs = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(dpi->regs)) + return dev_err_probe(dev, PTR_ERR(dpi->regs), + "Failed to ioremap mem resource\n"); dpi->engine_clk = devm_clk_get(dev, "engine"); - if (IS_ERR(dpi->engine_clk)) { - ret = PTR_ERR(dpi->engine_clk); - if (ret != -EPROBE_DEFER) - dev_err(dev, "Failed to get engine clock: %d\n", ret); - - return ret; - } + if (IS_ERR(dpi->engine_clk)) + return dev_err_probe(dev, PTR_ERR(dpi->engine_clk), + "Failed to get engine clock\n"); dpi->pixel_clk = devm_clk_get(dev, "pixel"); - if (IS_ERR(dpi->pixel_clk)) { - ret = PTR_ERR(dpi->pixel_clk); - if (ret != -EPROBE_DEFER) - dev_err(dev, "Failed to get pixel clock: %d\n", ret); - - return ret; - } + if (IS_ERR(dpi->pixel_clk)) + return dev_err_probe(dev, PTR_ERR(dpi->pixel_clk), + "Failed to get pixel clock\n"); dpi->tvd_clk = devm_clk_get(dev, "pll"); - if (IS_ERR(dpi->tvd_clk)) { - ret = PTR_ERR(dpi->tvd_clk); - if (ret != -EPROBE_DEFER) - dev_err(dev, "Failed to get tvdpll clock: %d\n", ret); - - return ret; - } + if (IS_ERR(dpi->tvd_clk)) + return dev_err_probe(dev, PTR_ERR(dpi->tvd_clk), + "Failed to get tvdpll clock\n"); dpi->irq = platform_get_irq(pdev, 0); - if (dpi->irq <= 0) - return -EINVAL; + if (dpi->irq < 0) + return dpi->irq; - ret = drm_of_find_panel_or_bridge(dev->of_node, 0, 0, - NULL, &dpi->next_bridge); - if (ret) - return ret; + dpi->next_bridge = devm_drm_of_get_bridge(dev, dev->of_node, 0, 0); + if (IS_ERR(dpi->next_bridge)) + return dev_err_probe(dev, PTR_ERR(dpi->next_bridge), + "Failed to get bridge\n"); dev_info(dev, "Found bridge node: %pOF\n", dpi->next_bridge->of_node); @@ -1089,49 +1073,31 @@ static int mtk_dpi_probe(struct platform_device *pdev) dpi->bridge.of_node = dev->of_node; dpi->bridge.type = DRM_MODE_CONNECTOR_DPI; - drm_bridge_add(&dpi->bridge); + ret = devm_drm_bridge_add(dev, &dpi->bridge); + if (ret) + return ret; ret = component_add(dev, &mtk_dpi_component_ops); - if (ret) { - drm_bridge_remove(&dpi->bridge); - dev_err(dev, "Failed to add component: %d\n", ret); - return ret; - } + if (ret) + return dev_err_probe(dev, ret, "Failed to add component.\n"); return 0; } static void mtk_dpi_remove(struct platform_device *pdev) { - struct mtk_dpi *dpi = platform_get_drvdata(pdev); - component_del(&pdev->dev, &mtk_dpi_component_ops); - drm_bridge_remove(&dpi->bridge); } static const struct of_device_id mtk_dpi_of_ids[] = { - { .compatible = "mediatek,mt2701-dpi", - .data = &mt2701_conf, - }, - { .compatible = "mediatek,mt8173-dpi", - .data = &mt8173_conf, - }, - { .compatible = "mediatek,mt8183-dpi", - .data = &mt8183_conf, - }, - { .compatible = "mediatek,mt8186-dpi", - .data = &mt8186_conf, - }, - { .compatible = "mediatek,mt8188-dp-intf", - .data = &mt8188_dpintf_conf, - }, - { .compatible = "mediatek,mt8192-dpi", - .data = &mt8192_conf, - }, - { .compatible = "mediatek,mt8195-dp-intf", - .data = &mt8195_dpintf_conf, - }, - { }, + { .compatible = "mediatek,mt2701-dpi", .data = &mt2701_conf }, + { .compatible = "mediatek,mt8173-dpi", .data = &mt8173_conf }, + { .compatible = "mediatek,mt8183-dpi", .data = &mt8183_conf }, + { .compatible = "mediatek,mt8186-dpi", .data = &mt8186_conf }, + { .compatible = "mediatek,mt8188-dp-intf", .data = &mt8188_dpintf_conf }, + { .compatible = "mediatek,mt8192-dpi", .data = &mt8192_conf }, + { .compatible = "mediatek,mt8195-dp-intf", .data = &mt8195_dpintf_conf }, + { /* sentinel */ }, }; MODULE_DEVICE_TABLE(of, mtk_dpi_of_ids); diff --git a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c index 8a43656ecc30..b6fa4ad2f94d 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_crtc.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_crtc.c @@ -117,10 +117,9 @@ static int mtk_drm_cmdq_pkt_create(struct cmdq_client *client, struct cmdq_pkt * dma_addr_t dma_addr; pkt->va_base = kzalloc(size, GFP_KERNEL); - if (!pkt->va_base) { - kfree(pkt); + if (!pkt->va_base) return -ENOMEM; - } + pkt->buf_size = size; pkt->cl = (void *)client; @@ -130,7 +129,6 @@ static int mtk_drm_cmdq_pkt_create(struct cmdq_client *client, struct cmdq_pkt * if (dma_mapping_error(dev, dma_addr)) { dev_err(dev, "dma map failed, size=%u\n", (u32)(u64)size); kfree(pkt->va_base); - kfree(pkt); return -ENOMEM; } @@ -146,7 +144,6 @@ static void mtk_drm_cmdq_pkt_destroy(struct cmdq_pkt *pkt) dma_unmap_single(client->chan->mbox->dev, pkt->pa_base, pkt->buf_size, DMA_TO_DEVICE); kfree(pkt->va_base); - kfree(pkt); } #endif diff --git a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c index f114da4d36a9..771f4e173353 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_ddp_comp.c @@ -563,14 +563,15 @@ int mtk_ddp_comp_init(struct device_node *node, struct mtk_ddp_comp *comp, /* Not all drm components have a DTS device node, such as ovl_adaptor, * which is the drm bring up sub driver */ - if (node) { - comp_pdev = of_find_device_by_node(node); - if (!comp_pdev) { - DRM_INFO("Waiting for device %s\n", node->full_name); - return -EPROBE_DEFER; - } - comp->dev = &comp_pdev->dev; + if (!node) + return 0; + + comp_pdev = of_find_device_by_node(node); + if (!comp_pdev) { + DRM_INFO("Waiting for device %s\n", node->full_name); + return -EPROBE_DEFER; } + comp->dev = &comp_pdev->dev; if (type == MTK_DISP_AAL || type == MTK_DISP_BLS || @@ -580,7 +581,6 @@ int mtk_ddp_comp_init(struct device_node *node, struct mtk_ddp_comp *comp, type == MTK_DISP_MERGE || type == MTK_DISP_OVL || type == MTK_DISP_OVL_2L || - type == MTK_DISP_OVL_ADAPTOR || type == MTK_DISP_PWM || type == MTK_DISP_RDMA || type == MTK_DPI || diff --git a/drivers/gpu/drm/mediatek/mtk_drm_drv.c b/drivers/gpu/drm/mediatek/mtk_drm_drv.c index d16cc8219105..93552d76b6e7 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_drv.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_drv.c @@ -355,7 +355,7 @@ static bool mtk_drm_get_all_drm_priv(struct device *dev) const struct of_device_id *of_id; struct device_node *node; struct device *drm_dev; - int cnt = 0; + unsigned int cnt = 0; int i, j; for_each_child_of_node(phandle->parent, node) { @@ -376,6 +376,9 @@ static bool mtk_drm_get_all_drm_priv(struct device *dev) all_drm_priv[cnt] = dev_get_drvdata(drm_dev); if (all_drm_priv[cnt] && all_drm_priv[cnt]->mtk_drm_bound) cnt++; + + if (cnt == MAX_CRTC) + break; } if (drm_priv->data->mmsys_dev_num == cnt) { @@ -827,7 +830,7 @@ static int mtk_drm_probe(struct platform_device *pdev) continue; } - comp_type = (enum mtk_ddp_comp_type)of_id->data; + comp_type = (enum mtk_ddp_comp_type)(uintptr_t)of_id->data; if (comp_type == MTK_DISP_MUTEX) { int id; diff --git a/drivers/gpu/drm/mediatek/mtk_drm_gem.c b/drivers/gpu/drm/mediatek/mtk_drm_gem.c index a25b28d3ee90..9f364df52478 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_gem.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_gem.c @@ -247,7 +247,11 @@ int mtk_drm_gem_prime_vmap(struct drm_gem_object *obj, struct iosys_map *map) mtk_gem->kvaddr = vmap(mtk_gem->pages, npages, VM_MAP, pgprot_writecombine(PAGE_KERNEL)); - + if (!mtk_gem->kvaddr) { + kfree(sgt); + kfree(mtk_gem->pages); + return -ENOMEM; + } out: kfree(sgt); iosys_map_set_vaddr(map, mtk_gem->kvaddr); diff --git a/drivers/gpu/drm/mediatek/mtk_drm_plane.c b/drivers/gpu/drm/mediatek/mtk_drm_plane.c index 31f9420aff6f..db2f70ae060d 100644 --- a/drivers/gpu/drm/mediatek/mtk_drm_plane.c +++ b/drivers/gpu/drm/mediatek/mtk_drm_plane.c @@ -122,11 +122,7 @@ static int mtk_plane_atomic_async_check(struct drm_plane *plane, if (ret) return ret; - if (state) - crtc_state = drm_atomic_get_existing_crtc_state(state, - new_plane_state->crtc); - else /* Special case for asynchronous cursor updates. */ - crtc_state = new_plane_state->crtc->state; + crtc_state = drm_atomic_get_existing_crtc_state(state, new_plane_state->crtc); return drm_atomic_helper_check_plane_state(plane->state, crtc_state, DRM_PLANE_NO_SCALING, diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c index c67089a7ebc1..0d8133f3174b 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -205,7 +205,7 @@ static int a2xx_hw_init(struct msm_gpu *gpu) A2XX_MH_INTERRUPT_MASK_MMU_PAGE_FAULT); for (i = 3; i <= 5; i++) - if ((SZ_16K << i) == adreno_gpu->gmem) + if ((SZ_16K << i) == adreno_gpu->info->gmem) break; gpu_write(gpu, REG_A2XX_RB_EDRAM_INFO, i); @@ -540,6 +540,10 @@ struct msm_gpu *a2xx_gpu_init(struct drm_device *dev) gpu->perfcntrs = perfcntrs; gpu->num_perfcntrs = ARRAY_SIZE(perfcntrs); + ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); + if (ret) + goto fail; + if (adreno_is_a20x(adreno_gpu)) adreno_gpu->registers = a200_registers; else if (adreno_is_a225(adreno_gpu)) @@ -547,10 +551,6 @@ struct msm_gpu *a2xx_gpu_init(struct drm_device *dev) else adreno_gpu->registers = a220_registers; - ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); - if (ret) - goto fail; - if (!gpu->aspace) { dev_err(dev->dev, "No memory protection without MMU\n"); if (!allow_vram_carveout) { diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index 715436cb3996..8b4cdf95f445 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -145,7 +145,7 @@ static void a4xx_enable_hwcg(struct msm_gpu *gpu) gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ, 0x00220000); /* Early A430's have a timing issue with SP/TP power collapse; disabling HW clock gating prevents it. */ - if (adreno_is_a430(adreno_gpu) && adreno_gpu->rev.patchid < 2) + if (adreno_is_a430(adreno_gpu) && adreno_patchid(adreno_gpu) < 2) gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0); else gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0xAAAAAAAA); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index a99310b68793..e5916c106796 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -66,7 +66,7 @@ void a5xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring, static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit) { struct msm_ringbuffer *ring = submit->ring; - struct msm_gem_object *obj; + struct drm_gem_object *obj; uint32_t *ptr, dwords; unsigned int i; @@ -83,13 +83,13 @@ static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit obj = submit->bos[submit->cmd[i].idx].obj; dwords = submit->cmd[i].size; - ptr = msm_gem_get_vaddr(&obj->base); + ptr = msm_gem_get_vaddr(obj); /* _get_vaddr() shouldn't fail at this point, * since we've already mapped it once in * submit_reloc() */ - if (WARN_ON(!ptr)) + if (WARN_ON(IS_ERR_OR_NULL(ptr))) return; for (i = 0; i < dwords; i++) { @@ -103,7 +103,7 @@ static void a5xx_submit_in_rb(struct msm_gpu *gpu, struct msm_gem_submit *submit OUT_RING(ring, ptr[i]); } - msm_gem_put_vaddr(&obj->base); + msm_gem_put_vaddr(obj); break; } @@ -749,7 +749,7 @@ static int a5xx_hw_init(struct msm_gpu *gpu) gpu_write(gpu, REG_A5XX_UCHE_GMEM_RANGE_MIN_LO, 0x00100000); gpu_write(gpu, REG_A5XX_UCHE_GMEM_RANGE_MIN_HI, 0x00000000); gpu_write(gpu, REG_A5XX_UCHE_GMEM_RANGE_MAX_LO, - 0x00100000 + adreno_gpu->gmem - 1); + 0x00100000 + adreno_gpu->info->gmem - 1); gpu_write(gpu, REG_A5XX_UCHE_GMEM_RANGE_MAX_HI, 0x00000000); if (adreno_is_a506(adreno_gpu) || adreno_is_a508(adreno_gpu) || @@ -1770,7 +1770,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) nr_rings = 4; - if (adreno_cmp_rev(ADRENO_REV(5, 1, 0, ANY_ID), config->rev)) + if (config->info->revn == 510) nr_rings = 1; ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, nr_rings); diff --git a/drivers/gpu/drm/msm/adreno/a5xx_power.c b/drivers/gpu/drm/msm/adreno/a5xx_power.c index 0e63a1429189..7705f8010484 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_power.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_power.c @@ -179,7 +179,7 @@ static void a540_lm_setup(struct msm_gpu *gpu) /* The battery current limiter isn't enabled for A540 */ config = AGC_LM_CONFIG_BCL_DISABLED; - config |= adreno_gpu->rev.patchid << AGC_LM_CONFIG_GPU_VERSION_SHIFT; + config |= adreno_patchid(adreno_gpu) << AGC_LM_CONFIG_GPU_VERSION_SHIFT; /* For now disable GPMU side throttling */ config |= AGC_LM_CONFIG_THROTTLE_DISABLE; diff --git a/drivers/gpu/drm/msm/adreno/a6xx.xml.h b/drivers/gpu/drm/msm/adreno/a6xx.xml.h index 4dc3be6ed45d..1c051535fd4a 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a6xx.xml.h @@ -1166,6 +1166,9 @@ static inline uint32_t A6XX_CP_ROQ_THRESHOLDS_2_ROQ_SIZE(uint32_t val) #define REG_A6XX_CP_DBG_ECO_CNTL 0x00000843 #define REG_A6XX_CP_PROTECT_CNTL 0x0000084f +#define A6XX_CP_PROTECT_CNTL_LAST_SPAN_INF_RANGE 0x00000008 +#define A6XX_CP_PROTECT_CNTL_ACCESS_FAULT_ON_VIOL_EN 0x00000002 +#define A6XX_CP_PROTECT_CNTL_ACCESS_PROT_EN 0x00000001 static inline uint32_t REG_A6XX_CP_SCRATCH(uint32_t i0) { return 0x00000883 + 0x1*i0; } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index b20ef6c8ea26..7923129363b0 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -678,12 +678,6 @@ struct block_header { u32 data[]; }; -/* this should be a general kernel helper */ -static int in_range(u32 addr, u32 start, u32 size) -{ - return addr >= start && addr < start + size; -} - static bool fw_block_mem(struct a6xx_gmu_bo *bo, const struct block_header *blk) { if (!in_range(blk->addr, bo->iova, bo->size)) @@ -792,10 +786,22 @@ static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state) gmu_write(gmu, REG_A6XX_GMU_AHB_FENCE_RANGE_0, (1 << 31) | (0xa << 18) | (0xa0)); - chipid = adreno_gpu->rev.core << 24; - chipid |= adreno_gpu->rev.major << 16; - chipid |= adreno_gpu->rev.minor << 12; - chipid |= adreno_gpu->rev.patchid << 8; + /* + * Snapshots toggle the NMI bit which will result in a jump to the NMI + * handler instead of __main. Set the M3 config value to avoid that. + */ + gmu_write(gmu, REG_A6XX_GMU_CM3_CFG, 0x4052); + + /* + * Note that the GMU has a slightly different layout for + * chip_id, for whatever reason, so a bit of massaging + * is needed. The upper 16b are the same, but minor and + * patchid are packed in four bits each with the lower + * 8b unused: + */ + chipid = adreno_gpu->chip_id & 0xffff0000; + chipid |= (adreno_gpu->chip_id << 4) & 0xf000; /* minor */ + chipid |= (adreno_gpu->chip_id << 8) & 0x0f00; /* patchid */ gmu_write(gmu, REG_A6XX_GMU_HFI_SFR_ADDR, chipid); @@ -889,6 +895,13 @@ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu) /* Make sure there are no outstanding RPMh votes */ a6xx_gmu_rpmh_off(gmu); + /* Clear the WRITEDROPPED fields and put fence into allow mode */ + gmu_write(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS_CLR, 0x7); + gmu_write(gmu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, 0); + + /* Make sure the above writes go through */ + wmb(); + /* Halt the gmu cm3 core */ gmu_write(gmu, REG_A6XX_GMU_CM3_SYSRESET, 1); @@ -1437,8 +1450,15 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) struct a6xx_gmu *gmu = &a6xx_gpu->gmu; struct platform_device *pdev = to_platform_device(gmu->dev); - if (!gmu->initialized) + mutex_lock(&gmu->lock); + if (!gmu->initialized) { + mutex_unlock(&gmu->lock); return; + } + + gmu->initialized = false; + + mutex_unlock(&gmu->lock); pm_runtime_force_suspend(gmu->dev); @@ -1468,8 +1488,6 @@ void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu) /* Drop reference taken in of_find_device_by_node */ put_device(gmu->dev); - - gmu->initialized = false; } static int cxpd_notifier_cb(struct notifier_block *nb, @@ -1619,7 +1637,7 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) SZ_256K - SZ_16K, 0x44000, "dcache"); if (ret) goto err_memory; - } else if (adreno_is_a630(adreno_gpu) || adreno_is_a615_family(adreno_gpu)) { + } else if (adreno_is_a630_family(adreno_gpu)) { /* HFI v1, has sptprac */ gmu->legacy = true; @@ -1629,13 +1647,13 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node) goto err_memory; } - /* Allocate memory for for the HFI queues */ - ret = a6xx_gmu_memory_alloc(gmu, &gmu->hfi, SZ_16K, 0, "hfi"); + /* Allocate memory for the GMU log region */ + ret = a6xx_gmu_memory_alloc(gmu, &gmu->log, SZ_16K, 0, "log"); if (ret) goto err_memory; - /* Allocate memory for the GMU log region */ - ret = a6xx_gmu_memory_alloc(gmu, &gmu->log, SZ_4K, 0, "log"); + /* Allocate memory for for the HFI queues */ + ret = a6xx_gmu_memory_alloc(gmu, &gmu->hfi, SZ_16K, 0, "hfi"); if (ret) goto err_memory; diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h index 9ab15d91aced..fcd9eb53baf8 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h @@ -425,6 +425,8 @@ static inline uint32_t A6XX_GMU_GPU_NAP_CTRL_SID(uint32_t val) #define REG_A6XX_GMU_AHB_FENCE_STATUS 0x00009313 +#define REG_A6XX_GMU_AHB_FENCE_STATUS_CLR 0x00009314 + #define REG_A6XX_GMU_RBBM_INT_UNMASKED_STATUS 0x00009315 #define REG_A6XX_GMU_AO_SPARE_CNTL 0x00009316 diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index b3ada1e7b598..d4e85e24002f 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -930,10 +930,16 @@ static void a6xx_set_cp_protect(struct msm_gpu *gpu) * protect violation and select the last span to protect from the start * address all the way to the end of the register address space */ - gpu_write(gpu, REG_A6XX_CP_PROTECT_CNTL, BIT(0) | BIT(1) | BIT(3)); - - for (i = 0; i < count - 1; i++) - gpu_write(gpu, REG_A6XX_CP_PROTECT(i), regs[i]); + gpu_write(gpu, REG_A6XX_CP_PROTECT_CNTL, + A6XX_CP_PROTECT_CNTL_ACCESS_PROT_EN | + A6XX_CP_PROTECT_CNTL_ACCESS_FAULT_ON_VIOL_EN | + A6XX_CP_PROTECT_CNTL_LAST_SPAN_INF_RANGE); + + for (i = 0; i < count - 1; i++) { + /* Intentionally skip writing to some registers */ + if (regs[i]) + gpu_write(gpu, REG_A6XX_CP_PROTECT(i), regs[i]); + } /* last CP_PROTECT to have "infinite" length on the last entry */ gpu_write(gpu, REG_A6XX_CP_PROTECT(count_max - 1), regs[i]); } @@ -1195,7 +1201,9 @@ static int hw_init(struct msm_gpu *gpu) if (!adreno_has_gmu_wrapper(adreno_gpu)) { /* Make sure the GMU keeps the GPU on while we set it up */ - a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + ret = a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET); + if (ret) + return ret; } /* Clear GBIF halt in case GX domain was not collapsed */ @@ -1270,7 +1278,7 @@ static int hw_init(struct msm_gpu *gpu) gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MIN, 0x00100000); gpu_write64(gpu, REG_A6XX_UCHE_GMEM_RANGE_MAX, - 0x00100000 + adreno_gpu->gmem - 1); + 0x00100000 + adreno_gpu->info->gmem - 1); } gpu_write(gpu, REG_A6XX_UCHE_FILTER_CNTL, 0x804); @@ -1729,16 +1737,6 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu) return IRQ_HANDLED; } -static void a6xx_llc_rmw(struct a6xx_gpu *a6xx_gpu, u32 reg, u32 mask, u32 or) -{ - return msm_rmw(a6xx_gpu->llc_mmio + (reg << 2), mask, or); -} - -static void a6xx_llc_write(struct a6xx_gpu *a6xx_gpu, u32 reg, u32 value) -{ - msm_writel(value, a6xx_gpu->llc_mmio + (reg << 2)); -} - static void a6xx_llc_deactivate(struct a6xx_gpu *a6xx_gpu) { llcc_slice_deactivate(a6xx_gpu->llc_slice); @@ -2091,9 +2089,7 @@ static void a6xx_destroy(struct msm_gpu *gpu) a6xx_llc_slices_destroy(a6xx_gpu); - mutex_lock(&a6xx_gpu->gmu.lock); a6xx_gmu_remove(a6xx_gpu); - mutex_unlock(&a6xx_gpu->gmu.lock); adreno_gpu_cleanup(adreno_gpu); @@ -2204,159 +2200,19 @@ static bool a6xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring) return progress; } -static u32 a610_get_speed_bin(u32 fuse) -{ - /* - * There are (at least) three SoCs implementing A610: SM6125 (trinket), - * SM6115 (bengal) and SM6225 (khaje). Trinket does not have speedbinning, - * as only a single SKU exists and we don't support khaje upstream yet. - * Hence, this matching table is only valid for bengal and can be easily - * expanded if need be. - */ - - if (fuse == 0) - return 0; - else if (fuse == 206) - return 1; - else if (fuse == 200) - return 2; - else if (fuse == 157) - return 3; - else if (fuse == 127) - return 4; - - return UINT_MAX; -} - -static u32 a618_get_speed_bin(u32 fuse) -{ - if (fuse == 0) - return 0; - else if (fuse == 169) - return 1; - else if (fuse == 174) - return 2; - - return UINT_MAX; -} - -static u32 a619_holi_get_speed_bin(u32 fuse) -{ - /* - * There are (at least) two SoCs implementing A619_holi: SM4350 (holi) - * and SM6375 (blair). Limit the fuse matching to the corresponding - * SoC to prevent bogus frequency setting (as improbable as it may be, - * given unexpected fuse values are.. unexpected! But still possible.) - */ - - if (fuse == 0) - return 0; - - if (of_machine_is_compatible("qcom,sm4350")) { - if (fuse == 138) - return 1; - else if (fuse == 92) - return 2; - } else if (of_machine_is_compatible("qcom,sm6375")) { - if (fuse == 190) - return 1; - else if (fuse == 177) - return 2; - } else - pr_warn("Unknown SoC implementing A619_holi!\n"); - - return UINT_MAX; -} - -static u32 a619_get_speed_bin(u32 fuse) -{ - if (fuse == 0) - return 0; - else if (fuse == 120) - return 4; - else if (fuse == 138) - return 3; - else if (fuse == 169) - return 2; - else if (fuse == 180) - return 1; - - return UINT_MAX; -} - -static u32 a640_get_speed_bin(u32 fuse) +static u32 fuse_to_supp_hw(const struct adreno_info *info, u32 fuse) { - if (fuse == 0) - return 0; - else if (fuse == 1) - return 1; - - return UINT_MAX; -} - -static u32 a650_get_speed_bin(u32 fuse) -{ - if (fuse == 0) - return 0; - else if (fuse == 1) - return 1; - /* Yep, 2 and 3 are swapped! :/ */ - else if (fuse == 2) - return 3; - else if (fuse == 3) - return 2; - - return UINT_MAX; -} + if (!info->speedbins) + return UINT_MAX; -static u32 adreno_7c3_get_speed_bin(u32 fuse) -{ - if (fuse == 0) - return 0; - else if (fuse == 117) - return 0; - else if (fuse == 190) - return 1; + for (int i = 0; info->speedbins[i].fuse != SHRT_MAX; i++) + if (info->speedbins[i].fuse == fuse) + return BIT(info->speedbins[i].speedbin); return UINT_MAX; } -static u32 fuse_to_supp_hw(struct device *dev, struct adreno_gpu *adreno_gpu, u32 fuse) -{ - u32 val = UINT_MAX; - - if (adreno_is_a610(adreno_gpu)) - val = a610_get_speed_bin(fuse); - - if (adreno_is_a618(adreno_gpu)) - val = a618_get_speed_bin(fuse); - - else if (adreno_is_a619_holi(adreno_gpu)) - val = a619_holi_get_speed_bin(fuse); - - else if (adreno_is_a619(adreno_gpu)) - val = a619_get_speed_bin(fuse); - - else if (adreno_is_7c3(adreno_gpu)) - val = adreno_7c3_get_speed_bin(fuse); - - else if (adreno_is_a640(adreno_gpu)) - val = a640_get_speed_bin(fuse); - - else if (adreno_is_a650(adreno_gpu)) - val = a650_get_speed_bin(fuse); - - if (val == UINT_MAX) { - DRM_DEV_ERROR(dev, - "missing support for speed-bin: %u. Some OPPs may not be supported by hardware\n", - fuse); - return UINT_MAX; - } - - return (1 << val); -} - -static int a6xx_set_supported_hw(struct device *dev, struct adreno_gpu *adreno_gpu) +static int a6xx_set_supported_hw(struct device *dev, const struct adreno_info *info) { u32 supp_hw; u32 speedbin; @@ -2375,7 +2231,14 @@ static int a6xx_set_supported_hw(struct device *dev, struct adreno_gpu *adreno_g return ret; } - supp_hw = fuse_to_supp_hw(dev, adreno_gpu, speedbin); + supp_hw = fuse_to_supp_hw(info, speedbin); + + if (supp_hw == UINT_MAX) { + DRM_DEV_ERROR(dev, + "missing support for speed-bin: %u. Some OPPs may not be supported by hardware\n", + speedbin); + return UINT_MAX; + } ret = devm_pm_opp_set_supported_hw(dev, &supp_hw, 1); if (ret) @@ -2449,7 +2312,6 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) struct msm_drm_private *priv = dev->dev_private; struct platform_device *pdev = priv->gpu_pdev; struct adreno_platform_config *config = pdev->dev.platform_data; - const struct adreno_info *info; struct device_node *node; struct a6xx_gpu *a6xx_gpu; struct adreno_gpu *adreno_gpu; @@ -2474,29 +2336,12 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) adreno_gpu->gmu_is_wrapper = of_device_is_compatible(node, "qcom,adreno-gmu-wrapper"); - /* - * We need to know the platform type before calling into adreno_gpu_init - * so that the hw_apriv flag can be correctly set. Snoop into the info - * and grab the revision number - */ - info = adreno_info(config->rev); - if (!info) - return ERR_PTR(-EINVAL); - - /* Assign these early so that we can use the is_aXYZ helpers */ - /* Numeric revision IDs (e.g. 630) */ - adreno_gpu->revn = info->revn; - /* New-style ADRENO_REV()-only */ - adreno_gpu->rev = info->rev; - /* Quirk data */ - adreno_gpu->info = info; - - if (adreno_is_a650(adreno_gpu) || adreno_is_a660_family(adreno_gpu)) - adreno_gpu->base.hw_apriv = true; + adreno_gpu->base.hw_apriv = + !!(config->info->quirks & ADRENO_QUIRK_HAS_HW_APRIV); a6xx_llc_slices_init(pdev, a6xx_gpu); - ret = a6xx_set_supported_hw(&pdev->dev, adreno_gpu); + ret = a6xx_set_supported_hw(&pdev->dev, config->info); if (ret) { a6xx_destroy(&(a6xx_gpu->base.base)); return ERR_PTR(ret); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h index c788b06e72da..34822b080759 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h @@ -39,8 +39,8 @@ struct a6xx_gpu { /* * Given a register and a count, return a value to program into - * REG_CP_PROTECT_REG(n) - this will block both reads and writes for _len - * registers starting at _reg. + * REG_CP_PROTECT_REG(n) - this will block both reads and writes for + * _len + 1 registers starting at _reg. */ #define A6XX_PROTECT_NORDWR(_reg, _len) \ ((1 << 31) | \ @@ -62,6 +62,21 @@ static inline bool a6xx_has_gbif(struct adreno_gpu *gpu) return true; } +static inline void a6xx_llc_rmw(struct a6xx_gpu *a6xx_gpu, u32 reg, u32 mask, u32 or) +{ + return msm_rmw(a6xx_gpu->llc_mmio + (reg << 2), mask, or); +} + +static inline u32 a6xx_llc_read(struct a6xx_gpu *a6xx_gpu, u32 reg) +{ + return msm_readl(a6xx_gpu->llc_mmio + (reg << 2)); +} + +static inline void a6xx_llc_write(struct a6xx_gpu *a6xx_gpu, u32 reg, u32 value) +{ + msm_writel(value, a6xx_gpu->llc_mmio + (reg << 2)); +} + #define shadowptr(_a6xx_gpu, _ring) ((_a6xx_gpu)->shadow_iova + \ ((_ring)->id * sizeof(uint32_t))) diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h index 790f55e24533..e788ed72eb0d 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h @@ -206,7 +206,7 @@ static const struct a6xx_shader_block { SHADER(A6XX_SP_LB_3_DATA, 0x800), SHADER(A6XX_SP_LB_4_DATA, 0x800), SHADER(A6XX_SP_LB_5_DATA, 0x200), - SHADER(A6XX_SP_CB_BINDLESS_DATA, 0x2000), + SHADER(A6XX_SP_CB_BINDLESS_DATA, 0x800), SHADER(A6XX_SP_CB_LEGACY_DATA, 0x280), SHADER(A6XX_SP_UAV_DATA, 0x80), SHADER(A6XX_SP_INST_TAG, 0x80), diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 26a27cb039e9..fa527935ffd4 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -22,9 +22,9 @@ module_param_named(allow_vram_carveout, allow_vram_carveout, bool, 0600); static const struct adreno_info gpulist[] = { { - .rev = ADRENO_REV(2, 0, 0, 0), + .chip_ids = ADRENO_CHIP_IDS(0x02000000), + .family = ADRENO_2XX_GEN1, .revn = 200, - .name = "A200", .fw = { [ADRENO_FW_PM4] = "yamato_pm4.fw", [ADRENO_FW_PFP] = "yamato_pfp.fw", @@ -33,9 +33,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a2xx_gpu_init, }, { /* a200 on i.mx51 has only 128kib gmem */ - .rev = ADRENO_REV(2, 0, 0, 1), + .chip_ids = ADRENO_CHIP_IDS(0x02000001), + .family = ADRENO_2XX_GEN1, .revn = 201, - .name = "A200", .fw = { [ADRENO_FW_PM4] = "yamato_pm4.fw", [ADRENO_FW_PFP] = "yamato_pfp.fw", @@ -44,9 +44,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a2xx_gpu_init, }, { - .rev = ADRENO_REV(2, 2, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x02020000), + .family = ADRENO_2XX_GEN2, .revn = 220, - .name = "A220", .fw = { [ADRENO_FW_PM4] = "leia_pm4_470.fw", [ADRENO_FW_PFP] = "leia_pfp_470.fw", @@ -55,9 +55,12 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a2xx_gpu_init, }, { - .rev = ADRENO_REV(3, 0, 5, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS( + 0x03000512, + 0x03000520 + ), + .family = ADRENO_3XX, .revn = 305, - .name = "A305", .fw = { [ADRENO_FW_PM4] = "a300_pm4.fw", [ADRENO_FW_PFP] = "a300_pfp.fw", @@ -66,9 +69,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a3xx_gpu_init, }, { - .rev = ADRENO_REV(3, 0, 6, 0), + .chip_ids = ADRENO_CHIP_IDS(0x03000600), + .family = ADRENO_3XX, .revn = 307, /* because a305c is revn==306 */ - .name = "A306", .fw = { [ADRENO_FW_PM4] = "a300_pm4.fw", [ADRENO_FW_PFP] = "a300_pfp.fw", @@ -77,9 +80,13 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a3xx_gpu_init, }, { - .rev = ADRENO_REV(3, 2, ANY_ID, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS( + 0x03020000, + 0x03020001, + 0x03020002 + ), + .family = ADRENO_3XX, .revn = 320, - .name = "A320", .fw = { [ADRENO_FW_PM4] = "a300_pm4.fw", [ADRENO_FW_PFP] = "a300_pfp.fw", @@ -88,9 +95,13 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a3xx_gpu_init, }, { - .rev = ADRENO_REV(3, 3, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS( + 0x03030000, + 0x03030001, + 0x03030002 + ), + .family = ADRENO_3XX, .revn = 330, - .name = "A330", .fw = { [ADRENO_FW_PM4] = "a330_pm4.fw", [ADRENO_FW_PFP] = "a330_pfp.fw", @@ -99,9 +110,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a3xx_gpu_init, }, { - .rev = ADRENO_REV(4, 0, 5, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x04000500), + .family = ADRENO_4XX, .revn = 405, - .name = "A405", .fw = { [ADRENO_FW_PM4] = "a420_pm4.fw", [ADRENO_FW_PFP] = "a420_pfp.fw", @@ -110,9 +121,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a4xx_gpu_init, }, { - .rev = ADRENO_REV(4, 2, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x04020000), + .family = ADRENO_4XX, .revn = 420, - .name = "A420", .fw = { [ADRENO_FW_PM4] = "a420_pm4.fw", [ADRENO_FW_PFP] = "a420_pfp.fw", @@ -121,9 +132,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a4xx_gpu_init, }, { - .rev = ADRENO_REV(4, 3, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x04030002), + .family = ADRENO_4XX, .revn = 430, - .name = "A430", .fw = { [ADRENO_FW_PM4] = "a420_pm4.fw", [ADRENO_FW_PFP] = "a420_pfp.fw", @@ -132,9 +143,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a4xx_gpu_init, }, { - .rev = ADRENO_REV(5, 0, 6, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x05000600), + .family = ADRENO_5XX, .revn = 506, - .name = "A506", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -150,9 +161,9 @@ static const struct adreno_info gpulist[] = { .init = a5xx_gpu_init, .zapfw = "a506_zap.mdt", }, { - .rev = ADRENO_REV(5, 0, 8, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x05000800), + .family = ADRENO_5XX, .revn = 508, - .name = "A508", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -167,9 +178,9 @@ static const struct adreno_info gpulist[] = { .init = a5xx_gpu_init, .zapfw = "a508_zap.mdt", }, { - .rev = ADRENO_REV(5, 0, 9, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x05000900), + .family = ADRENO_5XX, .revn = 509, - .name = "A509", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -185,9 +196,9 @@ static const struct adreno_info gpulist[] = { /* Adreno 509 uses the same ZAP as 512 */ .zapfw = "a512_zap.mdt", }, { - .rev = ADRENO_REV(5, 1, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x05010000), + .family = ADRENO_5XX, .revn = 510, - .name = "A510", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -200,9 +211,9 @@ static const struct adreno_info gpulist[] = { .inactive_period = 250, .init = a5xx_gpu_init, }, { - .rev = ADRENO_REV(5, 1, 2, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x05010200), + .family = ADRENO_5XX, .revn = 512, - .name = "A512", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -217,9 +228,12 @@ static const struct adreno_info gpulist[] = { .init = a5xx_gpu_init, .zapfw = "a512_zap.mdt", }, { - .rev = ADRENO_REV(5, 3, 0, 2), + .chip_ids = ADRENO_CHIP_IDS( + 0x05030002, + 0x05030004 + ), + .family = ADRENO_5XX, .revn = 530, - .name = "A530", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -236,9 +250,9 @@ static const struct adreno_info gpulist[] = { .init = a5xx_gpu_init, .zapfw = "a530_zap.mdt", }, { - .rev = ADRENO_REV(5, 4, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x05040001), + .family = ADRENO_5XX, .revn = 540, - .name = "A540", .fw = { [ADRENO_FW_PM4] = "a530_pm4.fw", [ADRENO_FW_PFP] = "a530_pfp.fw", @@ -254,9 +268,9 @@ static const struct adreno_info gpulist[] = { .init = a5xx_gpu_init, .zapfw = "a540_zap.mdt", }, { - .rev = ADRENO_REV(6, 1, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06010000), + .family = ADRENO_6XX_GEN1, .revn = 610, - .name = "A610", .fw = { [ADRENO_FW_SQE] = "a630_sqe.fw", }, @@ -265,21 +279,61 @@ static const struct adreno_info gpulist[] = { .init = a6xx_gpu_init, .zapfw = "a610_zap.mdt", .hwcg = a612_hwcg, + /* + * There are (at least) three SoCs implementing A610: SM6125 + * (trinket), SM6115 (bengal) and SM6225 (khaje). Trinket does + * not have speedbinning, as only a single SKU exists and we + * don't support khaje upstream yet. Hence, this matching + * table is only valid for bengal. + */ + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 206, 1 }, + { 200, 2 }, + { 157, 3 }, + { 127, 4 }, + ), }, { - .rev = ADRENO_REV(6, 1, 8, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06010800), + .family = ADRENO_6XX_GEN1, .revn = 618, - .name = "A618", .fw = { [ADRENO_FW_SQE] = "a630_sqe.fw", [ADRENO_FW_GMU] = "a630_gmu.bin", }, .gmem = SZ_512K, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT, + .init = a6xx_gpu_init, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 169, 1 }, + { 174, 2 }, + ), + }, { + .machine = "qcom,sm4350", + .chip_ids = ADRENO_CHIP_IDS(0x06010900), + .family = ADRENO_6XX_GEN1, + .revn = 619, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a619_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, .init = a6xx_gpu_init, + .zapfw = "a615_zap.mdt", + .hwcg = a615_hwcg, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 138, 1 }, + { 92, 2 }, + ), }, { - .rev = ADRENO_REV(6, 1, 9, ANY_ID), + .machine = "qcom,sm6375", + .chip_ids = ADRENO_CHIP_IDS(0x06010900), + .family = ADRENO_6XX_GEN1, .revn = 619, - .name = "A619", .fw = { [ADRENO_FW_SQE] = "a630_sqe.fw", [ADRENO_FW_GMU] = "a619_gmu.bin", @@ -289,94 +343,149 @@ static const struct adreno_info gpulist[] = { .init = a6xx_gpu_init, .zapfw = "a615_zap.mdt", .hwcg = a615_hwcg, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 190, 1 }, + { 177, 2 }, + ), }, { - .rev = ADRENO_REV(6, 3, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06010900), + .family = ADRENO_6XX_GEN1, + .revn = 619, + .fw = { + [ADRENO_FW_SQE] = "a630_sqe.fw", + [ADRENO_FW_GMU] = "a619_gmu.bin", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT, + .init = a6xx_gpu_init, + .zapfw = "a615_zap.mdt", + .hwcg = a615_hwcg, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 120, 4 }, + { 138, 3 }, + { 169, 2 }, + { 180, 1 }, + ), + }, { + .chip_ids = ADRENO_CHIP_IDS( + 0x06030001, + 0x06030002 + ), + .family = ADRENO_6XX_GEN1, .revn = 630, - .name = "A630", .fw = { [ADRENO_FW_SQE] = "a630_sqe.fw", [ADRENO_FW_GMU] = "a630_gmu.bin", }, .gmem = SZ_1M, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT, .init = a6xx_gpu_init, .zapfw = "a630_zap.mdt", .hwcg = a630_hwcg, }, { - .rev = ADRENO_REV(6, 4, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06040001), + .family = ADRENO_6XX_GEN2, .revn = 640, - .name = "A640", .fw = { [ADRENO_FW_SQE] = "a630_sqe.fw", [ADRENO_FW_GMU] = "a640_gmu.bin", }, .gmem = SZ_1M, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT, .init = a6xx_gpu_init, .zapfw = "a640_zap.mdt", .hwcg = a640_hwcg, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 1, 1 }, + ), }, { - .rev = ADRENO_REV(6, 5, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06050002), + .family = ADRENO_6XX_GEN3, .revn = 650, - .name = "A650", .fw = { [ADRENO_FW_SQE] = "a650_sqe.fw", [ADRENO_FW_GMU] = "a650_gmu.bin", }, .gmem = SZ_1M + SZ_128K, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, .init = a6xx_gpu_init, .zapfw = "a650_zap.mdt", .hwcg = a650_hwcg, .address_space_size = SZ_16G, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 1, 1 }, + { 2, 3 }, /* Yep, 2 and 3 are swapped! :/ */ + { 3, 2 }, + ), }, { - .rev = ADRENO_REV(6, 6, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06060001), + .family = ADRENO_6XX_GEN4, .revn = 660, - .name = "A660", .fw = { [ADRENO_FW_SQE] = "a660_sqe.fw", [ADRENO_FW_GMU] = "a660_gmu.bin", }, .gmem = SZ_1M + SZ_512K, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, .init = a6xx_gpu_init, .zapfw = "a660_zap.mdt", .hwcg = a660_hwcg, .address_space_size = SZ_16G, }, { - .rev = ADRENO_REV(6, 3, 5, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06030500), + .family = ADRENO_6XX_GEN4, .fw = { [ADRENO_FW_SQE] = "a660_sqe.fw", [ADRENO_FW_GMU] = "a660_gmu.bin", }, .gmem = SZ_512K, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, .init = a6xx_gpu_init, .hwcg = a660_hwcg, .address_space_size = SZ_16G, + .speedbins = ADRENO_SPEEDBINS( + { 0, 0 }, + { 117, 0 }, + { 190, 1 }, + ), }, { - .rev = ADRENO_REV(6, 8, 0, ANY_ID), + .chip_ids = ADRENO_CHIP_IDS(0x06080000), + .family = ADRENO_6XX_GEN2, .revn = 680, - .name = "A680", .fw = { [ADRENO_FW_SQE] = "a630_sqe.fw", [ADRENO_FW_GMU] = "a640_gmu.bin", }, .gmem = SZ_2M, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT, .init = a6xx_gpu_init, .zapfw = "a640_zap.mdt", .hwcg = a640_hwcg, }, { - .rev = ADRENO_REV(6, 9, 0, ANY_ID), - .revn = 690, - .name = "A690", + .chip_ids = ADRENO_CHIP_IDS(0x06090000), + .family = ADRENO_6XX_GEN4, .fw = { [ADRENO_FW_SQE] = "a660_sqe.fw", - [ADRENO_FW_GMU] = "a690_gmu.bin", + [ADRENO_FW_GMU] = "a660_gmu.bin", }, .gmem = SZ_4M, .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT | + ADRENO_QUIRK_HAS_HW_APRIV, .init = a6xx_gpu_init, .zapfw = "a690_zap.mdt", .hwcg = a690_hwcg, @@ -397,34 +506,31 @@ MODULE_FIRMWARE("qcom/a530_zap.mdt"); MODULE_FIRMWARE("qcom/a530_zap.b00"); MODULE_FIRMWARE("qcom/a530_zap.b01"); MODULE_FIRMWARE("qcom/a530_zap.b02"); +MODULE_FIRMWARE("qcom/a540_gpmu.fw2"); MODULE_FIRMWARE("qcom/a619_gmu.bin"); MODULE_FIRMWARE("qcom/a630_sqe.fw"); MODULE_FIRMWARE("qcom/a630_gmu.bin"); MODULE_FIRMWARE("qcom/a630_zap.mbn"); - -static inline bool _rev_match(uint8_t entry, uint8_t id) -{ - return (entry == ANY_ID) || (entry == id); -} - -bool adreno_cmp_rev(struct adreno_rev rev1, struct adreno_rev rev2) -{ - - return _rev_match(rev1.core, rev2.core) && - _rev_match(rev1.major, rev2.major) && - _rev_match(rev1.minor, rev2.minor) && - _rev_match(rev1.patchid, rev2.patchid); -} - -const struct adreno_info *adreno_info(struct adreno_rev rev) +MODULE_FIRMWARE("qcom/a640_gmu.bin"); +MODULE_FIRMWARE("qcom/a650_gmu.bin"); +MODULE_FIRMWARE("qcom/a650_sqe.fw"); +MODULE_FIRMWARE("qcom/a660_gmu.bin"); +MODULE_FIRMWARE("qcom/a660_sqe.fw"); +MODULE_FIRMWARE("qcom/leia_pfp_470.fw"); +MODULE_FIRMWARE("qcom/leia_pm4_470.fw"); +MODULE_FIRMWARE("qcom/yamato_pfp.fw"); +MODULE_FIRMWARE("qcom/yamato_pm4.fw"); + +static const struct adreno_info *adreno_info(uint32_t chip_id) { - int i; - /* identify gpu: */ - for (i = 0; i < ARRAY_SIZE(gpulist); i++) { + for (int i = 0; i < ARRAY_SIZE(gpulist); i++) { const struct adreno_info *info = &gpulist[i]; - if (adreno_cmp_rev(info->rev, rev)) - return info; + if (info->machine && !of_machine_is_compatible(info->machine)) + continue; + for (int j = 0; info->chip_ids[j]; j++) + if (info->chip_ids[j] == chip_id) + return info; } return NULL; @@ -504,12 +610,11 @@ err_disable_rpm: return NULL; } -static int find_chipid(struct device *dev, struct adreno_rev *rev) +static int find_chipid(struct device *dev, uint32_t *chipid) { struct device_node *node = dev->of_node; const char *compat; int ret; - u32 chipid; /* first search the compat strings for qcom,adreno-XYZ.W: */ ret = of_property_read_string_index(node, "compatible", 0, &compat); @@ -518,32 +623,34 @@ static int find_chipid(struct device *dev, struct adreno_rev *rev) if (sscanf(compat, "qcom,adreno-%u.%u", &r, &patch) == 2 || sscanf(compat, "amd,imageon-%u.%u", &r, &patch) == 2) { - rev->core = r / 100; + uint32_t core, major, minor; + + core = r / 100; r %= 100; - rev->major = r / 10; + major = r / 10; r %= 10; - rev->minor = r; - rev->patchid = patch; + minor = r; + + *chipid = (core << 24) | + (major << 16) | + (minor << 8) | + patch; return 0; } + + if (sscanf(compat, "qcom,adreno-%08x", chipid) == 1) + return 0; } /* and if that fails, fall back to legacy "qcom,chipid" property: */ - ret = of_property_read_u32(node, "qcom,chipid", &chipid); + ret = of_property_read_u32(node, "qcom,chipid", chipid); if (ret) { DRM_DEV_ERROR(dev, "could not parse qcom,chipid: %d\n", ret); return ret; } - rev->core = (chipid >> 24) & 0xff; - rev->major = (chipid >> 16) & 0xff; - rev->minor = (chipid >> 8) & 0xff; - rev->patchid = (chipid & 0xff); - dev_warn(dev, "Using legacy qcom,chipid binding!\n"); - dev_warn(dev, "Use compatible qcom,adreno-%u%u%u.%u instead.\n", - rev->core, rev->major, rev->minor, rev->patchid); return 0; } @@ -557,26 +664,27 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) struct msm_gpu *gpu; int ret; - ret = find_chipid(dev, &config.rev); + ret = find_chipid(dev, &config.chip_id); if (ret) return ret; dev->platform_data = &config; priv->gpu_pdev = to_platform_device(dev); - info = adreno_info(config.rev); - + info = adreno_info(config.chip_id); if (!info) { - dev_warn(drm->dev, "Unknown GPU revision: %u.%u.%u.%u\n", - config.rev.core, config.rev.major, - config.rev.minor, config.rev.patchid); + dev_warn(drm->dev, "Unknown GPU revision: %"ADRENO_CHIPID_FMT"\n", + ADRENO_CHIPID_ARGS(config.chip_id)); return -ENXIO; } - DBG("Found GPU: %u.%u.%u.%u", config.rev.core, config.rev.major, - config.rev.minor, config.rev.patchid); + config.info = info; - priv->is_a2xx = config.rev.core == 2; + DBG("Found GPU: %"ADRENO_CHIPID_FMT, ADRENO_CHIPID_ARGS(config.chip_id)); + + priv->is_a2xx = info->family < ADRENO_3XX; + priv->has_cached_coherent = + !!(info->quirks & ADRENO_QUIRK_HAS_CACHED_COHERENT); gpu = info->init(drm); if (IS_ERR(gpu)) { @@ -588,10 +696,6 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) if (ret) return ret; - if (config.rev.core >= 6) - if (!adreno_has_gmu_wrapper(to_adreno_gpu(gpu))) - priv->has_cached_coherent = true; - return 0; } diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 5c5901d65950..8090dde03280 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -320,16 +320,13 @@ int adreno_get_param(struct msm_gpu *gpu, struct msm_file_private *ctx, *value = adreno_gpu->info->revn; return 0; case MSM_PARAM_GMEM_SIZE: - *value = adreno_gpu->gmem; + *value = adreno_gpu->info->gmem; return 0; case MSM_PARAM_GMEM_BASE: *value = !adreno_is_a650_family(adreno_gpu) ? 0x100000 : 0; return 0; case MSM_PARAM_CHIP_ID: - *value = (uint64_t)adreno_gpu->rev.patchid | - ((uint64_t)adreno_gpu->rev.minor << 8) | - ((uint64_t)adreno_gpu->rev.major << 16) | - ((uint64_t)adreno_gpu->rev.core << 24); + *value = adreno_gpu->chip_id; if (!adreno_gpu->info->revn) *value |= ((uint64_t) adreno_gpu->speedbin) << 32; return 0; @@ -400,17 +397,9 @@ int adreno_set_param(struct msm_gpu *gpu, struct msm_file_private *ctx, case MSM_PARAM_CMDLINE: { char *str, **paramp; - str = kmalloc(len + 1, GFP_KERNEL); - if (!str) - return -ENOMEM; - - if (copy_from_user(str, u64_to_user_ptr(value), len)) { - kfree(str); - return -EFAULT; - } - - /* Ensure string is null terminated: */ - str[len] = '\0'; + str = memdup_user_nul(u64_to_user_ptr(value), len); + if (IS_ERR(str)) + return PTR_ERR(str); mutex_lock(&gpu->lock); @@ -847,10 +836,9 @@ void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, if (IS_ERR_OR_NULL(state)) return; - drm_printf(p, "revision: %d (%d.%d.%d.%d)\n", - adreno_gpu->info->revn, adreno_gpu->rev.core, - adreno_gpu->rev.major, adreno_gpu->rev.minor, - adreno_gpu->rev.patchid); + drm_printf(p, "revision: %u (%"ADRENO_CHIPID_FMT")\n", + adreno_gpu->info->revn, + ADRENO_CHIPID_ARGS(adreno_gpu->chip_id)); /* * If this is state collected due to iova fault, so fault related info * @@ -921,10 +909,9 @@ void adreno_dump_info(struct msm_gpu *gpu) struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); int i; - printk("revision: %d (%d.%d.%d.%d)\n", - adreno_gpu->info->revn, adreno_gpu->rev.core, - adreno_gpu->rev.major, adreno_gpu->rev.minor, - adreno_gpu->rev.patchid); + printk("revision: %u (%"ADRENO_CHIPID_FMT")\n", + adreno_gpu->info->revn, + ADRENO_CHIPID_ARGS(adreno_gpu->chip_id)); for (i = 0; i < gpu->nr_rings; i++) { struct msm_ringbuffer *ring = gpu->rb[i]; @@ -1041,14 +1028,16 @@ int adreno_gpu_ocmem_init(struct device *dev, struct adreno_gpu *adreno_gpu, return PTR_ERR(ocmem); } - ocmem_hdl = ocmem_allocate(ocmem, OCMEM_GRAPHICS, adreno_gpu->gmem); + ocmem_hdl = ocmem_allocate(ocmem, OCMEM_GRAPHICS, adreno_gpu->info->gmem); if (IS_ERR(ocmem_hdl)) return PTR_ERR(ocmem_hdl); adreno_ocmem->ocmem = ocmem; adreno_ocmem->base = ocmem_hdl->addr; adreno_ocmem->hdl = ocmem_hdl; - adreno_gpu->gmem = ocmem_hdl->len; + + if (WARN_ON(ocmem_hdl->len != adreno_gpu->info->gmem)) + return -ENOMEM; return 0; } @@ -1073,13 +1062,19 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, struct adreno_platform_config *config = dev->platform_data; struct msm_gpu_config adreno_gpu_config = { 0 }; struct msm_gpu *gpu = &adreno_gpu->base; - struct adreno_rev *rev = &config->rev; const char *gpu_name; u32 speedbin; int ret; + adreno_gpu->funcs = funcs; + adreno_gpu->info = config->info; + adreno_gpu->chip_id = config->chip_id; + + gpu->allow_relocs = config->info->family < ADRENO_6XX_GEN1; + /* Only handle the core clock when GMU is not in use (or is absent). */ - if (adreno_has_gmu_wrapper(adreno_gpu) || config->rev.core < 6) { + if (adreno_has_gmu_wrapper(adreno_gpu) || + adreno_gpu->info->family < ADRENO_6XX_GEN1) { /* * This can only be done before devm_pm_opp_of_add_table(), or * dev_pm_opp_set_config() will WARN_ON() @@ -1095,24 +1090,14 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, devm_pm_opp_set_clkname(dev, "core"); } - adreno_gpu->funcs = funcs; - adreno_gpu->info = adreno_info(config->rev); - adreno_gpu->gmem = adreno_gpu->info->gmem; - adreno_gpu->revn = adreno_gpu->info->revn; - adreno_gpu->rev = *rev; - if (adreno_read_speedbin(dev, &speedbin) || !speedbin) speedbin = 0xffff; adreno_gpu->speedbin = (uint16_t) (0xffff & speedbin); - gpu_name = adreno_gpu->info->name; - if (!gpu_name) { - gpu_name = devm_kasprintf(dev, GFP_KERNEL, "%d.%d.%d.%d", - rev->core, rev->major, rev->minor, - rev->patchid); - if (!gpu_name) - return -ENOMEM; - } + gpu_name = devm_kasprintf(dev, GFP_KERNEL, "%"ADRENO_CHIPID_FMT, + ADRENO_CHIPID_ARGS(config->chip_id)); + if (!gpu_name) + return -ENOMEM; adreno_gpu_config.ioname = "kgsl_3d0_reg_memory"; diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index d8c9e8cc3753..49f38edf9854 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -29,21 +29,40 @@ enum { ADRENO_FW_MAX, }; +/** + * @enum adreno_family: identify generation and possibly sub-generation + * + * In some cases there are distinct sub-generations within a major revision + * so it helps to be able to group the GPU devices by generation and if + * necessary sub-generation. + */ +enum adreno_family { + ADRENO_2XX_GEN1, /* a20x */ + ADRENO_2XX_GEN2, /* a22x */ + ADRENO_3XX, + ADRENO_4XX, + ADRENO_5XX, + ADRENO_6XX_GEN1, /* a630 family */ + ADRENO_6XX_GEN2, /* a640 family */ + ADRENO_6XX_GEN3, /* a650 family */ + ADRENO_6XX_GEN4, /* a660 family */ +}; + #define ADRENO_QUIRK_TWO_PASS_USE_WFI BIT(0) #define ADRENO_QUIRK_FAULT_DETECT_MASK BIT(1) #define ADRENO_QUIRK_LMLOADKILL_DISABLE BIT(2) +#define ADRENO_QUIRK_HAS_HW_APRIV BIT(3) +#define ADRENO_QUIRK_HAS_CACHED_COHERENT BIT(4) -struct adreno_rev { - uint8_t core; - uint8_t major; - uint8_t minor; - uint8_t patchid; -}; - -#define ANY_ID 0xff - -#define ADRENO_REV(core, major, minor, patchid) \ - ((struct adreno_rev){ core, major, minor, patchid }) +/* Helper for formating the chip_id in the way that userspace tools like + * crashdec expect. + */ +#define ADRENO_CHIPID_FMT "u.%u.%u.%u" +#define ADRENO_CHIPID_ARGS(_c) \ + (((_c) >> 24) & 0xff), \ + (((_c) >> 16) & 0xff), \ + (((_c) >> 8) & 0xff), \ + ((_c) & 0xff) struct adreno_gpu_funcs { struct msm_gpu_funcs base; @@ -58,10 +77,21 @@ struct adreno_reglist { extern const struct adreno_reglist a612_hwcg[], a615_hwcg[], a630_hwcg[], a640_hwcg[], a650_hwcg[]; extern const struct adreno_reglist a660_hwcg[], a690_hwcg[]; +struct adreno_speedbin { + uint16_t fuse; + uint16_t speedbin; +}; + struct adreno_info { - struct adreno_rev rev; + const char *machine; + /** + * @chipids: Table of matching chip-ids + * + * Terminated with 0 sentinal + */ + uint32_t *chip_ids; + enum adreno_family family; uint32_t revn; - const char *name; const char *fw[ADRENO_FW_MAX]; uint32_t gmem; u64 quirks; @@ -70,16 +100,39 @@ struct adreno_info { u32 inactive_period; const struct adreno_reglist *hwcg; u64 address_space_size; + /** + * @speedbins: Optional table of fuse to speedbin mappings + * + * Consists of pairs of fuse, index mappings, terminated with + * {SHRT_MAX, 0} sentinal. + */ + struct adreno_speedbin *speedbins; }; -const struct adreno_info *adreno_info(struct adreno_rev rev); +#define ADRENO_CHIP_IDS(tbl...) (uint32_t[]) { tbl, 0 } + +/* + * Helper to build a speedbin table, ie. the table: + * fuse | speedbin + * -----+--------- + * 0 | 0 + * 169 | 1 + * 174 | 2 + * + * would be declared as: + * + * .speedbins = ADRENO_SPEEDBINS( + * { 0, 0 }, + * { 169, 1 }, + * { 174, 2 }, + * ), + */ +#define ADRENO_SPEEDBINS(tbl...) (struct adreno_speedbin[]) { tbl {SHRT_MAX, 0} } struct adreno_gpu { struct msm_gpu base; - struct adreno_rev rev; const struct adreno_info *info; - uint32_t gmem; /* actual gmem size */ - uint32_t revn; /* numeric revision name */ + uint32_t chip_id; uint16_t speedbin; const struct adreno_gpu_funcs *funcs; @@ -128,7 +181,8 @@ struct adreno_ocmem { /* platform config data (ie. from DT, or pdata) */ struct adreno_platform_config { - struct adreno_rev rev; + uint32_t chip_id; + const struct adreno_info *info; }; #define ADRENO_IDLE_TIMEOUT msecs_to_jiffies(1000) @@ -145,13 +199,21 @@ struct adreno_platform_config { __ret; \ }) -bool adreno_cmp_rev(struct adreno_rev rev1, struct adreno_rev rev2); +static inline uint8_t adreno_patchid(const struct adreno_gpu *gpu) +{ + /* It is probably ok to assume legacy "adreno_rev" format + * for all a6xx devices, but probably best to limit this + * to older things. + */ + WARN_ON_ONCE(gpu->info->family >= ADRENO_6XX_GEN1); + return gpu->chip_id & 0xff; +} static inline bool adreno_is_revn(const struct adreno_gpu *gpu, uint32_t revn) { - WARN_ON_ONCE(!gpu->revn); - - return gpu->revn == revn; + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->revn == revn; } static inline bool adreno_has_gmu_wrapper(const struct adreno_gpu *gpu) @@ -161,16 +223,16 @@ static inline bool adreno_has_gmu_wrapper(const struct adreno_gpu *gpu) static inline bool adreno_is_a2xx(const struct adreno_gpu *gpu) { - WARN_ON_ONCE(!gpu->revn); - - return (gpu->revn < 300); + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family <= ADRENO_2XX_GEN2; } static inline bool adreno_is_a20x(const struct adreno_gpu *gpu) { - WARN_ON_ONCE(!gpu->revn); - - return (gpu->revn < 210); + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_2XX_GEN1; } static inline bool adreno_is_a225(const struct adreno_gpu *gpu) @@ -201,7 +263,7 @@ static inline bool adreno_is_a330(const struct adreno_gpu *gpu) static inline bool adreno_is_a330v2(const struct adreno_gpu *gpu) { - return adreno_is_a330(gpu) && (gpu->rev.patchid > 0); + return adreno_is_a330(gpu) && (adreno_patchid(gpu) > 0); } static inline int adreno_is_a405(const struct adreno_gpu *gpu) @@ -291,8 +353,7 @@ static inline int adreno_is_a650(const struct adreno_gpu *gpu) static inline int adreno_is_7c3(const struct adreno_gpu *gpu) { - /* The order of args is important here to handle ANY_ID correctly */ - return adreno_cmp_rev(ADRENO_REV(6, 3, 5, ANY_ID), gpu->rev); + return gpu->info->chip_ids[0] == 0x06030500; } static inline int adreno_is_a660(const struct adreno_gpu *gpu) @@ -307,34 +368,37 @@ static inline int adreno_is_a680(const struct adreno_gpu *gpu) static inline int adreno_is_a690(const struct adreno_gpu *gpu) { - return adreno_is_revn(gpu, 690); -}; + return gpu->info->chip_ids[0] == 0x06090000; +} -/* check for a615, a616, a618, a619 or any derivatives */ -static inline int adreno_is_a615_family(const struct adreno_gpu *gpu) +/* check for a615, a616, a618, a619 or any a630 derivatives */ +static inline int adreno_is_a630_family(const struct adreno_gpu *gpu) { - return adreno_is_revn(gpu, 615) || - adreno_is_revn(gpu, 616) || - adreno_is_revn(gpu, 618) || - adreno_is_revn(gpu, 619); + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN1; } static inline int adreno_is_a660_family(const struct adreno_gpu *gpu) { - return adreno_is_a660(gpu) || adreno_is_a690(gpu) || adreno_is_7c3(gpu); + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN4; } /* check for a650, a660, or any derivatives */ static inline int adreno_is_a650_family(const struct adreno_gpu *gpu) { - return adreno_is_revn(gpu, 650) || - adreno_is_revn(gpu, 620) || - adreno_is_a660_family(gpu); + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family >= ADRENO_6XX_GEN3; } static inline int adreno_is_a640_family(const struct adreno_gpu *gpu) { - return adreno_is_a640(gpu) || adreno_is_a680(gpu); + if (WARN_ON_ONCE(!gpu->info)) + return false; + return gpu->info->family == ADRENO_6XX_GEN2; } u64 adreno_private_address_space_size(struct msm_gpu *gpu); diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h index 7d0d0e74c3b0..43c47a19cd94 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_3_0_msm8998.h @@ -21,136 +21,257 @@ static const struct dpu_caps msm8998_dpu_caps = { .max_vdeci_exp = MAX_VERT_DECIMATION, }; -static const struct dpu_ubwc_cfg msm8998_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_10, - .highest_bank_bit = 0x2, -}; - -static const struct dpu_mdp_cfg msm8998_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg msm8998_mdp = { + .name = "top_0", .base = 0x0, .len = 0x458, .features = BIT(DPU_MDP_VSYNC_SEL), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 12 }, - .clk_ctrls[DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, - .clk_ctrls[DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 16 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 12 }, + [DPU_CLK_CTRL_CURSOR0] = { .reg_off = 0x3a8, .bit_off = 16 }, + [DPU_CLK_CTRL_CURSOR1] = { .reg_off = 0x3b0, .bit_off = 16 }, }, }; static const struct dpu_ctl_cfg msm8998_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x94, - .features = BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0x94, - .features = 0, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0x94, - .features = BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x1600, .len = 0x94, - .features = 0, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x1800, .len = 0x94, - .features = 0, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x94, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x94, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x94, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x94, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, }; static const struct dpu_sspp_cfg msm8998_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1ac, VIG_MSM8998_MASK, - msm8998_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x1ac, VIG_MSM8998_MASK, - msm8998_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x1ac, VIG_MSM8998_MASK, - msm8998_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x1ac, VIG_MSM8998_MASK, - msm8998_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1ac, DMA_MSM8998_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1ac, DMA_MSM8998_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1ac, DMA_CURSOR_MSM8998_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x1ac, DMA_CURSOR_MSM8998_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &msm8998_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &msm8998_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &msm8998_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1ac, + .features = VIG_MSM8998_MASK, + .sblk = &msm8998_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1ac, + .features = DMA_MSM8998_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1ac, + .features = DMA_CURSOR_MSM8998_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1ac, + .features = DMA_CURSOR_MSM8998_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg msm8998_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_2, LM_5, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_NONE, 0, 0), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_NONE, 0, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_MSM8998_MASK, - &msm8998_lm_sblk, PINGPONG_3, LM_2, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .pingpong = PINGPONG_NONE, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .pingpong = PINGPONG_NONE, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_MSM8998_MASK, + .sblk = &msm8998_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, }; static const struct dpu_pingpong_cfg msm8998_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SDM845_TE2_MASK, + .sblk = &sdm845_pp_sblk_te, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SDM845_TE2_MASK, + .sblk = &sdm845_pp_sblk_te, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .features = PINGPONG_SDM845_MASK, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .features = PINGPONG_SDM845_MASK, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), + }, }; static const struct dpu_dsc_cfg msm8998_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, 0), - DSC_BLK("dsc_1", DSC_1, 0x80400, 0), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, }; static const struct dpu_dspp_cfg msm8998_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &msm8998_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &msm8998_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &msm8998_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &msm8998_dspp_sblk, + }, }; static const struct dpu_intf_cfg msm8998_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 21, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 21, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 21, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_HDMI, 0, 21, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_HDMI, + .prog_fetch_lines_worst_case = 21, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg msm8998_perf_data = { @@ -189,11 +310,15 @@ static const struct dpu_perf_cfg msm8998_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version msm8998_mdss_ver = { + .core_major_ver = 3, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_msm8998_cfg = { + .mdss_ver = &msm8998_mdss_ver, .caps = &msm8998_dpu_caps, - .ubwc = &msm8998_ubwc_cfg, - .mdp_count = ARRAY_SIZE(msm8998_mdp), - .mdp = msm8998_mdp, + .mdp = &msm8998_mdp, .ctl_count = ARRAY_SIZE(msm8998_ctl), .ctl = msm8998_ctl, .sspp_count = ARRAY_SIZE(msm8998_sspp), @@ -211,14 +336,6 @@ const struct dpu_mdss_cfg dpu_msm8998_cfg = { .vbif_count = ARRAY_SIZE(msm8998_vbif), .vbif = msm8998_vbif, .perf = &msm8998_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF2_INTR) | \ - BIT(MDP_INTF3_INTR) | \ - BIT(MDP_INTF4_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h index b6098141bb9b..88a5177dfdb7 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_4_0_sdm845.h @@ -21,140 +21,274 @@ static const struct dpu_caps sdm845_dpu_caps = { .max_vdeci_exp = MAX_VERT_DECIMATION, }; -static const struct dpu_ubwc_cfg sdm845_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_20, - .highest_bank_bit = 0x2, -}; - -static const struct dpu_mdp_cfg sdm845_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sdm845_mdp = { + .name = "top_0", .base = 0x0, .len = 0x45c, .features = BIT(DPU_MDP_AUDIO_SELECT) | BIT(DPU_MDP_VSYNC_SEL), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, }, }; static const struct dpu_ctl_cfg sdm845_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0xe4, - .features = BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0xe4, - .features = BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0xe4, - .features = 0, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x1600, .len = 0xe4, - .features = 0, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x1800, .len = 0xe4, - .features = 0, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0xe4, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0xe4, + .features = BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0xe4, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0xe4, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0xe4, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), }, }; static const struct dpu_sspp_cfg sdm845_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1c8, VIG_SDM845_MASK_SDMA, - sdm845_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x1c8, VIG_SDM845_MASK_SDMA, - sdm845_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x1c8, VIG_SDM845_MASK_SDMA, - sdm845_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x1c8, VIG_SDM845_MASK_SDMA, - sdm845_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1c8, DMA_SDM845_MASK_SDMA, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1c8, DMA_SDM845_MASK_SDMA, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1c8, DMA_CURSOR_SDM845_MASK_SDMA, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x1c8, DMA_CURSOR_SDM845_MASK_SDMA, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &sdm845_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &sdm845_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &sdm845_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1c8, + .features = VIG_SDM845_MASK_SDMA, + .sblk = &sdm845_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1c8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1c8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1c8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1c8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sdm845_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_5, DSPP_2), - LM_BLK("lm_3", LM_3, 0x0, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_NONE, 0, DSPP_3), - LM_BLK("lm_4", LM_4, 0x0, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_NONE, 0, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x0, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_NONE, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x0, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_NONE, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, }; static const struct dpu_dspp_cfg sdm845_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sdm845_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12)), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SDM845_TE2_MASK, 0, sdm845_pp_sblk_te, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14)), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SDM845_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15)), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SDM845_TE2_MASK, + .sblk = &sdm845_pp_sblk_te, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SDM845_TE2_MASK, + .sblk = &sdm845_pp_sblk_te, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .features = PINGPONG_SDM845_MASK, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 14), + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .features = PINGPONG_SDM845_MASK, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 15), + }, }; static const struct dpu_dsc_cfg sdm845_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, 0), - DSC_BLK("dsc_1", DSC_1, 0x80400, 0), - DSC_BLK("dsc_2", DSC_2, 0x80800, 0), - DSC_BLK("dsc_3", DSC_3, 0x80c00, 0), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + }, }; static const struct dpu_intf_cfg sdm845_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK("intf_1", INTF_1, 0x6a800, 0x280, INTF_DSI, 0, 24, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27)), - INTF_BLK("intf_2", INTF_2, 0x6b000, 0x280, INTF_DSI, 1, 24, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29)), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SDM845_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x280, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sdm845_perf_data = { @@ -193,11 +327,15 @@ static const struct dpu_perf_cfg sdm845_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sdm845_mdss_ver = { + .core_major_ver = 4, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_sdm845_cfg = { + .mdss_ver = &sdm845_mdss_ver, .caps = &sdm845_dpu_caps, - .ubwc = &sdm845_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sdm845_mdp), - .mdp = sdm845_mdp, + .mdp = &sdm845_mdp, .ctl_count = ARRAY_SIZE(sdm845_ctl), .ctl = sdm845_ctl, .sspp_count = ARRAY_SIZE(sdm845_sspp), @@ -215,15 +353,6 @@ const struct dpu_mdss_cfg dpu_sdm845_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sdm845_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF2_INTR) | \ - BIT(MDP_INTF3_INTR) | \ - BIT(MDP_AD4_0_INTR) | \ - BIT(MDP_AD4_1_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h index b5f751354267..99acaf917e43 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_0_sm8150.h @@ -21,161 +21,324 @@ static const struct dpu_caps sm8150_dpu_caps = { .max_vdeci_exp = MAX_VERT_DECIMATION, }; -static const struct dpu_ubwc_cfg sm8150_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_30, - .highest_bank_bit = 0x2, -}; - -static const struct dpu_mdp_cfg sm8150_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm8150_mdp = { + .name = "top_0", .base = 0x0, .len = 0x45c, .features = BIT(DPU_MDP_AUDIO_SELECT), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, }, }; /* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sm8150_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sm8150_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f0, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f0, DMA_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f0, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x1f0, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sm8150_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sm8150_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm8150_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), - PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), - -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x72000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x72800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sm8150_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x83000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x83100), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x83200), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x83200, .len = 0x8, + }, }; static const struct dpu_dsc_cfg sm8150_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_1", DSC_1, 0x80400, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_2", DSC_2, 0x80800, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_3", DSC_3, 0x80c00, BIT(DPU_DSC_OUTPUT_CTRL)), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, }; static const struct dpu_intf_cfg sm8150_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2bc, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2bc, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sm8150_perf_data = { @@ -207,11 +370,15 @@ static const struct dpu_perf_cfg sm8150_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm8150_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_sm8150_cfg = { + .mdss_ver = &sm8150_mdss_ver, .caps = &sm8150_dpu_caps, - .ubwc = &sm8150_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm8150_mdp), - .mdp = sm8150_mdp, + .mdp = &sm8150_mdp, .ctl_count = ARRAY_SIZE(sm8150_ctl), .ctl = sm8150_ctl, .sspp_count = ARRAY_SIZE(sm8150_sspp), @@ -231,17 +398,6 @@ const struct dpu_mdss_cfg dpu_sm8150_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sm8150_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR) | \ - BIT(MDP_INTF2_INTR) | \ - BIT(MDP_INTF2_TEAR_INTR) | \ - BIT(MDP_INTF3_INTR) | \ - BIT(MDP_AD4_0_INTR) | \ - BIT(MDP_AD4_1_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h index 8ed2b263c5ea..f3de21025ca7 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_1_sc8180x.h @@ -21,169 +21,353 @@ static const struct dpu_caps sc8180x_dpu_caps = { .max_vdeci_exp = MAX_VERT_DECIMATION, }; -static const struct dpu_ubwc_cfg sc8180x_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_30, - .highest_bank_bit = 0x3, -}; - -static const struct dpu_mdp_cfg sc8180x_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sc8180x_mdp = { + .name = "top_0", .base = 0x0, .len = 0x45c, .features = BIT(DPU_MDP_AUDIO_SELECT), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, }, }; static const struct dpu_ctl_cfg sc8180x_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sc8180x_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x1f0, VIG_SDM845_MASK, - sdm845_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f0, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f0, DMA_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f0, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x1f0, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f0, + .features = VIG_SDM845_MASK, + .sblk = &sdm845_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f0, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sc8180x_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sc8180x_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sc8180x_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), - PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), - -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x72000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x72800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sc8180x_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x83000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x83100), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x83200), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x83200, .len = 0x8, + }, }; static const struct dpu_dsc_cfg sc8180x_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_1", DSC_1, 0x80400, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_2", DSC_2, 0x80800, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_3", DSC_3, 0x80c00, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_4", DSC_4, 0x81000, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_5", DSC_5, 0x81400, BIT(DPU_DSC_OUTPUT_CTRL)), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_4", .id = DSC_4, + .base = 0x81000, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_5", .id = DSC_5, + .base = 0x81400, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, }; static const struct dpu_intf_cfg sc8180x_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2bc, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2bc, INTF_DSI, 1, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2bc, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2bc, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, /* INTF_3 is for MST, wired to INTF_DP 0 and 1, use dummy index until this is supported */ - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 999, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), - INTF_BLK("intf_4", INTF_4, 0x6c000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21)), - INTF_BLK("intf_5", INTF_5, 0x6c800, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)), + { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = 999, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x6c000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x6c800, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sc8180x_perf_data = { @@ -213,11 +397,15 @@ static const struct dpu_perf_cfg sc8180x_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sc8180x_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 1, +}; + const struct dpu_mdss_cfg dpu_sc8180x_cfg = { + .mdss_ver = &sc8180x_mdss_ver, .caps = &sc8180x_dpu_caps, - .ubwc = &sc8180x_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sc8180x_mdp), - .mdp = sc8180x_mdp, + .mdp = &sc8180x_mdp, .ctl_count = ARRAY_SIZE(sc8180x_ctl), .ctl = sc8180x_ctl, .sspp_count = ARRAY_SIZE(sc8180x_sspp), @@ -237,19 +425,6 @@ const struct dpu_mdss_cfg dpu_sc8180x_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sc8180x_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR) | \ - BIT(MDP_INTF2_INTR) | \ - BIT(MDP_INTF2_TEAR_INTR) | \ - BIT(MDP_INTF3_INTR) | \ - BIT(MDP_INTF4_INTR) | \ - BIT(MDP_INTF5_INTR) | \ - BIT(MDP_AD4_0_INTR) | \ - BIT(MDP_AD4_1_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h new file mode 100644 index 000000000000..2491eed10039 --- /dev/null +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_5_4_sm6125.h @@ -0,0 +1,223 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2023 Marijn Suijten <marijn.suijten@somainline.org>. All rights reserved. + * Copyright (c) 2022. Qualcomm Innovation Center, Inc. All rights reserved. + * Copyright (c) 2015-2018, 2020 The Linux Foundation. All rights reserved. + */ + +#ifndef _DPU_5_4_SM6125_H +#define _DPU_5_4_SM6125_H + +static const struct dpu_caps sm6125_dpu_caps = { + .max_mixer_width = DEFAULT_DPU_OUTPUT_LINE_WIDTH, + .max_mixer_blendstages = 0x6, + .has_dim_layer = true, + .has_idle_pc = true, + .max_linewidth = 2160, + .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, + .max_hdeci_exp = MAX_HORZ_DECIMATION, + .max_vdeci_exp = MAX_VERT_DECIMATION, +}; + +static const struct dpu_mdp_cfg sm6125_mdp = { + .name = "top_0", + .base = 0x0, .len = 0x45c, + .features = 0, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + }, +}; + +static const struct dpu_ctl_cfg sm6125_ctl[] = { + { + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + }, +}; + +static const struct dpu_sspp_cfg sm6125_sspp[] = { + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f0, + .features = VIG_SM6125_MASK, + .sblk = &sm6125_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f0, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f0, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, +}; + +static const struct dpu_lm_cfg sm6125_lm[] = { + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_QCM2290_MASK, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + .lm_pair = LM_1, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_QCM2290_MASK, + .sblk = &sdm845_lm_sblk, + .pingpong = PINGPONG_1, + .dspp = 0, + .lm_pair = LM_0, + }, +}; + +static const struct dpu_dspp_cfg sm6125_dspp[] = { + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, +}; + +static const struct dpu_pingpong_cfg sm6125_pp[] = { + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .merge_3d = 0, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .merge_3d = 0, + .sblk = &sdm845_pp_sblk, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, +}; + +static const struct dpu_intf_cfg sm6125_intf[] = { + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = 0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, +}; + +static const struct dpu_perf_cfg sm6125_perf_data = { + .max_bw_low = 4100000, + .max_bw_high = 4100000, + .min_core_ib = 2400000, + .min_llcc_ib = 0, /* No LLCC on this SoC */ + .min_dram_ib = 800000, + .min_prefill_lines = 24, + .danger_lut_tbl = {0xf, 0xffff, 0x0}, + .safe_lut_tbl = {0xfff8, 0xf000, 0xffff}, + .qos_lut_tbl = { + {.nentry = ARRAY_SIZE(sm8150_qos_linear), + .entries = sm8150_qos_linear + }, + {.nentry = ARRAY_SIZE(sc7180_qos_macrotile), + .entries = sc7180_qos_macrotile + }, + {.nentry = ARRAY_SIZE(sc7180_qos_nrt), + .entries = sc7180_qos_nrt + }, + /* TODO: macrotile-qseed is different from macrotile */ + }, + .cdp_cfg = { + {.rd_enable = 1, .wr_enable = 1}, + {.rd_enable = 1, .wr_enable = 0} + }, + .clk_inefficiency_factor = 105, + .bw_inefficiency_factor = 120, +}; + +static const struct dpu_mdss_version sm6125_mdss_ver = { + .core_major_ver = 5, + .core_minor_ver = 4, +}; + +const struct dpu_mdss_cfg dpu_sm6125_cfg = { + .mdss_ver = &sm6125_mdss_ver, + .caps = &sm6125_dpu_caps, + .mdp = &sm6125_mdp, + .ctl_count = ARRAY_SIZE(sm6125_ctl), + .ctl = sm6125_ctl, + .sspp_count = ARRAY_SIZE(sm6125_sspp), + .sspp = sm6125_sspp, + .mixer_count = ARRAY_SIZE(sm6125_lm), + .mixer = sm6125_lm, + .dspp_count = ARRAY_SIZE(sm6125_dspp), + .dspp = sm6125_dspp, + .pingpong_count = ARRAY_SIZE(sm6125_pp), + .pingpong = sm6125_pp, + .intf_count = ARRAY_SIZE(sm6125_intf), + .intf = sm6125_intf, + .vbif_count = ARRAY_SIZE(sdm845_vbif), + .vbif = sdm845_vbif, + .perf = &sm6125_perf_data, +}; + +#endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h index daebd2170041..5f9b437b82a6 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_0_sm8250.h @@ -19,169 +19,340 @@ static const struct dpu_caps sm8250_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm8250_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_40, - .highest_bank_bit = 0x3, /* TODO: 2 for LP_DDR4 */ - .ubwc_swizzle = 0x6, -}; - -static const struct dpu_mdp_cfg sm8250_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm8250_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, - .clk_ctrls[DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, }, }; /* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sm8250_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x1600, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x1800, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a00, .len = 0x1e0, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG) | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x1800, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a00, .len = 0x1e0, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sm8250_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK_SDMA, - sm8250_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x1f8, VIG_SC7180_MASK_SDMA, - sm8250_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x1f8, VIG_SC7180_MASK_SDMA, - sm8250_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x1f8, VIG_SC7180_MASK_SDMA, - sm8250_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK_SDMA, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_SDM845_MASK_SDMA, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK_SDMA, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x1f8, DMA_CURSOR_SDM845_MASK_SDMA, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7180_MASK_SDMA, + .sblk = &sm8250_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f8, + .features = VIG_SC7180_MASK_SDMA, + .sblk = &sm8250_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f8, + .features = VIG_SC7180_MASK_SDMA, + .sblk = &sm8250_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f8, + .features = VIG_SC7180_MASK_SDMA, + .sblk = &sm8250_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sm8250_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sm8250_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm8250_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, MERGE_3D_0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK("pingpong_2", PINGPONG_2, 0x71000, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK("pingpong_3", PINGPONG_3, 0x71800, PINGPONG_SM8150_MASK, MERGE_3D_1, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), - PP_BLK("pingpong_4", PINGPONG_4, 0x72000, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), - -1), - PP_BLK("pingpong_5", PINGPONG_5, 0x72800, PINGPONG_SM8150_MASK, MERGE_3D_2, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x71000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x71800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x72000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x72800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sm8250_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x83000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x83100), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x83200), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x83000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x83100, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x83200, .len = 0x8, + }, }; static const struct dpu_dsc_cfg sm8250_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_1", DSC_1, 0x80400, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_2", DSC_2, 0x80800, BIT(DPU_DSC_OUTPUT_CTRL)), - DSC_BLK("dsc_3", DSC_3, 0x80c00, BIT(DPU_DSC_OUTPUT_CTRL)), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_1", .id = DSC_1, + .base = 0x80400, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_2", .id = DSC_2, + .base = 0x80800, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, { + .name = "dsc_3", .id = DSC_3, + .base = 0x80c00, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, }; static const struct dpu_intf_cfg sm8250_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x6b000, 0x2c0, INTF_DSI, 1, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2)), - INTF_BLK("intf_3", INTF_3, 0x6b800, 0x280, INTF_DP, 1, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x6b000, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x6b800, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_wb_cfg sm8250_wb[] = { - WB_BLK("wb_2", WB_2, 0x65000, WB_SM8250_MASK, DPU_CLK_CTRL_WB2, 6, - VBIF_RT, MDP_SSPP_TOP0_INTR, 4096, 4), + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SM8250_MASK, + .format_list = wb2_formats, + .num_formats = ARRAY_SIZE(wb2_formats), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .vbif_idx = VBIF_RT, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, }; static const struct dpu_perf_cfg sm8250_perf_data = { @@ -213,11 +384,15 @@ static const struct dpu_perf_cfg sm8250_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm8250_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_sm8250_cfg = { + .mdss_ver = &sm8250_mdss_ver, .caps = &sm8250_dpu_caps, - .ubwc = &sm8250_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm8250_mdp), - .mdp = sm8250_mdp, + .mdp = &sm8250_mdp, .ctl_count = ARRAY_SIZE(sm8250_ctl), .ctl = sm8250_ctl, .sspp_count = ARRAY_SIZE(sm8250_sspp), @@ -239,16 +414,6 @@ const struct dpu_mdss_cfg dpu_sm8250_cfg = { .wb_count = ARRAY_SIZE(sm8250_wb), .wb = sm8250_wb, .perf = &sm8250_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR) | \ - BIT(MDP_INTF2_INTR) | \ - BIT(MDP_INTF2_TEAR_INTR) | \ - BIT(MDP_INTF3_INTR) | \ - BIT(MDP_INTF4_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h index 67566b07195a..d030c08636b4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_2_sc7180.h @@ -17,90 +17,158 @@ static const struct dpu_caps sc7180_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sc7180_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_20, - .highest_bank_bit = 0x3, -}; - -static const struct dpu_mdp_cfg sc7180_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sc7180_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, }, }; static const struct dpu_ctl_cfg sc7180_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), }, }; static const struct dpu_sspp_cfg sc7180_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, - sc7180_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sc7180_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, }; static const struct dpu_lm_cfg sc7180_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_1, LM_0, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + }, }; static const struct dpu_dspp_cfg sc7180_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sc7180_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, }; static const struct dpu_intf_cfg sc7180_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, }; static const struct dpu_wb_cfg sc7180_wb[] = { - WB_BLK("wb_2", WB_2, 0x65000, WB_SM8250_MASK, DPU_CLK_CTRL_WB2, 6, - VBIF_RT, MDP_SSPP_TOP0_INTR, 4096, 4), + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SM8250_MASK, + .format_list = wb2_formats, + .num_formats = ARRAY_SIZE(wb2_formats), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .vbif_idx = VBIF_RT, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, }; static const struct dpu_perf_cfg sc7180_perf_data = { @@ -131,11 +199,15 @@ static const struct dpu_perf_cfg sc7180_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sc7180_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 2, +}; + const struct dpu_mdss_cfg dpu_sc7180_cfg = { + .mdss_ver = &sc7180_mdss_ver, .caps = &sc7180_dpu_caps, - .ubwc = &sc7180_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sc7180_mdp), - .mdp = sc7180_mdp, + .mdp = &sc7180_mdp, .ctl_count = ARRAY_SIZE(sc7180_ctl), .ctl = sc7180_ctl, .sspp_count = ARRAY_SIZE(sc7180_sspp), @@ -153,12 +225,6 @@ const struct dpu_mdss_cfg dpu_sc7180_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sc7180_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h index 031fc8dae3c6..7e6000167ef0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_3_sm6115.h @@ -17,59 +17,88 @@ static const struct dpu_caps sm6115_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm6115_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_10, - .highest_bank_bit = 0x1, - .ubwc_swizzle = 0x7, -}; - -static const struct dpu_mdp_cfg sm6115_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm6115_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, }, }; static const struct dpu_ctl_cfg sm6115_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, }; static const struct dpu_sspp_cfg sm6115_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, - sm6115_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sm6115_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, }; static const struct dpu_lm_cfg sm6115_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_QCM2290_MASK, - &qcm2290_lm_sblk, PINGPONG_0, 0, DSPP_0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_QCM2290_MASK, + .sblk = &qcm2290_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, }; static const struct dpu_dspp_cfg sm6115_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm6115_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, }; static const struct dpu_intf_cfg sm6115_intf[] = { - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, }; static const struct dpu_perf_cfg sm6115_perf_data = { @@ -101,11 +130,15 @@ static const struct dpu_perf_cfg sm6115_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm6115_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 3, +}; + const struct dpu_mdss_cfg dpu_sm6115_cfg = { + .mdss_ver = &sm6115_mdss_ver, .caps = &sm6115_dpu_caps, - .ubwc = &sm6115_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm6115_mdp), - .mdp = sm6115_mdp, + .mdp = &sm6115_mdp, .ctl_count = ARRAY_SIZE(sm6115_ctl), .ctl = sm6115_ctl, .sspp_count = ARRAY_SIZE(sm6115_sspp), @@ -121,11 +154,6 @@ const struct dpu_mdss_cfg dpu_sm6115_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sm6115_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h index 06eba23b0236..cf5db6f296bd 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_4_sm6350.h @@ -19,96 +19,157 @@ static const struct dpu_caps sm6350_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm6350_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_20, - .ubwc_swizzle = 6, - .highest_bank_bit = 1, -}; - -static const struct dpu_mdp_cfg sm6350_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm6350_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, }; static const struct dpu_ctl_cfg sm6350_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x1200, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x1400, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x1600, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x1200, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x1400, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x1600, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, }; static const struct dpu_sspp_cfg sm6350_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, - sc7180_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sc7180_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, }; static const struct dpu_lm_cfg sm6350_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_1, LM_0, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = 0, + }, }; static const struct dpu_dspp_cfg sm6350_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static struct dpu_pingpong_cfg sm6350_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK("pingpong_1", PINGPONG_1, 0x70800, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x70800, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, }; static const struct dpu_dsc_cfg sm6350_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, }; static const struct dpu_intf_cfg sm6350_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x6a000, 0x280, INTF_DP, 0, 35, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 35, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x6a000, .len = 0x280, + .features = INTF_SC7180_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 35, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 35, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, }; static const struct dpu_perf_cfg sm6350_perf_data = { @@ -140,11 +201,15 @@ static const struct dpu_perf_cfg sm6350_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm6350_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 4, +}; + const struct dpu_mdss_cfg dpu_sm6350_cfg = { + .mdss_ver = &sm6350_mdss_ver, .caps = &sm6350_dpu_caps, - .ubwc = &sm6350_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm6350_mdp), - .mdp = sm6350_mdp, + .mdp = &sm6350_mdp, .ctl_count = ARRAY_SIZE(sm6350_ctl), .ctl = sm6350_ctl, .sspp_count = ARRAY_SIZE(sm6350_sspp), @@ -162,12 +227,6 @@ const struct dpu_mdss_cfg dpu_sm6350_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sm6350_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h index f2808098af39..87a03aa16554 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_5_qcm2290.h @@ -16,57 +16,88 @@ static const struct dpu_caps qcm2290_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg qcm2290_ubwc_cfg = { - .highest_bank_bit = 0x2, -}; - -static const struct dpu_mdp_cfg qcm2290_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg qcm2290_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, }, }; static const struct dpu_ctl_cfg qcm2290_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, }; static const struct dpu_sspp_cfg qcm2290_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_QCM2290_MASK, - qcm2290_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, - qcm2290_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_QCM2290_MASK, + .sblk = &qcm2290_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &qcm2290_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, }; static const struct dpu_lm_cfg qcm2290_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_QCM2290_MASK, - &qcm2290_lm_sblk, PINGPONG_0, 0, DSPP_0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_QCM2290_MASK, + .sblk = &qcm2290_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, }; static const struct dpu_dspp_cfg qcm2290_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg qcm2290_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, }; static const struct dpu_intf_cfg qcm2290_intf[] = { - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, }; static const struct dpu_perf_cfg qcm2290_perf_data = { @@ -91,11 +122,15 @@ static const struct dpu_perf_cfg qcm2290_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version qcm2290_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 5, +}; + const struct dpu_mdss_cfg dpu_qcm2290_cfg = { + .mdss_ver = &qcm2290_mdss_ver, .caps = &qcm2290_dpu_caps, - .ubwc = &qcm2290_ubwc_cfg, - .mdp_count = ARRAY_SIZE(qcm2290_mdp), - .mdp = qcm2290_mdp, + .mdp = &qcm2290_mdp, .ctl_count = ARRAY_SIZE(qcm2290_ctl), .ctl = qcm2290_ctl, .sspp_count = ARRAY_SIZE(qcm2290_sspp), @@ -111,11 +146,6 @@ const struct dpu_mdss_cfg dpu_qcm2290_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &qcm2290_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h index 241fa6746674..a327e21c90fb 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_6_9_sm6375.h @@ -18,63 +18,97 @@ static const struct dpu_caps sm6375_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm6375_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_20, - .ubwc_swizzle = 6, - .highest_bank_bit = 1, -}; - -static const struct dpu_mdp_cfg sm6375_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm6375_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, }, }; static const struct dpu_ctl_cfg sm6375_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x1000, .len = 0x1dc, - .features = BIT(DPU_CTL_ACTIVE_CFG), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + .name = "ctl_0", .id = CTL_0, + .base = 0x1000, .len = 0x1dc, + .features = BIT(DPU_CTL_ACTIVE_CFG), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), }, }; static const struct dpu_sspp_cfg sm6375_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, - sm6115_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sm6115_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, }; static const struct dpu_lm_cfg sm6375_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_QCM2290_MASK, - &qcm2290_lm_sblk, PINGPONG_0, 0, DSPP_0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_QCM2290_MASK, + .sblk = &qcm2290_lm_sblk, + .lm_pair = 0, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, }; static const struct dpu_dspp_cfg sm6375_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm6375_pp[] = { - PP_BLK("pingpong_0", PINGPONG_0, 0x70000, PINGPONG_SM8150_MASK, 0, sdm845_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x70000, .len = 0xd4, + .features = PINGPONG_SM8150_MASK, + .sblk = &sdm845_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, }; static const struct dpu_dsc_cfg sm6375_dsc[] = { - DSC_BLK("dsc_0", DSC_0, 0x80000, BIT(DPU_DSC_OUTPUT_CTRL)), + { + .name = "dsc_0", .id = DSC_0, + .base = 0x80000, .len = 0x140, + .features = BIT(DPU_DSC_OUTPUT_CTRL), + }, }; static const struct dpu_intf_cfg sm6375_intf[] = { - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x6a800, 0x2c0, INTF_DSI, 0, 24, INTF_SC7180_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2)), + { + .name = "intf_1", .id = INTF_1, + .base = 0x6a800, .len = 0x2c0, + .features = INTF_SC7180_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, }; static const struct dpu_perf_cfg sm6375_perf_data = { @@ -106,11 +140,15 @@ static const struct dpu_perf_cfg sm6375_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm6375_mdss_ver = { + .core_major_ver = 6, + .core_minor_ver = 9, +}; + const struct dpu_mdss_cfg dpu_sm6375_cfg = { + .mdss_ver = &sm6375_mdss_ver, .caps = &sm6375_dpu_caps, - .ubwc = &sm6375_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm6375_mdp), - .mdp = sm6375_mdp, + .mdp = &sm6375_mdp, .ctl_count = ARRAY_SIZE(sm6375_ctl), .ctl = sm6375_ctl, .sspp_count = ARRAY_SIZE(sm6375_sspp), @@ -128,11 +166,6 @@ const struct dpu_mdss_cfg dpu_sm6375_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sm6375_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF1_INTR) | \ - BIT(MDP_INTF1_TEAR_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h index 8da424eaee6a..f8d16f9bf528 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_0_sm8350.h @@ -19,138 +19,260 @@ static const struct dpu_caps sm8350_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm8350_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_40, - .highest_bank_bit = 0x3, /* TODO: 2 for LP_DDR4 */ -}; - -static const struct dpu_mdp_cfg sm8350_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm8350_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, - .features = 0, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, }; /* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sm8350_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x15000, .len = 0x1e8, - .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x16000, .len = 0x1e8, - .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x17000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x18000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x19000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1e8, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1e8, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sm8350_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7180_MASK, - sm8250_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x1f8, VIG_SC7180_MASK, - sm8250_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x1f8, VIG_SC7180_MASK, - sm8250_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x1f8, VIG_SC7180_MASK, - sm8250_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x1f8, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x1f8, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sm8350_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sm8350_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm8350_pp[] = { - PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), - PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), - -1), - PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sm8350_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x4e000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x4f000), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, }; /* @@ -159,27 +281,71 @@ static const struct dpu_merge_3d_cfg sm8350_merge_3d[] = { * its own different sub block address. */ static const struct dpu_dsc_cfg sm8350_dsc[] = { - DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), - DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), - DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), - DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, }; static const struct dpu_intf_cfg sm8350_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x2c4, INTF_DSI, 1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x2c4, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x2c4, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sm8350_perf_data = { @@ -212,11 +378,15 @@ static const struct dpu_perf_cfg sm8350_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm8350_mdss_ver = { + .core_major_ver = 7, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_sm8350_cfg = { + .mdss_ver = &sm8350_mdss_ver, .caps = &sm8350_dpu_caps, - .ubwc = &sm8350_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm8350_mdp), - .mdp = sm8350_mdp, + .mdp = &sm8350_mdp, .ctl_count = ARRAY_SIZE(sm8350_ctl), .ctl = sm8350_ctl, .sspp_count = ARRAY_SIZE(sm8350_sspp), @@ -236,15 +406,6 @@ const struct dpu_mdss_cfg dpu_sm8350_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sm8350_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF2_7xxx_INTR) | \ - BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF3_7xxx_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h index 900fee410e11..3b5061c4402a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_7_2_sc7280.h @@ -17,112 +17,205 @@ static const struct dpu_caps sc7280_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sc7280_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_30, - .highest_bank_bit = 0x1, - .ubwc_swizzle = 0x6, -}; - -static const struct dpu_mdp_cfg sc7280_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sc7280_mdp = { + .name = "top_0", .base = 0x0, .len = 0x2014, - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_WB2] = { .reg_off = 0x3b8, .bit_off = 24 }, }, }; static const struct dpu_ctl_cfg sc7280_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x15000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x16000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x17000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x18000, .len = 0x1e8, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x1e8, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), }, }; static const struct dpu_sspp_cfg sc7280_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x1f8, VIG_SC7280_MASK_SDMA, - sc7280_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x1f8, DMA_SDM845_MASK_SDMA, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x1f8, DMA_CURSOR_SDM845_MASK_SDMA, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x1f8, DMA_CURSOR_SDM845_MASK_SDMA, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x1f8, + .features = VIG_SC7280_MASK_SDMA, + .sblk = &sc7280_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x1f8, + .features = DMA_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x1f8, + .features = DMA_CURSOR_SDM845_MASK_SDMA, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, }; static const struct dpu_lm_cfg sc7280_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_0, 0, DSPP_0), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_2, LM_3, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sc7180_lm_sblk, PINGPONG_3, LM_2, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sc7180_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, }; static const struct dpu_dspp_cfg sc7280_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sc7280_pp[] = { - PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, 0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, 0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, 0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, 0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = 0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, }; /* NOTE: sc7280 only has one DSC hard slice encoder */ static const struct dpu_dsc_cfg sc7280_dsc[] = { - DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, }; static const struct dpu_wb_cfg sc7280_wb[] = { - WB_BLK("wb_2", WB_2, 0x65000, WB_SM8250_MASK, DPU_CLK_CTRL_WB2, 6, - VBIF_RT, MDP_SSPP_TOP0_INTR, 4096, 4), + { + .name = "wb_2", .id = WB_2, + .base = 0x65000, .len = 0x2c8, + .features = WB_SM8250_MASK, + .format_list = wb2_formats, + .num_formats = ARRAY_SIZE(wb2_formats), + .clk_ctrl = DPU_CLK_CTRL_WB2, + .xin_id = 6, + .vbif_idx = VBIF_RT, + .maxlinewidth = 4096, + .intr_wb_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 4), + }, }; static const struct dpu_intf_cfg sc7280_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x2c4, INTF_DSI, 0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), - INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x2c4, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x39000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sc7280_perf_data = { @@ -153,11 +246,15 @@ static const struct dpu_perf_cfg sc7280_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sc7280_mdss_ver = { + .core_major_ver = 7, + .core_minor_ver = 2, +}; + const struct dpu_mdss_cfg dpu_sc7280_cfg = { + .mdss_ver = &sc7280_mdss_ver, .caps = &sc7280_dpu_caps, - .ubwc = &sc7280_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sc7280_mdp), - .mdp = sc7280_mdp, + .mdp = &sc7280_mdp, .ctl_count = ARRAY_SIZE(sc7280_ctl), .ctl = sc7280_ctl, .sspp_count = ARRAY_SIZE(sc7280_sspp), @@ -177,13 +274,6 @@ const struct dpu_mdss_cfg dpu_sc7280_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sc7280_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF5_7xxx_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h index f6ce6b090f71..58f5e25679b1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_0_sc8280xp.h @@ -19,127 +19,263 @@ static const struct dpu_caps sc8280xp_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sc8280xp_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_40, - .highest_bank_bit = 2, - .ubwc_swizzle = 6, -}; - -static const struct dpu_mdp_cfg sc8280xp_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sc8280xp_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, .features = BIT(DPU_MDP_PERIPH_0_REMOVED), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, }; /* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sc8280xp_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x15000, .len = 0x204, - .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x16000, .len = 0x204, - .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x17000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x18000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x19000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x204, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x204, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sc8280xp_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x2ac, VIG_SC7180_MASK, - sm8250_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x2ac, VIG_SC7180_MASK, - sm8250_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x2ac, VIG_SC7180_MASK, - sm8250_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x2ac, VIG_SC7180_MASK, - sm8250_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x2ac, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x2ac, DMA_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x2ac, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x2ac, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x2ac, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x2ac, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x2ac, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x2ac, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x2ac, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x2ac, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x2ac, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x2ac, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sc8280xp_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_2, LM_3, DSPP_2), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_3, LM_2, DSPP_3), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sc8280xp_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sc8280xp_pp[] = { - PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), -1), - PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), -1), - PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), -1), - PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), -1), - PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), -1), - PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x4e000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x4f000), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, }; /* @@ -148,45 +284,132 @@ static const struct dpu_merge_3d_cfg sc8280xp_merge_3d[] = { * its own different sub block address. */ static const struct dpu_dsc_cfg sc8280xp_dsc[] = { - DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), - DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), - DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), - DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), - DSC_BLK_1_2("dce_2_0", DSC_4, 0x82000, 0x29c, 0, dsc_sblk_0), - DSC_BLK_1_2("dce_2_1", DSC_5, 0x82000, 0x29c, 0, dsc_sblk_1), + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_2_0", .id = DSC_4, + .base = 0x82000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_2_1", .id = DSC_5, + .base = 0x82000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_1, + }, }; /* TODO: INTF 3, 8 and 7 are used for MST, marked as INTF_NONE for now */ static const struct dpu_intf_cfg sc8280xp_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), - INTF_BLK("intf_4", INTF_4, 0x38000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21)), - INTF_BLK("intf_5", INTF_5, 0x39000, 0x280, INTF_DP, MSM_DP_CONTROLLER_3, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23)), - INTF_BLK("intf_6", INTF_6, 0x3a000, 0x280, INTF_DP, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17)), - INTF_BLK("intf_7", INTF_7, 0x3b000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_2, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19)), - INTF_BLK("intf_8", INTF_8, 0x3c000, 0x280, INTF_NONE, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_NONE, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_4", .id = INTF_4, + .base = 0x38000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 20), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 21), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_5", .id = INTF_5, + .base = 0x39000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_3, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 22), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 23), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_6", .id = INTF_6, + .base = 0x3a000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 16), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 17), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_7", .id = INTF_7, + .base = 0x3b000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_NONE, + .controller_id = MSM_DP_CONTROLLER_2, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 18), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 19), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_8", .id = INTF_8, + .base = 0x3c000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_NONE, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 12), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 13), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sc8280xp_perf_data = { @@ -216,11 +439,15 @@ static const struct dpu_perf_cfg sc8280xp_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sc8280xp_mdss_ver = { + .core_major_ver = 8, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_sc8280xp_cfg = { + .mdss_ver = &sc8280xp_mdss_ver, .caps = &sc8280xp_dpu_caps, - .ubwc = &sc8280xp_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sc8280xp_mdp), - .mdp = sc8280xp_mdp, + .mdp = &sc8280xp_mdp, .ctl_count = ARRAY_SIZE(sc8280xp_ctl), .ctl = sc8280xp_ctl, .sspp_count = ARRAY_SIZE(sc8280xp_sspp), @@ -240,20 +467,6 @@ const struct dpu_mdss_cfg dpu_sc8280xp_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sc8280xp_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF2_7xxx_INTR) | \ - BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF3_7xxx_INTR) | \ - BIT(MDP_INTF4_7xxx_INTR) | \ - BIT(MDP_INTF5_7xxx_INTR) | \ - BIT(MDP_INTF6_7xxx_INTR) | \ - BIT(MDP_INTF7_7xxx_INTR) | \ - BIT(MDP_INTF8_7xxx_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h index 8d13c369213c..1b12178dfbca 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_8_1_sm8450.h @@ -19,146 +19,282 @@ static const struct dpu_caps sm8450_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm8450_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_40, - .highest_bank_bit = 0x3, /* TODO: 2 for LP_DDR4 */ - .ubwc_swizzle = 0x6, -}; - -static const struct dpu_mdp_cfg sm8450_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm8450_mdp = { + .name = "top_0", .base = 0x0, .len = 0x494, .features = BIT(DPU_MDP_PERIPH_0_REMOVED), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, - .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x2ac, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x2b4, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x2bc, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0x2c4, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x2ac, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x2b4, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x2bc, .bit_off = 8 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2c4, .bit_off = 8 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, }; /* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sm8450_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x15000, .len = 0x204, - .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x16000, .len = 0x204, - .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x17000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x18000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x19000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a000, .len = 0x204, - .features = CTL_SC7280_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x204, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x204, + .features = BIT(DPU_CTL_SPLIT_DISPLAY) | CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x204, + .features = CTL_SC7280_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sm8450_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x32c, VIG_SC7180_MASK, - sm8250_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x32c, VIG_SC7180_MASK, - sm8250_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x32c, VIG_SC7180_MASK, - sm8250_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x32c, VIG_SC7180_MASK, - sm8250_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x32c, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x32c, DMA_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x32c, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x32c, DMA_CURSOR_SDM845_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x32c, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x32c, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x32c, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x32c, + .features = VIG_SC7180_MASK, + .sblk = &sm8250_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x32c, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x32c, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x32c, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x32c, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, }; static const struct dpu_lm_cfg sm8450_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, DSPP_2), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, DSPP_3), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + .dspp = DSPP_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + .dspp = DSPP_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sm8450_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm8450_pp[] = { - PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), - PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), - -1), - PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), - -1), - PP_BLK_DITHER("pingpong_6", PINGPONG_6, 0x65800, MERGE_3D_3, sc7280_pp_sblk, - -1, - -1), - PP_BLK_DITHER("pingpong_7", PINGPONG_7, 0x65c00, MERGE_3D_3, sc7280_pp_sblk, - -1, - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, { + .name = "pingpong_6", .id = PINGPONG_6, + .base = 0x65800, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = -1, + .intr_rdptr = -1, + }, { + .name = "pingpong_7", .id = PINGPONG_7, + .base = 0x65c00, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = -1, + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sm8450_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x4e000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x4f000), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000), - MERGE_3D_BLK("merge_3d_3", MERGE_3D_3, 0x65f00), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x65f00, .len = 0x8, + }, }; /* @@ -167,27 +303,71 @@ static const struct dpu_merge_3d_cfg sm8450_merge_3d[] = { * its own different sub block address. */ static const struct dpu_dsc_cfg sm8450_dsc[] = { - DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), - DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), - DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), - DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, }; static const struct dpu_intf_cfg sm8450_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sm8450_perf_data = { @@ -220,11 +400,15 @@ static const struct dpu_perf_cfg sm8450_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm8450_mdss_ver = { + .core_major_ver = 8, + .core_minor_ver = 1, +}; + const struct dpu_mdss_cfg dpu_sm8450_cfg = { + .mdss_ver = &sm8450_mdss_ver, .caps = &sm8450_dpu_caps, - .ubwc = &sm8450_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm8450_mdp), - .mdp = sm8450_mdp, + .mdp = &sm8450_mdp, .ctl_count = ARRAY_SIZE(sm8450_ctl), .ctl = sm8450_ctl, .sspp_count = ARRAY_SIZE(sm8450_sspp), @@ -244,15 +428,6 @@ const struct dpu_mdss_cfg dpu_sm8450_cfg = { .vbif_count = ARRAY_SIZE(sdm845_vbif), .vbif = sdm845_vbif, .perf = &sm8450_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF2_7xxx_INTR) | \ - BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF3_7xxx_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h index f17b9a7fee85..7bed819dfc39 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h +++ b/drivers/gpu/drm/msm/disp/dpu1/catalog/dpu_9_0_sm8550.h @@ -19,150 +19,297 @@ static const struct dpu_caps sm8550_dpu_caps = { .pixel_ram_size = DEFAULT_PIXEL_RAM_SIZE, }; -static const struct dpu_ubwc_cfg sm8550_ubwc_cfg = { - .ubwc_version = DPU_HW_UBWC_VER_40, - .highest_bank_bit = 0x3, /* TODO: 2 for LP_DDR4 */ -}; - -static const struct dpu_mdp_cfg sm8550_mdp[] = { - { - .name = "top_0", .id = MDP_TOP, +static const struct dpu_mdp_cfg sm8550_mdp = { + .name = "top_0", .base = 0, .len = 0x494, .features = BIT(DPU_MDP_PERIPH_0_REMOVED), - .clk_ctrls[DPU_CLK_CTRL_VIG0] = { .reg_off = 0x4330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG1] = { .reg_off = 0x6330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG2] = { .reg_off = 0x8330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_VIG3] = { .reg_off = 0xa330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA0] = { .reg_off = 0x24330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA1] = { .reg_off = 0x26330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA2] = { .reg_off = 0x28330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2a330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA4] = { .reg_off = 0x2c330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_DMA5] = { .reg_off = 0x2e330, .bit_off = 0 }, - .clk_ctrls[DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, + .clk_ctrls = { + [DPU_CLK_CTRL_VIG0] = { .reg_off = 0x4330, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG1] = { .reg_off = 0x6330, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG2] = { .reg_off = 0x8330, .bit_off = 0 }, + [DPU_CLK_CTRL_VIG3] = { .reg_off = 0xa330, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA0] = { .reg_off = 0x24330, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA1] = { .reg_off = 0x26330, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA2] = { .reg_off = 0x28330, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA3] = { .reg_off = 0x2a330, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA4] = { .reg_off = 0x2c330, .bit_off = 0 }, + [DPU_CLK_CTRL_DMA5] = { .reg_off = 0x2e330, .bit_off = 0 }, + [DPU_CLK_CTRL_REG_DMA] = { .reg_off = 0x2bc, .bit_off = 20 }, }, }; /* FIXME: get rid of DPU_CTL_SPLIT_DISPLAY in favour of proper ACTIVE_CTL support */ static const struct dpu_ctl_cfg sm8550_ctl[] = { { - .name = "ctl_0", .id = CTL_0, - .base = 0x15000, .len = 0x290, - .features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), - }, - { - .name = "ctl_1", .id = CTL_1, - .base = 0x16000, .len = 0x290, - .features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY), - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), - }, - { - .name = "ctl_2", .id = CTL_2, - .base = 0x17000, .len = 0x290, - .features = CTL_SM8550_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), - }, - { - .name = "ctl_3", .id = CTL_3, - .base = 0x18000, .len = 0x290, - .features = CTL_SM8550_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), - }, - { - .name = "ctl_4", .id = CTL_4, - .base = 0x19000, .len = 0x290, - .features = CTL_SM8550_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), - }, - { - .name = "ctl_5", .id = CTL_5, - .base = 0x1a000, .len = 0x290, - .features = CTL_SM8550_MASK, - .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), + .name = "ctl_0", .id = CTL_0, + .base = 0x15000, .len = 0x290, + .features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 9), + }, { + .name = "ctl_1", .id = CTL_1, + .base = 0x16000, .len = 0x290, + .features = CTL_SM8550_MASK | BIT(DPU_CTL_SPLIT_DISPLAY), + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 10), + }, { + .name = "ctl_2", .id = CTL_2, + .base = 0x17000, .len = 0x290, + .features = CTL_SM8550_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 11), + }, { + .name = "ctl_3", .id = CTL_3, + .base = 0x18000, .len = 0x290, + .features = CTL_SM8550_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 12), + }, { + .name = "ctl_4", .id = CTL_4, + .base = 0x19000, .len = 0x290, + .features = CTL_SM8550_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 13), + }, { + .name = "ctl_5", .id = CTL_5, + .base = 0x1a000, .len = 0x290, + .features = CTL_SM8550_MASK, + .intr_start = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 23), }, }; static const struct dpu_sspp_cfg sm8550_sspp[] = { - SSPP_BLK("sspp_0", SSPP_VIG0, 0x4000, 0x344, VIG_SC7180_MASK, - sm8550_vig_sblk_0, 0, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG0), - SSPP_BLK("sspp_1", SSPP_VIG1, 0x6000, 0x344, VIG_SC7180_MASK, - sm8550_vig_sblk_1, 4, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG1), - SSPP_BLK("sspp_2", SSPP_VIG2, 0x8000, 0x344, VIG_SC7180_MASK, - sm8550_vig_sblk_2, 8, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG2), - SSPP_BLK("sspp_3", SSPP_VIG3, 0xa000, 0x344, VIG_SC7180_MASK, - sm8550_vig_sblk_3, 12, SSPP_TYPE_VIG, DPU_CLK_CTRL_VIG3), - SSPP_BLK("sspp_8", SSPP_DMA0, 0x24000, 0x344, DMA_SDM845_MASK, - sdm845_dma_sblk_0, 1, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA0), - SSPP_BLK("sspp_9", SSPP_DMA1, 0x26000, 0x344, DMA_SDM845_MASK, - sdm845_dma_sblk_1, 5, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA1), - SSPP_BLK("sspp_10", SSPP_DMA2, 0x28000, 0x344, DMA_SDM845_MASK, - sdm845_dma_sblk_2, 9, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA2), - SSPP_BLK("sspp_11", SSPP_DMA3, 0x2a000, 0x344, DMA_SDM845_MASK, - sdm845_dma_sblk_3, 13, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA3), - SSPP_BLK("sspp_12", SSPP_DMA4, 0x2c000, 0x344, DMA_CURSOR_SDM845_MASK, - sm8550_dma_sblk_4, 14, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA4), - SSPP_BLK("sspp_13", SSPP_DMA5, 0x2e000, 0x344, DMA_CURSOR_SDM845_MASK, - sm8550_dma_sblk_5, 15, SSPP_TYPE_DMA, DPU_CLK_CTRL_DMA5), + { + .name = "sspp_0", .id = SSPP_VIG0, + .base = 0x4000, .len = 0x344, + .features = VIG_SC7180_MASK, + .sblk = &sm8550_vig_sblk_0, + .xin_id = 0, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG0, + }, { + .name = "sspp_1", .id = SSPP_VIG1, + .base = 0x6000, .len = 0x344, + .features = VIG_SC7180_MASK, + .sblk = &sm8550_vig_sblk_1, + .xin_id = 4, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG1, + }, { + .name = "sspp_2", .id = SSPP_VIG2, + .base = 0x8000, .len = 0x344, + .features = VIG_SC7180_MASK, + .sblk = &sm8550_vig_sblk_2, + .xin_id = 8, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG2, + }, { + .name = "sspp_3", .id = SSPP_VIG3, + .base = 0xa000, .len = 0x344, + .features = VIG_SC7180_MASK, + .sblk = &sm8550_vig_sblk_3, + .xin_id = 12, + .type = SSPP_TYPE_VIG, + .clk_ctrl = DPU_CLK_CTRL_VIG3, + }, { + .name = "sspp_8", .id = SSPP_DMA0, + .base = 0x24000, .len = 0x344, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_0, + .xin_id = 1, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA0, + }, { + .name = "sspp_9", .id = SSPP_DMA1, + .base = 0x26000, .len = 0x344, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_1, + .xin_id = 5, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA1, + }, { + .name = "sspp_10", .id = SSPP_DMA2, + .base = 0x28000, .len = 0x344, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_2, + .xin_id = 9, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA2, + }, { + .name = "sspp_11", .id = SSPP_DMA3, + .base = 0x2a000, .len = 0x344, + .features = DMA_SDM845_MASK, + .sblk = &sdm845_dma_sblk_3, + .xin_id = 13, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA3, + }, { + .name = "sspp_12", .id = SSPP_DMA4, + .base = 0x2c000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sm8550_dma_sblk_4, + .xin_id = 14, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA4, + }, { + .name = "sspp_13", .id = SSPP_DMA5, + .base = 0x2e000, .len = 0x344, + .features = DMA_CURSOR_SDM845_MASK, + .sblk = &sm8550_dma_sblk_5, + .xin_id = 15, + .type = SSPP_TYPE_DMA, + .clk_ctrl = DPU_CLK_CTRL_DMA5, + }, }; static const struct dpu_lm_cfg sm8550_lm[] = { - LM_BLK("lm_0", LM_0, 0x44000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_0, LM_1, DSPP_0), - LM_BLK("lm_1", LM_1, 0x45000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_1, LM_0, DSPP_1), - LM_BLK("lm_2", LM_2, 0x46000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_2, LM_3, 0), - LM_BLK("lm_3", LM_3, 0x47000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_3, LM_2, 0), - LM_BLK("lm_4", LM_4, 0x48000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_4, LM_5, 0), - LM_BLK("lm_5", LM_5, 0x49000, MIXER_SDM845_MASK, - &sdm845_lm_sblk, PINGPONG_5, LM_4, 0), + { + .name = "lm_0", .id = LM_0, + .base = 0x44000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_1, + .pingpong = PINGPONG_0, + .dspp = DSPP_0, + }, { + .name = "lm_1", .id = LM_1, + .base = 0x45000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_0, + .pingpong = PINGPONG_1, + .dspp = DSPP_1, + }, { + .name = "lm_2", .id = LM_2, + .base = 0x46000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_3, + .pingpong = PINGPONG_2, + }, { + .name = "lm_3", .id = LM_3, + .base = 0x47000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_2, + .pingpong = PINGPONG_3, + }, { + .name = "lm_4", .id = LM_4, + .base = 0x48000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_5, + .pingpong = PINGPONG_4, + }, { + .name = "lm_5", .id = LM_5, + .base = 0x49000, .len = 0x320, + .features = MIXER_SDM845_MASK, + .sblk = &sdm845_lm_sblk, + .lm_pair = LM_4, + .pingpong = PINGPONG_5, + }, }; static const struct dpu_dspp_cfg sm8550_dspp[] = { - DSPP_BLK("dspp_0", DSPP_0, 0x54000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_1", DSPP_1, 0x56000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_2", DSPP_2, 0x58000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), - DSPP_BLK("dspp_3", DSPP_3, 0x5a000, DSPP_SC7180_MASK, - &sm8150_dspp_sblk), + { + .name = "dspp_0", .id = DSPP_0, + .base = 0x54000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_1", .id = DSPP_1, + .base = 0x56000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_2", .id = DSPP_2, + .base = 0x58000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, { + .name = "dspp_3", .id = DSPP_3, + .base = 0x5a000, .len = 0x1800, + .features = DSPP_SC7180_MASK, + .sblk = &sdm845_dspp_sblk, + }, }; static const struct dpu_pingpong_cfg sm8550_pp[] = { - PP_BLK_DITHER("pingpong_0", PINGPONG_0, 0x69000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), - -1), - PP_BLK_DITHER("pingpong_1", PINGPONG_1, 0x6a000, MERGE_3D_0, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), - -1), - PP_BLK_DITHER("pingpong_2", PINGPONG_2, 0x6b000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), - -1), - PP_BLK_DITHER("pingpong_3", PINGPONG_3, 0x6c000, MERGE_3D_1, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), - -1), - PP_BLK_DITHER("pingpong_4", PINGPONG_4, 0x6d000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), - -1), - PP_BLK_DITHER("pingpong_5", PINGPONG_5, 0x6e000, MERGE_3D_2, sc7280_pp_sblk, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), - -1), - PP_BLK_DITHER("pingpong_6", PINGPONG_6, 0x66000, MERGE_3D_3, sc7280_pp_sblk, - -1, - -1), - PP_BLK_DITHER("pingpong_7", PINGPONG_7, 0x66400, MERGE_3D_3, sc7280_pp_sblk, - -1, - -1), + { + .name = "pingpong_0", .id = PINGPONG_0, + .base = 0x69000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 8), + .intr_rdptr = -1, + }, { + .name = "pingpong_1", .id = PINGPONG_1, + .base = 0x6a000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_0, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 9), + .intr_rdptr = -1, + }, { + .name = "pingpong_2", .id = PINGPONG_2, + .base = 0x6b000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 10), + .intr_rdptr = -1, + }, { + .name = "pingpong_3", .id = PINGPONG_3, + .base = 0x6c000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_1, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 11), + .intr_rdptr = -1, + }, { + .name = "pingpong_4", .id = PINGPONG_4, + .base = 0x6d000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 30), + .intr_rdptr = -1, + }, { + .name = "pingpong_5", .id = PINGPONG_5, + .base = 0x6e000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_2, + .intr_done = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR2, 31), + .intr_rdptr = -1, + }, { + .name = "pingpong_6", .id = PINGPONG_6, + .base = 0x66000, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = -1, + .intr_rdptr = -1, + }, { + .name = "pingpong_7", .id = PINGPONG_7, + .base = 0x66400, .len = 0, + .features = BIT(DPU_PINGPONG_DITHER), + .sblk = &sc7280_pp_sblk, + .merge_3d = MERGE_3D_3, + .intr_done = -1, + .intr_rdptr = -1, + }, }; static const struct dpu_merge_3d_cfg sm8550_merge_3d[] = { - MERGE_3D_BLK("merge_3d_0", MERGE_3D_0, 0x4e000), - MERGE_3D_BLK("merge_3d_1", MERGE_3D_1, 0x4f000), - MERGE_3D_BLK("merge_3d_2", MERGE_3D_2, 0x50000), - MERGE_3D_BLK("merge_3d_3", MERGE_3D_3, 0x66700), + { + .name = "merge_3d_0", .id = MERGE_3D_0, + .base = 0x4e000, .len = 0x8, + }, { + .name = "merge_3d_1", .id = MERGE_3D_1, + .base = 0x4f000, .len = 0x8, + }, { + .name = "merge_3d_2", .id = MERGE_3D_2, + .base = 0x50000, .len = 0x8, + }, { + .name = "merge_3d_3", .id = MERGE_3D_3, + .base = 0x66700, .len = 0x8, + }, }; /* @@ -171,27 +318,71 @@ static const struct dpu_merge_3d_cfg sm8550_merge_3d[] = { * its own different sub block address. */ static const struct dpu_dsc_cfg sm8550_dsc[] = { - DSC_BLK_1_2("dce_0_0", DSC_0, 0x80000, 0x29c, 0, dsc_sblk_0), - DSC_BLK_1_2("dce_0_1", DSC_1, 0x80000, 0x29c, 0, dsc_sblk_1), - DSC_BLK_1_2("dce_1_0", DSC_2, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_0), - DSC_BLK_1_2("dce_1_1", DSC_3, 0x81000, 0x29c, BIT(DPU_DSC_NATIVE_42x_EN), dsc_sblk_1), + { + .name = "dce_0_0", .id = DSC_0, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_0_1", .id = DSC_1, + .base = 0x80000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2), + .sblk = &dsc_sblk_1, + }, { + .name = "dce_1_0", .id = DSC_2, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_0, + }, { + .name = "dce_1_1", .id = DSC_3, + .base = 0x81000, .len = 0x4, + .features = BIT(DPU_DSC_HW_REV_1_2) | BIT(DPU_DSC_NATIVE_42x_EN), + .sblk = &dsc_sblk_1, + }, }; static const struct dpu_intf_cfg sm8550_intf[] = { - INTF_BLK("intf_0", INTF_0, 0x34000, 0x280, INTF_DP, MSM_DP_CONTROLLER_0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25)), - INTF_BLK_DSI_TE("intf_1", INTF_1, 0x35000, 0x300, INTF_DSI, 0, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), - DPU_IRQ_IDX(MDP_INTF1_7xxx_TEAR_INTR, 2)), - INTF_BLK_DSI_TE("intf_2", INTF_2, 0x36000, 0x300, INTF_DSI, 1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), - DPU_IRQ_IDX(MDP_INTF2_7xxx_TEAR_INTR, 2)), - INTF_BLK("intf_3", INTF_3, 0x37000, 0x280, INTF_DP, MSM_DP_CONTROLLER_1, 24, INTF_SC7280_MASK, - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), - DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31)), + { + .name = "intf_0", .id = INTF_0, + .base = 0x34000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 24), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 25), + .intr_tear_rd_ptr = -1, + }, { + .name = "intf_1", .id = INTF_1, + .base = 0x35000, .len = 0x300, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_0, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 26), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 27), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF1_TEAR_INTR, 2), + }, { + .name = "intf_2", .id = INTF_2, + .base = 0x36000, .len = 0x300, + .features = INTF_SC7280_MASK, + .type = INTF_DSI, + .controller_id = MSM_DSI_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 28), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 29), + .intr_tear_rd_ptr = DPU_IRQ_IDX(MDP_INTF2_TEAR_INTR, 2), + }, { + .name = "intf_3", .id = INTF_3, + .base = 0x37000, .len = 0x280, + .features = INTF_SC7280_MASK, + .type = INTF_DP, + .controller_id = MSM_DP_CONTROLLER_1, + .prog_fetch_lines_worst_case = 24, + .intr_underrun = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 30), + .intr_vsync = DPU_IRQ_IDX(MDP_SSPP_TOP0_INTR, 31), + .intr_tear_rd_ptr = -1, + }, }; static const struct dpu_perf_cfg sm8550_perf_data = { @@ -224,11 +415,15 @@ static const struct dpu_perf_cfg sm8550_perf_data = { .bw_inefficiency_factor = 120, }; +static const struct dpu_mdss_version sm8550_mdss_ver = { + .core_major_ver = 9, + .core_minor_ver = 0, +}; + const struct dpu_mdss_cfg dpu_sm8550_cfg = { + .mdss_ver = &sm8550_mdss_ver, .caps = &sm8550_dpu_caps, - .ubwc = &sm8550_ubwc_cfg, - .mdp_count = ARRAY_SIZE(sm8550_mdp), - .mdp = sm8550_mdp, + .mdp = &sm8550_mdp, .ctl_count = ARRAY_SIZE(sm8550_ctl), .ctl = sm8550_ctl, .sspp_count = ARRAY_SIZE(sm8550_sspp), @@ -245,18 +440,9 @@ const struct dpu_mdss_cfg dpu_sm8550_cfg = { .merge_3d = sm8550_merge_3d, .intf_count = ARRAY_SIZE(sm8550_intf), .intf = sm8550_intf, - .vbif_count = ARRAY_SIZE(sdm845_vbif), - .vbif = sdm845_vbif, + .vbif_count = ARRAY_SIZE(sm8550_vbif), + .vbif = sm8550_vbif, .perf = &sm8550_perf_data, - .mdss_irqs = BIT(MDP_SSPP_TOP0_INTR) | \ - BIT(MDP_SSPP_TOP0_INTR2) | \ - BIT(MDP_SSPP_TOP0_HIST_INTR) | \ - BIT(MDP_INTF0_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_INTR) | \ - BIT(MDP_INTF1_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF2_7xxx_INTR) | \ - BIT(MDP_INTF2_7xxx_TEAR_INTR) | \ - BIT(MDP_INTF3_7xxx_INTR), }; #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c index 1d9d83d7b99e..ef871239adb2 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c @@ -33,11 +33,11 @@ enum dpu_perf_mode { /** * _dpu_core_perf_calc_bw() - to calculate BW per crtc - * @kms: pointer to the dpu_kms + * @perf_cfg: performance configuration * @crtc: pointer to a crtc * Return: returns aggregated BW for all planes in crtc. */ -static u64 _dpu_core_perf_calc_bw(struct dpu_kms *kms, +static u64 _dpu_core_perf_calc_bw(const struct dpu_perf_cfg *perf_cfg, struct drm_crtc *crtc) { struct drm_plane *plane; @@ -53,7 +53,7 @@ static u64 _dpu_core_perf_calc_bw(struct dpu_kms *kms, crtc_plane_bw += pstate->plane_fetch_bw; } - bw_factor = kms->catalog->perf->bw_inefficiency_factor; + bw_factor = perf_cfg->bw_inefficiency_factor; if (bw_factor) { crtc_plane_bw *= bw_factor; do_div(crtc_plane_bw, 100); @@ -64,12 +64,12 @@ static u64 _dpu_core_perf_calc_bw(struct dpu_kms *kms, /** * _dpu_core_perf_calc_clk() - to calculate clock per crtc - * @kms: pointer to the dpu_kms + * @perf_cfg: performance configuration * @crtc: pointer to a crtc * @state: pointer to a crtc state * Return: returns max clk for all planes in crtc. */ -static u64 _dpu_core_perf_calc_clk(struct dpu_kms *kms, +static u64 _dpu_core_perf_calc_clk(const struct dpu_perf_cfg *perf_cfg, struct drm_crtc *crtc, struct drm_crtc_state *state) { struct drm_plane *plane; @@ -90,7 +90,7 @@ static u64 _dpu_core_perf_calc_clk(struct dpu_kms *kms, crtc_clk = max(pstate->plane_clk, crtc_clk); } - clk_factor = kms->catalog->perf->clk_inefficiency_factor; + clk_factor = perf_cfg->clk_inefficiency_factor; if (clk_factor) { crtc_clk *= clk_factor; do_div(crtc_clk, 100); @@ -106,30 +106,32 @@ static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc) return to_dpu_kms(priv->kms); } -static void _dpu_core_perf_calc_crtc(struct dpu_kms *kms, - struct drm_crtc *crtc, - struct drm_crtc_state *state, - struct dpu_core_perf_params *perf) +static void _dpu_core_perf_calc_crtc(const struct dpu_core_perf *core_perf, + struct drm_crtc *crtc, + struct drm_crtc_state *state, + struct dpu_core_perf_params *perf) { - if (!kms || !kms->catalog || !crtc || !state || !perf) { + const struct dpu_perf_cfg *perf_cfg = core_perf->perf_cfg; + + if (!perf_cfg || !crtc || !state || !perf) { DPU_ERROR("invalid parameters\n"); return; } memset(perf, 0, sizeof(struct dpu_core_perf_params)); - if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) { + if (core_perf->perf_tune.mode == DPU_PERF_MODE_MINIMUM) { perf->bw_ctl = 0; perf->max_per_pipe_ib = 0; perf->core_clk_rate = 0; - } else if (kms->perf.perf_tune.mode == DPU_PERF_MODE_FIXED) { - perf->bw_ctl = kms->perf.fix_core_ab_vote; - perf->max_per_pipe_ib = kms->perf.fix_core_ib_vote; - perf->core_clk_rate = kms->perf.fix_core_clk_rate; + } else if (core_perf->perf_tune.mode == DPU_PERF_MODE_FIXED) { + perf->bw_ctl = core_perf->fix_core_ab_vote; + perf->max_per_pipe_ib = core_perf->fix_core_ib_vote; + perf->core_clk_rate = core_perf->fix_core_clk_rate; } else { - perf->bw_ctl = _dpu_core_perf_calc_bw(kms, crtc); - perf->max_per_pipe_ib = kms->catalog->perf->min_dram_ib; - perf->core_clk_rate = _dpu_core_perf_calc_clk(kms, crtc, state); + perf->bw_ctl = _dpu_core_perf_calc_bw(perf_cfg, crtc); + perf->max_per_pipe_ib = perf_cfg->min_dram_ib; + perf->core_clk_rate = _dpu_core_perf_calc_clk(perf_cfg, crtc, state); } DRM_DEBUG_ATOMIC( @@ -154,10 +156,6 @@ int dpu_core_perf_crtc_check(struct drm_crtc *crtc, } kms = _dpu_crtc_get_kms(crtc); - if (!kms->catalog) { - DPU_ERROR("invalid parameters\n"); - return 0; - } /* we only need bandwidth check on real-time clients (interfaces) */ if (dpu_crtc_get_client_type(crtc) == NRT_CLIENT) @@ -166,30 +164,30 @@ int dpu_core_perf_crtc_check(struct drm_crtc *crtc, dpu_cstate = to_dpu_crtc_state(state); /* obtain new values */ - _dpu_core_perf_calc_crtc(kms, crtc, state, &dpu_cstate->new_perf); + _dpu_core_perf_calc_crtc(&kms->perf, crtc, state, &dpu_cstate->new_perf); bw_sum_of_intfs = dpu_cstate->new_perf.bw_ctl; curr_client_type = dpu_crtc_get_client_type(crtc); drm_for_each_crtc(tmp_crtc, crtc->dev) { if (tmp_crtc->enabled && - (dpu_crtc_get_client_type(tmp_crtc) == - curr_client_type) && (tmp_crtc != crtc)) { + dpu_crtc_get_client_type(tmp_crtc) == curr_client_type && + tmp_crtc != crtc) { struct dpu_crtc_state *tmp_cstate = to_dpu_crtc_state(tmp_crtc->state); DRM_DEBUG_ATOMIC("crtc:%d bw:%llu ctrl:%d\n", - tmp_crtc->base.id, tmp_cstate->new_perf.bw_ctl, - tmp_cstate->bw_control); + tmp_crtc->base.id, tmp_cstate->new_perf.bw_ctl, + tmp_cstate->bw_control); - bw_sum_of_intfs += tmp_cstate->new_perf.bw_ctl; + bw_sum_of_intfs += tmp_cstate->new_perf.bw_ctl; } /* convert bandwidth to kb */ bw = DIV_ROUND_UP_ULL(bw_sum_of_intfs, 1000); DRM_DEBUG_ATOMIC("calculated bandwidth=%uk\n", bw); - threshold = kms->catalog->perf->max_bw_high; + threshold = kms->perf.perf_cfg->max_bw_high; DRM_DEBUG_ATOMIC("final threshold bw limit = %d\n", threshold); @@ -217,6 +215,9 @@ static int _dpu_core_perf_crtc_update_bus(struct dpu_kms *kms, int i, ret = 0; u64 avg_bw; + if (!kms->num_paths) + return 0; + drm_for_each_crtc(tmp_crtc, crtc->dev) { if (tmp_crtc->enabled && curr_client_type == @@ -234,9 +235,6 @@ static int _dpu_core_perf_crtc_update_bus(struct dpu_kms *kms, } } - if (!kms->num_paths) - return 0; - avg_bw = perf.bw_ctl; do_div(avg_bw, (kms->num_paths * 1000)); /*Bps_to_icc*/ @@ -265,11 +263,6 @@ void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc) } kms = _dpu_crtc_get_kms(crtc); - if (!kms->catalog) { - DPU_ERROR("invalid kms\n"); - return; - } - dpu_crtc = to_dpu_crtc(crtc); if (atomic_dec_return(&kms->bandwidth_ref) > 0) @@ -286,30 +279,30 @@ void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc) static u64 _dpu_core_perf_get_core_clk_rate(struct dpu_kms *kms) { - u64 clk_rate = kms->perf.perf_tune.min_core_clk; + u64 clk_rate; struct drm_crtc *crtc; struct dpu_crtc_state *dpu_cstate; + if (kms->perf.perf_tune.mode == DPU_PERF_MODE_FIXED) + return kms->perf.fix_core_clk_rate; + + if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) + return kms->perf.max_core_clk_rate; + + clk_rate = 0; drm_for_each_crtc(crtc, kms->dev) { if (crtc->enabled) { dpu_cstate = to_dpu_crtc_state(crtc->state); clk_rate = max(dpu_cstate->new_perf.core_clk_rate, clk_rate); - clk_rate = clk_round_rate(kms->perf.core_clk, - clk_rate); } } - if (kms->perf.perf_tune.mode == DPU_PERF_MODE_FIXED) - clk_rate = kms->perf.fix_core_clk_rate; - - DRM_DEBUG_ATOMIC("clk:%llu\n", clk_rate); - return clk_rate; } int dpu_core_perf_crtc_update(struct drm_crtc *crtc, - int params_changed, bool stop_req) + int params_changed) { struct dpu_core_perf_params *new, *old; bool update_bus = false, update_clk = false; @@ -325,21 +318,17 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, } kms = _dpu_crtc_get_kms(crtc); - if (!kms->catalog) { - DPU_ERROR("invalid kms\n"); - return -EINVAL; - } dpu_crtc = to_dpu_crtc(crtc); dpu_cstate = to_dpu_crtc_state(crtc->state); - DRM_DEBUG_ATOMIC("crtc:%d stop_req:%d core_clk:%llu\n", - crtc->base.id, stop_req, kms->perf.core_clk_rate); + DRM_DEBUG_ATOMIC("crtc:%d enabled:%d core_clk:%llu\n", + crtc->base.id, crtc->enabled, kms->perf.core_clk_rate); old = &dpu_crtc->cur_perf; new = &dpu_cstate->new_perf; - if (crtc->enabled && !stop_req) { + if (crtc->enabled) { /* * cases for bus bandwidth update. * 1. new bandwidth vote - "ab or ib vote" is higher @@ -359,10 +348,8 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, update_bus = true; } - if ((params_changed && - (new->core_clk_rate > old->core_clk_rate)) || - (!params_changed && - (new->core_clk_rate < old->core_clk_rate))) { + if ((params_changed && new->core_clk_rate > old->core_clk_rate) || + (!params_changed && new->core_clk_rate < old->core_clk_rate)) { old->core_clk_rate = new->core_clk_rate; update_clk = true; } @@ -374,7 +361,7 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, } trace_dpu_perf_crtc_update(crtc->base.id, new->bw_ctl, - new->core_clk_rate, stop_req, update_bus, update_clk); + new->core_clk_rate, !crtc->enabled, update_bus, update_clk); if (update_bus) { ret = _dpu_core_perf_crtc_update_bus(kms, crtc); @@ -392,7 +379,9 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, if (update_clk) { clk_rate = _dpu_core_perf_get_core_clk_rate(kms); - trace_dpu_core_perf_update_clk(kms->dev, stop_req, clk_rate); + DRM_DEBUG_ATOMIC("clk:%llu\n", clk_rate); + + trace_dpu_core_perf_update_clk(kms->dev, !crtc->enabled, clk_rate); clk_rate = min(clk_rate, kms->perf.max_core_clk_rate); ret = dev_pm_opp_set_rate(&kms->pdev->dev, clk_rate); @@ -413,7 +402,6 @@ static ssize_t _dpu_core_perf_mode_write(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) { struct dpu_core_perf *perf = file->private_data; - const struct dpu_perf_cfg *cfg = perf->catalog->perf; u32 perf_mode = 0; int ret; @@ -428,14 +416,9 @@ static ssize_t _dpu_core_perf_mode_write(struct file *file, DRM_INFO("fix performance mode\n"); } else if (perf_mode == DPU_PERF_MODE_MINIMUM) { /* run the driver with max clk and BW vote */ - perf->perf_tune.min_core_clk = perf->max_core_clk_rate; - perf->perf_tune.min_bus_vote = - (u64) cfg->max_bw_high * 1000; DRM_INFO("minimum performance mode\n"); } else if (perf_mode == DPU_PERF_MODE_NORMAL) { /* reset the perf tune params to 0 */ - perf->perf_tune.min_core_clk = 0; - perf->perf_tune.min_bus_vote = 0; DRM_INFO("normal performance mode\n"); } perf->perf_tune.mode = perf_mode; @@ -451,10 +434,8 @@ static ssize_t _dpu_core_perf_mode_read(struct file *file, char buf[128]; len = scnprintf(buf, sizeof(buf), - "mode %d min_mdp_clk %llu min_bus_vote %llu\n", - perf->perf_tune.mode, - perf->perf_tune.min_core_clk, - perf->perf_tune.min_bus_vote); + "mode %d\n", + perf->perf_tune.mode); return simple_read_from_buffer(buff, count, ppos, buf, len); } @@ -468,7 +449,6 @@ static const struct file_operations dpu_core_perf_mode_fops = { int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent) { struct dpu_core_perf *perf = &dpu_kms->perf; - const struct dpu_mdss_cfg *catalog = perf->catalog; struct dentry *entry; entry = debugfs_create_dir("core_perf", parent); @@ -480,15 +460,15 @@ int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent) debugfs_create_u32("enable_bw_release", 0600, entry, (u32 *)&perf->enable_bw_release); debugfs_create_u32("threshold_low", 0600, entry, - (u32 *)&catalog->perf->max_bw_low); + (u32 *)&perf->perf_cfg->max_bw_low); debugfs_create_u32("threshold_high", 0600, entry, - (u32 *)&catalog->perf->max_bw_high); + (u32 *)&perf->perf_cfg->max_bw_high); debugfs_create_u32("min_core_ib", 0600, entry, - (u32 *)&catalog->perf->min_core_ib); + (u32 *)&perf->perf_cfg->min_core_ib); debugfs_create_u32("min_llcc_ib", 0600, entry, - (u32 *)&catalog->perf->min_llcc_ib); + (u32 *)&perf->perf_cfg->min_llcc_ib); debugfs_create_u32("min_dram_ib", 0600, entry, - (u32 *)&catalog->perf->min_dram_ib); + (u32 *)&perf->perf_cfg->min_dram_ib); debugfs_create_file("perf_mode", 0600, entry, (u32 *)perf, &dpu_core_perf_mode_fops); debugfs_create_u64("fix_core_clk_rate", 0600, entry, @@ -502,33 +482,12 @@ int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent) } #endif -void dpu_core_perf_destroy(struct dpu_core_perf *perf) -{ - if (!perf) { - DPU_ERROR("invalid parameters\n"); - return; - } - - perf->max_core_clk_rate = 0; - perf->core_clk = NULL; - perf->catalog = NULL; - perf->dev = NULL; -} - int dpu_core_perf_init(struct dpu_core_perf *perf, - struct drm_device *dev, - const struct dpu_mdss_cfg *catalog, - struct clk *core_clk) + const struct dpu_perf_cfg *perf_cfg, + unsigned long max_core_clk_rate) { - perf->dev = dev; - perf->catalog = catalog; - perf->core_clk = core_clk; - - perf->max_core_clk_rate = clk_get_rate(core_clk); - if (!perf->max_core_clk_rate) { - DPU_DEBUG("optional max core clk rate, use default\n"); - perf->max_core_clk_rate = DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE; - } + perf->perf_cfg = perf_cfg; + perf->max_core_clk_rate = max_core_clk_rate; return 0; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h index e3795995e145..4186977390bd 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h @@ -12,21 +12,6 @@ #include "dpu_hw_catalog.h" -#define DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE 412500000 - -/** - * enum dpu_core_perf_data_bus_id - data bus identifier - * @DPU_CORE_PERF_DATA_BUS_ID_MNOC: DPU/MNOC data bus - * @DPU_CORE_PERF_DATA_BUS_ID_LLCC: MNOC/LLCC data bus - * @DPU_CORE_PERF_DATA_BUS_ID_EBI: LLCC/EBI data bus - */ -enum dpu_core_perf_data_bus_id { - DPU_CORE_PERF_DATA_BUS_ID_MNOC, - DPU_CORE_PERF_DATA_BUS_ID_LLCC, - DPU_CORE_PERF_DATA_BUS_ID_EBI, - DPU_CORE_PERF_DATA_BUS_ID_MAX, -}; - /** * struct dpu_core_perf_params - definition of performance parameters * @max_per_pipe_ib: maximum instantaneous bandwidth request @@ -42,21 +27,14 @@ struct dpu_core_perf_params { /** * struct dpu_core_perf_tune - definition of performance tuning control * @mode: performance mode - * @min_core_clk: minimum core clock - * @min_bus_vote: minimum bus vote */ struct dpu_core_perf_tune { u32 mode; - u64 min_core_clk; - u64 min_bus_vote; }; /** * struct dpu_core_perf - definition of core performance context - * @dev: Pointer to drm device - * @debugfs_root: top level debug folder - * @catalog: Pointer to catalog configuration - * @core_clk: Pointer to the core clock + * @perf_cfg: Platform-specific performance configuration * @core_clk_rate: current core clock rate * @max_core_clk_rate: maximum allowable core clock rate * @perf_tune: debug control for performance tuning @@ -66,10 +44,7 @@ struct dpu_core_perf_tune { * @fix_core_ab_vote: fixed core ab vote in bps used in mode 2 */ struct dpu_core_perf { - struct drm_device *dev; - struct dentry *debugfs_root; - const struct dpu_mdss_cfg *catalog; - struct clk *core_clk; + const struct dpu_perf_cfg *perf_cfg; u64 core_clk_rate; u64 max_core_clk_rate; struct dpu_core_perf_tune perf_tune; @@ -92,11 +67,10 @@ int dpu_core_perf_crtc_check(struct drm_crtc *crtc, * dpu_core_perf_crtc_update - update performance of the given crtc * @crtc: Pointer to crtc * @params_changed: true if crtc parameters are modified - * @stop_req: true if this is a stop request * return: zero if success, or error code otherwise */ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, - int params_changed, bool stop_req); + int params_changed); /** * dpu_core_perf_crtc_release_bw - release bandwidth of the given crtc @@ -105,22 +79,14 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc); /** - * dpu_core_perf_destroy - destroy the given core performance context - * @perf: Pointer to core performance context - */ -void dpu_core_perf_destroy(struct dpu_core_perf *perf); - -/** * dpu_core_perf_init - initialize the given core performance context * @perf: Pointer to core performance context - * @dev: Pointer to drm device - * @catalog: Pointer to catalog - * @core_clk: pointer to core clock + * @perf_cfg: Pointer to platform performance configuration + * @max_core_clk_rate: Maximum core clock rate */ int dpu_core_perf_init(struct dpu_core_perf *perf, - struct drm_device *dev, - const struct dpu_mdss_cfg *catalog, - struct clk *core_clk); + const struct dpu_perf_cfg *perf_cfg, + unsigned long max_core_clk_rate); struct dpu_kms; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index 1edf2b6b0a26..8ce7586e2ddf 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -718,7 +718,7 @@ static void dpu_crtc_frame_event_cb(void *data, u32 event) void dpu_crtc_complete_commit(struct drm_crtc *crtc) { trace_dpu_crtc_complete_commit(DRMID(crtc)); - dpu_core_perf_crtc_update(crtc, 0, false); + dpu_core_perf_crtc_update(crtc, 0); _dpu_crtc_complete_flip(crtc); } @@ -884,7 +884,7 @@ static void dpu_crtc_atomic_flush(struct drm_crtc *crtc, return; /* update performance setting before crtc kickoff */ - dpu_core_perf_crtc_update(crtc, 1, false); + dpu_core_perf_crtc_update(crtc, 1); /* * Final plane updates: Give each plane a chance to complete all @@ -1100,7 +1100,7 @@ static void dpu_crtc_disable(struct drm_crtc *crtc, atomic_set(&dpu_crtc->frame_pending, 0); } - dpu_core_perf_crtc_update(crtc, 0, true); + dpu_core_perf_crtc_update(crtc, 0); drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) dpu_encoder_register_frame_event_callback(encoder, NULL, NULL); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index 493905a5b63a..d34e684a4178 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -152,7 +152,6 @@ enum dpu_enc_rc_states { * @crtc_frame_event_cb_data: callback handler private data * @frame_done_timeout_ms: frame done timeout in ms * @frame_done_timer: watchdog timer for frame done event - * @vsync_event_timer: vsync timer * @disp_info: local copy of msm_display_info struct * @idle_pc_supported: indicate if idle power collaps is supported * @rc_lock: resource control mutex lock to protect @@ -160,7 +159,6 @@ enum dpu_enc_rc_states { * @rc_state: resource controller state * @delayed_off_work: delayed worker to schedule disabling of * clks and resources after IDLE_TIMEOUT time. - * @vsync_event_work: worker to handle vsync event for autorefresh * @topology: topology of the display * @idle_timeout: idle timeout duration in milliseconds * @wide_bus_en: wide bus is enabled on this interface @@ -194,7 +192,6 @@ struct dpu_encoder_virt { atomic_t frame_done_timeout_ms; struct timer_list frame_done_timer; - struct timer_list vsync_event_timer; struct msm_display_info disp_info; @@ -202,7 +199,6 @@ struct dpu_encoder_virt { struct mutex rc_lock; enum dpu_enc_rc_states rc_state; struct delayed_work delayed_off_work; - struct kthread_work vsync_event_work; struct msm_display_topology topology; u32 idle_timeout; @@ -543,11 +539,24 @@ bool dpu_encoder_use_dsc_merge(struct drm_encoder *drm_enc) return (num_dsc > 0) && (num_dsc > intf_count); } +static struct drm_dsc_config *dpu_encoder_get_dsc_config(struct drm_encoder *drm_enc) +{ + struct msm_drm_private *priv = drm_enc->dev->dev_private; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + int index = dpu_enc->disp_info.h_tile_instance[0]; + + if (dpu_enc->disp_info.intf_type == INTF_DSI) + return msm_dsi_get_dsc_config(priv->dsi[index]); + + return NULL; +} + static struct msm_display_topology dpu_encoder_get_topology( struct dpu_encoder_virt *dpu_enc, struct dpu_kms *dpu_kms, struct drm_display_mode *mode, - struct drm_crtc_state *crtc_state) + struct drm_crtc_state *crtc_state, + struct drm_dsc_config *dsc) { struct msm_display_topology topology = {0}; int i, intf_count = 0; @@ -579,7 +588,7 @@ static struct msm_display_topology dpu_encoder_get_topology( topology.num_intf = intf_count; - if (dpu_enc->dsc) { + if (dsc) { /* * In case of Display Stream Compression (DSC), we would use * 2 DSC encoders, 2 layer mixers and 1 interface @@ -605,6 +614,7 @@ static int dpu_encoder_virt_atomic_check( struct drm_display_mode *adj_mode; struct msm_display_topology topology; struct dpu_global_state *global_state; + struct drm_dsc_config *dsc; int i = 0; int ret = 0; @@ -640,7 +650,9 @@ static int dpu_encoder_virt_atomic_check( } } - topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode, crtc_state); + dsc = dpu_encoder_get_dsc_config(drm_enc); + + topology = dpu_encoder_get_topology(dpu_enc, dpu_kms, adj_mode, crtc_state, dsc); /* * Release and Allocate resources on every modeset @@ -1072,14 +1084,12 @@ static void dpu_encoder_virt_atomic_mode_set(struct drm_encoder *drm_enc, dpu_enc->hw_pp[i] = i < num_pp ? to_dpu_hw_pingpong(hw_pp[i]) : NULL; - if (dpu_enc->dsc) { - num_dsc = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, - drm_enc->base.id, DPU_HW_BLK_DSC, - hw_dsc, ARRAY_SIZE(hw_dsc)); - for (i = 0; i < num_dsc; i++) { - dpu_enc->hw_dsc[i] = to_dpu_hw_dsc(hw_dsc[i]); - dsc_mask |= BIT(dpu_enc->hw_dsc[i]->idx - DSC_0); - } + num_dsc = dpu_rm_get_assigned_resources(&dpu_kms->rm, global_state, + drm_enc->base.id, DPU_HW_BLK_DSC, + hw_dsc, ARRAY_SIZE(hw_dsc)); + for (i = 0; i < num_dsc; i++) { + dpu_enc->hw_dsc[i] = to_dpu_hw_dsc(hw_dsc[i]); + dsc_mask |= BIT(dpu_enc->hw_dsc[i]->idx - DSC_0); } dpu_enc->dsc_mask = dsc_mask; @@ -1187,6 +1197,8 @@ static void dpu_encoder_virt_atomic_enable(struct drm_encoder *drm_enc, dpu_enc = to_dpu_encoder_virt(drm_enc); + dpu_enc->dsc = dpu_encoder_get_dsc_config(drm_enc); + mutex_lock(&dpu_enc->enc_lock); cur_mode = &dpu_enc->base.crtc->state->adjusted_mode; @@ -1754,49 +1766,6 @@ int dpu_encoder_vsync_time(struct drm_encoder *drm_enc, ktime_t *wakeup_time) return 0; } -static void dpu_encoder_vsync_event_handler(struct timer_list *t) -{ - struct dpu_encoder_virt *dpu_enc = from_timer(dpu_enc, t, - vsync_event_timer); - struct drm_encoder *drm_enc = &dpu_enc->base; - struct msm_drm_private *priv; - struct msm_drm_thread *event_thread; - - if (!drm_enc->dev || !drm_enc->crtc) { - DPU_ERROR("invalid parameters\n"); - return; - } - - priv = drm_enc->dev->dev_private; - - if (drm_enc->crtc->index >= ARRAY_SIZE(priv->event_thread)) { - DPU_ERROR("invalid crtc index\n"); - return; - } - event_thread = &priv->event_thread[drm_enc->crtc->index]; - if (!event_thread) { - DPU_ERROR("event_thread not found for crtc:%d\n", - drm_enc->crtc->index); - return; - } - - del_timer(&dpu_enc->vsync_event_timer); -} - -static void dpu_encoder_vsync_event_work_handler(struct kthread_work *work) -{ - struct dpu_encoder_virt *dpu_enc = container_of(work, - struct dpu_encoder_virt, vsync_event_work); - ktime_t wakeup_time; - - if (dpu_encoder_vsync_time(&dpu_enc->base, &wakeup_time)) - return; - - trace_dpu_enc_vsync_event_work(DRMID(&dpu_enc->base), wakeup_time); - mod_timer(&dpu_enc->vsync_event_timer, - nsecs_to_jiffies(ktime_to_ns(wakeup_time))); -} - static u32 dpu_encoder_dsc_initial_line_calc(struct drm_dsc_config *dsc, u32 enc_ip_width) @@ -1956,7 +1925,6 @@ void dpu_encoder_kickoff(struct drm_encoder *drm_enc) { struct dpu_encoder_virt *dpu_enc; struct dpu_encoder_phys *phys; - ktime_t wakeup_time; unsigned long timeout_ms; unsigned int i; @@ -1982,14 +1950,6 @@ void dpu_encoder_kickoff(struct drm_encoder *drm_enc) phys->ops.handle_post_kickoff(phys); } - if (dpu_enc->disp_info.intf_type == INTF_DSI && - !dpu_encoder_vsync_time(drm_enc, &wakeup_time)) { - trace_dpu_enc_early_kickoff(DRMID(drm_enc), - ktime_to_ms(wakeup_time)); - mod_timer(&dpu_enc->vsync_event_timer, - nsecs_to_jiffies(ktime_to_ns(wakeup_time))); - } - DPU_ATRACE_END("encoder_kickoff"); } @@ -2108,8 +2068,10 @@ void dpu_encoder_helper_phys_cleanup(struct dpu_encoder_phys *phys_enc) phys_enc->hw_pp->merge_3d->idx); } - if (dpu_enc->dsc) + if (dpu_enc->dsc) { dpu_encoder_unprep_dsc(dpu_enc); + dpu_enc->dsc = NULL; + } intf_cfg.stream_sel = 0; /* Don't care value for video mode */ intf_cfg.mode_3d = dpu_encoder_helper_get_3d_blend_mode(phys_enc); @@ -2290,8 +2252,6 @@ static int dpu_encoder_setup_display(struct dpu_encoder_virt *dpu_enc, dpu_enc->idle_pc_supported = dpu_kms->catalog->caps->has_idle_pc; - dpu_enc->dsc = disp_info->dsc; - mutex_lock(&dpu_enc->enc_lock); for (i = 0; i < disp_info->num_of_h_tiles && !ret; i++) { /* @@ -2423,11 +2383,7 @@ struct drm_encoder *dpu_encoder_init(struct drm_device *dev, timer_setup(&dpu_enc->frame_done_timer, dpu_encoder_frame_done_timeout, 0); - if (disp_info->intf_type == INTF_DSI) - timer_setup(&dpu_enc->vsync_event_timer, - dpu_encoder_vsync_event_handler, - 0); - else if (disp_info->intf_type == INTF_DP) + if (disp_info->intf_type == INTF_DP) dpu_enc->wide_bus_en = msm_dp_wide_bus_available( priv->dp[disp_info->h_tile_instance[0]]); @@ -2435,9 +2391,6 @@ struct drm_encoder *dpu_encoder_init(struct drm_device *dev, dpu_encoder_off_work); dpu_enc->idle_timeout = IDLE_TIMEOUT; - kthread_init_work(&dpu_enc->vsync_event_work, - dpu_encoder_vsync_event_work_handler); - memcpy(&dpu_enc->disp_info, disp_info, sizeof(*disp_info)); DPU_DEBUG_ENC(dpu_enc, "created\n"); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h index 90e1925d7770..4c05fd5e9ed1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h @@ -28,7 +28,6 @@ * @is_cmd_mode Boolean to indicate if the CMD mode is requested * @is_te_using_watchdog_timer: Boolean to indicate watchdog TE is * used instead of panel TE in cmd mode panels - * @dsc: DSC configuration data for DSC-enabled displays */ struct msm_display_info { enum dpu_intf_type intf_type; @@ -36,7 +35,6 @@ struct msm_display_info { uint32_t h_tile_instance[MAX_H_TILES_PER_DISPLAY]; bool is_cmd_mode; bool is_te_using_watchdog_timer; - struct drm_dsc_config *dsc; }; /** diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c index b856c6286c85..df88358e7037 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -50,6 +50,7 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg( to_dpu_encoder_phys_cmd(phys_enc); struct dpu_hw_ctl *ctl; struct dpu_hw_intf_cfg intf_cfg = { 0 }; + struct dpu_hw_intf_cmd_mode_cfg cmd_mode_cfg = {}; ctl = phys_enc->hw_ctl; if (!ctl->ops.setup_intf_cfg) @@ -68,8 +69,11 @@ static void _dpu_encoder_phys_cmd_update_intf_cfg( phys_enc->hw_intf, phys_enc->hw_pp->idx); - if (intf_cfg.dsc != 0 && phys_enc->hw_intf->ops.enable_compression) - phys_enc->hw_intf->ops.enable_compression(phys_enc->hw_intf); + if (intf_cfg.dsc != 0) + cmd_mode_cfg.data_compress = true; + + if (phys_enc->hw_intf->ops.program_intf_cmd_cfg) + phys_enc->hw_intf->ops.program_intf_cmd_cfg(phys_enc->hw_intf, &cmd_mode_cfg); } static void dpu_encoder_phys_cmd_pp_tx_done_irq(void *arg, int irq_idx) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c index 662d74ded1b9..c2189e58de6a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -40,7 +40,7 @@ static bool dpu_encoder_phys_vid_is_master( static void drm_mode_to_intf_timing_params( const struct dpu_encoder_phys *phys_enc, const struct drm_display_mode *mode, - struct intf_timing_params *timing) + struct dpu_hw_intf_timing_params *timing) { memset(timing, 0, sizeof(*timing)); @@ -114,7 +114,7 @@ static void drm_mode_to_intf_timing_params( } } -static u32 get_horizontal_total(const struct intf_timing_params *timing) +static u32 get_horizontal_total(const struct dpu_hw_intf_timing_params *timing) { u32 active = timing->xres; u32 inactive = @@ -123,7 +123,7 @@ static u32 get_horizontal_total(const struct intf_timing_params *timing) return active + inactive; } -static u32 get_vertical_total(const struct intf_timing_params *timing) +static u32 get_vertical_total(const struct dpu_hw_intf_timing_params *timing) { u32 active = timing->yres; u32 inactive = @@ -148,7 +148,7 @@ static u32 get_vertical_total(const struct intf_timing_params *timing) */ static u32 programmable_fetch_get_num_lines( struct dpu_encoder_phys *phys_enc, - const struct intf_timing_params *timing) + const struct dpu_hw_intf_timing_params *timing) { u32 worst_case_needed_lines = phys_enc->hw_intf->cap->prog_fetch_lines_worst_case; @@ -196,9 +196,9 @@ static u32 programmable_fetch_get_num_lines( * @timing: Pointer to the intf timing information for the requested mode */ static void programmable_fetch_config(struct dpu_encoder_phys *phys_enc, - const struct intf_timing_params *timing) + const struct dpu_hw_intf_timing_params *timing) { - struct intf_prog_fetch f = { 0 }; + struct dpu_hw_intf_prog_fetch f = { 0 }; u32 vfp_fetch_lines = 0; u32 horiz_total = 0; u32 vert_total = 0; @@ -231,7 +231,7 @@ static void dpu_encoder_phys_vid_setup_timing_engine( struct dpu_encoder_phys *phys_enc) { struct drm_display_mode mode; - struct intf_timing_params timing_params = { 0 }; + struct dpu_hw_intf_timing_params timing_params = { 0 }; const struct dpu_format *fmt = NULL; u32 fmt_fourcc = DRM_FORMAT_RGB888; unsigned long lock_flags; @@ -522,7 +522,7 @@ static void dpu_encoder_phys_vid_disable(struct dpu_encoder_phys *phys_enc) { unsigned long lock_flags; int ret; - struct intf_status intf_status = {0}; + struct dpu_hw_intf_status intf_status = {0}; if (!phys_enc->parent || !phys_enc->parent->dev) { DPU_ERROR("invalid encoder/device\n"); @@ -651,7 +651,7 @@ static int dpu_encoder_phys_vid_get_line_count( static int dpu_encoder_phys_vid_get_frame_count( struct dpu_encoder_phys *phys_enc) { - struct intf_status s = {0}; + struct dpu_hw_intf_status s = {0}; u32 fetch_start = 0; struct drm_display_mode mode; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c index a466ff70a4d6..78037a697633 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c @@ -446,7 +446,8 @@ static int dpu_encoder_phys_wb_wait_for_commit_done( wait_info.atomic_cnt = &phys_enc->pending_kickoff_cnt; wait_info.timeout_ms = KICKOFF_TIMEOUT_MS; - ret = dpu_encoder_helper_wait_for_irq(phys_enc, INTR_IDX_WB_DONE, + ret = dpu_encoder_helper_wait_for_irq(phys_enc, + phys_enc->irq[INTR_IDX_WB_DONE], dpu_encoder_phys_wb_done_irq, &wait_info); if (ret == -ETIMEDOUT) _dpu_encoder_phys_wb_handle_wbdone_timeout(phys_enc); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c index 0de507d4d7b7..713dfc079718 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c @@ -33,6 +33,9 @@ #define VIG_SC7180_MASK \ (VIG_MASK | BIT(DPU_SSPP_QOS_8LVL) | BIT(DPU_SSPP_SCALER_QSEED4)) +#define VIG_SM6125_MASK \ + (VIG_MASK | BIT(DPU_SSPP_QOS_8LVL) | BIT(DPU_SSPP_SCALER_QSEED3LITE)) + #define VIG_SC7180_MASK_SDMA \ (VIG_SC7180_MASK | BIT(DPU_SSPP_SMART_DMA_V2)) @@ -93,19 +96,15 @@ #define CTL_SM8550_MASK \ (CTL_SC7280_MASK | BIT(DPU_CTL_HAS_LAYER_EXT4)) -#define MERGE_3D_SM8150_MASK (0) - #define DSPP_SC7180_MASK BIT(DPU_DSPP_PCC) -#define INTF_SDM845_MASK (0) - #define INTF_SC7180_MASK \ (BIT(DPU_INTF_INPUT_CTRL) | \ BIT(DPU_INTF_TE) | \ BIT(DPU_INTF_STATUS_SUPPORTED) | \ BIT(DPU_DATA_HCTL_EN)) -#define INTF_SC7280_MASK (INTF_SC7180_MASK | BIT(DPU_INTF_DATA_COMPRESS)) +#define INTF_SC7280_MASK (INTF_SC7180_MASK) #define WB_SM8250_MASK (BIT(DPU_WB_LINE_MODE) | \ BIT(DPU_WB_UBWC) | \ @@ -252,15 +251,15 @@ static const uint32_t wb2_formats[] = { *************************************************************/ /* SSPP common configuration */ -#define _VIG_SBLK(num, sdma_pri, qseed_ver) \ +#define _VIG_SBLK(sdma_pri, qseed_ver) \ { \ .maxdwnscale = MAX_DOWNSCALE_RATIO, \ .maxupscale = MAX_UPSCALE_RATIO, \ .smart_dma_priority = sdma_pri, \ - .scaler_blk = {.name = STRCAT("sspp_scaler", num), \ + .scaler_blk = {.name = "scaler", \ .id = qseed_ver, \ .base = 0xa00, .len = 0xa0,}, \ - .csc_blk = {.name = STRCAT("sspp_csc", num), \ + .csc_blk = {.name = "csc", \ .id = DPU_SSPP_CSC_10BIT, \ .base = 0x1a00, .len = 0x100,}, \ .format_list = plane_formats_yuv, \ @@ -270,15 +269,15 @@ static const uint32_t wb2_formats[] = { .rotation_cfg = NULL, \ } -#define _VIG_SBLK_ROT(num, sdma_pri, qseed_ver, rot_cfg) \ +#define _VIG_SBLK_ROT(sdma_pri, qseed_ver, rot_cfg) \ { \ .maxdwnscale = MAX_DOWNSCALE_RATIO, \ .maxupscale = MAX_UPSCALE_RATIO, \ .smart_dma_priority = sdma_pri, \ - .scaler_blk = {.name = STRCAT("sspp_scaler", num), \ + .scaler_blk = {.name = "scaler", \ .id = qseed_ver, \ .base = 0xa00, .len = 0xa0,}, \ - .csc_blk = {.name = STRCAT("sspp_csc", num), \ + .csc_blk = {.name = "csc", \ .id = DPU_SSPP_CSC_10BIT, \ .base = 0x1a00, .len = 0x100,}, \ .format_list = plane_formats_yuv, \ @@ -288,7 +287,7 @@ static const uint32_t wb2_formats[] = { .rotation_cfg = rot_cfg, \ } -#define _DMA_SBLK(num, sdma_pri) \ +#define _DMA_SBLK(sdma_pri) \ { \ .maxdwnscale = SSPP_UNITY_SCALE, \ .maxupscale = SSPP_UNITY_SCALE, \ @@ -300,13 +299,13 @@ static const uint32_t wb2_formats[] = { } static const struct dpu_sspp_sub_blks msm8998_vig_sblk_0 = - _VIG_SBLK("0", 0, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(0, DPU_SSPP_SCALER_QSEED3); static const struct dpu_sspp_sub_blks msm8998_vig_sblk_1 = - _VIG_SBLK("1", 0, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(0, DPU_SSPP_SCALER_QSEED3); static const struct dpu_sspp_sub_blks msm8998_vig_sblk_2 = - _VIG_SBLK("2", 0, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(0, DPU_SSPP_SCALER_QSEED3); static const struct dpu_sspp_sub_blks msm8998_vig_sblk_3 = - _VIG_SBLK("3", 0, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(0, DPU_SSPP_SCALER_QSEED3); static const struct dpu_rotation_cfg dpu_rot_sc7280_cfg_v2 = { .rot_maxheight = 1088, @@ -315,61 +314,52 @@ static const struct dpu_rotation_cfg dpu_rot_sc7280_cfg_v2 = { }; static const struct dpu_sspp_sub_blks sdm845_vig_sblk_0 = - _VIG_SBLK("0", 5, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(5, DPU_SSPP_SCALER_QSEED3); static const struct dpu_sspp_sub_blks sdm845_vig_sblk_1 = - _VIG_SBLK("1", 6, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(6, DPU_SSPP_SCALER_QSEED3); static const struct dpu_sspp_sub_blks sdm845_vig_sblk_2 = - _VIG_SBLK("2", 7, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(7, DPU_SSPP_SCALER_QSEED3); static const struct dpu_sspp_sub_blks sdm845_vig_sblk_3 = - _VIG_SBLK("3", 8, DPU_SSPP_SCALER_QSEED3); + _VIG_SBLK(8, DPU_SSPP_SCALER_QSEED3); -static const struct dpu_sspp_sub_blks sdm845_dma_sblk_0 = _DMA_SBLK("8", 1); -static const struct dpu_sspp_sub_blks sdm845_dma_sblk_1 = _DMA_SBLK("9", 2); -static const struct dpu_sspp_sub_blks sdm845_dma_sblk_2 = _DMA_SBLK("10", 3); -static const struct dpu_sspp_sub_blks sdm845_dma_sblk_3 = _DMA_SBLK("11", 4); - -#define SSPP_BLK(_name, _id, _base, _len, _features, \ - _sblk, _xinid, _type, _clkctrl) \ - { \ - .name = _name, .id = _id, \ - .base = _base, .len = _len, \ - .features = _features, \ - .sblk = &_sblk, \ - .xin_id = _xinid, \ - .type = _type, \ - .clk_ctrl = _clkctrl \ - } +static const struct dpu_sspp_sub_blks sdm845_dma_sblk_0 = _DMA_SBLK(1); +static const struct dpu_sspp_sub_blks sdm845_dma_sblk_1 = _DMA_SBLK(2); +static const struct dpu_sspp_sub_blks sdm845_dma_sblk_2 = _DMA_SBLK(3); +static const struct dpu_sspp_sub_blks sdm845_dma_sblk_3 = _DMA_SBLK(4); static const struct dpu_sspp_sub_blks sc7180_vig_sblk_0 = - _VIG_SBLK("0", 4, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(4, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sc7280_vig_sblk_0 = - _VIG_SBLK_ROT("0", 4, DPU_SSPP_SCALER_QSEED4, &dpu_rot_sc7280_cfg_v2); + _VIG_SBLK_ROT(4, DPU_SSPP_SCALER_QSEED4, &dpu_rot_sc7280_cfg_v2); static const struct dpu_sspp_sub_blks sm6115_vig_sblk_0 = - _VIG_SBLK("0", 2, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(2, DPU_SSPP_SCALER_QSEED4); + +static const struct dpu_sspp_sub_blks sm6125_vig_sblk_0 = + _VIG_SBLK(3, DPU_SSPP_SCALER_QSEED3LITE); static const struct dpu_sspp_sub_blks sm8250_vig_sblk_0 = - _VIG_SBLK("0", 5, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(5, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8250_vig_sblk_1 = - _VIG_SBLK("1", 6, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(6, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8250_vig_sblk_2 = - _VIG_SBLK("2", 7, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(7, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8250_vig_sblk_3 = - _VIG_SBLK("3", 8, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(8, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8550_vig_sblk_0 = - _VIG_SBLK("0", 7, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(7, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8550_vig_sblk_1 = - _VIG_SBLK("1", 8, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(8, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8550_vig_sblk_2 = - _VIG_SBLK("2", 9, DPU_SSPP_SCALER_QSEED4); + _VIG_SBLK(9, DPU_SSPP_SCALER_QSEED4); static const struct dpu_sspp_sub_blks sm8550_vig_sblk_3 = - _VIG_SBLK("3", 10, DPU_SSPP_SCALER_QSEED4); -static const struct dpu_sspp_sub_blks sm8550_dma_sblk_4 = _DMA_SBLK("12", 5); -static const struct dpu_sspp_sub_blks sm8550_dma_sblk_5 = _DMA_SBLK("13", 6); + _VIG_SBLK(10, DPU_SSPP_SCALER_QSEED4); +static const struct dpu_sspp_sub_blks sm8550_dma_sblk_4 = _DMA_SBLK(5); +static const struct dpu_sspp_sub_blks sm8550_dma_sblk_5 = _DMA_SBLK(6); -#define _VIG_SBLK_NOSCALE(num, sdma_pri) \ +#define _VIG_SBLK_NOSCALE(sdma_pri) \ { \ .maxdwnscale = SSPP_UNITY_SCALE, \ .maxupscale = SSPP_UNITY_SCALE, \ @@ -380,24 +370,13 @@ static const struct dpu_sspp_sub_blks sm8550_dma_sblk_5 = _DMA_SBLK("13", 6); .virt_num_formats = ARRAY_SIZE(plane_formats), \ } -static const struct dpu_sspp_sub_blks qcm2290_vig_sblk_0 = _VIG_SBLK_NOSCALE("0", 2); -static const struct dpu_sspp_sub_blks qcm2290_dma_sblk_0 = _DMA_SBLK("8", 1); +static const struct dpu_sspp_sub_blks qcm2290_vig_sblk_0 = _VIG_SBLK_NOSCALE(2); +static const struct dpu_sspp_sub_blks qcm2290_dma_sblk_0 = _DMA_SBLK(1); /************************************************************* * MIXER sub blocks config *************************************************************/ -#define LM_BLK(_name, _id, _base, _fmask, _sblk, _pp, _lmpair, _dspp) \ - { \ - .name = _name, .id = _id, \ - .base = _base, .len = 0x320, \ - .features = _fmask, \ - .sblk = _sblk, \ - .pingpong = _pp, \ - .lm_pair_mask = (1 << _lmpair), \ - .dspp = _dspp \ - } - /* MSM8998 */ static const struct dpu_lm_sub_blks msm8998_lm_sblk = { @@ -444,151 +423,48 @@ static const struct dpu_lm_sub_blks qcm2290_lm_sblk = { * DSPP sub blocks config *************************************************************/ static const struct dpu_dspp_sub_blks msm8998_dspp_sblk = { - .pcc = {.id = DPU_DSPP_PCC, .base = 0x1700, + .pcc = {.name = "pcc", .id = DPU_DSPP_PCC, .base = 0x1700, .len = 0x90, .version = 0x10007}, }; -static const struct dpu_dspp_sub_blks sm8150_dspp_sblk = { - .pcc = {.id = DPU_DSPP_PCC, .base = 0x1700, +static const struct dpu_dspp_sub_blks sdm845_dspp_sblk = { + .pcc = {.name = "pcc", .id = DPU_DSPP_PCC, .base = 0x1700, .len = 0x90, .version = 0x40000}, }; -#define DSPP_BLK(_name, _id, _base, _mask, _sblk) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = 0x1800, \ - .features = _mask, \ - .sblk = _sblk \ - } - /************************************************************* * PINGPONG sub blocks config *************************************************************/ static const struct dpu_pingpong_sub_blks sdm845_pp_sblk_te = { - .te2 = {.id = DPU_PINGPONG_TE2, .base = 0x2000, .len = 0x0, + .te2 = {.name = "te2", .id = DPU_PINGPONG_TE2, .base = 0x2000, .len = 0x0, .version = 0x1}, - .dither = {.id = DPU_PINGPONG_DITHER, .base = 0x30e0, + .dither = {.name = "dither", .id = DPU_PINGPONG_DITHER, .base = 0x30e0, .len = 0x20, .version = 0x10000}, }; static const struct dpu_pingpong_sub_blks sdm845_pp_sblk = { - .dither = {.id = DPU_PINGPONG_DITHER, .base = 0x30e0, + .dither = {.name = "dither", .id = DPU_PINGPONG_DITHER, .base = 0x30e0, .len = 0x20, .version = 0x10000}, }; static const struct dpu_pingpong_sub_blks sc7280_pp_sblk = { - .dither = {.id = DPU_PINGPONG_DITHER, .base = 0xe0, + .dither = {.name = "dither", .id = DPU_PINGPONG_DITHER, .base = 0xe0, .len = 0x20, .version = 0x20000}, }; -#define PP_BLK_DITHER(_name, _id, _base, _merge_3d, _sblk, _done, _rdptr) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = 0, \ - .features = BIT(DPU_PINGPONG_DITHER), \ - .merge_3d = _merge_3d, \ - .sblk = &_sblk, \ - .intr_done = _done, \ - .intr_rdptr = _rdptr, \ - } -#define PP_BLK(_name, _id, _base, _features, _merge_3d, _sblk, _done, _rdptr) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = 0xd4, \ - .features = _features, \ - .merge_3d = _merge_3d, \ - .sblk = &_sblk, \ - .intr_done = _done, \ - .intr_rdptr = _rdptr, \ - } - -/************************************************************* - * MERGE_3D sub blocks config - *************************************************************/ -#define MERGE_3D_BLK(_name, _id, _base) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = 0x8, \ - .features = MERGE_3D_SM8150_MASK, \ - .sblk = NULL \ - } - /************************************************************* * DSC sub blocks config *************************************************************/ static const struct dpu_dsc_sub_blks dsc_sblk_0 = { - .enc = {.base = 0x100, .len = 0x100}, - .ctl = {.base = 0xF00, .len = 0x10}, + .enc = {.name = "enc", .base = 0x100, .len = 0x9c}, + .ctl = {.name = "ctl", .base = 0xF00, .len = 0x10}, }; static const struct dpu_dsc_sub_blks dsc_sblk_1 = { - .enc = {.base = 0x200, .len = 0x100}, - .ctl = {.base = 0xF80, .len = 0x10}, + .enc = {.name = "enc", .base = 0x200, .len = 0x9c}, + .ctl = {.name = "ctl", .base = 0xF80, .len = 0x10}, }; -#define DSC_BLK(_name, _id, _base, _features) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = 0x140, \ - .features = _features, \ - } - -#define DSC_BLK_1_2(_name, _id, _base, _len, _features, _sblk) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = _len, \ - .features = BIT(DPU_DSC_HW_REV_1_2) | _features, \ - .sblk = &_sblk, \ - } - -/************************************************************* - * INTF sub blocks config - *************************************************************/ -#define INTF_BLK(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _underrun, _vsync) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = _len, \ - .features = _features, \ - .type = _type, \ - .controller_id = _ctrl_id, \ - .prog_fetch_lines_worst_case = _progfetch, \ - .intr_underrun = _underrun, \ - .intr_vsync = _vsync, \ - .intr_tear_rd_ptr = -1, \ - } - -/* DSI Interface sub-block with TEAR registers (since DPU 5.0.0) */ -#define INTF_BLK_DSI_TE(_name, _id, _base, _len, _type, _ctrl_id, _progfetch, _features, _underrun, _vsync, _tear_rd_ptr) \ - {\ - .name = _name, .id = _id, \ - .base = _base, .len = _len, \ - .features = _features, \ - .type = _type, \ - .controller_id = _ctrl_id, \ - .prog_fetch_lines_worst_case = _progfetch, \ - .intr_underrun = _underrun, \ - .intr_vsync = _vsync, \ - .intr_tear_rd_ptr = _tear_rd_ptr, \ - } - -/************************************************************* - * Writeback blocks config - *************************************************************/ -#define WB_BLK(_name, _id, _base, _features, _clk_ctrl, \ - __xin_id, vbif_id, _reg, _max_linewidth, _wb_done_bit) \ - { \ - .name = _name, .id = _id, \ - .base = _base, .len = 0x2c8, \ - .features = _features, \ - .format_list = wb2_formats, \ - .num_formats = ARRAY_SIZE(wb2_formats), \ - .clk_ctrl = _clk_ctrl, \ - .xin_id = __xin_id, \ - .vbif_idx = vbif_id, \ - .maxlinewidth = _max_linewidth, \ - .intr_wb_done = DPU_IRQ_IDX(_reg, _wb_done_bit) \ - } - /************************************************************* * VBIF sub blocks config *************************************************************/ @@ -663,6 +539,26 @@ static const struct dpu_vbif_cfg sdm845_vbif[] = { }, }; +static const struct dpu_vbif_cfg sm8550_vbif[] = { + { + .name = "vbif_rt", .id = VBIF_RT, + .base = 0, .len = 0x1040, + .features = BIT(DPU_VBIF_QOS_REMAP), + .xin_halt_timeout = 0x4000, + .qos_rp_remap_size = 0x40, + .qos_rt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_rt_pri_lvl), + .priority_lvl = sdm845_rt_pri_lvl, + }, + .qos_nrt_tbl = { + .npriority_lvl = ARRAY_SIZE(sdm845_nrt_pri_lvl), + .priority_lvl = sdm845_nrt_pri_lvl, + }, + .memtype_count = 16, + .memtype = {3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3}, + }, +}; + /************************************************************* * PERF data config *************************************************************/ @@ -762,6 +658,7 @@ static const struct dpu_qos_lut_entry sc7180_qos_nrt[] = { #include "catalog/dpu_5_0_sm8150.h" #include "catalog/dpu_5_1_sc8180x.h" +#include "catalog/dpu_5_4_sm6125.h" #include "catalog/dpu_6_0_sm8250.h" #include "catalog/dpu_6_2_sc7180.h" diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h index b860784ade72..6c9634209e9f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h @@ -29,23 +29,9 @@ #define MAX_XIN_COUNT 16 /** - * Supported UBWC feature versions - */ -enum { - DPU_HW_UBWC_VER_10 = 0x100, - DPU_HW_UBWC_VER_20 = 0x200, - DPU_HW_UBWC_VER_30 = 0x300, - DPU_HW_UBWC_VER_40 = 0x400, -}; - -/** * MDP TOP BLOCK features * @DPU_MDP_PANIC_PER_PIPE Panic configuration needs to be done per pipe * @DPU_MDP_10BIT_SUPPORT, Chipset supports 10 bit pixel formats - * @DPU_MDP_BWC, MDSS HW supports Bandwidth compression. - * @DPU_MDP_UBWC_1_0, This chipsets supports Universal Bandwidth - * compression initial revision - * @DPU_MDP_UBWC_1_5, Universal Bandwidth compression version 1.5 * @DPU_MDP_PERIPH_0_REMOVED Indicates that access to periph top0 block results * in a failure * @DPU_MDP_VSYNC_SEL Enables vsync source selection via MDP_VSYNC_SEL register @@ -56,9 +42,6 @@ enum { enum { DPU_MDP_PANIC_PER_PIPE = 0x1, DPU_MDP_10BIT_SUPPORT, - DPU_MDP_BWC, - DPU_MDP_UBWC_1_0, - DPU_MDP_UBWC_1_5, DPU_MDP_AUDIO_SELECT, DPU_MDP_PERIPH_0_REMOVED, DPU_MDP_VSYNC_SEL, @@ -181,7 +164,6 @@ enum { * @DPU_DATA_HCTL_EN Allows data to be transferred at different rate * than video timing * @DPU_INTF_STATUS_SUPPORTED INTF block has INTF_STATUS register - * @DPU_INTF_DATA_COMPRESS INTF block has DATA_COMPRESS register * @DPU_INTF_MAX */ enum { @@ -189,7 +171,6 @@ enum { DPU_INTF_TE, DPU_DATA_HCTL_EN, DPU_INTF_STATUS_SUPPORTED, - DPU_INTF_DATA_COMPRESS, DPU_INTF_MAX }; @@ -505,19 +486,6 @@ struct dpu_mdp_cfg { struct dpu_clk_ctrl_reg clk_ctrls[DPU_CLK_CTRL_MAX]; }; -/** - * struct dpu_ubwc_cfg - UBWC and memory configuration - * - * @ubwc_version UBWC feature version (0x0 for not supported) - * @highest_bank_bit: UBWC parameter - * @ubwc_swizzle: ubwc default swizzle setting - */ -struct dpu_ubwc_cfg { - u32 ubwc_version; - u32 highest_bank_bit; - u32 ubwc_swizzle; -}; - /* struct dpu_ctl_cfg : MDP CTL instance info * @id: index identifying this block * @base: register base offset to mdss @@ -554,14 +522,14 @@ struct dpu_sspp_cfg { * @features bit mask identifying sub-blocks/features * @sblk: LM Sub-blocks information * @pingpong: ID of connected PingPong, PINGPONG_NONE if unsupported - * @lm_pair_mask: Bitmask of LMs that can be controlled by same CTL + * @lm_pair: ID of LM that can be controlled by same CTL */ struct dpu_lm_cfg { DPU_HW_BLK_INFO; const struct dpu_lm_sub_blks *sblk; u32 pingpong; u32 dspp; - unsigned long lm_pair_mask; + unsigned long lm_pair; }; /** @@ -747,6 +715,16 @@ struct dpu_perf_cdp_cfg { }; /** + * struct dpu_mdss_version - DPU's major and minor versions + * @core_major_ver: DPU core's major version + * @core_minor_ver: DPU core's minor version + */ +struct dpu_mdss_version { + u8 core_major_ver; + u8 core_minor_ver; +}; + +/** * struct dpu_perf_cfg - performance control settings * @max_bw_low low threshold of maximum bandwidth (kbps) * @max_bw_high high threshold of maximum bandwidth (kbps) @@ -796,20 +774,19 @@ struct dpu_perf_cfg { /** * struct dpu_mdss_cfg - information of MDSS HW * This is the main catalog data structure representing - * this HW version. Contains number of instances, - * register offsets, capabilities of the all MDSS HW sub-blocks. + * this HW version. Contains dpu's major and minor versions, + * number of instances, register offsets, capabilities of the + * all MDSS HW sub-blocks. * * @dma_formats Supported formats for dma pipe * @cursor_formats Supported formats for cursor pipe * @vig_formats Supported formats for vig pipe - * @mdss_irqs: Bitmap with the irqs supported by the target */ struct dpu_mdss_cfg { - const struct dpu_caps *caps; + const struct dpu_mdss_version *mdss_ver; - const struct dpu_ubwc_cfg *ubwc; + const struct dpu_caps *caps; - u32 mdp_count; const struct dpu_mdp_cfg *mdp; u32 ctl_count; @@ -850,8 +827,6 @@ struct dpu_mdss_cfg { const struct dpu_format_extended *dma_formats; const struct dpu_format_extended *cursor_formats; const struct dpu_format_extended *vig_formats; - - unsigned long mdss_irqs; }; extern const struct dpu_mdss_cfg dpu_msm8998_cfg; @@ -861,6 +836,7 @@ extern const struct dpu_mdss_cfg dpu_sc8180x_cfg; extern const struct dpu_mdss_cfg dpu_sm8250_cfg; extern const struct dpu_mdss_cfg dpu_sc7180_cfg; extern const struct dpu_mdss_cfg dpu_sm6115_cfg; +extern const struct dpu_mdss_cfg dpu_sm6125_cfg; extern const struct dpu_mdss_cfg dpu_sm6350_cfg; extern const struct dpu_mdss_cfg dpu_qcm2290_cfg; extern const struct dpu_mdss_cfg dpu_sm6375_cfg; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c index c278fb9d2b5b..86182c734606 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c @@ -51,7 +51,7 @@ static const u32 fetch_tbl[SSPP_MAX] = {CTL_INVALID_BIT, 16, 17, 18, 19, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, CTL_INVALID_BIT, 0, - 1, 2, 3, CTL_INVALID_BIT, CTL_INVALID_BIT}; + 1, 2, 3, 4, 5}; static int _mixer_stages(const struct dpu_lm_cfg *mixer, int count, enum dpu_lm lm) @@ -198,6 +198,12 @@ static void dpu_hw_ctl_update_pending_flush_sspp(struct dpu_hw_ctl *ctx, case SSPP_DMA3: ctx->pending_flush_mask |= BIT(25); break; + case SSPP_DMA4: + ctx->pending_flush_mask |= BIT(13); + break; + case SSPP_DMA5: + ctx->pending_flush_mask |= BIT(14); + break; case SSPP_CURSOR0: ctx->pending_flush_mask |= BIT(22); break; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c index 5e2d68ebb113..e3c50439f80a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.c @@ -51,11 +51,9 @@ struct dpu_intr_reg { }; /* - * struct dpu_intr_reg - List of DPU interrupt registers - * - * When making changes be sure to sync with dpu_hw_intr_reg + * dpu_intr_set_legacy - List of DPU interrupt registers for DPU <= 6.x */ -static const struct dpu_intr_reg dpu_intr_set[] = { +static const struct dpu_intr_reg dpu_intr_set_legacy[] = { [MDP_SSPP_TOP0_INTR] = { INTR_CLEAR, INTR_EN, @@ -121,57 +119,78 @@ static const struct dpu_intr_reg dpu_intr_set[] = { MDP_AD4_INTR_EN_OFF(1), MDP_AD4_INTR_STATUS_OFF(1), }, - [MDP_INTF0_7xxx_INTR] = { +}; + +/* + * dpu_intr_set_7xxx - List of DPU interrupt registers for DPU >= 7.0 + */ +static const struct dpu_intr_reg dpu_intr_set_7xxx[] = { + [MDP_SSPP_TOP0_INTR] = { + INTR_CLEAR, + INTR_EN, + INTR_STATUS + }, + [MDP_SSPP_TOP0_INTR2] = { + INTR2_CLEAR, + INTR2_EN, + INTR2_STATUS + }, + [MDP_SSPP_TOP0_HIST_INTR] = { + HIST_INTR_CLEAR, + HIST_INTR_EN, + HIST_INTR_STATUS + }, + [MDP_INTF0_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(0), MDP_INTF_REV_7xxx_INTR_EN(0), MDP_INTF_REV_7xxx_INTR_STATUS(0) }, - [MDP_INTF1_7xxx_INTR] = { + [MDP_INTF1_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(1), MDP_INTF_REV_7xxx_INTR_EN(1), MDP_INTF_REV_7xxx_INTR_STATUS(1) }, - [MDP_INTF1_7xxx_TEAR_INTR] = { + [MDP_INTF1_TEAR_INTR] = { MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(1), MDP_INTF_REV_7xxx_INTR_TEAR_EN(1), MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(1) }, - [MDP_INTF2_7xxx_INTR] = { + [MDP_INTF2_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(2), MDP_INTF_REV_7xxx_INTR_EN(2), MDP_INTF_REV_7xxx_INTR_STATUS(2) }, - [MDP_INTF2_7xxx_TEAR_INTR] = { + [MDP_INTF2_TEAR_INTR] = { MDP_INTF_REV_7xxx_INTR_TEAR_CLEAR(2), MDP_INTF_REV_7xxx_INTR_TEAR_EN(2), MDP_INTF_REV_7xxx_INTR_TEAR_STATUS(2) }, - [MDP_INTF3_7xxx_INTR] = { + [MDP_INTF3_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(3), MDP_INTF_REV_7xxx_INTR_EN(3), MDP_INTF_REV_7xxx_INTR_STATUS(3) }, - [MDP_INTF4_7xxx_INTR] = { + [MDP_INTF4_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(4), MDP_INTF_REV_7xxx_INTR_EN(4), MDP_INTF_REV_7xxx_INTR_STATUS(4) }, - [MDP_INTF5_7xxx_INTR] = { + [MDP_INTF5_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(5), MDP_INTF_REV_7xxx_INTR_EN(5), MDP_INTF_REV_7xxx_INTR_STATUS(5) }, - [MDP_INTF6_7xxx_INTR] = { + [MDP_INTF6_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(6), MDP_INTF_REV_7xxx_INTR_EN(6), MDP_INTF_REV_7xxx_INTR_STATUS(6) }, - [MDP_INTF7_7xxx_INTR] = { + [MDP_INTF7_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(7), MDP_INTF_REV_7xxx_INTR_EN(7), MDP_INTF_REV_7xxx_INTR_STATUS(7) }, - [MDP_INTF8_7xxx_INTR] = { + [MDP_INTF8_INTR] = { MDP_INTF_REV_7xxx_INTR_CLEAR(8), MDP_INTF_REV_7xxx_INTR_EN(8), MDP_INTF_REV_7xxx_INTR_STATUS(8) @@ -216,19 +235,19 @@ irqreturn_t dpu_core_irq(struct msm_kms *kms) return IRQ_NONE; spin_lock_irqsave(&intr->irq_lock, irq_flags); - for (reg_idx = 0; reg_idx < ARRAY_SIZE(dpu_intr_set); reg_idx++) { + for (reg_idx = 0; reg_idx < MDP_INTR_MAX; reg_idx++) { if (!test_bit(reg_idx, &intr->irq_mask)) continue; /* Read interrupt status */ - irq_status = DPU_REG_READ(&intr->hw, dpu_intr_set[reg_idx].status_off); + irq_status = DPU_REG_READ(&intr->hw, intr->intr_set[reg_idx].status_off); /* Read enable mask */ - enable_mask = DPU_REG_READ(&intr->hw, dpu_intr_set[reg_idx].en_off); + enable_mask = DPU_REG_READ(&intr->hw, intr->intr_set[reg_idx].en_off); /* and clear the interrupt */ if (irq_status) - DPU_REG_WRITE(&intr->hw, dpu_intr_set[reg_idx].clr_off, + DPU_REG_WRITE(&intr->hw, intr->intr_set[reg_idx].clr_off, irq_status); /* Finally update IRQ status based on enable mask */ @@ -285,7 +304,11 @@ static int dpu_hw_intr_enable_irq_locked(struct dpu_hw_intr *intr, int irq_idx) assert_spin_locked(&intr->irq_lock); reg_idx = DPU_IRQ_REG(irq_idx); - reg = &dpu_intr_set[reg_idx]; + reg = &intr->intr_set[reg_idx]; + + /* Is this interrupt register supported on the platform */ + if (WARN_ON(!reg->en_off)) + return -EINVAL; cache_irq_mask = intr->cache_irq_mask[reg_idx]; if (cache_irq_mask & DPU_IRQ_MASK(irq_idx)) { @@ -334,7 +357,7 @@ static int dpu_hw_intr_disable_irq_locked(struct dpu_hw_intr *intr, int irq_idx) assert_spin_locked(&intr->irq_lock); reg_idx = DPU_IRQ_REG(irq_idx); - reg = &dpu_intr_set[reg_idx]; + reg = &intr->intr_set[reg_idx]; cache_irq_mask = intr->cache_irq_mask[reg_idx]; if ((cache_irq_mask & DPU_IRQ_MASK(irq_idx)) == 0) { @@ -368,10 +391,10 @@ static void dpu_clear_irqs(struct dpu_kms *dpu_kms) if (!intr) return; - for (i = 0; i < ARRAY_SIZE(dpu_intr_set); i++) { + for (i = 0; i < MDP_INTR_MAX; i++) { if (test_bit(i, &intr->irq_mask)) DPU_REG_WRITE(&intr->hw, - dpu_intr_set[i].clr_off, 0xffffffff); + intr->intr_set[i].clr_off, 0xffffffff); } /* ensure register writes go through */ @@ -386,10 +409,10 @@ static void dpu_disable_all_irqs(struct dpu_kms *dpu_kms) if (!intr) return; - for (i = 0; i < ARRAY_SIZE(dpu_intr_set); i++) { + for (i = 0; i < MDP_INTR_MAX; i++) { if (test_bit(i, &intr->irq_mask)) DPU_REG_WRITE(&intr->hw, - dpu_intr_set[i].en_off, 0x00000000); + intr->intr_set[i].en_off, 0x00000000); } /* ensure register writes go through */ @@ -421,10 +444,10 @@ u32 dpu_core_irq_read(struct dpu_kms *dpu_kms, int irq_idx) reg_idx = DPU_IRQ_REG(irq_idx); intr_status = DPU_REG_READ(&intr->hw, - dpu_intr_set[reg_idx].status_off) & + intr->intr_set[reg_idx].status_off) & DPU_IRQ_MASK(irq_idx); if (intr_status) - DPU_REG_WRITE(&intr->hw, dpu_intr_set[reg_idx].clr_off, + DPU_REG_WRITE(&intr->hw, intr->intr_set[reg_idx].clr_off, intr_status); /* ensure register writes go through */ @@ -435,17 +458,12 @@ u32 dpu_core_irq_read(struct dpu_kms *dpu_kms, int irq_idx) return intr_status; } -static void __intr_offset(const struct dpu_mdss_cfg *m, - void __iomem *addr, struct dpu_hw_blk_reg_map *hw) -{ - hw->blk_addr = addr + m->mdp[0].base; -} - struct dpu_hw_intr *dpu_hw_intr_init(void __iomem *addr, const struct dpu_mdss_cfg *m) { struct dpu_hw_intr *intr; int nirq = MDP_INTR_MAX * 32; + unsigned int i; if (!addr || !m) return ERR_PTR(-EINVAL); @@ -454,11 +472,29 @@ struct dpu_hw_intr *dpu_hw_intr_init(void __iomem *addr, if (!intr) return ERR_PTR(-ENOMEM); - __intr_offset(m, addr, &intr->hw); + if (m->mdss_ver->core_major_ver >= 7) + intr->intr_set = dpu_intr_set_7xxx; + else + intr->intr_set = dpu_intr_set_legacy; + + intr->hw.blk_addr = addr + m->mdp[0].base; intr->total_irqs = nirq; - intr->irq_mask = m->mdss_irqs; + intr->irq_mask = BIT(MDP_SSPP_TOP0_INTR) | + BIT(MDP_SSPP_TOP0_INTR2) | + BIT(MDP_SSPP_TOP0_HIST_INTR); + for (i = 0; i < m->intf_count; i++) { + const struct dpu_intf_cfg *intf = &m->intf[i]; + + if (intf->type == INTF_NONE) + continue; + + intr->irq_mask |= BIT(MDP_INTFn_INTR(intf->id)); + + if (intf->intr_tear_rd_ptr != -1) + intr->irq_mask |= BIT(DPU_IRQ_REG(intf->intr_tear_rd_ptr)); + } spin_lock_init(&intr->irq_lock); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h index 1f2dabc54c22..dab761e54863 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_interrupts.h @@ -17,30 +17,25 @@ enum dpu_hw_intr_reg { MDP_SSPP_TOP0_INTR, MDP_SSPP_TOP0_INTR2, MDP_SSPP_TOP0_HIST_INTR, + /* All MDP_INTFn_INTR should come sequentially */ MDP_INTF0_INTR, MDP_INTF1_INTR, MDP_INTF2_INTR, MDP_INTF3_INTR, MDP_INTF4_INTR, MDP_INTF5_INTR, + MDP_INTF6_INTR, + MDP_INTF7_INTR, + MDP_INTF8_INTR, MDP_INTF1_TEAR_INTR, MDP_INTF2_TEAR_INTR, MDP_AD4_0_INTR, MDP_AD4_1_INTR, - MDP_INTF0_7xxx_INTR, - MDP_INTF1_7xxx_INTR, - MDP_INTF1_7xxx_TEAR_INTR, - MDP_INTF2_7xxx_INTR, - MDP_INTF2_7xxx_TEAR_INTR, - MDP_INTF3_7xxx_INTR, - MDP_INTF4_7xxx_INTR, - MDP_INTF5_7xxx_INTR, - MDP_INTF6_7xxx_INTR, - MDP_INTF7_7xxx_INTR, - MDP_INTF8_7xxx_INTR, MDP_INTR_MAX, }; +#define MDP_INTFn_INTR(intf) (MDP_INTF0_INTR + (intf - INTF_0)) + #define DPU_IRQ_IDX(reg_idx, offset) (reg_idx * 32 + offset) /** @@ -60,6 +55,7 @@ struct dpu_hw_intr { u32 total_irqs; spinlock_t irq_lock; unsigned long irq_mask; + const struct dpu_intr_reg *intr_set; struct { void (*cb)(void *arg, int irq_idx); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c index 5b0f6627e29b..8ec6505d9e78 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c @@ -95,7 +95,7 @@ static void dpu_hw_intf_setup_timing_engine(struct dpu_hw_intf *ctx, - const struct intf_timing_params *p, + const struct dpu_hw_intf_timing_params *p, const struct dpu_format *fmt) { struct dpu_hw_blk_reg_map *c = &ctx->hw; @@ -244,7 +244,7 @@ static void dpu_hw_intf_enable_timing_engine( static void dpu_hw_intf_setup_prg_fetch( struct dpu_hw_intf *intf, - const struct intf_prog_fetch *fetch) + const struct dpu_hw_intf_prog_fetch *fetch) { struct dpu_hw_blk_reg_map *c = &intf->hw; int fetch_enable; @@ -286,7 +286,7 @@ static void dpu_hw_intf_bind_pingpong_blk( static void dpu_hw_intf_get_status( struct dpu_hw_intf *intf, - struct intf_status *s) + struct dpu_hw_intf_status *s) { struct dpu_hw_blk_reg_map *c = &intf->hw; unsigned long cap = intf->cap->features; @@ -513,17 +513,19 @@ static void dpu_hw_intf_disable_autorefresh(struct dpu_hw_intf *intf, } -static void dpu_hw_intf_enable_compression(struct dpu_hw_intf *ctx) +static void dpu_hw_intf_program_intf_cmd_cfg(struct dpu_hw_intf *ctx, + struct dpu_hw_intf_cmd_mode_cfg *cmd_mode_cfg) { u32 intf_cfg2 = DPU_REG_READ(&ctx->hw, INTF_CONFIG2); - intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS; + if (cmd_mode_cfg->data_compress) + intf_cfg2 |= INTF_CFG2_DCE_DATA_COMPRESS; DPU_REG_WRITE(&ctx->hw, INTF_CONFIG2, intf_cfg2); } static void _setup_intf_ops(struct dpu_hw_intf_ops *ops, - unsigned long cap) + unsigned long cap, const struct dpu_mdss_version *mdss_rev) { ops->setup_timing_gen = dpu_hw_intf_setup_timing_engine; ops->setup_prg_fetch = dpu_hw_intf_setup_prg_fetch; @@ -543,12 +545,12 @@ static void _setup_intf_ops(struct dpu_hw_intf_ops *ops, ops->disable_autorefresh = dpu_hw_intf_disable_autorefresh; } - if (cap & BIT(DPU_INTF_DATA_COMPRESS)) - ops->enable_compression = dpu_hw_intf_enable_compression; + if (mdss_rev->core_major_ver >= 7) + ops->program_intf_cmd_cfg = dpu_hw_intf_program_intf_cmd_cfg; } struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg, - void __iomem *addr) + void __iomem *addr, const struct dpu_mdss_version *mdss_rev) { struct dpu_hw_intf *c; @@ -569,7 +571,7 @@ struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg, */ c->idx = cfg->id; c->cap = cfg; - _setup_intf_ops(&c->ops, c->cap->features); + _setup_intf_ops(&c->ops, c->cap->features, mdss_rev); return c; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h index 99e21c4137f9..77f80531782b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -14,7 +14,7 @@ struct dpu_hw_intf; /* intf timing settings */ -struct intf_timing_params { +struct dpu_hw_intf_timing_params { u32 width; /* active width */ u32 height; /* active height */ u32 xres; /* Display panel width */ @@ -35,19 +35,23 @@ struct intf_timing_params { bool wide_bus_en; }; -struct intf_prog_fetch { +struct dpu_hw_intf_prog_fetch { u8 enable; /* vsync counter for the front porch pixel line */ u32 fetch_start; }; -struct intf_status { +struct dpu_hw_intf_status { u8 is_en; /* interface timing engine is enabled or not */ u8 is_prog_fetch_en; /* interface prog fetch counter is enabled or not */ u32 frame_count; /* frame count since timing engine enabled */ u32 line_count; /* current line count including blanking */ }; +struct dpu_hw_intf_cmd_mode_cfg { + u8 data_compress; /* enable data compress between dpu and dsi */ +}; + /** * struct dpu_hw_intf_ops : Interface to the interface Hw driver functions * Assumption is these functions will be called after clocks are enabled @@ -70,21 +74,21 @@ struct intf_status { * @get_autorefresh: Retrieve autorefresh config from hardware * Return: 0 on success, -ETIMEDOUT on timeout * @vsync_sel: Select vsync signal for tear-effect configuration - * @enable_compression: Enable data compression + * @program_intf_cmd_cfg: Program the DPU to interface datapath for command mode */ struct dpu_hw_intf_ops { void (*setup_timing_gen)(struct dpu_hw_intf *intf, - const struct intf_timing_params *p, + const struct dpu_hw_intf_timing_params *p, const struct dpu_format *fmt); void (*setup_prg_fetch)(struct dpu_hw_intf *intf, - const struct intf_prog_fetch *fetch); + const struct dpu_hw_intf_prog_fetch *fetch); void (*enable_timing)(struct dpu_hw_intf *intf, u8 enable); void (*get_status)(struct dpu_hw_intf *intf, - struct intf_status *status); + struct dpu_hw_intf_status *status); u32 (*get_line_count)(struct dpu_hw_intf *intf); @@ -108,7 +112,8 @@ struct dpu_hw_intf_ops { */ void (*disable_autorefresh)(struct dpu_hw_intf *intf, uint32_t encoder_id, u16 vdisplay); - void (*enable_compression)(struct dpu_hw_intf *intf); + void (*program_intf_cmd_cfg)(struct dpu_hw_intf *intf, + struct dpu_hw_intf_cmd_mode_cfg *cmd_mode_cfg); }; struct dpu_hw_intf { @@ -127,9 +132,10 @@ struct dpu_hw_intf { * interface catalog entry. * @cfg: interface catalog entry for which driver object is required * @addr: mapped register io address of MDP + * @mdss_rev: dpu core's major and minor versions */ struct dpu_hw_intf *dpu_hw_intf_init(const struct dpu_intf_cfg *cfg, - void __iomem *addr); + void __iomem *addr, const struct dpu_mdss_version *mdss_rev); /** * dpu_hw_intf_destroy(): Destroys INTF driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h index 02a0f48aac94..d85157acfbf8 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_mdss.h @@ -101,11 +101,6 @@ enum dpu_hw_blk_type { DPU_HW_BLK_MAX, }; -enum dpu_mdp { - MDP_TOP = 0x1, - MDP_MAX, -}; - enum dpu_sspp { SSPP_NONE, SSPP_VIG0, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c index b364cf75bb3f..f2192de93713 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c @@ -8,6 +8,8 @@ #include "dpu_hw_sspp.h" #include "dpu_kms.h" +#include "msm_mdss.h" + #include <drm/drm_file.h> #define DPU_FETCH_CONFIG_RESET_VALUE 0x00000087 @@ -270,26 +272,26 @@ static void dpu_hw_sspp_setup_format(struct dpu_sw_pipe *pipe, DPU_REG_WRITE(c, SSPP_FETCH_CONFIG, DPU_FETCH_CONFIG_RESET_VALUE | ctx->ubwc->highest_bank_bit << 18); - switch (ctx->ubwc->ubwc_version) { - case DPU_HW_UBWC_VER_10: + switch (ctx->ubwc->ubwc_enc_version) { + case UBWC_1_0: fast_clear = fmt->alpha_enable ? BIT(31) : 0; DPU_REG_WRITE(c, SSPP_UBWC_STATIC_CTRL, fast_clear | (ctx->ubwc->ubwc_swizzle & 0x1) | BIT(8) | (ctx->ubwc->highest_bank_bit << 4)); break; - case DPU_HW_UBWC_VER_20: + case UBWC_2_0: fast_clear = fmt->alpha_enable ? BIT(31) : 0; DPU_REG_WRITE(c, SSPP_UBWC_STATIC_CTRL, fast_clear | (ctx->ubwc->ubwc_swizzle) | (ctx->ubwc->highest_bank_bit << 4)); break; - case DPU_HW_UBWC_VER_30: + case UBWC_3_0: DPU_REG_WRITE(c, SSPP_UBWC_STATIC_CTRL, BIT(30) | (ctx->ubwc->ubwc_swizzle) | (ctx->ubwc->highest_bank_bit << 4)); break; - case DPU_HW_UBWC_VER_40: + case UBWC_4_0: DPU_REG_WRITE(c, SSPP_UBWC_STATIC_CTRL, DPU_FORMAT_IS_YUV(fmt) ? 0 : BIT(30)); break; @@ -670,11 +672,11 @@ int _dpu_hw_sspp_init_debugfs(struct dpu_hw_sspp *hw_pipe, struct dpu_kms *kms, #endif struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg, - void __iomem *addr, const struct dpu_ubwc_cfg *ubwc) + void __iomem *addr, const struct msm_mdss_data *mdss_data) { struct dpu_hw_sspp *hw_pipe; - if (!addr || !ubwc) + if (!addr) return ERR_PTR(-EINVAL); hw_pipe = kzalloc(sizeof(*hw_pipe), GFP_KERNEL); @@ -685,7 +687,7 @@ struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg, hw_pipe->hw.log_mask = DPU_DBG_MASK_SSPP; /* Assign ops */ - hw_pipe->ubwc = ubwc; + hw_pipe->ubwc = mdss_data; hw_pipe->idx = cfg->id; hw_pipe->cap = cfg; _setup_layer_ops(hw_pipe, hw_pipe->cap->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h index 085f34bc6b88..cbf4f95ff0fd 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h @@ -317,7 +317,7 @@ struct dpu_hw_sspp_ops { struct dpu_hw_sspp { struct dpu_hw_blk base; struct dpu_hw_blk_reg_map hw; - const struct dpu_ubwc_cfg *ubwc; + const struct msm_mdss_data *ubwc; /* Pipe */ enum dpu_sspp idx; @@ -333,10 +333,10 @@ struct dpu_kms; * Should be called once before accessing every pipe. * @cfg: Pipe catalog entry for which driver object is required * @addr: Mapped register io address of MDP - * @ubwc: UBWC configuration data + * @mdss_data: UBWC / MDSS configuration data */ struct dpu_hw_sspp *dpu_hw_sspp_init(const struct dpu_sspp_cfg *cfg, - void __iomem *addr, const struct dpu_ubwc_cfg *ubwc); + void __iomem *addr, const struct msm_mdss_data *mdss_data); /** * dpu_hw_sspp_destroy(): Destroys SSPP driver context diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c index 963bdb5e0252..cff48763ce25 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c @@ -268,51 +268,25 @@ static void _setup_mdp_ops(struct dpu_hw_mdp_ops *ops, ops->intf_audio_select = dpu_hw_intf_audio_select; } -static const struct dpu_mdp_cfg *_top_offset(enum dpu_mdp mdp, - const struct dpu_mdss_cfg *m, - void __iomem *addr, - struct dpu_hw_blk_reg_map *b) -{ - int i; - - if (!m || !addr || !b) - return ERR_PTR(-EINVAL); - - for (i = 0; i < m->mdp_count; i++) { - if (mdp == m->mdp[i].id) { - b->blk_addr = addr + m->mdp[i].base; - b->log_mask = DPU_DBG_MASK_TOP; - return &m->mdp[i]; - } - } - - return ERR_PTR(-EINVAL); -} - -struct dpu_hw_mdp *dpu_hw_mdptop_init(enum dpu_mdp idx, +struct dpu_hw_mdp *dpu_hw_mdptop_init(const struct dpu_mdp_cfg *cfg, void __iomem *addr, const struct dpu_mdss_cfg *m) { struct dpu_hw_mdp *mdp; - const struct dpu_mdp_cfg *cfg; - if (!addr || !m) + if (!addr) return ERR_PTR(-EINVAL); mdp = kzalloc(sizeof(*mdp), GFP_KERNEL); if (!mdp) return ERR_PTR(-ENOMEM); - cfg = _top_offset(idx, m, addr, &mdp->hw); - if (IS_ERR_OR_NULL(cfg)) { - kfree(mdp); - return ERR_PTR(-EINVAL); - } + mdp->hw.blk_addr = addr + cfg->base; + mdp->hw.log_mask = DPU_DBG_MASK_TOP; /* * Assign ops */ - mdp->idx = idx; mdp->caps = cfg; _setup_mdp_ops(&mdp->ops, mdp->caps->features); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h index a1a9e44bed36..8b1463d2b2f0 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h @@ -137,7 +137,6 @@ struct dpu_hw_mdp { struct dpu_hw_blk_reg_map hw; /* top */ - enum dpu_mdp idx; const struct dpu_mdp_cfg *caps; /* ops */ @@ -145,12 +144,12 @@ struct dpu_hw_mdp { }; /** - * dpu_hw_mdptop_init - initializes the top driver for the passed idx - * @idx: Interface index for which driver object is required + * dpu_hw_mdptop_init - initializes the top driver for the passed config + * @cfg: MDP TOP configuration from catalog * @addr: Mapped register io address of MDP * @m: Pointer to mdss catalog data */ -struct dpu_hw_mdp *dpu_hw_mdptop_init(enum dpu_mdp idx, +struct dpu_hw_mdp *dpu_hw_mdptop_init(const struct dpu_mdp_cfg *cfg, void __iomem *addr, const struct dpu_mdss_cfg *m); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index a16ae7db6245..82381d12414d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -22,6 +22,7 @@ #include "msm_drv.h" #include "msm_mmu.h" +#include "msm_mdss.h" #include "msm_gem.h" #include "disp/msm_disp_snapshot.h" @@ -544,8 +545,6 @@ static int _dpu_kms_initialize_dsi(struct drm_device *dev, info.is_cmd_mode = msm_dsi_is_cmd_mode(priv->dsi[i]); - info.dsc = msm_dsi_get_dsc_config(priv->dsi[i]); - encoder = dpu_encoder_init(dev, DRM_MODE_ENCODER_DSI, &info); if (IS_ERR(encoder)) { DPU_ERROR("encoder init failed for dsi display\n"); @@ -794,7 +793,7 @@ static int _dpu_kms_drm_obj_init(struct dpu_kms *dpu_kms) ret = PTR_ERR(crtc); return ret; } - priv->crtcs[priv->num_crtcs++] = crtc; + priv->num_crtcs++; } /* All CRTCs are compatible with all encoders */ @@ -890,6 +889,7 @@ static void dpu_kms_mdp_snapshot(struct msm_disp_state *disp_state, struct msm_k int i; struct dpu_kms *dpu_kms; const struct dpu_mdss_cfg *cat; + void __iomem *base; dpu_kms = to_dpu_kms(kms); @@ -900,37 +900,67 @@ static void dpu_kms_mdp_snapshot(struct msm_disp_state *disp_state, struct msm_k /* dump CTL sub-blocks HW regs info */ for (i = 0; i < cat->ctl_count; i++) msm_disp_snapshot_add_block(disp_state, cat->ctl[i].len, - dpu_kms->mmio + cat->ctl[i].base, "ctl_%d", i); + dpu_kms->mmio + cat->ctl[i].base, cat->ctl[i].name); /* dump DSPP sub-blocks HW regs info */ - for (i = 0; i < cat->dspp_count; i++) - msm_disp_snapshot_add_block(disp_state, cat->dspp[i].len, - dpu_kms->mmio + cat->dspp[i].base, "dspp_%d", i); + for (i = 0; i < cat->dspp_count; i++) { + base = dpu_kms->mmio + cat->dspp[i].base; + msm_disp_snapshot_add_block(disp_state, cat->dspp[i].len, base, cat->dspp[i].name); + + if (cat->dspp[i].sblk && cat->dspp[i].sblk->pcc.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->dspp[i].sblk->pcc.len, + base + cat->dspp[i].sblk->pcc.base, "%s_%s", + cat->dspp[i].name, + cat->dspp[i].sblk->pcc.name); + } /* dump INTF sub-blocks HW regs info */ for (i = 0; i < cat->intf_count; i++) msm_disp_snapshot_add_block(disp_state, cat->intf[i].len, - dpu_kms->mmio + cat->intf[i].base, "intf_%d", i); + dpu_kms->mmio + cat->intf[i].base, cat->intf[i].name); /* dump PP sub-blocks HW regs info */ - for (i = 0; i < cat->pingpong_count; i++) - msm_disp_snapshot_add_block(disp_state, cat->pingpong[i].len, - dpu_kms->mmio + cat->pingpong[i].base, "pingpong_%d", i); + for (i = 0; i < cat->pingpong_count; i++) { + base = dpu_kms->mmio + cat->pingpong[i].base; + msm_disp_snapshot_add_block(disp_state, cat->pingpong[i].len, base, + cat->pingpong[i].name); + + /* TE2 sub-block has length of 0, so will not print it */ + + if (cat->pingpong[i].sblk && cat->pingpong[i].sblk->dither.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->pingpong[i].sblk->dither.len, + base + cat->pingpong[i].sblk->dither.base, + "%s_%s", cat->pingpong[i].name, + cat->pingpong[i].sblk->dither.name); + } /* dump SSPP sub-blocks HW regs info */ - for (i = 0; i < cat->sspp_count; i++) - msm_disp_snapshot_add_block(disp_state, cat->sspp[i].len, - dpu_kms->mmio + cat->sspp[i].base, "sspp_%d", i); + for (i = 0; i < cat->sspp_count; i++) { + base = dpu_kms->mmio + cat->sspp[i].base; + msm_disp_snapshot_add_block(disp_state, cat->sspp[i].len, base, cat->sspp[i].name); + + if (cat->sspp[i].sblk && cat->sspp[i].sblk->scaler_blk.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->sspp[i].sblk->scaler_blk.len, + base + cat->sspp[i].sblk->scaler_blk.base, + "%s_%s", cat->sspp[i].name, + cat->sspp[i].sblk->scaler_blk.name); + + if (cat->sspp[i].sblk && cat->sspp[i].sblk->csc_blk.len > 0) + msm_disp_snapshot_add_block(disp_state, cat->sspp[i].sblk->csc_blk.len, + base + cat->sspp[i].sblk->csc_blk.base, + "%s_%s", cat->sspp[i].name, + cat->sspp[i].sblk->csc_blk.name); + } /* dump LM sub-blocks HW regs info */ for (i = 0; i < cat->mixer_count; i++) msm_disp_snapshot_add_block(disp_state, cat->mixer[i].len, - dpu_kms->mmio + cat->mixer[i].base, "lm_%d", i); + dpu_kms->mmio + cat->mixer[i].base, cat->mixer[i].name); /* dump WB sub-blocks HW regs info */ for (i = 0; i < cat->wb_count; i++) msm_disp_snapshot_add_block(disp_state, cat->wb[i].len, - dpu_kms->mmio + cat->wb[i].base, "wb_%d", i); + dpu_kms->mmio + cat->wb[i].base, cat->wb[i].name); if (cat->mdp[0].features & BIT(DPU_MDP_PERIPH_0_REMOVED)) { msm_disp_snapshot_add_block(disp_state, MDP_PERIPH_TOP0, @@ -943,9 +973,20 @@ static void dpu_kms_mdp_snapshot(struct msm_disp_state *disp_state, struct msm_k } /* dump DSC sub-blocks HW regs info */ - for (i = 0; i < cat->dsc_count; i++) - msm_disp_snapshot_add_block(disp_state, cat->dsc[i].len, - dpu_kms->mmio + cat->dsc[i].base, "dsc_%d", i); + for (i = 0; i < cat->dsc_count; i++) { + base = dpu_kms->mmio + cat->dsc[i].base; + msm_disp_snapshot_add_block(disp_state, cat->dsc[i].len, base, cat->dsc[i].name); + + if (cat->dsc[i].features & BIT(DPU_DSC_HW_REV_1_2)) { + struct dpu_dsc_blk enc = cat->dsc[i].sblk->enc; + struct dpu_dsc_blk ctl = cat->dsc[i].sblk->ctl; + + msm_disp_snapshot_add_block(disp_state, enc.len, base + enc.base, "%s_%s", + cat->dsc[i].name, enc.name); + msm_disp_snapshot_add_block(disp_state, ctl.len, base + ctl.base, "%s_%s", + cat->dsc[i].name, ctl.name); + } + } pm_runtime_put_sync(&dpu_kms->pdev->dev); } @@ -1011,11 +1052,14 @@ unsigned long dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name) return clk_get_rate(clk); } +#define DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE 412500000 + static int dpu_kms_hw_init(struct msm_kms *kms) { struct dpu_kms *dpu_kms; struct drm_device *dev; int i, rc = -EINVAL; + unsigned long max_core_clk_rate; u32 core_rev; if (!kms) { @@ -1084,7 +1128,20 @@ static int dpu_kms_hw_init(struct msm_kms *kms) goto power_error; } - rc = dpu_rm_init(&dpu_kms->rm, dpu_kms->catalog, dpu_kms->mmio); + dpu_kms->mdss = msm_mdss_get_mdss_data(dpu_kms->pdev->dev.parent); + if (IS_ERR(dpu_kms->mdss)) { + rc = PTR_ERR(dpu_kms->mdss); + DPU_ERROR("failed to get MDSS data: %d\n", rc); + goto power_error; + } + + if (!dpu_kms->mdss) { + rc = -EINVAL; + DPU_ERROR("NULL MDSS data\n"); + goto power_error; + } + + rc = dpu_rm_init(&dpu_kms->rm, dpu_kms->catalog, dpu_kms->mdss, dpu_kms->mmio); if (rc) { DPU_ERROR("rm init failed: %d\n", rc); goto power_error; @@ -1092,7 +1149,8 @@ static int dpu_kms_hw_init(struct msm_kms *kms) dpu_kms->rm_init = true; - dpu_kms->hw_mdp = dpu_hw_mdptop_init(MDP_TOP, dpu_kms->mmio, + dpu_kms->hw_mdp = dpu_hw_mdptop_init(dpu_kms->catalog->mdp, + dpu_kms->mmio, dpu_kms->catalog); if (IS_ERR(dpu_kms->hw_mdp)) { rc = PTR_ERR(dpu_kms->hw_mdp); @@ -1115,8 +1173,14 @@ static int dpu_kms_hw_init(struct msm_kms *kms) dpu_kms->hw_vbif[vbif->id] = hw; } - rc = dpu_core_perf_init(&dpu_kms->perf, dev, dpu_kms->catalog, - msm_clk_bulk_get_clock(dpu_kms->clocks, dpu_kms->num_clocks, "core")); + /* TODO: use the same max_freq as in dpu_kms_hw_init */ + max_core_clk_rate = dpu_kms_get_clk_rate(dpu_kms, "core"); + if (!max_core_clk_rate) { + DPU_DEBUG("max core clk rate not determined, using default\n"); + max_core_clk_rate = DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE; + } + + rc = dpu_core_perf_init(&dpu_kms->perf, dpu_kms->catalog->perf, max_core_clk_rate); if (rc) { DPU_ERROR("failed to init perf %d\n", rc); goto perf_err; @@ -1162,7 +1226,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms) return 0; drm_obj_init_err: - dpu_core_perf_destroy(&dpu_kms->perf); hw_intr_init_err: perf_err: power_error: @@ -1303,6 +1366,7 @@ static const struct of_device_id dpu_dt_match[] = { { .compatible = "qcom,sc8180x-dpu", .data = &dpu_sc8180x_cfg, }, { .compatible = "qcom,sc8280xp-dpu", .data = &dpu_sc8280xp_cfg, }, { .compatible = "qcom,sm6115-dpu", .data = &dpu_sm6115_cfg, }, + { .compatible = "qcom,sm6125-dpu", .data = &dpu_sm6125_cfg, }, { .compatible = "qcom,sm6350-dpu", .data = &dpu_sm6350_cfg, }, { .compatible = "qcom,sm6375-dpu", .data = &dpu_sm6375_cfg, }, { .compatible = "qcom,sm8150-dpu", .data = &dpu_sm8150_cfg, }, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index f3bdd4f11108..b6f53ca6e962 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -67,6 +67,7 @@ struct dpu_kms { struct msm_kms base; struct drm_device *dev; const struct dpu_mdss_cfg *catalog; + const struct msm_mdss_data *mdss; /* io/register spaces: */ void __iomem *mmio, *vbif[VBIF_MAX]; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c index 471842bbb950..f9215643c71a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c @@ -101,6 +101,7 @@ int dpu_rm_destroy(struct dpu_rm *rm) int dpu_rm_init(struct dpu_rm *rm, const struct dpu_mdss_cfg *cat, + const struct msm_mdss_data *mdss_data, void __iomem *mmio) { int rc, i; @@ -161,7 +162,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_intf *hw; const struct dpu_intf_cfg *intf = &cat->intf[i]; - hw = dpu_hw_intf_init(intf, mmio); + hw = dpu_hw_intf_init(intf, mmio, cat->mdss_ver); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed intf object creation: err %d\n", rc); @@ -230,7 +231,7 @@ int dpu_rm_init(struct dpu_rm *rm, struct dpu_hw_sspp *hw; const struct dpu_sspp_cfg *sspp = &cat->sspp[i]; - hw = dpu_hw_sspp_init(sspp, mmio, cat->ubwc); + hw = dpu_hw_sspp_init(sspp, mmio, mdss_data); if (IS_ERR(hw)) { rc = PTR_ERR(hw); DPU_ERROR("failed sspp object creation: err %d\n", rc); @@ -253,28 +254,19 @@ static bool _dpu_rm_needs_split_display(const struct msm_display_topology *top) } /** - * _dpu_rm_check_lm_peer - check if a mixer is a peer of the primary + * _dpu_rm_get_lm_peer - get the id of a mixer which is a peer of the primary * @rm: dpu resource manager handle * @primary_idx: index of primary mixer in rm->mixer_blks[] - * @peer_idx: index of other mixer in rm->mixer_blks[] - * Return: true if rm->mixer_blks[peer_idx] is a peer of - * rm->mixer_blks[primary_idx] */ -static bool _dpu_rm_check_lm_peer(struct dpu_rm *rm, int primary_idx, - int peer_idx) +static int _dpu_rm_get_lm_peer(struct dpu_rm *rm, int primary_idx) { const struct dpu_lm_cfg *prim_lm_cfg; - const struct dpu_lm_cfg *peer_cfg; prim_lm_cfg = to_dpu_hw_mixer(rm->mixer_blks[primary_idx])->cap; - peer_cfg = to_dpu_hw_mixer(rm->mixer_blks[peer_idx])->cap; - if (!test_bit(peer_cfg->id, &prim_lm_cfg->lm_pair_mask)) { - DPU_DEBUG("lm %d not peer of lm %d\n", peer_cfg->id, - peer_cfg->id); - return false; - } - return true; + if (prim_lm_cfg->lm_pair >= LM_0 && prim_lm_cfg->lm_pair < LM_MAX) + return prim_lm_cfg->lm_pair - LM_0; + return -EINVAL; } /** @@ -351,7 +343,7 @@ static int _dpu_rm_reserve_lms(struct dpu_rm *rm, int lm_idx[MAX_BLOCKS]; int pp_idx[MAX_BLOCKS]; int dspp_idx[MAX_BLOCKS] = {0}; - int i, j, lm_count = 0; + int i, lm_count = 0; if (!reqs->topology.num_lm) { DPU_ERROR("invalid number of lm: %d\n", reqs->topology.num_lm); @@ -376,16 +368,15 @@ static int _dpu_rm_reserve_lms(struct dpu_rm *rm, ++lm_count; /* Valid primary mixer found, find matching peers */ - for (j = i + 1; j < ARRAY_SIZE(rm->mixer_blks) && - lm_count < reqs->topology.num_lm; j++) { - if (!rm->mixer_blks[j]) + if (lm_count < reqs->topology.num_lm) { + int j = _dpu_rm_get_lm_peer(rm, i); + + /* ignore the peer if there is an error or if the peer was already processed */ + if (j < 0 || j < i) continue; - if (!_dpu_rm_check_lm_peer(rm, i, j)) { - DPU_DEBUG("lm %d not peer of lm %d\n", LM_0 + j, - LM_0 + i); + if (!rm->mixer_blks[j]) continue; - } if (!_dpu_rm_check_lm_and_get_connected_blks(rm, global_state, enc_id, j, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h index d62c2edb2460..2b551566cbf4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h @@ -40,11 +40,13 @@ struct dpu_rm { * for all HW blocks. * @rm: DPU Resource Manager handle * @cat: Pointer to hardware catalog + * @mdss_data: Pointer to MDSS / UBWC configuration * @mmio: mapped register io address of MDP * @Return: 0 on Success otherwise -ERROR */ int dpu_rm_init(struct dpu_rm *rm, const struct dpu_mdss_cfg *cat, + const struct msm_mdss_data *mdss_data, void __iomem *mmio); /** diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h index 1a92d21094f4..c74b9be25e68 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h @@ -453,29 +453,6 @@ TRACE_EVENT(dpu_enc_trigger_flush, __entry->extra_flush_bits, __entry->pending_flush_ret) ); -DECLARE_EVENT_CLASS(dpu_enc_ktime_template, - TP_PROTO(uint32_t drm_id, ktime_t time), - TP_ARGS(drm_id, time), - TP_STRUCT__entry( - __field( uint32_t, drm_id ) - __field( ktime_t, time ) - ), - TP_fast_assign( - __entry->drm_id = drm_id; - __entry->time = time; - ), - TP_printk("id=%u, time=%lld", __entry->drm_id, - ktime_to_ms(__entry->time)) -); -DEFINE_EVENT(dpu_enc_ktime_template, dpu_enc_vsync_event_work, - TP_PROTO(uint32_t drm_id, ktime_t time), - TP_ARGS(drm_id, time) -); -DEFINE_EVENT(dpu_enc_ktime_template, dpu_enc_early_kickoff, - TP_PROTO(uint32_t drm_id, ktime_t time), - TP_ARGS(drm_id, time) -); - DECLARE_EVENT_CLASS(dpu_id_event_template, TP_PROTO(uint32_t drm_id, u32 event), TP_ARGS(drm_id, event), diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c index 3c4258d1784b..e5012fa6771f 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c @@ -332,7 +332,7 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) goto fail; } - priv->crtcs[priv->num_crtcs++] = crtc; + priv->num_crtcs++; } /* @@ -417,7 +417,6 @@ static int mdp4_kms_init(struct drm_device *dev) irq = platform_get_irq(pdev, 0); if (irq < 0) { ret = irq; - DRM_DEV_ERROR(dev->dev, "failed to get irq: %d\n", ret); goto fail; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c index e8aebad5f208..8a7b44376bc6 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -497,7 +497,7 @@ static int modeset_init(struct mdp5_kms *mdp5_kms) DRM_DEV_ERROR(dev->dev, "failed to construct crtc %d (%d)\n", i, ret); goto fail; } - priv->crtcs[priv->num_crtcs++] = crtc; + priv->num_crtcs++; } /* diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c index bd2c4ac45601..0d5ff03cb091 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c @@ -130,8 +130,7 @@ static void mdp5_plane_destroy_state(struct drm_plane *plane, { struct mdp5_plane_state *pstate = to_mdp5_plane_state(state); - if (state->fb) - drm_framebuffer_put(state->fb); + __drm_atomic_helper_plane_destroy_state(state); kfree(pstate); } diff --git a/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c b/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c index acfe1b31e079..add72bbc28b1 100644 --- a/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c +++ b/drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c @@ -192,5 +192,5 @@ void msm_disp_snapshot_add_block(struct msm_disp_state *disp_state, u32 len, new_blk->base_addr = base_addr; msm_disp_state_dump_regs(&new_blk->state, new_blk->size, base_addr); - list_add(&new_blk->node, &disp_state->blocks); + list_add_tail(&new_blk->node, &disp_state->blocks); } diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c index 8a5fb6df7210..1f98ff74ceb0 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c +++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c @@ -160,6 +160,7 @@ static const char * const dsi_v2_4_clk_names[] = { static const struct regulator_bulk_data dsi_v2_4_regulators[] = { { .supply = "vdda", .init_load_uA = 21800 }, /* 1.2 V */ + { .supply = "refgen" }, }; static const struct msm_dsi_config sdm845_dsi_cfg = { @@ -191,6 +192,7 @@ static const struct msm_dsi_config sm8550_dsi_cfg = { static const struct regulator_bulk_data sc7280_dsi_regulators[] = { { .supply = "vdda", .init_load_uA = 8350 }, /* 1.2 V */ + { .supply = "refgen" }, }; static const struct msm_dsi_config sc7280_dsi_cfg = { diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 4c6d73e24bb5..5d9ec27c89d3 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -752,6 +752,13 @@ static void dsi_ctrl_enable(struct msm_dsi_host *msm_host, /* Always insert DCS command */ data |= DSI_CMD_CFG1_INSERT_DCS_COMMAND; dsi_write(msm_host, REG_DSI_CMD_CFG1, data); + + if (msm_host->cfg_hnd->major == MSM_DSI_VER_MAJOR_6G && + msm_host->cfg_hnd->minor >= MSM_DSI_6G_VER_MINOR_V1_3) { + data = dsi_read(msm_host, REG_DSI_CMD_MODE_MDP_CTRL2); + data |= DSI_CMD_MODE_MDP_CTRL2_BURST_MODE; + dsi_write(msm_host, REG_DSI_CMD_MODE_MDP_CTRL2, data); + } } dsi_write(msm_host, REG_DSI_CMD_DMA_CTRL, diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index 9d5795c58a98..05621e5e7d63 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -561,6 +561,8 @@ static const struct of_device_id dsi_phy_dt_match[] = { .data = &dsi_phy_14nm_660_cfgs }, { .compatible = "qcom,dsi-phy-14nm-8953", .data = &dsi_phy_14nm_8953_cfgs }, + { .compatible = "qcom,sm6125-dsi-phy-14nm", + .data = &dsi_phy_14nm_2290_cfgs }, #endif #ifdef CONFIG_DRM_MSM_DSI_10NM_PHY { .compatible = "qcom,dsi-phy-10nm", diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c index 3ce45b023e63..31deda1c664a 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c @@ -1087,8 +1087,6 @@ const struct msm_dsi_phy_cfg dsi_phy_14nm_8953_cfgs = { const struct msm_dsi_phy_cfg dsi_phy_14nm_2290_cfgs = { .has_phy_lane = true, - .regulator_data = dsi_phy_14nm_17mA_regulators, - .num_regulators = ARRAY_SIZE(dsi_phy_14nm_17mA_regulators), .ops = { .enable = dsi_14nm_phy_enable, .disable = dsi_14nm_phy_disable, diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 551b0bc28095..a428951ee539 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -155,7 +155,7 @@ static void msm_irq_uninstall(struct drm_device *dev) struct msm_vblank_work { struct work_struct work; - int crtc_id; + struct drm_crtc *crtc; bool enable; struct msm_drm_private *priv; }; @@ -168,15 +168,15 @@ static void vblank_ctrl_worker(struct work_struct *work) struct msm_kms *kms = priv->kms; if (vbl_work->enable) - kms->funcs->enable_vblank(kms, priv->crtcs[vbl_work->crtc_id]); + kms->funcs->enable_vblank(kms, vbl_work->crtc); else - kms->funcs->disable_vblank(kms, priv->crtcs[vbl_work->crtc_id]); + kms->funcs->disable_vblank(kms, vbl_work->crtc); kfree(vbl_work); } static int vblank_ctrl_queue_work(struct msm_drm_private *priv, - int crtc_id, bool enable) + struct drm_crtc *crtc, bool enable) { struct msm_vblank_work *vbl_work; @@ -186,7 +186,7 @@ static int vblank_ctrl_queue_work(struct msm_drm_private *priv, INIT_WORK(&vbl_work->work, vblank_ctrl_worker); - vbl_work->crtc_id = crtc_id; + vbl_work->crtc = crtc; vbl_work->enable = enable; vbl_work->priv = priv; @@ -407,7 +407,8 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) struct msm_drm_private *priv = dev_get_drvdata(dev); struct drm_device *ddev; struct msm_kms *kms; - int ret, i; + struct drm_crtc *crtc; + int ret; if (drm_firmware_drivers_only()) return -ENODEV; @@ -494,20 +495,21 @@ static int msm_drm_init(struct device *dev, const struct drm_driver *drv) ddev->mode_config.funcs = &mode_config_funcs; ddev->mode_config.helper_private = &mode_config_helper_funcs; - for (i = 0; i < priv->num_crtcs; i++) { + drm_for_each_crtc(crtc, ddev) { + struct msm_drm_thread *ev_thread; + /* initialize event thread */ - priv->event_thread[i].crtc_id = priv->crtcs[i]->base.id; - priv->event_thread[i].dev = ddev; - priv->event_thread[i].worker = kthread_create_worker(0, - "crtc_event:%d", priv->event_thread[i].crtc_id); - if (IS_ERR(priv->event_thread[i].worker)) { - ret = PTR_ERR(priv->event_thread[i].worker); + ev_thread = &priv->event_thread[drm_crtc_index(crtc)]; + ev_thread->dev = ddev; + ev_thread->worker = kthread_create_worker(0, "crtc_event:%d", crtc->base.id); + if (IS_ERR(ev_thread->worker)) { + ret = PTR_ERR(ev_thread->worker); DRM_DEV_ERROR(dev, "failed to create crtc_event kthread\n"); - priv->event_thread[i].worker = NULL; + ev_thread->worker = NULL; goto err_msm_uninit; } - sched_set_fifo(priv->event_thread[i].worker->task); + sched_set_fifo(ev_thread->worker->task); } ret = drm_vblank_init(ddev, priv->num_crtcs); @@ -639,25 +641,23 @@ static void msm_postclose(struct drm_device *dev, struct drm_file *file) int msm_crtc_enable_vblank(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; - unsigned int pipe = crtc->index; struct msm_drm_private *priv = dev->dev_private; struct msm_kms *kms = priv->kms; if (!kms) return -ENXIO; - drm_dbg_vbl(dev, "crtc=%u", pipe); - return vblank_ctrl_queue_work(priv, pipe, true); + drm_dbg_vbl(dev, "crtc=%u", crtc->base.id); + return vblank_ctrl_queue_work(priv, crtc, true); } void msm_crtc_disable_vblank(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; - unsigned int pipe = crtc->index; struct msm_drm_private *priv = dev->dev_private; struct msm_kms *kms = priv->kms; if (!kms) return; - drm_dbg_vbl(dev, "crtc=%u", pipe); - vblank_ctrl_queue_work(priv, pipe, false); + drm_dbg_vbl(dev, "crtc=%u", crtc->base.id); + vblank_ctrl_queue_work(priv, crtc, false); } /* diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 44c9e06f2dff..02fd6c7d0bb7 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -65,6 +65,12 @@ enum msm_dp_controller { MSM_DP_CONTROLLER_COUNT, }; +enum msm_dsi_controller { + MSM_DSI_CONTROLLER_0, + MSM_DSI_CONTROLLER_1, + MSM_DSI_CONTROLLER_COUNT, +}; + #define MSM_GPU_MAX_RINGS 4 #define MAX_H_TILES_PER_DISPLAY 2 @@ -97,7 +103,6 @@ struct msm_display_topology { /* Commit/Event thread specific structure */ struct msm_drm_thread { struct drm_device *dev; - unsigned int crtc_id; struct kthread_worker *worker; }; @@ -117,7 +122,7 @@ struct msm_drm_private { struct hdmi *hdmi; /* DSI is shared by mdp4 and mdp5 */ - struct msm_dsi *dsi[2]; + struct msm_dsi *dsi[MSM_DSI_CONTROLLER_COUNT]; struct msm_dp *dp[MSM_DP_CONTROLLER_COUNT]; @@ -198,7 +203,6 @@ struct msm_drm_private { struct workqueue_struct *wq; unsigned int num_crtcs; - struct drm_crtc *crtcs[MAX_CRTCS]; struct msm_drm_thread event_thread[MAX_CRTCS]; diff --git a/drivers/gpu/drm/msm/msm_fence.c b/drivers/gpu/drm/msm/msm_fence.c index 96599ec3eb78..1a5d4f1c8b42 100644 --- a/drivers/gpu/drm/msm/msm_fence.c +++ b/drivers/gpu/drm/msm/msm_fence.c @@ -191,6 +191,12 @@ msm_fence_init(struct dma_fence *fence, struct msm_fence_context *fctx) f->fctx = fctx; + /* + * Until this point, the fence was just some pre-allocated memory, + * no-one should have taken a reference to it yet. + */ + WARN_ON(kref_read(&fence->refcount)); + dma_fence_init(&f->base, &msm_fence_ops, &fctx->spinlock, fctx->context, ++fctx->last_fence); } diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index 635744bc4765..db1e748daa75 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -222,9 +222,7 @@ static void put_pages(struct drm_gem_object *obj) static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj, unsigned madv) { - struct msm_drm_private *priv = obj->dev->dev_private; struct msm_gem_object *msm_obj = to_msm_bo(obj); - struct page **p; msm_gem_assert_locked(obj); @@ -234,16 +232,29 @@ static struct page **msm_gem_pin_pages_locked(struct drm_gem_object *obj, return ERR_PTR(-EBUSY); } - p = get_pages(obj); - if (IS_ERR(p)) - return p; + return get_pages(obj); +} + +/* + * Update the pin count of the object, call under lru.lock + */ +void msm_gem_pin_obj_locked(struct drm_gem_object *obj) +{ + struct msm_drm_private *priv = obj->dev->dev_private; + + msm_gem_assert_locked(obj); + + to_msm_bo(obj)->pin_count++; + drm_gem_lru_move_tail_locked(&priv->lru.pinned, obj); +} + +static void pin_obj_locked(struct drm_gem_object *obj) +{ + struct msm_drm_private *priv = obj->dev->dev_private; mutex_lock(&priv->lru.lock); - msm_obj->pin_count++; - update_lru_locked(obj); + msm_gem_pin_obj_locked(obj); mutex_unlock(&priv->lru.lock); - - return p; } struct page **msm_gem_pin_pages(struct drm_gem_object *obj) @@ -252,6 +263,8 @@ struct page **msm_gem_pin_pages(struct drm_gem_object *obj) msm_gem_lock(obj); p = msm_gem_pin_pages_locked(obj, MSM_MADV_WILLNEED); + if (!IS_ERR(p)) + pin_obj_locked(obj); msm_gem_unlock(obj); return p; @@ -463,7 +476,7 @@ int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma) { struct msm_gem_object *msm_obj = to_msm_bo(obj); struct page **pages; - int ret, prot = IOMMU_READ; + int prot = IOMMU_READ; if (!(msm_obj->flags & MSM_BO_GPU_READONLY)) prot |= IOMMU_WRITE; @@ -480,11 +493,7 @@ int msm_gem_pin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vma) if (IS_ERR(pages)) return PTR_ERR(pages); - ret = msm_gem_vma_map(vma, prot, msm_obj->sgt, obj->size); - if (ret) - msm_gem_unpin_locked(obj); - - return ret; + return msm_gem_vma_map(vma, prot, msm_obj->sgt, obj->size); } void msm_gem_unpin_locked(struct drm_gem_object *obj) @@ -509,14 +518,11 @@ void msm_gem_unpin_locked(struct drm_gem_object *obj) */ void msm_gem_unpin_active(struct drm_gem_object *obj) { - struct msm_drm_private *priv = obj->dev->dev_private; struct msm_gem_object *msm_obj = to_msm_bo(obj); - mutex_lock(&priv->lru.lock); msm_obj->pin_count--; GEM_WARN_ON(msm_obj->pin_count < 0); update_lru_active(obj); - mutex_unlock(&priv->lru.lock); } struct msm_gem_vma *msm_gem_get_vma_locked(struct drm_gem_object *obj, @@ -539,8 +545,10 @@ static int get_and_pin_iova_range_locked(struct drm_gem_object *obj, return PTR_ERR(vma); ret = msm_gem_pin_vma_locked(obj, vma); - if (!ret) + if (!ret) { *iova = vma->iova; + pin_obj_locked(obj); + } return ret; } @@ -599,9 +607,6 @@ static int clear_iova(struct drm_gem_object *obj, if (!vma) return 0; - if (msm_gem_vma_inuse(vma)) - return -EBUSY; - msm_gem_vma_purge(vma); msm_gem_vma_close(vma); del_vma(vma); @@ -652,7 +657,6 @@ void msm_gem_unpin_iova(struct drm_gem_object *obj, msm_gem_lock(obj); vma = lookup_vma(obj, aspace); if (!GEM_WARN_ON(!vma)) { - msm_gem_vma_unpin(vma); msm_gem_unpin_locked(obj); } msm_gem_unlock(obj); @@ -703,6 +707,8 @@ static void *get_vaddr(struct drm_gem_object *obj, unsigned madv) if (IS_ERR(pages)) return ERR_CAST(pages); + pin_obj_locked(obj); + /* increment vmap_count *before* vmap() call, so shrinker can * check vmap_count (is_vunmapable()) outside of msm_obj lock. * This guarantees that we won't try to msm_gem_vunmap() this @@ -981,11 +987,10 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m, } else { name = comm = NULL; } - seq_printf(m, " [%s%s%s: aspace=%p, %08llx,%s,inuse=%d]", + seq_printf(m, " [%s%s%s: aspace=%p, %08llx,%s]", name, comm ? ":" : "", comm ? comm : "", vma->aspace, vma->iova, - vma->mapped ? "mapped" : "unmapped", - msm_gem_vma_inuse(vma)); + vma->mapped ? "mapped" : "unmapped"); kfree(comm); } diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index 2bd6846c83a9..8ddef5443140 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -59,24 +59,16 @@ struct msm_fence_context; struct msm_gem_vma { struct drm_mm_node node; - spinlock_t lock; uint64_t iova; struct msm_gem_address_space *aspace; struct list_head list; /* node in msm_gem_object::vmas */ bool mapped; - int inuse; - uint32_t fence_mask; - uint32_t fence[MSM_GPU_MAX_RINGS]; - struct msm_fence_context *fctx[MSM_GPU_MAX_RINGS]; }; struct msm_gem_vma *msm_gem_vma_new(struct msm_gem_address_space *aspace); int msm_gem_vma_init(struct msm_gem_vma *vma, int size, u64 range_start, u64 range_end); -bool msm_gem_vma_inuse(struct msm_gem_vma *vma); void msm_gem_vma_purge(struct msm_gem_vma *vma); -void msm_gem_vma_unpin(struct msm_gem_vma *vma); -void msm_gem_vma_unpin_fenced(struct msm_gem_vma *vma, struct msm_fence_context *fctx); int msm_gem_vma_map(struct msm_gem_vma *vma, int prot, struct sg_table *sgt, int size); void msm_gem_vma_close(struct msm_gem_vma *vma); @@ -142,6 +134,7 @@ int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace, uint64_t *iova); void msm_gem_unpin_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace); +void msm_gem_pin_obj_locked(struct drm_gem_object *obj); struct page **msm_gem_pin_pages(struct drm_gem_object *obj); void msm_gem_unpin_pages(struct drm_gem_object *obj); int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, @@ -297,15 +290,13 @@ struct msm_gem_submit { /* make sure these don't conflict w/ MSM_SUBMIT_BO_x */ #define BO_VALID 0x8000 /* is current addr in cmdstream correct/valid? */ #define BO_LOCKED 0x4000 /* obj lock is held */ -#define BO_OBJ_PINNED 0x2000 /* obj (pages) is pinned and on active list */ -#define BO_VMA_PINNED 0x1000 /* vma (virtual address) is pinned */ +#define BO_PINNED 0x2000 /* obj (pages) is pinned and on active list */ uint32_t flags; union { - struct msm_gem_object *obj; + struct drm_gem_object *obj; uint32_t handle; }; uint64_t iova; - struct msm_gem_vma *vma; } bos[]; }; diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index 3f1aa4de3b87..99744de6c05a 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -86,7 +86,19 @@ void __msm_gem_submit_destroy(struct kref *kref) } dma_fence_put(submit->user_fence); - dma_fence_put(submit->hw_fence); + + /* + * If the submit is freed before msm_job_run(), then hw_fence is + * just some pre-allocated memory, not a reference counted fence. + * Once the job runs and the hw_fence is initialized, it will + * have a refcount of at least one, since the submit holds a ref + * to the hw_fence. + */ + if (kref_read(&submit->hw_fence->refcount) == 0) { + kfree(submit->hw_fence); + } else { + dma_fence_put(submit->hw_fence); + } put_pid(submit->pid); msm_submitqueue_put(submit->queue); @@ -153,7 +165,7 @@ static int submit_lookup_objects(struct msm_gem_submit *submit, drm_gem_object_get(obj); - submit->bos[i].obj = to_msm_bo(obj); + submit->bos[i].obj = obj; } out_unlock: @@ -239,7 +251,7 @@ out: static void submit_cleanup_bo(struct msm_gem_submit *submit, int i, unsigned cleanup_flags) { - struct drm_gem_object *obj = &submit->bos[i].obj->base; + struct drm_gem_object *obj = submit->bos[i].obj; unsigned flags = submit->bos[i].flags & cleanup_flags; /* @@ -249,10 +261,7 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i, */ submit->bos[i].flags &= ~cleanup_flags; - if (flags & BO_VMA_PINNED) - msm_gem_vma_unpin(submit->bos[i].vma); - - if (flags & BO_OBJ_PINNED) + if (flags & BO_PINNED) msm_gem_unpin_locked(obj); if (flags & BO_LOCKED) @@ -261,7 +270,7 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i, static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, int i) { - unsigned cleanup_flags = BO_VMA_PINNED | BO_OBJ_PINNED | BO_LOCKED; + unsigned cleanup_flags = BO_PINNED | BO_LOCKED; submit_cleanup_bo(submit, i, cleanup_flags); if (!(submit->bos[i].flags & BO_VALID)) @@ -275,7 +284,7 @@ static int submit_lock_objects(struct msm_gem_submit *submit) retry: for (i = 0; i < submit->nr_bos; i++) { - struct msm_gem_object *msm_obj = submit->bos[i].obj; + struct drm_gem_object *obj = submit->bos[i].obj; if (slow_locked == i) slow_locked = -1; @@ -283,7 +292,7 @@ retry: contended = i; if (!(submit->bos[i].flags & BO_LOCKED)) { - ret = dma_resv_lock_interruptible(msm_obj->base.resv, + ret = dma_resv_lock_interruptible(obj->resv, &submit->ticket); if (ret) goto fail; @@ -309,9 +318,9 @@ fail: submit_unlock_unpin_bo(submit, slow_locked); if (ret == -EDEADLK) { - struct msm_gem_object *msm_obj = submit->bos[contended].obj; + struct drm_gem_object *obj = submit->bos[contended].obj; /* we lost out in a seqno race, lock and retry.. */ - ret = dma_resv_lock_slow_interruptible(msm_obj->base.resv, + ret = dma_resv_lock_slow_interruptible(obj->resv, &submit->ticket); if (!ret) { submit->bos[contended].flags |= BO_LOCKED; @@ -334,7 +343,7 @@ static int submit_fence_sync(struct msm_gem_submit *submit, bool no_implicit) int i, ret = 0; for (i = 0; i < submit->nr_bos; i++) { - struct drm_gem_object *obj = &submit->bos[i].obj->base; + struct drm_gem_object *obj = submit->bos[i].obj; bool write = submit->bos[i].flags & MSM_SUBMIT_BO_WRITE; /* NOTE: _reserve_shared() must happen before @@ -372,12 +381,13 @@ static int submit_fence_sync(struct msm_gem_submit *submit, bool no_implicit) static int submit_pin_objects(struct msm_gem_submit *submit) { + struct msm_drm_private *priv = submit->dev->dev_private; int i, ret = 0; submit->valid = true; for (i = 0; i < submit->nr_bos; i++) { - struct drm_gem_object *obj = &submit->bos[i].obj->base; + struct drm_gem_object *obj = submit->bos[i].obj; struct msm_gem_vma *vma; /* if locking succeeded, pin bo: */ @@ -391,9 +401,6 @@ static int submit_pin_objects(struct msm_gem_submit *submit) if (ret) break; - submit->bos[i].flags |= BO_OBJ_PINNED | BO_VMA_PINNED; - submit->bos[i].vma = vma; - if (vma->iova == submit->bos[i].iova) { submit->bos[i].flags |= BO_VALID; } else { @@ -404,6 +411,20 @@ static int submit_pin_objects(struct msm_gem_submit *submit) } } + /* + * A second loop while holding the LRU lock (a) avoids acquiring/dropping + * the LRU lock for each individual bo, while (b) avoiding holding the + * LRU lock while calling msm_gem_pin_vma_locked() (which could trigger + * get_pages() which could trigger reclaim.. and if we held the LRU lock + * could trigger deadlock with the shrinker). + */ + mutex_lock(&priv->lru.lock); + for (i = 0; i < submit->nr_bos; i++) { + msm_gem_pin_obj_locked(submit->bos[i].obj); + submit->bos[i].flags |= BO_PINNED; + } + mutex_unlock(&priv->lru.lock); + return ret; } @@ -412,7 +433,7 @@ static void submit_attach_object_fences(struct msm_gem_submit *submit) int i; for (i = 0; i < submit->nr_bos; i++) { - struct drm_gem_object *obj = &submit->bos[i].obj->base; + struct drm_gem_object *obj = submit->bos[i].obj; if (submit->bos[i].flags & MSM_SUBMIT_BO_WRITE) dma_resv_add_fence(obj->resv, submit->user_fence, @@ -424,7 +445,7 @@ static void submit_attach_object_fences(struct msm_gem_submit *submit) } static int submit_bo(struct msm_gem_submit *submit, uint32_t idx, - struct msm_gem_object **obj, uint64_t *iova, bool *valid) + struct drm_gem_object **obj, uint64_t *iova, bool *valid) { if (idx >= submit->nr_bos) { DRM_ERROR("invalid buffer index: %u (out of %u)\n", @@ -443,7 +464,7 @@ static int submit_bo(struct msm_gem_submit *submit, uint32_t idx, } /* process the reloc's and patch up the cmdstream as needed: */ -static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *obj, +static int submit_reloc(struct msm_gem_submit *submit, struct drm_gem_object *obj, uint32_t offset, uint32_t nr_relocs, struct drm_msm_gem_submit_reloc *relocs) { uint32_t i, last_offset = 0; @@ -461,7 +482,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob /* For now, just map the entire thing. Eventually we probably * to do it page-by-page, w/ kmap() if not vmap()d.. */ - ptr = msm_gem_get_vaddr_locked(&obj->base); + ptr = msm_gem_get_vaddr_locked(obj); if (IS_ERR(ptr)) { ret = PTR_ERR(ptr); @@ -485,7 +506,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob /* offset in dwords: */ off = submit_reloc.submit_offset / 4; - if ((off >= (obj->base.size / 4)) || + if ((off >= (obj->size / 4)) || (off < last_offset)) { DRM_ERROR("invalid offset %u at reloc %u\n", off, i); ret = -EINVAL; @@ -512,7 +533,7 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob } out: - msm_gem_put_vaddr_locked(&obj->base); + msm_gem_put_vaddr_locked(obj); return ret; } @@ -527,13 +548,13 @@ static void submit_cleanup(struct msm_gem_submit *submit, bool error) unsigned i; if (error) - cleanup_flags |= BO_VMA_PINNED | BO_OBJ_PINNED; + cleanup_flags |= BO_PINNED; for (i = 0; i < submit->nr_bos; i++) { - struct msm_gem_object *msm_obj = submit->bos[i].obj; + struct drm_gem_object *obj = submit->bos[i].obj; submit_cleanup_bo(submit, i, cleanup_flags); if (error) - drm_gem_object_put(&msm_obj->base); + drm_gem_object_put(obj); } } @@ -542,7 +563,7 @@ void msm_submit_retire(struct msm_gem_submit *submit) int i; for (i = 0; i < submit->nr_bos; i++) { - struct drm_gem_object *obj = &submit->bos[i].obj->base; + struct drm_gem_object *obj = submit->bos[i].obj; drm_gem_object_put(obj); } @@ -849,17 +870,17 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, goto out; for (i = 0; i < args->nr_cmds; i++) { - struct msm_gem_object *msm_obj; + struct drm_gem_object *obj; uint64_t iova; ret = submit_bo(submit, submit->cmd[i].idx, - &msm_obj, &iova, NULL); + &obj, &iova, NULL); if (ret) goto out; if (!submit->cmd[i].size || ((submit->cmd[i].size + submit->cmd[i].offset) > - msm_obj->base.size / 4)) { + obj->size / 4)) { DRM_ERROR("invalid cmdstream size: %u\n", submit->cmd[i].size * 4); ret = -EINVAL; goto out; @@ -870,7 +891,17 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (submit->valid) continue; - ret = submit_reloc(submit, msm_obj, submit->cmd[i].offset * 4, + if (!gpu->allow_relocs) { + if (submit->cmd[i].nr_relocs) { + DRM_ERROR("relocs not allowed\n"); + ret = -EINVAL; + goto out; + } + + continue; + } + + ret = submit_reloc(submit, obj, submit->cmd[i].offset * 4, submit->cmd[i].nr_relocs, submit->cmd[i].relocs); if (ret) goto out; @@ -889,7 +920,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, * after the job is armed */ if ((args->flags & MSM_SUBMIT_FENCE_SN_IN) && - idr_find(&queue->fence_idr, args->fence)) { + (!args->fence || idr_find(&queue->fence_idr, args->fence))) { spin_unlock(&queue->idr_lock); idr_preload_end(); ret = -EINVAL; diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c index 98287ed99960..11e842dda73c 100644 --- a/drivers/gpu/drm/msm/msm_gem_vma.c +++ b/drivers/gpu/drm/msm/msm_gem_vma.c @@ -38,41 +38,12 @@ msm_gem_address_space_get(struct msm_gem_address_space *aspace) return aspace; } -bool msm_gem_vma_inuse(struct msm_gem_vma *vma) -{ - bool ret = true; - - spin_lock(&vma->lock); - - if (vma->inuse > 0) - goto out; - - while (vma->fence_mask) { - unsigned idx = ffs(vma->fence_mask) - 1; - - if (!msm_fence_completed(vma->fctx[idx], vma->fence[idx])) - goto out; - - vma->fence_mask &= ~BIT(idx); - } - - ret = false; - -out: - spin_unlock(&vma->lock); - - return ret; -} - /* Actually unmap memory for the vma */ void msm_gem_vma_purge(struct msm_gem_vma *vma) { struct msm_gem_address_space *aspace = vma->aspace; unsigned size = vma->node.size; - /* Print a message if we try to purge a vma in use */ - GEM_WARN_ON(msm_gem_vma_inuse(vma)); - /* Don't do anything if the memory isn't mapped */ if (!vma->mapped) return; @@ -82,33 +53,6 @@ void msm_gem_vma_purge(struct msm_gem_vma *vma) vma->mapped = false; } -static void vma_unpin_locked(struct msm_gem_vma *vma) -{ - if (GEM_WARN_ON(!vma->inuse)) - return; - if (!GEM_WARN_ON(!vma->iova)) - vma->inuse--; -} - -/* Remove reference counts for the mapping */ -void msm_gem_vma_unpin(struct msm_gem_vma *vma) -{ - spin_lock(&vma->lock); - vma_unpin_locked(vma); - spin_unlock(&vma->lock); -} - -/* Replace pin reference with fence: */ -void msm_gem_vma_unpin_fenced(struct msm_gem_vma *vma, struct msm_fence_context *fctx) -{ - spin_lock(&vma->lock); - vma->fctx[fctx->index] = fctx; - vma->fence[fctx->index] = fctx->last_fence; - vma->fence_mask |= BIT(fctx->index); - vma_unpin_locked(vma); - spin_unlock(&vma->lock); -} - /* Map and pin vma: */ int msm_gem_vma_map(struct msm_gem_vma *vma, int prot, @@ -120,11 +64,6 @@ msm_gem_vma_map(struct msm_gem_vma *vma, int prot, if (GEM_WARN_ON(!vma->iova)) return -EINVAL; - /* Increase the usage counter */ - spin_lock(&vma->lock); - vma->inuse++; - spin_unlock(&vma->lock); - if (vma->mapped) return 0; @@ -146,9 +85,6 @@ msm_gem_vma_map(struct msm_gem_vma *vma, int prot, if (ret) { vma->mapped = false; - spin_lock(&vma->lock); - vma->inuse--; - spin_unlock(&vma->lock); } return ret; @@ -159,7 +95,7 @@ void msm_gem_vma_close(struct msm_gem_vma *vma) { struct msm_gem_address_space *aspace = vma->aspace; - GEM_WARN_ON(msm_gem_vma_inuse(vma) || vma->mapped); + GEM_WARN_ON(vma->mapped); spin_lock(&aspace->lock); if (vma->iova) @@ -179,7 +115,6 @@ struct msm_gem_vma *msm_gem_vma_new(struct msm_gem_address_space *aspace) if (!vma) return NULL; - spin_lock_init(&vma->lock); vma->aspace = aspace; return vma; diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 52db90e34ead..7f64c6667300 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -219,36 +219,36 @@ static void msm_gpu_devcoredump_free(void *data) } static void msm_gpu_crashstate_get_bo(struct msm_gpu_state *state, - struct msm_gem_object *obj, u64 iova, bool full) + struct drm_gem_object *obj, u64 iova, bool full) { struct msm_gpu_state_bo *state_bo = &state->bos[state->nr_bos]; /* Don't record write only objects */ - state_bo->size = obj->base.size; + state_bo->size = obj->size; state_bo->iova = iova; - BUILD_BUG_ON(sizeof(state_bo->name) != sizeof(obj->name)); + BUILD_BUG_ON(sizeof(state_bo->name) != sizeof(to_msm_bo(obj)->name)); - memcpy(state_bo->name, obj->name, sizeof(state_bo->name)); + memcpy(state_bo->name, to_msm_bo(obj)->name, sizeof(state_bo->name)); if (full) { void *ptr; - state_bo->data = kvmalloc(obj->base.size, GFP_KERNEL); + state_bo->data = kvmalloc(obj->size, GFP_KERNEL); if (!state_bo->data) goto out; - msm_gem_lock(&obj->base); - ptr = msm_gem_get_vaddr_active(&obj->base); - msm_gem_unlock(&obj->base); + msm_gem_lock(obj); + ptr = msm_gem_get_vaddr_active(obj); + msm_gem_unlock(obj); if (IS_ERR(ptr)) { kvfree(state_bo->data); state_bo->data = NULL; goto out; } - memcpy(state_bo->data, ptr, obj->base.size); - msm_gem_put_vaddr(&obj->base); + memcpy(state_bo->data, ptr, obj->size); + msm_gem_put_vaddr(obj); } out: state->nr_bos++; @@ -749,13 +749,11 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) struct msm_ringbuffer *ring = submit->ring; unsigned long flags; - WARN_ON(!mutex_is_locked(&gpu->lock)); - pm_runtime_get_sync(&gpu->pdev->dev); - msm_gpu_hw_init(gpu); + mutex_lock(&gpu->lock); - submit->seqno = submit->hw_fence->seqno; + msm_gpu_hw_init(gpu); update_sw_cntrs(gpu); @@ -781,8 +779,11 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) gpu->funcs->submit(gpu, submit); gpu->cur_ctx_seqno = submit->queue->ctx->seqno; - pm_runtime_put(&gpu->pdev->dev); hangcheck_timer_reset(gpu); + + mutex_unlock(&gpu->lock); + + pm_runtime_put(&gpu->pdev->dev); } /* @@ -897,7 +898,6 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, gpu->irq = platform_get_irq(pdev, 0); if (gpu->irq < 0) { ret = gpu->irq; - DRM_DEV_ERROR(drm->dev, "failed to get irq: %d\n", ret); goto fail; } diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index 7a4fa1b8655b..4252e3839fbc 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -285,6 +285,15 @@ struct msm_gpu { /* True if the hardware supports expanded apriv (a650 and newer) */ bool hw_apriv; + /** + * @allow_relocs: allow relocs in SUBMIT ioctl + * + * Mesa won't use relocs for driver version 1.4.0 and later. This + * switch-over happened early enough in mesa a6xx bringup that we + * can disallow relocs for a6xx and newer. + */ + bool allow_relocs; + struct thermal_cooling_device *cooling; }; diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c index 268fb3d490de..a429b704059e 100644 --- a/drivers/gpu/drm/msm/msm_mdss.c +++ b/drivers/gpu/drm/msm/msm_mdss.c @@ -15,7 +15,7 @@ #include <linux/pm_runtime.h> #include <linux/reset.h> -#include "msm_drv.h" +#include "msm_mdss.h" #include "msm_kms.h" #define HW_REV 0x0 @@ -28,16 +28,6 @@ #define MIN_IB_BW 400000000UL /* Min ib vote 400MB */ -struct msm_mdss_data { - u32 ubwc_version; - /* can be read from register 0x58 */ - u32 ubwc_dec_version; - u32 ubwc_swizzle; - u32 ubwc_static; - u32 highest_bank_bit; - u32 macrotile_mode; -}; - struct msm_mdss { struct device *dev; @@ -187,11 +177,6 @@ static int _msm_mdss_irq_domain_add(struct msm_mdss *msm_mdss) return 0; } -#define UBWC_1_0 0x10000000 -#define UBWC_2_0 0x20000000 -#define UBWC_3_0 0x30000000 -#define UBWC_4_0 0x40000000 - static void msm_mdss_setup_ubwc_dec_20(struct msm_mdss *msm_mdss) { const struct msm_mdss_data *data = msm_mdss->mdss_data; @@ -206,10 +191,10 @@ static void msm_mdss_setup_ubwc_dec_30(struct msm_mdss *msm_mdss) (data->highest_bank_bit & 0x3) << 4 | (data->macrotile_mode & 0x1) << 12; - if (data->ubwc_version == UBWC_3_0) + if (data->ubwc_enc_version == UBWC_3_0) value |= BIT(10); - if (data->ubwc_version == UBWC_1_0) + if (data->ubwc_enc_version == UBWC_1_0) value |= BIT(8); writel_relaxed(value, msm_mdss->mmio + UBWC_STATIC); @@ -225,15 +210,30 @@ static void msm_mdss_setup_ubwc_dec_40(struct msm_mdss *msm_mdss) writel_relaxed(value, msm_mdss->mmio + UBWC_STATIC); - if (data->ubwc_version == UBWC_3_0) { + if (data->ubwc_enc_version == UBWC_3_0) { writel_relaxed(1, msm_mdss->mmio + UBWC_CTRL_2); writel_relaxed(0, msm_mdss->mmio + UBWC_PREDICTION_MODE); } else { - writel_relaxed(2, msm_mdss->mmio + UBWC_CTRL_2); + if (data->ubwc_dec_version == UBWC_4_3) + writel_relaxed(3, msm_mdss->mmio + UBWC_CTRL_2); + else + writel_relaxed(2, msm_mdss->mmio + UBWC_CTRL_2); writel_relaxed(1, msm_mdss->mmio + UBWC_PREDICTION_MODE); } } +const struct msm_mdss_data *msm_mdss_get_mdss_data(struct device *dev) +{ + struct msm_mdss *mdss; + + if (!dev) + return ERR_PTR(-EINVAL); + + mdss = dev_get_drvdata(dev); + + return mdss->mdss_data; +} + static int msm_mdss_enable(struct msm_mdss *msm_mdss) { int ret; @@ -266,6 +266,10 @@ static int msm_mdss_enable(struct msm_mdss *msm_mdss) * UBWC_n and the rest of params comes from hw data. */ switch (msm_mdss->mdss_data->ubwc_dec_version) { + case 0: /* no UBWC */ + case UBWC_1_0: + /* do nothing */ + break; case UBWC_2_0: msm_mdss_setup_ubwc_dec_20(msm_mdss); break; @@ -273,6 +277,7 @@ static int msm_mdss_enable(struct msm_mdss *msm_mdss) msm_mdss_setup_ubwc_dec_30(msm_mdss); break; case UBWC_4_0: + case UBWC_4_3: msm_mdss_setup_ubwc_dec_40(msm_mdss); break; default: @@ -501,14 +506,26 @@ static void mdss_remove(struct platform_device *pdev) msm_mdss_destroy(mdss); } +static const struct msm_mdss_data msm8998_data = { + .ubwc_enc_version = UBWC_1_0, + .ubwc_dec_version = UBWC_1_0, + .highest_bank_bit = 1, +}; + +static const struct msm_mdss_data qcm2290_data = { + /* no UBWC */ + .highest_bank_bit = 0x2, +}; + static const struct msm_mdss_data sc7180_data = { - .ubwc_version = UBWC_2_0, + .ubwc_enc_version = UBWC_2_0, .ubwc_dec_version = UBWC_2_0, .ubwc_static = 0x1e, + .highest_bank_bit = 0x3, }; static const struct msm_mdss_data sc7280_data = { - .ubwc_version = UBWC_3_0, + .ubwc_enc_version = UBWC_3_0, .ubwc_dec_version = UBWC_4_0, .ubwc_swizzle = 6, .ubwc_static = 1, @@ -517,14 +534,14 @@ static const struct msm_mdss_data sc7280_data = { }; static const struct msm_mdss_data sc8180x_data = { - .ubwc_version = UBWC_3_0, + .ubwc_enc_version = UBWC_3_0, .ubwc_dec_version = UBWC_3_0, .highest_bank_bit = 3, .macrotile_mode = 1, }; static const struct msm_mdss_data sc8280xp_data = { - .ubwc_version = UBWC_4_0, + .ubwc_enc_version = UBWC_4_0, .ubwc_dec_version = UBWC_4_0, .ubwc_swizzle = 6, .ubwc_static = 1, @@ -533,13 +550,13 @@ static const struct msm_mdss_data sc8280xp_data = { }; static const struct msm_mdss_data sdm845_data = { - .ubwc_version = UBWC_2_0, + .ubwc_enc_version = UBWC_2_0, .ubwc_dec_version = UBWC_2_0, .highest_bank_bit = 2, }; static const struct msm_mdss_data sm6350_data = { - .ubwc_version = UBWC_2_0, + .ubwc_enc_version = UBWC_2_0, .ubwc_dec_version = UBWC_2_0, .ubwc_swizzle = 6, .ubwc_static = 0x1e, @@ -547,20 +564,28 @@ static const struct msm_mdss_data sm6350_data = { }; static const struct msm_mdss_data sm8150_data = { - .ubwc_version = UBWC_3_0, + .ubwc_enc_version = UBWC_3_0, .ubwc_dec_version = UBWC_3_0, .highest_bank_bit = 2, }; static const struct msm_mdss_data sm6115_data = { - .ubwc_version = UBWC_1_0, + .ubwc_enc_version = UBWC_1_0, .ubwc_dec_version = UBWC_2_0, .ubwc_swizzle = 7, .ubwc_static = 0x11f, + .highest_bank_bit = 0x1, +}; + +static const struct msm_mdss_data sm6125_data = { + .ubwc_enc_version = UBWC_1_0, + .ubwc_dec_version = UBWC_3_0, + .ubwc_swizzle = 1, + .highest_bank_bit = 1, }; static const struct msm_mdss_data sm8250_data = { - .ubwc_version = UBWC_4_0, + .ubwc_enc_version = UBWC_4_0, .ubwc_dec_version = UBWC_4_0, .ubwc_swizzle = 6, .ubwc_static = 1, @@ -569,23 +594,33 @@ static const struct msm_mdss_data sm8250_data = { .macrotile_mode = 1, }; +static const struct msm_mdss_data sm8550_data = { + .ubwc_enc_version = UBWC_4_0, + .ubwc_dec_version = UBWC_4_3, + .ubwc_swizzle = 6, + .ubwc_static = 1, + /* TODO: highest_bank_bit = 2 for LP_DDR4 */ + .highest_bank_bit = 3, + .macrotile_mode = 1, +}; static const struct of_device_id mdss_dt_match[] = { { .compatible = "qcom,mdss" }, - { .compatible = "qcom,msm8998-mdss" }, - { .compatible = "qcom,qcm2290-mdss" }, + { .compatible = "qcom,msm8998-mdss", .data = &msm8998_data }, + { .compatible = "qcom,qcm2290-mdss", .data = &qcm2290_data }, { .compatible = "qcom,sdm845-mdss", .data = &sdm845_data }, { .compatible = "qcom,sc7180-mdss", .data = &sc7180_data }, { .compatible = "qcom,sc7280-mdss", .data = &sc7280_data }, { .compatible = "qcom,sc8180x-mdss", .data = &sc8180x_data }, { .compatible = "qcom,sc8280xp-mdss", .data = &sc8280xp_data }, { .compatible = "qcom,sm6115-mdss", .data = &sm6115_data }, + { .compatible = "qcom,sm6125-mdss", .data = &sm6125_data }, { .compatible = "qcom,sm6350-mdss", .data = &sm6350_data }, { .compatible = "qcom,sm6375-mdss", .data = &sm6350_data }, { .compatible = "qcom,sm8150-mdss", .data = &sm8150_data }, { .compatible = "qcom,sm8250-mdss", .data = &sm8250_data }, { .compatible = "qcom,sm8350-mdss", .data = &sm8250_data }, { .compatible = "qcom,sm8450-mdss", .data = &sm8250_data }, - { .compatible = "qcom,sm8550-mdss", .data = &sm8250_data }, + { .compatible = "qcom,sm8550-mdss", .data = &sm8550_data }, {} }; MODULE_DEVICE_TABLE(of, mdss_dt_match); diff --git a/drivers/gpu/drm/msm/msm_mdss.h b/drivers/gpu/drm/msm/msm_mdss.h new file mode 100644 index 000000000000..02bbab42adbc --- /dev/null +++ b/drivers/gpu/drm/msm/msm_mdss.h @@ -0,0 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018, The Linux Foundation + */ + +#ifndef __MSM_MDSS_H__ +#define __MSM_MDSS_H__ + +struct msm_mdss_data { + u32 ubwc_enc_version; + /* can be read from register 0x58 */ + u32 ubwc_dec_version; + u32 ubwc_swizzle; + u32 ubwc_static; + u32 highest_bank_bit; + u32 macrotile_mode; +}; + +#define UBWC_1_0 0x10000000 +#define UBWC_2_0 0x20000000 +#define UBWC_3_0 0x30000000 +#define UBWC_4_0 0x40000000 +#define UBWC_4_3 0x40030000 + +const struct msm_mdss_data *msm_mdss_get_mdss_data(struct device *dev); + +#endif /* __MSM_MDSS_H__ */ diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index 8d5687d5ed78..5adc51f7ab59 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -310,7 +310,7 @@ static void snapshot_buf(struct msm_rd_state *rd, struct msm_gem_submit *submit, int idx, uint64_t iova, uint32_t size, bool full) { - struct msm_gem_object *obj = submit->bos[idx].obj; + struct drm_gem_object *obj = submit->bos[idx].obj; unsigned offset = 0; const char *buf; @@ -318,7 +318,7 @@ static void snapshot_buf(struct msm_rd_state *rd, offset = iova - submit->bos[idx].iova; } else { iova = submit->bos[idx].iova; - size = obj->base.size; + size = obj->size; } /* @@ -335,7 +335,7 @@ static void snapshot_buf(struct msm_rd_state *rd, if (!(submit->bos[idx].flags & MSM_SUBMIT_BO_READ)) return; - buf = msm_gem_get_vaddr_active(&obj->base); + buf = msm_gem_get_vaddr_active(obj); if (IS_ERR(buf)) return; @@ -343,7 +343,7 @@ static void snapshot_buf(struct msm_rd_state *rd, rd_write_section(rd, RD_BUFFER_CONTENTS, buf, size); - msm_gem_put_vaddr_locked(&obj->base); + msm_gem_put_vaddr_locked(obj); } /* called under gpu->lock */ diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.c b/drivers/gpu/drm/msm/msm_ringbuffer.c index b60199184409..40c0bc35a44c 100644 --- a/drivers/gpu/drm/msm/msm_ringbuffer.c +++ b/drivers/gpu/drm/msm/msm_ringbuffer.c @@ -16,25 +16,26 @@ static struct dma_fence *msm_job_run(struct drm_sched_job *job) struct msm_gem_submit *submit = to_msm_submit(job); struct msm_fence_context *fctx = submit->ring->fctx; struct msm_gpu *gpu = submit->gpu; + struct msm_drm_private *priv = gpu->dev->dev_private; int i; msm_fence_init(submit->hw_fence, fctx); + submit->seqno = submit->hw_fence->seqno; + + mutex_lock(&priv->lru.lock); + for (i = 0; i < submit->nr_bos; i++) { - struct drm_gem_object *obj = &submit->bos[i].obj->base; + struct drm_gem_object *obj = submit->bos[i].obj; - msm_gem_vma_unpin_fenced(submit->bos[i].vma, fctx); msm_gem_unpin_active(obj); - submit->bos[i].flags &= ~(BO_VMA_PINNED | BO_OBJ_PINNED); + submit->bos[i].flags &= ~BO_PINNED; } - /* TODO move submit path over to using a per-ring lock.. */ - mutex_lock(&gpu->lock); + mutex_unlock(&priv->lru.lock); msm_gpu_submit(gpu, submit); - mutex_unlock(&gpu->lock); - return dma_fence_get(submit->hw_fence); } diff --git a/drivers/gpu/drm/nouveau/dispnv04/crtc.c b/drivers/gpu/drm/nouveau/dispnv04/crtc.c index a34924523133..a34917b048f9 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/crtc.c +++ b/drivers/gpu/drm/nouveau/dispnv04/crtc.c @@ -1122,18 +1122,11 @@ nv04_page_flip_emit(struct nouveau_channel *chan, PUSH_NVSQ(push, NV_SW, NV_SW_PAGE_FLIP, 0x00000000); PUSH_KICK(push); - ret = nouveau_fence_new(pfence); + ret = nouveau_fence_new(pfence, chan); if (ret) goto fail; - ret = nouveau_fence_emit(*pfence, chan); - if (ret) - goto fail_fence_unref; - return 0; - -fail_fence_unref: - nouveau_fence_unref(pfence); fail: spin_lock_irqsave(&dev->event_lock, flags); list_del(&s->head); diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index 39e9ba4139c7..bba01fa0780c 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -1878,6 +1878,8 @@ nv50_pior_destroy(struct drm_encoder *encoder) nvif_outp_dtor(&nv_encoder->outp); drm_encoder_cleanup(encoder); + + mutex_destroy(&nv_encoder->dp.hpd_irq_lock); kfree(encoder); } @@ -1922,6 +1924,8 @@ nv50_pior_create(struct drm_connector *connector, struct dcb_output *dcbe) nv_encoder->i2c = ddc; nv_encoder->aux = aux; + mutex_init(&nv_encoder->dp.hpd_irq_lock); + encoder = to_drm_encoder(nv_encoder); encoder->possible_crtcs = dcbe->heads; encoder->possible_clones = 0; diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h index 40a1065ae626..ef441dfdea09 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h @@ -16,7 +16,7 @@ struct nvkm_i2c_bus { const struct nvkm_i2c_bus_func *func; struct nvkm_i2c_pad *pad; #define NVKM_I2C_BUS_CCB(n) /* 'n' is ccb index */ (n) -#define NVKM_I2C_BUS_EXT(n) /* 'n' is dcb external encoder type */ ((n) + 0x100) +#define NVKM_I2C_BUS_EXT(n) /* 'n' is dcb external encoder type */ ((n) + 0x10) #define NVKM_I2C_BUS_PRI /* ccb primary comm. port */ -1 #define NVKM_I2C_BUS_SEC /* ccb secondary comm. port */ -2 int id; @@ -38,7 +38,7 @@ struct nvkm_i2c_aux { const struct nvkm_i2c_aux_func *func; struct nvkm_i2c_pad *pad; #define NVKM_I2C_AUX_CCB(n) /* 'n' is ccb index */ (n) -#define NVKM_I2C_AUX_EXT(n) /* 'n' is dcb external encoder type */ ((n) + 0x100) +#define NVKM_I2C_AUX_EXT(n) /* 'n' is dcb external encoder type */ ((n) + 0x10) int id; struct mutex mutex; diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 19cab37ac69c..0f3bd187ede6 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -875,16 +875,10 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int evict, if (ret) goto out_unlock; - ret = nouveau_fence_new(&fence); + ret = nouveau_fence_new(&fence, chan); if (ret) goto out_unlock; - ret = nouveau_fence_emit(fence, chan); - if (ret) { - nouveau_fence_unref(&fence); - goto out_unlock; - } - /* TODO: figure out a better solution here * * wait on the fence here explicitly as going through diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c index 1fd5ccf41128..bb3d6e5c122f 100644 --- a/drivers/gpu/drm/nouveau/nouveau_chan.c +++ b/drivers/gpu/drm/nouveau/nouveau_chan.c @@ -70,11 +70,9 @@ nouveau_channel_idle(struct nouveau_channel *chan) struct nouveau_fence *fence = NULL; int ret; - ret = nouveau_fence_new(&fence); + ret = nouveau_fence_new(&fence, chan); if (!ret) { - ret = nouveau_fence_emit(fence, chan); - if (!ret) - ret = nouveau_fence_wait(fence, false, false); + ret = nouveau_fence_wait(fence, false, false); nouveau_fence_unref(&fence); } diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index 68b4fb4bec63..79ea30aac31f 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -970,7 +970,7 @@ nouveau_connector_get_modes(struct drm_connector *connector) /* Determine display colour depth for everything except LVDS now, * DP requires this before mode_valid() is called. */ - if (connector->connector_type != DRM_MODE_CONNECTOR_LVDS && nv_connector->native_mode) + if (connector->connector_type != DRM_MODE_CONNECTOR_LVDS) nouveau_connector_detect_depth(connector); /* Find the native mode if this is a digital panel, if we didn't @@ -1411,8 +1411,7 @@ nouveau_connector_create(struct drm_device *dev, ret = nvif_conn_ctor(&disp->disp, nv_connector->base.name, nv_connector->index, &nv_connector->conn); if (ret) { - kfree(nv_connector); - return ERR_PTR(ret); + goto drm_conn_err; } ret = nvif_conn_event_ctor(&nv_connector->conn, "kmsHotplug", @@ -1429,8 +1428,7 @@ nouveau_connector_create(struct drm_device *dev, if (ret) { nvif_event_dtor(&nv_connector->hpd); nvif_conn_dtor(&nv_connector->conn); - kfree(nv_connector); - return ERR_PTR(ret); + goto drm_conn_err; } } } @@ -1478,4 +1476,9 @@ nouveau_connector_create(struct drm_device *dev, drm_connector_register(connector); return connector; + +drm_conn_err: + drm_connector_cleanup(connector); + kfree(nv_connector); + return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c index 61e84562094a..12feecf71e75 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -209,8 +209,7 @@ static vm_fault_t nouveau_dmem_migrate_to_ram(struct vm_fault *vmf) goto done; } - if (!nouveau_fence_new(&fence)) - nouveau_fence_emit(fence, dmem->migrate.chan); + nouveau_fence_new(&fence, dmem->migrate.chan); migrate_vma_pages(&args); nouveau_dmem_fence_done(&fence); dma_unmap_page(drm->dev->dev, dma_addr, PAGE_SIZE, DMA_BIDIRECTIONAL); @@ -403,8 +402,7 @@ nouveau_dmem_evict_chunk(struct nouveau_dmem_chunk *chunk) } } - if (!nouveau_fence_new(&fence)) - nouveau_fence_emit(fence, chunk->drm->dmem->migrate.chan); + nouveau_fence_new(&fence, chunk->drm->dmem->migrate.chan); migrate_device_pages(src_pfns, dst_pfns, npages); nouveau_dmem_fence_done(&fence); migrate_device_finalize(src_pfns, dst_pfns, npages); @@ -677,8 +675,7 @@ static void nouveau_dmem_migrate_chunk(struct nouveau_drm *drm, addr += PAGE_SIZE; } - if (!nouveau_fence_new(&fence)) - nouveau_fence_emit(fence, drm->dmem->migrate.chan); + nouveau_fence_new(&fence, drm->dmem->migrate.chan); migrate_vma_pages(args); nouveau_dmem_fence_done(&fence); nouveau_pfns_map(svmm, args->vma->vm_mm, args->start, pfns, i); diff --git a/drivers/gpu/drm/nouveau/nouveau_exec.c b/drivers/gpu/drm/nouveau/nouveau_exec.c index a90c4cd8cbb2..19024ce21fbb 100644 --- a/drivers/gpu/drm/nouveau/nouveau_exec.c +++ b/drivers/gpu/drm/nouveau/nouveau_exec.c @@ -96,7 +96,8 @@ nouveau_exec_job_submit(struct nouveau_job *job) unsigned long index; int ret; - ret = nouveau_fence_new(&exec_job->fence); + /* Create a new fence, but do not emit yet. */ + ret = nouveau_fence_create(&exec_job->fence, exec_job->chan); if (ret) return ret; @@ -170,13 +171,17 @@ nouveau_exec_job_run(struct nouveau_job *job) nv50_dma_push(chan, p->va, p->va_len, no_prefetch); } - ret = nouveau_fence_emit(fence, chan); + ret = nouveau_fence_emit(fence); if (ret) { + nouveau_fence_unref(&exec_job->fence); NV_PRINTK(err, job->cli, "error fencing pushbuf: %d\n", ret); WIND_RING(chan); return ERR_PTR(ret); } + /* The fence was emitted successfully, set the job's fence pointer to + * NULL in order to avoid freeing it up when the job is cleaned up. + */ exec_job->fence = NULL; return &fence->base; @@ -189,7 +194,7 @@ nouveau_exec_job_free(struct nouveau_job *job) nouveau_job_free(job); - nouveau_fence_unref(&exec_job->fence); + kfree(exec_job->fence); kfree(exec_job->push.s); kfree(exec_job); } diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 77c739a55b19..61d9e70da9fd 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -205,16 +205,13 @@ nouveau_fence_context_new(struct nouveau_channel *chan, struct nouveau_fence_cha } int -nouveau_fence_emit(struct nouveau_fence *fence, struct nouveau_channel *chan) +nouveau_fence_emit(struct nouveau_fence *fence) { + struct nouveau_channel *chan = fence->channel; struct nouveau_fence_chan *fctx = chan->fence; struct nouveau_fence_priv *priv = (void*)chan->drm->fence; int ret; - if (unlikely(!chan->fence)) - return -ENODEV; - - fence->channel = chan; fence->timeout = jiffies + (15 * HZ); if (priv->uevent) @@ -406,18 +403,41 @@ nouveau_fence_unref(struct nouveau_fence **pfence) } int -nouveau_fence_new(struct nouveau_fence **pfence) +nouveau_fence_create(struct nouveau_fence **pfence, + struct nouveau_channel *chan) { struct nouveau_fence *fence; + if (unlikely(!chan->fence)) + return -ENODEV; + fence = kzalloc(sizeof(*fence), GFP_KERNEL); if (!fence) return -ENOMEM; + fence->channel = chan; + *pfence = fence; return 0; } +int +nouveau_fence_new(struct nouveau_fence **pfence, + struct nouveau_channel *chan) +{ + int ret = 0; + + ret = nouveau_fence_create(pfence, chan); + if (ret) + return ret; + + ret = nouveau_fence_emit(*pfence); + if (ret) + nouveau_fence_unref(pfence); + + return ret; +} + static const char *nouveau_fence_get_get_driver_name(struct dma_fence *fence) { return "nouveau"; diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouveau/nouveau_fence.h index 2c72d96ef17d..64d33ae7f356 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.h +++ b/drivers/gpu/drm/nouveau/nouveau_fence.h @@ -17,10 +17,11 @@ struct nouveau_fence { unsigned long timeout; }; -int nouveau_fence_new(struct nouveau_fence **); +int nouveau_fence_create(struct nouveau_fence **, struct nouveau_channel *); +int nouveau_fence_new(struct nouveau_fence **, struct nouveau_channel *); void nouveau_fence_unref(struct nouveau_fence **); -int nouveau_fence_emit(struct nouveau_fence *, struct nouveau_channel *); +int nouveau_fence_emit(struct nouveau_fence *); bool nouveau_fence_done(struct nouveau_fence *); int nouveau_fence_wait(struct nouveau_fence *, bool lazy, bool intr); int nouveau_fence_sync(struct nouveau_bo *, struct nouveau_channel *, bool exclusive, bool intr); diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index c0b10d8d3d03..a0d303e5ce3d 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -914,11 +914,8 @@ revalidate: } } - ret = nouveau_fence_new(&fence); - if (!ret) - ret = nouveau_fence_emit(fence, chan); + ret = nouveau_fence_new(&fence, chan); if (ret) { - nouveau_fence_unref(&fence); NV_PRINTK(err, cli, "error fencing pushbuf: %d\n", ret); WIND_RING(chan); goto out; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/uconn.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/uconn.c index dad942be6679..46b057fe1412 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/uconn.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/uconn.c @@ -81,20 +81,29 @@ nvkm_uconn_uevent(struct nvkm_object *object, void *argv, u32 argc, struct nvkm_ return -ENOSYS; list_for_each_entry(outp, &conn->disp->outps, head) { - if (outp->info.connector == conn->index && outp->dp.aux) { - if (args->v0.types & NVIF_CONN_EVENT_V0_PLUG ) bits |= NVKM_I2C_PLUG; - if (args->v0.types & NVIF_CONN_EVENT_V0_UNPLUG) bits |= NVKM_I2C_UNPLUG; - if (args->v0.types & NVIF_CONN_EVENT_V0_IRQ ) bits |= NVKM_I2C_IRQ; + if (outp->info.connector == conn->index) + break; + } - return nvkm_uevent_add(uevent, &device->i2c->event, outp->dp.aux->id, bits, - nvkm_uconn_uevent_aux); - } + if (&outp->head == &conn->disp->outps) + return -EINVAL; + + if (outp->dp.aux && !outp->info.location) { + if (args->v0.types & NVIF_CONN_EVENT_V0_PLUG ) bits |= NVKM_I2C_PLUG; + if (args->v0.types & NVIF_CONN_EVENT_V0_UNPLUG) bits |= NVKM_I2C_UNPLUG; + if (args->v0.types & NVIF_CONN_EVENT_V0_IRQ ) bits |= NVKM_I2C_IRQ; + + return nvkm_uevent_add(uevent, &device->i2c->event, outp->dp.aux->id, bits, + nvkm_uconn_uevent_aux); } if (args->v0.types & NVIF_CONN_EVENT_V0_PLUG ) bits |= NVKM_GPIO_HI; if (args->v0.types & NVIF_CONN_EVENT_V0_UNPLUG) bits |= NVKM_GPIO_LO; - if (args->v0.types & NVIF_CONN_EVENT_V0_IRQ) - return -EINVAL; + if (args->v0.types & NVIF_CONN_EVENT_V0_IRQ) { + /* TODO: support DP IRQ on ANX9805 and remove this hack. */ + if (!outp->info.location) + return -EINVAL; + } return nvkm_uevent_add(uevent, &device->gpio->event, conn->info.hpd, bits, nvkm_uconn_uevent_gpio); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h index 00dbeda7e346..de161e7a04aa 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h @@ -117,6 +117,7 @@ void gk104_grctx_generate_r418800(struct gf100_gr *); extern const struct gf100_grctx_func gk110_grctx; void gk110_grctx_generate_r419eb0(struct gf100_gr *); +void gk110_grctx_generate_r419f78(struct gf100_gr *); extern const struct gf100_grctx_func gk110b_grctx; extern const struct gf100_grctx_func gk208_grctx; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c index 94233d0119df..52a234b1ef01 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk104.c @@ -906,7 +906,9 @@ static void gk104_grctx_generate_r419f78(struct gf100_gr *gr) { struct nvkm_device *device = gr->base.engine.subdev.device; - nvkm_mask(device, 0x419f78, 0x00000001, 0x00000000); + + /* bit 3 set disables loads in fp helper invocations, we need it enabled */ + nvkm_mask(device, 0x419f78, 0x00000009, 0x00000000); } void diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c index 4391458e1fb2..3acdd9eeb74a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110.c @@ -820,6 +820,15 @@ gk110_grctx_generate_r419eb0(struct gf100_gr *gr) nvkm_mask(device, 0x419eb0, 0x00001000, 0x00001000); } +void +gk110_grctx_generate_r419f78(struct gf100_gr *gr) +{ + struct nvkm_device *device = gr->base.engine.subdev.device; + + /* bit 3 set disables loads in fp helper invocations, we need it enabled */ + nvkm_mask(device, 0x419f78, 0x00000008, 0x00000000); +} + const struct gf100_grctx_func gk110_grctx = { .main = gf100_grctx_generate_main, @@ -854,4 +863,5 @@ gk110_grctx = { .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, .r418800 = gk104_grctx_generate_r418800, .r419eb0 = gk110_grctx_generate_r419eb0, + .r419f78 = gk110_grctx_generate_r419f78, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c index 7b9a34f9ec3c..5597e87624ac 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk110b.c @@ -103,4 +103,5 @@ gk110b_grctx = { .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, .r418800 = gk104_grctx_generate_r418800, .r419eb0 = gk110_grctx_generate_r419eb0, + .r419f78 = gk110_grctx_generate_r419f78, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c index c78d07a8bb7d..612656496541 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgk208.c @@ -568,4 +568,5 @@ gk208_grctx = { .dist_skip_table = gf117_grctx_generate_dist_skip_table, .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, .r418800 = gk104_grctx_generate_r418800, + .r419f78 = gk110_grctx_generate_r419f78, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c index beac66eb2a80..9906974ac3f0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm107.c @@ -988,4 +988,5 @@ gm107_grctx = { .r406500 = gm107_grctx_generate_r406500, .gpc_tpc_nr = gk104_grctx_generate_gpc_tpc_nr, .r419e00 = gm107_grctx_generate_r419e00, + .r419f78 = gk110_grctx_generate_r419f78, }; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c index 3b6c8100a242..a7775aa18541 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/tu102.c @@ -206,19 +206,6 @@ tu102_gr_av_to_init_veid(struct nvkm_blob *blob, struct gf100_gr_pack **ppack) return gk20a_gr_av_to_init_(blob, 64, 0x00100000, ppack); } -int -tu102_gr_load(struct gf100_gr *gr, int ver, const struct gf100_gr_fwif *fwif) -{ - int ret; - - ret = gm200_gr_load(gr, ver, fwif); - if (ret) - return ret; - - return gk20a_gr_load_net(gr, "gr/", "sw_veid_bundle_init", ver, tu102_gr_av_to_init_veid, - &gr->bundle_veid); -} - static const struct gf100_gr_fwif tu102_gr_fwif[] = { { 0, gm200_gr_load, &tu102_gr, &gp108_gr_fecs_acr, &gp108_gr_gpccs_acr }, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c index 976539de4220..731b2f68d3db 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/base.c @@ -260,10 +260,11 @@ nvkm_i2c_new_(const struct nvkm_i2c_func *func, struct nvkm_device *device, { struct nvkm_bios *bios = device->bios; struct nvkm_i2c *i2c; + struct nvkm_i2c_aux *aux; struct dcb_i2c_entry ccbE; struct dcb_output dcbE; u8 ver, hdr; - int ret, i; + int ret, i, ids; if (!(i2c = *pi2c = kzalloc(sizeof(*i2c), GFP_KERNEL))) return -ENOMEM; @@ -406,5 +407,11 @@ nvkm_i2c_new_(const struct nvkm_i2c_func *func, struct nvkm_device *device, } } - return nvkm_event_init(&nvkm_i2c_intr_func, &i2c->subdev, 4, i, &i2c->event); + ids = 0; + list_for_each_entry(aux, &i2c->aux, head) + ids = max(ids, aux->id + 1); + if (!ids) + return 0; + + return nvkm_event_init(&nvkm_i2c_intr_func, &i2c->subdev, 4, ids, &i2c->event); } diff --git a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c index 0583360b1280..ea5a85779382 100644 --- a/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c +++ b/drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c @@ -567,6 +567,7 @@ static const struct of_device_id s6d7aa0_of_match[] = { }, { /* sentinel */ } }; +MODULE_DEVICE_TABLE(of, s6d7aa0_of_match); static struct mipi_dsi_driver s6d7aa0_driver = { .probe = s6d7aa0_probe, diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c index bb89e6d047bc..857bc01591db 100644 --- a/drivers/gpu/drm/panel/panel-simple.c +++ b/drivers/gpu/drm/panel/panel-simple.c @@ -995,21 +995,21 @@ static const struct panel_desc auo_g104sn02 = { .connector_type = DRM_MODE_CONNECTOR_LVDS, }; -static const struct drm_display_mode auo_g121ean01_mode = { - .clock = 66700, - .hdisplay = 1280, - .hsync_start = 1280 + 58, - .hsync_end = 1280 + 58 + 8, - .htotal = 1280 + 58 + 8 + 70, - .vdisplay = 800, - .vsync_start = 800 + 6, - .vsync_end = 800 + 6 + 4, - .vtotal = 800 + 6 + 4 + 10, +static const struct display_timing auo_g121ean01_timing = { + .pixelclock = { 60000000, 74400000, 90000000 }, + .hactive = { 1280, 1280, 1280 }, + .hfront_porch = { 20, 50, 100 }, + .hback_porch = { 20, 50, 100 }, + .hsync_len = { 30, 100, 200 }, + .vactive = { 800, 800, 800 }, + .vfront_porch = { 2, 10, 25 }, + .vback_porch = { 2, 10, 25 }, + .vsync_len = { 4, 18, 50 }, }; static const struct panel_desc auo_g121ean01 = { - .modes = &auo_g121ean01_mode, - .num_modes = 1, + .timings = &auo_g121ean01_timing, + .num_timings = 1, .bpc = 8, .size = { .width = 261, @@ -2377,7 +2377,7 @@ static const struct panel_desc innolux_g121x1_l03 = { }; static const struct display_timing innolux_g156hce_l01_timings = { - .pixelclock = { 120000000, 144000000, 150000000 }, + .pixelclock = { 120000000, 141860000, 150000000 }, .hactive = { 1920, 1920, 1920 }, .hfront_porch = { 80, 90, 100 }, .hback_porch = { 80, 90, 100 }, diff --git a/drivers/gpu/drm/panfrost/panfrost_devfreq.c b/drivers/gpu/drm/panfrost/panfrost_devfreq.c index 58dfb15a8757..e78de99e9933 100644 --- a/drivers/gpu/drm/panfrost/panfrost_devfreq.c +++ b/drivers/gpu/drm/panfrost/panfrost_devfreq.c @@ -96,7 +96,7 @@ static int panfrost_read_speedbin(struct device *dev) * keep going without it; any other error means that we are * supposed to read the bin value, but we failed doing so. */ - if (ret != -ENOENT) { + if (ret != -ENOENT && ret != -EOPNOTSUPP) { DRM_DEV_ERROR(dev, "Cannot read speed-bin (%d).", ret); return ret; } diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h index ea993d7162e8..307a890fde13 100644 --- a/drivers/gpu/drm/qxl/qxl_drv.h +++ b/drivers/gpu/drm/qxl/qxl_drv.h @@ -310,7 +310,7 @@ int qxl_gem_object_create_with_handle(struct qxl_device *qdev, u32 domain, size_t size, struct qxl_surface *surf, - struct qxl_bo **qobj, + struct drm_gem_object **gobj, uint32_t *handle); void qxl_gem_object_free(struct drm_gem_object *gobj); int qxl_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_priv); diff --git a/drivers/gpu/drm/qxl/qxl_dumb.c b/drivers/gpu/drm/qxl/qxl_dumb.c index d636ba685451..17df5c7ccf69 100644 --- a/drivers/gpu/drm/qxl/qxl_dumb.c +++ b/drivers/gpu/drm/qxl/qxl_dumb.c @@ -34,6 +34,7 @@ int qxl_mode_dumb_create(struct drm_file *file_priv, { struct qxl_device *qdev = to_qxl(dev); struct qxl_bo *qobj; + struct drm_gem_object *gobj; uint32_t handle; int r; struct qxl_surface surf; @@ -62,11 +63,13 @@ int qxl_mode_dumb_create(struct drm_file *file_priv, r = qxl_gem_object_create_with_handle(qdev, file_priv, QXL_GEM_DOMAIN_CPU, - args->size, &surf, &qobj, + args->size, &surf, &gobj, &handle); if (r) return r; + qobj = gem_to_qxl_bo(gobj); qobj->is_dumb = true; + drm_gem_object_put(gobj); args->pitch = pitch; args->handle = handle; return 0; diff --git a/drivers/gpu/drm/qxl/qxl_gem.c b/drivers/gpu/drm/qxl/qxl_gem.c index a08da0bd9098..fc5e3763c359 100644 --- a/drivers/gpu/drm/qxl/qxl_gem.c +++ b/drivers/gpu/drm/qxl/qxl_gem.c @@ -72,32 +72,41 @@ int qxl_gem_object_create(struct qxl_device *qdev, int size, return 0; } +/* + * If the caller passed a valid gobj pointer, it is responsible to call + * drm_gem_object_put() when it no longer needs to acess the object. + * + * If gobj is NULL, it is handled internally. + */ int qxl_gem_object_create_with_handle(struct qxl_device *qdev, struct drm_file *file_priv, u32 domain, size_t size, struct qxl_surface *surf, - struct qxl_bo **qobj, + struct drm_gem_object **gobj, uint32_t *handle) { - struct drm_gem_object *gobj; int r; + struct drm_gem_object *local_gobj; - BUG_ON(!qobj); BUG_ON(!handle); r = qxl_gem_object_create(qdev, size, 0, domain, false, false, surf, - &gobj); + &local_gobj); if (r) return -ENOMEM; - r = drm_gem_handle_create(file_priv, gobj, handle); + r = drm_gem_handle_create(file_priv, local_gobj, handle); if (r) return r; - /* drop reference from allocate - handle holds it now */ - *qobj = gem_to_qxl_bo(gobj); - drm_gem_object_put(gobj); + + if (gobj) + *gobj = local_gobj; + else + /* drop reference from allocate - handle holds it now */ + drm_gem_object_put(local_gobj); + return 0; } diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c index 30f58b21372a..dd0f834d881c 100644 --- a/drivers/gpu/drm/qxl/qxl_ioctl.c +++ b/drivers/gpu/drm/qxl/qxl_ioctl.c @@ -38,7 +38,6 @@ int qxl_alloc_ioctl(struct drm_device *dev, void *data, struct drm_file *file_pr struct qxl_device *qdev = to_qxl(dev); struct drm_qxl_alloc *qxl_alloc = data; int ret; - struct qxl_bo *qobj; uint32_t handle; u32 domain = QXL_GEM_DOMAIN_VRAM; @@ -50,7 +49,7 @@ int qxl_alloc_ioctl(struct drm_device *dev, void *data, struct drm_file *file_pr domain, qxl_alloc->size, NULL, - &qobj, &handle); + NULL, &handle); if (ret) { DRM_ERROR("%s: failed to create gem ret=%d\n", __func__, ret); @@ -386,7 +385,6 @@ int qxl_alloc_surf_ioctl(struct drm_device *dev, void *data, struct drm_file *fi { struct qxl_device *qdev = to_qxl(dev); struct drm_qxl_alloc_surf *param = data; - struct qxl_bo *qobj; int handle; int ret; int size, actual_stride; @@ -406,7 +404,7 @@ int qxl_alloc_surf_ioctl(struct drm_device *dev, void *data, struct drm_file *fi QXL_GEM_DOMAIN_SURFACE, size, &surf, - &qobj, &handle); + NULL, &handle); if (ret) { DRM_ERROR("%s: failed to create gem ret=%d\n", __func__, ret); diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c index c1bbfbe28bda..ceb6d772ef94 100644 --- a/drivers/gpu/drm/radeon/atom.c +++ b/drivers/gpu/drm/radeon/atom.c @@ -68,8 +68,8 @@ typedef struct { } atom_exec_context; int atom_debug = 0; -static int atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t * params); -int atom_execute_table(struct atom_context *ctx, int index, uint32_t * params); +static int atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t *params); +int atom_execute_table(struct atom_context *ctx, int index, uint32_t *params); static uint32_t atom_arg_mask[8] = { 0xFFFFFFFF, 0x0000FFFF, 0x00FFFF00, 0xFFFF0000, @@ -163,13 +163,9 @@ static uint32_t atom_iio_execute(struct atom_context *ctx, int base, ~((0xFFFFFFFF >> (32 - CU8(base + 1))) << CU8(base + 3)); temp |= - ((ctx-> - io_attr >> CU8(base + 2)) & (0xFFFFFFFF >> (32 - - CU8 - (base - + - 1)))) - << CU8(base + 3); + ((ctx->io_attr >> CU8(base + 2)) & + (0xFFFFFFFF >> (32 - CU8(base + 1)))) << + CU8(base + 3); base += 4; break; case ATOM_IIO_END: @@ -1156,7 +1152,7 @@ static struct { atom_op_shr, ATOM_ARG_MC}, { atom_op_debug, 0},}; -static int atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t * params) +static int atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t *params) { int base = CU16(ctx->cmd_table + 4 + 2 * index); int len, ws, ps, ptr; @@ -1216,7 +1212,7 @@ free: return ret; } -int atom_execute_table_scratch_unlocked(struct atom_context *ctx, int index, uint32_t * params) +int atom_execute_table_scratch_unlocked(struct atom_context *ctx, int index, uint32_t *params) { int r; @@ -1237,7 +1233,7 @@ int atom_execute_table_scratch_unlocked(struct atom_context *ctx, int index, uin return r; } -int atom_execute_table(struct atom_context *ctx, int index, uint32_t * params) +int atom_execute_table(struct atom_context *ctx, int index, uint32_t *params) { int r; mutex_lock(&ctx->scratch_mutex); @@ -1359,8 +1355,8 @@ void atom_destroy(struct atom_context *ctx) } bool atom_parse_data_header(struct atom_context *ctx, int index, - uint16_t * size, uint8_t * frev, uint8_t * crev, - uint16_t * data_start) + uint16_t *size, uint8_t *frev, uint8_t *crev, + uint16_t *data_start) { int offset = index * 2 + 4; int idx = CU16(ctx->data_table + offset); @@ -1379,8 +1375,8 @@ bool atom_parse_data_header(struct atom_context *ctx, int index, return true; } -bool atom_parse_cmd_header(struct atom_context *ctx, int index, uint8_t * frev, - uint8_t * crev) +bool atom_parse_cmd_header(struct atom_context *ctx, int index, uint8_t *frev, + uint8_t *crev) { int offset = index * 2 + 4; int idx = CU16(ctx->cmd_table + offset); diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c index 5d6b81a6578e..10be30366c2b 100644 --- a/drivers/gpu/drm/radeon/cik.c +++ b/drivers/gpu/drm/radeon/cik.c @@ -9534,17 +9534,8 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &bridge_cfg); - pcie_capability_read_word(rdev->pdev, PCI_EXP_LNKCTL, - &gpu_cfg); - - tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); - - tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(rdev->pdev, PCI_EXP_LNKCTL, - tmp16); + pcie_capability_set_word(root, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); + pcie_capability_set_word(rdev->pdev, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); tmp = RREG32_PCIE_PORT(PCIE_LC_STATUS1); max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT; @@ -9591,21 +9582,14 @@ static void cik_pcie_gen3_enable(struct radeon_device *rdev) msleep(100); /* linkctl */ - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(root, PCI_EXP_LNKCTL, - tmp16); - - pcie_capability_read_word(rdev->pdev, - PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(rdev->pdev, - PCI_EXP_LNKCTL, - tmp16); + pcie_capability_clear_and_set_word(root, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + bridge_cfg & + PCI_EXP_LNKCTL_HAWD); + pcie_capability_clear_and_set_word(rdev->pdev, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + gpu_cfg & + PCI_EXP_LNKCTL_HAWD); /* linkctl2 */ pcie_capability_read_word(root, PCI_EXP_LNKCTL2, diff --git a/drivers/gpu/drm/radeon/clearstate_si.h b/drivers/gpu/drm/radeon/clearstate_si.h index 356219c6c7f2..7da8418704fe 100644 --- a/drivers/gpu/drm/radeon/clearstate_si.h +++ b/drivers/gpu/drm/radeon/clearstate_si.h @@ -23,8 +23,7 @@ #include "clearstate_defs.h" -static const u32 si_SECT_CONTEXT_def_1[] = -{ +static const u32 si_SECT_CONTEXT_def_1[] = { 0x00000000, // DB_RENDER_CONTROL 0x00000000, // DB_COUNT_CONTROL 0x00000000, // DB_DEPTH_VIEW diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index 9c1a92fa2af6..25201b9a5aae 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -249,7 +249,7 @@ void r300_ring_start(struct radeon_device *rdev, struct radeon_ring *ring) /* Sub pixel 1/12 so we can have 4K rendering according to doc */ gb_tile_config = (R300_ENABLE_TILING | R300_TILE_SIZE_16); - switch(rdev->num_gb_pipes) { + switch (rdev->num_gb_pipes) { case 2: gb_tile_config |= R300_PIPE_COUNT_R300; break; @@ -638,7 +638,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p, track = (struct r100_cs_track *)p->track; idx_value = radeon_get_ib_value(p, idx); - switch(reg) { + switch (reg) { case AVIVO_D1MODE_VLINE_START_END: case RADEON_CRTC_GUI_TRIG_VLINE: r = r100_cs_packet_parse_vline(p); @@ -1180,7 +1180,7 @@ static int r300_packet3_check(struct radeon_cs_parser *p, ib = p->ib.ptr; idx = pkt->idx + 1; track = (struct r100_cs_track *)p->track; - switch(pkt->opcode) { + switch (pkt->opcode) { case PACKET3_3D_LOAD_VBPNTR: r = r100_packet3_load_vbpntr(p, pkt, idx); if (r) diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c index 5771d1fcb073..603a78e41ba5 100644 --- a/drivers/gpu/drm/radeon/radeon_acpi.c +++ b/drivers/gpu/drm/radeon/radeon_acpi.c @@ -618,7 +618,7 @@ int radeon_acpi_pcie_performance_request(struct radeon_device *rdev, atcs_input.size = sizeof(struct atcs_pref_req_input); /* client id (bit 2-0: func num, 7-3: dev num, 15-8: bus num) */ - atcs_input.client_id = rdev->pdev->devfn | (rdev->pdev->bus->number << 8); + atcs_input.client_id = pci_dev_id(rdev->pdev); atcs_input.valid_flags_mask = ATCS_VALID_FLAGS_MASK; atcs_input.flags = ATCS_WAIT_FOR_COMPLETION; if (advertise) diff --git a/drivers/gpu/drm/radeon/radeon_agp.c b/drivers/gpu/drm/radeon/radeon_agp.c index d124600b5f58..a3d749e350f9 100644 --- a/drivers/gpu/drm/radeon/radeon_agp.c +++ b/drivers/gpu/drm/radeon/radeon_agp.c @@ -130,7 +130,7 @@ static struct radeon_agpmode_quirk radeon_agpmode_quirk_list[] = { struct radeon_agp_head *radeon_agp_head_init(struct drm_device *dev) { struct pci_dev *pdev = to_pci_dev(dev->dev); - struct radeon_agp_head *head = NULL; + struct radeon_agp_head *head; head = kzalloc(sizeof(*head), GFP_KERNEL); if (!head) diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index bf3c411a55c5..85c4bb186203 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -1389,7 +1389,7 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev, num_indices = (size - sizeof(ATOM_COMMON_TABLE_HEADER)) / sizeof(ATOM_SPREAD_SPECTRUM_ASSIGNMENT); - ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT*) + ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT *) ((u8 *)&ss_info->asSS_Info[0]); for (i = 0; i < num_indices; i++) { if (ss_assign->ucSS_Id == id) { @@ -1402,7 +1402,7 @@ bool radeon_atombios_get_ppll_ss_info(struct radeon_device *rdev, ss->refdiv = ss_assign->ucRecommendedRef_Div; return true; } - ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT*) + ss_assign = (struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT *) ((u8 *)ss_assign + sizeof(struct _ATOM_SPREAD_SPECTRUM_ASSIGNMENT)); } } @@ -3406,7 +3406,7 @@ static ATOM_VOLTAGE_OBJECT_V2 *atom_lookup_voltage_object_v2(ATOM_VOLTAGE_OBJECT { u32 size = le16_to_cpu(v2->sHeader.usStructureSize); u32 offset = offsetof(ATOM_VOLTAGE_OBJECT_INFO_V2, asVoltageObj[0]); - u8 *start = (u8*)v2; + u8 *start = (u8 *)v2; while (offset < size) { ATOM_VOLTAGE_OBJECT_V2 *vo = (ATOM_VOLTAGE_OBJECT_V2 *)(start + offset); @@ -3423,7 +3423,7 @@ static ATOM_VOLTAGE_OBJECT_V3 *atom_lookup_voltage_object_v3(ATOM_VOLTAGE_OBJECT { u32 size = le16_to_cpu(v3->sHeader.usStructureSize); u32 offset = offsetof(ATOM_VOLTAGE_OBJECT_INFO_V3_1, asVoltageObj[0]); - u8 *start = (u8*)v3; + u8 *start = (u8 *)v3; while (offset < size) { ATOM_VOLTAGE_OBJECT_V3 *vo = (ATOM_VOLTAGE_OBJECT_V3 *)(start + offset); diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c index d0b450a06506..595354e3ce0b 100644 --- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c +++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c @@ -115,7 +115,7 @@ static union acpi_object *radeon_atpx_call(acpi_handle handle, int function, /* Fail only if calling the method fails and ATPX is supported */ if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) { - printk("failed to evaluate ATPX got %s\n", + pr_err("failed to evaluate ATPX got %s\n", acpi_format_exception(status)); kfree(buffer.pointer); return NULL; @@ -171,7 +171,7 @@ static int radeon_atpx_validate(struct radeon_atpx *atpx) size = *(u16 *) info->buffer.pointer; if (size < 10) { - printk("ATPX buffer is too small: %zu\n", size); + pr_err("ATPX buffer is too small: %zu\n", size); kfree(info); return -EINVAL; } @@ -202,7 +202,7 @@ static int radeon_atpx_validate(struct radeon_atpx *atpx) atpx->is_hybrid = false; if (valid_bits & ATPX_MS_HYBRID_GFX_SUPPORTED) { - printk("ATPX Hybrid Graphics\n"); + pr_info("ATPX Hybrid Graphics\n"); /* * Disable legacy PM methods only when pcie port PM is usable, * otherwise the device might fail to power off or power on. @@ -239,7 +239,7 @@ static int radeon_atpx_verify_interface(struct radeon_atpx *atpx) size = *(u16 *) info->buffer.pointer; if (size < 8) { - printk("ATPX buffer is too small: %zu\n", size); + pr_err("ATPX buffer is too small: %zu\n", size); err = -EINVAL; goto out; } @@ -248,8 +248,8 @@ static int radeon_atpx_verify_interface(struct radeon_atpx *atpx) memcpy(&output, info->buffer.pointer, size); /* TODO: check version? */ - printk("ATPX version %u, functions 0x%08x\n", - output.version, output.function_bits); + pr_info("ATPX version %u, functions 0x%08x\n", + output.version, output.function_bits); radeon_atpx_parse_functions(&atpx->functions, output.function_bits); diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index 795c3667f6d6..2620efc7c675 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -863,7 +863,7 @@ struct radeon_encoder_primary_dac *radeon_combios_get_primary_dac_info(struct struct radeon_device *rdev = dev->dev_private; uint16_t dac_info; uint8_t rev, bg, dac; - struct radeon_encoder_primary_dac *p_dac = NULL; + struct radeon_encoder_primary_dac *p_dac; int found = 0; p_dac = kzalloc(sizeof(struct radeon_encoder_primary_dac), @@ -1014,7 +1014,7 @@ struct radeon_encoder_tv_dac *radeon_combios_get_tv_dac_info(struct struct radeon_device *rdev = dev->dev_private; uint16_t dac_info; uint8_t rev, bg, dac; - struct radeon_encoder_tv_dac *tv_dac = NULL; + struct radeon_encoder_tv_dac *tv_dac; int found = 0; tv_dac = kzalloc(sizeof(struct radeon_encoder_tv_dac), GFP_KERNEL); @@ -1100,7 +1100,7 @@ static struct radeon_encoder_lvds *radeon_legacy_get_lvds_info_from_regs(struct radeon_device *rdev) { - struct radeon_encoder_lvds *lvds = NULL; + struct radeon_encoder_lvds *lvds; uint32_t fp_vert_stretch, fp_horz_stretch; uint32_t ppll_div_sel, ppll_val; uint32_t lvds_ss_gen_cntl = RREG32(RADEON_LVDS_SS_GEN_CNTL); diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 07193cd0c417..d2f02c3dfce2 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -198,8 +198,7 @@ int radeon_get_monitor_bpc(struct drm_connector *connector) DRM_DEBUG("%s: HDMI deep color 10 bpc exceeds max tmds clock. Using %d bpc.\n", connector->name, bpc); } - } - else if (bpc > 8) { + } else if (bpc > 8) { /* max_tmds_clock missing, but hdmi spec mandates it for deep color. */ DRM_DEBUG("%s: Required max tmds clock for HDMI deep color missing. Using 8 bpc.\n", connector->name); @@ -334,10 +333,8 @@ static void radeon_connector_free_edid(struct drm_connector *connector) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); - if (radeon_connector->edid) { - kfree(radeon_connector->edid); - radeon_connector->edid = NULL; - } + kfree(radeon_connector->edid); + radeon_connector->edid = NULL; } static int radeon_ddc_get_modes(struct drm_connector *connector) @@ -1372,7 +1369,7 @@ radeon_dvi_detect(struct drm_connector *connector, bool force) /* assume digital unless load detected otherwise */ radeon_connector->use_digital = true; lret = encoder_funcs->detect(encoder, connector); - DRM_DEBUG_KMS("load_detect %x returned: %x\n",encoder->encoder_type,lret); + DRM_DEBUG_KMS("load_detect %x returned: %x\n", encoder->encoder_type, lret); if (lret == connector_status_connected) radeon_connector->use_digital = false; } diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 39cdede460b5..fa531493b111 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -113,59 +113,32 @@ #define KMS_DRIVER_MAJOR 2 #define KMS_DRIVER_MINOR 50 #define KMS_DRIVER_PATCHLEVEL 0 -int radeon_suspend_kms(struct drm_device *dev, bool suspend, - bool fbcon, bool freeze); -int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon); -extern int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int crtc, - unsigned int flags, int *vpos, int *hpos, - ktime_t *stime, ktime_t *etime, - const struct drm_display_mode *mode); -extern bool radeon_is_px(struct drm_device *dev); -int radeon_mode_dumb_mmap(struct drm_file *filp, - struct drm_device *dev, - uint32_t handle, uint64_t *offset_p); -int radeon_mode_dumb_create(struct drm_file *file_priv, - struct drm_device *dev, - struct drm_mode_create_dumb *args); - -/* atpx handler */ -#if defined(CONFIG_VGA_SWITCHEROO) -void radeon_register_atpx_handler(void); -void radeon_unregister_atpx_handler(void); -bool radeon_has_atpx_dgpu_power_cntl(void); -bool radeon_is_atpx_hybrid(void); -#else -static inline void radeon_register_atpx_handler(void) {} -static inline void radeon_unregister_atpx_handler(void) {} -static inline bool radeon_has_atpx_dgpu_power_cntl(void) { return false; } -static inline bool radeon_is_atpx_hybrid(void) { return false; } -#endif int radeon_no_wb; int radeon_modeset = -1; int radeon_dynclks = -1; -int radeon_r4xx_atom = 0; +int radeon_r4xx_atom; int radeon_agpmode = -1; -int radeon_vram_limit = 0; +int radeon_vram_limit; int radeon_gart_size = -1; /* auto */ -int radeon_benchmarking = 0; -int radeon_testing = 0; -int radeon_connector_table = 0; +int radeon_benchmarking; +int radeon_testing; +int radeon_connector_table; int radeon_tv = 1; int radeon_audio = -1; -int radeon_disp_priority = 0; -int radeon_hw_i2c = 0; +int radeon_disp_priority; +int radeon_hw_i2c; int radeon_pcie_gen2 = -1; int radeon_msi = -1; int radeon_lockup_timeout = 10000; -int radeon_fastfb = 0; +int radeon_fastfb; int radeon_dpm = -1; int radeon_aspm = -1; int radeon_runtime_pm = -1; -int radeon_hard_reset = 0; +int radeon_hard_reset; int radeon_vm_size = 8; int radeon_vm_block_size = -1; -int radeon_deep_color = 0; +int radeon_deep_color; int radeon_use_pflipirq = 2; int radeon_bapm = -1; int radeon_backlight = -1; @@ -384,6 +357,7 @@ radeon_pci_shutdown(struct pci_dev *pdev) static int radeon_pmops_suspend(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); + return radeon_suspend_kms(drm_dev, true, true, false); } @@ -404,12 +378,14 @@ static int radeon_pmops_resume(struct device *dev) static int radeon_pmops_freeze(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); + return radeon_suspend_kms(drm_dev, false, true, true); } static int radeon_pmops_thaw(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); + return radeon_resume_kms(drm_dev, false, true); } @@ -494,6 +470,7 @@ long radeon_drm_ioctl(struct file *filp, struct drm_file *file_priv = filp->private_data; struct drm_device *dev; long ret; + dev = file_priv->minor->dev; ret = pm_runtime_get_sync(dev->dev); if (ret < 0) { diff --git a/drivers/gpu/drm/radeon/radeon_drv.h b/drivers/gpu/drm/radeon/radeon_drv.h index 2ffe0975ee54..34a1c73d3938 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.h +++ b/drivers/gpu/drm/radeon/radeon_drv.h @@ -124,4 +124,17 @@ int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv); void radeon_driver_postclose_kms(struct drm_device *dev, struct drm_file *file_priv); +/* atpx handler */ +#if defined(CONFIG_VGA_SWITCHEROO) +void radeon_register_atpx_handler(void); +void radeon_unregister_atpx_handler(void); +bool radeon_has_atpx_dgpu_power_cntl(void); +bool radeon_is_atpx_hybrid(void); +#else +static inline void radeon_register_atpx_handler(void) {} +static inline void radeon_unregister_atpx_handler(void) {} +static inline bool radeon_has_atpx_dgpu_power_cntl(void) { return false; } +static inline bool radeon_is_atpx_hybrid(void) { return false; } +#endif + #endif /* __RADEON_DRV_H__ */ diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c index b3518a8f95a0..9cb6401fe97e 100644 --- a/drivers/gpu/drm/radeon/radeon_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_encoders.c @@ -58,6 +58,7 @@ static uint32_t radeon_encoder_clones(struct drm_encoder *encoder) count = -1; list_for_each_entry(clone_encoder, &dev->mode_config.encoder_list, head) { struct radeon_encoder *radeon_clone = to_radeon_encoder(clone_encoder); + count++; if (clone_encoder == encoder) @@ -108,9 +109,10 @@ radeon_get_encoder_enum(struct drm_device *dev, uint32_t supported_device, uint8 if (ASIC_IS_AVIVO(rdev)) ret = ENCODER_INTERNAL_KLDSCP_DAC2_ENUM_ID1; else { - /*if (rdev->family == CHIP_R200) - ret = ENCODER_INTERNAL_DVO1_ENUM_ID1; - else*/ + /* if (rdev->family == CHIP_R200) + * ret = ENCODER_INTERNAL_DVO1_ENUM_ID1; + * else + */ ret = ENCODER_INTERNAL_DAC2_ENUM_ID1; } break; @@ -234,6 +236,7 @@ void radeon_encoder_set_active_device(struct drm_encoder *encoder) list_for_each_entry(connector, &dev->mode_config.connector_list, head) { if (connector->encoder == encoder) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); + radeon_encoder->active_device = radeon_encoder->devices & radeon_connector->devices; DRM_DEBUG_KMS("setting active device to %08x from %08x %08x for encoder %d\n", radeon_encoder->active_device, radeon_encoder->devices, @@ -320,12 +323,12 @@ void radeon_panel_mode_fixup(struct drm_encoder *encoder, struct drm_device *dev = encoder->dev; struct radeon_device *rdev = dev->dev_private; struct drm_display_mode *native_mode = &radeon_encoder->native_mode; - unsigned hblank = native_mode->htotal - native_mode->hdisplay; - unsigned vblank = native_mode->vtotal - native_mode->vdisplay; - unsigned hover = native_mode->hsync_start - native_mode->hdisplay; - unsigned vover = native_mode->vsync_start - native_mode->vdisplay; - unsigned hsync_width = native_mode->hsync_end - native_mode->hsync_start; - unsigned vsync_width = native_mode->vsync_end - native_mode->vsync_start; + unsigned int hblank = native_mode->htotal - native_mode->hdisplay; + unsigned int vblank = native_mode->vtotal - native_mode->vdisplay; + unsigned int hover = native_mode->hsync_start - native_mode->hdisplay; + unsigned int vover = native_mode->vsync_start - native_mode->vdisplay; + unsigned int hsync_width = native_mode->hsync_end - native_mode->hsync_start; + unsigned int vsync_width = native_mode->vsync_end - native_mode->vsync_start; adjusted_mode->clock = native_mode->clock; adjusted_mode->flags = native_mode->flags; @@ -424,6 +427,7 @@ bool radeon_dig_monitor_is_duallink(struct drm_encoder *encoder, bool radeon_encoder_is_digital(struct drm_encoder *encoder) { struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); + switch (radeon_encoder->encoder_id) { case ENCODER_OBJECT_ID_INTERNAL_LVDS: case ENCODER_OBJECT_ID_INTERNAL_TMDS1: diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index 04109a2a6fd7..4bb242437ff6 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -74,9 +74,9 @@ int radeon_gart_table_ram_alloc(struct radeon_device *rdev) ptr = dma_alloc_coherent(&rdev->pdev->dev, rdev->gart.table_size, &rdev->gart.table_addr, GFP_KERNEL); - if (ptr == NULL) { + if (!ptr) return -ENOMEM; - } + #ifdef CONFIG_X86 if (rdev->family == CHIP_RS400 || rdev->family == CHIP_RS480 || rdev->family == CHIP_RS690 || rdev->family == CHIP_RS740) { @@ -99,9 +99,9 @@ int radeon_gart_table_ram_alloc(struct radeon_device *rdev) */ void radeon_gart_table_ram_free(struct radeon_device *rdev) { - if (rdev->gart.ptr == NULL) { + if (!rdev->gart.ptr) return; - } + #ifdef CONFIG_X86 if (rdev->family == CHIP_RS400 || rdev->family == CHIP_RS480 || rdev->family == CHIP_RS690 || rdev->family == CHIP_RS740) { @@ -133,9 +133,8 @@ int radeon_gart_table_vram_alloc(struct radeon_device *rdev) r = radeon_bo_create(rdev, rdev->gart.table_size, PAGE_SIZE, true, RADEON_GEM_DOMAIN_VRAM, 0, NULL, NULL, &rdev->gart.robj); - if (r) { + if (r) return r; - } } return 0; } @@ -197,9 +196,9 @@ void radeon_gart_table_vram_unpin(struct radeon_device *rdev) { int r; - if (rdev->gart.robj == NULL) { + if (!rdev->gart.robj) return; - } + r = radeon_bo_reserve(rdev->gart.robj, false); if (likely(r == 0)) { radeon_bo_kunmap(rdev->gart.robj); @@ -220,9 +219,9 @@ void radeon_gart_table_vram_unpin(struct radeon_device *rdev) */ void radeon_gart_table_vram_free(struct radeon_device *rdev) { - if (rdev->gart.robj == NULL) { + if (!rdev->gart.robj) return; - } + radeon_bo_unref(&rdev->gart.robj); } @@ -239,11 +238,10 @@ void radeon_gart_table_vram_free(struct radeon_device *rdev) * Unbinds the requested pages from the gart page table and * replaces them with the dummy page (all asics). */ -void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset, +void radeon_gart_unbind(struct radeon_device *rdev, unsigned int offset, int pages) { - unsigned t; - unsigned p; + unsigned int t, p; int i, j; if (!rdev->gart.ready) { @@ -284,12 +282,11 @@ void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset, * (all asics). * Returns 0 for success, -EINVAL for failure. */ -int radeon_gart_bind(struct radeon_device *rdev, unsigned offset, +int radeon_gart_bind(struct radeon_device *rdev, unsigned int offset, int pages, struct page **pagelist, dma_addr_t *dma_addr, uint32_t flags) { - unsigned t; - unsigned p; + unsigned int t, p; uint64_t page_base, page_entry; int i, j; @@ -307,9 +304,9 @@ int radeon_gart_bind(struct radeon_device *rdev, unsigned offset, for (j = 0; j < (PAGE_SIZE / RADEON_GPU_PAGE_SIZE); j++, t++) { page_entry = radeon_gart_get_page_entry(page_base, flags); rdev->gart.pages_entry[t] = page_entry; - if (rdev->gart.ptr) { + if (rdev->gart.ptr) radeon_gart_set_page(rdev, t, page_entry); - } + page_base += RADEON_GPU_PAGE_SIZE; } } @@ -332,9 +329,9 @@ int radeon_gart_init(struct radeon_device *rdev) { int r, i; - if (rdev->gart.pages) { + if (rdev->gart.pages) return 0; - } + /* We need PAGE_SIZE >= RADEON_GPU_PAGE_SIZE */ if (PAGE_SIZE < RADEON_GPU_PAGE_SIZE) { DRM_ERROR("Page size is smaller than GPU page size!\n"); diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index d0119c5f7eb3..358d19242f4b 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -316,7 +316,7 @@ int radeon_gem_pread_ioctl(struct drm_device *dev, void *data, { /* TODO: implement */ DRM_ERROR("unimplemented %s\n", __func__); - return -ENOSYS; + return -EOPNOTSUPP; } int radeon_gem_pwrite_ioctl(struct drm_device *dev, void *data, @@ -324,7 +324,7 @@ int radeon_gem_pwrite_ioctl(struct drm_device *dev, void *data, { /* TODO: implement */ DRM_ERROR("unimplemented %s\n", __func__); - return -ENOSYS; + return -EOPNOTSUPP; } int radeon_gem_create_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index e0214cf1b43b..a16590c6247f 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -444,7 +444,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) DRM_DEBUG_KMS("timestamp is r6xx+ only!\n"); return -EINVAL; } - value = (uint32_t*)&value64; + value = (uint32_t *)&value64; value_size = sizeof(uint64_t); value64 = radeon_get_gpu_clock_counter(rdev); break; @@ -543,18 +543,18 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) *value = rdev->vce.fb_version; break; case RADEON_INFO_NUM_BYTES_MOVED: - value = (uint32_t*)&value64; + value = (uint32_t *)&value64; value_size = sizeof(uint64_t); value64 = atomic64_read(&rdev->num_bytes_moved); break; case RADEON_INFO_VRAM_USAGE: - value = (uint32_t*)&value64; + value = (uint32_t *)&value64; value_size = sizeof(uint64_t); man = ttm_manager_type(&rdev->mman.bdev, TTM_PL_VRAM); value64 = ttm_resource_manager_usage(man); break; case RADEON_INFO_GTT_USAGE: - value = (uint32_t*)&value64; + value = (uint32_t *)&value64; value_size = sizeof(uint64_t); man = ttm_manager_type(&rdev->mman.bdev, TTM_PL_TT); value64 = ttm_resource_manager_usage(man); @@ -614,7 +614,7 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) DRM_DEBUG_KMS("Invalid request %d\n", info->request); return -EINVAL; } - if (copy_to_user(value_ptr, (char*)value, value_size)) { + if (copy_to_user(value_ptr, (char *)value, value_size)) { DRM_ERROR("copy_to_user %s:%u\n", __func__, __LINE__); return -EFAULT; } diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c index 601d35d34eab..c4350ac2b3d2 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c @@ -1692,7 +1692,7 @@ static struct radeon_encoder_int_tmds *radeon_legacy_get_tmds_info(struct radeon { struct drm_device *dev = encoder->base.dev; struct radeon_device *rdev = dev->dev_private; - struct radeon_encoder_int_tmds *tmds = NULL; + struct radeon_encoder_int_tmds *tmds; bool ret; tmds = kzalloc(sizeof(struct radeon_encoder_int_tmds), GFP_KERNEL); @@ -1715,7 +1715,7 @@ static struct radeon_encoder_ext_tmds *radeon_legacy_get_ext_tmds_info(struct ra { struct drm_device *dev = encoder->base.dev; struct radeon_device *rdev = dev->dev_private; - struct radeon_encoder_ext_tmds *tmds = NULL; + struct radeon_encoder_ext_tmds *tmds; bool ret; if (rdev->is_atom_bios) diff --git a/drivers/gpu/drm/radeon/radeon_legacy_tv.c b/drivers/gpu/drm/radeon/radeon_legacy_tv.c index 12e180b119ac..7883e9ec0bae 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_tv.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_tv.c @@ -724,12 +724,14 @@ void radeon_legacy_tv_mode_set(struct drm_encoder *encoder, } for (i = 0; i < MAX_H_CODE_TIMING_LEN; i++) { - if ((tv_dac->tv.h_code_timing[i] = hor_timing[i]) == 0) + tv_dac->tv.h_code_timing[i] = hor_timing[i]; + if (tv_dac->tv.h_code_timing[i] == 0) break; } for (i = 0; i < MAX_V_CODE_TIMING_LEN; i++) { - if ((tv_dac->tv.v_code_timing[i] = vert_timing[i]) == 0) + tv_dac->tv.v_code_timing[i] = vert_timing[i]; + if (tv_dac->tv.v_code_timing[i] == 0) break; } diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c index c87a57c9c592..22dd8b445685 100644 --- a/drivers/gpu/drm/radeon/radeon_sa.c +++ b/drivers/gpu/drm/radeon/radeon_sa.c @@ -123,7 +123,7 @@ int radeon_sa_bo_new(struct radeon_sa_manager *sa_manager, unsigned int size, unsigned int align) { struct drm_suballoc *sa = drm_suballoc_new(&sa_manager->base, size, - GFP_KERNEL, true, align); + GFP_KERNEL, false, align); if (IS_ERR(sa)) { *sa_bo = NULL; diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c index a5e1d2139e80..c9fef9b61ced 100644 --- a/drivers/gpu/drm/radeon/radeon_test.c +++ b/drivers/gpu/drm/radeon/radeon_test.c @@ -156,10 +156,10 @@ static void radeon_do_test_moves(struct radeon_device *rdev, int flag) i, *vram_start, gtt_start, (unsigned long long) (gtt_addr - rdev->mc.gtt_start + - (void*)gtt_start - gtt_map), + (void *)gtt_start - gtt_map), (unsigned long long) (vram_addr - rdev->mc.vram_start + - (void*)gtt_start - gtt_map)); + (void *)gtt_start - gtt_map)); radeon_bo_kunmap(vram_obj); goto out_lclean_unpin; } @@ -207,10 +207,10 @@ static void radeon_do_test_moves(struct radeon_device *rdev, int flag) i, *gtt_start, vram_start, (unsigned long long) (vram_addr - rdev->mc.vram_start + - (void*)vram_start - vram_map), + (void *)vram_start - vram_map), (unsigned long long) (gtt_addr - rdev->mc.gtt_start + - (void*)vram_start - vram_map)); + (void *)vram_start - vram_map)); radeon_bo_kunmap(gtt_obj[i]); goto out_lclean_unpin; } diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index ca4a36464340..d1871af967d4 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -95,7 +95,7 @@ int radeon_vce_init(struct radeon_device *rdev) size = rdev->vce_fw->size - strlen(fw_version) - 9; c = rdev->vce_fw->data; - for (;size > 0; --size, ++c) + for (; size > 0; --size, ++c) if (strncmp(c, fw_version, strlen(fw_version)) == 0) break; @@ -110,7 +110,7 @@ int radeon_vce_init(struct radeon_device *rdev) size = rdev->vce_fw->size - strlen(fb_version) - 3; c = rdev->vce_fw->data; - for (;size > 0; --size, ++c) + for (; size > 0; --size, ++c) if (strncmp(c, fb_version, strlen(fb_version)) == 0) break; diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index 26fa9b095514..9ce12fa3c356 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -136,8 +136,7 @@ int rv770_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk) return 0; } -static const u32 r7xx_golden_registers[] = -{ +static const u32 r7xx_golden_registers[] = { 0x8d00, 0xffffffff, 0x0e0e0074, 0x8d04, 0xffffffff, 0x013a2b34, 0x9508, 0xffffffff, 0x00000002, @@ -152,8 +151,7 @@ static const u32 r7xx_golden_registers[] = 0x7300, 0xffffffff, 0x001000f0 }; -static const u32 r7xx_golden_dyn_gpr_registers[] = -{ +static const u32 r7xx_golden_dyn_gpr_registers[] = { 0x8db0, 0xffffffff, 0x98989898, 0x8db4, 0xffffffff, 0x98989898, 0x8db8, 0xffffffff, 0x98989898, @@ -165,8 +163,7 @@ static const u32 r7xx_golden_dyn_gpr_registers[] = 0x88c4, 0xffffffff, 0x00000082 }; -static const u32 rv770_golden_registers[] = -{ +static const u32 rv770_golden_registers[] = { 0x562c, 0xffffffff, 0, 0x3f90, 0xffffffff, 0, 0x9148, 0xffffffff, 0, @@ -175,8 +172,7 @@ static const u32 rv770_golden_registers[] = 0x9698, 0x18000000, 0x18000000 }; -static const u32 rv770ce_golden_registers[] = -{ +static const u32 rv770ce_golden_registers[] = { 0x562c, 0xffffffff, 0, 0x3f90, 0xffffffff, 0x00cc0000, 0x9148, 0xffffffff, 0x00cc0000, @@ -187,8 +183,7 @@ static const u32 rv770ce_golden_registers[] = 0x9698, 0x18000000, 0x18000000 }; -static const u32 rv770_mgcg_init[] = -{ +static const u32 rv770_mgcg_init[] = { 0x8bcc, 0xffffffff, 0x130300f9, 0x5448, 0xffffffff, 0x100, 0x55e4, 0xffffffff, 0x100, @@ -345,8 +340,7 @@ static const u32 rv770_mgcg_init[] = 0x92a4, 0xffffffff, 0x00080007 }; -static const u32 rv710_golden_registers[] = -{ +static const u32 rv710_golden_registers[] = { 0x3f90, 0x00ff0000, 0x00fc0000, 0x9148, 0x00ff0000, 0x00fc0000, 0x3f94, 0x00ff0000, 0x00fc0000, @@ -355,8 +349,7 @@ static const u32 rv710_golden_registers[] = 0xa180, 0xffffffff, 0x00003f3f }; -static const u32 rv710_mgcg_init[] = -{ +static const u32 rv710_mgcg_init[] = { 0x8bcc, 0xffffffff, 0x13030040, 0x5448, 0xffffffff, 0x100, 0x55e4, 0xffffffff, 0x100, @@ -414,8 +407,7 @@ static const u32 rv710_mgcg_init[] = 0x9150, 0xffffffff, 0x4d940000 }; -static const u32 rv730_golden_registers[] = -{ +static const u32 rv730_golden_registers[] = { 0x3f90, 0x00ff0000, 0x00f00000, 0x9148, 0x00ff0000, 0x00f00000, 0x3f94, 0x00ff0000, 0x00f00000, @@ -425,8 +417,7 @@ static const u32 rv730_golden_registers[] = 0xa180, 0xffffffff, 0x00003f3f }; -static const u32 rv730_mgcg_init[] = -{ +static const u32 rv730_mgcg_init[] = { 0x8bcc, 0xffffffff, 0x130300f9, 0x5448, 0xffffffff, 0x100, 0x55e4, 0xffffffff, 0x100, @@ -547,8 +538,7 @@ static const u32 rv730_mgcg_init[] = 0x92a4, 0xffffffff, 0x00000005 }; -static const u32 rv740_golden_registers[] = -{ +static const u32 rv740_golden_registers[] = { 0x88c4, 0xffffffff, 0x00000082, 0x28a50, 0xfffffffc, 0x00000004, 0x2650, 0x00040000, 0, @@ -584,8 +574,7 @@ static const u32 rv740_golden_registers[] = 0x9698, 0x18000000, 0x18000000 }; -static const u32 rv740_mgcg_init[] = -{ +static const u32 rv740_mgcg_init[] = { 0x8bcc, 0xffffffff, 0x13030100, 0x5448, 0xffffffff, 0x100, 0x55e4, 0xffffffff, 0x100, diff --git a/drivers/gpu/drm/radeon/rv770_smc.c b/drivers/gpu/drm/radeon/rv770_smc.c index 45575c0d0a1d..09fa7f5e7c41 100644 --- a/drivers/gpu/drm/radeon/rv770_smc.c +++ b/drivers/gpu/drm/radeon/rv770_smc.c @@ -34,8 +34,7 @@ #define FIRST_SMC_INT_VECT_REG 0xFFD8 #define FIRST_INT_VECT_S19 0xFFC0 -static const u8 rv770_smc_int_vectors[] = -{ +static const u8 rv770_smc_int_vectors[] = { 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, @@ -54,8 +53,7 @@ static const u8 rv770_smc_int_vectors[] = 0x03, 0x51, 0x03, 0x51 }; -static const u8 rv730_smc_int_vectors[] = -{ +static const u8 rv730_smc_int_vectors[] = { 0x08, 0x15, 0x08, 0x15, 0x08, 0x15, 0x08, 0x15, 0x08, 0x15, 0x08, 0x15, @@ -74,8 +72,7 @@ static const u8 rv730_smc_int_vectors[] = 0x03, 0x56, 0x03, 0x56 }; -static const u8 rv710_smc_int_vectors[] = -{ +static const u8 rv710_smc_int_vectors[] = { 0x08, 0x04, 0x08, 0x04, 0x08, 0x04, 0x08, 0x04, 0x08, 0x04, 0x08, 0x04, @@ -94,8 +91,7 @@ static const u8 rv710_smc_int_vectors[] = 0x03, 0x51, 0x03, 0x51 }; -static const u8 rv740_smc_int_vectors[] = -{ +static const u8 rv740_smc_int_vectors[] = { 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, 0x08, 0x10, @@ -114,8 +110,7 @@ static const u8 rv740_smc_int_vectors[] = 0x03, 0x51, 0x03, 0x51 }; -static const u8 cedar_smc_int_vectors[] = -{ +static const u8 cedar_smc_int_vectors[] = { 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, @@ -134,8 +129,7 @@ static const u8 cedar_smc_int_vectors[] = 0x04, 0xF6, 0x04, 0xF6 }; -static const u8 redwood_smc_int_vectors[] = -{ +static const u8 redwood_smc_int_vectors[] = { 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, @@ -154,8 +148,7 @@ static const u8 redwood_smc_int_vectors[] = 0x04, 0xF6, 0x04, 0xF6 }; -static const u8 juniper_smc_int_vectors[] = -{ +static const u8 juniper_smc_int_vectors[] = { 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, @@ -174,8 +167,7 @@ static const u8 juniper_smc_int_vectors[] = 0x04, 0xF6, 0x04, 0xF6 }; -static const u8 cypress_smc_int_vectors[] = -{ +static const u8 cypress_smc_int_vectors[] = { 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, 0x0B, 0x05, @@ -194,8 +186,7 @@ static const u8 cypress_smc_int_vectors[] = 0x04, 0xF6, 0x04, 0xF6 }; -static const u8 barts_smc_int_vectors[] = -{ +static const u8 barts_smc_int_vectors[] = { 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, @@ -214,8 +205,7 @@ static const u8 barts_smc_int_vectors[] = 0x05, 0x0A, 0x05, 0x0A }; -static const u8 turks_smc_int_vectors[] = -{ +static const u8 turks_smc_int_vectors[] = { 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, @@ -234,8 +224,7 @@ static const u8 turks_smc_int_vectors[] = 0x05, 0x0A, 0x05, 0x0A }; -static const u8 caicos_smc_int_vectors[] = -{ +static const u8 caicos_smc_int_vectors[] = { 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, 0x0C, 0x14, @@ -254,8 +243,7 @@ static const u8 caicos_smc_int_vectors[] = 0x05, 0x0A, 0x05, 0x0A }; -static const u8 cayman_smc_int_vectors[] = -{ +static const u8 cayman_smc_int_vectors[] = { 0x12, 0x05, 0x12, 0x05, 0x12, 0x05, 0x12, 0x05, 0x12, 0x05, 0x12, 0x05, diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 8d5e4b25609d..a91012447b56 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -7131,17 +7131,8 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) u16 bridge_cfg2, gpu_cfg2; u32 max_lw, current_lw, tmp; - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &bridge_cfg); - pcie_capability_read_word(rdev->pdev, PCI_EXP_LNKCTL, - &gpu_cfg); - - tmp16 = bridge_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(root, PCI_EXP_LNKCTL, tmp16); - - tmp16 = gpu_cfg | PCI_EXP_LNKCTL_HAWD; - pcie_capability_write_word(rdev->pdev, PCI_EXP_LNKCTL, - tmp16); + pcie_capability_set_word(root, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); + pcie_capability_set_word(rdev->pdev, PCI_EXP_LNKCTL, PCI_EXP_LNKCTL_HAWD); tmp = RREG32_PCIE(PCIE_LC_STATUS1); max_lw = (tmp & LC_DETECTED_LINK_WIDTH_MASK) >> LC_DETECTED_LINK_WIDTH_SHIFT; @@ -7188,22 +7179,14 @@ static void si_pcie_gen3_enable(struct radeon_device *rdev) msleep(100); /* linkctl */ - pcie_capability_read_word(root, PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (bridge_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(root, - PCI_EXP_LNKCTL, - tmp16); - - pcie_capability_read_word(rdev->pdev, - PCI_EXP_LNKCTL, - &tmp16); - tmp16 &= ~PCI_EXP_LNKCTL_HAWD; - tmp16 |= (gpu_cfg & PCI_EXP_LNKCTL_HAWD); - pcie_capability_write_word(rdev->pdev, - PCI_EXP_LNKCTL, - tmp16); + pcie_capability_clear_and_set_word(root, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + bridge_cfg & + PCI_EXP_LNKCTL_HAWD); + pcie_capability_clear_and_set_word(rdev->pdev, PCI_EXP_LNKCTL, + PCI_EXP_LNKCTL_HAWD, + gpu_cfg & + PCI_EXP_LNKCTL_HAWD); /* linkctl2 */ pcie_capability_read_word(root, PCI_EXP_LNKCTL2, diff --git a/drivers/gpu/drm/radeon/sislands_smc.h b/drivers/gpu/drm/radeon/sislands_smc.h index 4ea1cb2e45a3..4b7dee3cf58b 100644 --- a/drivers/gpu/drm/radeon/sislands_smc.h +++ b/drivers/gpu/drm/radeon/sislands_smc.h @@ -89,8 +89,7 @@ struct PP_SIslands_PAPMStatus }; typedef struct PP_SIslands_PAPMStatus PP_SIslands_PAPMStatus; -struct PP_SIslands_PAPMParameters -{ +struct PP_SIslands_PAPMParameters { uint32_t NearTDPLimitTherm; uint32_t NearTDPLimitPAPM; uint32_t PlatformPowerLimit; @@ -100,8 +99,7 @@ struct PP_SIslands_PAPMParameters }; typedef struct PP_SIslands_PAPMParameters PP_SIslands_PAPMParameters; -struct SISLANDS_SMC_SCLK_VALUE -{ +struct SISLANDS_SMC_SCLK_VALUE { uint32_t vCG_SPLL_FUNC_CNTL; uint32_t vCG_SPLL_FUNC_CNTL_2; uint32_t vCG_SPLL_FUNC_CNTL_3; @@ -113,8 +111,7 @@ struct SISLANDS_SMC_SCLK_VALUE typedef struct SISLANDS_SMC_SCLK_VALUE SISLANDS_SMC_SCLK_VALUE; -struct SISLANDS_SMC_MCLK_VALUE -{ +struct SISLANDS_SMC_MCLK_VALUE { uint32_t vMPLL_FUNC_CNTL; uint32_t vMPLL_FUNC_CNTL_1; uint32_t vMPLL_FUNC_CNTL_2; @@ -129,8 +126,7 @@ struct SISLANDS_SMC_MCLK_VALUE typedef struct SISLANDS_SMC_MCLK_VALUE SISLANDS_SMC_MCLK_VALUE; -struct SISLANDS_SMC_VOLTAGE_VALUE -{ +struct SISLANDS_SMC_VOLTAGE_VALUE { uint16_t value; uint8_t index; uint8_t phase_settings; @@ -138,8 +134,7 @@ struct SISLANDS_SMC_VOLTAGE_VALUE typedef struct SISLANDS_SMC_VOLTAGE_VALUE SISLANDS_SMC_VOLTAGE_VALUE; -struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL -{ +struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL { uint8_t ACIndex; uint8_t displayWatermark; uint8_t gen2PCIE; @@ -180,8 +175,7 @@ struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL typedef struct SISLANDS_SMC_HW_PERFORMANCE_LEVEL SISLANDS_SMC_HW_PERFORMANCE_LEVEL; -struct SISLANDS_SMC_SWSTATE -{ +struct SISLANDS_SMC_SWSTATE { uint8_t flags; uint8_t levelCount; uint8_t padding2; @@ -205,8 +199,7 @@ struct SISLANDS_SMC_SWSTATE_SINGLE { #define SISLANDS_SMC_VOLTAGEMASK_VDDC_PHASE_SHEDDING 3 #define SISLANDS_SMC_VOLTAGEMASK_MAX 4 -struct SISLANDS_SMC_VOLTAGEMASKTABLE -{ +struct SISLANDS_SMC_VOLTAGEMASKTABLE { uint32_t lowMask[SISLANDS_SMC_VOLTAGEMASK_MAX]; }; @@ -214,8 +207,7 @@ typedef struct SISLANDS_SMC_VOLTAGEMASKTABLE SISLANDS_SMC_VOLTAGEMASKTABLE; #define SISLANDS_MAX_NO_VREG_STEPS 32 -struct SISLANDS_SMC_STATETABLE -{ +struct SISLANDS_SMC_STATETABLE { uint8_t thermalProtectType; uint8_t systemFlags; uint8_t maxVDDCIndexInPPTable; @@ -254,8 +246,7 @@ typedef struct SISLANDS_SMC_STATETABLE SISLANDS_SMC_STATETABLE; #define SI_SMC_SOFT_REGISTER_svi_rework_gpio_id_svd 0x11c #define SI_SMC_SOFT_REGISTER_svi_rework_gpio_id_svc 0x120 -struct PP_SIslands_FanTable -{ +struct PP_SIslands_FanTable { uint8_t fdo_mode; uint8_t padding; int16_t temp_min; @@ -285,8 +276,7 @@ typedef struct PP_SIslands_FanTable PP_SIslands_FanTable; #define SMC_SISLANDS_SCALE_I 7 #define SMC_SISLANDS_SCALE_R 12 -struct PP_SIslands_CacConfig -{ +struct PP_SIslands_CacConfig { uint16_t cac_lkge_lut[SMC_SISLANDS_LKGE_LUT_NUM_OF_TEMP_ENTRIES][SMC_SISLANDS_LKGE_LUT_NUM_OF_VOLT_ENTRIES]; uint32_t lkge_lut_V0; uint32_t lkge_lut_Vstep; @@ -308,23 +298,20 @@ typedef struct PP_SIslands_CacConfig PP_SIslands_CacConfig; #define SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE 16 #define SMC_SISLANDS_MC_REGISTER_ARRAY_SET_COUNT 20 -struct SMC_SIslands_MCRegisterAddress -{ +struct SMC_SIslands_MCRegisterAddress { uint16_t s0; uint16_t s1; }; typedef struct SMC_SIslands_MCRegisterAddress SMC_SIslands_MCRegisterAddress; -struct SMC_SIslands_MCRegisterSet -{ +struct SMC_SIslands_MCRegisterSet { uint32_t value[SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE]; }; typedef struct SMC_SIslands_MCRegisterSet SMC_SIslands_MCRegisterSet; -struct SMC_SIslands_MCRegisters -{ +struct SMC_SIslands_MCRegisters { uint8_t last; uint8_t reserved[3]; SMC_SIslands_MCRegisterAddress address[SMC_SISLANDS_MC_REGISTER_ARRAY_SIZE]; @@ -333,8 +320,7 @@ struct SMC_SIslands_MCRegisters typedef struct SMC_SIslands_MCRegisters SMC_SIslands_MCRegisters; -struct SMC_SIslands_MCArbDramTimingRegisterSet -{ +struct SMC_SIslands_MCArbDramTimingRegisterSet { uint32_t mc_arb_dram_timing; uint32_t mc_arb_dram_timing2; uint8_t mc_arb_rfsh_rate; @@ -344,8 +330,7 @@ struct SMC_SIslands_MCArbDramTimingRegisterSet typedef struct SMC_SIslands_MCArbDramTimingRegisterSet SMC_SIslands_MCArbDramTimingRegisterSet; -struct SMC_SIslands_MCArbDramTimingRegisters -{ +struct SMC_SIslands_MCArbDramTimingRegisters { uint8_t arb_current; uint8_t reserved[3]; SMC_SIslands_MCArbDramTimingRegisterSet data[16]; @@ -353,8 +338,7 @@ struct SMC_SIslands_MCArbDramTimingRegisters typedef struct SMC_SIslands_MCArbDramTimingRegisters SMC_SIslands_MCArbDramTimingRegisters; -struct SMC_SISLANDS_SPLL_DIV_TABLE -{ +struct SMC_SISLANDS_SPLL_DIV_TABLE { uint32_t freq[256]; uint32_t ss[256]; }; @@ -374,8 +358,7 @@ typedef struct SMC_SISLANDS_SPLL_DIV_TABLE SMC_SISLANDS_SPLL_DIV_TABLE; #define SMC_SISLANDS_DTE_MAX_TEMPERATURE_DEPENDENT_ARRAY_SIZE 16 -struct Smc_SIslands_DTE_Configuration -{ +struct Smc_SIslands_DTE_Configuration { uint32_t tau[SMC_SISLANDS_DTE_MAX_FILTER_STAGES]; uint32_t R[SMC_SISLANDS_DTE_MAX_FILTER_STAGES]; uint32_t K; diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c index 1dd722fe8631..dee530e4c8b2 100644 --- a/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_drv.c @@ -696,6 +696,10 @@ static int rcar_du_probe(struct platform_device *pdev) /* DRM/KMS objects */ ret = rcar_du_modeset_init(rcdu); if (ret < 0) { + /* + * Don't use dev_err_probe(), as it would overwrite the probe + * deferral reason recorded in rcar_du_modeset_init(). + */ if (ret != -EPROBE_DEFER) dev_err(&pdev->dev, "failed to initialize DRM/KMS (%d)\n", ret); @@ -710,7 +714,7 @@ static int rcar_du_probe(struct platform_device *pdev) if (ret) goto error; - DRM_INFO("Device %s probed\n", dev_name(&pdev->dev)); + drm_info(&rcdu->ddev, "Device %s probed\n", dev_name(&pdev->dev)); drm_fbdev_generic_setup(&rcdu->ddev, 32); diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c index 9ff4537c26c8..70d8ad065bfa 100644 --- a/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_kms.c @@ -935,7 +935,8 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu) /* Initialize the Color Management Modules. */ ret = rcar_du_cmm_init(rcdu); if (ret) - return ret; + return dev_err_probe(rcdu->dev, ret, + "failed to initialize CMM\n"); /* Create the CRTCs. */ for (swindex = 0, hwindex = 0; swindex < rcdu->num_crtcs; ++hwindex) { @@ -955,7 +956,8 @@ int rcar_du_modeset_init(struct rcar_du_device *rcdu) /* Initialize the encoders. */ ret = rcar_du_encoders_init(rcdu); if (ret < 0) - return ret; + return dev_err_probe(rcdu->dev, ret, + "failed to initialize encoders\n"); if (ret == 0) { dev_err(rcdu->dev, "error: no encoder could be initialized\n"); diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c b/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c index 9cbb5e6e2cba..7aa0373563a4 100644 --- a/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_du_vsp.c @@ -123,6 +123,8 @@ static const u32 rcar_du_vsp_formats[] = { DRM_FORMAT_RGB888, DRM_FORMAT_BGRA8888, DRM_FORMAT_BGRX8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_XBGR8888, DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB8888, DRM_FORMAT_UYVY, @@ -155,6 +157,8 @@ static const u32 rcar_du_vsp_formats_gen4[] = { DRM_FORMAT_RGB888, DRM_FORMAT_BGRA8888, DRM_FORMAT_BGRX8888, + DRM_FORMAT_ABGR8888, + DRM_FORMAT_XBGR8888, DRM_FORMAT_ARGB8888, DRM_FORMAT_XRGB8888, DRM_FORMAT_RGBX1010102, @@ -177,6 +181,41 @@ static const u32 rcar_du_vsp_formats_gen4[] = { DRM_FORMAT_Y212, }; +static u32 rcar_du_vsp_state_get_format(struct rcar_du_vsp_plane_state *state) +{ + u32 fourcc = state->format->fourcc; + + if (state->state.pixel_blend_mode == DRM_MODE_BLEND_PIXEL_NONE) { + switch (fourcc) { + case DRM_FORMAT_ARGB1555: + fourcc = DRM_FORMAT_XRGB1555; + break; + + case DRM_FORMAT_ARGB4444: + fourcc = DRM_FORMAT_XRGB4444; + break; + + case DRM_FORMAT_ARGB8888: + fourcc = DRM_FORMAT_XRGB8888; + break; + + case DRM_FORMAT_ABGR8888: + fourcc = DRM_FORMAT_XBGR8888; + break; + + case DRM_FORMAT_BGRA8888: + fourcc = DRM_FORMAT_BGRX8888; + break; + + case DRM_FORMAT_RGBA1010102: + fourcc = DRM_FORMAT_RGBX1010102; + break; + } + } + + return fourcc; +} + static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane) { struct rcar_du_vsp_plane_state *state = @@ -190,7 +229,7 @@ static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane) .alpha = state->state.alpha >> 8, .zpos = state->state.zpos, }; - u32 fourcc = state->format->fourcc; + u32 fourcc = rcar_du_vsp_state_get_format(state); unsigned int i; cfg.src.left = state->state.src.x1 >> 16; @@ -207,22 +246,6 @@ static void rcar_du_vsp_plane_setup(struct rcar_du_vsp_plane *plane) cfg.mem[i] = sg_dma_address(state->sg_tables[i].sgl) + fb->offsets[i]; - if (state->state.pixel_blend_mode == DRM_MODE_BLEND_PIXEL_NONE) { - switch (fourcc) { - case DRM_FORMAT_ARGB1555: - fourcc = DRM_FORMAT_XRGB1555; - break; - - case DRM_FORMAT_ARGB4444: - fourcc = DRM_FORMAT_XRGB4444; - break; - - case DRM_FORMAT_ARGB8888: - fourcc = DRM_FORMAT_XRGB8888; - break; - } - } - format = rcar_du_format_info(fourcc); cfg.pixelformat = format->v4l2; diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c index 586c5c4ebb14..2dba7c5ffd2c 100644 --- a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c +++ b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c @@ -1001,7 +1001,6 @@ static int rcar_mipi_dsi_get_clocks(struct rcar_mipi_dsi *dsi) static int rcar_mipi_dsi_probe(struct platform_device *pdev) { struct rcar_mipi_dsi *dsi; - struct resource *mem; int ret; dsi = devm_kzalloc(&pdev->dev, sizeof(*dsi), GFP_KERNEL); @@ -1018,8 +1017,7 @@ static int rcar_mipi_dsi_probe(struct platform_device *pdev) return ret; /* Acquire resources. */ - mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); - dsi->mmio = devm_ioremap_resource(dsi->dev, mem); + dsi->mmio = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(dsi->mmio)) return PTR_ERR(dsi->mmio); diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c index 4872aecda066..e1ea50c8fd39 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -827,12 +827,12 @@ static int vop_plane_atomic_check(struct drm_plane *plane, * need align with 2 pixel. */ if (fb->format->is_yuv && ((new_plane_state->src.x1 >> 16) % 2)) { - DRM_ERROR("Invalid Source: Yuv format not support odd xpos\n"); + DRM_DEBUG_KMS("Invalid Source: Yuv format not support odd xpos\n"); return -EINVAL; } if (fb->format->is_yuv && new_plane_state->rotation & DRM_MODE_REFLECT_Y) { - DRM_ERROR("Invalid Source: Yuv format does not support this rotation\n"); + DRM_DEBUG_KMS("Invalid Source: Yuv format does not support this rotation\n"); return -EINVAL; } @@ -840,7 +840,7 @@ static int vop_plane_atomic_check(struct drm_plane *plane, struct vop *vop = to_vop(crtc); if (!vop->data->afbc) { - DRM_ERROR("vop does not support AFBC\n"); + DRM_DEBUG_KMS("vop does not support AFBC\n"); return -EINVAL; } @@ -849,15 +849,16 @@ static int vop_plane_atomic_check(struct drm_plane *plane, return ret; if (new_plane_state->src.x1 || new_plane_state->src.y1) { - DRM_ERROR("AFBC does not support offset display, xpos=%d, ypos=%d, offset=%d\n", - new_plane_state->src.x1, - new_plane_state->src.y1, fb->offsets[0]); + DRM_DEBUG_KMS("AFBC does not support offset display, " \ + "xpos=%d, ypos=%d, offset=%d\n", + new_plane_state->src.x1, new_plane_state->src.y1, + fb->offsets[0]); return -EINVAL; } if (new_plane_state->rotation && new_plane_state->rotation != DRM_MODE_ROTATE_0) { - DRM_ERROR("No rotation support in AFBC, rotation=%d\n", - new_plane_state->rotation); + DRM_DEBUG_KMS("No rotation support in AFBC, rotation=%d\n", + new_plane_state->rotation); return -EINVAL; } } diff --git a/drivers/gpu/drm/sun4i/sun8i_tcon_top.c b/drivers/gpu/drm/sun4i/sun8i_tcon_top.c index 6f076cf4b403..a1ca3916f42b 100644 --- a/drivers/gpu/drm/sun4i/sun8i_tcon_top.c +++ b/drivers/gpu/drm/sun4i/sun8i_tcon_top.c @@ -141,6 +141,7 @@ static int sun8i_tcon_top_bind(struct device *dev, struct device *master, GFP_KERNEL); if (!clk_data) return -ENOMEM; + clk_data->num = CLK_NUM; tcon_top->clk_data = clk_data; spin_lock_init(&tcon_top->reg_lock); @@ -213,8 +214,6 @@ static int sun8i_tcon_top_bind(struct device *dev, struct device *master, goto err_unregister_gates; } - clk_data->num = CLK_NUM; - ret = of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get, clk_data); if (ret) diff --git a/drivers/gpu/drm/tests/drm_kunit_helpers.c b/drivers/gpu/drm/tests/drm_kunit_helpers.c index 3d624ff2f651..c1dfbfcaa000 100644 --- a/drivers/gpu/drm/tests/drm_kunit_helpers.c +++ b/drivers/gpu/drm/tests/drm_kunit_helpers.c @@ -156,7 +156,7 @@ static void action_drm_release_context(void *ptr) } /** - * drm_kunit_helper_context_alloc - Allocates an acquire context + * drm_kunit_helper_acquire_ctx_alloc - Allocates an acquire context * @test: The test context object * * Allocates and initializes a modeset acquire context. diff --git a/drivers/gpu/drm/tiny/gm12u320.c b/drivers/gpu/drm/tiny/gm12u320.c index c5bb683e440c..0187539ff5ea 100644 --- a/drivers/gpu/drm/tiny/gm12u320.c +++ b/drivers/gpu/drm/tiny/gm12u320.c @@ -70,10 +70,10 @@ MODULE_PARM_DESC(eco_mode, "Turn on Eco mode (less bright, more silent)"); #define READ_STATUS_SIZE 13 #define MISC_VALUE_SIZE 4 -#define CMD_TIMEOUT msecs_to_jiffies(200) -#define DATA_TIMEOUT msecs_to_jiffies(1000) -#define IDLE_TIMEOUT msecs_to_jiffies(2000) -#define FIRST_FRAME_TIMEOUT msecs_to_jiffies(2000) +#define CMD_TIMEOUT 200 +#define DATA_TIMEOUT 1000 +#define IDLE_TIMEOUT 2000 +#define FIRST_FRAME_TIMEOUT 2000 #define MISC_REQ_GET_SET_ECO_A 0xff #define MISC_REQ_GET_SET_ECO_B 0x35 @@ -389,7 +389,7 @@ static void gm12u320_fb_update_work(struct work_struct *work) * switches back to showing its logo. */ queue_delayed_work(system_long_wq, &gm12u320->fb_update.work, - IDLE_TIMEOUT); + msecs_to_jiffies(IDLE_TIMEOUT)); return; err: diff --git a/drivers/gpu/drm/ttm/tests/ttm_pool_test.c b/drivers/gpu/drm/ttm/tests/ttm_pool_test.c index 8d90870fb199..2d9cae8cd984 100644 --- a/drivers/gpu/drm/ttm/tests/ttm_pool_test.c +++ b/drivers/gpu/drm/ttm/tests/ttm_pool_test.c @@ -228,8 +228,8 @@ static void ttm_pool_alloc_basic_dma_addr(struct kunit *test) dma1 = tt->dma_address[0]; dma2 = tt->dma_address[tt->num_pages - 1]; - KUNIT_ASSERT_NOT_NULL(test, (void *)dma1); - KUNIT_ASSERT_NOT_NULL(test, (void *)dma2); + KUNIT_ASSERT_NOT_NULL(test, (void *)(uintptr_t)dma1); + KUNIT_ASSERT_NOT_NULL(test, (void *)(uintptr_t)dma2); ttm_pool_free(pool, tt); ttm_tt_fini(tt); diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index aa8ab1a418af..e58b7e249816 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -520,7 +520,8 @@ static bool ttm_bo_evict_swapout_allowable(struct ttm_buffer_object *bo, if (bo->pin_count) { *locked = false; - *busy = false; + if (busy) + *busy = false; return false; } diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c index a488625773dc..25c9c71256d3 100644 --- a/drivers/gpu/drm/vc4/vc4_hdmi.c +++ b/drivers/gpu/drm/vc4/vc4_hdmi.c @@ -2615,9 +2615,13 @@ static int vc4_hdmi_audio_cpu_dai_probe(struct snd_soc_dai *dai) return 0; } +static const struct snd_soc_dai_ops vc4_snd_dai_ops = { + .probe = vc4_hdmi_audio_cpu_dai_probe, +}; + static struct snd_soc_dai_driver vc4_hdmi_audio_cpu_dai_drv = { .name = "vc4-hdmi-cpu-dai", - .probe = vc4_hdmi_audio_cpu_dai_probe, + .ops = &vc4_snd_dai_ops, .playback = { .stream_name = "Playback", .channels_min = 1, diff --git a/drivers/gpu/drm/vkms/vkms_composer.c b/drivers/gpu/drm/vkms/vkms_composer.c index d5d4f642d367..3c99fb8b54e2 100644 --- a/drivers/gpu/drm/vkms/vkms_composer.c +++ b/drivers/gpu/drm/vkms/vkms_composer.c @@ -408,15 +408,10 @@ void vkms_set_composer(struct vkms_output *out, bool enabled) if (enabled) drm_crtc_vblank_get(&out->crtc); - mutex_lock(&out->enabled_lock); + spin_lock_irq(&out->lock); old_enabled = out->composer_enabled; out->composer_enabled = enabled; - - /* the composition wasn't enabled, so unlock the lock to make sure the lock - * will be balanced even if we have a failed commit - */ - if (!out->composer_enabled) - mutex_unlock(&out->enabled_lock); + spin_unlock_irq(&out->lock); if (old_enabled) drm_crtc_vblank_put(&out->crtc); diff --git a/drivers/gpu/drm/vkms/vkms_crtc.c b/drivers/gpu/drm/vkms/vkms_crtc.c index 3c5ebf106b66..61e500b8c9da 100644 --- a/drivers/gpu/drm/vkms/vkms_crtc.c +++ b/drivers/gpu/drm/vkms/vkms_crtc.c @@ -16,7 +16,7 @@ static enum hrtimer_restart vkms_vblank_simulate(struct hrtimer *timer) struct drm_crtc *crtc = &output->crtc; struct vkms_crtc_state *state; u64 ret_overrun; - bool ret, fence_cookie, composer_enabled; + bool ret, fence_cookie; fence_cookie = dma_fence_begin_signalling(); @@ -25,15 +25,15 @@ static enum hrtimer_restart vkms_vblank_simulate(struct hrtimer *timer) if (ret_overrun != 1) pr_warn("%s: vblank timer overrun\n", __func__); + spin_lock(&output->lock); ret = drm_crtc_handle_vblank(crtc); if (!ret) DRM_ERROR("vkms failure on handling vblank"); state = output->composer_state; - composer_enabled = output->composer_enabled; - mutex_unlock(&output->enabled_lock); + spin_unlock(&output->lock); - if (state && composer_enabled) { + if (state && output->composer_enabled) { u64 frame = drm_crtc_accurate_vblank_count(crtc); /* update frame_start only if a queued vkms_composer_worker() @@ -295,7 +295,6 @@ int vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, spin_lock_init(&vkms_out->lock); spin_lock_init(&vkms_out->composer_lock); - mutex_init(&vkms_out->enabled_lock); vkms_out->composer_workq = alloc_ordered_workqueue("vkms_composer", 0); if (!vkms_out->composer_workq) diff --git a/drivers/gpu/drm/vkms/vkms_drv.h b/drivers/gpu/drm/vkms/vkms_drv.h index c7ae6c2ba1df..8f5710debb1e 100644 --- a/drivers/gpu/drm/vkms/vkms_drv.h +++ b/drivers/gpu/drm/vkms/vkms_drv.h @@ -108,10 +108,8 @@ struct vkms_output { struct workqueue_struct *composer_workq; /* protects concurrent access to composer */ spinlock_t lock; - /* guarantees that if the composer is enabled, a job will be queued */ - struct mutex enabled_lock; - /* protected by @enabled_lock */ + /* protected by @lock */ bool composer_enabled; struct vkms_crtc_state *composer_state; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c index 82094c137855..c43853597776 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.c @@ -497,10 +497,9 @@ static int vmw_user_bo_synccpu_release(struct drm_file *filp, if (!(flags & drm_vmw_synccpu_allow_cs)) { atomic_dec(&vmw_bo->cpu_writers); } - ttm_bo_put(&vmw_bo->tbo); + vmw_user_bo_unref(vmw_bo); } - drm_gem_object_put(&vmw_bo->tbo.base); return ret; } @@ -540,8 +539,7 @@ int vmw_user_bo_synccpu_ioctl(struct drm_device *dev, void *data, return ret; ret = vmw_user_bo_synccpu_grab(vbo, arg->flags); - vmw_bo_unreference(&vbo); - drm_gem_object_put(&vbo->tbo.base); + vmw_user_bo_unref(vbo); if (unlikely(ret != 0)) { if (ret == -ERESTARTSYS || ret == -EBUSY) return -EBUSY; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h index 50a836e70994..1d433fceed3d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_bo.h @@ -195,6 +195,14 @@ static inline struct vmw_bo *vmw_bo_reference(struct vmw_bo *buf) return buf; } +static inline void vmw_user_bo_unref(struct vmw_bo *vbo) +{ + if (vbo) { + ttm_bo_put(&vbo->tbo); + drm_gem_object_put(&vbo->tbo.base); + } +} + static inline struct vmw_bo *to_vmw_bo(struct drm_gem_object *gobj) { return container_of((gobj), struct vmw_bo, tbo.base); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 3810a9984a7f..58bfdf203eca 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -1513,4 +1513,16 @@ static inline bool vmw_has_fences(struct vmw_private *vmw) return (vmw_fifo_caps(vmw) & SVGA_FIFO_CAP_FENCE) != 0; } +static inline bool vmw_shadertype_is_valid(enum vmw_sm_type shader_model, + u32 shader_type) +{ + SVGA3dShaderType max_allowed = SVGA3D_SHADERTYPE_PREDX_MAX; + + if (shader_model >= VMW_SM_5) + max_allowed = SVGA3D_SHADERTYPE_MAX; + else if (shader_model >= VMW_SM_4) + max_allowed = SVGA3D_SHADERTYPE_DX10_MAX; + return shader_type >= SVGA3D_SHADERTYPE_MIN && shader_type < max_allowed; +} + #endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 6b9aa2b4ef54..98e0723ca6f5 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1164,8 +1164,7 @@ static int vmw_translate_mob_ptr(struct vmw_private *dev_priv, } vmw_bo_placement_set(vmw_bo, VMW_BO_DOMAIN_MOB, VMW_BO_DOMAIN_MOB); ret = vmw_validation_add_bo(sw_context->ctx, vmw_bo); - ttm_bo_put(&vmw_bo->tbo); - drm_gem_object_put(&vmw_bo->tbo.base); + vmw_user_bo_unref(vmw_bo); if (unlikely(ret != 0)) return ret; @@ -1221,8 +1220,7 @@ static int vmw_translate_guest_ptr(struct vmw_private *dev_priv, vmw_bo_placement_set(vmw_bo, VMW_BO_DOMAIN_GMR | VMW_BO_DOMAIN_VRAM, VMW_BO_DOMAIN_GMR | VMW_BO_DOMAIN_VRAM); ret = vmw_validation_add_bo(sw_context->ctx, vmw_bo); - ttm_bo_put(&vmw_bo->tbo); - drm_gem_object_put(&vmw_bo->tbo.base); + vmw_user_bo_unref(vmw_bo); if (unlikely(ret != 0)) return ret; @@ -1992,7 +1990,7 @@ static int vmw_cmd_set_shader(struct vmw_private *dev_priv, cmd = container_of(header, typeof(*cmd), header); - if (cmd->body.type >= SVGA3D_SHADERTYPE_PREDX_MAX) { + if (!vmw_shadertype_is_valid(VMW_SM_LEGACY, cmd->body.type)) { VMW_DEBUG_USER("Illegal shader type %u.\n", (unsigned int) cmd->body.type); return -EINVAL; @@ -2115,8 +2113,6 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetSingleConstantBuffer); - SVGA3dShaderType max_shader_num = has_sm5_context(dev_priv) ? - SVGA3D_NUM_SHADERTYPE : SVGA3D_NUM_SHADERTYPE_DX10; struct vmw_resource *res = NULL; struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); @@ -2133,6 +2129,14 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, if (unlikely(ret != 0)) return ret; + if (!vmw_shadertype_is_valid(dev_priv->sm_type, cmd->body.type) || + cmd->body.slot >= SVGA3D_DX_MAX_CONSTBUFFERS) { + VMW_DEBUG_USER("Illegal const buffer shader %u slot %u.\n", + (unsigned int) cmd->body.type, + (unsigned int) cmd->body.slot); + return -EINVAL; + } + binding.bi.ctx = ctx_node->ctx; binding.bi.res = res; binding.bi.bt = vmw_ctx_binding_cb; @@ -2141,14 +2145,6 @@ vmw_cmd_dx_set_single_constant_buffer(struct vmw_private *dev_priv, binding.size = cmd->body.sizeInBytes; binding.slot = cmd->body.slot; - if (binding.shader_slot >= max_shader_num || - binding.slot >= SVGA3D_DX_MAX_CONSTBUFFERS) { - VMW_DEBUG_USER("Illegal const buffer shader %u slot %u.\n", - (unsigned int) cmd->body.type, - (unsigned int) binding.slot); - return -EINVAL; - } - vmw_binding_add(ctx_node->staged, &binding.bi, binding.shader_slot, binding.slot); @@ -2207,15 +2203,13 @@ static int vmw_cmd_dx_set_shader_res(struct vmw_private *dev_priv, { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShaderResources) = container_of(header, typeof(*cmd), header); - SVGA3dShaderType max_allowed = has_sm5_context(dev_priv) ? - SVGA3D_SHADERTYPE_MAX : SVGA3D_SHADERTYPE_DX10_MAX; u32 num_sr_view = (cmd->header.size - sizeof(cmd->body)) / sizeof(SVGA3dShaderResourceViewId); if ((u64) cmd->body.startView + (u64) num_sr_view > (u64) SVGA3D_DX_MAX_SRVIEWS || - cmd->body.type >= max_allowed) { + !vmw_shadertype_is_valid(dev_priv->sm_type, cmd->body.type)) { VMW_DEBUG_USER("Invalid shader binding.\n"); return -EINVAL; } @@ -2239,8 +2233,6 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, SVGA3dCmdHeader *header) { VMW_DECLARE_CMD_VAR(*cmd, SVGA3dCmdDXSetShader); - SVGA3dShaderType max_allowed = has_sm5_context(dev_priv) ? - SVGA3D_SHADERTYPE_MAX : SVGA3D_SHADERTYPE_DX10_MAX; struct vmw_resource *res = NULL; struct vmw_ctx_validation_info *ctx_node = VMW_GET_CTX_NODE(sw_context); struct vmw_ctx_bindinfo_shader binding; @@ -2251,8 +2243,7 @@ static int vmw_cmd_dx_set_shader(struct vmw_private *dev_priv, cmd = container_of(header, typeof(*cmd), header); - if (cmd->body.type >= max_allowed || - cmd->body.type < SVGA3D_SHADERTYPE_MIN) { + if (!vmw_shadertype_is_valid(dev_priv->sm_type, cmd->body.type)) { VMW_DEBUG_USER("Illegal shader type %u.\n", (unsigned int) cmd->body.type); return -EINVAL; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index b62207be3363..1489ad73c103 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -1665,10 +1665,8 @@ static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev, err_out: /* vmw_user_lookup_handle takes one ref so does new_fb */ - if (bo) { - vmw_bo_unreference(&bo); - drm_gem_object_put(&bo->tbo.base); - } + if (bo) + vmw_user_bo_unref(bo); if (surface) vmw_surface_unreference(&surface); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c index 7e112319a23c..fb85f244c3d0 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_overlay.c @@ -451,8 +451,7 @@ int vmw_overlay_ioctl(struct drm_device *dev, void *data, ret = vmw_overlay_update_stream(dev_priv, buf, arg, true); - vmw_bo_unreference(&buf); - drm_gem_object_put(&buf->tbo.base); + vmw_user_bo_unref(buf); out_unlock: mutex_unlock(&overlay->mutex); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c index e7226db8b242..1e81ff2422cf 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_shader.c @@ -809,8 +809,7 @@ static int vmw_shader_define(struct drm_device *dev, struct drm_file *file_priv, shader_type, num_input_sig, num_output_sig, tfile, shader_handle); out_bad_arg: - vmw_bo_unreference(&buffer); - drm_gem_object_put(&buffer->tbo.base); + vmw_user_bo_unref(buffer); return ret; } diff --git a/drivers/gpu/drm/xlnx/zynqmp_disp.c b/drivers/gpu/drm/xlnx/zynqmp_disp.c index 3b87eebddc97..407bc07cec69 100644 --- a/drivers/gpu/drm/xlnx/zynqmp_disp.c +++ b/drivers/gpu/drm/xlnx/zynqmp_disp.c @@ -1094,8 +1094,8 @@ static int zynqmp_disp_layer_request_dma(struct zynqmp_disp *disp, "%s%u", dma_names[layer->id], i); dma->chan = dma_request_chan(disp->dev, dma_channel_name); if (IS_ERR(dma->chan)) { - dev_err(disp->dev, "failed to request dma channel\n"); - ret = PTR_ERR(dma->chan); + ret = dev_err_probe(disp->dev, PTR_ERR(dma->chan), + "failed to request dma channel\n"); dma->chan = NULL; return ret; } @@ -1228,7 +1228,6 @@ int zynqmp_disp_probe(struct zynqmp_dpsub *dpsub) { struct platform_device *pdev = to_platform_device(dpsub->dev); struct zynqmp_disp *disp; - struct resource *res; int ret; disp = kzalloc(sizeof(*disp), GFP_KERNEL); @@ -1238,22 +1237,19 @@ int zynqmp_disp_probe(struct zynqmp_dpsub *dpsub) disp->dev = &pdev->dev; disp->dpsub = dpsub; - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "blend"); - disp->blend.base = devm_ioremap_resource(disp->dev, res); + disp->blend.base = devm_platform_ioremap_resource_byname(pdev, "blend"); if (IS_ERR(disp->blend.base)) { ret = PTR_ERR(disp->blend.base); goto error; } - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "av_buf"); - disp->avbuf.base = devm_ioremap_resource(disp->dev, res); + disp->avbuf.base = devm_platform_ioremap_resource_byname(pdev, "av_buf"); if (IS_ERR(disp->avbuf.base)) { ret = PTR_ERR(disp->avbuf.base); goto error; } - res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "aud"); - disp->audio.base = devm_ioremap_resource(disp->dev, res); + disp->audio.base = devm_platform_ioremap_resource_byname(pdev, "aud"); if (IS_ERR(disp->audio.base)) { ret = PTR_ERR(disp->audio.base); goto error; diff --git a/drivers/gpu/drm/xlnx/zynqmp_dp.c b/drivers/gpu/drm/xlnx/zynqmp_dp.c index 0a7b466446fb..a0606fab0e22 100644 --- a/drivers/gpu/drm/xlnx/zynqmp_dp.c +++ b/drivers/gpu/drm/xlnx/zynqmp_dp.c @@ -784,7 +784,7 @@ static int zynqmp_dp_link_train_ce(struct zynqmp_dp *dp) } /** - * zynqmp_dp_link_train - Train the link + * zynqmp_dp_train - Train the link * @dp: DisplayPort IP core structure * * Return: 0 if all trains are done successfully, or corresponding error code. diff --git a/drivers/gpu/drm/xlnx/zynqmp_dpsub.c b/drivers/gpu/drm/xlnx/zynqmp_dpsub.c index 9d64a8d98372..88eb33acd5f0 100644 --- a/drivers/gpu/drm/xlnx/zynqmp_dpsub.c +++ b/drivers/gpu/drm/xlnx/zynqmp_dpsub.c @@ -227,7 +227,9 @@ static int zynqmp_dpsub_probe(struct platform_device *pdev) dpsub->dev = &pdev->dev; platform_set_drvdata(pdev, dpsub); - dma_set_mask(dpsub->dev, DMA_BIT_MASK(ZYNQMP_DISP_MAX_DMA_BIT)); + ret = dma_set_mask(dpsub->dev, DMA_BIT_MASK(ZYNQMP_DISP_MAX_DMA_BIT)); + if (ret) + return ret; /* Try the reserved memory. Proceed if there's none. */ of_reserved_mem_device_init(&pdev->dev); |