diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu')
42 files changed, 429 insertions, 240 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig index a04f2fc7bf37..f6e5c0282fc1 100644 --- a/drivers/gpu/drm/amd/amdgpu/Kconfig +++ b/drivers/gpu/drm/amd/amdgpu/Kconfig @@ -27,8 +27,7 @@ config DRM_AMDGPU_CIK config DRM_AMDGPU_USERPTR bool "Always enable userptr write support" depends on DRM_AMDGPU - depends on ARCH_HAS_HMM - select HMM_MIRROR + depends on HMM_MIRROR help This option selects CONFIG_HMM and CONFIG_HMM_MIRROR if it isn't already selected to enabled full userptr support. diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 7bbcf1b20cfd..56e084367b93 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -54,7 +54,9 @@ amdgpu-y += amdgpu_device.o amdgpu_kms.o \ amdgpu_gtt_mgr.o amdgpu_vram_mgr.o amdgpu_virt.o amdgpu_atomfirmware.o \ amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \ amdgpu_gmc.o amdgpu_xgmi.o amdgpu_csa.o amdgpu_ras.o amdgpu_vm_cpu.o \ - amdgpu_vm_sdma.o amdgpu_pmu.o amdgpu_discovery.o + amdgpu_vm_sdma.o amdgpu_discovery.o + +amdgpu-$(CONFIG_PERF_EVENTS) += amdgpu_pmu.o # add asic specific block amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 596f7e07b5a8..8199d201b43a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -164,6 +164,7 @@ extern int amdgpu_async_gfx_ring; extern int amdgpu_mcbp; extern int amdgpu_discovery; extern int amdgpu_mes; +extern int amdgpu_noretry; #ifdef CONFIG_DRM_AMDGPU_SI extern int amdgpu_si_support; @@ -775,6 +776,7 @@ struct amdgpu_device { struct mutex grbm_idx_mutex; struct dev_pm_domain vga_pm_domain; bool have_disp_power_ref; + bool have_atomics_support; /* BIOS */ bool is_atom_fw; @@ -1216,6 +1218,10 @@ int amdgpu_dm_display_resume(struct amdgpu_device *adev ); static inline int amdgpu_dm_display_resume(struct amdgpu_device *adev) { return 0; } #endif + +void amdgpu_register_gpu_instance(struct amdgpu_device *adev); +void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev); + #include "amdgpu_object.h" /* used by df_v3_6.c and amdgpu_pmu.c */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index fab3eb173b05..9fa4f25a3745 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -668,6 +668,13 @@ bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid) return false; } +bool amdgpu_amdkfd_have_atomics_support(struct kgd_dev *kgd) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + + return adev->have_atomics_support; +} + #ifndef CONFIG_HSA_AMD bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 93a25c799d75..b6076d19e442 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -135,6 +135,7 @@ int amdgpu_amdkfd_submit_ib(struct kgd_dev *kgd, enum kgd_engine_type engine, uint32_t vmid, uint64_t gpu_addr, uint32_t *ib_cmd, uint32_t ib_len); void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle); +bool amdgpu_amdkfd_have_atomics_support(struct kgd_dev *kgd); struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void); struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 0aa81456ec32..1d3ee9c42f7e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -504,7 +504,7 @@ static int init_user_pages(struct kgd_mem *mem, struct mm_struct *mm, goto out; } - ret = amdgpu_ttm_tt_get_user_pages(bo->tbo.ttm, bo->tbo.ttm->pages); + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); if (ret) { pr_err("%s: Failed to get user pages: %d\n", __func__, ret); goto unregister_out; @@ -813,7 +813,7 @@ static int process_sync_pds_resv(struct amdkfd_process_info *process_info, ret = amdgpu_sync_resv(NULL, sync, pd->tbo.resv, - AMDGPU_FENCE_OWNER_UNDEFINED, false); + AMDGPU_FENCE_OWNER_KFD, false); if (ret) return ret; } @@ -1729,8 +1729,7 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, bo = mem->bo; /* Get updated user pages */ - ret = amdgpu_ttm_tt_get_user_pages(bo->tbo.ttm, - bo->tbo.ttm->pages); + ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); if (ret) { pr_debug("%s: Failed to get user pages: %d\n", __func__, ret); @@ -1740,6 +1739,12 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info, } amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm); + + /* Mark the BO as valid unless it was invalidated + * again concurrently. + */ + if (atomic_cmpxchg(&mem->invalid, invalid, 0) != invalid) + return -EAGAIN; } return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 37adce981fa3..e069de8b54e6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -633,7 +633,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, return -ENOMEM; } - r = amdgpu_ttm_tt_get_user_pages(bo->tbo.ttm, e->user_pages); + r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages); if (r) { kvfree(e->user_pages); e->user_pages = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 20ce158490db..6d54decef7f8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -106,10 +106,10 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, ssize_t result = 0; int r; bool pm_pg_lock, use_bank, use_ring; - unsigned instance_bank, sh_bank, se_bank, me, pipe, queue; + unsigned instance_bank, sh_bank, se_bank, me, pipe, queue, vmid; pm_pg_lock = use_bank = use_ring = false; - instance_bank = sh_bank = se_bank = me = pipe = queue = 0; + instance_bank = sh_bank = se_bank = me = pipe = queue = vmid = 0; if (size & 0x3 || *pos & 0x3 || ((*pos & (1ULL << 62)) && (*pos & (1ULL << 61)))) @@ -135,6 +135,7 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, me = (*pos & GENMASK_ULL(33, 24)) >> 24; pipe = (*pos & GENMASK_ULL(43, 34)) >> 34; queue = (*pos & GENMASK_ULL(53, 44)) >> 44; + vmid = (*pos & GENMASK_ULL(58, 54)) >> 54; use_ring = 1; } else { @@ -152,7 +153,7 @@ static int amdgpu_debugfs_process_reg_op(bool read, struct file *f, sh_bank, instance_bank); } else if (use_ring) { mutex_lock(&adev->srbm_mutex); - amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue); + amdgpu_gfx_select_me_pipe_q(adev, me, pipe, queue, vmid); } if (pm_pg_lock) @@ -185,7 +186,7 @@ end: amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); mutex_unlock(&adev->grbm_idx_mutex); } else if (use_ring) { - amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0); + amdgpu_gfx_select_me_pipe_q(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index e886be292f86..5a7f893cf724 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2537,6 +2537,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, hash_init(adev->mn_hash); mutex_init(&adev->lock_reset); mutex_init(&adev->virt.dpm_mutex); + mutex_init(&adev->psp.mutex); r = amdgpu_device_check_arguments(adev); if (r) @@ -2596,6 +2597,17 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (adev->rio_mem == NULL) DRM_INFO("PCI I/O BAR is not found.\n"); + /* enable PCIE atomic ops */ + r = pci_enable_atomic_ops_to_root(adev->pdev, + PCI_EXP_DEVCAP2_ATOMIC_COMP32 | + PCI_EXP_DEVCAP2_ATOMIC_COMP64); + if (r) { + adev->have_atomics_support = false; + DRM_INFO("PCIE atomic ops is not supported\n"); + } else { + adev->have_atomics_support = true; + } + amdgpu_device_get_pcie_info(adev); if (amdgpu_mcbp) @@ -2604,7 +2616,7 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (amdgpu_mes && adev->asic_type >= CHIP_NAVI10) adev->enable_mes = true; - if (amdgpu_discovery) { + if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) { r = amdgpu_discovery_init(adev); if (r) { dev_err(adev->dev, "amdgpu_discovery_init failed\n"); @@ -2798,7 +2810,8 @@ fence_driver_init: return r; } - r = amdgpu_pmu_init(adev); + if (IS_ENABLED(CONFIG_PERF_EVENTS)) + r = amdgpu_pmu_init(adev); if (r) dev_err(adev->dev, "amdgpu_pmu_init failed\n"); @@ -2870,9 +2883,10 @@ void amdgpu_device_fini(struct amdgpu_device *adev) amdgpu_debugfs_regs_cleanup(adev); device_remove_file(adev->dev, &dev_attr_pcie_replay_count); amdgpu_ucode_sysfs_fini(adev); - amdgpu_pmu_fini(adev); + if (IS_ENABLED(CONFIG_PERF_EVENTS)) + amdgpu_pmu_fini(adev); amdgpu_debugfs_preempt_cleanup(adev); - if (amdgpu_discovery) + if (amdgpu_discovery && adev->asic_type >= CHIP_NAVI10) amdgpu_discovery_fini(adev); } @@ -3559,6 +3573,12 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, if (vram_lost) amdgpu_device_fill_reset_magic(tmp_adev); + /* + * Add this ASIC as tracked as reset was already + * complete successfully. + */ + amdgpu_register_gpu_instance(tmp_adev); + r = amdgpu_device_ip_late_init(tmp_adev); if (r) goto out; @@ -3693,8 +3713,19 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, device_list_handle = &device_list; } + /* + * Mark these ASICs to be reseted as untracked first + * And add them back after reset completed + */ + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) + amdgpu_unregister_gpu_instance(tmp_adev); + /* block all schedulers and reset given job's ring */ list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + /* disable ras on ALL IPs */ + if (amdgpu_device_ip_need_full_reset(tmp_adev)) + amdgpu_ras_suspend(tmp_adev); + for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = tmp_adev->rings[i]; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index e049ae6a76fb..1481899f86c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -123,7 +123,7 @@ static int hw_id_map[MAX_HWIP] = { [UVD_HWIP] = UVD_HWID, [VCE_HWIP] = VCE_HWID, [DF_HWIP] = DF_HWID, - [DCE_HWIP] = DCEAZ_HWID, + [DCE_HWIP] = DMU_HWID, [OSSSYS_HWIP] = OSSSYS_HWID, [SMUIO_HWIP] = SMUIO_HWID, [PWR_HWIP] = PWR_HWID, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 3913a75924c6..f2e8b4238efd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -140,8 +140,9 @@ uint amdgpu_smu_memory_pool_size = 0; uint amdgpu_dc_feature_mask = 0; int amdgpu_async_gfx_ring = 1; int amdgpu_mcbp = 0; -int amdgpu_discovery = 0; +int amdgpu_discovery = -1; int amdgpu_mes = 0; +int amdgpu_noretry; struct amdgpu_mgpu_info mgpu_info = { .mutex = __MUTEX_INITIALIZER(mgpu_info.mutex), @@ -247,7 +248,8 @@ module_param_named(msi, amdgpu_msi, int, 0444); * By default(with no lockup_timeout settings), the timeout for all non-compute(GFX, SDMA and Video) * jobs is 10000. And there is no timeout enforced on compute jobs. */ -MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: 10000 for non-compute jobs and no timeout for compute jobs), " +MODULE_PARM_DESC(lockup_timeout, "GPU lockup timeout in ms (default: 10000 for non-compute jobs and infinity timeout for compute jobs." + " 0: keep default value. negative: infinity timeout), " "format is [Non-Compute] or [GFX,Compute,SDMA,Video]"); module_param_string(lockup_timeout, amdgpu_lockup_timeout, sizeof(amdgpu_lockup_timeout), 0444); @@ -581,18 +583,36 @@ MODULE_PARM_DESC(async_gfx_ring, "Asynchronous GFX rings that could be configured with either different priorities (HP3D ring and LP3D ring), or equal priorities (0 = disabled, 1 = enabled (default))"); module_param_named(async_gfx_ring, amdgpu_async_gfx_ring, int, 0444); +/** + * DOC: mcbp (int) + * It is used to enable mid command buffer preemption. (0 = disabled (default), 1 = enabled) + */ MODULE_PARM_DESC(mcbp, "Enable Mid-command buffer preemption (0 = disabled (default), 1 = enabled)"); module_param_named(mcbp, amdgpu_mcbp, int, 0444); +/** + * DOC: discovery (int) + * Allow driver to discover hardware IP information from IP Discovery table at the top of VRAM. + * (-1 = auto (default), 0 = disabled, 1 = enabled) + */ MODULE_PARM_DESC(discovery, "Allow driver to discover hardware IPs from IP Discovery table at the top of VRAM"); module_param_named(discovery, amdgpu_discovery, int, 0444); +/** + * DOC: mes (int) + * Enable Micro Engine Scheduler. This is a new hw scheduling engine for gfx, sdma, and compute. + * (0 = disabled (default), 1 = enabled) + */ MODULE_PARM_DESC(mes, "Enable Micro Engine Scheduler (0 = disabled (default), 1 = enabled)"); module_param_named(mes, amdgpu_mes, int, 0444); +MODULE_PARM_DESC(noretry, + "Disable retry faults (0 = retry enabled (default), 1 = retry disabled)"); +module_param_named(noretry, amdgpu_noretry, int, 0644); + #ifdef CONFIG_HSA_AMD /** * DOC: sched_policy (int) @@ -669,17 +689,6 @@ MODULE_PARM_DESC(ignore_crat, "Ignore CRAT table during KFD initialization (0 = use CRAT (default), 1 = ignore CRAT)"); /** - * DOC: noretry (int) - * This parameter sets sh_mem_config.retry_disable. Default value, 0, enables retry. - * Setting 1 disables retry. - * Retry is needed for recoverable page faults. - */ -int noretry; -module_param(noretry, int, 0644); -MODULE_PARM_DESC(noretry, - "Set sh_mem_config.retry_disable on Vega10 (0 = retry enabled (default), 1 = retry disabled)"); - -/** * 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. @@ -1302,7 +1311,8 @@ int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev) * By default timeout for non compute jobs is 10000. * And there is no timeout enforced on compute jobs. */ - adev->gfx_timeout = adev->sdma_timeout = adev->video_timeout = 10000; + adev->gfx_timeout = msecs_to_jiffies(10000); + adev->sdma_timeout = adev->video_timeout = adev->gfx_timeout; adev->compute_timeout = MAX_SCHEDULE_TIMEOUT; if (strnlen(input, AMDGPU_MAX_TIMEOUT_PARAM_LENTH)) { @@ -1312,10 +1322,13 @@ int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev) if (ret) return ret; - /* Invalidate 0 and negative values */ - if (timeout <= 0) { + if (timeout == 0) { index++; continue; + } else if (timeout < 0) { + timeout = MAX_SCHEDULE_TIMEOUT; + } else { + timeout = msecs_to_jiffies(timeout); } switch (index++) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h index dad2186f4ed5..df8a23554831 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gds.h @@ -31,7 +31,8 @@ struct amdgpu_gds { uint32_t gds_size; uint32_t gws_size; uint32_t oa_size; - uint32_t gds_compute_max_wave_id; + uint32_t gds_compute_max_wave_id; + uint32_t vgt_gs_max_wave_id; }; struct amdgpu_gds_reg_offset { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 1f9f27061e2f..939f8305511b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -327,8 +327,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data, } if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) { - r = amdgpu_ttm_tt_get_user_pages(bo->tbo.ttm, - bo->tbo.ttm->pages); + r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages); if (r) goto release_object; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index f96407ba9770..1199b5828b90 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -195,7 +195,7 @@ struct amdgpu_gfx_funcs { uint32_t wave, uint32_t start, uint32_t size, uint32_t *dst); void (*select_me_pipe_q)(struct amdgpu_device *adev, u32 me, u32 pipe, - u32 queue); + u32 queue, u32 vmid); }; struct amdgpu_ngg_buf { @@ -327,7 +327,7 @@ struct amdgpu_gfx { #define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev)) #define amdgpu_gfx_select_se_sh(adev, se, sh, instance) (adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance)) -#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q)) +#define amdgpu_gfx_select_me_pipe_q(adev, me, pipe, q, vmid) (adev)->gfx.funcs->select_me_pipe_q((adev), (me), (pipe), (q), (vmid)) /** * amdgpu_gfx_create_bitmask - create a bitmask diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 5832cd8f4ff1..0cf7e8606fd3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -44,7 +44,7 @@ #include "amdgpu_display.h" #include "amdgpu_ras.h" -static void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev) +void amdgpu_unregister_gpu_instance(struct amdgpu_device *adev) { struct amdgpu_gpu_instance *gpu_instance; int i; @@ -105,7 +105,7 @@ done_free: dev->dev_private = NULL; } -static void amdgpu_register_gpu_instance(struct amdgpu_device *adev) +void amdgpu_register_gpu_instance(struct amdgpu_device *adev) { struct amdgpu_gpu_instance *gpu_instance; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c index 623f56a1485f..3971c201f320 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c @@ -45,49 +45,12 @@ #include <linux/firmware.h> #include <linux/module.h> -#include <linux/hmm.h> -#include <linux/interval_tree.h> - #include <drm/drm.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" /** - * struct amdgpu_mn - * - * @adev: amdgpu device pointer - * @mm: process address space - * @type: type of MMU notifier - * @work: destruction work item - * @node: hash table node to find structure by adev and mn - * @lock: rw semaphore protecting the notifier nodes - * @objects: interval tree containing amdgpu_mn_nodes - * @mirror: HMM mirror function support - * - * Data for each amdgpu device and process address space. - */ -struct amdgpu_mn { - /* constant after initialisation */ - struct amdgpu_device *adev; - struct mm_struct *mm; - enum amdgpu_mn_type type; - - /* only used on destruction */ - struct work_struct work; - - /* protected by adev->mn_lock */ - struct hlist_node node; - - /* objects protected by lock */ - struct rw_semaphore lock; - struct rb_root_cached objects; - - /* HMM mirror */ - struct hmm_mirror mirror; -}; - -/** * struct amdgpu_mn_node * * @it: interval node defining start-last of the affected address range diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.h index f5b67c63ed6b..b8ed68943625 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.h @@ -24,17 +24,53 @@ #ifndef __AMDGPU_MN_H__ #define __AMDGPU_MN_H__ -/* - * HMM mirror - */ -struct amdgpu_mn; -struct hmm_range; +#include <linux/types.h> +#include <linux/hmm.h> +#include <linux/rwsem.h> +#include <linux/workqueue.h> +#include <linux/interval_tree.h> enum amdgpu_mn_type { AMDGPU_MN_TYPE_GFX, AMDGPU_MN_TYPE_HSA, }; +/** + * struct amdgpu_mn + * + * @adev: amdgpu device pointer + * @mm: process address space + * @type: type of MMU notifier + * @work: destruction work item + * @node: hash table node to find structure by adev and mn + * @lock: rw semaphore protecting the notifier nodes + * @objects: interval tree containing amdgpu_mn_nodes + * @mirror: HMM mirror function support + * + * Data for each amdgpu device and process address space. + */ +struct amdgpu_mn { + /* constant after initialisation */ + struct amdgpu_device *adev; + struct mm_struct *mm; + enum amdgpu_mn_type type; + + /* only used on destruction */ + struct work_struct work; + + /* protected by adev->mn_lock */ + struct hlist_node node; + + /* objects protected by lock */ + struct rw_semaphore lock; + struct rb_root_cached objects; + +#ifdef CONFIG_HMM_MIRROR + /* HMM mirror */ + struct hmm_mirror mirror; +#endif +}; + #if defined(CONFIG_HMM_MIRROR) void amdgpu_mn_lock(struct amdgpu_mn *mn); void amdgpu_mn_unlock(struct amdgpu_mn *mn); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index 193d53720d9b..8b7efd0a7028 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -2077,11 +2077,6 @@ static ssize_t amdgpu_hwmon_show_sclk(struct device *dev, (ddev->switch_power_state != DRM_SWITCH_POWER_ON)) return -EINVAL; - /* sanity check PP is enabled */ - if (!(adev->powerplay.pp_funcs && - adev->powerplay.pp_funcs->read_sensor)) - return -EINVAL; - /* get the sclk */ r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_SCLK, (void *)&sclk, &size); @@ -2112,11 +2107,6 @@ static ssize_t amdgpu_hwmon_show_mclk(struct device *dev, (ddev->switch_power_state != DRM_SWITCH_POWER_ON)) return -EINVAL; - /* sanity check PP is enabled */ - if (!(adev->powerplay.pp_funcs && - adev->powerplay.pp_funcs->read_sensor)) - return -EINVAL; - /* get the sclk */ r = amdgpu_dpm_read_sensor(adev, AMDGPU_PP_SENSOR_GFX_MCLK, (void *)&mclk, &size); @@ -2996,13 +2986,10 @@ void amdgpu_pm_compute_clocks(struct amdgpu_device *adev) } if (is_support_sw_smu(adev)) { - struct smu_context *smu = &adev->smu; struct smu_dpm_context *smu_dpm = &adev->smu.smu_dpm; - mutex_lock(&(smu->mutex)); smu_handle_task(&adev->smu, smu_dpm->dpm_level, AMD_PP_TASK_DISPLAY_CONFIG_CHANGE); - mutex_unlock(&(smu->mutex)); } else { if (adev->powerplay.pp_funcs->dispatch_tasks) { if (!amdgpu_device_has_dc_support(adev)) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index e69ad6e089c5..c027e5e7713e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -130,6 +130,8 @@ psp_cmd_submit_buf(struct psp_context *psp, int index; int timeout = 2000; + mutex_lock(&psp->mutex); + memset(psp->cmd_buf_mem, 0, PSP_CMD_BUFFER_SIZE); memcpy(psp->cmd_buf_mem, cmd, sizeof(struct psp_gfx_cmd_resp)); @@ -139,6 +141,7 @@ psp_cmd_submit_buf(struct psp_context *psp, fence_mc_addr, index); if (ret) { atomic_dec(&psp->fence_value); + mutex_unlock(&psp->mutex); return ret; } @@ -161,8 +164,10 @@ psp_cmd_submit_buf(struct psp_context *psp, ucode->ucode_id); DRM_WARN("psp command failed and response status is (%d)\n", psp->cmd_buf_mem->resp.status); - if (!timeout) + if (!timeout) { + mutex_unlock(&psp->mutex); return -EINVAL; + } } /* get xGMI session id from response buffer */ @@ -172,6 +177,7 @@ psp_cmd_submit_buf(struct psp_context *psp, ucode->tmr_mc_addr_lo = psp->cmd_buf_mem->resp.fw_addr_lo; ucode->tmr_mc_addr_hi = psp->cmd_buf_mem->resp.fw_addr_hi; } + mutex_unlock(&psp->mutex); return ret; } @@ -763,6 +769,15 @@ static int psp_hw_start(struct psp_context *psp) int ret; if (!amdgpu_sriov_vf(adev) || !adev->in_gpu_reset) { + if (psp->kdb_bin_size && + (psp->funcs->bootloader_load_kdb != NULL)) { + ret = psp_bootloader_load_kdb(psp); + if (ret) { + DRM_ERROR("PSP load kdb failed!\n"); + return ret; + } + } + ret = psp_bootloader_load_sysdrv(psp); if (ret) { DRM_ERROR("PSP load sysdrv failed!\n"); @@ -1188,10 +1203,16 @@ failed: int psp_gpu_reset(struct amdgpu_device *adev) { + int ret; + if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) return 0; - return psp_mode1_reset(&adev->psp); + mutex_lock(&adev->psp.mutex); + ret = psp_mode1_reset(&adev->psp); + mutex_unlock(&adev->psp.mutex); + + return ret; } int psp_rlc_autoload_start(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 6039acc84346..e0fc2a790e53 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -42,6 +42,12 @@ struct psp_context; struct psp_xgmi_node_info; struct psp_xgmi_topology_info; +enum psp_bootloader_cmd { + PSP_BL__LOAD_SYSDRV = 0x10000, + PSP_BL__LOAD_SOSDRV = 0x20000, + PSP_BL__LOAD_KEY_DATABASE = 0x80000, +}; + enum psp_ring_type { PSP_RING_TYPE__INVALID = 0, @@ -73,6 +79,7 @@ enum psp_reg_prog_id { struct psp_funcs { int (*init_microcode)(struct psp_context *psp); + int (*bootloader_load_kdb)(struct psp_context *psp); int (*bootloader_load_sysdrv)(struct psp_context *psp); int (*bootloader_load_sos)(struct psp_context *psp); int (*ring_init)(struct psp_context *psp, enum psp_ring_type ring_type); @@ -156,9 +163,11 @@ struct psp_context uint32_t sys_bin_size; uint32_t sos_bin_size; uint32_t toc_bin_size; + uint32_t kdb_bin_size; uint8_t *sys_start_addr; uint8_t *sos_start_addr; uint8_t *toc_start_addr; + uint8_t *kdb_start_addr; /* tmr buffer */ struct amdgpu_bo *tmr_bo; @@ -201,6 +210,7 @@ struct psp_context uint8_t *ta_ras_start_addr; struct psp_xgmi_context xgmi_context; struct psp_ras_context ras; + struct mutex mutex; }; struct amdgpu_psp_funcs { @@ -219,6 +229,8 @@ struct amdgpu_psp_funcs { (psp)->funcs->compare_sram_data((psp), (ucode), (type)) #define psp_init_microcode(psp) \ ((psp)->funcs->init_microcode ? (psp)->funcs->init_microcode((psp)) : 0) +#define psp_bootloader_load_kdb(psp) \ + ((psp)->funcs->bootloader_load_kdb ? (psp)->funcs->bootloader_load_kdb((psp)) : 0) #define psp_bootloader_load_sysdrv(psp) \ ((psp)->funcs->bootloader_load_sysdrv ? (psp)->funcs->bootloader_load_sysdrv((psp)) : 0) #define psp_bootloader_load_sos(psp) \ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index c9faa69cd677..e51b48ac48eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -391,6 +391,7 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, src_node_start = amdgpu_mm_node_addr(src->bo, ++src_mm, src->mem); src_node_size = (src_mm->size << PAGE_SHIFT); + src_page_offset = 0; } else { src_node_start += cur_size; src_page_offset = src_node_start & (PAGE_SIZE - 1); @@ -400,6 +401,7 @@ int amdgpu_ttm_copy_mem_to_mem(struct amdgpu_device *adev, dst_node_start = amdgpu_mm_node_addr(dst->bo, ++dst_mm, dst->mem); dst_node_size = (dst_mm->size << PAGE_SHIFT); + dst_page_offset = 0; } else { dst_node_start += cur_size; dst_page_offset = dst_node_start & (PAGE_SIZE - 1); @@ -487,6 +489,7 @@ static int amdgpu_move_vram_ram(struct ttm_buffer_object *bo, bool evict, placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT; r = ttm_bo_mem_space(bo, &placement, &tmp_mem, ctx); if (unlikely(r)) { + pr_err("Failed to find GTT space for blit from VRAM\n"); return r; } @@ -545,6 +548,7 @@ static int amdgpu_move_ram_vram(struct ttm_buffer_object *bo, bool evict, placements.flags = TTM_PL_MASK_CACHING | TTM_PL_FLAG_TT; r = ttm_bo_mem_space(bo, &placement, &tmp_mem, ctx); if (unlikely(r)) { + pr_err("Failed to find GTT space for blit to VRAM\n"); return r; } @@ -565,6 +569,30 @@ out_cleanup: } /** + * amdgpu_mem_visible - Check that memory can be accessed by ttm_bo_move_memcpy + * + * Called by amdgpu_bo_move() + */ +static bool amdgpu_mem_visible(struct amdgpu_device *adev, + struct ttm_mem_reg *mem) +{ + struct drm_mm_node *nodes = mem->mm_node; + + if (mem->mem_type == TTM_PL_SYSTEM || + mem->mem_type == TTM_PL_TT) + return true; + if (mem->mem_type != TTM_PL_VRAM) + return false; + + /* ttm_mem_reg_ioremap only supports contiguous memory */ + if (nodes->size != mem->num_pages) + return false; + + return ((nodes->start + nodes->size) << PAGE_SHIFT) + <= adev->gmc.visible_vram_size; +} + +/** * amdgpu_bo_move - Move a buffer object to a new memory location * * Called by ttm_bo_handle_move_mem() @@ -608,8 +636,10 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict, return 0; } - if (!adev->mman.buffer_funcs_enabled) + if (!adev->mman.buffer_funcs_enabled) { + r = -ENODEV; goto memcpy; + } if (old_mem->mem_type == TTM_PL_VRAM && new_mem->mem_type == TTM_PL_SYSTEM) { @@ -624,10 +654,16 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict, if (r) { memcpy: - r = ttm_bo_move_memcpy(bo, ctx, new_mem); - if (r) { + /* Check that all memory is CPU accessible */ + if (!amdgpu_mem_visible(adev, old_mem) || + !amdgpu_mem_visible(adev, new_mem)) { + pr_err("Move buffer fallback to memcpy unavailable\n"); return r; } + + r = ttm_bo_move_memcpy(bo, ctx, new_mem); + if (r) + return r; } if (bo->type == ttm_bo_type_device && @@ -731,8 +767,10 @@ struct amdgpu_ttm_tt { #define MAX_RETRY_HMM_RANGE_FAULT 16 -int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) +int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages) { + struct hmm_mirror *mirror = bo->mn ? &bo->mn->mirror : NULL; + struct ttm_tt *ttm = bo->tbo.ttm; struct amdgpu_ttm_tt *gtt = (void *)ttm; struct mm_struct *mm = gtt->usertask->mm; unsigned long start = gtt->userptr; @@ -746,6 +784,12 @@ int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) if (!mm) /* Happens during process shutdown */ return -ESRCH; + if (unlikely(!mirror)) { + DRM_DEBUG_DRIVER("Failed to get hmm_mirror\n"); + r = -EFAULT; + goto out; + } + vma = find_vma(mm, start); if (unlikely(!vma || start < vma->vm_start)) { r = -EFAULT; @@ -775,7 +819,7 @@ int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) 0 : range->flags[HMM_PFN_WRITE]; range->pfn_flags_mask = 0; range->pfns = pfns; - hmm_range_register(range, mm, start, + hmm_range_register(range, mirror, start, start + ttm->num_pages * PAGE_SIZE, PAGE_SHIFT); retry: @@ -2051,9 +2095,9 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, mm_node = bo->tbo.mem.mm_node; num_loops = 0; while (num_pages) { - uint32_t byte_count = mm_node->size << PAGE_SHIFT; + uint64_t byte_count = mm_node->size << PAGE_SHIFT; - num_loops += DIV_ROUND_UP(byte_count, max_bytes); + num_loops += DIV_ROUND_UP_ULL(byte_count, max_bytes); num_pages -= mm_node->size; ++mm_node; } @@ -2079,12 +2123,13 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo, mm_node = bo->tbo.mem.mm_node; while (num_pages) { - uint32_t byte_count = mm_node->size << PAGE_SHIFT; + uint64_t byte_count = mm_node->size << PAGE_SHIFT; uint64_t dst_addr; dst_addr = amdgpu_mm_node_addr(&bo->tbo, mm_node, &bo->tbo.mem); while (byte_count) { - uint32_t cur_size_in_bytes = min(byte_count, max_bytes); + uint32_t cur_size_in_bytes = min_t(uint64_t, byte_count, + max_bytes); amdgpu_emit_fill_buffer(adev, &job->ibs[0], src_data, dst_addr, cur_size_in_bytes); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index c2b7669004ba..caa76c693700 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -102,10 +102,11 @@ int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo); int amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo); #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR) -int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages); +int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages); bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm); #else -static inline int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages) +static inline int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, + struct page **pages) { return -EPERM; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c index c352a519ddd4..bfaa0eac3213 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.c @@ -262,6 +262,12 @@ void amdgpu_ucode_print_psp_hdr(const struct common_firmware_header *hdr) le32_to_cpu(psp_hdr_v1_1->toc_offset_bytes)); DRM_DEBUG("toc_size_bytes: %u\n", le32_to_cpu(psp_hdr_v1_1->toc_size_bytes)); + DRM_DEBUG("kdb_header_version: %u\n", + le32_to_cpu(psp_hdr_v1_1->kdb_header_version)); + DRM_DEBUG("kdb_offset_bytes: %u\n", + le32_to_cpu(psp_hdr_v1_1->kdb_offset_bytes)); + DRM_DEBUG("kdb_size_bytes: %u\n", + le32_to_cpu(psp_hdr_v1_1->kdb_size_bytes)); } } else { DRM_ERROR("Unknown PSP ucode version: %u.%u\n", diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h index f46944453c6e..c1fb6dc86440 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h @@ -85,6 +85,9 @@ struct psp_firmware_header_v1_1 { uint32_t toc_header_version; uint32_t toc_offset_bytes; uint32_t toc_size_bytes; + uint32_t kdb_header_version; + uint32_t kdb_offset_bytes; + uint32_t kdb_size_bytes; }; /* version_major=1, version_minor=0 */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 07a7e3820b7b..59dd204498c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -390,7 +390,8 @@ static uint32_t parse_clk(char *buf, bool min) if (!ptr) break; ptr+=2; - clk = simple_strtoul(ptr, NULL, 10); + if (kstrtou32(ptr, 10, &clk)) + return 0; } while (!min); return clk * 100; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 2932ade7dbd0..1675d5837c3c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -109,6 +109,13 @@ static const struct soc15_reg_golden golden_settings_gc_10_0_nv10[] = /* Pending on emulation bring up */ }; +#define DEFAULT_SH_MEM_CONFIG \ + ((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \ + (SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \ + (SH_MEM_RETRY_MODE_ALL << SH_MEM_CONFIG__RETRY_MODE__SHIFT) | \ + (3 << SH_MEM_CONFIG__INITIAL_INST_PREFETCH__SHIFT)) + + static void gfx_v10_0_set_ring_funcs(struct amdgpu_device *adev); static void gfx_v10_0_set_irq_funcs(struct amdgpu_device *adev); static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev); @@ -995,6 +1002,12 @@ static void gfx_v10_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, start + SQIND_WAVE_VGPRS_OFFSET, size, dst); } +static void gfx_v10_0_select_me_pipe_q(struct amdgpu_device *adev, + u32 me, u32 pipe, u32 q, u32 vm) + { + nv_grbm_select(adev, me, pipe, q, vm); + } + static const struct amdgpu_gfx_funcs gfx_v10_0_gfx_funcs = { .get_gpu_clock_counter = &gfx_v10_0_get_gpu_clock_counter, @@ -1002,6 +1015,7 @@ static const struct amdgpu_gfx_funcs gfx_v10_0_gfx_funcs = { .read_wave_data = &gfx_v10_0_read_wave_data, .read_wave_sgprs = &gfx_v10_0_read_wave_sgprs, .read_wave_vgprs = &gfx_v10_0_read_wave_vgprs, + .select_me_pipe_q = &gfx_v10_0_select_me_pipe_q, }; static void gfx_v10_0_gpu_early_init(struct amdgpu_device *adev) @@ -1408,7 +1422,6 @@ static u32 gfx_v10_0_init_pa_sc_tile_steering_override(struct amdgpu_device *ade static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev) { int i; - uint32_t sh_mem_config; uint32_t sh_mem_bases; /* @@ -1419,15 +1432,11 @@ static void gfx_v10_0_init_compute_vmid(struct amdgpu_device *adev) */ sh_mem_bases = DEFAULT_SH_MEM_BASES | (DEFAULT_SH_MEM_BASES << 16); - sh_mem_config = SH_MEM_ADDRESS_MODE_64 | - SH_MEM_ALIGNMENT_MODE_UNALIGNED << - SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT; - mutex_lock(&adev->srbm_mutex); for (i = FIRST_COMPUTE_VMID; i < LAST_COMPUTE_VMID; i++) { nv_grbm_select(adev, 0, 0, 0, i); /* CP and shaders */ - WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, sh_mem_config); + WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG); WREG32_SOC15(GC, 0, mmSH_MEM_BASES, sh_mem_bases); } nv_grbm_select(adev, 0, 0, 0, 0); @@ -1520,17 +1529,8 @@ static void gfx_v10_0_constants_init(struct amdgpu_device *adev) for (i = 0; i < adev->vm_manager.id_mgr[AMDGPU_GFXHUB].num_ids; i++) { nv_grbm_select(adev, 0, 0, 0, i); /* CP and shaders */ - if (i == 0) { - tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, - SH_MEM_ALIGNMENT_MODE_UNALIGNED); - tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_MODE, 0); - WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, tmp); - WREG32_SOC15(GC, 0, mmSH_MEM_BASES, 0); - } else { - tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, - SH_MEM_ALIGNMENT_MODE_UNALIGNED); - tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_MODE, 0); - WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, tmp); + WREG32_SOC15(GC, 0, mmSH_MEM_CONFIG, DEFAULT_SH_MEM_CONFIG); + if (i != 0) { tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE, (adev->gmc.private_aperture_start >> 48)); tmp = REG_SET_FIELD(tmp, SH_MEM_BASES, SHARED_BASE, @@ -1544,24 +1544,6 @@ static void gfx_v10_0_constants_init(struct amdgpu_device *adev) gfx_v10_0_init_compute_vmid(adev); - mutex_lock(&adev->grbm_idx_mutex); - /* - * making sure that the following register writes will be broadcasted - * to all the shaders - */ - gfx_v10_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); - - tmp = REG_SET_FIELD(0, PA_SC_FIFO_SIZE, SC_FRONTEND_PRIM_FIFO_SIZE, - adev->gfx.config.sc_prim_fifo_size_frontend); - tmp = REG_SET_FIELD(tmp, PA_SC_FIFO_SIZE, SC_BACKEND_PRIM_FIFO_SIZE, - adev->gfx.config.sc_prim_fifo_size_backend); - tmp = REG_SET_FIELD(tmp, PA_SC_FIFO_SIZE, SC_HIZ_TILE_FIFO_SIZE, - adev->gfx.config.sc_hiz_tile_fifo_size); - tmp = REG_SET_FIELD(tmp, PA_SC_FIFO_SIZE, SC_EARLYZ_TILE_FIFO_SIZE, - adev->gfx.config.sc_earlyz_tile_fifo_size); - WREG32_SOC15(GC, 0, mmPA_SC_FIFO_SIZE, tmp); - - mutex_unlock(&adev->grbm_idx_mutex); } static void gfx_v10_0_enable_gui_idle_interrupt(struct amdgpu_device *adev, @@ -4215,6 +4197,15 @@ static void gfx_v10_0_ring_emit_ib_gfx(struct amdgpu_ring *ring, unsigned vmid = AMDGPU_JOB_GET_VMID(job); u32 header, control = 0; + /* Prevent a hw deadlock due to a wave ID mismatch between ME and GDS. + * This resets the wave ID counters. (needed by transform feedback) + * TODO: This might only be needed on a VMID switch when we change + * the GDS OA mapping, not sure. + */ + amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); + amdgpu_ring_write(ring, mmVGT_GS_MAX_WAVE_ID); + amdgpu_ring_write(ring, ring->adev->gds.vgt_gs_max_wave_id); + if (ib->flags & AMDGPU_IB_FLAG_CE) header = PACKET3(PACKET3_INDIRECT_BUFFER_CNST, 2); else @@ -4252,6 +4243,22 @@ static void gfx_v10_0_ring_emit_ib_compute(struct amdgpu_ring *ring, unsigned vmid = AMDGPU_JOB_GET_VMID(job); u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24); + /* Currently, there is a high possibility to get wave ID mismatch + * between ME and GDS, leading to a hw deadlock, because ME generates + * different wave IDs than the GDS expects. This situation happens + * randomly when at least 5 compute pipes use GDS ordered append. + * The wave IDs generated by ME are also wrong after suspend/resume. + * Those are probably bugs somewhere else in the kernel driver. + * + * Writing GDS_COMPUTE_MAX_WAVE_ID resets wave ID counters in ME and + * GDS to 0 for this ring (me/pipe). + */ + if (ib->flags & AMDGPU_IB_FLAG_RESET_GDS_MAX_WAVE_ID) { + amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); + amdgpu_ring_write(ring, mmGDS_COMPUTE_MAX_WAVE_ID); + amdgpu_ring_write(ring, ring->adev->gds.gds_compute_max_wave_id); + } + amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2)); BUG_ON(ib->gpu_addr & 0x3); /* Dword align */ amdgpu_ring_write(ring, @@ -4278,11 +4285,7 @@ static void gfx_v10_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, amdgpu_ring_write(ring, PACKET3(PACKET3_RELEASE_MEM, 6)); amdgpu_ring_write(ring, (PACKET3_RELEASE_MEM_GCR_SEQ | PACKET3_RELEASE_MEM_GCR_GL2_WB | - PACKET3_RELEASE_MEM_GCR_GL2_INV | - PACKET3_RELEASE_MEM_GCR_GL2_US | - PACKET3_RELEASE_MEM_GCR_GL1_INV | - PACKET3_RELEASE_MEM_GCR_GLV_INV | - PACKET3_RELEASE_MEM_GCR_GLM_INV | + PACKET3_RELEASE_MEM_GCR_GLM_INV | /* must be set with GLM_WB */ PACKET3_RELEASE_MEM_GCR_GLM_WB | PACKET3_RELEASE_MEM_CACHE_POLICY(3) | PACKET3_RELEASE_MEM_EVENT_TYPE(CACHE_FLUSH_AND_INV_TS_EVENT) | @@ -4948,7 +4951,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { 5 + /* HDP_INVL */ 8 + 8 + /* FENCE x2 */ 2, /* SWITCH_BUFFER */ - .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_gfx */ + .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_gfx */ .emit_ib = gfx_v10_0_ring_emit_ib_gfx, .emit_fence = gfx_v10_0_ring_emit_fence, .emit_pipeline_sync = gfx_v10_0_ring_emit_pipeline_sync, @@ -4987,7 +4990,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v10_0_ring_emit_vm_flush */ 8 + 8 + 8, /* gfx_v10_0_ring_emit_fence x3 for user fence, vm fence */ - .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_compute */ + .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_compute */ .emit_ib = gfx_v10_0_ring_emit_ib_compute, .emit_fence = gfx_v10_0_ring_emit_fence, .emit_pipeline_sync = gfx_v10_0_ring_emit_pipeline_sync, @@ -5020,7 +5023,7 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_kiq = { SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v10_0_ring_emit_vm_flush */ 8 + 8 + 8, /* gfx_v10_0_ring_emit_fence_kiq x3 for user fence, vm fence */ - .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_compute */ + .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_compute */ .emit_ib = gfx_v10_0_ring_emit_ib_compute, .emit_fence = gfx_v10_0_ring_emit_fence_kiq, .test_ring = gfx_v10_0_ring_test_ring, @@ -5096,10 +5099,10 @@ static void gfx_v10_0_set_gds_init(struct amdgpu_device *adev) /* init asic gds info */ switch (adev->asic_type) { case CHIP_NAVI10: - adev->gds.gds_size = 0x10000; - break; default: adev->gds.gds_size = 0x10000; + adev->gds.gds_compute_max_wave_id = 0x4ff; + adev->gds.vgt_gs_max_wave_id = 0x3ff; break; } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index 789e900905e9..7f0a63628c43 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -3043,7 +3043,7 @@ static void gfx_v6_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v6_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q) + u32 me, u32 pipe, u32 q, u32 vm) { DRM_INFO("Not implemented\n"); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index 341b5024e598..0db9f488da7e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -4169,9 +4169,9 @@ static void gfx_v7_0_read_wave_sgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v7_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q) + u32 me, u32 pipe, u32 q, u32 vm) { - cik_srbm_select(adev, me, pipe, q, 0); + cik_srbm_select(adev, me, pipe, q, vm); } static const struct amdgpu_gfx_funcs gfx_v7_0_gfx_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index a57b5abf96a0..5f401b41ef7c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -3436,9 +3436,9 @@ static void gfx_v8_0_select_se_sh(struct amdgpu_device *adev, } static void gfx_v8_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q) + u32 me, u32 pipe, u32 q, u32 vm) { - vi_srbm_select(adev, me, pipe, q, 0); + vi_srbm_select(adev, me, pipe, q, vm); } static u32 gfx_v8_0_get_rb_active_bitmap(struct amdgpu_device *adev) @@ -3925,11 +3925,10 @@ static int gfx_v8_0_init_save_restore_list(struct amdgpu_device *adev) int list_size; unsigned int *register_list_format = - kmalloc(adev->gfx.rlc.reg_list_format_size_bytes, GFP_KERNEL); + kmemdup(adev->gfx.rlc.register_list_format, + adev->gfx.rlc.reg_list_format_size_bytes, GFP_KERNEL); if (!register_list_format) return -ENOMEM; - memcpy(register_list_format, adev->gfx.rlc.register_list_format, - adev->gfx.rlc.reg_list_format_size_bytes); gfx_v8_0_parse_ind_reg_list(register_list_format, RLC_FormatDirectRegListLength, diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 9b413f6fa588..f4c4eea62526 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -1313,9 +1313,9 @@ static void gfx_v9_0_read_wave_vgprs(struct amdgpu_device *adev, uint32_t simd, } static void gfx_v9_0_select_me_pipe_q(struct amdgpu_device *adev, - u32 me, u32 pipe, u32 q) + u32 me, u32 pipe, u32 q, u32 vm) { - soc15_grbm_select(adev, me, pipe, q, 0); + soc15_grbm_select(adev, me, pipe, q, vm); } static const struct amdgpu_gfx_funcs gfx_v9_0_gfx_funcs = { @@ -1942,11 +1942,15 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev) if (i == 0) { tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, SH_MEM_ALIGNMENT_MODE_UNALIGNED); + tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE, + !!amdgpu_noretry); WREG32_SOC15_RLC(GC, 0, mmSH_MEM_CONFIG, tmp); WREG32_SOC15_RLC(GC, 0, mmSH_MEM_BASES, 0); } else { tmp = REG_SET_FIELD(0, SH_MEM_CONFIG, ALIGNMENT_MODE, SH_MEM_ALIGNMENT_MODE_UNALIGNED); + tmp = REG_SET_FIELD(tmp, SH_MEM_CONFIG, RETRY_DISABLE, + !!amdgpu_noretry); WREG32_SOC15_RLC(GC, 0, mmSH_MEM_CONFIG, tmp); tmp = REG_SET_FIELD(0, SH_MEM_BASES, PRIVATE_BASE, (adev->gmc.private_aperture_start >> 48)); @@ -1960,25 +1964,6 @@ static void gfx_v9_0_constants_init(struct amdgpu_device *adev) mutex_unlock(&adev->srbm_mutex); gfx_v9_0_init_compute_vmid(adev); - - mutex_lock(&adev->grbm_idx_mutex); - /* - * making sure that the following register writes will be broadcasted - * to all the shaders - */ - gfx_v9_0_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff); - - WREG32_SOC15_RLC(GC, 0, mmPA_SC_FIFO_SIZE, - (adev->gfx.config.sc_prim_fifo_size_frontend << - PA_SC_FIFO_SIZE__SC_FRONTEND_PRIM_FIFO_SIZE__SHIFT) | - (adev->gfx.config.sc_prim_fifo_size_backend << - PA_SC_FIFO_SIZE__SC_BACKEND_PRIM_FIFO_SIZE__SHIFT) | - (adev->gfx.config.sc_hiz_tile_fifo_size << - PA_SC_FIFO_SIZE__SC_HIZ_TILE_FIFO_SIZE__SHIFT) | - (adev->gfx.config.sc_earlyz_tile_fifo_size << - PA_SC_FIFO_SIZE__SC_EARLYZ_TILE_FIFO_SIZE__SHIFT)); - mutex_unlock(&adev->grbm_idx_mutex); - } static void gfx_v9_0_wait_for_rlc_serdes(struct amdgpu_device *adev) @@ -2093,11 +2078,10 @@ static int gfx_v9_1_init_rlc_save_restore_list(struct amdgpu_device *adev) u32 tmp = 0; u32 *register_list_format = - kmalloc(adev->gfx.rlc.reg_list_format_size_bytes, GFP_KERNEL); + kmemdup(adev->gfx.rlc.register_list_format, + adev->gfx.rlc.reg_list_format_size_bytes, GFP_KERNEL); if (!register_list_format) return -ENOMEM; - memcpy(register_list_format, adev->gfx.rlc.register_list_format, - adev->gfx.rlc.reg_list_format_size_bytes); /* setup unique_indirect_regs array and indirect_start_offsets array */ unique_indirect_reg_count = ARRAY_SIZE(unique_indirect_regs); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index 9f0f189fc111..15986748f59f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -236,7 +236,8 @@ static void gfxhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) block_size); /* Send no-retry XNACK on fault to suppress VM fault storm. */ tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 1); + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, + !amdgpu_noretry); WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_CNTL, i, tmp); WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, i*2, 0); WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, i*2, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index b7de60a15623..d605b4963f8a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -215,7 +215,8 @@ static void gfxhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) adev->vm_manager.block_size - 9); /* Send no-retry XNACK on fault to suppress VM fault storm. */ tmp = REG_SET_FIELD(tmp, GCVM_CONTEXT1_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0); + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, + !amdgpu_noretry); WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_CNTL, i, tmp); WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, i*2, 0); WREG32_SOC15_OFFSET(GC, 0, mmGCVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, i*2, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index cec7c1fb14bf..5eeb72fcc123 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -245,8 +245,9 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, mutex_lock(&adev->mman.gtt_window_lock); gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB, 0); - if (!adev->mman.buffer_funcs_enabled || !adev->ib_pool_ready || - adev->asic_type != CHIP_NAVI10) { + if (!adev->mman.buffer_funcs_enabled || + !adev->ib_pool_ready || + adev->in_gpu_reset) { gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB, 0); mutex_unlock(&adev->mman.gtt_window_lock); return; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.h b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.h index 17b9b53fa892..9afd6ddb01e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v10_1.h +++ b/drivers/gpu/drm/amd/amdgpu/mes_v10_1.h @@ -22,7 +22,7 @@ */ #ifndef __MES_V10_1_H__ -#define __MES_v10_1_H__ +#define __MES_V10_1_H__ extern const struct amdgpu_ip_block_version mes_v10_1_ip_block; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index 05d1d448c8f5..dc5ce03034d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -265,7 +265,8 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) block_size); /* Send no-retry XNACK on fault to suppress VM fault storm. */ tmp = REG_SET_FIELD(tmp, VM_CONTEXT1_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 1); + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, + !amdgpu_noretry); WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_CNTL, i, tmp); WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, i*2, 0); WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, i*2, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index 37a1a318ae63..0f9549f19ade 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -205,7 +205,8 @@ static void mmhub_v2_0_setup_vmid_config(struct amdgpu_device *adev) adev->vm_manager.block_size - 9); /* Send no-retry XNACK on fault to suppress VM fault storm. */ tmp = REG_SET_FIELD(tmp, MMVM_CONTEXT1_CNTL, - RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, 0); + RETRY_PERMISSION_OR_INVALID_PAGE_FAULT, + !amdgpu_noretry); WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT1_CNTL, i, tmp); WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_LO32, i*2, 0); WREG32_SOC15_OFFSET(MMHUB, 0, mmMMVM_CONTEXT1_PAGE_TABLE_START_ADDR_HI32, i*2, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index af20ffb55c54..662612f89c70 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -31,6 +31,7 @@ #include "amdgpu_vce.h" #include "amdgpu_ucode.h" #include "amdgpu_psp.h" +#include "amdgpu_smu.h" #include "atom.h" #include "amd_pcie.h" @@ -255,6 +256,39 @@ static void nv_gpu_pci_config_reset(struct amdgpu_device *adev) } #endif +static int nv_asic_mode1_reset(struct amdgpu_device *adev) +{ + u32 i; + int ret = 0; + + amdgpu_atombios_scratch_regs_engine_hung(adev, true); + + dev_info(adev->dev, "GPU mode1 reset\n"); + + /* disable BM */ + pci_clear_master(adev->pdev); + + pci_save_state(adev->pdev); + + ret = psp_gpu_reset(adev); + if (ret) + dev_err(adev->dev, "GPU mode1 reset failed\n"); + + pci_restore_state(adev->pdev); + + /* wait for asic to come out of reset */ + for (i = 0; i < adev->usec_timeout; i++) { + u32 memsize = adev->nbio_funcs->get_memsize(adev); + + if (memsize != 0xffffffff) + break; + udelay(1); + } + + amdgpu_atombios_scratch_regs_engine_hung(adev, false); + + return ret; +} static int nv_asic_reset(struct amdgpu_device *adev) { @@ -266,8 +300,15 @@ static int nv_asic_reset(struct amdgpu_device *adev) amdgpu_atombios_scratch_regs_engine_hung(adev, false); #endif + int ret = 0; + struct smu_context *smu = &adev->smu; - return 0; + if (smu_baco_is_support(smu)) + ret = smu_baco_reset(smu); + else + ret = nv_asic_mode1_reset(adev); + + return ret; } static int nv_set_uvd_clocks(struct amdgpu_device *adev, u32 vclk, u32 dclk) @@ -348,8 +389,10 @@ int nv_set_ip_blocks(struct amdgpu_device *adev) amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block); if (adev->enable_virtual_display || amdgpu_sriov_vf(adev)) amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block); +#if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) amdgpu_device_ip_block_add(adev, &dm_ip_block); +#endif amdgpu_device_ip_block_add(adev, &gfx_v10_0_ip_block); amdgpu_device_ip_block_add(adev, &sdma_v5_0_ip_block); if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT && diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index 759b2768c51c..41b72588adcf 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -103,6 +103,9 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) adev->psp.toc_bin_size = le32_to_cpu(sos_hdr_v1_1->toc_size_bytes); adev->psp.toc_start_addr = (uint8_t *)adev->psp.sys_start_addr + le32_to_cpu(sos_hdr_v1_1->toc_offset_bytes); + adev->psp.kdb_bin_size = le32_to_cpu(sos_hdr_v1_1->kdb_size_bytes); + adev->psp.kdb_start_addr = (uint8_t *)adev->psp.sys_start_addr + + le32_to_cpu(sos_hdr_v1_1->kdb_offset_bytes); } break; default: @@ -177,6 +180,48 @@ out: return err; } +static int psp_v11_0_bootloader_load_kdb(struct psp_context *psp) +{ + int ret; + uint32_t psp_gfxdrv_command_reg = 0; + struct amdgpu_device *adev = psp->adev; + uint32_t sol_reg; + + /* Check tOS sign of life register to confirm sys driver and sOS + * are already been loaded. + */ + sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); + if (sol_reg) { + psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); + dev_info(adev->dev, "sos fw version = 0x%x.\n", psp->sos_fw_version); + return 0; + } + + /* 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, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + if (ret) + return ret; + + memset(psp->fw_pri_buf, 0, PSP_1_MEG); + + /* Copy PSP KDB binary to memory */ + memcpy(psp->fw_pri_buf, psp->kdb_start_addr, psp->kdb_bin_size); + + /* Provide the sys driver to bootloader */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, + (uint32_t)(psp->fw_pri_mc_addr >> 20)); + psp_gfxdrv_command_reg = PSP_BL__LOAD_KEY_DATABASE; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, + psp_gfxdrv_command_reg); + + /* 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, mmMP0_SMN_C2PMSG_35), + 0x80000000, 0x80000000, false); + + return ret; +} + static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp) { int ret; @@ -190,7 +235,7 @@ static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp) sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); if (sol_reg) { psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); - printk("sos fw version = 0x%x.\n", psp->sos_fw_version); + dev_info(adev->dev, "sos fw version = 0x%x.\n", psp->sos_fw_version); return 0; } @@ -208,7 +253,7 @@ static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp) /* Provide the sys driver to bootloader */ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = 1 << 16; + psp_gfxdrv_command_reg = PSP_BL__LOAD_SYSDRV; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, psp_gfxdrv_command_reg); @@ -249,7 +294,7 @@ static int psp_v11_0_bootloader_load_sos(struct psp_context *psp) /* Provide the PSP secure OS to bootloader */ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = 2 << 16; + psp_gfxdrv_command_reg = PSP_BL__LOAD_SOSDRV; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, psp_gfxdrv_command_reg); @@ -534,7 +579,7 @@ psp_v11_0_sram_map(struct amdgpu_device *adev, case AMDGPU_UCODE_ID_RLC_G: *sram_offset = 0x2000; - if (adev->asic_type != CHIP_NAVI10) { + if (adev->asic_type < CHIP_NAVI10) { *sram_addr_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_ADDR); *sram_data_reg_offset = SOC15_REG_OFFSET(GC, 0, mmRLC_GPM_UCODE_DATA); } else { @@ -545,7 +590,7 @@ psp_v11_0_sram_map(struct amdgpu_device *adev, case AMDGPU_UCODE_ID_SDMA0: *sram_offset = 0x0; - if (adev->asic_type != CHIP_NAVI10) { + if (adev->asic_type < CHIP_NAVI10) { *sram_addr_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_ADDR); *sram_data_reg_offset = SOC15_REG_OFFSET(SDMA0, 0, mmSDMA0_UCODE_DATA); } else { @@ -822,6 +867,7 @@ static int psp_v11_0_rlc_autoload_start(struct psp_context *psp) static const struct psp_funcs psp_v11_0_funcs = { .init_microcode = psp_v11_0_init_microcode, + .bootloader_load_kdb = psp_v11_0_bootloader_load_kdb, .bootloader_load_sysdrv = psp_v11_0_bootloader_load_sysdrv, .bootloader_load_sos = psp_v11_0_bootloader_load_sos, .ring_init = psp_v11_0_ring_init, diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index 2ea772692037..019c47feee42 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -155,7 +155,7 @@ static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp) /* Provide the sys driver to bootloader */ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = 1 << 16; + psp_gfxdrv_command_reg = PSP_BL__LOAD_SYSDRV; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, psp_gfxdrv_command_reg); @@ -218,7 +218,7 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp) /* Provide the PSP secure OS to bootloader */ WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36, (uint32_t)(psp->fw_pri_mc_addr >> 20)); - psp_gfxdrv_command_reg = 2 << 16; + psp_gfxdrv_command_reg = PSP_BL__LOAD_SOSDRV; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_35, psp_gfxdrv_command_reg); diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 87152d8ef0df..23265414d448 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -649,8 +649,6 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) #if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) amdgpu_device_ip_block_add(adev, &dm_ip_block); -#else -# warning "Enable CONFIG_DRM_AMD_DC for display support on SOC15." #endif if (!(adev->asic_type == CHIP_VEGA20 && amdgpu_sriov_vf(adev))) { amdgpu_device_ip_block_add(adev, &uvd_v7_0_ip_block); @@ -671,8 +669,6 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) #if defined(CONFIG_DRM_AMD_DC) else if (amdgpu_device_has_dc_support(adev)) amdgpu_device_ip_block_add(adev, &dm_ip_block); -#else -# warning "Enable CONFIG_DRM_AMD_DC for display support on SOC15." #endif amdgpu_device_ip_block_add(adev, &vcn_v1_0_ip_block); break; @@ -717,9 +713,15 @@ static void soc15_get_pcie_usage(struct amdgpu_device *adev, uint64_t *count0, return; /* Set the 2 events that we wish to watch, defined above */ - /* Reg 40 is # received msgs, Reg 104 is # of posted requests sent */ + /* Reg 40 is # received msgs */ perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT0_SEL, 40); - perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, EVENT1_SEL, 104); + /* Pre-VG20, Reg 104 is # of posted requests sent. On VG20 it's 108 */ + if (adev->asic_type == CHIP_VEGA20) + perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, + EVENT1_SEL, 108); + else + perfctr = REG_SET_FIELD(perfctr, PCIE_PERF_CNTL_TXCLK, + EVENT1_SEL, 104); /* Write to enable desired perf counters */ WREG32_PCIE(smnPCIE_PERF_CNTL_TXCLK, perfctr); diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index d40ed1a828dd..6575ddcfcf00 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -59,7 +59,6 @@ #include "vid.h" #include "vi.h" -#include "vi_dpm.h" #include "gmc_v8_0.h" #include "gmc_v7_0.h" #include "gfx_v8_0.h" diff --git a/drivers/gpu/drm/amd/amdgpu/vi_dpm.h b/drivers/gpu/drm/amd/amdgpu/vi_dpm.h deleted file mode 100644 index c43e03fddfba..000000000000 --- a/drivers/gpu/drm/amd/amdgpu/vi_dpm.h +++ /dev/null @@ -1,32 +0,0 @@ -/* - * Copyright 2014 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 __VI_DPM_H__ -#define __VI_DPM_H__ - -extern const struct amd_ip_funcs cz_dpm_ip_funcs; -int cz_smu_init(struct amdgpu_device *adev); -int cz_smu_start(struct amdgpu_device *adev); -int cz_smu_fini(struct amdgpu_device *adev); - -#endif |