diff options
407 files changed, 9428 insertions, 3844 deletions
diff --git a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv index 882d2518f8ed..3825f00ca9fe 100644 --- a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv +++ b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv @@ -16,6 +16,7 @@ Radeon (RX|TM) (PRO|WX) Vega /MI25 /V320 /V340L /8200 /9100 /SSG MxGPU, VEGA10, AMD Radeon (Pro) VII /MI50 /MI60, VEGA20, DCE 12, 9.4.0, VCE 4.1.0 / UVD 7.2.0, 4.2.0 MI100, ARCTURUS, *, 9.4.1, VCN 2.5.0, 4.2.2 MI200, ALDEBARAN, *, 9.4.2, VCN 2.6.0, 4.4.0 +MI300, AQUA_VANGARAM, *, 9.4.3, VCN 4.0.3, 4.4.2 AMD Radeon (RX|Pro) 5600(M|XT) /5700 (M|XT|XTB) /W5700, NAVI10, DCN 2.0.0, 10.1.10, VCN 2.0.0, 5.0.0 AMD Radeon (Pro) 5300 /5500XTB/5500(XT|M) /W5500M /W5500, NAVI14, DCN 2.0.0, 10.1.1, VCN 2.0.2, 5.0.2 AMD Radeon RX 6800(XT) /6900(XT) /W6800, SIENNA_CICHLID, DCN 3.0.0, 10.3.0, VCN 3.0.0, 5.2.0 @@ -23,4 +24,5 @@ AMD Radeon RX 6700 XT / 6800M / 6700M, NAVY_FLOUNDER, DCN 3.0.0, 10.3.2, VCN 3.0 AMD Radeon RX 6600(XT) /6600M /W6600 /W6600M, DIMGREY_CAVEFISH, DCN 3.0.2, 10.3.4, VCN 3.0.16, 5.2.4 AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3.0.33, 5.2.5 AMD Radeon RX 7900 XT /XTX, , DCN 3.2.0, 11.0.0, VCN 4.0.0, 6.0.0 +AMD Radeon RX 7800 XT, , DCN 3.2.0, 11.0.3, VCN 4.0.0, 6.0.3 AMD Radeon RX 7600M (XT) /7700S /7600S, , DCN 3.2.1, 11.0.2, VCN 4.0.4, 6.0.2 diff --git a/Documentation/gpu/amdgpu/display/dcn-blocks.rst b/Documentation/gpu/amdgpu/display/dcn-blocks.rst new file mode 100644 index 000000000000..a3fbd3ea028b --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dcn-blocks.rst @@ -0,0 +1,78 @@ +========== +DCN Blocks +========== + +In this section, you will find some extra details about some of the DCN blocks +and the code documentation when it is automatically generated. + +DCHUBBUB +-------- + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :export: + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :internal: + +HUBP +---- + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :export: + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :internal: + +DPP +--- + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :export: + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h + :internal: + +MPC +--- + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h + :export: + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h + :internal: + +OPP +--- + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/opp.h + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/opp.h + :export: + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/opp.h + :internal: + +DIO +--- + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h + :export: + +.. kernel-doc:: drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h + :internal: diff --git a/Documentation/gpu/amdgpu/display/display-contributing.rst b/Documentation/gpu/amdgpu/display/display-contributing.rst new file mode 100644 index 000000000000..fdb2bea01d53 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/display-contributing.rst @@ -0,0 +1,168 @@ +.. _display_todos: + +============================== +AMDGPU - Display Contributions +============================== + +First of all, if you are here, you probably want to give some technical +contribution to the display code, and for that, we say thank you :) + +This page summarizes some of the issues you can help with; keep in mind that +this is a static page, and it is always a good idea to try to reach developers +in the amdgfx or some of the maintainers. Finally, this page follows the DRM +way of creating a TODO list; for more information, check +'Documentation/gpu/todo.rst'. + +Gitlab issues +============= + +Users can report issues associated with AMD GPUs at: + +- https://gitlab.freedesktop.org/drm/amd + +Usually, we try to add a proper label to all new tickets to make it easy to +filter issues. If you can reproduce any problem, you could help by adding more +information or fixing the issue. + +Level: diverse + +IGT +=== + +`IGT`_ provides many integration tests that can be run on your GPU. We always +want to pass a large set of tests to increase the test coverage in our CI. If +you wish to contribute to the display code but are unsure where a good place +is, we recommend you run all IGT tests and try to fix any failure you see in +your hardware. Keep in mind that this failure can be an IGT problem or a kernel +issue; it is necessary to analyze case-by-case. + +Level: diverse + +.. _IGT: https://gitlab.freedesktop.org/drm/igt-gpu-tools + +Compilation +=========== + +Fix compilation warnings +------------------------ + +Enable the W1 or W2 warning level in the kernel compilation and try to fix the +issues on the display side. + +Level: Starter + +Fix compilation issues when using um architecture +------------------------------------------------- + +Linux has a User-mode Linux (UML) feature, and the kernel can be compiled to +the **um** architecture. Compiling for **um** can bring multiple advantages +from the test perspective. We currently have some compilation issues in this +area that we need to fix. + +Level: Intermediate + +Code Refactor +============= + +Add prefix to DC functions to improve the debug with ftrace +----------------------------------------------------------- + +The Ftrace debug feature (check 'Documentation/trace/ftrace.rst') is a +fantastic way to check the code path when developers try to make sense of a +bug. Ftrace provides a filter mechanism that can be useful when the developer +has some hunch of which part of the code can cause the issue; for this reason, +if a set of functions has a proper prefix, it becomes easy to create a good +filter. Additionally, prefixes can improve stack trace readability. + +The DC code does not follow some prefix rules, which makes the Ftrace filter +more complicated and reduces the readability of the stack trace. If you want +something simple to start contributing to the display, you can make patches for +adding prefixes to DC functions. To create those prefixes, use part of the file +name as a prefix for all functions in the target file. Check the +'amdgpu_dm_crtc.c` and `amdgpu_dm_plane.c` for some references. However, we +strongly advise not to send huge patches changing these prefixes; otherwise, it +will be hard to review and test, which can generate second thoughts from +maintainers. Try small steps; in case of double, you can ask before you put in +effort. We recommend first looking at folders like dceXYZ, dcnXYZ, basics, +bios, core, clk_mgr, hwss, resource, and irq. + +Level: Starter + +Reduce code duplication +----------------------- + +AMD has an extensive portfolio with various dGPUs and APUs that amdgpu +supports. To maintain the new hardware release cadence, DCE/DCN was designed in +a modular design, making the bring-up for new hardware fast. Over the years, +amdgpu accumulated some technical debt in the code duplication area. For this +task, it would be a good idea to find a tool that can discover code duplication +(including patterns) and use it as guidance to reduce duplications. + +Level: Intermediate + +Make atomic_commit_[check|tail] more readable +--------------------------------------------- + +The functions responsible for atomic commit and tail are intricate and +extensive. In particular `amdgpu_dm_atomic_commit_tail` is a long function and +could benefit from being split into smaller helpers. Improvements in this area +are more than welcome, but keep in mind that changes in this area will affect +all ASICs, meaning that refactoring requires a comprehensive verification; in +other words, this effort can take some time for validation. + +Level: Advanced + +Documentation +============= + +Expand kernel-doc +----------------- + +Many DC functions do not have a proper kernel-doc; understanding a function and +adding documentation is a great way to learn more about the amdgpu driver and +also leave an outstanding contribution to the entire community. + +Level: Starter + +Beyond AMDGPU +============= + +AMDGPU provides features that are not yet enabled in the userspace. This +section highlights some of the coolest display features, which could be enabled +with the userspace developer helper. + +Enable underlay +--------------- + +AMD display has this feature called underlay (which you can read more about at +'Documentation/GPU/amdgpu/display/mpo-overview.rst') which is intended to +save power when playing a video. The basic idea is to put a video in the +underlay plane at the bottom and the desktop in the plane above it with a hole +in the video area. This feature is enabled in ChromeOS, and from our data +measurement, it can save power. + +Level: Unknown + +Adaptive Backlight Modulation (ABM) +----------------------------------- + +ABM is a feature that adjusts the display panel's backlight level and pixel +values depending on the displayed image. This power-saving feature can be very +useful when the system starts to run off battery; since this will impact the +display output fidelity, it would be good if this option was something that +users could turn on or off. + +Level: Unknown + + +HDR & Color management & VRR +---------------------------- + +HDR, Color Management, and VRR are huge topics and it's hard to put these into +concise ToDos. If you are interested in this topic, we recommend checking some +blog posts from the community developers to better understand some of the +specific challenges and people working on the subject. If anyone wants to work +on some particular part, we can try to help with some basic guidance. Finally, +keep in mind that we already have some kernel-doc in place for those areas. + +Level: Unknown diff --git a/Documentation/gpu/amdgpu/display/display-manager.rst b/Documentation/gpu/amdgpu/display/display-manager.rst index be2651ecdd7f..67a811e6891f 100644 --- a/Documentation/gpu/amdgpu/display/display-manager.rst +++ b/Documentation/gpu/amdgpu/display/display-manager.rst @@ -132,9 +132,6 @@ The DRM blend mode and its elements are then mapped by AMDGPU display manager (MPC), as follows: .. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h - :doc: mpc-overview - -.. kernel-doc:: drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h :functions: mpcc_blnd_cfg Therefore, the blending configuration for a single MPCC instance on the MPC diff --git a/Documentation/gpu/amdgpu/display/index.rst b/Documentation/gpu/amdgpu/display/index.rst index f8a4f53d70d8..f0c342e00a39 100644 --- a/Documentation/gpu/amdgpu/display/index.rst +++ b/Documentation/gpu/amdgpu/display/index.rst @@ -7,18 +7,80 @@ drm/amd/display - Display Core (DC) AMD display engine is partially shared with other operating systems; for this reason, our Display Core Driver is divided into two pieces: -1. **Display Core (DC)** contains the OS-agnostic components. Things like +#. **Display Core (DC)** contains the OS-agnostic components. Things like hardware programming and resource management are handled here. -2. **Display Manager (DM)** contains the OS-dependent components. Hooks to the - amdgpu base driver and DRM are implemented here. +#. **Display Manager (DM)** contains the OS-dependent components. Hooks to the + amdgpu base driver and DRM are implemented here. For example, you can check + display/amdgpu_dm/ folder. + +------------------ +DC Code validation +------------------ + +Maintaining the same code base across multiple OSes requires a lot of +synchronization effort between repositories and exhaustive validation. In the +DC case, we maintain a tree to centralize code from different parts. The shared +repository has integration tests with our Internal Linux CI farm, and we run a +comprehensive set of IGT tests in various AMD GPUs/APUs (mostly recent dGPUs +and APUs). Our CI also checks ARM64/32, PPC64/32, and x86_64/32 compilation +with DCN enabled and disabled. + +When we upstream a new feature or some patches, we pack them in a patchset with +the prefix **DC Patches for <DATE>**, which is created based on the latest +`amd-staging-drm-next <https://gitlab.freedesktop.org/agd5f/linux>`_. All of +those patches are under a DC version tested as follows: + +* Ensure that every patch compiles and the entire series pass our set of IGT + test in different hardware. +* Prepare a branch with those patches for our validation team. If there is an + error, a developer will debug as fast as possible; usually, a simple bisect + in the series is enough to point to a bad change, and two possible actions + emerge: fix the issue or drop the patch. If it is not an easy fix, the bad + patch is dropped. +* Finally, developers wait a few days for community feedback before we merge + the series. + +It is good to stress that the test phase is something that we take extremely +seriously, and we never merge anything that fails our validation. Follows an +overview of our test set: + +#. Manual test + * Multiple Hotplugs with DP and HDMI. + * Stress test with multiple display configuration changes via the user interface. + * Validate VRR behaviour. + * Check PSR. + * Validate MPO when playing video. + * Test more than two displays connected at the same time. + * Check suspend/resume. + * Validate FPO. + * Check MST. +#. Automated test + * IGT tests in a farm with GPUs and APUs that support DCN and DCE. + * Compilation validation with the latest GCC and Clang from LTS distro. + * Cross-compilation for PowerPC 64/32, ARM 64/32, and x86 32. + +In terms of test setup for CI and manual tests, we usually use: + +#. The latest Ubuntu LTS. +#. In terms of userspace, we only use fully updated open-source components + provided by the distribution official package manager. +#. Regarding IGT, we use the latest code from the upstream. +#. Most of the manual tests are conducted in the GNome but we also use KDE. + +Notice that someone from our test team will always reply to the cover letter +with the test report. + +-------------- +DC Information +-------------- The display pipe is responsible for "scanning out" a rendered frame from the GPU memory (also called VRAM, FrameBuffer, etc.) to a display. In other words, it would: -1. Read frame information from memory; -2. Perform required transformation; -3. Send pixel data to sink devices. +#. Read frame information from memory; +#. Perform required transformation; +#. Send pixel data to sink devices. If you want to learn more about our driver details, take a look at the below table of content: @@ -26,7 +88,9 @@ table of content: .. toctree:: display-manager.rst - dc-debug.rst dcn-overview.rst + dcn-blocks.rst mpo-overview.rst + dc-debug.rst + display-contributing.rst dc-glossary.rst diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile index 260e32ef7bae..4c989da4d2f3 100644 --- a/drivers/gpu/drm/amd/amdgpu/Makefile +++ b/drivers/gpu/drm/amd/amdgpu/Makefile @@ -80,7 +80,7 @@ amdgpu-y += amdgpu_device.o amdgpu_doorbell_mgr.o amdgpu_kms.o \ amdgpu_umc.o smu_v11_0_i2c.o amdgpu_fru_eeprom.o amdgpu_rap.o \ amdgpu_fw_attestation.o amdgpu_securedisplay.o \ amdgpu_eeprom.o amdgpu_mca.o amdgpu_psp_ta.o amdgpu_lsdma.o \ - amdgpu_ring_mux.o amdgpu_xcp.o amdgpu_seq64.o + amdgpu_ring_mux.o amdgpu_xcp.o amdgpu_seq64.o amdgpu_aca.o amdgpu-$(CONFIG_PROC_FS) += amdgpu_fdinfo.o diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 3d8a48f46b01..312dfaec7b4a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -107,6 +107,7 @@ #include "amdgpu_smuio.h" #include "amdgpu_fdinfo.h" #include "amdgpu_mca.h" +#include "amdgpu_aca.h" #include "amdgpu_ras.h" #include "amdgpu_xcp.h" #include "amdgpu_seq64.h" @@ -114,14 +115,12 @@ #define MAX_GPU_INSTANCE 64 -struct amdgpu_gpu_instance -{ +struct amdgpu_gpu_instance { struct amdgpu_device *adev; int mgpu_fan_enabled; }; -struct amdgpu_mgpu_info -{ +struct amdgpu_mgpu_info { struct amdgpu_gpu_instance gpu_ins[MAX_GPU_INSTANCE]; struct mutex mutex; uint32_t num_gpu; @@ -140,8 +139,7 @@ enum amdgpu_ss { AMDGPU_SS_DRV_UNLOAD }; -struct amdgpu_watchdog_timer -{ +struct amdgpu_watchdog_timer { bool timeout_fatal_disable; uint32_t period; /* maxCycles = (1 << period), the number of cycles before a timeout */ }; @@ -1045,6 +1043,9 @@ struct amdgpu_device { /* MCA */ struct amdgpu_mca mca; + /* ACA */ + struct amdgpu_aca aca; + struct amdgpu_ip_block ip_blocks[AMDGPU_MAX_IP_NUM]; uint32_t harvest_ip_mask; int num_ip_blocks; @@ -1078,6 +1079,8 @@ struct amdgpu_device { bool in_s3; bool in_s4; bool in_s0ix; + /* indicate amdgpu suspension status */ + bool suspend_complete; enum pp_mp1_state mp1_state; struct amdgpu_doorbell_index doorbell_index; @@ -1329,6 +1332,7 @@ int emu_soc_asic_init(struct amdgpu_device *adev); #define WREG32_FIELD_OFFSET(reg, offset, field, val) \ WREG32(mm##reg + offset, (RREG32(mm##reg + offset) & ~REG_FIELD_MASK(reg, field)) | (val) << REG_FIELD_SHIFT(reg, field)) +#define AMDGPU_GET_REG_FIELD(x, h, l) (((x) & GENMASK_ULL(h, l)) >> (l)) /* * BIOS helpers. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c new file mode 100644 index 000000000000..493982f94649 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c @@ -0,0 +1,879 @@ +/* + * 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/list.h> +#include "amdgpu.h" +#include "amdgpu_aca.h" +#include "amdgpu_ras.h" + +#define ACA_BANK_HWID(type, hwid, mcatype) [ACA_HWIP_TYPE_##type] = {hwid, mcatype} + +typedef int bank_handler_t(struct aca_handle *handle, struct aca_bank *bank, enum aca_error_type type, void *data); + +struct aca_banks { + int nr_banks; + struct list_head list; +}; + +struct aca_hwip { + int hwid; + int mcatype; +}; + +static struct aca_hwip aca_hwid_mcatypes[ACA_HWIP_TYPE_COUNT] = { + ACA_BANK_HWID(SMU, 0x01, 0x01), + ACA_BANK_HWID(PCS_XGMI, 0x50, 0x00), + ACA_BANK_HWID(UMC, 0x96, 0x00), +}; + +static void aca_banks_init(struct aca_banks *banks) +{ + if (!banks) + return; + + memset(banks, 0, sizeof(*banks)); + INIT_LIST_HEAD(&banks->list); +} + +static int aca_banks_add_bank(struct aca_banks *banks, struct aca_bank *bank) +{ + struct aca_bank_node *node; + + if (!bank) + return -EINVAL; + + node = kvzalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + + memcpy(&node->bank, bank, sizeof(*bank)); + + INIT_LIST_HEAD(&node->node); + list_add_tail(&node->node, &banks->list); + + banks->nr_banks++; + + return 0; +} + +static void aca_banks_release(struct aca_banks *banks) +{ + struct aca_bank_node *node, *tmp; + + list_for_each_entry_safe(node, tmp, &banks->list, node) { + list_del(&node->node); + kvfree(node); + } +} + +static int aca_smu_get_valid_aca_count(struct amdgpu_device *adev, enum aca_error_type type, u32 *count) +{ + struct amdgpu_aca *aca = &adev->aca; + const struct aca_smu_funcs *smu_funcs = aca->smu_funcs; + + if (!count) + return -EINVAL; + + if (!smu_funcs || !smu_funcs->get_valid_aca_count) + return -EOPNOTSUPP; + + return smu_funcs->get_valid_aca_count(adev, type, count); +} + +static struct aca_regs_dump { + const char *name; + int reg_idx; +} aca_regs[] = { + {"CONTROL", ACA_REG_IDX_CTL}, + {"STATUS", ACA_REG_IDX_STATUS}, + {"ADDR", ACA_REG_IDX_ADDR}, + {"MISC", ACA_REG_IDX_MISC0}, + {"CONFIG", ACA_REG_IDX_CONFG}, + {"IPID", ACA_REG_IDX_IPID}, + {"SYND", ACA_REG_IDX_SYND}, + {"DESTAT", ACA_REG_IDX_DESTAT}, + {"DEADDR", ACA_REG_IDX_DEADDR}, + {"CONTROL_MASK", ACA_REG_IDX_CTL_MASK}, +}; + +static void aca_smu_bank_dump(struct amdgpu_device *adev, int idx, int total, struct aca_bank *bank) +{ + int i; + + dev_info(adev->dev, HW_ERR "Accelerator Check Architecture events logged\n"); + /* plus 1 for output format, e.g: ACA[08/08]: xxxx */ + for (i = 0; i < ARRAY_SIZE(aca_regs); i++) + dev_info(adev->dev, HW_ERR "ACA[%02d/%02d].%s=0x%016llx\n", + idx + 1, total, aca_regs[i].name, bank->regs[aca_regs[i].reg_idx]); +} + +static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_error_type type, + int start, int count, + struct aca_banks *banks) +{ + struct amdgpu_aca *aca = &adev->aca; + const struct aca_smu_funcs *smu_funcs = aca->smu_funcs; + struct aca_bank bank; + int i, max_count, ret; + + if (!count) + return 0; + + if (!smu_funcs || !smu_funcs->get_valid_aca_bank) + return -EOPNOTSUPP; + + switch (type) { + case ACA_ERROR_TYPE_UE: + max_count = smu_funcs->max_ue_bank_count; + break; + case ACA_ERROR_TYPE_CE: + max_count = smu_funcs->max_ce_bank_count; + break; + case ACA_ERROR_TYPE_DEFERRED: + default: + return -EINVAL; + } + + if (start + count >= max_count) + return -EINVAL; + + count = min_t(int, count, max_count); + for (i = 0; i < count; i++) { + memset(&bank, 0, sizeof(bank)); + ret = smu_funcs->get_valid_aca_bank(adev, type, start + i, &bank); + if (ret) + return ret; + + aca_smu_bank_dump(adev, i, count, &bank); + + ret = aca_banks_add_bank(banks, &bank); + if (ret) + return ret; + } + + return 0; +} + +static bool aca_bank_hwip_is_matched(struct aca_bank *bank, enum aca_hwip_type type) +{ + + struct aca_hwip *hwip; + int hwid, mcatype; + u64 ipid; + + if (!bank || type == ACA_HWIP_TYPE_UNKNOW) + return false; + + hwip = &aca_hwid_mcatypes[type]; + if (!hwip->hwid) + return false; + + ipid = bank->regs[ACA_REG_IDX_IPID]; + hwid = ACA_REG__IPID__HARDWAREID(ipid); + mcatype = ACA_REG__IPID__MCATYPE(ipid); + + return hwip->hwid == hwid && hwip->mcatype == mcatype; +} + +static bool aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, enum aca_error_type type) +{ + const struct aca_bank_ops *bank_ops = handle->bank_ops; + + if (!aca_bank_hwip_is_matched(bank, handle->hwip)) + return false; + + if (!bank_ops->aca_bank_is_valid) + return true; + + return bank_ops->aca_bank_is_valid(handle, bank, type, handle->data); +} + +static struct aca_bank_error *new_bank_error(struct aca_error *aerr, struct aca_bank_info *info) +{ + struct aca_bank_error *bank_error; + + bank_error = kvzalloc(sizeof(*bank_error), GFP_KERNEL); + if (!bank_error) + return NULL; + + INIT_LIST_HEAD(&bank_error->node); + memcpy(&bank_error->info, info, sizeof(*info)); + + mutex_lock(&aerr->lock); + list_add_tail(&bank_error->node, &aerr->list); + mutex_unlock(&aerr->lock); + + return bank_error; +} + +static struct aca_bank_error *find_bank_error(struct aca_error *aerr, struct aca_bank_info *info) +{ + struct aca_bank_error *bank_error = NULL; + struct aca_bank_info *tmp_info; + bool found = false; + + mutex_lock(&aerr->lock); + list_for_each_entry(bank_error, &aerr->list, node) { + tmp_info = &bank_error->info; + if (tmp_info->socket_id == info->socket_id && + tmp_info->die_id == info->die_id) { + found = true; + goto out_unlock; + } + } + +out_unlock: + mutex_unlock(&aerr->lock); + + return found ? bank_error : NULL; +} + +static void aca_bank_error_remove(struct aca_error *aerr, struct aca_bank_error *bank_error) +{ + if (!aerr || !bank_error) + return; + + list_del(&bank_error->node); + aerr->nr_errors--; + + kvfree(bank_error); +} + +static struct aca_bank_error *get_bank_error(struct aca_error *aerr, struct aca_bank_info *info) +{ + struct aca_bank_error *bank_error; + + if (!aerr || !info) + return NULL; + + bank_error = find_bank_error(aerr, info); + if (bank_error) + return bank_error; + + return new_bank_error(aerr, info); +} + +static int aca_log_errors(struct aca_handle *handle, enum aca_error_type type, + struct aca_bank_report *report) +{ + struct aca_error_cache *error_cache = &handle->error_cache; + struct aca_bank_error *bank_error; + struct aca_error *aerr; + + if (!handle || !report) + return -EINVAL; + + if (!report->count[type]) + return 0; + + aerr = &error_cache->errors[type]; + bank_error = get_bank_error(aerr, &report->info); + if (!bank_error) + return -ENOMEM; + + bank_error->count[type] += report->count[type]; + + return 0; +} + +static int aca_generate_bank_report(struct aca_handle *handle, struct aca_bank *bank, + enum aca_error_type type, struct aca_bank_report *report) +{ + const struct aca_bank_ops *bank_ops = handle->bank_ops; + + if (!bank || !report) + return -EINVAL; + + if (!bank_ops->aca_bank_generate_report) + return -EOPNOTSUPP; + + memset(report, 0, sizeof(*report)); + return bank_ops->aca_bank_generate_report(handle, bank, type, + report, handle->data); +} + +static int handler_aca_log_bank_error(struct aca_handle *handle, struct aca_bank *bank, + enum aca_error_type type, void *data) +{ + struct aca_bank_report report; + int ret; + + ret = aca_generate_bank_report(handle, bank, type, &report); + if (ret) + return ret; + + if (!report.count[type]) + return 0; + + ret = aca_log_errors(handle, type, &report); + if (ret) + return ret; + + return 0; +} + +static int aca_dispatch_bank(struct aca_handle_manager *mgr, struct aca_bank *bank, + enum aca_error_type type, bank_handler_t handler, void *data) +{ + struct aca_handle *handle; + int ret; + + if (list_empty(&mgr->list)) + return 0; + + list_for_each_entry(handle, &mgr->list, node) { + if (!aca_bank_is_valid(handle, bank, type)) + continue; + + ret = handler(handle, bank, type, data); + if (ret) + return ret; + } + + return 0; +} + +static int aca_dispatch_banks(struct aca_handle_manager *mgr, struct aca_banks *banks, + enum aca_error_type type, bank_handler_t handler, void *data) +{ + struct aca_bank_node *node; + struct aca_bank *bank; + int ret; + + if (!mgr || !banks) + return -EINVAL; + + /* pre check to avoid unnecessary operations */ + if (list_empty(&mgr->list) || list_empty(&banks->list)) + return 0; + + list_for_each_entry(node, &banks->list, node) { + bank = &node->bank; + + ret = aca_dispatch_bank(mgr, bank, type, handler, data); + if (ret) + return ret; + } + + return 0; +} + +static int aca_banks_update(struct amdgpu_device *adev, enum aca_error_type type, + bank_handler_t handler, void *data) +{ + struct amdgpu_aca *aca = &adev->aca; + struct aca_banks banks; + u32 count = 0; + int ret; + + if (list_empty(&aca->mgr.list)) + return 0; + + /* NOTE: pmfw is only support UE and CE */ + if (type == ACA_ERROR_TYPE_DEFERRED) + type = ACA_ERROR_TYPE_CE; + + ret = aca_smu_get_valid_aca_count(adev, type, &count); + if (ret) + return ret; + + if (!count) + return 0; + + aca_banks_init(&banks); + + ret = aca_smu_get_valid_aca_banks(adev, type, 0, count, &banks); + if (ret) + goto err_release_banks; + + if (list_empty(&banks.list)) { + ret = 0; + goto err_release_banks; + } + + ret = aca_dispatch_banks(&aca->mgr, &banks, type, + handler, data); + if (ret) + goto err_release_banks; + +err_release_banks: + aca_banks_release(&banks); + + return ret; +} + +static int aca_log_aca_error_data(struct aca_bank_error *bank_error, enum aca_error_type type, struct ras_err_data *err_data) +{ + struct aca_bank_info *info; + struct amdgpu_smuio_mcm_config_info mcm_info; + u64 count; + + if (type >= ACA_ERROR_TYPE_COUNT) + return -EINVAL; + + count = bank_error->count[type]; + if (!count) + return 0; + + info = &bank_error->info; + mcm_info.die_id = info->die_id; + mcm_info.socket_id = info->socket_id; + + switch (type) { + case ACA_ERROR_TYPE_UE: + amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, count); + break; + case ACA_ERROR_TYPE_CE: + amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, NULL, count); + break; + case ACA_ERROR_TYPE_DEFERRED: + default: + break; + } + + return 0; +} + +static int aca_log_aca_error(struct aca_handle *handle, enum aca_error_type type, struct ras_err_data *err_data) +{ + struct aca_error_cache *error_cache = &handle->error_cache; + struct aca_error *aerr = &error_cache->errors[type]; + struct aca_bank_error *bank_error, *tmp; + + mutex_lock(&aerr->lock); + + if (list_empty(&aerr->list)) + goto out_unlock; + + list_for_each_entry_safe(bank_error, tmp, &aerr->list, node) { + aca_log_aca_error_data(bank_error, type, err_data); + aca_bank_error_remove(aerr, bank_error); + } + +out_unlock: + mutex_unlock(&aerr->lock); + + return 0; +} + +static int __aca_get_error_data(struct amdgpu_device *adev, struct aca_handle *handle, enum aca_error_type type, + struct ras_err_data *err_data) +{ + int ret; + + /* udpate aca bank to aca source error_cache first */ + ret = aca_banks_update(adev, type, handler_aca_log_bank_error, NULL); + if (ret) + return ret; + + return aca_log_aca_error(handle, type, err_data); +} + +static bool aca_handle_is_valid(struct aca_handle *handle) +{ + if (!handle->mask || !list_empty(&handle->node)) + return false; + + return true; +} + +int amdgpu_aca_get_error_data(struct amdgpu_device *adev, struct aca_handle *handle, + enum aca_error_type type, void *data) +{ + struct ras_err_data *err_data = (struct ras_err_data *)data; + + if (!handle || !err_data) + return -EINVAL; + + if (aca_handle_is_valid(handle)) + return -EOPNOTSUPP; + + if (!(BIT(type) & handle->mask)) + return 0; + + return __aca_get_error_data(adev, handle, type, err_data); +} + +static void aca_error_init(struct aca_error *aerr, enum aca_error_type type) +{ + mutex_init(&aerr->lock); + INIT_LIST_HEAD(&aerr->list); + aerr->type = type; + aerr->nr_errors = 0; +} + +static void aca_init_error_cache(struct aca_handle *handle) +{ + struct aca_error_cache *error_cache = &handle->error_cache; + int type; + + for (type = ACA_ERROR_TYPE_UE; type < ACA_ERROR_TYPE_COUNT; type++) + aca_error_init(&error_cache->errors[type], type); +} + +static void aca_error_fini(struct aca_error *aerr) +{ + struct aca_bank_error *bank_error, *tmp; + + mutex_lock(&aerr->lock); + list_for_each_entry_safe(bank_error, tmp, &aerr->list, node) + aca_bank_error_remove(aerr, bank_error); + + mutex_destroy(&aerr->lock); +} + +static void aca_fini_error_cache(struct aca_handle *handle) +{ + struct aca_error_cache *error_cache = &handle->error_cache; + int type; + + for (type = ACA_ERROR_TYPE_UE; type < ACA_ERROR_TYPE_COUNT; type++) + aca_error_fini(&error_cache->errors[type]); +} + +static int add_aca_handle(struct amdgpu_device *adev, struct aca_handle_manager *mgr, struct aca_handle *handle, + const char *name, const struct aca_info *ras_info, void *data) +{ + memset(handle, 0, sizeof(*handle)); + + handle->adev = adev; + handle->mgr = mgr; + handle->name = name; + handle->hwip = ras_info->hwip; + handle->mask = ras_info->mask; + handle->bank_ops = ras_info->bank_ops; + handle->data = data; + aca_init_error_cache(handle); + + INIT_LIST_HEAD(&handle->node); + list_add_tail(&handle->node, &mgr->list); + mgr->nr_handles++; + + return 0; +} + +static ssize_t aca_sysfs_read(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct aca_handle *handle = container_of(attr, struct aca_handle, aca_attr); + + /* NOTE: the aca cache will be auto cleared once read, + * So the driver should unify the query entry point, forward request to ras query interface directly */ + return amdgpu_ras_aca_sysfs_read(dev, attr, handle, buf, handle->data); +} + +static int add_aca_sysfs(struct amdgpu_device *adev, struct aca_handle *handle) +{ + struct device_attribute *aca_attr = &handle->aca_attr; + + snprintf(handle->attr_name, sizeof(handle->attr_name) - 1, "aca_%s", handle->name); + aca_attr->show = aca_sysfs_read; + aca_attr->attr.name = handle->attr_name; + aca_attr->attr.mode = S_IRUGO; + sysfs_attr_init(&aca_attr->attr); + + return sysfs_add_file_to_group(&adev->dev->kobj, + &aca_attr->attr, + "ras"); +} + +int amdgpu_aca_add_handle(struct amdgpu_device *adev, struct aca_handle *handle, + const char *name, const struct aca_info *ras_info, void *data) +{ + struct amdgpu_aca *aca = &adev->aca; + int ret; + + if (!amdgpu_aca_is_enabled(adev)) + return 0; + + ret = add_aca_handle(adev, &aca->mgr, handle, name, ras_info, data); + if (ret) + return ret; + + return add_aca_sysfs(adev, handle); +} + +static void remove_aca_handle(struct aca_handle *handle) +{ + struct aca_handle_manager *mgr = handle->mgr; + + aca_fini_error_cache(handle); + list_del(&handle->node); + mgr->nr_handles--; +} + +static void remove_aca_sysfs(struct aca_handle *handle) +{ + struct amdgpu_device *adev = handle->adev; + struct device_attribute *aca_attr = &handle->aca_attr; + + if (adev->dev->kobj.sd) + sysfs_remove_file_from_group(&adev->dev->kobj, + &aca_attr->attr, + "ras"); +} + +void amdgpu_aca_remove_handle(struct aca_handle *handle) +{ + if (!handle || list_empty(&handle->node)) + return; + + remove_aca_sysfs(handle); + remove_aca_handle(handle); +} + +static int aca_manager_init(struct aca_handle_manager *mgr) +{ + INIT_LIST_HEAD(&mgr->list); + mgr->nr_handles = 0; + + return 0; +} + +static void aca_manager_fini(struct aca_handle_manager *mgr) +{ + struct aca_handle *handle, *tmp; + + list_for_each_entry_safe(handle, tmp, &mgr->list, node) + amdgpu_aca_remove_handle(handle); +} + +bool amdgpu_aca_is_enabled(struct amdgpu_device *adev) +{ + return adev->aca.is_enabled; +} + +int amdgpu_aca_init(struct amdgpu_device *adev) +{ + struct amdgpu_aca *aca = &adev->aca; + int ret; + + ret = aca_manager_init(&aca->mgr); + if (ret) + return ret; + + return 0; +} + +void amdgpu_aca_fini(struct amdgpu_device *adev) +{ + struct amdgpu_aca *aca = &adev->aca; + + aca_manager_fini(&aca->mgr); +} + +int amdgpu_aca_reset(struct amdgpu_device *adev) +{ + amdgpu_aca_fini(adev); + + return amdgpu_aca_init(adev); +} + +void amdgpu_aca_set_smu_funcs(struct amdgpu_device *adev, const struct aca_smu_funcs *smu_funcs) +{ + struct amdgpu_aca *aca = &adev->aca; + + WARN_ON(aca->smu_funcs); + aca->smu_funcs = smu_funcs; +} + +int aca_bank_info_decode(struct aca_bank *bank, struct aca_bank_info *info) +{ + u64 ipid; + u32 instidhi, instidlo; + + if (!bank || !info) + return -EINVAL; + + ipid = bank->regs[ACA_REG_IDX_IPID]; + info->hwid = ACA_REG__IPID__HARDWAREID(ipid); + info->mcatype = ACA_REG__IPID__MCATYPE(ipid); + /* + * Unfied DieID Format: SAASS. A:AID, S:Socket. + * Unfied DieID[4:4] = InstanceId[0:0] + * Unfied DieID[0:3] = InstanceIdHi[0:3] + */ + instidhi = ACA_REG__IPID__INSTANCEIDHI(ipid); + instidlo = ACA_REG__IPID__INSTANCEIDLO(ipid); + info->die_id = ((instidhi >> 2) & 0x03); + info->socket_id = ((instidlo & 0x1) << 2) | (instidhi & 0x03); + + return 0; +} + +static int aca_bank_get_error_code(struct amdgpu_device *adev, struct aca_bank *bank) +{ + int error_code; + + switch (amdgpu_ip_version(adev, MP1_HWIP, 0)) { + case IP_VERSION(13, 0, 6): + if (!(adev->flags & AMD_IS_APU) && adev->pm.fw_version >= 0x00555600) { + error_code = ACA_REG__SYND__ERRORINFORMATION(bank->regs[ACA_REG_IDX_SYND]); + return error_code & 0xff; + } + break; + default: + break; + } + + /* NOTE: the true error code is encoded in status.errorcode[0:7] */ + error_code = ACA_REG__STATUS__ERRORCODE(bank->regs[ACA_REG_IDX_STATUS]); + + return error_code & 0xff; +} + +int aca_bank_check_error_codes(struct amdgpu_device *adev, struct aca_bank *bank, int *err_codes, int size) +{ + int i, error_code; + + if (!bank || !err_codes) + return -EINVAL; + + error_code = aca_bank_get_error_code(adev, bank); + for (i = 0; i < size; i++) { + if (err_codes[i] == error_code) + return 0; + } + + return -EINVAL; +} + +int amdgpu_aca_smu_set_debug_mode(struct amdgpu_device *adev, bool en) +{ + struct amdgpu_aca *aca = &adev->aca; + const struct aca_smu_funcs *smu_funcs = aca->smu_funcs; + + if (!smu_funcs || !smu_funcs->set_debug_mode) + return -EOPNOTSUPP; + + return smu_funcs->set_debug_mode(adev, en); +} + +#if defined(CONFIG_DEBUG_FS) +static int amdgpu_aca_smu_debug_mode_set(void *data, u64 val) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)data; + int ret; + + ret = amdgpu_ras_set_aca_debug_mode(adev, val ? true : false); + if (ret) + return ret; + + dev_info(adev->dev, "amdgpu set smu aca debug mode %s success\n", val ? "on" : "off"); + + return 0; +} + +static void aca_dump_entry(struct seq_file *m, struct aca_bank *bank, enum aca_error_type type, int idx) +{ + struct aca_bank_info info; + int i, ret; + + ret = aca_bank_info_decode(bank, &info); + if (ret) + return; + + seq_printf(m, "aca entry[%d].type: %s\n", idx, type == ACA_ERROR_TYPE_UE ? "UE" : "CE"); + seq_printf(m, "aca entry[%d].info: socketid:%d aid:%d hwid:0x%03x mcatype:0x%04x\n", + idx, info.socket_id, info.die_id, info.hwid, info.mcatype); + + for (i = 0; i < ARRAY_SIZE(aca_regs); i++) + seq_printf(m, "aca entry[%d].regs[%d]: 0x%016llx\n", idx, aca_regs[i].reg_idx, bank->regs[aca_regs[i].reg_idx]); +} + +struct aca_dump_context { + struct seq_file *m; + int idx; +}; + +static int handler_aca_bank_dump(struct aca_handle *handle, struct aca_bank *bank, + enum aca_error_type type, void *data) +{ + struct aca_dump_context *ctx = (struct aca_dump_context *)data; + + aca_dump_entry(ctx->m, bank, type, ctx->idx++); + + return handler_aca_log_bank_error(handle, bank, type, NULL); +} + +static int aca_dump_show(struct seq_file *m, enum aca_error_type type) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)m->private; + struct aca_dump_context context = { + .m = m, + .idx = 0, + }; + + return aca_banks_update(adev, type, handler_aca_bank_dump, (void *)&context); +} + +static int aca_dump_ce_show(struct seq_file *m, void *unused) +{ + return aca_dump_show(m, ACA_ERROR_TYPE_CE); +} + +static int aca_dump_ce_open(struct inode *inode, struct file *file) +{ + return single_open(file, aca_dump_ce_show, inode->i_private); +} + +static const struct file_operations aca_ce_dump_debug_fops = { + .owner = THIS_MODULE, + .open = aca_dump_ce_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int aca_dump_ue_show(struct seq_file *m, void *unused) +{ + return aca_dump_show(m, ACA_ERROR_TYPE_UE); +} + +static int aca_dump_ue_open(struct inode *inode, struct file *file) +{ + return single_open(file, aca_dump_ue_show, inode->i_private); +} + +static const struct file_operations aca_ue_dump_debug_fops = { + .owner = THIS_MODULE, + .open = aca_dump_ue_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +DEFINE_DEBUGFS_ATTRIBUTE(aca_debug_mode_fops, NULL, amdgpu_aca_smu_debug_mode_set, "%llu\n"); +#endif + +void amdgpu_aca_smu_debugfs_init(struct amdgpu_device *adev, struct dentry *root) +{ +#if defined(CONFIG_DEBUG_FS) + if (!root || adev->ip_versions[MP1_HWIP][0] != IP_VERSION(13, 0, 6)) + return; + + debugfs_create_file("aca_debug_mode", 0200, root, adev, &aca_debug_mode_fops); + debugfs_create_file("aca_ue_dump", 0400, root, adev, &aca_ue_dump_debug_fops); + debugfs_create_file("aca_ce_dump", 0400, root, adev, &aca_ce_dump_debug_fops); +#endif +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h new file mode 100644 index 000000000000..2da50e095883 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.h @@ -0,0 +1,202 @@ +/* + * 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 __AMDGPU_ACA_H__ +#define __AMDGPU_ACA_H__ + +#include <linux/list.h> + +#define ACA_MAX_REGS_COUNT (16) + +#define ACA_REG_FIELD(x, h, l) (((x) & GENMASK_ULL(h, l)) >> l) +#define ACA_REG__STATUS__VAL(x) ACA_REG_FIELD(x, 63, 63) +#define ACA_REG__STATUS__OVERFLOW(x) ACA_REG_FIELD(x, 62, 62) +#define ACA_REG__STATUS__UC(x) ACA_REG_FIELD(x, 61, 61) +#define ACA_REG__STATUS__EN(x) ACA_REG_FIELD(x, 60, 60) +#define ACA_REG__STATUS__MISCV(x) ACA_REG_FIELD(x, 59, 59) +#define ACA_REG__STATUS__ADDRV(x) ACA_REG_FIELD(x, 58, 58) +#define ACA_REG__STATUS__PCC(x) ACA_REG_FIELD(x, 57, 57) +#define ACA_REG__STATUS__ERRCOREIDVAL(x) ACA_REG_FIELD(x, 56, 56) +#define ACA_REG__STATUS__TCC(x) ACA_REG_FIELD(x, 55, 55) +#define ACA_REG__STATUS__SYNDV(x) ACA_REG_FIELD(x, 53, 53) +#define ACA_REG__STATUS__CECC(x) ACA_REG_FIELD(x, 46, 46) +#define ACA_REG__STATUS__UECC(x) ACA_REG_FIELD(x, 45, 45) +#define ACA_REG__STATUS__DEFERRED(x) ACA_REG_FIELD(x, 44, 44) +#define ACA_REG__STATUS__POISON(x) ACA_REG_FIELD(x, 43, 43) +#define ACA_REG__STATUS__SCRUB(x) ACA_REG_FIELD(x, 40, 40) +#define ACA_REG__STATUS__ERRCOREID(x) ACA_REG_FIELD(x, 37, 32) +#define ACA_REG__STATUS__ADDRLSB(x) ACA_REG_FIELD(x, 29, 24) +#define ACA_REG__STATUS__ERRORCODEEXT(x) ACA_REG_FIELD(x, 21, 16) +#define ACA_REG__STATUS__ERRORCODE(x) ACA_REG_FIELD(x, 15, 0) + +#define ACA_REG__IPID__MCATYPE(x) ACA_REG_FIELD(x, 63, 48) +#define ACA_REG__IPID__INSTANCEIDHI(x) ACA_REG_FIELD(x, 47, 44) +#define ACA_REG__IPID__HARDWAREID(x) ACA_REG_FIELD(x, 43, 32) +#define ACA_REG__IPID__INSTANCEIDLO(x) ACA_REG_FIELD(x, 31, 0) + +#define ACA_REG__MISC0__VALID(x) ACA_REG_FIELD(x, 63, 63) +#define ACA_REG__MISC0__OVRFLW(x) ACA_REG_FIELD(x, 48, 48) +#define ACA_REG__MISC0__ERRCNT(x) ACA_REG_FIELD(x, 43, 32) + +#define ACA_REG__SYND__ERRORINFORMATION(x) ACA_REG_FIELD(x, 17, 0) + +/* NOTE: The following codes refers to the smu header file */ +#define ACA_EXTERROR_CODE_CE 0x3a +#define ACA_EXTERROR_CODE_FAULT 0x3b + +#define ACA_ERROR_UE_MASK BIT_MASK(ACA_ERROR_TYPE_UE) +#define ACA_ERROR_CE_MASK BIT_MASK(ACA_ERROR_TYPE_CE) +#define ACA_ERROR_DEFERRED_MASK BIT_MASK(ACA_ERROR_TYPE_DEFERRED) + +enum aca_reg_idx { + ACA_REG_IDX_CTL = 0, + ACA_REG_IDX_STATUS = 1, + ACA_REG_IDX_ADDR = 2, + ACA_REG_IDX_MISC0 = 3, + ACA_REG_IDX_CONFG = 4, + ACA_REG_IDX_IPID = 5, + ACA_REG_IDX_SYND = 6, + ACA_REG_IDX_DESTAT = 8, + ACA_REG_IDX_DEADDR = 9, + ACA_REG_IDX_CTL_MASK = 10, + ACA_REG_IDX_COUNT = 16, +}; + +enum aca_hwip_type { + ACA_HWIP_TYPE_UNKNOW = -1, + ACA_HWIP_TYPE_PSP = 0, + ACA_HWIP_TYPE_UMC, + ACA_HWIP_TYPE_SMU, + ACA_HWIP_TYPE_PCS_XGMI, + ACA_HWIP_TYPE_COUNT, +}; + +enum aca_error_type { + ACA_ERROR_TYPE_INVALID = -1, + ACA_ERROR_TYPE_UE = 0, + ACA_ERROR_TYPE_CE, + ACA_ERROR_TYPE_DEFERRED, + ACA_ERROR_TYPE_COUNT +}; + +struct aca_bank { + u64 regs[ACA_MAX_REGS_COUNT]; +}; + +struct aca_bank_node { + struct aca_bank bank; + struct list_head node; +}; + +struct aca_bank_info { + int die_id; + int socket_id; + int hwid; + int mcatype; +}; + +struct aca_bank_report { + struct aca_bank_info info; + u64 count[ACA_ERROR_TYPE_COUNT]; +}; + +struct aca_bank_error { + struct list_head node; + struct aca_bank_info info; + u64 count[ACA_ERROR_TYPE_COUNT]; +}; + +struct aca_error { + struct list_head list; + struct mutex lock; + enum aca_error_type type; + int nr_errors; +}; + +struct aca_handle_manager { + struct list_head list; + int nr_handles; +}; + +struct aca_error_cache { + struct aca_error errors[ACA_ERROR_TYPE_COUNT]; +}; + +struct aca_handle { + struct list_head node; + enum aca_hwip_type hwip; + struct amdgpu_device *adev; + struct aca_handle_manager *mgr; + struct aca_error_cache error_cache; + const struct aca_bank_ops *bank_ops; + struct device_attribute aca_attr; + char attr_name[64]; + const char *name; + u32 mask; + void *data; +}; + +struct aca_bank_ops { + int (*aca_bank_generate_report)(struct aca_handle *handle, struct aca_bank *bank, enum aca_error_type type, + struct aca_bank_report *report, void *data); + bool (*aca_bank_is_valid)(struct aca_handle *handle, struct aca_bank *bank, enum aca_error_type type, + void *data); +}; + +struct aca_smu_funcs { + int max_ue_bank_count; + int max_ce_bank_count; + int (*set_debug_mode)(struct amdgpu_device *adev, bool enable); + int (*get_valid_aca_count)(struct amdgpu_device *adev, enum aca_error_type type, u32 *count); + int (*get_valid_aca_bank)(struct amdgpu_device *adev, enum aca_error_type type, int idx, struct aca_bank *bank); +}; + +struct amdgpu_aca { + struct aca_handle_manager mgr; + const struct aca_smu_funcs *smu_funcs; + bool is_enabled; +}; + +struct aca_info { + enum aca_hwip_type hwip; + const struct aca_bank_ops *bank_ops; + u32 mask; +}; + +int amdgpu_aca_init(struct amdgpu_device *adev); +void amdgpu_aca_fini(struct amdgpu_device *adev); +int amdgpu_aca_reset(struct amdgpu_device *adev); +void amdgpu_aca_set_smu_funcs(struct amdgpu_device *adev, const struct aca_smu_funcs *smu_funcs); +bool amdgpu_aca_is_enabled(struct amdgpu_device *adev); + +int aca_bank_info_decode(struct aca_bank *bank, struct aca_bank_info *info); +int aca_bank_check_error_codes(struct amdgpu_device *adev, struct aca_bank *bank, int *err_codes, int size); + +int amdgpu_aca_add_handle(struct amdgpu_device *adev, struct aca_handle *handle, + const char *name, const struct aca_info *aca_info, void *data); +void amdgpu_aca_remove_handle(struct aca_handle *handle); +int amdgpu_aca_get_error_data(struct amdgpu_device *adev, struct aca_handle *handle, + enum aca_error_type type, void *data); +int amdgpu_aca_smu_set_debug_mode(struct amdgpu_device *adev, bool en); +void amdgpu_aca_smu_debugfs_init(struct amdgpu_device *adev, struct dentry *root); +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 41db030ddc4e..190039f14c30 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -742,9 +742,10 @@ void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev) amdgpu_device_flush_hdp(adev, NULL); } -void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, bool reset) +void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, + enum amdgpu_ras_block block, bool reset) { - amdgpu_umc_poison_handler(adev, reset); + amdgpu_umc_poison_handler(adev, block, reset); } int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index 27c61c535e29..e60f63ccf79a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -193,6 +193,9 @@ struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f); int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo); int amdgpu_amdkfd_evict_userptr(struct mmu_interval_notifier *mni, unsigned long cur_seq, struct kgd_mem *mem); +int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo, + uint32_t domain, + struct dma_fence *fence); #else static inline bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm) @@ -218,6 +221,13 @@ int amdgpu_amdkfd_evict_userptr(struct mmu_interval_notifier *mni, { return 0; } +static inline +int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo, + uint32_t domain, + struct dma_fence *fence) +{ + return 0; +} #endif /* Shared API */ int amdgpu_amdkfd_alloc_gtt_mem(struct amdgpu_device *adev, size_t size, @@ -326,7 +336,7 @@ void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev); int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev, struct tile_config *config); void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev, - bool reset); + enum amdgpu_ras_block block, bool reset); bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem); void amdgpu_amdkfd_block_mmu_notifications(void *p); int amdgpu_amdkfd_criu_resume(void *p); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 231fd927dcfb..5cd84f72bf26 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -426,9 +426,9 @@ validate_fail: return ret; } -static int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo, - uint32_t domain, - struct dma_fence *fence) +int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo, + uint32_t domain, + struct dma_fence *fence) { int ret = amdgpu_bo_reserve(bo, false); @@ -464,13 +464,15 @@ static int amdgpu_amdkfd_validate_vm_bo(void *_unused, struct amdgpu_bo *bo) * again. Page directories are only updated after updating page * tables. */ -static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm) +static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm, + struct ww_acquire_ctx *ticket) { struct amdgpu_bo *pd = vm->root.bo; struct amdgpu_device *adev = amdgpu_ttm_adev(pd->tbo.bdev); int ret; - ret = amdgpu_vm_validate_pt_bos(adev, vm, amdgpu_amdkfd_validate_vm_bo, NULL); + ret = amdgpu_vm_validate(adev, vm, ticket, + amdgpu_amdkfd_validate_vm_bo, NULL); if (ret) { pr_err("failed to validate PT BOs\n"); return ret; @@ -1310,14 +1312,15 @@ update_gpuvm_pte_failed: return ret; } -static int process_validate_vms(struct amdkfd_process_info *process_info) +static int process_validate_vms(struct amdkfd_process_info *process_info, + struct ww_acquire_ctx *ticket) { struct amdgpu_vm *peer_vm; int ret; list_for_each_entry(peer_vm, &process_info->vm_list_head, vm_list_node) { - ret = vm_validate_pt_pd_bos(peer_vm); + ret = vm_validate_pt_pd_bos(peer_vm, ticket); if (ret) return ret; } @@ -1402,7 +1405,7 @@ static int init_kfd_vm(struct amdgpu_vm *vm, void **process_info, ret = amdgpu_bo_reserve(vm->root.bo, true); if (ret) goto reserve_pd_fail; - ret = vm_validate_pt_pd_bos(vm); + ret = vm_validate_pt_pd_bos(vm, NULL); if (ret) { pr_err("validate_pt_pd_bos() failed\n"); goto validate_pd_fail; @@ -2043,7 +2046,7 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu( bo->tbo.resource->mem_type == TTM_PL_SYSTEM) is_invalid_userptr = true; - ret = vm_validate_pt_pd_bos(avm); + ret = vm_validate_pt_pd_bos(avm, NULL); if (unlikely(ret)) goto out_unreserve; @@ -2136,7 +2139,7 @@ int amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu( goto unreserve_out; } - ret = vm_validate_pt_pd_bos(avm); + ret = vm_validate_pt_pd_bos(avm, NULL); if (unlikely(ret)) goto unreserve_out; @@ -2634,7 +2637,7 @@ static int validate_invalid_user_pages(struct amdkfd_process_info *process_info) } } - ret = process_validate_vms(process_info); + ret = process_validate_vms(process_info, NULL); if (ret) goto unreserve_out; @@ -2894,11 +2897,6 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence __rcu * amdgpu_sync_create(&sync_obj); - /* Validate PDs and PTs */ - ret = process_validate_vms(process_info); - if (ret) - goto validate_map_fail; - /* Validate BOs and map them to GPUVM (update VM page tables). */ list_for_each_entry(mem, &process_info->kfd_bo_list, validate_list) { @@ -2949,6 +2947,13 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence __rcu * if (failed_size) pr_debug("0x%lx/0x%lx in system\n", failed_size, total_size); + /* Validate PDs, PTs and evicted DMABuf imports last. Otherwise BO + * validations above would invalidate DMABuf imports again. + */ + ret = process_validate_vms(process_info, &exec.ticket); + if (ret) + goto validate_map_fail; + /* Update mappings not managed by KFD */ list_for_each_entry(peer_vm, &process_info->vm_list_head, vm_list_node) { @@ -3020,7 +3025,7 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence __rcu * &process_info->eviction_fence->base, DMA_RESV_USAGE_BOOKKEEP); } - /* Attach eviction fence to PD / PT BOs */ + /* Attach eviction fence to PD / PT BOs and DMABuf imports */ list_for_each_entry(peer_vm, &process_info->vm_list_head, vm_list_node) { struct amdgpu_bo *bo = peer_vm->root.bo; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c index dce9e7d5e4ec..52b12c1718eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c @@ -1018,7 +1018,8 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, if (clock_type == COMPUTE_ENGINE_PLL_PARAM) { args.v3.ulClockParams = cpu_to_le32((clock_type << 24) | clock); - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); dividers->post_div = args.v3.ucPostDiv; dividers->enable_post_div = (args.v3.ucCntlFlag & @@ -1038,7 +1039,8 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, if (strobe_mode) args.v5.ucInputFlag = ATOM_PLL_INPUT_FLAG_PLL_STROBE_MODE_EN; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); dividers->post_div = args.v5.ucPostDiv; dividers->enable_post_div = (args.v5.ucCntlFlag & @@ -1056,7 +1058,8 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, /* fusion */ args.v4.ulClock = cpu_to_le32(clock); /* 10 khz */ - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); dividers->post_divider = dividers->post_div = args.v4.ucPostDiv; dividers->real_clock = le32_to_cpu(args.v4.ulClock); @@ -1067,7 +1070,8 @@ int amdgpu_atombios_get_clock_dividers(struct amdgpu_device *adev, args.v6_in.ulClock.ulComputeClockFlag = clock_type; args.v6_in.ulClock.ulClockFreq = cpu_to_le32(clock); /* 10 khz */ - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); dividers->whole_fb_div = le16_to_cpu(args.v6_out.ulFbDiv.usFbDiv); dividers->frac_fb_div = le16_to_cpu(args.v6_out.ulFbDiv.usFbDivFrac); @@ -1109,7 +1113,8 @@ int amdgpu_atombios_get_memory_pll_dividers(struct amdgpu_device *adev, if (strobe_mode) args.ucInputFlag |= MPLL_INPUT_FLAG_STROBE_MODE_EN; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); mpll_param->clkfrac = le16_to_cpu(args.ulFbDiv.usFbDivFrac); mpll_param->clkf = le16_to_cpu(args.ulFbDiv.usFbDiv); @@ -1151,7 +1156,8 @@ void amdgpu_atombios_set_engine_dram_timings(struct amdgpu_device *adev, if (mem_clock) args.sReserved.ulClock = cpu_to_le32(mem_clock & SET_CLOCK_FREQ_MASK); - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); } void amdgpu_atombios_get_default_voltages(struct amdgpu_device *adev, @@ -1205,7 +1211,8 @@ int amdgpu_atombios_get_max_vddc(struct amdgpu_device *adev, u8 voltage_type, args.v2.ucVoltageMode = 0; args.v2.usVoltageLevel = 0; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); *voltage = le16_to_cpu(args.v2.usVoltageLevel); break; @@ -1214,7 +1221,8 @@ int amdgpu_atombios_get_max_vddc(struct amdgpu_device *adev, u8 voltage_type, args.v3.ucVoltageMode = ATOM_GET_VOLTAGE_LEVEL; args.v3.usVoltageLevel = cpu_to_le16(voltage_id); - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, + sizeof(args)); *voltage = le16_to_cpu(args.v3.usVoltageLevel); break; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index fb2681dd6b33..6857c586ded7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -941,5 +941,6 @@ int amdgpu_atomfirmware_asic_init(struct amdgpu_device *adev, bool fb_reset) return -EINVAL; } - return amdgpu_atom_execute_table(ctx, ATOM_CMD_INIT, (uint32_t *)&asic_init_ps_v2_1); + return amdgpu_atom_execute_table(ctx, ATOM_CMD_INIT, (uint32_t *)&asic_init_ps_v2_1, + sizeof(asic_init_ps_v2_1)); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h index c7eb2caec65a..649b5530d8ae 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.h @@ -36,7 +36,7 @@ int amdgpu_atomfirmware_get_clock_info(struct amdgpu_device *adev); int amdgpu_atomfirmware_get_gfx_info(struct amdgpu_device *adev); bool amdgpu_atomfirmware_mem_ecc_supported(struct amdgpu_device *adev); bool amdgpu_atomfirmware_sram_ecc_supported(struct amdgpu_device *adev); -bool amdgpu_atomfirmware_ras_rom_addr(struct amdgpu_device *adev, uint8_t* i2c_address); +bool amdgpu_atomfirmware_ras_rom_addr(struct amdgpu_device *adev, uint8_t *i2c_address); bool amdgpu_atomfirmware_mem_training_supported(struct amdgpu_device *adev); bool amdgpu_atomfirmware_dynamic_boot_config_supported(struct amdgpu_device *adev); int amdgpu_atomfirmware_get_fw_reserved_fb_size(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 6adeddfb3d56..0a4b09709cfb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -952,10 +952,10 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, p->bytes_moved = 0; p->bytes_moved_vis = 0; - r = amdgpu_vm_validate_pt_bos(p->adev, &fpriv->vm, - amdgpu_cs_bo_validate, p); + r = amdgpu_vm_validate(p->adev, &fpriv->vm, NULL, + amdgpu_cs_bo_validate, p); if (r) { - DRM_ERROR("amdgpu_vm_validate_pt_bos() failed.\n"); + DRM_ERROR("amdgpu_vm_validate() failed.\n"); goto out_free_user_pages; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c index 796fa6f1420b..b5ad56690a9d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c @@ -30,7 +30,7 @@ uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev) { uint64_t addr = adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT; - addr -= AMDGPU_VA_RESERVED_SIZE; + addr -= AMDGPU_VA_RESERVED_CSA_SIZE; addr = amdgpu_gmc_sign_extend(addr); return addr; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index fdde7488d0ed..d534e192e260 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -96,6 +96,9 @@ MODULE_FIRMWARE("amdgpu/navi12_gpu_info.bin"); #define AMDGPU_RESUME_MS 2000 #define AMDGPU_MAX_RETRY_LIMIT 2 #define AMDGPU_RETRY_SRIOV_RESET(r) ((r) == -EBUSY || (r) == -ETIMEDOUT || (r) == -EINVAL) +#define AMDGPU_PCIE_INDEX_FALLBACK (0x38 >> 2) +#define AMDGPU_PCIE_INDEX_HI_FALLBACK (0x44 >> 2) +#define AMDGPU_PCIE_DATA_FALLBACK (0x3C >> 2) static const struct drm_driver amdgpu_kms_driver; @@ -781,12 +784,22 @@ u32 amdgpu_device_indirect_rreg_ext(struct amdgpu_device *adev, void __iomem *pcie_index_hi_offset; void __iomem *pcie_data_offset; - pcie_index = adev->nbio.funcs->get_pcie_index_offset(adev); - pcie_data = adev->nbio.funcs->get_pcie_data_offset(adev); - if ((reg_addr >> 32) && (adev->nbio.funcs->get_pcie_index_hi_offset)) - pcie_index_hi = adev->nbio.funcs->get_pcie_index_hi_offset(adev); - else + if (unlikely(!adev->nbio.funcs)) { + pcie_index = AMDGPU_PCIE_INDEX_FALLBACK; + pcie_data = AMDGPU_PCIE_DATA_FALLBACK; + } else { + pcie_index = adev->nbio.funcs->get_pcie_index_offset(adev); + pcie_data = adev->nbio.funcs->get_pcie_data_offset(adev); + } + + if (reg_addr >> 32) { + if (unlikely(!adev->nbio.funcs)) + pcie_index_hi = AMDGPU_PCIE_INDEX_HI_FALLBACK; + else + pcie_index_hi = adev->nbio.funcs->get_pcie_index_hi_offset(adev); + } else { pcie_index_hi = 0; + } spin_lock_irqsave(&adev->pcie_idx_lock, flags); pcie_index_offset = (void __iomem *)adev->rmmio + pcie_index * 4; @@ -1218,8 +1231,6 @@ static int amdgpu_device_asic_init(struct amdgpu_device *adev) amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(11, 0, 0)) { amdgpu_psp_wait_for_bootloader(adev); ret = amdgpu_atomfirmware_asic_init(adev, true); - /* TODO: check the return val and stop device initialization if boot fails */ - amdgpu_psp_query_boot_status(adev); return ret; } else { return amdgpu_atom_asic_init(adev->mode_info.atom_context); @@ -1442,6 +1453,10 @@ int amdgpu_device_resize_fb_bar(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) return 0; + /* PCI_EXT_CAP_ID_VNDR extended capability is located at 0x100 */ + if (!pci_find_ext_capability(adev->pdev, PCI_EXT_CAP_ID_VNDR)) + DRM_WARN("System can't access extended configuration space,please check!!\n"); + /* skip if the bios has already enabled large BAR */ if (adev->gmc.real_vram_size && (pci_resource_len(adev->pdev, 0) >= adev->gmc.real_vram_size)) @@ -5680,6 +5695,7 @@ retry: /* Rest of adevs pre asic reset from XGMI hive. */ /* Aldebaran and gfx_11_0_3 support ras in SRIOV, so need resume ras during reset */ if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 2) || + amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) || amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(11, 0, 3)) amdgpu_ras_resume(adev); } else { @@ -6101,6 +6117,20 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev) struct amdgpu_reset_context reset_context; u32 memsize; struct list_head device_list; + struct amdgpu_hive_info *hive; + int hive_ras_recovery = 0; + struct amdgpu_ras *ras; + + /* PCI error slot reset should be skipped During RAS recovery */ + hive = amdgpu_get_xgmi_hive(adev); + if (hive) { + hive_ras_recovery = atomic_read(&hive->ras_recovery); + amdgpu_put_xgmi_hive(hive); + } + ras = amdgpu_ras_get_context(adev); + if ((amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3)) && + ras && (atomic_read(&ras->in_recovery) || hive_ras_recovery)) + return PCI_ERS_RESULT_RECOVERED; DRM_INFO("PCI error: slot reset callback!!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index c7d60dd0fb97..118288b64487 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -27,6 +27,7 @@ #include "amdgpu_discovery.h" #include "soc15_hw_ip.h" #include "discovery.h" +#include "amdgpu_ras.h" #include "soc15.h" #include "gfx_v9_0.h" @@ -98,6 +99,7 @@ #define FIRMWARE_IP_DISCOVERY "amdgpu/ip_discovery.bin" MODULE_FIRMWARE(FIRMWARE_IP_DISCOVERY); +#define mmIP_DISCOVERY_VERSION 0x16A00 #define mmRCC_CONFIG_MEMSIZE 0xde3 #define mmMP0_SMN_C2PMSG_33 0x16061 #define mmMM_INDEX 0x0 @@ -518,7 +520,9 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev) out: kfree(adev->mman.discovery_bin); adev->mman.discovery_bin = NULL; - + if ((amdgpu_discovery != 2) && + (RREG32(mmIP_DISCOVERY_VERSION) == 4)) + amdgpu_ras_query_boot_status(adev, 4); return r; } @@ -1278,11 +1282,10 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) * 0b10 : encode is disabled * 0b01 : decode is disabled */ - adev->vcn.vcn_config[adev->vcn.num_vcn_inst] = - ip->revision & 0xc0; - ip->revision &= ~0xc0; if (adev->vcn.num_vcn_inst < AMDGPU_MAX_VCN_INSTANCES) { + adev->vcn.vcn_config[adev->vcn.num_vcn_inst] = + ip->revision & 0xc0; adev->vcn.num_vcn_inst++; adev->vcn.inst_mask |= (1U << ip->instance_number); @@ -1293,6 +1296,7 @@ static int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) adev->vcn.num_vcn_inst + 1, AMDGPU_MAX_VCN_INSTANCES); } + ip->revision &= ~0xc0; } if (le16_to_cpu(ip->hw_id) == SDMA0_HWID || le16_to_cpu(ip->hw_id) == SDMA1_HWID || diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c index decbbe3d4f06..055ba2ea4c12 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c @@ -377,6 +377,10 @@ amdgpu_dma_buf_move_notify(struct dma_buf_attachment *attach) struct amdgpu_vm_bo_base *bo_base; int r; + /* FIXME: This should be after the "if", but needs a fix to make sure + * DMABuf imports are initialized in the right VM list. + */ + amdgpu_vm_bo_invalidate(adev, bo, false); if (!bo->tbo.resource || bo->tbo.resource->mem_type == TTM_PL_SYSTEM) return; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 971acf01bea6..161ecf9b4174 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -366,7 +366,7 @@ module_param_named(aspm, amdgpu_aspm, int, 0444); * 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, -2 = autowith displays)"); +MODULE_PARM_DESC(runpm, "PX runtime pm (2 = force enable with BAMACO, 1 = force enable with BACO, 0 = disable, -1 = auto, -2 = auto with displays)"); module_param_named(runpm, amdgpu_runtime_pm, int, 0444); /** @@ -593,7 +593,7 @@ module_param_named(timeout_period, amdgpu_watchdog_timer.period, uint, 0644); #ifdef CONFIG_DRM_AMDGPU_SI #if IS_ENABLED(CONFIG_DRM_RADEON) || IS_ENABLED(CONFIG_DRM_RADEON_MODULE) -int amdgpu_si_support = 0; +int amdgpu_si_support; MODULE_PARM_DESC(si_support, "SI support (1 = enabled, 0 = disabled (default))"); #else int amdgpu_si_support = 1; @@ -612,7 +612,7 @@ module_param_named(si_support, amdgpu_si_support, int, 0444); #ifdef CONFIG_DRM_AMDGPU_CIK #if IS_ENABLED(CONFIG_DRM_RADEON) || IS_ENABLED(CONFIG_DRM_RADEON_MODULE) -int amdgpu_cik_support = 0; +int amdgpu_cik_support; MODULE_PARM_DESC(cik_support, "CIK support (1 = enabled, 0 = disabled (default))"); #else int amdgpu_cik_support = 1; @@ -2476,6 +2476,7 @@ static int amdgpu_pmops_suspend(struct device *dev) struct drm_device *drm_dev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(drm_dev); + adev->suspend_complete = false; if (amdgpu_acpi_is_s0ix_active(adev)) adev->in_s0ix = true; else if (amdgpu_acpi_is_s3_active(adev)) @@ -2490,6 +2491,7 @@ static int amdgpu_pmops_suspend_noirq(struct device *dev) struct drm_device *drm_dev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(drm_dev); + adev->suspend_complete = true; if (amdgpu_acpi_should_gpu_reset(adev)) return amdgpu_asic_reset(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 49a5f1c73b3e..22aeee8adb71 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -187,7 +187,34 @@ static int amdgpu_gem_object_open(struct drm_gem_object *obj, else ++bo_va->ref_count; amdgpu_bo_unreserve(abo); - return 0; + + /* Validate and add eviction fence to DMABuf imports with dynamic + * attachment in compute VMs. Re-validation will be done by + * amdgpu_vm_validate. Fences are on the reservation shared with the + * export, which is currently required to be validated and fenced + * already by amdgpu_amdkfd_gpuvm_restore_process_bos. + * + * Nested locking below for the case that a GEM object is opened in + * kfd_mem_export_dmabuf. Since the lock below is only taken for imports, + * but not for export, this is a different lock class that cannot lead to + * circular lock dependencies. + */ + if (!vm->is_compute_context || !vm->process_info) + return 0; + if (!obj->import_attach || + !dma_buf_is_dynamic(obj->import_attach->dmabuf)) + return 0; + mutex_lock_nested(&vm->process_info->lock, 1); + if (!WARN_ON(!vm->process_info->eviction_fence)) { + r = amdgpu_amdkfd_bo_validate_and_fence(abo, AMDGPU_GEM_DOMAIN_GTT, + &vm->process_info->eviction_fence->base); + if (r) + dev_warn(adev->dev, "%d: validate_and_fence failed: %d\n", + vm->task_info.pid, r); + } + mutex_unlock(&vm->process_info->lock); + + return r; } static void amdgpu_gem_object_close(struct drm_gem_object *obj, @@ -682,10 +709,10 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, uint64_t vm_size; int r = 0; - if (args->va_address < AMDGPU_VA_RESERVED_SIZE) { + if (args->va_address < AMDGPU_VA_RESERVED_BOTTOM) { dev_dbg(dev->dev, "va_address 0x%llx is in reserved area 0x%llx\n", - args->va_address, AMDGPU_VA_RESERVED_SIZE); + args->va_address, AMDGPU_VA_RESERVED_BOTTOM); return -EINVAL; } @@ -701,7 +728,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, args->va_address &= AMDGPU_GMC_HOLE_MASK; vm_size = adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE; - vm_size -= AMDGPU_VA_RESERVED_SIZE; + vm_size -= AMDGPU_VA_RESERVED_TOP; if (args->va_address + args->map_size > vm_size) { dev_dbg(dev->dev, "va_address 0x%llx is in top reserved area 0x%llx\n", diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 82b4b2019fca..78229b61c233 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -643,8 +643,8 @@ int amdgpu_gfx_enable_kcq(struct amdgpu_device *adev, int xcc_id) kiq->pmf->kiq_set_resources(kiq_ring, queue_mask); for (i = 0; i < adev->gfx.num_compute_rings; i++) { j = i + xcc_id * adev->gfx.num_compute_rings; - kiq->pmf->kiq_map_queues(kiq_ring, - &adev->gfx.compute_ring[j]); + kiq->pmf->kiq_map_queues(kiq_ring, + &adev->gfx.compute_ring[j]); } r = amdgpu_ring_test_helper(kiq_ring); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 55784a9f26c4..d4a848c51a83 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -52,7 +52,7 @@ int amdgpu_gmc_pdb0_alloc(struct amdgpu_device *adev) struct amdgpu_bo_param bp; u64 vram_size = adev->gmc.xgmi.node_segment_size * adev->gmc.xgmi.num_physical_nodes; uint32_t pde0_page_shift = adev->gmc.vmid0_page_table_block_size + 21; - uint32_t npdes = (vram_size + (1ULL << pde0_page_shift) -1) >> pde0_page_shift; + uint32_t npdes = (vram_size + (1ULL << pde0_page_shift) - 1) >> pde0_page_shift; memset(&bp, 0, sizeof(bp)); bp.size = PAGE_ALIGN((npdes + 1) * 8); @@ -746,6 +746,59 @@ error_unlock_reset: return r; } +void amdgpu_gmc_fw_reg_write_reg_wait(struct amdgpu_device *adev, + uint32_t reg0, uint32_t reg1, + uint32_t ref, uint32_t mask, + uint32_t xcc_inst) +{ + struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_inst]; + struct amdgpu_ring *ring = &kiq->ring; + signed long r, cnt = 0; + unsigned long flags; + uint32_t seq; + + if (adev->mes.ring.sched.ready) { + amdgpu_mes_reg_write_reg_wait(adev, reg0, reg1, + ref, mask); + return; + } + + spin_lock_irqsave(&kiq->ring_lock, flags); + amdgpu_ring_alloc(ring, 32); + amdgpu_ring_emit_reg_write_reg_wait(ring, reg0, reg1, + ref, mask); + r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); + if (r) + goto failed_undo; + + amdgpu_ring_commit(ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); + + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + + /* don't wait anymore for IRQ context */ + if (r < 1 && in_interrupt()) + goto failed_kiq; + + might_sleep(); + while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) { + + msleep(MAX_KIQ_REG_BAILOUT_INTERVAL); + r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); + } + + if (cnt > MAX_KIQ_REG_TRY) + goto failed_kiq; + + return; + +failed_undo: + amdgpu_ring_undo(ring); + spin_unlock_irqrestore(&kiq->ring_lock, flags); +failed_kiq: + dev_err(adev->dev, "failed to write reg %x wait reg %x\n", reg0, reg1); +} + /** * amdgpu_gmc_tmz_set -- check and set if a device supports TMZ * @adev: amdgpu_device pointer diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index e699d1ca8deb..17f40ea1104b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -417,6 +417,10 @@ void amdgpu_gmc_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid, uint32_t flush_type, bool all_hub, uint32_t inst); +void amdgpu_gmc_fw_reg_write_reg_wait(struct amdgpu_device *adev, + uint32_t reg0, uint32_t reg1, + uint32_t ref, uint32_t mask, + uint32_t xcc_inst); extern void amdgpu_gmc_tmz_set(struct amdgpu_device *adev); extern void amdgpu_gmc_noretry_set(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index ddd0891da116..3d7fcdeaf8cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -62,9 +62,8 @@ int amdgpu_pasid_alloc(unsigned int bits) int pasid = -EINVAL; for (bits = min(bits, 31U); bits > 0; bits--) { - pasid = ida_simple_get(&amdgpu_pasid_ida, - 1U << (bits - 1), 1U << bits, - GFP_KERNEL); + pasid = ida_alloc_range(&amdgpu_pasid_ida, 1U << (bits - 1), + (1U << bits) - 1, GFP_KERNEL); if (pasid != -ENOSPC) break; } @@ -82,7 +81,7 @@ int amdgpu_pasid_alloc(unsigned int bits) void amdgpu_pasid_free(u32 pasid) { trace_amdgpu_pasid_freed(pasid); - ida_simple_remove(&amdgpu_pasid_ida, pasid); + ida_free(&amdgpu_pasid_ida, pasid); } static void amdgpu_pasid_free_cb(struct dma_fence *fence, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index bf4f48fe438d..a2df3025a754 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -894,14 +894,14 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) dev_info->ids_flags |= AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD; vm_size = adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE; - vm_size -= AMDGPU_VA_RESERVED_SIZE; + vm_size -= AMDGPU_VA_RESERVED_TOP; /* Older VCE FW versions are buggy and can handle only 40bits */ if (adev->vce.fw_version && adev->vce.fw_version < AMDGPU_VCE_FW_53_45) vm_size = min(vm_size, 1ULL << 40); - dev_info->virtual_address_offset = AMDGPU_VA_RESERVED_SIZE; + dev_info->virtual_address_offset = AMDGPU_VA_RESERVED_BOTTOM; dev_info->virtual_address_max = min(vm_size, AMDGPU_GMC_HOLE_START); @@ -1114,6 +1114,15 @@ int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) } ui32 >>= 8; break; + case AMDGPU_INFO_SENSOR_GPU_INPUT_POWER: + /* get input GPU power */ + if (amdgpu_dpm_read_sensor(adev, + AMDGPU_PP_SENSOR_GPU_INPUT_POWER, + (void *)&ui32, &ui32_size)) { + return -EINVAL; + } + ui32 >>= 8; + break; case AMDGPU_INFO_SENSOR_VDDNB: /* get VDDNB in millivolts */ if (amdgpu_dpm_read_sensor(adev, @@ -1370,6 +1379,10 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) goto error_vm; } + r = amdgpu_seq64_map(adev, &fpriv->vm, &fpriv->seq64_va); + if (r) + goto error_vm; + mutex_init(&fpriv->bo_list_lock); idr_init_base(&fpriv->bo_list_handles, 1); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c index 59fafb8392e0..24ad4b97177b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.c @@ -27,6 +27,16 @@ #include "umc/umc_6_7_0_offset.h" #include "umc/umc_6_7_0_sh_mask.h" +static bool amdgpu_mca_is_deferred_error(struct amdgpu_device *adev, + uint64_t mc_status) +{ + if (adev->umc.ras->check_ecc_err_status) + return adev->umc.ras->check_ecc_err_status(adev, + AMDGPU_MCA_ERROR_TYPE_DE, &mc_status); + + return false; +} + void amdgpu_mca_query_correctable_error_count(struct amdgpu_device *adev, uint64_t mc_status_addr, unsigned long *error_count) @@ -202,16 +212,16 @@ int amdgpu_mca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable) static void amdgpu_mca_smu_mca_bank_dump(struct amdgpu_device *adev, int idx, struct mca_bank_entry *entry) { - dev_info(adev->dev, "[Hardware error] Accelerator Check Architecture events logged\n"); - dev_info(adev->dev, "[Hardware error] aca entry[%02d].STATUS=0x%016llx\n", + dev_info(adev->dev, HW_ERR "Accelerator Check Architecture events logged\n"); + dev_info(adev->dev, HW_ERR "aca entry[%02d].STATUS=0x%016llx\n", idx, entry->regs[MCA_REG_IDX_STATUS]); - dev_info(adev->dev, "[Hardware error] aca entry[%02d].ADDR=0x%016llx\n", + dev_info(adev->dev, HW_ERR "aca entry[%02d].ADDR=0x%016llx\n", idx, entry->regs[MCA_REG_IDX_ADDR]); - dev_info(adev->dev, "[Hardware error] aca entry[%02d].MISC0=0x%016llx\n", + dev_info(adev->dev, HW_ERR "aca entry[%02d].MISC0=0x%016llx\n", idx, entry->regs[MCA_REG_IDX_MISC0]); - dev_info(adev->dev, "[Hardware error] aca entry[%02d].IPID=0x%016llx\n", + dev_info(adev->dev, HW_ERR "aca entry[%02d].IPID=0x%016llx\n", idx, entry->regs[MCA_REG_IDX_IPID]); - dev_info(adev->dev, "[Hardware error] aca entry[%02d].SYND=0x%016llx\n", + dev_info(adev->dev, HW_ERR "aca entry[%02d].SYND=0x%016llx\n", idx, entry->regs[MCA_REG_IDX_SYND]); } @@ -256,9 +266,14 @@ int amdgpu_mca_smu_log_ras_error(struct amdgpu_device *adev, enum amdgpu_ras_blo if (type == AMDGPU_MCA_ERROR_TYPE_UE) amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, &err_addr, (uint64_t)count); - else - amdgpu_ras_error_statistic_ce_count(err_data, - &mcm_info, &err_addr, (uint64_t)count); + else { + if (amdgpu_mca_is_deferred_error(adev, entry->regs[MCA_REG_IDX_STATUS])) + amdgpu_ras_error_statistic_de_count(err_data, + &mcm_info, &err_addr, (uint64_t)count); + else + amdgpu_ras_error_statistic_ce_count(err_data, + &mcm_info, &err_addr, (uint64_t)count); + } } out_mca_release: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h index b399f1b62887..b964110ed1e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mca.h @@ -65,6 +65,7 @@ enum amdgpu_mca_ip { enum amdgpu_mca_error_type { AMDGPU_MCA_ERROR_TYPE_UE = 0, AMDGPU_MCA_ERROR_TYPE_CE, + AMDGPU_MCA_ERROR_TYPE_DE, }; struct amdgpu_mca_ras_block { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index da48b6da0107..a98e03e0a51f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -1398,7 +1398,7 @@ int amdgpu_mes_self_test(struct amdgpu_device *adev) goto error_fini; } - ctx_data.meta_data_gpu_addr = AMDGPU_VA_RESERVED_SIZE; + ctx_data.meta_data_gpu_addr = AMDGPU_VA_RESERVED_BOTTOM; r = amdgpu_mes_ctx_map_meta_data(adev, vm, &ctx_data); if (r) { DRM_ERROR("failed to map ctx meta data\n"); @@ -1565,9 +1565,9 @@ void amdgpu_debugfs_mes_event_log_init(struct amdgpu_device *adev) #if defined(CONFIG_DEBUG_FS) struct drm_minor *minor = adev_to_drm(adev)->primary; struct dentry *root = minor->debugfs_root; - - debugfs_create_file("amdgpu_mes_event_log", 0444, root, - adev, &amdgpu_debugfs_mes_event_log_fops); + if (adev->enable_mes) + debugfs_create_file("amdgpu_mes_event_log", 0444, root, + adev, &amdgpu_debugfs_mes_event_log_fops); #endif } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 0328616473f8..d9e5eb24341d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -291,21 +291,22 @@ static int psp_memory_training_init(struct psp_context *psp) struct psp_memory_training_context *ctx = &psp->mem_train_ctx; if (ctx->init != PSP_MEM_TRAIN_RESERVE_SUCCESS) { - DRM_DEBUG("memory training is not supported!\n"); + dev_dbg(psp->adev->dev, "memory training is not supported!\n"); return 0; } ctx->sys_cache = kzalloc(ctx->train_data_size, GFP_KERNEL); if (ctx->sys_cache == NULL) { - DRM_ERROR("alloc mem_train_ctx.sys_cache failed!\n"); + dev_err(psp->adev->dev, "alloc mem_train_ctx.sys_cache failed!\n"); ret = -ENOMEM; goto Err_out; } - DRM_DEBUG("train_data_size:%llx,p2c_train_data_offset:%llx,c2p_train_data_offset:%llx.\n", - ctx->train_data_size, - ctx->p2c_train_data_offset, - ctx->c2p_train_data_offset); + dev_dbg(psp->adev->dev, + "train_data_size:%llx,p2c_train_data_offset:%llx,c2p_train_data_offset:%llx.\n", + ctx->train_data_size, + ctx->p2c_train_data_offset, + ctx->c2p_train_data_offset); ctx->init = PSP_MEM_TRAIN_INIT_SUCCESS; return 0; @@ -407,7 +408,7 @@ static int psp_sw_init(void *handle) psp->cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL); if (!psp->cmd) { - DRM_ERROR("Failed to allocate memory to command buffer!\n"); + dev_err(adev->dev, "Failed to allocate memory to command buffer!\n"); ret = -ENOMEM; } @@ -454,13 +455,13 @@ static int psp_sw_init(void *handle) if (mem_training_ctx->enable_mem_training) { ret = psp_memory_training_init(psp); if (ret) { - DRM_ERROR("Failed to initialize memory training!\n"); + dev_err(adev->dev, "Failed to initialize memory training!\n"); return ret; } ret = psp_mem_training(psp, PSP_MEM_TRAIN_COLD_BOOT); if (ret) { - DRM_ERROR("Failed to process memory training!\n"); + dev_err(adev->dev, "Failed to process memory training!\n"); return ret; } } @@ -675,9 +676,11 @@ psp_cmd_submit_buf(struct psp_context *psp, */ if (!skip_unsupport && (psp->cmd_buf_mem->resp.status || !timeout) && !ras_intr) { if (ucode) - DRM_WARN("failed to load ucode %s(0x%X) ", - amdgpu_ucode_name(ucode->ucode_id), ucode->ucode_id); - DRM_WARN("psp gfx command %s(0x%X) failed and response status is (0x%X)\n", + dev_warn(psp->adev->dev, + "failed to load ucode %s(0x%X) ", + amdgpu_ucode_name(ucode->ucode_id), ucode->ucode_id); + dev_warn(psp->adev->dev, + "psp gfx command %s(0x%X) failed and response status is (0x%X)\n", psp_gfx_cmd_name(psp->cmd_buf_mem->cmd_id), psp->cmd_buf_mem->cmd_id, psp->cmd_buf_mem->resp.status); /* If any firmware (including CAP) load fails under SRIOV, it should @@ -807,7 +810,7 @@ static int psp_tmr_init(struct psp_context *psp) psp->fw_pri_buf) { ret = psp_load_toc(psp, &tmr_size); if (ret) { - DRM_ERROR("Failed to load toc\n"); + dev_err(psp->adev->dev, "Failed to load toc\n"); return ret; } } @@ -855,7 +858,7 @@ static int psp_tmr_load(struct psp_context *psp) psp_prep_tmr_cmd_buf(psp, cmd, psp->tmr_mc_addr, psp->tmr_bo); if (psp->tmr_bo) - DRM_INFO("reserve 0x%lx from 0x%llx for PSP TMR\n", + dev_info(psp->adev->dev, "reserve 0x%lx from 0x%llx for PSP TMR\n", amdgpu_bo_size(psp->tmr_bo), psp->tmr_mc_addr); ret = psp_cmd_submit_buf(psp, NULL, cmd, @@ -1113,7 +1116,7 @@ int psp_reg_program(struct psp_context *psp, enum psp_reg_prog_id reg, psp_prep_reg_prog_cmd_buf(cmd, reg, value); ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr); if (ret) - DRM_ERROR("PSP failed to program reg id %d", reg); + dev_err(psp->adev->dev, "PSP failed to program reg id %d\n", reg); release_psp_cmd_buf(psp); @@ -1526,22 +1529,22 @@ static void psp_ras_ta_check_status(struct psp_context *psp) switch (ras_cmd->ras_status) { case TA_RAS_STATUS__ERROR_UNSUPPORTED_IP: dev_warn(psp->adev->dev, - "RAS WARNING: cmd failed due to unsupported ip\n"); + "RAS WARNING: cmd failed due to unsupported ip\n"); break; case TA_RAS_STATUS__ERROR_UNSUPPORTED_ERROR_INJ: dev_warn(psp->adev->dev, - "RAS WARNING: cmd failed due to unsupported error injection\n"); + "RAS WARNING: cmd failed due to unsupported error injection\n"); break; case TA_RAS_STATUS__SUCCESS: break; case TA_RAS_STATUS__TEE_ERROR_ACCESS_DENIED: if (ras_cmd->cmd_id == TA_RAS_COMMAND__TRIGGER_ERROR) dev_warn(psp->adev->dev, - "RAS WARNING: Inject error to critical region is not allowed\n"); + "RAS WARNING: Inject error to critical region is not allowed\n"); break; default: dev_warn(psp->adev->dev, - "RAS WARNING: ras status = 0x%X\n", ras_cmd->ras_status); + "RAS WARNING: ras status = 0x%X\n", ras_cmd->ras_status); break; } } @@ -1565,7 +1568,7 @@ int psp_ras_invoke(struct psp_context *psp, uint32_t ta_cmd_id) return ret; if (ras_cmd->if_version > RAS_TA_HOST_IF_VER) { - DRM_WARN("RAS: Unsupported Interface"); + dev_warn(psp->adev->dev, "RAS: Unsupported Interface\n"); return -EINVAL; } @@ -1715,7 +1718,7 @@ int psp_ras_initialize(struct psp_context *psp) psp->ras_context.context.initialized = true; else { if (ras_cmd->ras_status) - dev_warn(psp->adev->dev, "RAS Init Status: 0x%X\n", ras_cmd->ras_status); + dev_warn(adev->dev, "RAS Init Status: 0x%X\n", ras_cmd->ras_status); /* fail to load RAS TA */ psp->ras_context.context.initialized = false; @@ -1779,6 +1782,31 @@ int psp_ras_trigger_error(struct psp_context *psp, return 0; } + +int psp_ras_query_address(struct psp_context *psp, + struct ta_ras_query_address_input *addr_in, + struct ta_ras_query_address_output *addr_out) +{ + struct ta_ras_shared_memory *ras_cmd; + int ret; + + if (!psp->ras_context.context.initialized) + return -EINVAL; + + ras_cmd = (struct ta_ras_shared_memory *)psp->ras_context.context.mem_context.shared_buf; + memset(ras_cmd, 0, sizeof(struct ta_ras_shared_memory)); + + ras_cmd->cmd_id = TA_RAS_COMMAND__QUERY_ADDRESS; + ras_cmd->ras_in_message.address = *addr_in; + + ret = psp_ras_invoke(psp, ras_cmd->cmd_id); + if (ret || ras_cmd->ras_status || psp->cmd_buf_mem->resp.status) + return -EINVAL; + + *addr_out = ras_cmd->ras_out_message.address; + + return 0; +} // ras end // HDCP start @@ -2125,19 +2153,14 @@ int amdgpu_psp_wait_for_bootloader(struct amdgpu_device *adev) return ret; } -int amdgpu_psp_query_boot_status(struct amdgpu_device *adev) +bool amdgpu_psp_get_ras_capability(struct psp_context *psp) { - struct psp_context *psp = &adev->psp; - int ret = 0; - - if (amdgpu_sriov_vf(adev) || (adev->flags & AMD_IS_APU)) - return 0; - if (psp->funcs && - psp->funcs->query_boot_status) - ret = psp->funcs->query_boot_status(psp); - - return ret; + psp->funcs->get_ras_capability) { + return psp->funcs->get_ras_capability(psp); + } else { + return false; + } } static int psp_hw_start(struct psp_context *psp) @@ -2150,7 +2173,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_kdb != NULL)) { ret = psp_bootloader_load_kdb(psp); if (ret) { - DRM_ERROR("PSP load kdb failed!\n"); + dev_err(adev->dev, "PSP load kdb failed!\n"); return ret; } } @@ -2159,7 +2182,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_spl != NULL)) { ret = psp_bootloader_load_spl(psp); if (ret) { - DRM_ERROR("PSP load spl failed!\n"); + dev_err(adev->dev, "PSP load spl failed!\n"); return ret; } } @@ -2168,7 +2191,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_sysdrv != NULL)) { ret = psp_bootloader_load_sysdrv(psp); if (ret) { - DRM_ERROR("PSP load sys drv failed!\n"); + dev_err(adev->dev, "PSP load sys drv failed!\n"); return ret; } } @@ -2177,7 +2200,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_soc_drv != NULL)) { ret = psp_bootloader_load_soc_drv(psp); if (ret) { - DRM_ERROR("PSP load soc drv failed!\n"); + dev_err(adev->dev, "PSP load soc drv failed!\n"); return ret; } } @@ -2186,7 +2209,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_intf_drv != NULL)) { ret = psp_bootloader_load_intf_drv(psp); if (ret) { - DRM_ERROR("PSP load intf drv failed!\n"); + dev_err(adev->dev, "PSP load intf drv failed!\n"); return ret; } } @@ -2195,7 +2218,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_dbg_drv != NULL)) { ret = psp_bootloader_load_dbg_drv(psp); if (ret) { - DRM_ERROR("PSP load dbg drv failed!\n"); + dev_err(adev->dev, "PSP load dbg drv failed!\n"); return ret; } } @@ -2204,7 +2227,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_ras_drv != NULL)) { ret = psp_bootloader_load_ras_drv(psp); if (ret) { - DRM_ERROR("PSP load ras_drv failed!\n"); + dev_err(adev->dev, "PSP load ras_drv failed!\n"); return ret; } } @@ -2213,7 +2236,7 @@ static int psp_hw_start(struct psp_context *psp) (psp->funcs->bootloader_load_sos != NULL)) { ret = psp_bootloader_load_sos(psp); if (ret) { - DRM_ERROR("PSP load sos failed!\n"); + dev_err(adev->dev, "PSP load sos failed!\n"); return ret; } } @@ -2221,7 +2244,7 @@ static int psp_hw_start(struct psp_context *psp) ret = psp_ring_create(psp, PSP_RING_TYPE__KM); if (ret) { - DRM_ERROR("PSP create ring failed!\n"); + dev_err(adev->dev, "PSP create ring failed!\n"); return ret; } @@ -2231,7 +2254,7 @@ static int psp_hw_start(struct psp_context *psp) if (!psp_boottime_tmr(psp)) { ret = psp_tmr_init(psp); if (ret) { - DRM_ERROR("PSP tmr init failed!\n"); + dev_err(adev->dev, "PSP tmr init failed!\n"); return ret; } } @@ -2250,7 +2273,7 @@ skip_pin_bo: ret = psp_tmr_load(psp); if (ret) { - DRM_ERROR("PSP load tmr failed!\n"); + dev_err(adev->dev, "PSP load tmr failed!\n"); return ret; } @@ -2518,7 +2541,8 @@ static void psp_print_fw_hdr(struct psp_context *psp, } } -static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode, +static int psp_prep_load_ip_fw_cmd_buf(struct psp_context *psp, + struct amdgpu_firmware_info *ucode, struct psp_gfx_cmd_resp *cmd) { int ret; @@ -2531,7 +2555,7 @@ static int psp_prep_load_ip_fw_cmd_buf(struct amdgpu_firmware_info *ucode, ret = psp_get_fw_type(ucode, &cmd->cmd.cmd_load_ip_fw.fw_type); if (ret) - DRM_ERROR("Unknown firmware type\n"); + dev_err(psp->adev->dev, "Unknown firmware type\n"); return ret; } @@ -2542,7 +2566,7 @@ int psp_execute_ip_fw_load(struct psp_context *psp, int ret = 0; struct psp_gfx_cmd_resp *cmd = acquire_psp_cmd_buf(psp); - ret = psp_prep_load_ip_fw_cmd_buf(ucode, cmd); + ret = psp_prep_load_ip_fw_cmd_buf(psp, ucode, cmd); if (!ret) { ret = psp_cmd_submit_buf(psp, ucode, cmd, psp->fence_buf_mc_addr); @@ -2601,13 +2625,13 @@ static int psp_load_smu_fw(struct psp_context *psp) amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(11, 0, 2)))) { ret = amdgpu_dpm_set_mp1_state(adev, PP_MP1_STATE_UNLOAD); if (ret) - DRM_WARN("Failed to set MP1 state prepare for reload\n"); + dev_err(adev->dev, "Failed to set MP1 state prepare for reload\n"); } ret = psp_execute_ip_fw_load(psp, ucode); if (ret) - DRM_ERROR("PSP load smu failed!\n"); + dev_err(adev->dev, "PSP load smu failed!\n"); return ret; } @@ -2712,7 +2736,7 @@ static int psp_load_non_psp_fw(struct psp_context *psp) adev->virt.autoload_ucode_id : AMDGPU_UCODE_ID_RLC_G)) { ret = psp_rlc_autoload_start(psp); if (ret) { - DRM_ERROR("Failed to start rlc autoload\n"); + dev_err(adev->dev, "Failed to start rlc autoload\n"); return ret; } } @@ -2734,7 +2758,7 @@ static int psp_load_fw(struct amdgpu_device *adev) ret = psp_ring_init(psp, PSP_RING_TYPE__KM); if (ret) { - DRM_ERROR("PSP ring init failed!\n"); + dev_err(adev->dev, "PSP ring init failed!\n"); goto failed; } } @@ -2749,13 +2773,13 @@ static int psp_load_fw(struct amdgpu_device *adev) ret = psp_asd_initialize(psp); if (ret) { - DRM_ERROR("PSP load asd failed!\n"); + dev_err(adev->dev, "PSP load asd failed!\n"); goto failed1; } ret = psp_rl_load(adev); if (ret) { - DRM_ERROR("PSP load RL failed!\n"); + dev_err(adev->dev, "PSP load RL failed!\n"); goto failed1; } @@ -2775,7 +2799,7 @@ static int psp_load_fw(struct amdgpu_device *adev) ret = psp_ras_initialize(psp); if (ret) dev_err(psp->adev->dev, - "RAS: Failed to initialize RAS\n"); + "RAS: Failed to initialize RAS\n"); ret = psp_hdcp_initialize(psp); if (ret) @@ -2828,7 +2852,7 @@ static int psp_hw_init(void *handle) ret = psp_load_fw(adev); if (ret) { - DRM_ERROR("PSP firmware loading failed\n"); + dev_err(adev->dev, "PSP firmware loading failed\n"); goto failed; } @@ -2875,7 +2899,7 @@ static int psp_suspend(void *handle) psp->xgmi_context.context.initialized) { ret = psp_xgmi_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate xgmi ta\n"); + dev_err(adev->dev, "Failed to terminate xgmi ta\n"); goto out; } } @@ -2883,46 +2907,46 @@ static int psp_suspend(void *handle) if (psp->ta_fw) { ret = psp_ras_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate ras ta\n"); + dev_err(adev->dev, "Failed to terminate ras ta\n"); goto out; } ret = psp_hdcp_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate hdcp ta\n"); + dev_err(adev->dev, "Failed to terminate hdcp ta\n"); goto out; } ret = psp_dtm_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate dtm ta\n"); + dev_err(adev->dev, "Failed to terminate dtm ta\n"); goto out; } ret = psp_rap_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate rap ta\n"); + dev_err(adev->dev, "Failed to terminate rap ta\n"); goto out; } ret = psp_securedisplay_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate securedisplay ta\n"); + dev_err(adev->dev, "Failed to terminate securedisplay ta\n"); goto out; } } ret = psp_asd_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate asd\n"); + dev_err(adev->dev, "Failed to terminate asd\n"); goto out; } ret = psp_tmr_terminate(psp); if (ret) { - DRM_ERROR("Failed to terminate tmr\n"); + dev_err(adev->dev, "Failed to terminate tmr\n"); goto out; } ret = psp_ring_stop(psp, PSP_RING_TYPE__KM); if (ret) - DRM_ERROR("PSP ring stop failed\n"); + dev_err(adev->dev, "PSP ring stop failed\n"); out: return ret; @@ -2934,12 +2958,12 @@ static int psp_resume(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct psp_context *psp = &adev->psp; - DRM_INFO("PSP is resuming...\n"); + dev_info(adev->dev, "PSP is resuming...\n"); if (psp->mem_train_ctx.enable_mem_training) { ret = psp_mem_training(psp, PSP_MEM_TRAIN_RESUME); if (ret) { - DRM_ERROR("Failed to process memory training!\n"); + dev_err(adev->dev, "Failed to process memory training!\n"); return ret; } } @@ -2956,7 +2980,7 @@ static int psp_resume(void *handle) ret = psp_asd_initialize(psp); if (ret) { - DRM_ERROR("PSP load asd failed!\n"); + dev_err(adev->dev, "PSP load asd failed!\n"); goto failed; } @@ -2980,7 +3004,7 @@ static int psp_resume(void *handle) ret = psp_ras_initialize(psp); if (ret) dev_err(psp->adev->dev, - "RAS: Failed to initialize RAS\n"); + "RAS: Failed to initialize RAS\n"); ret = psp_hdcp_initialize(psp); if (ret) @@ -3008,7 +3032,7 @@ static int psp_resume(void *handle) return 0; failed: - DRM_ERROR("PSP resume failed\n"); + dev_err(adev->dev, "PSP resume failed\n"); mutex_unlock(&adev->firmware.mutex); return ret; } @@ -3069,9 +3093,11 @@ int psp_ring_cmd_submit(struct psp_context *psp, write_frame = ring_buffer_start + (psp_write_ptr_reg / rb_frame_size_dw); /* Check invalid write_frame ptr address */ if ((write_frame < ring_buffer_start) || (ring_buffer_end < write_frame)) { - DRM_ERROR("ring_buffer_start = %p; ring_buffer_end = %p; write_frame = %p\n", - ring_buffer_start, ring_buffer_end, write_frame); - DRM_ERROR("write_frame is pointing to address out of bounds\n"); + dev_err(adev->dev, + "ring_buffer_start = %p; ring_buffer_end = %p; write_frame = %p\n", + ring_buffer_start, ring_buffer_end, write_frame); + dev_err(adev->dev, + "write_frame is pointing to address out of bounds\n"); return -EINVAL; } @@ -3597,7 +3623,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_read(struct device *dev, int ret; if (!adev->ip_blocks[AMD_IP_BLOCK_TYPE_PSP].status.late_initialized) { - DRM_INFO("PSP block is not ready yet."); + dev_info(adev->dev, "PSP block is not ready yet\n."); return -EBUSY; } @@ -3606,7 +3632,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_read(struct device *dev, mutex_unlock(&adev->psp.mutex); if (ret) { - DRM_ERROR("Failed to read USBC PD FW, err = %d", ret); + dev_err(adev->dev, "Failed to read USBC PD FW, err = %d\n", ret); return ret; } @@ -3628,7 +3654,7 @@ static ssize_t psp_usbc_pd_fw_sysfs_write(struct device *dev, void *fw_pri_cpu_addr; if (!adev->ip_blocks[AMD_IP_BLOCK_TYPE_PSP].status.late_initialized) { - DRM_INFO("PSP block is not ready yet."); + dev_err(adev->dev, "PSP block is not ready yet."); return -EBUSY; } @@ -3661,7 +3687,7 @@ rel_buf: release_firmware(usbc_pd_fw); fail: if (ret) { - DRM_ERROR("Failed to load USBC PD FW, err = %d", ret); + dev_err(adev->dev, "Failed to load USBC PD FW, err = %d", ret); count = ret; } @@ -3708,7 +3734,7 @@ static ssize_t amdgpu_psp_vbflash_write(struct file *filp, struct kobject *kobj, /* Safeguard against memory drain */ if (adev->psp.vbflash_image_size > AMD_VBIOS_FILE_MAX_SIZE_B) { - dev_err(adev->dev, "File size cannot exceed %u", AMD_VBIOS_FILE_MAX_SIZE_B); + dev_err(adev->dev, "File size cannot exceed %u\n", AMD_VBIOS_FILE_MAX_SIZE_B); kvfree(adev->psp.vbflash_tmp_buf); adev->psp.vbflash_tmp_buf = NULL; adev->psp.vbflash_image_size = 0; @@ -3727,7 +3753,7 @@ 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_dbg(adev->dev, "IFWI staged for update"); + dev_dbg(adev->dev, "IFWI staged for update\n"); return count; } @@ -3747,7 +3773,7 @@ static ssize_t amdgpu_psp_vbflash_read(struct file *filp, struct kobject *kobj, if (adev->psp.vbflash_image_size == 0) return -EINVAL; - dev_dbg(adev->dev, "PSP IFWI flash process initiated"); + dev_dbg(adev->dev, "PSP IFWI flash process initiated\n"); ret = amdgpu_bo_create_kernel(adev, adev->psp.vbflash_image_size, AMDGPU_GPU_PAGE_SIZE, @@ -3772,11 +3798,11 @@ rel_buf: adev->psp.vbflash_image_size = 0; if (ret) { - dev_err(adev->dev, "Failed to load IFWI, err = %d", ret); + dev_err(adev->dev, "Failed to load IFWI, err = %d\n", ret); return ret; } - dev_dbg(adev->dev, "PSP IFWI flash process done"); + dev_dbg(adev->dev, "PSP IFWI flash process done\n"); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index c4d9cbde55b9..9951bdd022de 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -134,7 +134,7 @@ struct psp_funcs { 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); - int (*query_boot_status)(struct psp_context *psp); + bool (*get_ras_capability)(struct psp_context *psp); }; struct ta_funcs { @@ -502,6 +502,9 @@ int psp_ras_enable_features(struct psp_context *psp, int psp_ras_trigger_error(struct psp_context *psp, struct ta_ras_trigger_error_input *info, uint32_t instance_mask); int psp_ras_terminate(struct psp_context *psp); +int psp_ras_query_address(struct psp_context *psp, + struct ta_ras_query_address_input *addr_in, + struct ta_ras_query_address_output *addr_out); int psp_hdcp_invoke(struct psp_context *psp, uint32_t ta_cmd_id); int psp_dtm_invoke(struct psp_context *psp, uint32_t ta_cmd_id); @@ -538,7 +541,5 @@ int psp_spatial_partition(struct psp_context *psp, int mode); int is_psp_fw_valid(struct psp_bin_desc bin); int amdgpu_psp_wait_for_bootloader(struct amdgpu_device *adev); - -int amdgpu_psp_query_boot_status(struct amdgpu_device *adev); - +bool amdgpu_psp_get_ras_capability(struct psp_context *psp); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c index 468a67b302d4..ca5c86e5f7cd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c @@ -362,7 +362,7 @@ static ssize_t ta_if_invoke_debugfs_write(struct file *fp, const char *buf, size } } - if (copy_to_user((char *)buf, context->mem_context.shared_buf, shared_buf_len)) + if (copy_to_user((char *)&buf[copy_pos], context->mem_context.shared_buf, shared_buf_len)) ret = -EFAULT; err_free_shared_buf: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 31823a30dea2..46f3d1013e8c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -39,6 +39,7 @@ #include "nbio_v7_9.h" #include "atom.h" #include "amdgpu_reset.h" +#include "amdgpu_psp.h" #ifdef CONFIG_X86_MCE_AMD #include <asm/mce.h> @@ -73,6 +74,8 @@ const char *ras_block_string[] = { "mca", "vcn", "jpeg", + "ih", + "mpio", }; const char *ras_mca_block_string[] = { @@ -94,7 +97,8 @@ const char *get_ras_block_str(struct ras_common_if *ras_block) if (!ras_block) return "NULL"; - if (ras_block->block >= AMDGPU_RAS_BLOCK_COUNT) + if (ras_block->block >= AMDGPU_RAS_BLOCK_COUNT || + ras_block->block >= ARRAY_SIZE(ras_block_string)) return "OUT OF RANGE"; if (ras_block->block == AMDGPU_RAS_BLOCK__MCA) @@ -116,6 +120,8 @@ const char *get_ras_block_str(struct ras_common_if *ras_block) /* typical ECC bad page rate is 1 bad page per 100MB VRAM */ #define RAS_BAD_PAGE_COVER (100 * 1024 * 1024ULL) +#define MAX_UMC_POISON_POLLING_TIME_ASYNC 100 //ms + enum amdgpu_ras_retire_page_reservation { AMDGPU_RAS_RETIRE_PAGE_RESERVED, AMDGPU_RAS_RETIRE_PAGE_PENDING, @@ -628,8 +634,12 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev, dev_warn(obj->adev->dev, "Failed to reset error counter and error status"); } - return sysfs_emit(buf, "%s: %lu\n%s: %lu\n", "ue", info.ue_count, - "ce", info.ce_count); + if (info.head.block == AMDGPU_RAS_BLOCK__UMC) + return sysfs_emit(buf, "%s: %lu\n%s: %lu\n%s: %lu\n", "ue", info.ue_count, + "ce", info.ce_count, "de", info.de_count); + else + return sysfs_emit(buf, "%s: %lu\n%s: %lu\n", "ue", info.ue_count, + "ce", info.ce_count); } /* obj begin */ @@ -1036,7 +1046,8 @@ static void amdgpu_ras_error_print_error_data(struct amdgpu_device *adev, struct ras_manager *ras_mgr, struct ras_err_data *err_data, const char *blk_name, - bool is_ue) + bool is_ue, + bool is_de) { struct amdgpu_smuio_mcm_config_info *mcm_info; struct ras_err_node *err_node; @@ -1065,25 +1076,50 @@ static void amdgpu_ras_error_print_error_data(struct amdgpu_device *adev, } } else { - for_each_ras_error(err_node, err_data) { - err_info = &err_node->err_info; - mcm_info = &err_info->mcm_info; - if (err_info->ce_count) { + if (is_de) { + for_each_ras_error(err_node, err_data) { + err_info = &err_node->err_info; + mcm_info = &err_info->mcm_info; + if (err_info->de_count) { + dev_info(adev->dev, "socket: %d, die: %d, " + "%lld new deferred hardware errors detected in %s block\n", + mcm_info->socket_id, + mcm_info->die_id, + err_info->de_count, + blk_name); + } + } + + for_each_ras_error(err_node, &ras_mgr->err_data) { + err_info = &err_node->err_info; + mcm_info = &err_info->mcm_info; dev_info(adev->dev, "socket: %d, die: %d, " - "%lld new correctable hardware errors detected in %s block\n", - mcm_info->socket_id, - mcm_info->die_id, - err_info->ce_count, - blk_name); + "%lld deferred hardware errors detected in total in %s block\n", + mcm_info->socket_id, mcm_info->die_id, + err_info->de_count, blk_name); + } + } else { + for_each_ras_error(err_node, err_data) { + err_info = &err_node->err_info; + mcm_info = &err_info->mcm_info; + if (err_info->ce_count) { + dev_info(adev->dev, "socket: %d, die: %d, " + "%lld new correctable hardware errors detected in %s block\n", + mcm_info->socket_id, + mcm_info->die_id, + err_info->ce_count, + blk_name); + } } - } - for_each_ras_error(err_node, &ras_mgr->err_data) { - err_info = &err_node->err_info; - mcm_info = &err_info->mcm_info; - dev_info(adev->dev, "socket: %d, die: %d, " - "%lld correctable hardware errors detected in total in %s block\n", - mcm_info->socket_id, mcm_info->die_id, err_info->ce_count, blk_name); + for_each_ras_error(err_node, &ras_mgr->err_data) { + err_info = &err_node->err_info; + mcm_info = &err_info->mcm_info; + dev_info(adev->dev, "socket: %d, die: %d, " + "%lld correctable hardware errors detected in total in %s block\n", + mcm_info->socket_id, mcm_info->die_id, + err_info->ce_count, blk_name); + } } } } @@ -1102,7 +1138,8 @@ static void amdgpu_ras_error_generate_report(struct amdgpu_device *adev, if (err_data->ce_count) { if (err_data_has_source_info(err_data)) { - amdgpu_ras_error_print_error_data(adev, ras_mgr, err_data, blk_name, false); + amdgpu_ras_error_print_error_data(adev, ras_mgr, err_data, + blk_name, false, false); } else if (!adev->aid_mask && adev->smuio.funcs && adev->smuio.funcs->get_socket_id && @@ -1124,7 +1161,8 @@ static void amdgpu_ras_error_generate_report(struct amdgpu_device *adev, if (err_data->ue_count) { if (err_data_has_source_info(err_data)) { - amdgpu_ras_error_print_error_data(adev, ras_mgr, err_data, blk_name, true); + amdgpu_ras_error_print_error_data(adev, ras_mgr, err_data, + blk_name, true, false); } else if (!adev->aid_mask && adev->smuio.funcs && adev->smuio.funcs->get_socket_id && @@ -1144,6 +1182,28 @@ static void amdgpu_ras_error_generate_report(struct amdgpu_device *adev, } } + if (err_data->de_count) { + if (err_data_has_source_info(err_data)) { + amdgpu_ras_error_print_error_data(adev, ras_mgr, err_data, + blk_name, false, true); + } else 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 " + "%ld deferred hardware errors " + "detected in %s block\n", + adev->smuio.funcs->get_socket_id(adev), + adev->smuio.funcs->get_die_id(adev), + ras_mgr->err_data.de_count, + blk_name); + } else { + dev_info(adev->dev, "%ld deferred hardware errors " + "detected in %s block\n", + ras_mgr->err_data.de_count, + blk_name); + } + } } static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, struct ras_err_data *err_data) @@ -1154,7 +1214,8 @@ static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, s if (err_data_has_source_info(err_data)) { for_each_ras_error(err_node, err_data) { err_info = &err_node->err_info; - + amdgpu_ras_error_statistic_de_count(&obj->err_data, + &err_info->mcm_info, NULL, err_info->de_count); amdgpu_ras_error_statistic_ce_count(&obj->err_data, &err_info->mcm_info, NULL, err_info->ce_count); amdgpu_ras_error_statistic_ue_count(&obj->err_data, @@ -1164,9 +1225,72 @@ static void amdgpu_rasmgr_error_data_statistic_update(struct ras_manager *obj, s /* for legacy asic path which doesn't has error source info */ obj->err_data.ue_count += err_data->ue_count; obj->err_data.ce_count += err_data->ce_count; + obj->err_data.de_count += err_data->de_count; } } +static struct ras_manager *get_ras_manager(struct amdgpu_device *adev, enum amdgpu_ras_block blk) +{ + struct ras_common_if head; + + memset(&head, 0, sizeof(head)); + head.block = blk; + + return amdgpu_ras_find_obj(adev, &head); +} + +int amdgpu_ras_bind_aca(struct amdgpu_device *adev, enum amdgpu_ras_block blk, + const struct aca_info *aca_info, void *data) +{ + struct ras_manager *obj; + + obj = get_ras_manager(adev, blk); + if (!obj) + return -EINVAL; + + return amdgpu_aca_add_handle(adev, &obj->aca_handle, ras_block_str(blk), aca_info, data); +} + +int amdgpu_ras_unbind_aca(struct amdgpu_device *adev, enum amdgpu_ras_block blk) +{ + struct ras_manager *obj; + + obj = get_ras_manager(adev, blk); + if (!obj) + return -EINVAL; + + amdgpu_aca_remove_handle(&obj->aca_handle); + + return 0; +} + +static int amdgpu_aca_log_ras_error_data(struct amdgpu_device *adev, enum amdgpu_ras_block blk, + enum aca_error_type type, struct ras_err_data *err_data) +{ + struct ras_manager *obj; + + obj = get_ras_manager(adev, blk); + if (!obj) + return -EINVAL; + + return amdgpu_aca_get_error_data(adev, &obj->aca_handle, type, err_data); +} + +ssize_t amdgpu_ras_aca_sysfs_read(struct device *dev, struct device_attribute *attr, + struct aca_handle *handle, char *buf, void *data) +{ + struct ras_manager *obj = container_of(handle, struct ras_manager, aca_handle); + struct ras_query_if info = { + .head = obj->head, + }; + + if (amdgpu_ras_query_error_status(obj->adev, &info)) + return -EINVAL; + + return sysfs_emit(buf, "%s: %lu\n%s: %lu\n", "ue", info.ue_count, + "ce", info.ce_count); +} + static int amdgpu_ras_query_error_status_helper(struct amdgpu_device *adev, struct ras_query_if *info, struct ras_err_data *err_data, @@ -1174,6 +1298,7 @@ static int amdgpu_ras_query_error_status_helper(struct amdgpu_device *adev, { enum amdgpu_ras_block blk = info ? info->head.block : AMDGPU_RAS_BLOCK_COUNT; struct amdgpu_ras_block_object *block_obj = NULL; + int ret; if (blk == AMDGPU_RAS_BLOCK_COUNT) return -EINVAL; @@ -1203,9 +1328,19 @@ static int amdgpu_ras_query_error_status_helper(struct amdgpu_device *adev, } } } else { - /* FIXME: add code to check return value later */ - amdgpu_mca_smu_log_ras_error(adev, blk, AMDGPU_MCA_ERROR_TYPE_UE, err_data); - amdgpu_mca_smu_log_ras_error(adev, blk, AMDGPU_MCA_ERROR_TYPE_CE, err_data); + if (amdgpu_aca_is_enabled(adev)) { + ret = amdgpu_aca_log_ras_error_data(adev, blk, ACA_ERROR_TYPE_UE, err_data); + if (ret) + return ret; + + ret = amdgpu_aca_log_ras_error_data(adev, blk, ACA_ERROR_TYPE_CE, err_data); + if (ret) + return ret; + } else { + /* FIXME: add code to check return value later */ + amdgpu_mca_smu_log_ras_error(adev, blk, AMDGPU_MCA_ERROR_TYPE_UE, err_data); + amdgpu_mca_smu_log_ras_error(adev, blk, AMDGPU_MCA_ERROR_TYPE_CE, err_data); + } } return 0; @@ -1239,6 +1374,7 @@ int amdgpu_ras_query_error_status(struct amdgpu_device *adev, struct ras_query_i info->ue_count = obj->err_data.ue_count; info->ce_count = obj->err_data.ce_count; + info->de_count = obj->err_data.de_count; amdgpu_ras_error_generate_report(adev, info, &err_data); @@ -1254,6 +1390,7 @@ int amdgpu_ras_reset_error_count(struct amdgpu_device *adev, struct amdgpu_ras_block_object *block_obj = amdgpu_ras_get_ras_block(adev, block, 0); struct amdgpu_ras *ras = amdgpu_ras_get_context(adev); const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs; + const struct aca_smu_funcs *smu_funcs = adev->aca.smu_funcs; struct amdgpu_hive_info *hive; int hive_ras_recovery = 0; @@ -1264,7 +1401,7 @@ int amdgpu_ras_reset_error_count(struct amdgpu_device *adev, } if (!amdgpu_ras_is_supported(adev, block) || - !amdgpu_ras_get_mca_debug_mode(adev)) + !amdgpu_ras_get_aca_debug_mode(adev)) return -EOPNOTSUPP; hive = amdgpu_get_xgmi_hive(adev); @@ -1276,7 +1413,8 @@ int amdgpu_ras_reset_error_count(struct amdgpu_device *adev, /* skip ras error reset in gpu reset */ if ((amdgpu_in_reset(adev) || atomic_read(&ras->in_recovery) || hive_ras_recovery) && - mca_funcs && mca_funcs->mca_set_debug_mode) + ((smu_funcs && smu_funcs->set_debug_mode) || + (mca_funcs && mca_funcs->mca_set_debug_mode))) return -EOPNOTSUPP; if (block_obj->hw_ops->reset_ras_error_count) @@ -1772,7 +1910,10 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) } } - amdgpu_mca_smu_debugfs_init(adev, dir); + if (amdgpu_aca_is_enabled(adev)) + amdgpu_aca_smu_debugfs_init(adev, dir); + else + amdgpu_mca_smu_debugfs_init(adev, dir); } /* debugfs end */ @@ -1900,7 +2041,7 @@ static void amdgpu_ras_interrupt_poison_consumption_handler(struct ras_manager * } } - amdgpu_umc_poison_handler(adev, false); + amdgpu_umc_poison_handler(adev, obj->head.block, false); if (block_obj->hw_ops && block_obj->hw_ops->handle_poison_consumption) poison_stat = block_obj->hw_ops->handle_poison_consumption(adev); @@ -1951,6 +2092,7 @@ static void amdgpu_ras_interrupt_umc_handler(struct ras_manager *obj, */ obj->err_data.ue_count += err_data.ue_count; obj->err_data.ce_count += err_data.ce_count; + obj->err_data.de_count += err_data.de_count; } amdgpu_ras_error_data_fini(&err_data); @@ -2520,6 +2662,32 @@ static void amdgpu_ras_validate_threshold(struct amdgpu_device *adev, } } +static int amdgpu_ras_page_retirement_thread(void *param) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)param; + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + + while (!kthread_should_stop()) { + + wait_event_interruptible(con->page_retirement_wq, + kthread_should_stop() || + atomic_read(&con->page_retirement_req_cnt)); + + if (kthread_should_stop()) + break; + + dev_info(adev->dev, "Start processing page retirement. request:%d\n", + atomic_read(&con->page_retirement_req_cnt)); + + atomic_dec(&con->page_retirement_req_cnt); + + amdgpu_umc_bad_page_polling_timeout(adev, + false, MAX_UMC_POISON_POLLING_TIME_ASYNC); + } + + return 0; +} + int amdgpu_ras_recovery_init(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); @@ -2583,6 +2751,16 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) } } + mutex_init(&con->page_retirement_lock); + init_waitqueue_head(&con->page_retirement_wq); + atomic_set(&con->page_retirement_req_cnt, 0); + con->page_retirement_thread = + kthread_run(amdgpu_ras_page_retirement_thread, adev, "umc_page_retirement"); + if (IS_ERR(con->page_retirement_thread)) { + con->page_retirement_thread = NULL; + dev_warn(adev->dev, "Failed to create umc_page_retirement thread!!!\n"); + } + #ifdef CONFIG_X86_MCE_AMD if ((adev->asic_type == CHIP_ALDEBARAN) && (adev->gmc.xgmi.connected_to_cpu)) @@ -2618,6 +2796,11 @@ static int amdgpu_ras_recovery_fini(struct amdgpu_device *adev) if (!data) return 0; + if (con->page_retirement_thread) + kthread_stop(con->page_retirement_thread); + + atomic_set(&con->page_retirement_req_cnt, 0); + cancel_work_sync(&con->recovery_work); mutex_lock(&con->recovery_lock); @@ -2679,6 +2862,87 @@ static void amdgpu_ras_get_quirks(struct amdgpu_device *adev) adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__GFX); } +/* Query ras capablity via atomfirmware interface */ +static void amdgpu_ras_query_ras_capablity_from_vbios(struct amdgpu_device *adev) +{ + /* mem_ecc cap */ + if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { + dev_info(adev->dev, "MEM ECC is active.\n"); + adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else { + dev_info(adev->dev, "MEM ECC is not presented.\n"); + } + + /* sram_ecc cap */ + if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { + dev_info(adev->dev, "SRAM ECC is active.\n"); + if (!amdgpu_sriov_vf(adev)) + adev->ras_hw_enabled |= ~(1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + else + adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__PCIE_BIF | + 1 << AMDGPU_RAS_BLOCK__SDMA | + 1 << AMDGPU_RAS_BLOCK__GFX); + + /* + * VCN/JPEG RAS can be supported on both bare metal and + * SRIOV environment + */ + if (amdgpu_ip_version(adev, VCN_HWIP, 0) == IP_VERSION(2, 6, 0) || + amdgpu_ip_version(adev, VCN_HWIP, 0) == IP_VERSION(4, 0, 0) || + amdgpu_ip_version(adev, VCN_HWIP, 0) == IP_VERSION(4, 0, 3)) + adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__VCN | + 1 << AMDGPU_RAS_BLOCK__JPEG); + else + adev->ras_hw_enabled &= ~(1 << AMDGPU_RAS_BLOCK__VCN | + 1 << AMDGPU_RAS_BLOCK__JPEG); + + /* + * XGMI RAS is not supported if xgmi num physical nodes + * is zero + */ + if (!adev->gmc.xgmi.num_physical_nodes) + adev->ras_hw_enabled &= ~(1 << AMDGPU_RAS_BLOCK__XGMI_WAFL); + } else { + dev_info(adev->dev, "SRAM ECC is not presented.\n"); + } +} + +/* Query poison mode from umc/df IP callbacks */ +static void amdgpu_ras_query_poison_mode(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + bool df_poison, umc_poison; + + /* poison setting is useless on SRIOV guest */ + if (amdgpu_sriov_vf(adev) || !con) + return; + + /* Init poison supported flag, the default value is false */ + if (adev->gmc.xgmi.connected_to_cpu || + adev->gmc.is_app_apu) { + /* enabled by default when GPU is connected to CPU */ + con->poison_supported = true; + } else if (adev->df.funcs && + adev->df.funcs->query_ras_poison_mode && + adev->umc.ras && + adev->umc.ras->query_ras_poison_mode) { + df_poison = + adev->df.funcs->query_ras_poison_mode(adev); + umc_poison = + adev->umc.ras->query_ras_poison_mode(adev); + + /* Only poison is set in both DF and UMC, we can support it */ + if (df_poison && umc_poison) + con->poison_supported = true; + else if (df_poison != umc_poison) + dev_warn(adev->dev, + "Poison setting is inconsistent in DF/UMC(%d:%d)!\n", + df_poison, umc_poison); + } +} + /* * check hardware's ras ability which will be saved in hw_supported. * if hardware does not support ras, we can skip some ras initializtion and @@ -2695,49 +2959,13 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev) if (!amdgpu_ras_asic_supported(adev)) return; - if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) { - if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { - dev_info(adev->dev, "MEM ECC is active.\n"); - adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__UMC | - 1 << AMDGPU_RAS_BLOCK__DF); - } else { - dev_info(adev->dev, "MEM ECC is not presented.\n"); - } - - if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { - dev_info(adev->dev, "SRAM ECC is active.\n"); - if (!amdgpu_sriov_vf(adev)) - adev->ras_hw_enabled |= ~(1 << AMDGPU_RAS_BLOCK__UMC | - 1 << AMDGPU_RAS_BLOCK__DF); - else - adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__PCIE_BIF | - 1 << AMDGPU_RAS_BLOCK__SDMA | - 1 << AMDGPU_RAS_BLOCK__GFX); - - /* VCN/JPEG RAS can be supported on both bare metal and - * SRIOV environment - */ - if (amdgpu_ip_version(adev, VCN_HWIP, 0) == - IP_VERSION(2, 6, 0) || - amdgpu_ip_version(adev, VCN_HWIP, 0) == - IP_VERSION(4, 0, 0) || - amdgpu_ip_version(adev, VCN_HWIP, 0) == - IP_VERSION(4, 0, 3)) - adev->ras_hw_enabled |= (1 << AMDGPU_RAS_BLOCK__VCN | - 1 << AMDGPU_RAS_BLOCK__JPEG); - else - adev->ras_hw_enabled &= ~(1 << AMDGPU_RAS_BLOCK__VCN | - 1 << AMDGPU_RAS_BLOCK__JPEG); + /* query ras capability from psp */ + if (amdgpu_psp_get_ras_capability(&adev->psp)) + goto init_ras_enabled_flag; - /* - * XGMI RAS is not supported if xgmi num physical nodes - * is zero - */ - if (!adev->gmc.xgmi.num_physical_nodes) - adev->ras_hw_enabled &= ~(1 << AMDGPU_RAS_BLOCK__XGMI_WAFL); - } else { - dev_info(adev->dev, "SRAM ECC is not presented.\n"); - } + /* query ras capablity from bios */ + if (!adev->gmc.xgmi.connected_to_cpu && !adev->gmc.is_app_apu) { + amdgpu_ras_query_ras_capablity_from_vbios(adev); } else { /* driver only manages a few IP blocks RAS feature * when GPU is connected cpu through XGMI */ @@ -2746,13 +2974,21 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev) 1 << AMDGPU_RAS_BLOCK__MMHUB); } + /* apply asic specific settings (vega20 only for now) */ amdgpu_ras_get_quirks(adev); + /* query poison mode from umc/df ip callback */ + amdgpu_ras_query_poison_mode(adev); + +init_ras_enabled_flag: /* 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; + + /* aca is disabled by default */ + adev->aca.is_enabled = false; } static void amdgpu_ras_counte_dw(struct work_struct *work) @@ -2780,39 +3016,6 @@ Out: pm_runtime_put_autosuspend(dev->dev); } -static void amdgpu_ras_query_poison_mode(struct amdgpu_device *adev) -{ - struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - bool df_poison, umc_poison; - - /* poison setting is useless on SRIOV guest */ - if (amdgpu_sriov_vf(adev) || !con) - return; - - /* Init poison supported flag, the default value is false */ - if (adev->gmc.xgmi.connected_to_cpu || - adev->gmc.is_app_apu) { - /* enabled by default when GPU is connected to CPU */ - con->poison_supported = true; - } else if (adev->df.funcs && - adev->df.funcs->query_ras_poison_mode && - adev->umc.ras && - adev->umc.ras->query_ras_poison_mode) { - df_poison = - adev->df.funcs->query_ras_poison_mode(adev); - umc_poison = - adev->umc.ras->query_ras_poison_mode(adev); - - /* Only poison is set in both DF and UMC, we can support it */ - if (df_poison && umc_poison) - con->poison_supported = true; - else if (df_poison != umc_poison) - dev_warn(adev->dev, - "Poison setting is inconsistent in DF/UMC(%d:%d)!\n", - df_poison, umc_poison); - } -} - static int amdgpu_get_ras_schema(struct amdgpu_device *adev) { return amdgpu_ras_is_poison_mode_supported(adev) ? AMDGPU_RAS_ERROR__POISON : 0 | @@ -2917,12 +3120,11 @@ int amdgpu_ras_init(struct amdgpu_device *adev) goto release_con; } - amdgpu_ras_query_poison_mode(adev); - /* Packed socket_id to ras feature mask bits[31:29] */ if (adev->smuio.funcs && adev->smuio.funcs->get_socket_id) - con->features |= ((adev->smuio.funcs->get_socket_id(adev)) << 29); + con->features |= ((adev->smuio.funcs->get_socket_id(adev)) << + AMDGPU_RAS_FEATURES_SOCKETID_SHIFT); /* Get RAS schema for particular SOC */ con->schema = amdgpu_get_ras_schema(adev); @@ -3128,7 +3330,7 @@ void amdgpu_ras_suspend(struct amdgpu_device *adev) amdgpu_ras_disable_all_features(adev, 0); /* Make sure all ras objects are disabled. */ - if (con->features) + if (AMDGPU_RAS_GET_FEATURES(con->features)) amdgpu_ras_disable_all_features(adev, 1); } @@ -3142,15 +3344,29 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev)) return 0; - amdgpu_ras_set_mca_debug_mode(adev, false); + if (amdgpu_aca_is_enabled(adev)) { + if (amdgpu_in_reset(adev)) + r = amdgpu_aca_reset(adev); + else + r = amdgpu_aca_init(adev); + if (r) + return r; + + amdgpu_ras_set_aca_debug_mode(adev, false); + } else { + amdgpu_ras_set_mca_debug_mode(adev, false); + } list_for_each_entry_safe(node, tmp, &adev->ras_list, node) { - if (!node->ras_obj) { + obj = node->ras_obj; + if (!obj) { dev_warn(adev->dev, "Warning: abnormal ras list node.\n"); continue; } - obj = node->ras_obj; + if (!amdgpu_ras_is_supported(adev, obj->ras_comm.block)) + continue; + if (obj->ras_late_init) { r = obj->ras_late_init(adev, &obj->ras_comm); if (r) { @@ -3175,7 +3391,7 @@ int amdgpu_ras_pre_fini(struct amdgpu_device *adev) /* Need disable ras on all IPs here before ip [hw/sw]fini */ - if (con->features) + if (AMDGPU_RAS_GET_FEATURES(con->features)) amdgpu_ras_disable_all_features(adev, 0); amdgpu_ras_recovery_fini(adev); return 0; @@ -3208,10 +3424,13 @@ int amdgpu_ras_fini(struct amdgpu_device *adev) amdgpu_ras_fs_fini(adev); amdgpu_ras_interrupt_remove_all(adev); - WARN(con->features, "Feature mask is not cleared"); + if (amdgpu_aca_is_enabled(adev)) + amdgpu_aca_fini(adev); - if (con->features) - amdgpu_ras_disable_all_features(adev, 1); + WARN(AMDGPU_RAS_GET_FEATURES(con->features), "Feature mask is not cleared"); + + if (AMDGPU_RAS_GET_FEATURES(con->features)) + amdgpu_ras_disable_all_features(adev, 0); cancel_delayed_work_sync(&con->ras_counte_delay_work); @@ -3425,22 +3644,41 @@ int amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool enable) if (con) { ret = amdgpu_mca_smu_set_debug_mode(adev, enable); if (!ret) - con->is_mca_debug_mode = enable; + con->is_aca_debug_mode = enable; + } + + return ret; +} + +int amdgpu_ras_set_aca_debug_mode(struct amdgpu_device *adev, bool enable) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + int ret = 0; + + if (con) { + if (amdgpu_aca_is_enabled(adev)) + ret = amdgpu_aca_smu_set_debug_mode(adev, enable); + else + ret = amdgpu_mca_smu_set_debug_mode(adev, enable); + if (!ret) + con->is_aca_debug_mode = enable; } return ret; } -bool amdgpu_ras_get_mca_debug_mode(struct amdgpu_device *adev) +bool amdgpu_ras_get_aca_debug_mode(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + const struct aca_smu_funcs *smu_funcs = adev->aca.smu_funcs; const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs; if (!con) return false; - if (mca_funcs && mca_funcs->mca_set_debug_mode) - return con->is_mca_debug_mode; + if ((amdgpu_aca_is_enabled(adev) && smu_funcs && smu_funcs->set_debug_mode) || + (!amdgpu_aca_is_enabled(adev) && mca_funcs && mca_funcs->mca_set_debug_mode)) + return con->is_aca_debug_mode; else return true; } @@ -3450,15 +3688,16 @@ bool amdgpu_ras_get_error_query_mode(struct amdgpu_device *adev, { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); const struct amdgpu_mca_smu_funcs *mca_funcs = adev->mca.mca_funcs; + const struct aca_smu_funcs *smu_funcs = adev->aca.smu_funcs; if (!con) { *error_query_mode = AMDGPU_RAS_INVALID_ERROR_QUERY; return false; } - if (mca_funcs && mca_funcs->mca_set_debug_mode) + if ((smu_funcs && smu_funcs->set_debug_mode) || (mca_funcs && mca_funcs->mca_set_debug_mode)) *error_query_mode = - (con->is_mca_debug_mode) ? AMDGPU_RAS_DIRECT_ERROR_QUERY : AMDGPU_RAS_FIRMWARE_ERROR_QUERY; + (con->is_aca_debug_mode) ? AMDGPU_RAS_DIRECT_ERROR_QUERY : AMDGPU_RAS_FIRMWARE_ERROR_QUERY; else *error_query_mode = AMDGPU_RAS_DIRECT_ERROR_QUERY; @@ -3699,8 +3938,7 @@ static int ras_err_info_cmp(void *priv, const struct list_head *a, const struct } static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_data, - struct amdgpu_smuio_mcm_config_info *mcm_info, - struct ras_err_addr *err_addr) + struct amdgpu_smuio_mcm_config_info *mcm_info) { struct ras_err_node *err_node; @@ -3712,10 +3950,9 @@ static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_d if (!err_node) return NULL; - memcpy(&err_node->err_info.mcm_info, mcm_info, sizeof(*mcm_info)); + INIT_LIST_HEAD(&err_node->err_info.err_addr_list); - if (err_addr) - memcpy(&err_node->err_info.err_addr, err_addr, sizeof(*err_addr)); + memcpy(&err_node->err_info.mcm_info, mcm_info, sizeof(*mcm_info)); err_data->err_list_count++; list_add_tail(&err_node->node, &err_data->err_node_list); @@ -3724,6 +3961,29 @@ static struct ras_err_info *amdgpu_ras_error_get_info(struct ras_err_data *err_d return &err_node->err_info; } +void amdgpu_ras_add_mca_err_addr(struct ras_err_info *err_info, struct ras_err_addr *err_addr) +{ + struct ras_err_addr *mca_err_addr; + + mca_err_addr = kzalloc(sizeof(*mca_err_addr), GFP_KERNEL); + if (!mca_err_addr) + return; + + INIT_LIST_HEAD(&mca_err_addr->node); + + mca_err_addr->err_status = err_addr->err_status; + mca_err_addr->err_ipid = err_addr->err_ipid; + mca_err_addr->err_addr = err_addr->err_addr; + + list_add_tail(&mca_err_addr->node, &err_info->err_addr_list); +} + +void amdgpu_ras_del_mca_err_addr(struct ras_err_info *err_info, struct ras_err_addr *mca_err_addr) +{ + list_del(&mca_err_addr->node); + kfree(mca_err_addr); +} + int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, struct amdgpu_smuio_mcm_config_info *mcm_info, struct ras_err_addr *err_addr, u64 count) @@ -3736,10 +3996,13 @@ int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, if (!count) return 0; - err_info = amdgpu_ras_error_get_info(err_data, mcm_info, err_addr); + err_info = amdgpu_ras_error_get_info(err_data, mcm_info); if (!err_info) return -EINVAL; + if (err_addr && err_addr->err_status) + amdgpu_ras_add_mca_err_addr(err_info, err_addr); + err_info->ue_count += count; err_data->ue_count += count; @@ -3758,7 +4021,7 @@ int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data, if (!count) return 0; - err_info = amdgpu_ras_error_get_info(err_data, mcm_info, err_addr); + err_info = amdgpu_ras_error_get_info(err_data, mcm_info); if (!err_info) return -EINVAL; @@ -3767,3 +4030,135 @@ int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data, return 0; } + +int amdgpu_ras_error_statistic_de_count(struct ras_err_data *err_data, + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr, u64 count) +{ + struct ras_err_info *err_info; + + if (!err_data || !mcm_info) + return -EINVAL; + + if (!count) + return 0; + + err_info = amdgpu_ras_error_get_info(err_data, mcm_info); + if (!err_info) + return -EINVAL; + + if (err_addr && err_addr->err_status) + amdgpu_ras_add_mca_err_addr(err_info, err_addr); + + err_info->de_count += count; + err_data->de_count += count; + + return 0; +} + +#define mmMP0_SMN_C2PMSG_92 0x1609C +#define mmMP0_SMN_C2PMSG_126 0x160BE +static void amdgpu_ras_boot_time_error_reporting(struct amdgpu_device *adev, + u32 instance, u32 boot_error) +{ + u32 socket_id, aid_id, hbm_id; + u32 reg_data; + u64 reg_addr; + + socket_id = AMDGPU_RAS_GPU_ERR_SOCKET_ID(boot_error); + aid_id = AMDGPU_RAS_GPU_ERR_AID_ID(boot_error); + hbm_id = AMDGPU_RAS_GPU_ERR_HBM_ID(boot_error); + + /* The pattern for smn addressing in other SOC could be different from + * the one for aqua_vanjaram. We should revisit the code if the pattern + * is changed. In such case, replace the aqua_vanjaram implementation + * with more common helper */ + reg_addr = (mmMP0_SMN_C2PMSG_92 << 2) + + aqua_vanjaram_encode_ext_smn_addressing(instance); + + reg_data = amdgpu_device_indirect_rreg_ext(adev, reg_addr); + dev_err(adev->dev, "socket: %d, aid: %d, firmware boot failed, fw status is 0x%x\n", + socket_id, aid_id, reg_data); + + if (AMDGPU_RAS_GPU_ERR_MEM_TRAINING(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, hbm: %d, memory training failed\n", + socket_id, aid_id, hbm_id); + + if (AMDGPU_RAS_GPU_ERR_FW_LOAD(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, firmware load failed at boot time\n", + socket_id, aid_id); + + if (AMDGPU_RAS_GPU_ERR_WAFL_LINK_TRAINING(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, wafl link training failed\n", + socket_id, aid_id); + + if (AMDGPU_RAS_GPU_ERR_XGMI_LINK_TRAINING(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, xgmi link training failed\n", + socket_id, aid_id); + + if (AMDGPU_RAS_GPU_ERR_USR_CP_LINK_TRAINING(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, usr cp link training failed\n", + socket_id, aid_id); + + if (AMDGPU_RAS_GPU_ERR_USR_DP_LINK_TRAINING(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, usr dp link training failed\n", + socket_id, aid_id); + + if (AMDGPU_RAS_GPU_ERR_HBM_MEM_TEST(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, hbm: %d, hbm memory test failed\n", + socket_id, aid_id, hbm_id); + + if (AMDGPU_RAS_GPU_ERR_HBM_BIST_TEST(boot_error)) + dev_info(adev->dev, "socket: %d, aid: %d, hbm: %d, hbm bist test failed\n", + socket_id, aid_id, hbm_id); +} + +static int amdgpu_ras_wait_for_boot_complete(struct amdgpu_device *adev, + u32 instance, u32 *boot_error) +{ + u32 reg_addr; + u32 reg_data; + int retry_loop; + + reg_addr = (mmMP0_SMN_C2PMSG_92 << 2) + + aqua_vanjaram_encode_ext_smn_addressing(instance); + + for (retry_loop = 0; retry_loop < AMDGPU_RAS_BOOT_STATUS_POLLING_LIMIT; retry_loop++) { + reg_data = amdgpu_device_indirect_rreg_ext(adev, reg_addr); + if ((reg_data & AMDGPU_RAS_BOOT_STATUS_MASK) == AMDGPU_RAS_BOOT_STEADY_STATUS) { + *boot_error = AMDGPU_RAS_BOOT_SUCEESS; + return 0; + } + msleep(1); + } + + /* The pattern for smn addressing in other SOC could be different from + * the one for aqua_vanjaram. We should revisit the code if the pattern + * is changed. In such case, replace the aqua_vanjaram implementation + * with more common helper */ + reg_addr = (mmMP0_SMN_C2PMSG_126 << 2) + + aqua_vanjaram_encode_ext_smn_addressing(instance); + + for (retry_loop = 0; retry_loop < AMDGPU_RAS_BOOT_STATUS_POLLING_LIMIT; retry_loop++) { + reg_data = amdgpu_device_indirect_rreg_ext(adev, reg_addr); + if (AMDGPU_RAS_GPU_ERR_BOOT_STATUS(reg_data)) { + *boot_error = reg_data; + return 0; + } + msleep(1); + } + + *boot_error = reg_data; + return -ETIME; +} + +void amdgpu_ras_query_boot_status(struct amdgpu_device *adev, u32 num_instances) +{ + u32 boot_error = 0; + u32 i; + + for (i = 0; i < num_instances; i++) { + if (amdgpu_ras_wait_for_boot_complete(adev, i, &boot_error)) + amdgpu_ras_boot_time_error_reporting(adev, i, boot_error); + } +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 76fb85628716..d10e5bb0e52f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -29,9 +29,28 @@ #include "ta_ras_if.h" #include "amdgpu_ras_eeprom.h" #include "amdgpu_smuio.h" +#include "amdgpu_aca.h" struct amdgpu_iv_entry; +#define AMDGPU_RAS_GPU_ERR_MEM_TRAINING(x) AMDGPU_GET_REG_FIELD(x, 0, 0) +#define AMDGPU_RAS_GPU_ERR_FW_LOAD(x) AMDGPU_GET_REG_FIELD(x, 1, 1) +#define AMDGPU_RAS_GPU_ERR_WAFL_LINK_TRAINING(x) AMDGPU_GET_REG_FIELD(x, 2, 2) +#define AMDGPU_RAS_GPU_ERR_XGMI_LINK_TRAINING(x) AMDGPU_GET_REG_FIELD(x, 3, 3) +#define AMDGPU_RAS_GPU_ERR_USR_CP_LINK_TRAINING(x) AMDGPU_GET_REG_FIELD(x, 4, 4) +#define AMDGPU_RAS_GPU_ERR_USR_DP_LINK_TRAINING(x) AMDGPU_GET_REG_FIELD(x, 5, 5) +#define AMDGPU_RAS_GPU_ERR_HBM_MEM_TEST(x) AMDGPU_GET_REG_FIELD(x, 6, 6) +#define AMDGPU_RAS_GPU_ERR_HBM_BIST_TEST(x) AMDGPU_GET_REG_FIELD(x, 7, 7) +#define AMDGPU_RAS_GPU_ERR_SOCKET_ID(x) AMDGPU_GET_REG_FIELD(x, 10, 8) +#define AMDGPU_RAS_GPU_ERR_AID_ID(x) AMDGPU_GET_REG_FIELD(x, 12, 11) +#define AMDGPU_RAS_GPU_ERR_HBM_ID(x) AMDGPU_GET_REG_FIELD(x, 13, 13) +#define AMDGPU_RAS_GPU_ERR_BOOT_STATUS(x) AMDGPU_GET_REG_FIELD(x, 31, 31) + +#define AMDGPU_RAS_BOOT_STATUS_POLLING_LIMIT 1000 +#define AMDGPU_RAS_BOOT_STEADY_STATUS 0xBA +#define AMDGPU_RAS_BOOT_STATUS_MASK 0xFF +#define AMDGPU_RAS_BOOT_SUCEESS 0x80000000 + #define AMDGPU_RAS_FLAG_INIT_BY_VBIOS (0x1 << 0) /* position of instance value in sub_block_index of * ta_ras_trigger_error_input, the sub block uses lower 12 bits @@ -39,6 +58,12 @@ struct amdgpu_iv_entry; #define AMDGPU_RAS_INST_MASK 0xfffff000 #define AMDGPU_RAS_INST_SHIFT 0xc +#define AMDGPU_RAS_FEATURES_SOCKETID_SHIFT 29 +#define AMDGPU_RAS_FEATURES_SOCKETID_MASK 0xe0000000 + +/* The high three bits indicates socketid */ +#define AMDGPU_RAS_GET_FEATURES(val) ((val) & ~AMDGPU_RAS_FEATURES_SOCKETID_MASK) + enum amdgpu_ras_block { AMDGPU_RAS_BLOCK__UMC = 0, AMDGPU_RAS_BLOCK__SDMA, @@ -57,6 +82,8 @@ enum amdgpu_ras_block { AMDGPU_RAS_BLOCK__MCA, AMDGPU_RAS_BLOCK__VCN, AMDGPU_RAS_BLOCK__JPEG, + AMDGPU_RAS_BLOCK__IH, + AMDGPU_RAS_BLOCK__MPIO, AMDGPU_RAS_BLOCK__LAST }; @@ -441,10 +468,15 @@ struct amdgpu_ras { /* Indicates smu whether need update bad channel info */ bool update_channel_flag; /* Record status of smu mca debug mode */ - bool is_mca_debug_mode; + bool is_aca_debug_mode; /* Record special requirements of gpu reset caller */ uint32_t gpu_reset_flags; + + struct task_struct *page_retirement_thread; + wait_queue_head_t page_retirement_wq; + struct mutex page_retirement_lock; + atomic_t page_retirement_req_cnt; }; struct ras_fs_data { @@ -453,6 +485,7 @@ struct ras_fs_data { }; struct ras_err_addr { + struct list_head node; uint64_t err_status; uint64_t err_ipid; uint64_t err_addr; @@ -462,7 +495,8 @@ struct ras_err_info { struct amdgpu_smuio_mcm_config_info mcm_info; u64 ce_count; u64 ue_count; - struct ras_err_addr err_addr; + u64 de_count; + struct list_head err_addr_list; }; struct ras_err_node { @@ -473,6 +507,7 @@ struct ras_err_node { struct ras_err_data { unsigned long ue_count; unsigned long ce_count; + unsigned long de_count; unsigned long err_addr_cnt; struct eeprom_table_record *err_addr; u32 err_list_count; @@ -529,6 +564,8 @@ struct ras_manager { struct ras_ih_data ih_data; struct ras_err_data err_data; + + struct aca_handle aca_handle; }; struct ras_badpage { @@ -548,6 +585,7 @@ struct ras_query_if { struct ras_common_if head; unsigned long ue_count; unsigned long ce_count; + unsigned long de_count; }; struct ras_inject_if { @@ -781,7 +819,8 @@ struct amdgpu_ras* amdgpu_ras_get_context(struct amdgpu_device *adev); int amdgpu_ras_set_context(struct amdgpu_device *adev, struct amdgpu_ras *ras_con); int amdgpu_ras_set_mca_debug_mode(struct amdgpu_device *adev, bool enable); -bool amdgpu_ras_get_mca_debug_mode(struct amdgpu_device *adev); +int amdgpu_ras_set_aca_debug_mode(struct amdgpu_device *adev, bool enable); +bool amdgpu_ras_get_aca_debug_mode(struct amdgpu_device *adev); bool amdgpu_ras_get_error_query_mode(struct amdgpu_device *adev, unsigned int *mode); @@ -818,5 +857,20 @@ int amdgpu_ras_error_statistic_ce_count(struct ras_err_data *err_data, int amdgpu_ras_error_statistic_ue_count(struct ras_err_data *err_data, struct amdgpu_smuio_mcm_config_info *mcm_info, struct ras_err_addr *err_addr, u64 count); +int amdgpu_ras_error_statistic_de_count(struct ras_err_data *err_data, + struct amdgpu_smuio_mcm_config_info *mcm_info, + struct ras_err_addr *err_addr, u64 count); +void amdgpu_ras_query_boot_status(struct amdgpu_device *adev, u32 num_instances); +int amdgpu_ras_bind_aca(struct amdgpu_device *adev, enum amdgpu_ras_block blk, + const struct aca_info *aca_info, void *data); +int amdgpu_ras_unbind_aca(struct amdgpu_device *adev, enum amdgpu_ras_block blk); + +ssize_t amdgpu_ras_aca_sysfs_read(struct device *dev, struct device_attribute *attr, + struct aca_handle *handle, char *buf, void *data); + +void amdgpu_ras_add_mca_err_addr(struct ras_err_info *err_info, + struct ras_err_addr *err_addr); +void amdgpu_ras_del_mca_err_addr(struct ras_err_info *err_info, + struct ras_err_addr *mca_err_addr); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c index 2c3675d91614..db5791e1a7ce 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c @@ -241,7 +241,7 @@ void amdgpu_gfx_rlc_setup_cp_table(struct amdgpu_device *adev) table_size = le32_to_cpu(hdr->jt_size); } - for (i = 0; i < table_size; i ++) { + for (i = 0; i < table_size; i++) { dst_ptr[bo_offset + i] = cpu_to_le32(le32_to_cpu(fw_data[table_offset + i])); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h index b591d33af264..5a17e0ff2ab8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h @@ -169,7 +169,7 @@ struct amdgpu_rlc_funcs { void (*stop)(struct amdgpu_device *adev); void (*reset)(struct amdgpu_device *adev); void (*start)(struct amdgpu_device *adev); - void (*update_spm_vmid)(struct amdgpu_device *adev, unsigned vmid); + void (*update_spm_vmid)(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned vmid); bool (*is_rlcg_access_range)(struct amdgpu_device *adev, uint32_t reg); }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c index 7a6a67275404..e9081a98cf81 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c @@ -36,13 +36,28 @@ */ /** + * amdgpu_seq64_get_va_base - Get the seq64 va base address + * + * @adev: amdgpu_device pointer + * + * Returns: + * va base address on success + */ +static inline u64 amdgpu_seq64_get_va_base(struct amdgpu_device *adev) +{ + u64 addr = adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT; + + addr -= AMDGPU_VA_RESERVED_TOP; + + return addr; +} + +/** * amdgpu_seq64_map - Map the seq64 memory to VM * * @adev: amdgpu_device pointer * @vm: vm pointer * @bo_va: bo_va pointer - * @seq64_addr: seq64 vaddr start address - * @size: seq64 pool size * * Map the seq64 memory to the given VM. * @@ -50,11 +65,11 @@ * 0 on success or a negative error code on failure */ int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm, - struct amdgpu_bo_va **bo_va, u64 seq64_addr, - uint32_t size) + struct amdgpu_bo_va **bo_va) { struct amdgpu_bo *bo; struct drm_exec exec; + u64 seq64_addr; int r; bo = adev->seq64.sbo; @@ -77,9 +92,9 @@ int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm, goto error; } - r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0, size, - AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE | - AMDGPU_PTE_EXECUTABLE); + seq64_addr = amdgpu_seq64_get_va_base(adev); + r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0, AMDGPU_VA_RESERVED_SEQ64_SIZE, + AMDGPU_PTE_READABLE); if (r) { DRM_ERROR("failed to do bo_map on userq sem, err=%d\n", r); amdgpu_vm_bo_del(adev, *bo_va); @@ -144,31 +159,25 @@ error: * amdgpu_seq64_alloc - Allocate a 64 bit memory * * @adev: amdgpu_device pointer - * @gpu_addr: allocated gpu VA start address - * @cpu_addr: allocated cpu VA start address + * @va: VA to access the seq in process address space + * @cpu_addr: CPU address to access the seq * * Alloc a 64 bit memory from seq64 pool. * * Returns: * 0 on success or a negative error code on failure */ -int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *gpu_addr, - u64 **cpu_addr) +int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *va, u64 **cpu_addr) { unsigned long bit_pos; - u32 offset; bit_pos = find_first_zero_bit(adev->seq64.used, adev->seq64.num_sem); + if (bit_pos >= adev->seq64.num_sem) + return -ENOSPC; - if (bit_pos < adev->seq64.num_sem) { - __set_bit(bit_pos, adev->seq64.used); - offset = bit_pos << 6; /* convert to qw offset */ - } else { - return -EINVAL; - } - - *gpu_addr = offset + AMDGPU_SEQ64_VADDR_START; - *cpu_addr = offset + adev->seq64.cpu_base_addr; + __set_bit(bit_pos, adev->seq64.used); + *va = bit_pos * sizeof(u64) + amdgpu_seq64_get_va_base(adev); + *cpu_addr = bit_pos + adev->seq64.cpu_base_addr; return 0; } @@ -177,20 +186,17 @@ int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *gpu_addr, * amdgpu_seq64_free - Free the given 64 bit memory * * @adev: amdgpu_device pointer - * @gpu_addr: gpu start address to be freed + * @va: gpu start address to be freed * * Free the given 64 bit memory from seq64 pool. - * */ -void amdgpu_seq64_free(struct amdgpu_device *adev, u64 gpu_addr) +void amdgpu_seq64_free(struct amdgpu_device *adev, u64 va) { - u32 offset; - - offset = gpu_addr - AMDGPU_SEQ64_VADDR_START; + unsigned long bit_pos; - offset >>= 6; - if (offset < adev->seq64.num_sem) - __clear_bit(offset, adev->seq64.used); + bit_pos = (va - amdgpu_seq64_get_va_base(adev)) / sizeof(u64); + if (bit_pos < adev->seq64.num_sem) + __clear_bit(bit_pos, adev->seq64.used); } /** @@ -229,7 +235,7 @@ int amdgpu_seq64_init(struct amdgpu_device *adev) * AMDGPU_MAX_SEQ64_SLOTS * sizeof(u64) * 8 = AMDGPU_MAX_SEQ64_SLOTS * 64bit slots */ - r = amdgpu_bo_create_kernel(adev, AMDGPU_SEQ64_SIZE, + r = amdgpu_bo_create_kernel(adev, AMDGPU_VA_RESERVED_SEQ64_SIZE, PAGE_SIZE, AMDGPU_GEM_DOMAIN_GTT, &adev->seq64.sbo, NULL, (void **)&adev->seq64.cpu_base_addr); @@ -238,7 +244,7 @@ int amdgpu_seq64_init(struct amdgpu_device *adev) return r; } - memset(adev->seq64.cpu_base_addr, 0, AMDGPU_SEQ64_SIZE); + memset(adev->seq64.cpu_base_addr, 0, AMDGPU_VA_RESERVED_SEQ64_SIZE); adev->seq64.num_sem = AMDGPU_MAX_SEQ64_SLOTS; memset(&adev->seq64.used, 0, sizeof(adev->seq64.used)); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h index 2196e72be508..4203b2ab318d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.h @@ -25,10 +25,9 @@ #ifndef __AMDGPU_SEQ64_H__ #define __AMDGPU_SEQ64_H__ -#define AMDGPU_SEQ64_SIZE (2ULL << 20) -#define AMDGPU_MAX_SEQ64_SLOTS (AMDGPU_SEQ64_SIZE / (sizeof(u64) * 8)) -#define AMDGPU_SEQ64_VADDR_OFFSET 0x50000 -#define AMDGPU_SEQ64_VADDR_START (AMDGPU_VA_RESERVED_SIZE + AMDGPU_SEQ64_VADDR_OFFSET) +#include "amdgpu_vm.h" + +#define AMDGPU_MAX_SEQ64_SLOTS (AMDGPU_VA_RESERVED_SEQ64_SIZE / sizeof(u64)) struct amdgpu_seq64 { struct amdgpu_bo *sbo; @@ -42,7 +41,7 @@ int amdgpu_seq64_init(struct amdgpu_device *adev); int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *gpu_addr, u64 **cpu_addr); void amdgpu_seq64_free(struct amdgpu_device *adev, u64 gpu_addr); int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm, - struct amdgpu_bo_va **bo_va, u64 seq64_addr, uint32_t size); + struct amdgpu_bo_va **bo_va); void amdgpu_seq64_unmap(struct amdgpu_device *adev, struct amdgpu_fpriv *fpriv); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index d65e21914d8c..20436f81856a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -23,6 +23,7 @@ #include "amdgpu.h" #include "umc_v6_7.h" +#define MAX_UMC_POISON_POLLING_TIME_SYNC 20 //ms static int amdgpu_umc_convert_error_address(struct amdgpu_device *adev, struct ras_err_data *err_data, uint64_t err_addr, @@ -85,18 +86,21 @@ out_fini_err_data: return ret; } -static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, - void *ras_error_status, - struct amdgpu_iv_entry *entry, - bool reset) +static void amdgpu_umc_handle_bad_pages(struct amdgpu_device *adev, + void *ras_error_status) { struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + unsigned int error_query_mode; int ret = 0; + unsigned long err_count; - kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); + amdgpu_ras_get_error_query_mode(adev, &error_query_mode); + + mutex_lock(&con->page_retirement_lock); ret = amdgpu_dpm_get_ecc_info(adev, (void *)&(con->umc_ecc)); - if (ret == -EOPNOTSUPP) { + if (ret == -EOPNOTSUPP && + error_query_mode == AMDGPU_RAS_DIRECT_ERROR_QUERY) { if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops && adev->umc.ras->ras_block.hw_ops->query_ras_error_count) adev->umc.ras->ras_block.hw_ops->query_ras_error_count(adev, ras_error_status); @@ -120,7 +124,8 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, */ adev->umc.ras->ras_block.hw_ops->query_ras_error_address(adev, ras_error_status); } - } else if (!ret) { + } else if (error_query_mode == AMDGPU_RAS_FIRMWARE_ERROR_QUERY || + (!ret && error_query_mode == AMDGPU_RAS_DIRECT_ERROR_QUERY)) { if (adev->umc.ras && adev->umc.ras->ecc_info_query_ras_error_count) adev->umc.ras->ecc_info_query_ras_error_count(adev, ras_error_status); @@ -147,16 +152,13 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, } /* only uncorrectable error needs gpu reset */ - if (err_data->ue_count) { - dev_info(adev->dev, "%ld uncorrectable hardware errors " - "detected in UMC block\n", - err_data->ue_count); - + if (err_data->ue_count || err_data->de_count) { + err_count = err_data->ue_count + err_data->de_count; if ((amdgpu_bad_page_threshold != 0) && err_data->err_addr_cnt) { amdgpu_ras_add_bad_pages(adev, err_data->err_addr, err_data->err_addr_cnt); - amdgpu_ras_save_bad_pages(adev, &(err_data->ue_count)); + amdgpu_ras_save_bad_pages(adev, &err_count); amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); @@ -165,20 +167,87 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, con->update_channel_flag = false; } } - - if (reset) { - /* use mode-2 reset for poison consumption */ - if (!entry) - con->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET; - amdgpu_ras_reset_gpu(adev); - } } kfree(err_data->err_addr); + + mutex_unlock(&con->page_retirement_lock); +} + +static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev, + void *ras_error_status, + struct amdgpu_iv_entry *entry, + bool reset) +{ + struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + + kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); + amdgpu_umc_handle_bad_pages(adev, ras_error_status); + + if (err_data->ue_count && reset) { + /* use mode-2 reset for poison consumption */ + if (!entry) + con->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET; + amdgpu_ras_reset_gpu(adev); + } + return AMDGPU_RAS_SUCCESS; } -int amdgpu_umc_poison_handler(struct amdgpu_device *adev, bool reset) +int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev, + bool reset, uint32_t timeout_ms) +{ + struct ras_err_data err_data; + struct ras_common_if head = { + .block = AMDGPU_RAS_BLOCK__UMC, + }; + struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); + uint32_t timeout = timeout_ms; + + memset(&err_data, 0, sizeof(err_data)); + amdgpu_ras_error_data_init(&err_data); + + do { + + amdgpu_umc_handle_bad_pages(adev, &err_data); + + if (timeout && !err_data.de_count) { + msleep(1); + timeout--; + } + + } while (timeout && !err_data.de_count); + + if (!timeout) + dev_warn(adev->dev, "Can't find bad pages\n"); + + if (err_data.de_count) + dev_info(adev->dev, "%ld new deferred hardware errors detected\n", err_data.de_count); + + if (obj) { + obj->err_data.ue_count += err_data.ue_count; + obj->err_data.ce_count += err_data.ce_count; + obj->err_data.de_count += err_data.de_count; + } + + amdgpu_ras_error_data_fini(&err_data); + + kgd2kfd_set_sram_ecc_flag(adev->kfd.dev); + + if (reset) { + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + + /* use mode-2 reset for poison consumption */ + con->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE2_RESET; + amdgpu_ras_reset_gpu(adev); + } + + return 0; +} + +int amdgpu_umc_poison_handler(struct amdgpu_device *adev, + enum amdgpu_ras_block block, bool reset) { int ret = AMDGPU_RAS_SUCCESS; @@ -195,27 +264,41 @@ int amdgpu_umc_poison_handler(struct amdgpu_device *adev, bool reset) } if (!amdgpu_sriov_vf(adev)) { - struct ras_err_data err_data; - struct ras_common_if head = { - .block = AMDGPU_RAS_BLOCK__UMC, - }; - struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); + if (amdgpu_ip_version(adev, UMC_HWIP, 0) < IP_VERSION(12, 0, 0)) { + struct ras_err_data err_data; + struct ras_common_if head = { + .block = AMDGPU_RAS_BLOCK__UMC, + }; + struct ras_manager *obj = amdgpu_ras_find_obj(adev, &head); + + ret = amdgpu_ras_error_data_init(&err_data); + if (ret) + return ret; - ret = amdgpu_ras_error_data_init(&err_data); - if (ret) - return ret; + ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset); - ret = amdgpu_umc_do_page_retirement(adev, &err_data, NULL, reset); + if (ret == AMDGPU_RAS_SUCCESS && obj) { + obj->err_data.ue_count += err_data.ue_count; + obj->err_data.ce_count += err_data.ce_count; + obj->err_data.de_count += err_data.de_count; + } - if (ret == AMDGPU_RAS_SUCCESS && obj) { - obj->err_data.ue_count += err_data.ue_count; - obj->err_data.ce_count += err_data.ce_count; - } + amdgpu_ras_error_data_fini(&err_data); + } else { + if (reset) { + amdgpu_umc_bad_page_polling_timeout(adev, + reset, MAX_UMC_POISON_POLLING_TIME_SYNC); + } else { + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - amdgpu_ras_error_data_fini(&err_data); + atomic_inc(&con->page_retirement_req_cnt); + + wake_up(&con->page_retirement_wq); + } + } } else { if (adev->virt.ops && adev->virt.ops->ras_poison_handler) - adev->virt.ops->ras_poison_handler(adev); + adev->virt.ops->ras_poison_handler(adev, block); else dev_warn(adev->dev, "No ras_poison_handler interface in SRIOV!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h index 417a6726c71b..26d2ae498daf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.h @@ -21,7 +21,7 @@ #ifndef __AMDGPU_UMC_H__ #define __AMDGPU_UMC_H__ #include "amdgpu_ras.h" - +#include "amdgpu_mca.h" /* * (addr / 256) * 4096, the higher 26 bits in ErrorAddr * is the index of 4KB block @@ -64,6 +64,8 @@ struct amdgpu_umc_ras { void *ras_error_status); void (*ecc_info_query_ras_error_address)(struct amdgpu_device *adev, void *ras_error_status); + bool (*check_ecc_err_status)(struct amdgpu_device *adev, + enum amdgpu_mca_error_type type, void *ras_error_status); /* support different eeprom table version for different asic */ void (*set_eeprom_table_version)(struct amdgpu_ras_eeprom_table_header *hdr); }; @@ -100,7 +102,8 @@ struct amdgpu_umc { int amdgpu_umc_ras_sw_init(struct amdgpu_device *adev); int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block); -int amdgpu_umc_poison_handler(struct amdgpu_device *adev, bool reset); +int amdgpu_umc_poison_handler(struct amdgpu_device *adev, + enum amdgpu_ras_block block, bool reset); int amdgpu_umc_process_ecc_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry); @@ -118,4 +121,7 @@ int amdgpu_umc_page_retirement_mca(struct amdgpu_device *adev, int amdgpu_umc_loop_channels(struct amdgpu_device *adev, umc_func func, void *data); + +int amdgpu_umc_bad_page_polling_timeout(struct amdgpu_device *adev, + bool reset, uint32_t timeout_ms); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h index 107f9bb0e24f..5b27fc41ffbf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umr.h @@ -69,12 +69,12 @@ struct amdgpu_debugfs_gprwave_data { }; enum AMDGPU_DEBUGFS_REGS2_CMDS { - AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE=0, + AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE = 0, AMDGPU_DEBUGFS_REGS2_CMD_SET_STATE_V2, }; enum AMDGPU_DEBUGFS_GPRWAVE_CMDS { - AMDGPU_DEBUGFS_GPRWAVE_CMD_SET_STATE=0, + AMDGPU_DEBUGFS_GPRWAVE_CMD_SET_STATE = 0, }; //reg2 interface diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c index bfbf59326ee1..ab820cf52668 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c @@ -358,7 +358,7 @@ static int setup_umsch_mm_test(struct amdgpu_device *adev, memset(test->ring_data_cpu_addr, 0, sizeof(struct umsch_mm_test_ring_data)); - test->ring_data_gpu_addr = AMDGPU_VA_RESERVED_SIZE; + test->ring_data_gpu_addr = AMDGPU_VA_RESERVED_BOTTOM; r = map_ring_data(adev, test->vm, test->ring_data_obj, &test->bo_va, test->ring_data_gpu_addr, sizeof(struct umsch_mm_test_ring_data)); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index f4963330c772..f300d4a4457d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -1189,7 +1189,7 @@ int amdgpu_vcn_process_poison_irq(struct amdgpu_device *adev, amdgpu_ras_interrupt_dispatch(adev, &ih_data); } else { if (adev->virt.ops && adev->virt.ops->ras_poison_handler) - adev->virt.ops->ras_poison_handler(adev); + adev->virt.ops->ras_poison_handler(adev, ras_if->block); else dev_warn(adev->dev, "No ras_poison_handler interface in SRIOV for VCN!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index 0dcff2889e25..6ff7d3fb2008 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -71,59 +71,6 @@ void amdgpu_virt_init_setting(struct amdgpu_device *adev) amdgpu_num_kcq = 2; } -void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, - uint32_t reg0, uint32_t reg1, - uint32_t ref, uint32_t mask, - uint32_t xcc_inst) -{ - struct amdgpu_kiq *kiq = &adev->gfx.kiq[xcc_inst]; - struct amdgpu_ring *ring = &kiq->ring; - signed long r, cnt = 0; - unsigned long flags; - uint32_t seq; - - if (adev->mes.ring.sched.ready) { - amdgpu_mes_reg_write_reg_wait(adev, reg0, reg1, - ref, mask); - return; - } - - spin_lock_irqsave(&kiq->ring_lock, flags); - amdgpu_ring_alloc(ring, 32); - amdgpu_ring_emit_reg_write_reg_wait(ring, reg0, reg1, - ref, mask); - r = amdgpu_fence_emit_polling(ring, &seq, MAX_KIQ_REG_WAIT); - if (r) - goto failed_undo; - - amdgpu_ring_commit(ring); - spin_unlock_irqrestore(&kiq->ring_lock, flags); - - r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); - - /* don't wait anymore for IRQ context */ - if (r < 1 && in_interrupt()) - goto failed_kiq; - - might_sleep(); - while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) { - - msleep(MAX_KIQ_REG_BAILOUT_INTERVAL); - r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT); - } - - if (cnt > MAX_KIQ_REG_TRY) - goto failed_kiq; - - return; - -failed_undo: - amdgpu_ring_undo(ring); - spin_unlock_irqrestore(&kiq->ring_lock, flags); -failed_kiq: - dev_err(adev->dev, "failed to write reg %x wait reg %x\n", reg0, reg1); -} - /** * amdgpu_virt_request_full_gpu() - request full gpu access * @adev: amdgpu device. @@ -303,11 +250,11 @@ static int amdgpu_virt_init_ras_err_handler_data(struct amdgpu_device *adev) if (!*data) goto data_failure; - bps = kmalloc_array(align_space, sizeof((*data)->bps), GFP_KERNEL); + bps = kmalloc_array(align_space, sizeof(*(*data)->bps), GFP_KERNEL); if (!bps) goto bps_failure; - bps_bo = kmalloc_array(align_space, sizeof((*data)->bps_bo), GFP_KERNEL); + bps_bo = kmalloc_array(align_space, sizeof(*(*data)->bps_bo), GFP_KERNEL); if (!bps_bo) goto bps_bo_failure; @@ -340,8 +287,10 @@ static void amdgpu_virt_ras_release_bp(struct amdgpu_device *adev) for (i = data->last_reserved - 1; i >= 0; i--) { bo = data->bps_bo[i]; - amdgpu_bo_free_kernel(&bo, NULL, NULL); - data->bps_bo[i] = bo; + if (bo) { + amdgpu_bo_free_kernel(&bo, NULL, NULL); + data->bps_bo[i] = bo; + } data->last_reserved = i; } } @@ -381,6 +330,8 @@ static void amdgpu_virt_ras_reserve_bps(struct amdgpu_device *adev) { struct amdgpu_virt *virt = &adev->virt; struct amdgpu_virt_ras_err_handler_data *data = virt->virt_eh_data; + struct amdgpu_vram_mgr *mgr = &adev->mman.vram_mgr; + struct ttm_resource_manager *man = &mgr->manager; struct amdgpu_bo *bo = NULL; uint64_t bp; int i; @@ -396,12 +347,18 @@ static void amdgpu_virt_ras_reserve_bps(struct amdgpu_device *adev) * 2) a ras bad page has been reserved (duplicate error injection * for one page); */ - if (amdgpu_bo_create_kernel_at(adev, bp << AMDGPU_GPU_PAGE_SHIFT, - AMDGPU_GPU_PAGE_SIZE, - &bo, NULL)) - DRM_DEBUG("RAS WARN: reserve vram for retired page %llx fail\n", bp); - - data->bps_bo[i] = bo; + if (ttm_resource_manager_used(man)) { + amdgpu_vram_mgr_reserve_range(&adev->mman.vram_mgr, + bp << AMDGPU_GPU_PAGE_SHIFT, + AMDGPU_GPU_PAGE_SIZE); + data->bps_bo[i] = NULL; + } else { + if (amdgpu_bo_create_kernel_at(adev, bp << AMDGPU_GPU_PAGE_SHIFT, + AMDGPU_GPU_PAGE_SIZE, + &bo, NULL)) + DRM_DEBUG("RAS WARN: reserve vram for retired page %llx fail\n", bp); + data->bps_bo[i] = bo; + } data->last_reserved = i + 1; bo = NULL; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index d4207e44141f..fa7be5f277b9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -88,7 +88,8 @@ struct amdgpu_virt_ops { int (*wait_reset)(struct amdgpu_device *adev); void (*trans_msg)(struct amdgpu_device *adev, enum idh_request req, u32 data1, u32 data2, u32 data3); - void (*ras_poison_handler)(struct amdgpu_device *adev); + void (*ras_poison_handler)(struct amdgpu_device *adev, + enum amdgpu_ras_block block); }; /* @@ -332,10 +333,6 @@ static inline bool is_virtual_machine(void) ((adev)->virt.gim_feature & AMDGIM_FEATURE_VCN_RB_DECOUPLE) bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev); void amdgpu_virt_init_setting(struct amdgpu_device *adev); -void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev, - uint32_t reg0, uint32_t rreg1, - uint32_t ref, uint32_t mask, - uint32_t xcc_inst); int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_release_full_gpu(struct amdgpu_device *adev, bool init); int amdgpu_virt_reset_gpu(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c index 453a4b786cfc..8baa2e0935cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c @@ -660,8 +660,7 @@ static const struct amd_ip_funcs amdgpu_vkms_ip_funcs = { .set_powergating_state = amdgpu_vkms_set_powergating_state, }; -const struct amdgpu_ip_block_version amdgpu_vkms_ip_block = -{ +const struct amdgpu_ip_block_version amdgpu_vkms_ip_block = { .type = AMD_IP_BLOCK_TYPE_DCE, .major = 1, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index b8fcb6c55698..ed4a8c5d26d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -234,6 +234,22 @@ static void amdgpu_vm_bo_invalidated(struct amdgpu_vm_bo_base *vm_bo) } /** + * amdgpu_vm_bo_evicted_user - vm_bo is evicted + * + * @vm_bo: vm_bo which is evicted + * + * State for BOs used by user mode queues which are not at the location they + * should be. + */ +static void amdgpu_vm_bo_evicted_user(struct amdgpu_vm_bo_base *vm_bo) +{ + vm_bo->moved = true; + spin_lock(&vm_bo->vm->status_lock); + list_move(&vm_bo->vm_status, &vm_bo->vm->evicted_user); + spin_unlock(&vm_bo->vm->status_lock); +} + +/** * amdgpu_vm_bo_relocated - vm_bo is reloacted * * @vm_bo: vm_bo which is relocated @@ -427,21 +443,25 @@ uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm) } /** - * amdgpu_vm_validate_pt_bos - validate the page table BOs + * amdgpu_vm_validate - validate evicted BOs tracked in the VM * * @adev: amdgpu device pointer * @vm: vm providing the BOs + * @ticket: optional reservation ticket used to reserve the VM * @validate: callback to do the validation * @param: parameter for the validation callback * - * Validate the page table BOs on command submission if neccessary. + * Validate the page table BOs and per-VM BOs on command submission if + * necessary. If a ticket is given, also try to validate evicted user queue + * BOs. They must already be reserved with the given ticket. * * Returns: * Validation result. */ -int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, - int (*validate)(void *p, struct amdgpu_bo *bo), - void *param) +int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm, + struct ww_acquire_ctx *ticket, + int (*validate)(void *p, struct amdgpu_bo *bo), + void *param) { struct amdgpu_vm_bo_base *bo_base; struct amdgpu_bo *shadow; @@ -484,6 +504,28 @@ int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, } spin_lock(&vm->status_lock); } + while (ticket && !list_empty(&vm->evicted_user)) { + bo_base = list_first_entry(&vm->evicted_user, + struct amdgpu_vm_bo_base, + vm_status); + spin_unlock(&vm->status_lock); + + bo = bo_base->bo; + + if (dma_resv_locking_ctx(bo->tbo.base.resv) != ticket) { + pr_warn_ratelimited("Evicted user BO is not reserved in pid %d\n", + vm->task_info.pid); + return -EINVAL; + } + + r = validate(param, bo); + if (r) + return r; + + amdgpu_vm_bo_invalidated(bo_base); + + spin_lock(&vm->status_lock); + } spin_unlock(&vm->status_lock); amdgpu_vm_eviction_lock(vm); @@ -651,7 +693,7 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job, amdgpu_gmc_emit_pasid_mapping(ring, job->vmid, job->pasid); if (spm_update_needed && adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, job->vmid); + adev->gfx.rlc.funcs->update_spm_vmid(adev, ring, job->vmid); if (!ring->is_mes_queue && ring->funcs->emit_gds_switch && gds_switch_needed) { @@ -1426,11 +1468,21 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev, } r = amdgpu_vm_bo_update(adev, bo_va, clear); - if (r) - return r; if (unlock) dma_resv_unlock(resv); + if (r) + return r; + + /* Remember evicted DMABuf imports in compute VMs for later + * validation + */ + if (vm->is_compute_context && + bo_va->base.bo->tbo.base.import_attach && + (!bo_va->base.bo->tbo.resource || + bo_va->base.bo->tbo.resource->mem_type == TTM_PL_SYSTEM)) + amdgpu_vm_bo_evicted_user(&bo_va->base); + spin_lock(&vm->status_lock); } spin_unlock(&vm->status_lock); @@ -2196,6 +2248,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) vm->reserved_vmid[i] = NULL; INIT_LIST_HEAD(&vm->evicted); + INIT_LIST_HEAD(&vm->evicted_user); INIT_LIST_HEAD(&vm->relocated); INIT_LIST_HEAD(&vm->moved); INIT_LIST_HEAD(&vm->idle); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 4740dd65b99d..666698a57192 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -136,7 +136,11 @@ struct amdgpu_mem_stats; #define AMDGPU_IS_MMHUB1(x) ((x) >= AMDGPU_MMHUB1_START && (x) < AMDGPU_MAX_VMHUBS) /* Reserve 2MB at top/bottom of address space for kernel use */ -#define AMDGPU_VA_RESERVED_SIZE (2ULL << 20) +#define AMDGPU_VA_RESERVED_CSA_SIZE (2ULL << 20) +#define AMDGPU_VA_RESERVED_SEQ64_SIZE (2ULL << 20) +#define AMDGPU_VA_RESERVED_BOTTOM (2ULL << 20) +#define AMDGPU_VA_RESERVED_TOP (AMDGPU_VA_RESERVED_SEQ64_SIZE + \ + AMDGPU_VA_RESERVED_CSA_SIZE) /* See vm_update_mode */ #define AMDGPU_VM_USE_CPU_FOR_GFX (1 << 0) @@ -288,9 +292,12 @@ struct amdgpu_vm { /* Lock to protect vm_bo add/del/move on all lists of vm */ spinlock_t status_lock; - /* BOs who needs a validation */ + /* Per-VM and PT BOs who needs a validation */ struct list_head evicted; + /* BOs for user mode queues that need a validation */ + struct list_head evicted_user; + /* PT BOs which relocated and their parent need an update */ struct list_head relocated; @@ -434,9 +441,10 @@ int amdgpu_vm_lock_pd(struct amdgpu_vm *vm, struct drm_exec *exec, unsigned int num_fences); bool amdgpu_vm_ready(struct amdgpu_vm *vm); uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm); -int amdgpu_vm_validate_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm, - int (*callback)(void *p, struct amdgpu_bo *bo), - void *param); +int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm, + struct ww_acquire_ctx *ticket, + int (*callback)(void *p, struct amdgpu_bo *bo), + void *param); int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job, bool need_pipe_sync); int amdgpu_vm_update_pdes(struct amdgpu_device *adev, struct amdgpu_vm *vm, bool immediate); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index a6c88f2fe6e5..20d51f6c9bb8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -1035,15 +1035,74 @@ int amdgpu_xgmi_remove_device(struct amdgpu_device *adev) return 0; } +static int xgmi_v6_4_0_aca_bank_generate_report(struct aca_handle *handle, struct aca_bank *bank, enum aca_error_type type, + struct aca_bank_report *report, void *data) +{ + struct amdgpu_device *adev = handle->adev; + const char *error_str; + u64 status; + int ret, ext_error_code; + + ret = aca_bank_info_decode(bank, &report->info); + if (ret) + return ret; + + status = bank->regs[ACA_REG_IDX_STATUS]; + ext_error_code = ACA_REG__STATUS__ERRORCODEEXT(status); + + error_str = ext_error_code < ARRAY_SIZE(xgmi_v6_4_0_ras_error_code_ext) ? + xgmi_v6_4_0_ras_error_code_ext[ext_error_code] : NULL; + if (error_str) + dev_info(adev->dev, "%s detected\n", error_str); + + if ((type == ACA_ERROR_TYPE_UE && ext_error_code == 0) || + (type == ACA_ERROR_TYPE_CE && ext_error_code == 6)) + report->count[type] = ACA_REG__MISC0__ERRCNT(bank->regs[ACA_REG_IDX_MISC0]); + + return 0; +} + +static const struct aca_bank_ops xgmi_v6_4_0_aca_bank_ops = { + .aca_bank_generate_report = xgmi_v6_4_0_aca_bank_generate_report, +}; + +static const struct aca_info xgmi_v6_4_0_aca_info = { + .hwip = ACA_HWIP_TYPE_PCS_XGMI, + .mask = ACA_ERROR_UE_MASK | ACA_ERROR_CE_MASK, + .bank_ops = &xgmi_v6_4_0_aca_bank_ops, +}; + static int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) { + int r; + if (!adev->gmc.xgmi.supported || adev->gmc.xgmi.num_physical_nodes == 0) return 0; amdgpu_ras_reset_error_count(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL); - return amdgpu_ras_block_late_init(adev, ras_block); + r = amdgpu_ras_block_late_init(adev, ras_block); + if (r) + return r; + + switch (amdgpu_ip_version(adev, XGMI_HWIP, 0)) { + case IP_VERSION(6, 4, 0): + r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__XGMI_WAFL, + &xgmi_v6_4_0_aca_info, NULL); + if (r) + goto late_fini; + break; + default: + break; + } + + return 0; + +late_fini: + amdgpu_ras_block_late_fini(adev, ras_block); + + return r; } uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, @@ -1099,7 +1158,7 @@ static void amdgpu_xgmi_legacy_reset_ras_error_count(struct amdgpu_device *adev) static void __xgmi_v6_4_0_reset_error_count(struct amdgpu_device *adev, int xgmi_inst, u64 mca_base) { - WREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS, 0ULL); + WREG64_MCA(xgmi_inst, mca_base, ACA_REG_IDX_STATUS, 0ULL); } static void xgmi_v6_4_0_reset_error_count(struct amdgpu_device *adev, int xgmi_inst) @@ -1277,12 +1336,12 @@ static void amdgpu_xgmi_legacy_query_ras_error_count(struct amdgpu_device *adev, err_data->ce_count += ce_cnt; } -static enum amdgpu_mca_error_type xgmi_v6_4_0_pcs_mca_get_error_type(struct amdgpu_device *adev, u64 status) +static enum aca_error_type xgmi_v6_4_0_pcs_mca_get_error_type(struct amdgpu_device *adev, u64 status) { const char *error_str; int ext_error_code; - ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(status); + ext_error_code = ACA_REG__STATUS__ERRORCODEEXT(status); error_str = ext_error_code < ARRAY_SIZE(xgmi_v6_4_0_ras_error_code_ext) ? xgmi_v6_4_0_ras_error_code_ext[ext_error_code] : NULL; @@ -1291,9 +1350,9 @@ static enum amdgpu_mca_error_type xgmi_v6_4_0_pcs_mca_get_error_type(struct amdg switch (ext_error_code) { case 0: - return AMDGPU_MCA_ERROR_TYPE_UE; + return ACA_ERROR_TYPE_UE; case 6: - return AMDGPU_MCA_ERROR_TYPE_CE; + return ACA_ERROR_TYPE_CE; default: return -EINVAL; } @@ -1307,22 +1366,22 @@ static void __xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, struct a int xgmi_inst = mcm_info->die_id; u64 status = 0; - status = RREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS); - if (!MCA_REG__STATUS__VAL(status)) + status = RREG64_MCA(xgmi_inst, mca_base, ACA_REG_IDX_STATUS); + if (!ACA_REG__STATUS__VAL(status)) return; switch (xgmi_v6_4_0_pcs_mca_get_error_type(adev, status)) { - case AMDGPU_MCA_ERROR_TYPE_UE: + case ACA_ERROR_TYPE_UE: amdgpu_ras_error_statistic_ue_count(err_data, mcm_info, NULL, 1ULL); break; - case AMDGPU_MCA_ERROR_TYPE_CE: + case ACA_ERROR_TYPE_CE: amdgpu_ras_error_statistic_ce_count(err_data, mcm_info, NULL, 1ULL); break; default: break; } - WREG64_MCA(xgmi_inst, mca_base, MCA_REG_IDX_STATUS, 0ULL); + WREG64_MCA(xgmi_inst, mca_base, ACA_REG_IDX_STATUS, 0ULL); } static void xgmi_v6_4_0_query_error_count(struct amdgpu_device *adev, int xgmi_inst, struct ras_err_data *err_data) diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index a33e890c70d9..b888613f653f 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -62,6 +62,7 @@ typedef struct { struct atom_context *ctx; uint32_t *ps, *ws; + int ps_size, ws_size; int ps_shift; uint16_t start; unsigned last_jump; @@ -70,8 +71,8 @@ typedef struct { } atom_exec_context; int amdgpu_atom_debug; -static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t *params); -int amdgpu_atom_execute_table(struct atom_context *ctx, int index, uint32_t *params); +static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t *params, int params_size); +int amdgpu_atom_execute_table(struct atom_context *ctx, int index, uint32_t *params, int params_size); static uint32_t atom_arg_mask[8] = { 0xFFFFFFFF, 0xFFFF, 0xFFFF00, 0xFFFF0000, 0xFF, 0xFF00, 0xFF0000, @@ -223,7 +224,10 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, (*ptr)++; /* get_unaligned_le32 avoids unaligned accesses from atombios * tables, noticed on a DEC Alpha. */ - val = get_unaligned_le32((u32 *)&ctx->ps[idx]); + if (idx < ctx->ps_size) + val = get_unaligned_le32((u32 *)&ctx->ps[idx]); + else + pr_info("PS index out of range: %i > %i\n", idx, ctx->ps_size); if (print) DEBUG("PS[0x%02X,0x%04X]", idx, val); break; @@ -261,7 +265,10 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, val = gctx->reg_block; break; default: - val = ctx->ws[idx]; + if (idx < ctx->ws_size) + val = ctx->ws[idx]; + else + pr_info("WS index out of range: %i > %i\n", idx, ctx->ws_size); } break; case ATOM_ARG_ID: @@ -495,6 +502,10 @@ static void atom_put_dst(atom_exec_context *ctx, int arg, uint8_t attr, idx = U8(*ptr); (*ptr)++; DEBUG("PS[0x%02X]", idx); + if (idx >= ctx->ps_size) { + pr_info("PS index out of range: %i > %i\n", idx, ctx->ps_size); + return; + } ctx->ps[idx] = cpu_to_le32(val); break; case ATOM_ARG_WS: @@ -527,6 +538,10 @@ static void atom_put_dst(atom_exec_context *ctx, int arg, uint8_t attr, gctx->reg_block = val; break; default: + if (idx >= ctx->ws_size) { + pr_info("WS index out of range: %i > %i\n", idx, ctx->ws_size); + return; + } ctx->ws[idx] = val; } break; @@ -624,7 +639,7 @@ static void atom_op_calltable(atom_exec_context *ctx, int *ptr, int arg) else SDEBUG(" table: %d\n", idx); if (U16(ctx->ctx->cmd_table + 4 + 2 * idx)) - r = amdgpu_atom_execute_table_locked(ctx->ctx, idx, ctx->ps + ctx->ps_shift); + r = amdgpu_atom_execute_table_locked(ctx->ctx, idx, ctx->ps + ctx->ps_shift, ctx->ps_size - ctx->ps_shift); if (r) { ctx->abort = true; } @@ -1203,7 +1218,7 @@ static struct { atom_op_div32, ATOM_ARG_WS}, }; -static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t *params) +static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t *params, int params_size) { int base = CU16(ctx->cmd_table + 4 + 2 * index); int len, ws, ps, ptr; @@ -1225,12 +1240,16 @@ static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index, ectx.ps_shift = ps / 4; ectx.start = base; ectx.ps = params; + ectx.ps_size = params_size; ectx.abort = false; ectx.last_jump = 0; - if (ws) + if (ws) { ectx.ws = kcalloc(4, ws, GFP_KERNEL); - else + ectx.ws_size = ws; + } else { ectx.ws = NULL; + ectx.ws_size = 0; + } debug_depth++; while (1) { @@ -1264,7 +1283,7 @@ free: return ret; } -int amdgpu_atom_execute_table(struct atom_context *ctx, int index, uint32_t *params) +int amdgpu_atom_execute_table(struct atom_context *ctx, int index, uint32_t *params, int params_size) { int r; @@ -1280,7 +1299,7 @@ int amdgpu_atom_execute_table(struct atom_context *ctx, int index, uint32_t *par /* reset divmul */ ctx->divmul[0] = 0; ctx->divmul[1] = 0; - r = amdgpu_atom_execute_table_locked(ctx, index, params); + r = amdgpu_atom_execute_table_locked(ctx, index, params, params_size); mutex_unlock(&ctx->mutex); return r; } @@ -1552,7 +1571,7 @@ int amdgpu_atom_asic_init(struct atom_context *ctx) if (!CU16(ctx->cmd_table + 4 + 2 * ATOM_CMD_INIT)) return 1; - ret = amdgpu_atom_execute_table(ctx, ATOM_CMD_INIT, ps); + ret = amdgpu_atom_execute_table(ctx, ATOM_CMD_INIT, ps, 16); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/atom.h b/drivers/gpu/drm/amd/amdgpu/atom.h index c11cf18a0f18..b807f6639a4c 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.h +++ b/drivers/gpu/drm/amd/amdgpu/atom.h @@ -156,7 +156,7 @@ struct atom_context { extern int amdgpu_atom_debug; 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_execute_table(struct atom_context *ctx, int index, uint32_t *params, int params_size); 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, diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c b/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c index 10098fdd33fc..3dfc28840a7d 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_crtc.c @@ -77,7 +77,7 @@ void amdgpu_atombios_crtc_overscan_setup(struct drm_crtc *crtc, args.usOverscanTop = cpu_to_le16(amdgpu_crtc->v_border); break; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_scaler_setup(struct drm_crtc *crtc) @@ -106,7 +106,7 @@ void amdgpu_atombios_crtc_scaler_setup(struct drm_crtc *crtc) args.ucEnable = ATOM_SCALER_DISABLE; break; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_lock(struct drm_crtc *crtc, int lock) @@ -123,7 +123,7 @@ void amdgpu_atombios_crtc_lock(struct drm_crtc *crtc, int lock) args.ucCRTC = amdgpu_crtc->crtc_id; args.ucEnable = lock; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_enable(struct drm_crtc *crtc, int state) @@ -139,7 +139,7 @@ void amdgpu_atombios_crtc_enable(struct drm_crtc *crtc, int state) args.ucCRTC = amdgpu_crtc->crtc_id; args.ucEnable = state; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_blank(struct drm_crtc *crtc, int state) @@ -155,7 +155,7 @@ void amdgpu_atombios_crtc_blank(struct drm_crtc *crtc, int state) args.ucCRTC = amdgpu_crtc->crtc_id; args.ucBlanking = state; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_powergate(struct drm_crtc *crtc, int state) @@ -171,7 +171,7 @@ void amdgpu_atombios_crtc_powergate(struct drm_crtc *crtc, int state) args.ucDispPipeId = amdgpu_crtc->crtc_id; args.ucEnable = state; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_powergate_init(struct amdgpu_device *adev) @@ -183,7 +183,7 @@ void amdgpu_atombios_crtc_powergate_init(struct amdgpu_device *adev) args.ucEnable = ATOM_INIT; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_crtc_set_dtd_timing(struct drm_crtc *crtc, @@ -228,7 +228,7 @@ void amdgpu_atombios_crtc_set_dtd_timing(struct drm_crtc *crtc, args.susModeMiscInfo.usAccess = cpu_to_le16(misc); args.ucCRTC = amdgpu_crtc->crtc_id; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } union atom_enable_ss { @@ -293,7 +293,7 @@ static void amdgpu_atombios_crtc_program_ss(struct amdgpu_device *adev, args.v3.usSpreadSpectrumStep = cpu_to_le16(ss->step); args.v3.ucEnable = enable; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } union adjust_pixel_clock { @@ -395,7 +395,7 @@ static u32 amdgpu_atombios_crtc_adjust_pll(struct drm_crtc *crtc, ADJUST_DISPLAY_CONFIG_SS_ENABLE; amdgpu_atom_execute_table(adev->mode_info.atom_context, - index, (uint32_t *)&args); + index, (uint32_t *)&args, sizeof(args)); adjusted_clock = le16_to_cpu(args.v1.usPixelClock) * 10; break; case 3: @@ -428,7 +428,7 @@ static u32 amdgpu_atombios_crtc_adjust_pll(struct drm_crtc *crtc, args.v3.sInput.ucExtTransmitterID = 0; amdgpu_atom_execute_table(adev->mode_info.atom_context, - index, (uint32_t *)&args); + index, (uint32_t *)&args, sizeof(args)); adjusted_clock = le32_to_cpu(args.v3.sOutput.ulDispPllFreq) * 10; if (args.v3.sOutput.ucRefDiv) { amdgpu_crtc->pll_flags |= AMDGPU_PLL_USE_FRAC_FB_DIV; @@ -514,7 +514,7 @@ void amdgpu_atombios_crtc_set_disp_eng_pll(struct amdgpu_device *adev, DRM_ERROR("Unknown table version %d %d\n", frev, crev); return; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } union set_dce_clock { @@ -544,7 +544,7 @@ u32 amdgpu_atombios_crtc_set_dce_clock(struct amdgpu_device *adev, args.v2_1.asParam.ulDCEClkFreq = cpu_to_le32(freq); /* 10kHz units */ args.v2_1.asParam.ucDCEClkType = clk_type; args.v2_1.asParam.ucDCEClkSrc = clk_src; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); ret_freq = le32_to_cpu(args.v2_1.asParam.ulDCEClkFreq) * 10; break; default: @@ -740,7 +740,7 @@ void amdgpu_atombios_crtc_program_pll(struct drm_crtc *crtc, return; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } int amdgpu_atombios_crtc_prepare_pll(struct drm_crtc *crtc, diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c index 87c41e0e9b7c..622634c08c7b 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c @@ -83,7 +83,7 @@ static int amdgpu_atombios_dp_process_aux_ch(struct amdgpu_i2c_chan *chan, args.v2.ucDelay = delay / 10; args.v2.ucHPD_ID = chan->rec.hpd; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *ack = args.v2.ucReplyStatus; @@ -301,7 +301,7 @@ static u8 amdgpu_atombios_dp_encoder_service(struct amdgpu_device *adev, args.ucLaneNum = lane_num; args.ucStatus = 0; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); return args.ucStatus; } diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c index 7672abe6c140..25feab188dfe 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c @@ -335,7 +335,7 @@ amdgpu_atombios_encoder_setup_dac(struct drm_encoder *encoder, int action) args.ucDacStandard = ATOM_DAC1_PS2; args.usPixelClock = cpu_to_le16(amdgpu_encoder->pixel_clock / 10); - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } @@ -432,7 +432,7 @@ amdgpu_atombios_encoder_setup_dvo(struct drm_encoder *encoder, int action) break; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } int amdgpu_atombios_encoder_get_encoder_mode(struct drm_encoder *encoder) @@ -732,7 +732,7 @@ amdgpu_atombios_encoder_setup_dig_encoder(struct drm_encoder *encoder, break; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } @@ -1136,7 +1136,7 @@ amdgpu_atombios_encoder_setup_dig_transmitter(struct drm_encoder *encoder, int a break; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } bool @@ -1164,7 +1164,7 @@ amdgpu_atombios_encoder_set_edp_panel_power(struct drm_connector *connector, args.v1.ucAction = action; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); /* wait for the panel to power up */ if (action == ATOM_TRANSMITTER_ACTION_POWER_ON) { @@ -1288,7 +1288,7 @@ amdgpu_atombios_encoder_setup_external_encoder(struct drm_encoder *encoder, DRM_ERROR("Unknown table version: %d, %d\n", frev, crev); return; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void @@ -1633,7 +1633,7 @@ amdgpu_atombios_encoder_set_crtc_source(struct drm_encoder *encoder) return; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } /* This only needs to be called once at startup */ @@ -1706,7 +1706,7 @@ amdgpu_atombios_encoder_dac_load_detect(struct drm_encoder *encoder, args.sDacload.ucMisc = DAC_LOAD_MISC_YPrPb; } - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); return true; } else diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c b/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c index af0335535f82..a6501114322f 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_i2c.c @@ -86,7 +86,7 @@ static int amdgpu_atombios_i2c_process_i2c_ch(struct amdgpu_i2c_chan *chan, args.ucSlaveAddr = slave_addr << 1; args.ucLineNumber = chan->rec.i2c_id; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); /* error */ if (args.ucStatus != HW_ASSISTED_I2C_STATUS_SUCCESS) { @@ -172,5 +172,5 @@ void amdgpu_atombios_i2c_channel_trans(struct amdgpu_device *adev, u8 slave_addr args.ucSlaveAddr = slave_addr; args.ucLineNumber = line_number; - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args); + amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } diff --git a/drivers/gpu/drm/amd/amdgpu/clearstate_gfx9.h b/drivers/gpu/drm/amd/amdgpu/clearstate_gfx9.h index 567a904804bc..9c85ca6358c1 100644 --- a/drivers/gpu/drm/amd/amdgpu/clearstate_gfx9.h +++ b/drivers/gpu/drm/amd/amdgpu/clearstate_gfx9.h @@ -21,8 +21,7 @@ * */ -static const unsigned int gfx9_SECT_CONTEXT_def_1[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_1[] = { 0x00000000, // DB_RENDER_CONTROL 0x00000000, // DB_COUNT_CONTROL 0x00000000, // DB_DEPTH_VIEW @@ -236,8 +235,7 @@ static const unsigned int gfx9_SECT_CONTEXT_def_1[] = 0x00000000, // PA_SC_VPORT_ZMIN_15 0x3f800000, // PA_SC_VPORT_ZMAX_15 }; -static const unsigned int gfx9_SECT_CONTEXT_def_2[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_2[] = { 0x00000000, // PA_SC_SCREEN_EXTENT_CONTROL 0x00000000, // PA_SC_TILE_STEERING_OVERRIDE 0x00000000, // CP_PERFMON_CNTX_CNTL @@ -521,15 +519,13 @@ static const unsigned int gfx9_SECT_CONTEXT_def_2[] = 0x00000000, // CB_MRT6_EPITCH 0x00000000, // CB_MRT7_EPITCH }; -static const unsigned int gfx9_SECT_CONTEXT_def_3[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_3[] = { 0x00000000, // PA_CL_POINT_X_RAD 0x00000000, // PA_CL_POINT_Y_RAD 0x00000000, // PA_CL_POINT_SIZE 0x00000000, // PA_CL_POINT_CULL_RAD }; -static const unsigned int gfx9_SECT_CONTEXT_def_4[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_4[] = { 0x00000000, // DB_DEPTH_CONTROL 0x00000000, // DB_EQAA 0x00000000, // CB_COLOR_CONTROL @@ -688,17 +684,14 @@ static const unsigned int gfx9_SECT_CONTEXT_def_4[] = 0x00000000, // VGT_GS_OUT_PRIM_TYPE 0x00000000, // IA_ENHANCE }; -static const unsigned int gfx9_SECT_CONTEXT_def_5[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_5[] = { 0x00000000, // WD_ENHANCE 0x00000000, // VGT_PRIMITIVEID_EN }; -static const unsigned int gfx9_SECT_CONTEXT_def_6[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_6[] = { 0x00000000, // VGT_PRIMITIVEID_RESET }; -static const unsigned int gfx9_SECT_CONTEXT_def_7[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_7[] = { 0x00000000, // VGT_GS_MAX_PRIMS_PER_SUBGROUP 0x00000000, // VGT_DRAW_PAYLOAD_CNTL 0, // HOLE @@ -766,8 +759,7 @@ static const unsigned int gfx9_SECT_CONTEXT_def_7[] = 0x00000000, // VGT_STRMOUT_CONFIG 0x00000000, // VGT_STRMOUT_BUFFER_CONFIG }; -static const unsigned int gfx9_SECT_CONTEXT_def_8[] = -{ +static const unsigned int gfx9_SECT_CONTEXT_def_8[] = { 0x00000000, // PA_SC_CENTROID_PRIORITY_0 0x00000000, // PA_SC_CENTROID_PRIORITY_1 0x00001000, // PA_SC_LINE_CNTL @@ -924,8 +916,7 @@ static const unsigned int gfx9_SECT_CONTEXT_def_8[] = 0x00000000, // CB_COLOR7_DCC_BASE 0x00000000, // CB_COLOR7_DCC_BASE_EXT }; -static const struct cs_extent_def gfx9_SECT_CONTEXT_defs[] = -{ +static const struct cs_extent_def gfx9_SECT_CONTEXT_defs[] = { {gfx9_SECT_CONTEXT_def_1, 0x0000a000, 212 }, {gfx9_SECT_CONTEXT_def_2, 0x0000a0d6, 282 }, {gfx9_SECT_CONTEXT_def_3, 0x0000a1f5, 4 }, diff --git a/drivers/gpu/drm/amd/amdgpu/clearstate_si.h b/drivers/gpu/drm/amd/amdgpu/clearstate_si.h index 66e39cdb5cb0..5fd96ddd7f0f 100644 --- a/drivers/gpu/drm/amd/amdgpu/clearstate_si.h +++ b/drivers/gpu/drm/amd/amdgpu/clearstate_si.h @@ -21,8 +21,7 @@ * */ -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 @@ -236,8 +235,7 @@ static const u32 si_SECT_CONTEXT_def_1[] = 0x00000000, // PA_SC_VPORT_ZMIN_15 0x3f800000, // PA_SC_VPORT_ZMAX_15 }; -static const u32 si_SECT_CONTEXT_def_2[] = -{ +static const u32 si_SECT_CONTEXT_def_2[] = { 0x00000000, // CP_PERFMON_CNTX_CNTL 0x00000000, // CP_RINGID 0x00000000, // CP_VMID @@ -511,8 +509,7 @@ static const u32 si_SECT_CONTEXT_def_2[] = 0x00000000, // CB_BLEND6_CONTROL 0x00000000, // CB_BLEND7_CONTROL }; -static const u32 si_SECT_CONTEXT_def_3[] = -{ +static const u32 si_SECT_CONTEXT_def_3[] = { 0x00000000, // PA_CL_POINT_X_RAD 0x00000000, // PA_CL_POINT_Y_RAD 0x00000000, // PA_CL_POINT_SIZE @@ -520,8 +517,7 @@ static const u32 si_SECT_CONTEXT_def_3[] = 0x00000000, // VGT_DMA_BASE_HI 0x00000000, // VGT_DMA_BASE }; -static const u32 si_SECT_CONTEXT_def_4[] = -{ +static const u32 si_SECT_CONTEXT_def_4[] = { 0x00000000, // DB_DEPTH_CONTROL 0x00000000, // DB_EQAA 0x00000000, // CB_COLOR_CONTROL @@ -680,16 +676,13 @@ static const u32 si_SECT_CONTEXT_def_4[] = 0x00000000, // VGT_GS_OUT_PRIM_TYPE 0x00000000, // IA_ENHANCE }; -static const u32 si_SECT_CONTEXT_def_5[] = -{ +static const u32 si_SECT_CONTEXT_def_5[] = { 0x00000000, // VGT_PRIMITIVEID_EN }; -static const u32 si_SECT_CONTEXT_def_6[] = -{ +static const u32 si_SECT_CONTEXT_def_6[] = { 0x00000000, // VGT_PRIMITIVEID_RESET }; -static const u32 si_SECT_CONTEXT_def_7[] = -{ +static const u32 si_SECT_CONTEXT_def_7[] = { 0x00000000, // VGT_MULTI_PRIM_IB_RESET_EN 0, // HOLE 0, // HOLE @@ -924,8 +917,7 @@ static const u32 si_SECT_CONTEXT_def_7[] = 0x00000000, // CB_COLOR7_CLEAR_WORD0 0x00000000, // CB_COLOR7_CLEAR_WORD1 }; -static const struct cs_extent_def si_SECT_CONTEXT_defs[] = -{ +static const struct cs_extent_def si_SECT_CONTEXT_defs[] = { {si_SECT_CONTEXT_def_1, 0x0000a000, 212 }, {si_SECT_CONTEXT_def_2, 0x0000a0d8, 272 }, {si_SECT_CONTEXT_def_3, 0x0000a1f5, 6 }, diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 587ee632a3b8..221af054d874 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -52,6 +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 void dce_v10_0_hpd_int_ack(struct amdgpu_device *adev, int hpd); static const u32 crtc_offsets[] = { CRTC0_REGISTER_OFFSET, @@ -364,6 +365,7 @@ static void dce_v10_0_hpd_init(struct amdgpu_device *adev) AMDGPU_HPD_DISCONNECT_INT_DELAY_IN_MS); WREG32(mmDC_HPD_TOGGLE_FILT_CNTL + hpd_offsets[amdgpu_connector->hpd.hpd], tmp); + dce_v10_0_hpd_int_ack(adev, amdgpu_connector->hpd.hpd); dce_v10_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd); amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd); diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index f22ec27365bd..69e8b0db6cf7 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -52,6 +52,7 @@ static void dce_v11_0_set_display_funcs(struct amdgpu_device *adev); static void dce_v11_0_set_irq_funcs(struct amdgpu_device *adev); +static void dce_v11_0_hpd_int_ack(struct amdgpu_device *adev, int hpd); static const u32 crtc_offsets[] = { @@ -388,6 +389,7 @@ static void dce_v11_0_hpd_init(struct amdgpu_device *adev) AMDGPU_HPD_DISCONNECT_INT_DELAY_IN_MS); WREG32(mmDC_HPD_TOGGLE_FILT_CNTL + hpd_offsets[amdgpu_connector->hpd.hpd], tmp); + dce_v11_0_hpd_int_ack(adev, amdgpu_connector->hpd.hpd); dce_v11_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd); amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd); } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index 4dbe9b3259b5..60d40201fdd1 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -273,6 +273,21 @@ static void dce_v6_0_hpd_set_polarity(struct amdgpu_device *adev, WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], tmp); } +static void dce_v6_0_hpd_int_ack(struct amdgpu_device *adev, + int hpd) +{ + u32 tmp; + + if (hpd >= adev->mode_info.num_hpd) { + DRM_DEBUG("invalid hdp %d\n", hpd); + return; + } + + tmp = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd]); + tmp |= DC_HPD1_INT_CONTROL__DC_HPD1_INT_ACK_MASK; + WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], tmp); +} + /** * dce_v6_0_hpd_init - hpd setup callback. * @@ -312,6 +327,7 @@ static void dce_v6_0_hpd_init(struct amdgpu_device *adev) continue; } + dce_v6_0_hpd_int_ack(adev, amdgpu_connector->hpd.hpd); dce_v6_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd); amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd); } @@ -3089,7 +3105,7 @@ static int dce_v6_0_hpd_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - uint32_t disp_int, mask, tmp; + uint32_t disp_int, mask; unsigned hpd; if (entry->src_data[0] >= adev->mode_info.num_hpd) { @@ -3102,9 +3118,7 @@ static int dce_v6_0_hpd_irq(struct amdgpu_device *adev, mask = interrupt_status_offsets[hpd].hpd; if (disp_int & mask) { - tmp = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd]); - tmp |= DC_HPD1_INT_CONTROL__DC_HPD1_INT_ACK_MASK; - WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], tmp); + dce_v6_0_hpd_int_ack(adev, hpd); schedule_delayed_work(&adev->hotplug_work, 0); DRM_DEBUG("IH: HPD%d\n", hpd + 1); } diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 05bcce23385e..5a5fcc45e452 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -265,6 +265,21 @@ static void dce_v8_0_hpd_set_polarity(struct amdgpu_device *adev, WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], tmp); } +static void dce_v8_0_hpd_int_ack(struct amdgpu_device *adev, + int hpd) +{ + u32 tmp; + + if (hpd >= adev->mode_info.num_hpd) { + DRM_DEBUG("invalid hdp %d\n", hpd); + return; + } + + tmp = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd]); + tmp |= DC_HPD1_INT_CONTROL__DC_HPD1_INT_ACK_MASK; + WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], tmp); +} + /** * dce_v8_0_hpd_init - hpd setup callback. * @@ -304,6 +319,7 @@ static void dce_v8_0_hpd_init(struct amdgpu_device *adev) continue; } + dce_v8_0_hpd_int_ack(adev, amdgpu_connector->hpd.hpd); dce_v8_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd); amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd); } @@ -3177,7 +3193,7 @@ static int dce_v8_0_hpd_irq(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) { - uint32_t disp_int, mask, tmp; + uint32_t disp_int, mask; unsigned hpd; if (entry->src_data[0] >= adev->mode_info.num_hpd) { @@ -3190,9 +3206,7 @@ static int dce_v8_0_hpd_irq(struct amdgpu_device *adev, mask = interrupt_status_offsets[hpd].hpd; if (disp_int & mask) { - tmp = RREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd]); - tmp |= DC_HPD1_INT_CONTROL__DC_HPD1_INT_ACK_MASK; - WREG32(mmDC_HPD1_INT_CONTROL + hpd_offsets[hpd], tmp); + dce_v8_0_hpd_int_ack(adev, hpd); schedule_delayed_work(&adev->hotplug_work, 0); DRM_DEBUG("IH: HPD%d\n", hpd + 1); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index dcdecb18b230..b02d63328f1c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -7947,7 +7947,7 @@ static void gfx_v10_0_update_spm_vmid_internal(struct amdgpu_device *adev, WREG32_SOC15_NO_KIQ(GC, 0, mmRLC_SPM_MC_CNTL, data); } -static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, unsigned int vmid) +static void gfx_v10_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned int vmid) { amdgpu_gfx_off_ctrl(adev, false); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index 4f3bfdc75b37..2fb1342d5bd9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -727,7 +727,7 @@ static int gfx_v11_0_rlc_init(struct amdgpu_device *adev) /* init spm vmid with 0xf */ if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + adev->gfx.rlc.funcs->update_spm_vmid(adev, NULL, 0xf); return 0; } @@ -5027,7 +5027,7 @@ static int gfx_v11_0_update_gfx_clock_gating(struct amdgpu_device *adev, return 0; } -static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned vmid) { u32 data; @@ -5041,6 +5041,14 @@ static void gfx_v11_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) WREG32_SOC15_NO_KIQ(GC, 0, regRLC_SPM_MC_CNTL, data); amdgpu_gfx_off_ctrl(adev, true); + + if (ring + && amdgpu_sriov_is_pp_one_vf(adev) + && ((ring->funcs->type == AMDGPU_RING_TYPE_GFX) + || (ring->funcs->type == AMDGPU_RING_TYPE_COMPUTE))) { + uint32_t reg = SOC15_REG_OFFSET(GC, 0, regRLC_SPM_MC_CNTL); + amdgpu_ring_emit_wreg(ring, reg, data); + } } static const struct amdgpu_rlc_funcs gfx_v11_0_rlc_funcs = { @@ -6104,7 +6112,8 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = { .get_rptr = gfx_v11_0_ring_get_rptr_gfx, .get_wptr = gfx_v11_0_ring_get_wptr_gfx, .set_wptr = gfx_v11_0_ring_set_wptr_gfx, - .emit_frame_size = /* totally 242 maximum if 16 IBs */ + .emit_frame_size = /* totally 247 maximum if 16 IBs */ + 5 + /* update_spm_vmid */ 5 + /* COND_EXEC */ 9 + /* SET_Q_PREEMPTION_MODE */ 7 + /* PIPELINE_SYNC */ @@ -6154,6 +6163,7 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_compute = { .get_wptr = gfx_v11_0_ring_get_wptr_compute, .set_wptr = gfx_v11_0_ring_set_wptr_compute, .emit_frame_size = + 5 + /* update_spm_vmid */ 20 + /* gfx_v11_0_ring_emit_gds_switch */ 7 + /* gfx_v11_0_ring_emit_hdp_flush */ 5 + /* hdp invalidate */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c index 26d6286d86c9..9e7ce1e6bc06 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0_3.c @@ -69,7 +69,7 @@ static int gfx_v11_0_3_rlc_gc_fed_irq(struct amdgpu_device *adev, amdgpu_ras_interrupt_dispatch(adev, &ih_data); } else { if (adev->virt.ops && adev->virt.ops->ras_poison_handler) - adev->virt.ops->ras_poison_handler(adev); + adev->virt.ops->ras_poison_handler(adev, ras_if->block); else dev_warn(adev->dev, "No ras_poison_handler interface in SRIOV for %s!\n", ras_if->name); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index c2faf6b4c2fc..86a4865b1ae5 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -3274,7 +3274,7 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev) /* init spm vmid with 0xf */ if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + adev->gfx.rlc.funcs->update_spm_vmid(adev, NULL, 0xf); return 0; } @@ -3500,7 +3500,7 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev) return 0; } -static void gfx_v7_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +static void gfx_v7_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned vmid) { u32 data; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index 1943beb135c4..ea174b76ee70 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -1288,7 +1288,7 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev) /* init spm vmid with 0xf */ if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + adev->gfx.rlc.funcs->update_spm_vmid(adev, NULL, 0xf); return 0; } @@ -5579,7 +5579,7 @@ static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev, int xcc_id) } } -static void gfx_v8_0_update_spm_vmid(struct amdgpu_device *adev, unsigned vmid) +static void gfx_v8_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned vmid) { u32 data; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 69c500910746..169d45268ef6 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -3034,6 +3034,14 @@ static int gfx_v9_0_cp_gfx_start(struct amdgpu_device *adev) gfx_v9_0_cp_gfx_enable(adev, true); + /* Now only limit the quirk on the APU gfx9 series and already + * confirmed that the APU gfx10/gfx11 needn't such update. + */ + if (adev->flags & AMD_IS_APU && + adev->in_s3 && !adev->suspend_complete) { + DRM_INFO(" Will skip the CSB packet resubmit\n"); + return 0; + } r = amdgpu_ring_alloc(ring, gfx_v9_0_get_csb_size(adev) + 4 + 3); if (r) { DRM_ERROR("amdgpu: cp failed to lock ring (%d).\n", r); @@ -4894,7 +4902,7 @@ static void gfx_v9_0_update_spm_vmid_internal(struct amdgpu_device *adev, WREG32_SOC15(GC, 0, mmRLC_SPM_MC_CNTL, data); } -static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, unsigned int vmid) +static void gfx_v9_0_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned int vmid) { amdgpu_gfx_off_ctrl(adev, false); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c index bc8416afb62c..f53b379d8971 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4.c @@ -970,8 +970,9 @@ static void gfx_v9_4_reset_ras_error_count(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmATC_L2_CACHE_4K_DSM_INDEX, 255); } -static const struct soc15_reg_entry gfx_v9_4_ea_err_status_regs = - { SOC15_REG_ENTRY(GC, 0, mmGCEA_ERR_STATUS), 0, 1, 32 }; +static const struct soc15_reg_entry gfx_v9_4_ea_err_status_regs = { + SOC15_REG_ENTRY(GC, 0, mmGCEA_ERR_STATUS), 0, 1, 32 +}; static void gfx_v9_4_query_ras_error_status(struct amdgpu_device *adev) { 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 131cddbdda0d..aace4594a603 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -38,6 +38,7 @@ #include "gfx_v9_4_3.h" #include "amdgpu_xcp.h" +#include "amdgpu_aca.h" MODULE_FIRMWARE("amdgpu/gc_9_4_3_mec.bin"); MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin"); @@ -48,6 +49,10 @@ MODULE_FIRMWARE("amdgpu/gc_9_4_3_rlc.bin"); #define GOLDEN_GB_ADDR_CONFIG 0x2a114042 #define CP_HQD_PERSISTENT_STATE_DEFAULT 0xbe05301 +#define mmSMNAID_XCD0_MCA_SMU 0x36430400 /* SMN AID XCD0 */ +#define mmSMNAID_XCD1_MCA_SMU 0x38430400 /* SMN AID XCD1 */ +#define mmSMNXCD_XCD0_MCA_SMU 0x40430400 /* SMN XCD XCD0 */ + struct amdgpu_gfx_ras gfx_v9_4_3_ras; static void gfx_v9_4_3_set_ring_funcs(struct amdgpu_device *adev); @@ -675,6 +680,66 @@ static const struct amdgpu_gfx_funcs gfx_v9_4_3_gfx_funcs = { .ih_node_to_logical_xcc = &gfx_v9_4_3_ih_to_xcc_inst, }; +static int gfx_v9_4_3_aca_bank_generate_report(struct aca_handle *handle, + struct aca_bank *bank, enum aca_error_type type, + struct aca_bank_report *report, void *data) +{ + u64 status, misc0; + u32 instlo; + int ret; + + status = bank->regs[ACA_REG_IDX_STATUS]; + if ((type == ACA_ERROR_TYPE_UE && + ACA_REG__STATUS__ERRORCODEEXT(status) == ACA_EXTERROR_CODE_FAULT) || + (type == ACA_ERROR_TYPE_CE && + ACA_REG__STATUS__ERRORCODEEXT(status) == ACA_EXTERROR_CODE_CE)) { + + ret = aca_bank_info_decode(bank, &report->info); + if (ret) + return ret; + + /* NOTE: overwrite info.die_id with xcd id for gfx */ + instlo = ACA_REG__IPID__INSTANCEIDLO(bank->regs[ACA_REG_IDX_IPID]); + instlo &= GENMASK(31, 1); + report->info.die_id = instlo == mmSMNAID_XCD0_MCA_SMU ? 0 : 1; + + misc0 = bank->regs[ACA_REG_IDX_MISC0]; + report->count[type] = ACA_REG__MISC0__ERRCNT(misc0); + } + + return 0; +} + +static bool gfx_v9_4_3_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, + enum aca_error_type type, void *data) +{ + u32 instlo; + + instlo = ACA_REG__IPID__INSTANCEIDLO(bank->regs[ACA_REG_IDX_IPID]); + instlo &= GENMASK(31, 1); + switch (instlo) { + case mmSMNAID_XCD0_MCA_SMU: + case mmSMNAID_XCD1_MCA_SMU: + case mmSMNXCD_XCD0_MCA_SMU: + return true; + default: + break; + } + + return false; +} + +static const struct aca_bank_ops gfx_v9_4_3_aca_bank_ops = { + .aca_bank_generate_report = gfx_v9_4_3_aca_bank_generate_report, + .aca_bank_is_valid = gfx_v9_4_3_aca_bank_is_valid, +}; + +static const struct aca_info gfx_v9_4_3_aca_info = { + .hwip = ACA_HWIP_TYPE_SMU, + .mask = ACA_ERROR_UE_MASK | ACA_ERROR_CE_MASK, + .bank_ops = &gfx_v9_4_3_aca_bank_ops, +}; + static int gfx_v9_4_3_gpu_early_init(struct amdgpu_device *adev) { u32 gb_addr_config; @@ -1109,7 +1174,7 @@ static int gfx_v9_4_3_rlc_init(struct amdgpu_device *adev) { /* init spm vmid with 0xf */ if (adev->gfx.rlc.funcs->update_spm_vmid) - adev->gfx.rlc.funcs->update_spm_vmid(adev, 0xf); + adev->gfx.rlc.funcs->update_spm_vmid(adev, NULL, 0xf); return 0; } @@ -1320,7 +1385,7 @@ static int gfx_v9_4_3_rlc_resume(struct amdgpu_device *adev) return 0; } -static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, +static void gfx_v9_4_3_update_spm_vmid(struct amdgpu_device *adev, struct amdgpu_ring *ring, unsigned vmid) { u32 reg, data; @@ -4242,9 +4307,32 @@ struct amdgpu_ras_block_hw_ops gfx_v9_4_3_ras_ops = { .reset_ras_error_count = &gfx_v9_4_3_reset_ras_error_count, }; +static int gfx_v9_4_3_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) +{ + int r; + + r = amdgpu_ras_block_late_init(adev, ras_block); + if (r) + return r; + + r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__GFX, + &gfx_v9_4_3_aca_info, + NULL); + if (r) + goto late_fini; + + return 0; + +late_fini: + amdgpu_ras_block_late_fini(adev, ras_block); + + return r; +} + struct amdgpu_gfx_ras gfx_v9_4_3_ras = { .ras_block = { .hw_ops = &gfx_v9_4_3_ras_ops, + .ras_late_init = &gfx_v9_4_3_ras_late_init, }, .enable_watchdog_timer = &gfx_v9_4_3_enable_watchdog_timer, }; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 6c5185608854..db89d13bd80d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -262,16 +262,17 @@ static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, /* flush hdp cache */ adev->hdp.funcs->flush_hdp(adev, NULL); - /* For SRIOV run time, driver shouldn't access the register through MMIO - * Directly use kiq to do the vm invalidation instead + /* This is necessary for SRIOV as well as for GFXOFF to function + * properly under bare metal */ if (adev->gfx.kiq[0].ring.sched.ready && !adev->enable_mes && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) { - amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req, - 1 << vmid, GET_INST(GC, 0)); + amdgpu_gmc_fw_reg_write_reg_wait(adev, req, ack, inv_req, + 1 << vmid, GET_INST(GC, 0)); return; } + /* This path is needed before KIQ/MES/GFXOFF are set up */ hub_ip = (vmhub == AMDGPU_GFXHUB(0)) ? GC_HWIP : MMHUB_HWIP; spin_lock(&adev->gmc.invalidate_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index c9c653cfc765..6c68135cac9f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -223,16 +223,17 @@ static void gmc_v11_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, /* flush hdp cache */ adev->hdp.funcs->flush_hdp(adev, NULL); - /* For SRIOV run time, driver shouldn't access the register through MMIO - * Directly use kiq to do the vm invalidation instead + /* This is necessary for SRIOV as well as for GFXOFF to function + * properly under bare metal */ if ((adev->gfx.kiq[0].ring.sched.ready || adev->mes.ring.sched.ready) && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) { - amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req, - 1 << vmid, GET_INST(GC, 0)); + amdgpu_gmc_fw_reg_write_reg_wait(adev, req, ack, inv_req, + 1 << vmid, GET_INST(GC, 0)); return; } + /* This path is needed before KIQ/MES/GFXOFF are set up */ hub_ip = (vmhub == AMDGPU_GFXHUB(0)) ? GC_HWIP : MMHUB_HWIP; spin_lock(&adev->gmc.invalidate_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index 59d9215e5556..23b478639921 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -435,9 +435,10 @@ static void gmc_v6_0_set_prt(struct amdgpu_device *adev, bool enable) WREG32(mmVM_PRT_CNTL, tmp); if (enable) { - uint32_t low = AMDGPU_VA_RESERVED_SIZE >> AMDGPU_GPU_PAGE_SHIFT; + uint32_t low = AMDGPU_VA_RESERVED_BOTTOM >> + AMDGPU_GPU_PAGE_SHIFT; uint32_t high = adev->vm_manager.max_pfn - - (AMDGPU_VA_RESERVED_SIZE >> AMDGPU_GPU_PAGE_SHIFT); + (AMDGPU_VA_RESERVED_TOP >> AMDGPU_GPU_PAGE_SHIFT); WREG32(mmVM_PRT_APERTURE0_LOW_ADDR, low); WREG32(mmVM_PRT_APERTURE1_LOW_ADDR, low); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 45a2f8e031a2..3da7b6a2b00d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -563,9 +563,10 @@ static void gmc_v7_0_set_prt(struct amdgpu_device *adev, bool enable) WREG32(mmVM_PRT_CNTL, tmp); if (enable) { - uint32_t low = AMDGPU_VA_RESERVED_SIZE >> AMDGPU_GPU_PAGE_SHIFT; + uint32_t low = AMDGPU_VA_RESERVED_BOTTOM >> + AMDGPU_GPU_PAGE_SHIFT; uint32_t high = adev->vm_manager.max_pfn - - (AMDGPU_VA_RESERVED_SIZE >> AMDGPU_GPU_PAGE_SHIFT); + (AMDGPU_VA_RESERVED_TOP >> AMDGPU_GPU_PAGE_SHIFT); WREG32(mmVM_PRT_APERTURE0_LOW_ADDR, low); WREG32(mmVM_PRT_APERTURE1_LOW_ADDR, low); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 4422b27a3cc2..969a9e867170 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -777,9 +777,10 @@ static void gmc_v8_0_set_prt(struct amdgpu_device *adev, bool enable) WREG32(mmVM_PRT_CNTL, tmp); if (enable) { - uint32_t low = AMDGPU_VA_RESERVED_SIZE >> AMDGPU_GPU_PAGE_SHIFT; + uint32_t low = AMDGPU_VA_RESERVED_BOTTOM >> + AMDGPU_GPU_PAGE_SHIFT; uint32_t high = adev->vm_manager.max_pfn - - (AMDGPU_VA_RESERVED_SIZE >> AMDGPU_GPU_PAGE_SHIFT); + (AMDGPU_VA_RESERVED_TOP >> AMDGPU_GPU_PAGE_SHIFT); WREG32(mmVM_PRT_APERTURE0_LOW_ADDR, low); WREG32(mmVM_PRT_APERTURE1_LOW_ADDR, low); diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 40a00ea0009f..d442ae85162d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -829,23 +829,25 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid, req = hub->vm_inv_eng0_req + hub->eng_distance * eng; ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng; - /* This is necessary for a HW workaround under SRIOV as well - * as GFXOFF under bare metal - */ if (vmhub >= AMDGPU_MMHUB0(0)) inst = GET_INST(GC, 0); else inst = vmhub; + + /* This is necessary for SRIOV as well as for GFXOFF to function + * properly under bare metal + */ if (adev->gfx.kiq[inst].ring.sched.ready && (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) { uint32_t req = hub->vm_inv_eng0_req + hub->eng_distance * eng; uint32_t ack = hub->vm_inv_eng0_ack + hub->eng_distance * eng; - amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, inv_req, - 1 << vmid, inst); + amdgpu_gmc_fw_reg_write_reg_wait(adev, req, ack, inv_req, + 1 << vmid, inst); return; } + /* This path is needed before KIQ/MES/GFXOFF are set up */ spin_lock(&adev->gmc.invalidate_lock); /* @@ -1947,14 +1949,6 @@ static int gmc_v9_0_init_mem_ranges(struct amdgpu_device *adev) static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev) { - static const u32 regBIF_BIOS_SCRATCH_4 = 0x50; - u32 vram_info; - - /* Only for dGPU, vendor informaton is reliable */ - if (!amdgpu_sriov_vf(adev) && !(adev->flags & AMD_IS_APU)) { - 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; } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index e67a337457ed..99cd49ee8ef6 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -551,7 +551,7 @@ static int jpeg_v2_5_set_powergating_state(void *handle, struct amdgpu_device *adev = (struct amdgpu_device *)handle; int ret; - if(state == adev->jpeg.cur_state) + if (state == adev->jpeg.cur_state) return 0; if (state == AMD_PG_STATE_GATE) @@ -559,7 +559,7 @@ static int jpeg_v2_5_set_powergating_state(void *handle, else ret = jpeg_v2_5_start(adev); - if(!ret) + if (!ret) adev->jpeg.cur_state = state; return ret; @@ -754,8 +754,7 @@ static void jpeg_v2_5_set_irq_funcs(struct amdgpu_device *adev) } } -const struct amdgpu_ip_block_version jpeg_v2_5_ip_block = -{ +const struct amdgpu_ip_block_version jpeg_v2_5_ip_block = { .type = AMD_IP_BLOCK_TYPE_JPEG, .major = 2, .minor = 5, @@ -763,8 +762,7 @@ const struct amdgpu_ip_block_version jpeg_v2_5_ip_block = .funcs = &jpeg_v2_5_ip_funcs, }; -const struct amdgpu_ip_block_version jpeg_v2_6_ip_block = -{ +const struct amdgpu_ip_block_version jpeg_v2_6_ip_block = { .type = AMD_IP_BLOCK_TYPE_JPEG, .major = 2, .minor = 6, diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index bc38b90f8cf8..88ea58d5c4ab 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -674,14 +674,6 @@ static int jpeg_v4_0_set_powergating_state(void *handle, return ret; } -static int jpeg_v4_0_set_interrupt_state(struct amdgpu_device *adev, - struct amdgpu_irq_src *source, - unsigned type, - enum amdgpu_interrupt_state state) -{ - return 0; -} - static int jpeg_v4_0_set_ras_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, unsigned int type, @@ -765,7 +757,6 @@ static void jpeg_v4_0_set_dec_ring_funcs(struct amdgpu_device *adev) } static const struct amdgpu_irq_src_funcs jpeg_v4_0_irq_funcs = { - .set = jpeg_v4_0_set_interrupt_state, .process = jpeg_v4_0_process_interrupt, }; diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c index 6ede85b28cc8..78b74daf4eeb 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c @@ -181,7 +181,6 @@ static int jpeg_v4_0_5_hw_fini(void *handle) RREG32_SOC15(JPEG, 0, regUVD_JRBC_STATUS)) jpeg_v4_0_5_set_powergating_state(adev, AMD_PG_STATE_GATE); } - amdgpu_irq_put(adev, &adev->jpeg.inst->irq, 0); return 0; } @@ -516,14 +515,6 @@ static int jpeg_v4_0_5_set_powergating_state(void *handle, return ret; } -static int jpeg_v4_0_5_set_interrupt_state(struct amdgpu_device *adev, - struct amdgpu_irq_src *source, - unsigned type, - enum amdgpu_interrupt_state state) -{ - return 0; -} - static int jpeg_v4_0_5_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -603,7 +594,6 @@ static void jpeg_v4_0_5_set_dec_ring_funcs(struct amdgpu_device *adev) } static const struct amdgpu_irq_src_funcs jpeg_v4_0_5_irq_funcs = { - .set = jpeg_v4_0_5_set_interrupt_state, .process = jpeg_v4_0_5_process_interrupt, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c index fb53aacdcba2..c0fc44cdd658 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_8.c @@ -33,6 +33,7 @@ #define regVM_L2_CNTL3_DEFAULT 0x80100007 #define regVM_L2_CNTL4_DEFAULT 0x000000c1 +#define mmSMNAID_AID0_MCA_SMU 0x03b30400 static u64 mmhub_v1_8_get_fb_location(struct amdgpu_device *adev) { @@ -705,8 +706,94 @@ static const struct amdgpu_ras_block_hw_ops mmhub_v1_8_ras_hw_ops = { .reset_ras_error_count = mmhub_v1_8_reset_ras_error_count, }; +static int mmhub_v1_8_aca_bank_generate_report(struct aca_handle *handle, + struct aca_bank *bank, enum aca_error_type type, + struct aca_bank_report *report, void *data) +{ + u64 status, misc0; + int ret; + + status = bank->regs[ACA_REG_IDX_STATUS]; + if ((type == ACA_ERROR_TYPE_UE && + ACA_REG__STATUS__ERRORCODEEXT(status) == ACA_EXTERROR_CODE_FAULT) || + (type == ACA_ERROR_TYPE_CE && + ACA_REG__STATUS__ERRORCODEEXT(status) == ACA_EXTERROR_CODE_CE)) { + + ret = aca_bank_info_decode(bank, &report->info); + if (ret) + return ret; + + misc0 = bank->regs[ACA_REG_IDX_MISC0]; + report->count[type] = ACA_REG__MISC0__ERRCNT(misc0); + } + + return 0; +} + +/* reference to smu driver if header file */ +static int mmhub_v1_8_err_codes[] = { + 0, 1, 2, 3, 4, /* CODE_DAGB0 - 4 */ + 5, 6, 7, 8, 9, /* CODE_EA0 - 4 */ + 10, /* CODE_UTCL2_ROUTER */ + 11, /* CODE_VML2 */ + 12, /* CODE_VML2_WALKER */ + 13, /* CODE_MMCANE */ +}; + +static bool mmhub_v1_8_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, + enum aca_error_type type, void *data) +{ + u32 instlo; + + instlo = ACA_REG__IPID__INSTANCEIDLO(bank->regs[ACA_REG_IDX_IPID]); + instlo &= GENMASK(31, 1); + + if (instlo != mmSMNAID_AID0_MCA_SMU) + return false; + + if (aca_bank_check_error_codes(handle->adev, bank, + mmhub_v1_8_err_codes, + ARRAY_SIZE(mmhub_v1_8_err_codes))) + return false; + + return true; +} + +static const struct aca_bank_ops mmhub_v1_8_aca_bank_ops = { + .aca_bank_generate_report = mmhub_v1_8_aca_bank_generate_report, + .aca_bank_is_valid = mmhub_v1_8_aca_bank_is_valid, +}; + +static const struct aca_info mmhub_v1_8_aca_info = { + .hwip = ACA_HWIP_TYPE_SMU, + .mask = ACA_ERROR_UE_MASK, + .bank_ops = &mmhub_v1_8_aca_bank_ops, +}; + +static int mmhub_v1_8_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) +{ + int r; + + r = amdgpu_ras_block_late_init(adev, ras_block); + if (r) + return r; + + r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__MMHUB, + &mmhub_v1_8_aca_info, NULL); + if (r) + goto late_fini; + + return 0; + +late_fini: + amdgpu_ras_block_late_fini(adev, ras_block); + + return r; +} + struct amdgpu_mmhub_ras mmhub_v1_8_ras = { .ras_block = { .hw_ops = &mmhub_v1_8_ras_hw_ops, + .ras_late_init = mmhub_v1_8_ras_late_init, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index 63725b2ebc03..a2bd2c3b1ef9 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -404,7 +404,8 @@ static int xgpu_ai_request_init_data(struct amdgpu_device *adev) return xgpu_ai_send_access_requests(adev, IDH_REQ_GPU_INIT_DATA); } -static void xgpu_ai_ras_poison_handler(struct amdgpu_device *adev) +static void xgpu_ai_ras_poison_handler(struct amdgpu_device *adev, + enum amdgpu_ras_block block) { xgpu_ai_send_access_requests(adev, IDH_RAS_POISON); } diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index 6a68ee946f1c..77f5b55decf9 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -152,14 +152,14 @@ static void xgpu_nv_mailbox_trans_msg (struct amdgpu_device *adev, xgpu_nv_mailbox_set_valid(adev, false); } -static int xgpu_nv_send_access_requests(struct amdgpu_device *adev, - enum idh_request req) +static int xgpu_nv_send_access_requests_with_param(struct amdgpu_device *adev, + enum idh_request req, u32 data1, u32 data2, u32 data3) { int r, retry = 1; enum idh_event event = -1; send_request: - xgpu_nv_mailbox_trans_msg(adev, req, 0, 0, 0); + xgpu_nv_mailbox_trans_msg(adev, req, data1, data2, data3); switch (req) { case IDH_REQ_GPU_INIT_ACCESS: @@ -170,6 +170,10 @@ send_request: case IDH_REQ_GPU_INIT_DATA: event = IDH_REQ_GPU_INIT_DATA_READY; break; + case IDH_RAS_POISON: + if (data1 != 0) + event = IDH_RAS_POISON_READY; + break; default: break; } @@ -206,6 +210,13 @@ send_request: return 0; } +static int xgpu_nv_send_access_requests(struct amdgpu_device *adev, + enum idh_request req) +{ + return xgpu_nv_send_access_requests_with_param(adev, + req, 0, 0, 0); +} + static int xgpu_nv_request_reset(struct amdgpu_device *adev) { int ret, i = 0; @@ -424,9 +435,17 @@ void xgpu_nv_mailbox_put_irq(struct amdgpu_device *adev) amdgpu_irq_put(adev, &adev->virt.rcv_irq, 0); } -static void xgpu_nv_ras_poison_handler(struct amdgpu_device *adev) +static void xgpu_nv_ras_poison_handler(struct amdgpu_device *adev, + enum amdgpu_ras_block block) { - xgpu_nv_send_access_requests(adev, IDH_RAS_POISON); + if (amdgpu_ip_version(adev, UMC_HWIP, 0) < IP_VERSION(12, 0, 0)) { + xgpu_nv_send_access_requests(adev, IDH_RAS_POISON); + } else { + amdgpu_virt_fini_data_exchange(adev); + xgpu_nv_send_access_requests_with_param(adev, + IDH_RAS_POISON, block, 0, 0); + amdgpu_virt_init_data_exchange(adev); + } } const struct amdgpu_virt_ops xgpu_nv_virt_ops = { diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h index d0221ce08769..1e8fd90cab43 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h @@ -51,6 +51,7 @@ enum idh_event { IDH_FAIL, IDH_QUERY_ALIVE, IDH_REQ_GPU_INIT_DATA_READY, + IDH_RAS_POISON_READY, IDH_TEXT_MESSAGE = 255, }; diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index de93614726c9..4178f4e5dad7 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -728,8 +728,7 @@ static void navi10_ih_set_interrupt_funcs(struct amdgpu_device *adev) adev->irq.ih_funcs = &navi10_ih_funcs; } -const struct amdgpu_ip_block_version navi10_ih_ip_block = -{ +const struct amdgpu_ip_block_version navi10_ih_ip_block = { .type = AMD_IP_BLOCK_TYPE_IH, .major = 5, .minor = 0, diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index e90f33780803..b4723d68eab0 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -431,6 +431,12 @@ static void nbio_v7_9_init_registers(struct amdgpu_device *adev) u32 inst_mask; int i; + 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; WREG32_SOC15(NBIO, 0, regXCC_DOORBELL_FENCE, 0xff & ~(adev->gfx.xcc_mask)); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index df1844d0800f..722b6066ce07 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -27,6 +27,7 @@ #include "amdgpu_ucode.h" #include "soc15_common.h" #include "psp_v13_0.h" +#include "amdgpu_ras.h" #include "mp/mp_13_0_2_offset.h" #include "mp/mp_13_0_2_sh_mask.h" @@ -187,11 +188,18 @@ static int psp_v13_0_wait_for_bootloader(struct psp_context *psp) static int psp_v13_0_wait_for_bootloader_steady_state(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; + int ret; if (amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 6)) { - psp_v13_0_wait_for_vmbx_ready(psp); + ret = psp_v13_0_wait_for_vmbx_ready(psp); + if (ret) + amdgpu_ras_query_boot_status(adev, 4); + + ret = psp_v13_0_wait_for_bootloader(psp); + if (ret) + amdgpu_ras_query_boot_status(adev, 4); - return psp_v13_0_wait_for_bootloader(psp); + return ret; } return 0; @@ -763,81 +771,28 @@ static int psp_v13_0_fatal_error_recovery_quirk(struct psp_context *psp) return 0; } - -static void psp_v13_0_boot_error_reporting(struct amdgpu_device *adev, - uint32_t inst, - uint32_t boot_error) -{ - uint32_t socket_id; - uint32_t aid_id; - uint32_t hbm_id; - uint32_t reg_data; - - socket_id = REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, SOCKET_ID); - aid_id = REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, AID_ID); - hbm_id = REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, HBM_ID); - - reg_data = RREG32_SOC15(MP0, inst, regMP0_SMN_C2PMSG_109); - dev_info(adev->dev, "socket: %d, aid: %d, firmware boot failed, fw status is 0x%x\n", - socket_id, aid_id, reg_data); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_MEM_TRAINING)) - dev_info(adev->dev, "socket: %d, aid: %d, hbm: %d, memory training failed\n", - socket_id, aid_id, hbm_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_FW_LOAD)) - dev_info(adev->dev, "socket: %d, aid: %d, firmware load failed at boot time\n", - socket_id, aid_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_WAFL_LINK_TRAINING)) - dev_info(adev->dev, "socket: %d, aid: %d, wafl link training failed\n", - socket_id, aid_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_XGMI_LINK_TRAINING)) - dev_info(adev->dev, "socket: %d, aid: %d, xgmi link training failed\n", - socket_id, aid_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_USR_CP_LINK_TRAINING)) - dev_info(adev->dev, "socket: %d, aid: %d, usr cp link training failed\n", - socket_id, aid_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_USR_DP_LINK_TRAINING)) - dev_info(adev->dev, "socket: %d, aid: %d, usr dp link training failed\n", - socket_id, aid_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_HBM_MEM_TEST)) - dev_info(adev->dev, "socket: %d, aid: %d, hbm: %d, hbm memory test failed\n", - socket_id, aid_id, hbm_id); - - if (REG_GET_FIELD(boot_error, MP0_SMN_C2PMSG_126, GPU_ERR_HBM_BIST_TEST)) - dev_info(adev->dev, "socket: %d, aid: %d, hbm: %d, hbm bist test failed\n", - socket_id, aid_id, hbm_id); -} - -static int psp_v13_0_query_boot_status(struct psp_context *psp) +static bool psp_v13_0_get_ras_capability(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; - int inst_mask = adev->aid_mask; - uint32_t reg_data; - uint32_t i; - int ret = 0; + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + u32 reg_data; - if (amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(13, 0, 6)) - return 0; + /* query ras cap should be done from host side */ + if (amdgpu_sriov_vf(adev)) + return false; - if (RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_59) < 0x00a10109) - return 0; + if (!con) + return false; - for_each_inst(i, inst_mask) { - reg_data = RREG32_SOC15(MP0, i, regMP0_SMN_C2PMSG_126); - if (!REG_GET_FIELD(reg_data, MP0_SMN_C2PMSG_126, BOOT_STATUS)) { - psp_v13_0_boot_error_reporting(adev, i, reg_data); - ret = -EINVAL; - break; - } + if ((amdgpu_ip_version(adev, MP0_HWIP, 0) == IP_VERSION(13, 0, 6)) && + (!(adev->flags & AMD_IS_APU))) { + reg_data = RREG32_SOC15(MP0, 0, regMP0_SMN_C2PMSG_127); + adev->ras_hw_enabled = (reg_data & GENMASK_ULL(23, 0)); + con->poison_supported = ((reg_data & GENMASK_ULL(24, 24)) >> 24) ? true : false; + return true; + } else { + return false; } - - return ret; } static const struct psp_funcs psp_v13_0_funcs = { @@ -862,7 +817,7 @@ static const struct psp_funcs psp_v13_0_funcs = { .update_spirom = psp_v13_0_update_spirom, .vbflash_stat = psp_v13_0_vbflash_status, .fatal_error_recovery_quirk = psp_v13_0_fatal_error_recovery_quirk, - .query_boot_status = psp_v13_0_query_boot_status, + .get_ras_capability = psp_v13_0_get_ras_capability, }; void psp_v13_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index 8d5d86675a7f..07e19caf2bc1 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -57,22 +57,19 @@ static void sdma_v2_4_set_irq_funcs(struct amdgpu_device *adev); MODULE_FIRMWARE("amdgpu/topaz_sdma.bin"); MODULE_FIRMWARE("amdgpu/topaz_sdma1.bin"); -static const u32 sdma_offsets[SDMA_MAX_INSTANCE] = -{ +static const u32 sdma_offsets[SDMA_MAX_INSTANCE] = { SDMA0_REGISTER_OFFSET, SDMA1_REGISTER_OFFSET }; -static const u32 golden_settings_iceland_a11[] = -{ +static const u32 golden_settings_iceland_a11[] = { mmSDMA0_CHICKEN_BITS, 0xfc910007, 0x00810007, mmSDMA0_CLK_CTRL, 0xff000fff, 0x00000000, mmSDMA1_CHICKEN_BITS, 0xfc910007, 0x00810007, mmSDMA1_CLK_CTRL, 0xff000fff, 0x00000000, }; -static const u32 iceland_mgcg_cgcg_init[] = -{ +static const u32 iceland_mgcg_cgcg_init[] = { mmSDMA0_CLK_CTRL, 0xff000ff0, 0x00000100, mmSDMA1_CLK_CTRL, 0xff000ff0, 0x00000100 }; @@ -142,7 +139,8 @@ static int sdma_v2_4_init_microcode(struct amdgpu_device *adev) case CHIP_TOPAZ: chip_name = "topaz"; break; - default: BUG(); + default: + BUG(); } for (i = 0; i < adev->sdma.num_instances; i++) { @@ -1258,8 +1256,7 @@ static void sdma_v2_4_set_vm_pte_funcs(struct amdgpu_device *adev) adev->vm_manager.vm_pte_num_scheds = adev->sdma.num_instances; } -const struct amdgpu_ip_block_version sdma_v2_4_ip_block = -{ +const struct amdgpu_ip_block_version sdma_v2_4_ip_block = { .type = AMD_IP_BLOCK_TYPE_SDMA, .major = 2, .minor = 4, diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index 2d688dca26be..fec5a3d1c4bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -45,6 +45,8 @@ MODULE_FIRMWARE("amdgpu/sdma_4_4_2.bin"); +#define mmSMNAID_AID0_MCA_SMU 0x03b30400 + #define WREG32_SDMA(instance, offset, value) \ WREG32(sdma_v4_4_2_get_reg_offset(adev, (instance), (offset)), value) #define RREG32_SDMA(instance, offset) \ @@ -2204,9 +2206,79 @@ static const struct amdgpu_ras_block_hw_ops sdma_v4_4_2_ras_hw_ops = { .reset_ras_error_count = sdma_v4_4_2_reset_ras_error_count, }; +static int sdma_v4_4_2_aca_bank_generate_report(struct aca_handle *handle, + struct aca_bank *bank, enum aca_error_type type, + struct aca_bank_report *report, void *data) +{ + u64 status, misc0; + int ret; + + status = bank->regs[ACA_REG_IDX_STATUS]; + if ((type == ACA_ERROR_TYPE_UE && + ACA_REG__STATUS__ERRORCODEEXT(status) == ACA_EXTERROR_CODE_FAULT) || + (type == ACA_ERROR_TYPE_CE && + ACA_REG__STATUS__ERRORCODEEXT(status) == ACA_EXTERROR_CODE_CE)) { + + ret = aca_bank_info_decode(bank, &report->info); + if (ret) + return ret; + + misc0 = bank->regs[ACA_REG_IDX_MISC0]; + report->count[type] = ACA_REG__MISC0__ERRCNT(misc0); + } + + return 0; +} + +/* CODE_SDMA0 - CODE_SDMA4, reference to smu driver if header file */ +static int sdma_v4_4_2_err_codes[] = { 33, 34, 35, 36 }; + +static bool sdma_v4_4_2_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, + enum aca_error_type type, void *data) +{ + u32 instlo; + + instlo = ACA_REG__IPID__INSTANCEIDLO(bank->regs[ACA_REG_IDX_IPID]); + instlo &= GENMASK(31, 1); + + if (instlo != mmSMNAID_AID0_MCA_SMU) + return false; + + if (aca_bank_check_error_codes(handle->adev, bank, + sdma_v4_4_2_err_codes, + ARRAY_SIZE(sdma_v4_4_2_err_codes))) + return false; + + return true; +} + +static const struct aca_bank_ops sdma_v4_4_2_aca_bank_ops = { + .aca_bank_generate_report = sdma_v4_4_2_aca_bank_generate_report, + .aca_bank_is_valid = sdma_v4_4_2_aca_bank_is_valid, +}; + +static const struct aca_info sdma_v4_4_2_aca_info = { + .hwip = ACA_HWIP_TYPE_SMU, + .mask = ACA_ERROR_UE_MASK, + .bank_ops = &sdma_v4_4_2_aca_bank_ops, +}; + +static int sdma_v4_4_2_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) +{ + int r; + + r = amdgpu_sdma_ras_late_init(adev, ras_block); + if (r) + return r; + + return amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__SDMA, + &sdma_v4_4_2_aca_info, NULL); +} + static struct amdgpu_sdma_ras sdma_v4_4_2_ras = { .ras_block = { .hw_ops = &sdma_v4_4_2_ras_hw_ops, + .ras_late_init = sdma_v4_4_2_ras_late_init, }, }; diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 15033efec2ba..c64c01e2944a 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -1298,10 +1298,32 @@ static int soc15_common_suspend(void *handle) return soc15_common_hw_fini(adev); } +static bool soc15_need_reset_on_resume(struct amdgpu_device *adev) +{ + u32 sol_reg; + + sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); + + /* Will reset for the following suspend abort cases. + * 1) Only reset limit on APU side, dGPU hasn't checked yet. + * 2) S3 suspend abort and TOS already launched. + */ + if (adev->flags & AMD_IS_APU && adev->in_s3 && + !adev->suspend_complete && + sol_reg) + return true; + + return false; +} + static int soc15_common_resume(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (soc15_need_reset_on_resume(adev)) { + dev_info(adev->dev, "S3 suspend abort case, let's reset ASIC.\n"); + soc15_asic_reset(adev); + } return soc15_common_hw_init(adev); } diff --git a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h index 879bb7af297c..056d4df8fa1f 100644 --- a/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h +++ b/drivers/gpu/drm/amd/amdgpu/ta_ras_if.h @@ -36,6 +36,9 @@ enum ras_command { TA_RAS_COMMAND__ENABLE_FEATURES = 0, TA_RAS_COMMAND__DISABLE_FEATURES, TA_RAS_COMMAND__TRIGGER_ERROR, + TA_RAS_COMMAND__QUERY_BLOCK_INFO, + TA_RAS_COMMAND__QUERY_SUB_BLOCK_INFO, + TA_RAS_COMMAND__QUERY_ADDRESS, }; enum ta_ras_status { @@ -105,6 +108,11 @@ enum ta_ras_error_type { TA_RAS_ERROR__POISON = 8, }; +enum ta_ras_address_type { + TA_RAS_MCA_TO_PA, + TA_RAS_PA_TO_MCA, +}; + /* Input/output structures for RAS commands */ /**********************************************************/ @@ -133,12 +141,38 @@ struct ta_ras_init_flags { uint8_t channel_dis_num; }; +struct ta_ras_mca_addr { + uint64_t err_addr; + uint32_t ch_inst; + uint32_t umc_inst; + uint32_t node_inst; +}; + +struct ta_ras_phy_addr { + uint64_t pa; + uint32_t bank; + uint32_t channel_idx; +}; + +struct ta_ras_query_address_input { + enum ta_ras_address_type addr_type; + struct ta_ras_mca_addr ma; + struct ta_ras_phy_addr pa; +}; + struct ta_ras_output_flags { uint8_t ras_init_success_flag; uint8_t err_inject_switch_disable_flag; uint8_t reg_access_failure_flag; }; +struct ta_ras_query_address_output { + /* don't use the flags here */ + struct ta_ras_output_flags flags; + struct ta_ras_mca_addr ma; + struct ta_ras_phy_addr pa; +}; + /* Common input structure for RAS callbacks */ /**********************************************************/ union ta_ras_cmd_input { @@ -146,12 +180,14 @@ union ta_ras_cmd_input { struct ta_ras_enable_features_input enable_features; struct ta_ras_disable_features_input disable_features; struct ta_ras_trigger_error_input trigger_error; + struct ta_ras_query_address_input address; uint32_t reserve_pad[256]; }; union ta_ras_cmd_output { struct ta_ras_output_flags flags; + struct ta_ras_query_address_output address; uint32_t reserve_pad[256]; }; diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c index 7458a218e89d..14ef7a24be7b 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c @@ -89,12 +89,28 @@ static void umc_v12_0_reset_error_count(struct amdgpu_device *adev) umc_v12_0_reset_error_count_per_channel, NULL); } +bool umc_v12_0_is_deferred_error(struct amdgpu_device *adev, uint64_t mc_umc_status) +{ + dev_info(adev->dev, + "MCA_UMC_STATUS(0x%llx): Val:%llu, Poison:%llu, Deferred:%llu, PCC:%llu, UC:%llu, TCC:%llu\n", + mc_umc_status, + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val), + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Poison), + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred), + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC), + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC), + REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, TCC) + ); + + return (amdgpu_ras_is_poison_mode_supported(adev) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && + (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1)); +} + bool umc_v12_0_is_uncorrectable_error(struct amdgpu_device *adev, uint64_t mc_umc_status) { - if (amdgpu_ras_is_poison_mode_supported(adev) && - (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && - (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1)) - return true; + if (umc_v12_0_is_deferred_error(adev, mc_umc_status)) + return false; return ((REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, PCC) == 1 || @@ -104,9 +120,7 @@ bool umc_v12_0_is_uncorrectable_error(struct amdgpu_device *adev, uint64_t mc_um bool umc_v12_0_is_correctable_error(struct amdgpu_device *adev, uint64_t mc_umc_status) { - if (amdgpu_ras_is_poison_mode_supported(adev) && - (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1) && - (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Deferred) == 1)) + if (umc_v12_0_is_deferred_error(adev, mc_umc_status)) return false; return (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && @@ -119,9 +133,10 @@ bool umc_v12_0_is_correctable_error(struct amdgpu_device *adev, uint64_t mc_umc_ !(umc_v12_0_is_uncorrectable_error(adev, mc_umc_status))))); } -static void umc_v12_0_query_correctable_error_count(struct amdgpu_device *adev, +static void umc_v12_0_query_error_count_per_type(struct amdgpu_device *adev, uint64_t umc_reg_offset, - unsigned long *error_count) + unsigned long *error_count, + check_error_type_func error_type_func) { uint64_t mc_umc_status; uint64_t mc_umc_status_addr; @@ -129,31 +144,11 @@ static void umc_v12_0_query_correctable_error_count(struct amdgpu_device *adev, mc_umc_status_addr = SOC15_REG_OFFSET(UMC, 0, regMCA_UMC_UMC0_MCUMC_STATUST0); - /* Rely on MCUMC_STATUS for correctable error counter - * MCUMC_STATUS is a 64 bit register - */ + /* Check MCUMC_STATUS */ mc_umc_status = RREG64_PCIE_EXT((mc_umc_status_addr + umc_reg_offset) * 4); - if (umc_v12_0_is_correctable_error(adev, mc_umc_status)) - *error_count += 1; -} - -static void umc_v12_0_query_uncorrectable_error_count(struct amdgpu_device *adev, - uint64_t umc_reg_offset, - unsigned long *error_count) -{ - uint64_t mc_umc_status; - uint64_t mc_umc_status_addr; - - mc_umc_status_addr = - SOC15_REG_OFFSET(UMC, 0, regMCA_UMC_UMC0_MCUMC_STATUST0); - - /* Check the MCUMC_STATUS. */ - mc_umc_status = - RREG64_PCIE_EXT((mc_umc_status_addr + umc_reg_offset) * 4); - - if (umc_v12_0_is_uncorrectable_error(adev, mc_umc_status)) + if (error_type_func(adev, mc_umc_status)) *error_count += 1; } @@ -162,7 +157,7 @@ static int umc_v12_0_query_error_count(struct amdgpu_device *adev, uint32_t ch_inst, void *data) { struct ras_err_data *err_data = (struct ras_err_data *)data; - unsigned long ue_count = 0, ce_count = 0; + unsigned long ue_count = 0, ce_count = 0, de_count = 0; /* NOTE: node_inst is converted by adev->umc.active_mask and the range is [0-3], * which can be used as die ID directly */ @@ -174,11 +169,16 @@ static int umc_v12_0_query_error_count(struct amdgpu_device *adev, uint64_t umc_reg_offset = get_umc_v12_0_reg_offset(adev, node_inst, umc_inst, ch_inst); - umc_v12_0_query_correctable_error_count(adev, umc_reg_offset, &ce_count); - umc_v12_0_query_uncorrectable_error_count(adev, umc_reg_offset, &ue_count); + umc_v12_0_query_error_count_per_type(adev, umc_reg_offset, + &ce_count, umc_v12_0_is_correctable_error); + umc_v12_0_query_error_count_per_type(adev, umc_reg_offset, + &ue_count, umc_v12_0_is_uncorrectable_error); + umc_v12_0_query_error_count_per_type(adev, umc_reg_offset, + &de_count, umc_v12_0_is_deferred_error); amdgpu_ras_error_statistic_ue_count(err_data, &mcm_info, NULL, ue_count); amdgpu_ras_error_statistic_ce_count(err_data, &mcm_info, NULL, ce_count); + amdgpu_ras_error_statistic_de_count(err_data, &mcm_info, NULL, de_count); return 0; } @@ -203,14 +203,14 @@ static bool umc_v12_0_bit_wise_xor(uint32_t val) return result; } -static void umc_v12_0_convert_error_address(struct amdgpu_device *adev, - struct ras_err_data *err_data, uint64_t err_addr, - uint32_t ch_inst, uint32_t umc_inst, - uint32_t node_inst) +static void umc_v12_0_mca_addr_to_pa(struct amdgpu_device *adev, + uint64_t err_addr, uint32_t ch_inst, uint32_t umc_inst, + uint32_t node_inst, + struct ta_ras_query_address_output *addr_out) { uint32_t channel_index, i; - uint64_t soc_pa, na, retired_page, column; - uint32_t bank_hash0, bank_hash1, bank_hash2, bank_hash3, col, row, row_xor; + uint64_t na, soc_pa; + uint32_t bank_hash0, bank_hash1, bank_hash2, bank_hash3, col, row; uint32_t bank0, bank1, bank2, bank3, bank; bank_hash0 = (err_addr >> UMC_V12_0_MCA_B0_BIT) & 0x1ULL; @@ -260,12 +260,44 @@ static void umc_v12_0_convert_error_address(struct amdgpu_device *adev, /* the umc channel bits are not original values, they are hashed */ UMC_V12_0_SET_CHANNEL_HASH(channel_index, soc_pa); + addr_out->pa.pa = soc_pa; + addr_out->pa.bank = bank; + addr_out->pa.channel_idx = channel_index; +} + +static void umc_v12_0_convert_error_address(struct amdgpu_device *adev, + struct ras_err_data *err_data, uint64_t err_addr, + uint32_t ch_inst, uint32_t umc_inst, + uint32_t node_inst) +{ + uint32_t col, row, row_xor, bank, channel_index; + uint64_t soc_pa, retired_page, column; + struct ta_ras_query_address_input addr_in; + struct ta_ras_query_address_output addr_out; + + addr_in.addr_type = TA_RAS_MCA_TO_PA; + addr_in.ma.err_addr = err_addr; + addr_in.ma.ch_inst = ch_inst; + addr_in.ma.umc_inst = umc_inst; + addr_in.ma.node_inst = node_inst; + + if (psp_ras_query_address(&adev->psp, &addr_in, &addr_out)) + /* fallback to old path if fail to get pa from psp */ + umc_v12_0_mca_addr_to_pa(adev, err_addr, ch_inst, umc_inst, + node_inst, &addr_out); + + soc_pa = addr_out.pa.pa; + bank = addr_out.pa.bank; + channel_index = addr_out.pa.channel_idx; + + col = (err_addr >> 1) & 0x1fULL; + row = (err_addr >> 10) & 0x3fffULL; + row_xor = row ^ (0x1ULL << 13); /* clear [C3 C2] in soc physical address */ soc_pa &= ~(0x3ULL << UMC_V12_0_PA_C2_BIT); /* clear [C4] in soc physical address */ soc_pa &= ~(0x1ULL << UMC_V12_0_PA_C4_BIT); - row_xor = row ^ (0x1ULL << 13); /* loop for all possibilities of [C4 C3 C2] */ for (column = 0; column < UMC_V12_0_NA_MAP_PA_NUM; column++) { retired_page = soc_pa | ((column & 0x3) << UMC_V12_0_PA_C2_BIT); @@ -316,10 +348,7 @@ static int umc_v12_0_query_error_address(struct amdgpu_device *adev, } /* calculate error address if ue error is detected */ - if (REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, Val) == 1 && - REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, AddrV) == 1 && - REG_GET_FIELD(mc_umc_status, MCA_UMC_UMC0_MCUMC_STATUST0, UC) == 1) { - + if (umc_v12_0_is_uncorrectable_error(adev, mc_umc_status)) { mc_umc_addrt0 = SOC15_REG_OFFSET(UMC, 0, regMCA_UMC_UMC0_MCUMC_ADDRT0); @@ -385,45 +414,69 @@ static void umc_v12_0_ecc_info_query_ras_error_address(struct amdgpu_device *ade { struct ras_err_node *err_node; uint64_t mc_umc_status; + struct ras_err_info *err_info; + struct ras_err_addr *mca_err_addr, *tmp; struct ras_err_data *err_data = (struct ras_err_data *)ras_error_status; for_each_ras_error(err_node, err_data) { - mc_umc_status = err_node->err_info.err_addr.err_status; - if (!mc_umc_status) + err_info = &err_node->err_info; + if (list_empty(&err_info->err_addr_list)) continue; - if (umc_v12_0_is_uncorrectable_error(adev, mc_umc_status)) { - uint64_t mca_addr, err_addr, mca_ipid; - uint32_t InstanceIdLo; - struct amdgpu_smuio_mcm_config_info *mcm_info; - - mcm_info = &err_node->err_info.mcm_info; - mca_addr = err_node->err_info.err_addr.err_addr; - mca_ipid = err_node->err_info.err_addr.err_ipid; - - err_addr = REG_GET_FIELD(mca_addr, MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); - InstanceIdLo = REG_GET_FIELD(mca_ipid, MCMP1_IPIDT0, InstanceIdLo); - - dev_info(adev->dev, "UMC:IPID:0x%llx, aid:%d, inst:%d, ch:%d, err_addr:0x%llx\n", - mca_ipid, - mcm_info->die_id, - MCA_IPID_LO_2_UMC_INST(InstanceIdLo), - MCA_IPID_LO_2_UMC_CH(InstanceIdLo), - err_addr); - - umc_v12_0_convert_error_address(adev, - err_data, err_addr, - MCA_IPID_LO_2_UMC_CH(InstanceIdLo), - MCA_IPID_LO_2_UMC_INST(InstanceIdLo), - mcm_info->die_id); - - /* Clear umc error address content */ - memset(&err_node->err_info.err_addr, - 0, sizeof(err_node->err_info.err_addr)); + list_for_each_entry_safe(mca_err_addr, tmp, &err_info->err_addr_list, node) { + mc_umc_status = mca_err_addr->err_status; + if (mc_umc_status && + (umc_v12_0_is_uncorrectable_error(adev, mc_umc_status) || + umc_v12_0_is_deferred_error(adev, mc_umc_status))) { + uint64_t mca_addr, err_addr, mca_ipid; + uint32_t InstanceIdLo; + + mca_addr = mca_err_addr->err_addr; + mca_ipid = mca_err_addr->err_ipid; + + err_addr = REG_GET_FIELD(mca_addr, + MCA_UMC_UMC0_MCUMC_ADDRT0, ErrorAddr); + InstanceIdLo = REG_GET_FIELD(mca_ipid, MCMP1_IPIDT0, InstanceIdLo); + + dev_info(adev->dev, "UMC:IPID:0x%llx, aid:%d, inst:%d, ch:%d, err_addr:0x%llx\n", + mca_ipid, + err_info->mcm_info.die_id, + MCA_IPID_LO_2_UMC_INST(InstanceIdLo), + MCA_IPID_LO_2_UMC_CH(InstanceIdLo), + err_addr); + + umc_v12_0_convert_error_address(adev, + err_data, err_addr, + MCA_IPID_LO_2_UMC_CH(InstanceIdLo), + MCA_IPID_LO_2_UMC_INST(InstanceIdLo), + err_info->mcm_info.die_id); + } + + /* Delete error address node from list and free memory */ + amdgpu_ras_del_mca_err_addr(err_info, mca_err_addr); } } } +static bool umc_v12_0_check_ecc_err_status(struct amdgpu_device *adev, + enum amdgpu_mca_error_type type, void *ras_error_status) +{ + uint64_t mc_umc_status = *(uint64_t *)ras_error_status; + + switch (type) { + case AMDGPU_MCA_ERROR_TYPE_UE: + return umc_v12_0_is_uncorrectable_error(adev, mc_umc_status); + case AMDGPU_MCA_ERROR_TYPE_CE: + return umc_v12_0_is_correctable_error(adev, mc_umc_status); + case AMDGPU_MCA_ERROR_TYPE_DE: + return umc_v12_0_is_deferred_error(adev, mc_umc_status); + default: + return false; + } + + return false; +} + static void umc_v12_0_err_cnt_init(struct amdgpu_device *adev) { amdgpu_umc_loop_channels(adev, @@ -444,12 +497,71 @@ const struct amdgpu_ras_block_hw_ops umc_v12_0_ras_hw_ops = { .query_ras_error_address = umc_v12_0_query_ras_error_address, }; +static int umc_v12_0_aca_bank_generate_report(struct aca_handle *handle, struct aca_bank *bank, enum aca_error_type type, + struct aca_bank_report *report, void *data) +{ + struct amdgpu_device *adev = handle->adev; + u64 status; + int ret; + + ret = aca_bank_info_decode(bank, &report->info); + if (ret) + return ret; + + status = bank->regs[ACA_REG_IDX_STATUS]; + switch (type) { + case ACA_ERROR_TYPE_UE: + if (umc_v12_0_is_uncorrectable_error(adev, status)) { + report->count[type] = 1; + } + break; + case ACA_ERROR_TYPE_CE: + if (umc_v12_0_is_correctable_error(adev, status)) { + report->count[type] = 1; + } + break; + default: + return -EINVAL; + } + + return 0; +} + +static const struct aca_bank_ops umc_v12_0_aca_bank_ops = { + .aca_bank_generate_report = umc_v12_0_aca_bank_generate_report, +}; + +const struct aca_info umc_v12_0_aca_info = { + .hwip = ACA_HWIP_TYPE_UMC, + .mask = ACA_ERROR_UE_MASK | ACA_ERROR_CE_MASK, + .bank_ops = &umc_v12_0_aca_bank_ops, +}; + +static int umc_v12_0_ras_late_init(struct amdgpu_device *adev, struct ras_common_if *ras_block) +{ + int ret; + + ret = amdgpu_umc_ras_late_init(adev, ras_block); + if (ret) + return ret; + + ret = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__UMC, + &umc_v12_0_aca_info, NULL); + if (ret) + return ret; + + return 0; +} + struct amdgpu_umc_ras umc_v12_0_ras = { .ras_block = { .hw_ops = &umc_v12_0_ras_hw_ops, + .ras_late_init = umc_v12_0_ras_late_init, }, .err_cnt_init = umc_v12_0_err_cnt_init, .query_ras_poison_mode = umc_v12_0_query_ras_poison_mode, .ecc_info_query_ras_error_count = umc_v12_0_ecc_info_query_ras_error_count, .ecc_info_query_ras_error_address = umc_v12_0_ecc_info_query_ras_error_address, + .check_ecc_err_status = umc_v12_0_check_ecc_err_status, }; + diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h index e8de3a92251a..5973bfb14fce 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h +++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.h @@ -121,9 +121,12 @@ (((_ipid_lo) >> 12) & 0xF)) #define MCA_IPID_LO_2_UMC_INST(_ipid_lo) (((_ipid_lo) >> 21) & 0x7) +bool umc_v12_0_is_deferred_error(struct amdgpu_device *adev, uint64_t mc_umc_status); bool umc_v12_0_is_uncorrectable_error(struct amdgpu_device *adev, uint64_t mc_umc_status); bool umc_v12_0_is_correctable_error(struct amdgpu_device *adev, uint64_t mc_umc_status); +typedef bool (*check_error_type_func)(struct amdgpu_device *adev, uint64_t mc_umc_status); + extern const uint32_t umc_v12_0_channel_idx_tbl[] [UMC_V12_0_UMC_INSTANCE_NUM] diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v6_0.c index 0d6b50528d76..97fa88ed770c 100644 --- a/drivers/gpu/drm/amd/amdgpu/umc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/umc_v6_0.c @@ -25,7 +25,7 @@ static void umc_v6_0_init_registers(struct amdgpu_device *adev) { - unsigned i,j; + unsigned i, j; for (i = 0; i < 4; i++) for (j = 0; j < 4; j++) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index cd8e459201f1..002b08fa632f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -55,6 +55,7 @@ static struct kfd_gpu_cache_info kaveri_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -64,6 +65,7 @@ static struct kfd_gpu_cache_info kaveri_cache_info[] = { /* Scalar L1 Instruction Cache (in SQC module) per bank */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -73,6 +75,7 @@ static struct kfd_gpu_cache_info kaveri_cache_info[] = { /* Scalar L1 Data Cache (in SQC module) per bank */ .cache_size = 8, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -88,6 +91,7 @@ static struct kfd_gpu_cache_info carrizo_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -95,8 +99,9 @@ static struct kfd_gpu_cache_info carrizo_cache_info[] = { }, { /* Scalar L1 Instruction Cache (in SQC module) per bank */ - .cache_size = 8, + .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -104,8 +109,9 @@ static struct kfd_gpu_cache_info carrizo_cache_info[] = { }, { /* Scalar L1 Data Cache (in SQC module) per bank. */ - .cache_size = 4, + .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -135,6 +141,7 @@ static struct kfd_gpu_cache_info vega10_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -144,6 +151,7 @@ static struct kfd_gpu_cache_info vega10_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -153,6 +161,7 @@ static struct kfd_gpu_cache_info vega10_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -162,6 +171,7 @@ static struct kfd_gpu_cache_info vega10_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 4096, .cache_level = 2, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -174,6 +184,7 @@ static struct kfd_gpu_cache_info raven_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -183,6 +194,7 @@ static struct kfd_gpu_cache_info raven_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -192,6 +204,7 @@ static struct kfd_gpu_cache_info raven_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -201,6 +214,7 @@ static struct kfd_gpu_cache_info raven_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 1024, .cache_level = 2, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -213,6 +227,7 @@ static struct kfd_gpu_cache_info renoir_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -222,6 +237,7 @@ static struct kfd_gpu_cache_info renoir_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -231,6 +247,7 @@ static struct kfd_gpu_cache_info renoir_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -240,6 +257,7 @@ static struct kfd_gpu_cache_info renoir_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 1024, .cache_level = 2, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -252,6 +270,7 @@ static struct kfd_gpu_cache_info vega12_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -261,6 +280,7 @@ static struct kfd_gpu_cache_info vega12_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -270,6 +290,7 @@ static struct kfd_gpu_cache_info vega12_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -279,6 +300,7 @@ static struct kfd_gpu_cache_info vega12_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 2048, .cache_level = 2, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -291,6 +313,7 @@ static struct kfd_gpu_cache_info vega20_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -300,6 +323,7 @@ static struct kfd_gpu_cache_info vega20_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -309,6 +333,7 @@ static struct kfd_gpu_cache_info vega20_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -318,6 +343,7 @@ static struct kfd_gpu_cache_info vega20_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 8192, .cache_level = 2, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -330,6 +356,7 @@ static struct kfd_gpu_cache_info aldebaran_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -339,6 +366,7 @@ static struct kfd_gpu_cache_info aldebaran_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -348,6 +376,7 @@ static struct kfd_gpu_cache_info aldebaran_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -357,6 +386,7 @@ static struct kfd_gpu_cache_info aldebaran_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 8192, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -369,6 +399,7 @@ static struct kfd_gpu_cache_info navi10_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -378,6 +409,7 @@ static struct kfd_gpu_cache_info navi10_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -387,6 +419,7 @@ static struct kfd_gpu_cache_info navi10_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -396,6 +429,7 @@ static struct kfd_gpu_cache_info navi10_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -405,6 +439,7 @@ static struct kfd_gpu_cache_info navi10_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 4096, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -417,6 +452,7 @@ static struct kfd_gpu_cache_info vangogh_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -426,6 +462,7 @@ static struct kfd_gpu_cache_info vangogh_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -435,6 +472,7 @@ static struct kfd_gpu_cache_info vangogh_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -444,6 +482,7 @@ static struct kfd_gpu_cache_info vangogh_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -453,6 +492,7 @@ static struct kfd_gpu_cache_info vangogh_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 1024, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -465,6 +505,7 @@ static struct kfd_gpu_cache_info navi14_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -474,6 +515,7 @@ static struct kfd_gpu_cache_info navi14_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -483,6 +525,7 @@ static struct kfd_gpu_cache_info navi14_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -492,6 +535,7 @@ static struct kfd_gpu_cache_info navi14_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -501,6 +545,7 @@ static struct kfd_gpu_cache_info navi14_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 2048, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -513,6 +558,7 @@ static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -522,6 +568,7 @@ static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -531,6 +578,7 @@ static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -540,6 +588,7 @@ static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -549,6 +598,7 @@ static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 4096, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -558,6 +608,7 @@ static struct kfd_gpu_cache_info sienna_cichlid_cache_info[] = { /* L3 Data Cache per GPU */ .cache_size = 128*1024, .cache_level = 3, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -570,6 +621,7 @@ static struct kfd_gpu_cache_info navy_flounder_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -579,6 +631,7 @@ static struct kfd_gpu_cache_info navy_flounder_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -588,6 +641,7 @@ static struct kfd_gpu_cache_info navy_flounder_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -597,6 +651,7 @@ static struct kfd_gpu_cache_info navy_flounder_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -606,6 +661,7 @@ static struct kfd_gpu_cache_info navy_flounder_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 3072, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -615,6 +671,7 @@ static struct kfd_gpu_cache_info navy_flounder_cache_info[] = { /* L3 Data Cache per GPU */ .cache_size = 96*1024, .cache_level = 3, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -627,6 +684,7 @@ static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -636,6 +694,7 @@ static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -645,6 +704,7 @@ static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -654,6 +714,7 @@ static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -663,6 +724,7 @@ static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 2048, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -672,6 +734,7 @@ static struct kfd_gpu_cache_info dimgrey_cavefish_cache_info[] = { /* L3 Data Cache per GPU */ .cache_size = 32*1024, .cache_level = 3, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -684,6 +747,7 @@ static struct kfd_gpu_cache_info beige_goby_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -693,6 +757,7 @@ static struct kfd_gpu_cache_info beige_goby_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -702,6 +767,7 @@ static struct kfd_gpu_cache_info beige_goby_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -711,6 +777,7 @@ static struct kfd_gpu_cache_info beige_goby_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -720,6 +787,7 @@ static struct kfd_gpu_cache_info beige_goby_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 1024, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -729,6 +797,7 @@ static struct kfd_gpu_cache_info beige_goby_cache_info[] = { /* L3 Data Cache per GPU */ .cache_size = 16*1024, .cache_level = 3, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -741,6 +810,7 @@ static struct kfd_gpu_cache_info yellow_carp_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -750,6 +820,7 @@ static struct kfd_gpu_cache_info yellow_carp_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -759,6 +830,7 @@ static struct kfd_gpu_cache_info yellow_carp_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -768,6 +840,7 @@ static struct kfd_gpu_cache_info yellow_carp_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -777,6 +850,7 @@ static struct kfd_gpu_cache_info yellow_carp_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 2048, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -789,6 +863,7 @@ static struct kfd_gpu_cache_info gfx1037_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -798,6 +873,7 @@ static struct kfd_gpu_cache_info gfx1037_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -807,6 +883,7 @@ static struct kfd_gpu_cache_info gfx1037_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -816,6 +893,7 @@ static struct kfd_gpu_cache_info gfx1037_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -825,6 +903,7 @@ static struct kfd_gpu_cache_info gfx1037_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 256, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -837,6 +916,7 @@ static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -846,6 +926,7 @@ static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -855,6 +936,7 @@ static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -864,6 +946,7 @@ static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -873,6 +956,7 @@ static struct kfd_gpu_cache_info gc_10_3_6_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 256, .cache_level = 2, + .cache_line_size = 128, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -885,6 +969,7 @@ static struct kfd_gpu_cache_info dummy_cache_info[] = { /* TCP L1 Cache per CU */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -894,6 +979,7 @@ static struct kfd_gpu_cache_info dummy_cache_info[] = { /* Scalar L1 Instruction Cache per SQC */ .cache_size = 32, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_INST_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -903,6 +989,7 @@ static struct kfd_gpu_cache_info dummy_cache_info[] = { /* Scalar L1 Data Cache per SQC */ .cache_size = 16, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -912,6 +999,7 @@ static struct kfd_gpu_cache_info dummy_cache_info[] = { /* GL1 Data Cache per SA */ .cache_size = 128, .cache_level = 1, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), @@ -921,6 +1009,7 @@ static struct kfd_gpu_cache_info dummy_cache_info[] = { /* L2 Data Cache per GPU (Total Tex Cache) */ .cache_size = 2048, .cache_level = 2, + .cache_line_size = 64, .flags = (CRAT_CACHE_FLAGS_ENABLED | CRAT_CACHE_FLAGS_DATA_CACHE | CRAT_CACHE_FLAGS_SIMD_CACHE), diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h index 74c2d7a0d628..300634b9f668 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.h @@ -303,6 +303,7 @@ struct kfd_node; struct kfd_gpu_cache_info { uint32_t cache_size; uint32_t cache_level; + uint32_t cache_line_size; uint32_t flags; /* Indicates how many Compute Units share this cache * within a SA. Value = 1 indicates the cache is not shared diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c index 9ec750666382..d889e3545120 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_debug.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_debug.c @@ -1018,12 +1018,14 @@ int kfd_dbg_trap_device_snapshot(struct kfd_process *target, uint32_t *entry_size) { struct kfd_dbg_device_info_entry device_info; - uint32_t tmp_entry_size = *entry_size, tmp_num_devices; + uint32_t tmp_entry_size, tmp_num_devices; int i, r = 0; if (!(target && user_info && number_of_device_infos && entry_size)) return -EINVAL; + tmp_entry_size = *entry_size; + tmp_num_devices = min_t(size_t, *number_of_device_infos, target->n_pdds); *number_of_device_infos = target->n_pdds; *entry_size = min_t(size_t, *entry_size, sizeof(device_info)); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_events.c index 739721254a5d..9b33d9d2c9ad 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_events.c @@ -1285,8 +1285,10 @@ void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid) uint32_t id = KFD_FIRST_NONSIGNAL_EVENT_ID; int user_gpu_id; - if (!p) + if (!p) { + dev_warn(dev->adev->dev, "Not find process with pasid:%d\n", pasid); return; /* Presumably process exited. */ + } user_gpu_id = kfd_process_get_user_gpu_id(p, dev->id); if (unlikely(user_gpu_id == -EINVAL)) { @@ -1322,6 +1324,8 @@ void kfd_signal_poison_consumed_event(struct kfd_node *dev, u32 pasid) } } + dev_warn(dev->adev->dev, "Send SIGBUS to process %s(pasid:%d)\n", + p->lead_thread->comm, pasid); rcu_read_unlock(); /* user application will handle SIGBUS signal */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c index a7697ec8188e..9a06c6fb6605 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v10.c @@ -132,6 +132,7 @@ enum SQ_INTERRUPT_ERROR_TYPE { static void event_interrupt_poison_consumption(struct kfd_node *dev, uint16_t pasid, uint16_t client_id) { + enum amdgpu_ras_block block = 0; int old_poison, ret = -EINVAL; struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); @@ -151,12 +152,14 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev, case SOC15_IH_CLIENTID_SE3SH: case SOC15_IH_CLIENTID_UTCL2: ret = kfd_dqm_evict_pasid(dev->dqm, pasid); + block = AMDGPU_RAS_BLOCK__GFX; break; case SOC15_IH_CLIENTID_SDMA0: case SOC15_IH_CLIENTID_SDMA1: case SOC15_IH_CLIENTID_SDMA2: case SOC15_IH_CLIENTID_SDMA3: case SOC15_IH_CLIENTID_SDMA4: + block = AMDGPU_RAS_BLOCK__SDMA; break; default: break; @@ -171,12 +174,12 @@ static void event_interrupt_poison_consumption(struct kfd_node *dev, dev_warn(dev->adev->dev, "RAS poison consumption, unmap queue flow succeeded: client id %d\n", client_id); - amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, false); } else { dev_warn(dev->adev->dev, "RAS poison consumption, fall back to gpu reset flow: client id %d\n", client_id); - amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, true); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c index 2a65792fd116..7e2859736a55 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v11.c @@ -191,6 +191,7 @@ static void print_sq_intr_info_error(uint32_t context_id0, uint32_t context_id1) static void event_interrupt_poison_consumption_v11(struct kfd_node *dev, uint16_t pasid, uint16_t source_id) { + enum amdgpu_ras_block block = 0; int ret = -EINVAL; struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); @@ -210,9 +211,11 @@ static void event_interrupt_poison_consumption_v11(struct kfd_node *dev, case SOC15_INTSRC_SQ_INTERRUPT_MSG: if (dev->dqm->ops.reset_queues) ret = dev->dqm->ops.reset_queues(dev->dqm, pasid); + block = AMDGPU_RAS_BLOCK__GFX; break; case SOC21_INTSRC_SDMA_ECC: default: + block = AMDGPU_RAS_BLOCK__GFX; break; } @@ -221,9 +224,9 @@ static void event_interrupt_poison_consumption_v11(struct kfd_node *dev, /* resetting queue passes, do page retirement without gpu reset resetting queue fails, fallback to gpu reset solution */ if (!ret) - amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, false); else - amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, true); } static bool event_interrupt_isr_v11(struct kfd_node *dev, 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 27cdaea40501..91dd5e045b51 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -143,6 +143,7 @@ enum SQ_INTERRUPT_ERROR_TYPE { static void event_interrupt_poison_consumption_v9(struct kfd_node *dev, uint16_t pasid, uint16_t client_id) { + enum amdgpu_ras_block block = 0; int old_poison, ret = -EINVAL; struct kfd_process *p = kfd_lookup_process_by_pasid(pasid); @@ -162,12 +163,14 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev, case SOC15_IH_CLIENTID_SE3SH: case SOC15_IH_CLIENTID_UTCL2: ret = kfd_dqm_evict_pasid(dev->dqm, pasid); + block = AMDGPU_RAS_BLOCK__GFX; break; case SOC15_IH_CLIENTID_SDMA0: case SOC15_IH_CLIENTID_SDMA1: case SOC15_IH_CLIENTID_SDMA2: case SOC15_IH_CLIENTID_SDMA3: case SOC15_IH_CLIENTID_SDMA4: + block = AMDGPU_RAS_BLOCK__SDMA; break; default: break; @@ -182,12 +185,12 @@ static void event_interrupt_poison_consumption_v9(struct kfd_node *dev, dev_warn(dev->adev->dev, "RAS poison consumption, unmap queue flow succeeded: client id %d\n", client_id); - amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, false); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, false); } else { dev_warn(dev->adev->dev, "RAS poison consumption, fall back to gpu reset flow: client id %d\n", client_id); - amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, true); + amdgpu_amdkfd_ras_poison_consumption_handler(dev->adev, block, true); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index c50a0dc9c9c0..f0f7f48af413 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -1515,9 +1515,9 @@ static int svm_range_reserve_bos(struct svm_validate_context *ctx, bool intr) goto unreserve_out; } - r = amdgpu_vm_validate_pt_bos(pdd->dev->adev, - drm_priv_to_vm(pdd->drm_priv), - svm_range_bo_validate, NULL); + r = amdgpu_vm_validate(pdd->dev->adev, + drm_priv_to_vm(pdd->drm_priv), NULL, + svm_range_bo_validate, NULL); if (r) { pr_debug("failed %d validate pt bos\n", r); goto unreserve_out; @@ -1641,7 +1641,9 @@ static int svm_range_validate_and_map(struct mm_struct *mm, goto free_ctx; } - svm_range_reserve_bos(ctx, intr); + r = svm_range_reserve_bos(ctx, intr); + if (r) + goto free_ctx; p = container_of(prange->svms, struct kfd_process, svms); owner = kfd_svm_page_owner(p, find_first_bit(ctx->bitmap, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index e5f7c92eebcb..0136c27ef49f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -1564,6 +1564,7 @@ static int fill_in_l1_pcache(struct kfd_cache_properties **props_ext, pcache->processor_id_low = cu_processor_id + (first_active_cu - 1); pcache->cache_level = pcache_info[cache_type].cache_level; pcache->cache_size = pcache_info[cache_type].cache_size; + pcache->cacheline_size = pcache_info[cache_type].cache_line_size; if (pcache_info[cache_type].flags & CRAT_CACHE_FLAGS_DATA_CACHE) pcache->cache_type |= HSA_CACHE_TYPE_DATA; @@ -1632,6 +1633,7 @@ static int fill_in_l2_l3_pcache(struct kfd_cache_properties **props_ext, pcache->processor_id_low = cu_processor_id + (first_active_cu - 1); pcache->cache_level = pcache_info[cache_type].cache_level; + pcache->cacheline_size = pcache_info[cache_type].cache_line_size; if (KFD_GC_VERSION(knode) == IP_VERSION(9, 4, 3)) mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev); @@ -1705,6 +1707,7 @@ static void kfd_fill_cache_non_crat_info(struct kfd_topology_device *dev, struct gpu_processor_id = dev->node_props.simd_id_base; + memset(cache_info, 0, sizeof(cache_info)); pcache_info = cache_info; num_of_cache_types = kfd_get_gpu_cache_info(kdev, &pcache_info); if (!num_of_cache_types) { diff --git a/drivers/gpu/drm/amd/display/TODO b/drivers/gpu/drm/amd/display/TODO deleted file mode 100644 index a8a6c106e8c7..000000000000 --- a/drivers/gpu/drm/amd/display/TODO +++ /dev/null @@ -1,110 +0,0 @@ -=============================================================================== -TODOs -=============================================================================== - -1. Base this on drm-next - WIP - - -2. Cleanup commit history - - -3. WIP - Drop page flip helper and use DRM's version - - -4. DONE - Flatten all DC objects - * dc_stream/core_stream/stream should just be dc_stream - * Same for other DC objects - - "Is there any major reason to keep all those abstractions? - - Could you collapse everything into struct dc_stream? - - I haven't looked recently but I didn't get the impression there was a - lot of design around what was public/protected, more whatever needed - to be used by someone else was in public." - ~ Dave Airlie - - -5. DONE - Rename DC objects to align more with DRM - * dc_surface -> dc_plane_state - * dc_stream -> dc_stream_state - - -6. DONE - Per-plane and per-stream validation - - -7. WIP - Per-plane and per-stream commit - - -8. WIP - Split pipe_ctx into plane and stream resource structs - - -9. Attach plane and stream reources to state object instead of validate_context - - -10. Remove dc_edid_caps and drm_helpers_parse_edid_caps - * Use drm_display_info instead - * Remove DC's edid quirks and rely on DRM's quirks (add quirks if needed) - - "Making sure you use the sink-specific helper libraries and kernel - subsystems, since there's really no good reason to have 2nd - implementation of those in the kernel. Looks likes that's done for mst - and edid parsing. There's still a bit a midlayer feeling to the edid - parsing side (e.g. dc_edid_caps and dm_helpers_parse_edid_caps, I - think it'd be much better if you convert that over to reading stuff - from drm_display_info and if needed, push stuff into the core). Also, - I can't come up with a good reason why DC needs all this (except to - reimplement half of our edid quirk table, which really isn't a good - idea). Might be good if you put this onto the list of things to fix - long-term, but imo not a blocker. Definitely make sure new stuff - doesn't slip in (i.e. if you start adding edid quirks to DC instead of - the drm core, refactoring to use the core edid stuff was pointless)." - ~ Daniel Vetter - - -11. Remove dc/i2caux. This folder can be somewhat misleading. It's basically an -overy complicated HW programming function for sendind and receiving i2c/aux -commands. We can greatly simplify that and move it into dc/dceXYZ like other -HW blocks. - -12. drm_modeset_lock in MST should no longer be needed in recent kernels - * Adopt appropriate locking scheme - -13. get_modes and best_encoder callbacks look a bit funny. Can probably rip out -a few indirections, and consider removing entirely and using the -drm_atomic_helper_best_encoder default behaviour. - -14. core/dc_debug.c, consider switching to the atomic state debug helpers and -moving all your driver state printing into the various atomic_print_state -callbacks. There's also plans to expose this stuff in a standard way across all -drivers, to make debugging userspace compositors easier across different hw. - -15. Move DP/HDMI dual mode adaptors to drm_dp_dual_mode_helper.c. See -dal_ddc_service_i2c_query_dp_dual_mode_adaptor. - -16. Move to core SCDC helpers (I think those are new since initial DC review). - -17. There's still a pretty massive layer cake around dp aux and DPCD handling, -with like 3 levels of abstraction and using your own structures instead of the -stuff in drm_dp_helper.h. drm_dp_helper.h isn't really great and already has 2 -incompatible styles, just means more reasons not to add a third (or well third -one gets to do the cleanup refactor). - -18. There's a pile of sink handling code, both for DP and HDMI where I didn't -immediately recognize the standard. I think long term it'd be best for the drm -subsystem if we try to move as much of that into helpers/core as possible, and -share it with drivers. But that's a very long term goal, and by far not just an -issue with DC - other drivers, especially around DP sink handling, are equally -guilty. - -19. DONE - The DC logger is still a rather sore thing, but I know that the -DRM_DEBUG stuff just isn't up to the challenges either. We need to figure out -something that integrates better with DRM and linux debug printing, while not -being useless with filtering output. dynamic debug printing might be an option. - -20. Use kernel i2c device to program HDMI retimer. Some boards have an HDMI -retimer that we need to program to pass PHY compliance. Currently that's -bypassing the i2c device and goes directly to HW. This should be changed. - -21. Remove vector.c from dc/basics. It's used in DDC code which can probably -be simplified enough to no longer need a vector implementation. 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 d292f290cd6e..467796d97313 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -67,6 +67,7 @@ #include "amdgpu_dm_debugfs.h" #endif #include "amdgpu_dm_psr.h" +#include "amdgpu_dm_replay.h" #include "ivsrcid/ivsrcid_vislands30.h" @@ -2121,6 +2122,16 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) const struct dmcub_firmware_header_v1_0 *hdr; enum dmub_asic dmub_asic; enum dmub_status status; + static enum dmub_window_memory_type window_memory_type[DMUB_WINDOW_TOTAL] = { + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_0_INST_CONST + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_1_STACK + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_2_BSS_DATA + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_3_VBIOS + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_4_MAILBOX + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_5_TRACEBUFF + DMUB_WINDOW_MEMORY_TYPE_FB, //DMUB_WINDOW_6_FW_STATE + DMUB_WINDOW_MEMORY_TYPE_FB //DMUB_WINDOW_7_SCRATCH_MEM + }; int r; switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) { @@ -2218,7 +2229,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) adev->dm.dmub_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + PSP_HEADER_BYTES; - region_params.is_mailbox_in_inbox = false; + region_params.window_memory_type = window_memory_type; status = dmub_srv_calc_region_info(dmub_srv, ®ion_params, ®ion_info); @@ -2246,6 +2257,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) memory_params.cpu_fb_addr = adev->dm.dmub_bo_cpu_addr; memory_params.gpu_fb_addr = adev->dm.dmub_bo_gpu_addr; memory_params.region_info = ®ion_info; + memory_params.window_memory_type = window_memory_type; adev->dm.dmub_fb_info = kzalloc(sizeof(*adev->dm.dmub_fb_info), GFP_KERNEL); @@ -4399,6 +4411,7 @@ 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; @@ -4510,6 +4523,23 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev) } } + /* Determine whether to enable Replay support by default. */ + if (!(amdgpu_dc_debug_mask & DC_DISABLE_REPLAY)) { + switch (amdgpu_ip_version(adev, 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): + case IP_VERSION(3, 5, 0): + 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; @@ -4578,6 +4608,11 @@ 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) + if (amdgpu_dm_set_replay_caps(link, aconnector)) + psr_feature_enabled = false; + if (psr_feature_enabled) amdgpu_dm_set_psr_caps(link); @@ -6402,10 +6437,81 @@ int amdgpu_dm_connector_atomic_get_property(struct drm_connector *connector, return ret; } +/** + * DOC: panel power savings + * + * The display manager allows you to set your desired **panel power savings** + * level (between 0-4, with 0 representing off), e.g. using the following:: + * + * # echo 3 > /sys/class/drm/card0-eDP-1/amdgpu/panel_power_savings + * + * Modifying this value can have implications on color accuracy, so tread + * carefully. + */ + +static ssize_t panel_power_savings_show(struct device *device, + struct device_attribute *attr, + char *buf) +{ + struct drm_connector *connector = dev_get_drvdata(device); + struct drm_device *dev = connector->dev; + u8 val; + + drm_modeset_lock(&dev->mode_config.connection_mutex, NULL); + val = to_dm_connector_state(connector->state)->abm_level == + ABM_LEVEL_IMMEDIATE_DISABLE ? 0 : + to_dm_connector_state(connector->state)->abm_level; + drm_modeset_unlock(&dev->mode_config.connection_mutex); + + return sysfs_emit(buf, "%u\n", val); +} + +static ssize_t panel_power_savings_store(struct device *device, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct drm_connector *connector = dev_get_drvdata(device); + struct drm_device *dev = connector->dev; + long val; + int ret; + + ret = kstrtol(buf, 0, &val); + + if (ret) + return ret; + + if (val < 0 || val > 4) + return -EINVAL; + + drm_modeset_lock(&dev->mode_config.connection_mutex, NULL); + to_dm_connector_state(connector->state)->abm_level = val ?: + ABM_LEVEL_IMMEDIATE_DISABLE; + drm_modeset_unlock(&dev->mode_config.connection_mutex); + + drm_kms_helper_hotplug_event(dev); + + return count; +} + +static DEVICE_ATTR_RW(panel_power_savings); + +static struct attribute *amdgpu_attrs[] = { + &dev_attr_panel_power_savings.attr, + NULL +}; + +static const struct attribute_group amdgpu_group = { + .name = "amdgpu", + .attrs = amdgpu_attrs +}; + static void amdgpu_dm_connector_unregister(struct drm_connector *connector) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); + if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) + sysfs_remove_group(&connector->kdev->kobj, &amdgpu_group); + drm_dp_aux_unregister(&amdgpu_dm_connector->dm_dp_aux.aux); } @@ -6507,6 +6613,13 @@ amdgpu_dm_connector_late_register(struct drm_connector *connector) to_amdgpu_dm_connector(connector); int r; + if (connector->connector_type == DRM_MODE_CONNECTOR_eDP) { + r = sysfs_create_group(&connector->kdev->kobj, + &amdgpu_group); + if (r) + return r; + } + amdgpu_dm_register_backlight_device(amdgpu_dm_connector); if ((connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) || @@ -8526,10 +8639,22 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, dm_update_pflip_irq_state(drm_to_adev(dev), acrtc_attach); - if ((acrtc_state->update_type > UPDATE_TYPE_FAST) && - acrtc_state->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED && - !acrtc_state->stream->link->psr_settings.psr_feature_enabled) - amdgpu_dm_link_setup_psr(acrtc_state->stream); + if (acrtc_state->update_type > UPDATE_TYPE_FAST) { + if (acrtc_state->stream->link->replay_settings.config.replay_supported && + !acrtc_state->stream->link->replay_settings.replay_feature_enabled) { + struct amdgpu_dm_connector *aconn = + (struct amdgpu_dm_connector *)acrtc_state->stream->dm_stream_context; + amdgpu_dm_link_setup_replay(acrtc_state->stream->link, aconn); + } else if (acrtc_state->stream->link->psr_settings.psr_version != DC_PSR_VERSION_UNSUPPORTED && + !acrtc_state->stream->link->psr_settings.psr_feature_enabled) { + + struct amdgpu_dm_connector *aconn = (struct amdgpu_dm_connector *) + acrtc_state->stream->dm_stream_context; + + if (!aconn->disallow_edp_enter_psr) + amdgpu_dm_link_setup_psr(acrtc_state->stream); + } + } /* Decrement skip count when PSR is enabled and we're doing fast updates. */ if (acrtc_state->update_type == UPDATE_TYPE_FAST && @@ -8556,6 +8681,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, !amdgpu_dm_crc_window_is_activated(acrtc_state->base.crtc) && #endif !acrtc_state->stream->link->psr_settings.psr_allow_active && + !aconn->disallow_edp_enter_psr && (timestamp_ns - acrtc_state->stream->link->psr_settings.psr_dirty_rects_change_timestamp_ns) > 500000000) @@ -8818,11 +8944,12 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state, } } /* for_each_crtc_in_state() */ - /* if there mode set or reset, disable eDP PSR */ + /* if there mode set or reset, disable eDP PSR, Replay */ if (mode_set_reset_required) { if (dm->vblank_control_workqueue) flush_workqueue(dm->vblank_control_workqueue); + amdgpu_dm_replay_disable_all(dm); amdgpu_dm_psr_disable_all(dm); } @@ -10731,11 +10858,13 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, goto fail; } - ret = compute_mst_dsc_configs_for_state(state, dm_state->context, vars); - if (ret) { - DRM_DEBUG_DRIVER("compute_mst_dsc_configs_for_state() failed\n"); - ret = -EINVAL; - goto fail; + if (dc_resource_is_dsc_encoding_supported(dc)) { + ret = compute_mst_dsc_configs_for_state(state, dm_state->context, vars); + if (ret) { + DRM_DEBUG_DRIVER("compute_mst_dsc_configs_for_state() failed\n"); + ret = -EINVAL; + goto fail; + } } ret = dm_update_mst_vcpi_slots_for_dsc(state, dm_state->context, vars); 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 9c1871b866cc..09519b7abf67 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -693,6 +693,7 @@ struct amdgpu_dm_connector { struct drm_display_mode freesync_vid_base; int psr_skip_count; + bool disallow_edp_enter_psr; /* Record progress status of mst*/ uint8_t mst_status; 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 6e715ef3a556..e23a0a276e33 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" @@ -95,6 +96,61 @@ bool amdgpu_dm_crtc_vrr_active(struct dm_crtc_state *dm_state) dm_state->freesync_config.state == VRR_STATE_ACTIVE_FIXED; } +/** + * amdgpu_dm_crtc_set_panel_sr_feature() - Manage panel self-refresh features. + * + * @vblank_work: is a pointer to a struct vblank_control_work object. + * @vblank_enabled: indicates whether the DRM vblank counter is currently + * enabled (true) or disabled (false). + * @allow_sr_entry: represents whether entry into the self-refresh mode is + * allowed (true) or not allowed (false). + * + * The DRM vblank counter enable/disable action is used as the trigger to enable + * or disable various panel self-refresh features: + * + * Panel Replay and PSR SU + * - Enable when: + * - vblank counter is disabled + * - entry is allowed: usermode demonstrates an adequate number of fast + * commits) + * - CRC capture window isn't active + * - Keep enabled even when vblank counter gets enabled + * + * PSR1 + * - Enable condition same as above + * - Disable when vblank counter is enabled + */ +static void amdgpu_dm_crtc_set_panel_sr_feature( + struct vblank_control_work *vblank_work, + bool vblank_enabled, bool allow_sr_entry) +{ + struct dc_link *link = vblank_work->stream->link; + bool is_sr_active = (link->replay_settings.replay_allow_active || + link->psr_settings.psr_allow_active); + bool is_crc_window_active = false; + +#ifdef CONFIG_DRM_AMD_SECURE_DISPLAY + is_crc_window_active = + amdgpu_dm_crc_window_is_activated(&vblank_work->acrtc->base); +#endif + + if (link->replay_settings.replay_feature_enabled && + allow_sr_entry && !is_sr_active && !is_crc_window_active) { + amdgpu_dm_replay_enable(vblank_work->stream, true); + } else if (vblank_enabled) { + if (link->psr_settings.psr_version < DC_PSR_VERSION_SU_1 && is_sr_active) + amdgpu_dm_psr_disable(vblank_work->stream); + } else if (link->psr_settings.psr_feature_enabled && + allow_sr_entry && !is_sr_active && !is_crc_window_active) { + + struct amdgpu_dm_connector *aconn = + (struct amdgpu_dm_connector *) vblank_work->stream->dm_stream_context; + + if (!aconn->disallow_edp_enter_psr) + amdgpu_dm_psr_enable(vblank_work->stream); + } +} + static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work) { struct vblank_control_work *vblank_work = @@ -123,18 +179,10 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work) * fill_dc_dirty_rects(). */ if (vblank_work->stream && vblank_work->stream->link) { - 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); - } else if (vblank_work->stream->link->psr_settings.psr_feature_enabled && - !vblank_work->stream->link->psr_settings.psr_allow_active && -#ifdef CONFIG_DRM_AMD_SECURE_DISPLAY - !amdgpu_dm_crc_window_is_activated(&vblank_work->acrtc->base) && -#endif - vblank_work->acrtc->dm_irq_params.allow_psr_entry) { - amdgpu_dm_psr_enable(vblank_work->stream); - } + amdgpu_dm_crtc_set_panel_sr_feature( + vblank_work, vblank_work->enable, + vblank_work->acrtc->dm_irq_params.allow_psr_entry || + vblank_work->stream->link->replay_settings.replay_feature_enabled); } mutex_unlock(&dm->dc_lock); 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 68a846323912..eee4945653e2 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 @@ -1483,7 +1483,7 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf, const uint32_t rd_buf_size = 10; struct pipe_ctx *pipe_ctx; ssize_t result = 0; - int i, r, str_len = 30; + int i, r, str_len = 10; rd_buf = kcalloc(rd_buf_size, sizeof(char), GFP_KERNEL); @@ -2971,6 +2971,53 @@ static int allow_edp_hotplug_detection_set(void *data, u64 val) return 0; } +/* check if kernel disallow eDP enter psr state + * cat /sys/kernel/debug/dri/0/eDP-X/disallow_edp_enter_psr + * 0: allow edp enter psr; 1: disallow + */ +static int disallow_edp_enter_psr_get(void *data, u64 *val) +{ + struct amdgpu_dm_connector *aconnector = data; + + *val = (u64) aconnector->disallow_edp_enter_psr; + return 0; +} + +/* set kernel disallow eDP enter psr state + * echo 0x0 /sys/kernel/debug/dri/0/eDP-X/disallow_edp_enter_psr + * 0: allow edp enter psr; 1: disallow + * + * usage: test app read crc from PSR eDP rx. + * + * during kernel boot up, kernel write dpcd 0x170 = 5. + * this notify eDP rx psr enable and let rx check crc. + * rx fw will start checking crc for rx internal logic. + * crc read count within dpcd 0x246 is not updated and + * value is 0. when eDP tx driver wants to read rx crc + * from dpcd 0x246, 0x270, read count 0 lead tx driver + * timeout. + * + * to avoid this, we add this debugfs to let test app to disbable + * rx crc checking for rx internal logic. then test app can read + * non-zero crc read count. + * + * expected app sequence is as below: + * 1. disable eDP PHY and notify eDP rx with dpcd 0x600 = 2. + * 2. echo 0x1 /sys/kernel/debug/dri/0/eDP-X/disallow_edp_enter_psr + * 3. enable eDP PHY and notify eDP rx with dpcd 0x600 = 1 but + * without dpcd 0x170 = 5. + * 4. read crc from rx dpcd 0x270, 0x246, etc. + * 5. echo 0x0 /sys/kernel/debug/dri/0/eDP-X/disallow_edp_enter_psr. + * this will let eDP back to normal with psr setup dpcd 0x170 = 5. + */ +static int disallow_edp_enter_psr_set(void *data, u64 val) +{ + struct amdgpu_dm_connector *aconnector = data; + + aconnector->disallow_edp_enter_psr = val ? true : false; + return 0; +} + static int dmub_trace_mask_set(void *data, u64 val) { struct amdgpu_device *adev = data; @@ -3092,6 +3139,10 @@ DEFINE_DEBUGFS_ATTRIBUTE(allow_edp_hotplug_detection_fops, allow_edp_hotplug_detection_get, allow_edp_hotplug_detection_set, "%llu\n"); +DEFINE_DEBUGFS_ATTRIBUTE(disallow_edp_enter_psr_fops, + disallow_edp_enter_psr_get, + disallow_edp_enter_psr_set, "%llu\n"); + DEFINE_SHOW_ATTRIBUTE(current_backlight); DEFINE_SHOW_ATTRIBUTE(target_backlight); @@ -3265,6 +3316,8 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) &edp_ilr_debugfs_fops); debugfs_create_file("allow_edp_hotplug_detection", 0644, dir, connector, &allow_edp_hotplug_detection_fops); + debugfs_create_file("disallow_edp_enter_psr", 0644, dir, connector, + &disallow_edp_enter_psr_fops); } for (i = 0; i < ARRAY_SIZE(connector_debugfs_entries); i++) { 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 index 5ce542b1f860..738a58eebba7 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c @@ -60,21 +60,26 @@ static bool link_supports_replay(struct dc_link *link, struct amdgpu_dm_connecto if (!as_caps->dp_adap_sync_caps.bits.ADAPTIVE_SYNC_SDP_SUPPORT) return false; + // Sink shall populate line deviation information + if (dpcd_caps->pr_info.pixel_deviation_per_line == 0 || + dpcd_caps->pr_info.max_deviation_line == 0) + return false; + return true; } /* - * amdgpu_dm_setup_replay() - setup replay configuration + * amdgpu_dm_set_replay_caps() - setup Replay capabilities * @link: link * @aconnector: aconnector * */ -bool amdgpu_dm_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *aconnector) +bool amdgpu_dm_set_replay_caps(struct dc_link *link, struct amdgpu_dm_connector *aconnector) { - struct replay_config pr_config; + struct replay_config pr_config = { 0 }; union replay_debug_flags *debug_flags = NULL; - // For eDP, if Replay is supported, return true to skip checks + // If Replay is already set to support, return true to skip checks if (link->replay_settings.config.replay_supported) return true; @@ -87,27 +92,50 @@ bool amdgpu_dm_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *ac if (!link_supports_replay(link, aconnector)) return false; - // Mark Replay is supported in link and update related attributes + // Mark Replay is supported in pr_config 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; - - 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; - link->replay_settings.replay_feature_enabled = true; - init_replay_config(link, &pr_config); return true; } +/* + * amdgpu_dm_link_setup_replay() - configure replay link + * @link: link + * @aconnector: aconnector + * + */ +bool amdgpu_dm_link_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *aconnector) +{ + struct replay_config *pr_config; + + if (link == NULL || aconnector == NULL) + return false; + + pr_config = &link->replay_settings.config; + + if (!pr_config->replay_supported) + return false; + + pr_config->replay_power_opt_supported = 0x11; + pr_config->replay_smu_opt_supported = false; + pr_config->replay_enable_option |= pr_enable_option_static_screen; + pr_config->replay_support_fast_resync_in_ultra_sleep_mode = aconnector->max_vfreq >= 2 * aconnector->min_vfreq; + pr_config->replay_timing_sync_supported = false; + + if (!pr_config->replay_timing_sync_supported) + pr_config->replay_enable_option &= ~pr_enable_option_general_ui; + + link->replay_settings.replay_feature_enabled = true; + + return true; +} /* * amdgpu_dm_replay_enable() - enable replay f/w @@ -117,51 +145,23 @@ bool amdgpu_dm_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *ac */ 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 */ + if (link) { + link->dc->link_srv->edp_setup_replay(link, stream); + link->dc->link_srv->edp_set_coasting_vtotal(link, stream->timing.v_total); + DRM_DEBUG_DRIVER("Enabling replay...\n"); + link->dc->link_srv->edp_set_replay_allow_active(link, &replay_active, wait, false, NULL); + return true; } - return true; + return false; } /* @@ -172,12 +172,31 @@ bool amdgpu_dm_replay_enable(struct dc_stream_state *stream, bool wait) */ bool amdgpu_dm_replay_disable(struct dc_stream_state *stream) { + bool replay_active = false; + struct dc_link *link = NULL; - if (stream->link) { + if (stream == NULL) + return false; + + link = stream->link; + + if (link) { DRM_DEBUG_DRIVER("Disabling replay...\n"); - stream->link->dc->link_srv->edp_set_replay_allow_active(stream->link, NULL, false, false, NULL); + link->dc->link_srv->edp_set_replay_allow_active(stream->link, &replay_active, true, false, NULL); return true; } return false; } + +/* + * amdgpu_dm_replay_disable_all() - disable replay f/w + * if replay is enabled on any stream + * + * Return: true if success + */ +bool amdgpu_dm_replay_disable_all(struct amdgpu_display_manager *dm) +{ + DRM_DEBUG_DRIVER("Disabling replay if replay is enabled on any stream\n"); + return dc_set_replay_allow_active(dm->dc, 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 index 01cba3cd6246..f0d30eb47312 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h @@ -40,7 +40,9 @@ enum replay_enable_option { 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_set_replay_caps(struct dc_link *link, struct amdgpu_dm_connector *aconnector); +bool amdgpu_dm_link_setup_replay(struct dc_link *link, struct amdgpu_dm_connector *aconnector); bool amdgpu_dm_replay_disable(struct dc_stream_state *stream); +bool amdgpu_dm_replay_disable_all(struct amdgpu_display_manager *dm); #endif /* AMDGPU_DM_AMDGPU_DM_REPLAY_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.c b/drivers/gpu/drm/amd/display/dc/basics/conversion.c index 1090d235086a..bd1f60ecaba4 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.c +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.c @@ -101,6 +101,40 @@ void convert_float_matrix( } } +static struct fixed31_32 int_frac_to_fixed_point(uint16_t arg, + uint8_t integer_bits, + uint8_t fractional_bits) +{ + struct fixed31_32 result; + uint16_t sign_mask = 1 << (fractional_bits + integer_bits); + uint16_t value_mask = sign_mask - 1; + + result.value = (long long)(arg & value_mask) << + (FIXED31_32_BITS_PER_FRACTIONAL_PART - fractional_bits); + + if (arg & sign_mask) + result = dc_fixpt_neg(result); + + return result; +} + +/** + * convert_hw_matrix - converts HW values into fixed31_32 matrix. + * @matrix: fixed point 31.32 matrix + * @reg: array of register values + * @buffer_size: size of the array of register values + * + * Converts HW register spec defined format S2D13 into a fixed-point 31.32 + * matrix. + */ +void convert_hw_matrix(struct fixed31_32 *matrix, + uint16_t *reg, + uint32_t buffer_size) +{ + for (int i = 0; i < buffer_size; ++i) + matrix[i] = int_frac_to_fixed_point(reg[i], 2, 13); +} + static uint32_t find_gcd(uint32_t a, uint32_t b) { uint32_t remainder; diff --git a/drivers/gpu/drm/amd/display/dc/basics/conversion.h b/drivers/gpu/drm/amd/display/dc/basics/conversion.h index 81da4e6f7a1a..a433cef78496 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/conversion.h +++ b/drivers/gpu/drm/amd/display/dc/basics/conversion.h @@ -41,6 +41,10 @@ void convert_float_matrix( void reduce_fraction(uint32_t num, uint32_t den, uint32_t *out_num, uint32_t *out_den); +void convert_hw_matrix(struct fixed31_32 *matrix, + uint16_t *reg, + uint32_t buffer_size); + static inline unsigned int log_2(unsigned int num) { return ilog2(num); diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table.c b/drivers/gpu/drm/amd/display/dc/bios/command_table.c index 818a529cacc3..86f9198e7501 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.c @@ -37,7 +37,7 @@ #define EXEC_BIOS_CMD_TABLE(command, params)\ (amdgpu_atom_execute_table(((struct amdgpu_device *)bp->base.ctx->driver_context)->mode_info.atom_context, \ GetIndexIntoMasterTable(COMMAND, command), \ - (uint32_t *)¶ms) == 0) + (uint32_t *)¶ms, sizeof(params)) == 0) #define BIOS_CMD_TABLE_REVISION(command, frev, crev)\ amdgpu_atom_parse_cmd_header(((struct amdgpu_device *)bp->base.ctx->driver_context)->mode_info.atom_context, \ diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c index 293a919d605d..cbae1be7b009 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table2.c @@ -49,7 +49,7 @@ #define EXEC_BIOS_CMD_TABLE(fname, params)\ (amdgpu_atom_execute_table(((struct amdgpu_device *)bp->base.ctx->driver_context)->mode_info.atom_context, \ GET_INDEX_INTO_MASTER_TABLE(command, fname), \ - (uint32_t *)¶ms) == 0) + (uint32_t *)¶ms, sizeof(params)) == 0) #define BIOS_CMD_TABLE_REVISION(fname, frev, crev)\ amdgpu_atom_parse_cmd_header(((struct amdgpu_device *)bp->base.ctx->driver_context)->mode_info.atom_context, \ diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index 0c6a4ab72b1d..e3e1940198a9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -707,9 +707,7 @@ void rn_clk_mgr_construct( int is_green_sardine = 0; struct clk_log_info log_info = {0}; -#if defined(CONFIG_DRM_AMD_DC_FP) is_green_sardine = ASICREV_IS_GREEN_SARDINE(ctx->asic_id.hw_internal_rev); -#endif clk_mgr->base.ctx = ctx; clk_mgr->base.funcs = &dcn21_funcs; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c index 8c9d45e5b13b..d72acbb049b1 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c @@ -185,10 +185,6 @@ int rn_vbios_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int reque VBIOSSMC_MSG_SetHardMinDcfclkByFreq, khz_to_mhz_ceil(requested_dcfclk_khz)); -#ifdef DBG - smu_print("actual_dcfclk_set_mhz %d is set to : %d\n", actual_dcfclk_set_mhz, actual_dcfclk_set_mhz * 1000); -#endif - return actual_dcfclk_set_mhz * 1000; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c index e4f96b6fd79d..19e5b3be9275 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c @@ -180,10 +180,6 @@ int dcn301_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int request VBIOSSMC_MSG_SetHardMinDcfclkByFreq, khz_to_mhz_ceil(requested_dcfclk_khz)); -#ifdef DBG - smu_print("actual_dcfclk_set_mhz %d is set to : %d\n", actual_dcfclk_set_mhz, actual_dcfclk_set_mhz * 1000); -#endif - return actual_dcfclk_set_mhz * 1000; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c index 32279c5db724..6904e95113c1 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c @@ -202,10 +202,6 @@ int dcn31_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int requeste VBIOSSMC_MSG_SetHardMinDcfclkByFreq, khz_to_mhz_ceil(requested_dcfclk_khz)); -#ifdef DBG - smu_print("actual_dcfclk_set_mhz %d is set to : %d\n", actual_dcfclk_set_mhz, actual_dcfclk_set_mhz * 1000); -#endif - return actual_dcfclk_set_mhz * 1000; } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c index 07baa10a8647..c4af406146b7 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_smu.c @@ -220,12 +220,6 @@ int dcn314_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int request VBIOSSMC_MSG_SetHardMinDcfclkByFreq, khz_to_mhz_ceil(requested_dcfclk_khz)); -#ifdef DBG - smu_print("actual_dcfclk_set_mhz %d is set to : %d\n", - actual_dcfclk_set_mhz, - actual_dcfclk_set_mhz * 1000); -#endif - return actual_dcfclk_set_mhz * 1000; } 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 1042cf1a3ab0..879f1494c4cd 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 @@ -215,10 +215,6 @@ int dcn315_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int request VBIOSSMC_MSG_SetHardMinDcfclkByFreq, khz_to_mhz_ceil(requested_dcfclk_khz)); -#ifdef DBG - smu_print("actual_dcfclk_set_mhz %d is set to : %d\n", actual_dcfclk_set_mhz, actual_dcfclk_set_mhz * 1000); -#endif - return actual_dcfclk_set_mhz * 1000; } 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 3ed19197a755..8b82092b91cd 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 @@ -189,10 +189,6 @@ int dcn316_smu_set_hard_min_dcfclk(struct clk_mgr_internal *clk_mgr, int request VBIOSSMC_MSG_SetHardMinDcfclkByFreq, khz_to_mhz_ceil(requested_dcfclk_khz)); -#ifdef DBG - smu_print("actual_dcfclk_set_mhz %d is set to : %d\n", actual_dcfclk_set_mhz, actual_dcfclk_set_mhz * 1000); -#endif - return actual_dcfclk_set_mhz * 1000; } 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 aadd07bc68c5..e64e45e4c833 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 @@ -387,7 +387,15 @@ static void dcn32_update_clocks_update_dentist( uint32_t temp_dispclk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR * clk_mgr->base.dentist_vco_freq_khz) / temp_disp_divider; if (clk_mgr->smu_present) - dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DISPCLK, khz_to_mhz_ceil(temp_dispclk_khz)); + /* + * SMU uses discrete dispclk presets. We applied + * the same formula to increase our dppclk_khz + * to the next matching discrete value. By + * contract, we should use the preset dispclk + * floored in Mhz to describe the intended clock. + */ + dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DISPCLK, + khz_to_mhz_floor(temp_dispclk_khz)); if (dc->debug.override_dispclk_programming) { REG_GET(DENTIST_DISPCLK_CNTL, @@ -426,7 +434,15 @@ static void dcn32_update_clocks_update_dentist( /* do requested DISPCLK updates*/ if (clk_mgr->smu_present) - dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DISPCLK, khz_to_mhz_ceil(clk_mgr->base.clks.dispclk_khz)); + /* + * SMU uses discrete dispclk presets. We applied + * the same formula to increase our dppclk_khz + * to the next matching discrete value. By + * contract, we should use the preset dispclk + * floored in Mhz to describe the intended clock. + */ + dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DISPCLK, + khz_to_mhz_floor(clk_mgr->base.clks.dispclk_khz)); if (dc->debug.override_dispclk_programming) { REG_GET(DENTIST_DISPCLK_CNTL, @@ -493,6 +509,8 @@ static void dcn32_auto_dpm_test_log( } } + msleep(5); + mall_ss_size_bytes = context->bw_ctx.bw.dcn.mall_ss_size_bytes; dispclk_khz_reg = REG_READ(CLK1_CLK0_CURRENT_CNT); // DISPCLK @@ -734,7 +752,15 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz; if (clk_mgr->smu_present && !dpp_clock_lowered) - dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DPPCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dppclk_khz)); + /* + * SMU uses discrete dppclk presets. We applied + * the same formula to increase our dppclk_khz + * to the next matching discrete value. By + * contract, we should use the preset dppclk + * floored in Mhz to describe the intended clock. + */ + dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DPPCLK, + khz_to_mhz_floor(clk_mgr_base->clks.dppclk_khz)); update_dppclk = true; } @@ -765,7 +791,15 @@ static void dcn32_update_clocks(struct clk_mgr *clk_mgr_base, dcn32_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower); dcn32_update_clocks_update_dentist(clk_mgr, context); if (clk_mgr->smu_present) - dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DPPCLK, khz_to_mhz_ceil(clk_mgr_base->clks.dppclk_khz)); + /* + * SMU uses discrete dppclk presets. We applied + * the same formula to increase our dppclk_khz + * to the next matching discrete value. By + * contract, we should use the preset dppclk + * floored in Mhz to describe the intended clock. + */ + dcn32_smu_set_hard_min_by_freq(clk_mgr, PPCLK_DPPCLK, + khz_to_mhz_floor(clk_mgr_base->clks.dppclk_khz)); } else { /* if clock is being raised, increase refclk before lowering DTO */ if (update_dppclk || update_dispclk) 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 a34c258c19dc..c76352a817de 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 @@ -36,8 +36,7 @@ #define DALSMC_MSG_SetCabForUclkPstate 0x12 #define DALSMC_Result_OK 0x1 -void -dcn32_smu_send_fclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool enable); +void dcn32_smu_send_fclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool enable); void dcn32_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr); 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); diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c index 14cec1c7b718..06edca50a8fa 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c @@ -384,19 +384,6 @@ static void dcn35_enable_pme_wa(struct clk_mgr *clk_mgr_base) dcn35_smu_enable_pme_wa(clk_mgr); } -void dcn35_init_clocks(struct clk_mgr *clk_mgr) -{ - uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz; - - memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); - - // Assumption is that boot state always supports pstate - clk_mgr->clks.ref_dtbclk_khz = ref_dtbclk; // restore ref_dtbclk - clk_mgr->clks.p_state_change_support = true; - clk_mgr->clks.prev_p_state_change_support = true; - clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; - clk_mgr->clks.zstate_support = DCN_ZSTATE_SUPPORT_UNKNOWN; -} bool dcn35_are_clock_states_equal(struct dc_clocks *a, struct dc_clocks *b) @@ -422,6 +409,23 @@ static void dcn35_dump_clk_registers(struct clk_state_registers_and_bypass *regs { } +static void init_clk_states(struct clk_mgr *clk_mgr) +{ + uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz; + memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); + + clk_mgr->clks.dtbclk_en = true; + clk_mgr->clks.ref_dtbclk_khz = ref_dtbclk; // restore ref_dtbclk + clk_mgr->clks.p_state_change_support = true; + clk_mgr->clks.prev_p_state_change_support = true; + clk_mgr->clks.pwr_state = DCN_PWR_STATE_UNKNOWN; + clk_mgr->clks.zstate_support = DCN_ZSTATE_SUPPORT_UNKNOWN; +} + +void dcn35_init_clocks(struct clk_mgr *clk_mgr) +{ + init_clk_states(clk_mgr); +} static struct clk_bw_params dcn35_bw_params = { .vram_type = Ddr4MemType, .num_channels = 1, @@ -826,7 +830,7 @@ static void dcn35_set_low_power_state(struct clk_mgr *clk_mgr_base) } } -static void dcn35_set_idle_state(struct clk_mgr *clk_mgr_base, bool allow_idle) +static void dcn35_set_ips_idle_state(struct clk_mgr *clk_mgr_base, bool allow_idle) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct dc *dc = clk_mgr_base->ctx->dc; @@ -874,7 +878,7 @@ static bool dcn35_is_ips_supported(struct clk_mgr *clk_mgr_base) return ips_supported; } -static uint32_t dcn35_get_idle_state(struct clk_mgr *clk_mgr_base) +static uint32_t dcn35_get_ips_idle_state(struct clk_mgr *clk_mgr_base) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); @@ -883,7 +887,7 @@ static uint32_t dcn35_get_idle_state(struct clk_mgr *clk_mgr_base) static void dcn35_init_clocks_fpga(struct clk_mgr *clk_mgr) { - dcn35_init_clocks(clk_mgr); + init_clk_states(clk_mgr); /* TODO: Implement the functions and remove the ifndef guard */ } @@ -968,8 +972,8 @@ static struct clk_mgr_funcs dcn35_funcs = { .set_low_power_state = dcn35_set_low_power_state, .exit_low_power_state = dcn35_exit_low_power_state, .is_ips_supported = dcn35_is_ips_supported, - .set_idle_state = dcn35_set_idle_state, - .get_idle_state = dcn35_get_idle_state + .set_idle_state = dcn35_set_ips_idle_state, + .get_idle_state = dcn35_get_ips_idle_state }; struct clk_mgr_funcs dcn35_fpga_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c index 6d4a1ffab5ed..a07f7e685d28 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_smu.c @@ -447,6 +447,9 @@ void dcn35_smu_set_dtbclk(struct clk_mgr_internal *clk_mgr, bool enable) void dcn35_vbios_smu_enable_48mhz_tmdp_refclk_pwrdwn(struct clk_mgr_internal *clk_mgr, bool enable) { + if (!clk_mgr->smu_present) + return; + dcn35_smu_send_msg_with_param( clk_mgr, VBIOSSMC_MSG_EnableTmdp48MHzRefclkPwrDown, @@ -458,6 +461,9 @@ int dcn35_smu_exit_low_power_state(struct clk_mgr_internal *clk_mgr) { int retv; + if (!clk_mgr->smu_present) + return 0; + retv = dcn35_smu_send_msg_with_param( clk_mgr, VBIOSSMC_MSG_DispPsrExit, @@ -470,6 +476,9 @@ int dcn35_smu_get_ips_supported(struct clk_mgr_internal *clk_mgr) { int retv; + if (!clk_mgr->smu_present) + return 0; + retv = dcn35_smu_send_msg_with_param( clk_mgr, VBIOSSMC_MSG_QueryIPS2Support, @@ -481,6 +490,9 @@ int dcn35_smu_get_ips_supported(struct clk_mgr_internal *clk_mgr) void dcn35_smu_write_ips_scratch(struct clk_mgr_internal *clk_mgr, uint32_t param) { + if (!clk_mgr->smu_present) + return; + REG_WRITE(MP1_SMN_C2PMSG_71, param); //smu_print("%s: write_ips_scratch = %x\n", __func__, param); } @@ -489,6 +501,9 @@ uint32_t dcn35_smu_read_ips_scratch(struct clk_mgr_internal *clk_mgr) { uint32_t retv; + if (!clk_mgr->smu_present) + return 0; + retv = REG_READ(MP1_SMN_C2PMSG_71); //smu_print("%s: dcn35_smu_read_ips_scratch = %x\n", __func__, retv); return retv; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index aa7c02ba948e..72512903f88f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -414,6 +414,8 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc, if (dc->optimized_required || dc->wm_optimized_required) return false; + dc_exit_ips_for_hw_access(dc); + stream->adjust.v_total_max = adjust->v_total_max; stream->adjust.v_total_mid = adjust->v_total_mid; stream->adjust.v_total_mid_frame_num = adjust->v_total_mid_frame_num; @@ -454,6 +456,8 @@ bool dc_stream_get_last_used_drr_vtotal(struct dc *dc, int i = 0; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -484,6 +488,8 @@ bool dc_stream_get_crtc_position(struct dc *dc, bool ret = false; struct crtc_position position; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { struct pipe_ctx *pipe = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -603,6 +609,8 @@ bool dc_stream_configure_crc(struct dc *dc, struct dc_stream_state *stream, if (pipe == NULL) return false; + dc_exit_ips_for_hw_access(dc); + /* By default, capture the full frame */ param.windowa_x_start = 0; param.windowa_y_start = 0; @@ -662,6 +670,8 @@ bool dc_stream_get_crc(struct dc *dc, struct dc_stream_state *stream, struct pipe_ctx *pipe; struct timing_generator *tg; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { pipe = &dc->current_state->res_ctx.pipe_ctx[i]; if (pipe->stream == stream) @@ -686,6 +696,8 @@ void dc_stream_set_dyn_expansion(struct dc *dc, struct dc_stream_state *stream, int i; struct pipe_ctx *pipe_ctx; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { if (dc->current_state->res_ctx.pipe_ctx[i].stream == stream) { @@ -721,6 +733,8 @@ void dc_stream_set_dither_option(struct dc_stream_state *stream, if (option > DITHER_OPTION_MAX) return; + dc_exit_ips_for_hw_access(stream->ctx->dc); + stream->dither_option = option; memset(¶ms, 0, sizeof(params)); @@ -745,6 +759,8 @@ bool dc_stream_set_gamut_remap(struct dc *dc, const struct dc_stream_state *stre bool ret = false; struct pipe_ctx *pipes; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { if (dc->current_state->res_ctx.pipe_ctx[i].stream == stream) { pipes = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -762,6 +778,8 @@ bool dc_stream_program_csc_matrix(struct dc *dc, struct dc_stream_state *stream) bool ret = false; struct pipe_ctx *pipes; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { if (dc->current_state->res_ctx.pipe_ctx[i].stream == stream) { @@ -788,6 +806,8 @@ void dc_stream_set_static_screen_params(struct dc *dc, struct pipe_ctx *pipes_affected[MAX_PIPES]; int num_pipes_affected = 0; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < num_streams; i++) { struct dc_stream_state *stream = streams[i]; @@ -1766,6 +1786,8 @@ void dc_enable_stereo( int i, j; struct pipe_ctx *pipe; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { if (context != NULL) { pipe = &context->res_ctx.pipe_ctx[i]; @@ -1785,6 +1807,8 @@ void dc_enable_stereo( void dc_trigger_sync(struct dc *dc, struct dc_state *context) { if (context->stream_count > 1 && !dc->debug.disable_timing_sync) { + dc_exit_ips_for_hw_access(dc); + enable_timing_multisync(dc, context); program_timing_sync(dc, context); } @@ -2041,6 +2065,8 @@ enum dc_status dc_commit_streams(struct dc *dc, if (!streams_changed(dc, streams, stream_count)) return res; + dc_exit_ips_for_hw_access(dc); + DC_LOG_DC("%s: %d streams\n", __func__, stream_count); for (i = 0; i < stream_count; i++) { @@ -3067,6 +3093,10 @@ static bool update_planes_and_stream_state(struct dc *dc, if (otg_master && otg_master->stream->test_pattern.type != DP_TEST_PATTERN_VIDEO_MODE) resource_build_test_pattern_params(&context->res_ctx, otg_master); + + if (otg_master && (otg_master->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422 || + otg_master->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)) + resource_build_subsampling_params(&context->res_ctx, otg_master); } } @@ -3376,6 +3406,8 @@ static void commit_planes_for_stream_fast(struct dc *dc, int i, j; struct pipe_ctx *top_pipe_to_program = NULL; struct dc_stream_status *stream_status = NULL; + dc_exit_ips_for_hw_access(dc); + dc_z10_restore(dc); top_pipe_to_program = resource_get_otg_master_for_stream( @@ -3503,10 +3535,23 @@ 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_exit_ips_for_hw_access(dc); + dc_z10_restore(dc); if (update_type == UPDATE_TYPE_FULL) wait_for_outstanding_hw_updates(dc, context); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->plane_state) { + set_p_state_switch_method(dc, context, pipe); + + if (dc->debug.visual_confirm) + dc_update_visual_confirm_color(dc, context, pipe); + } + } + if (update_type == UPDATE_TYPE_FULL) { dc_allow_idle_optimizations(dc, false); @@ -3541,17 +3586,6 @@ static void commit_planes_for_stream(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 && pipe->plane_state) { - set_p_state_switch_method(dc, context, pipe); - - if (dc->debug.visual_confirm) - dc_update_visual_confirm_color(dc, context, pipe); - } - } - if (stream->test_pattern.type != DP_TEST_PATTERN_VIDEO_MODE) { struct pipe_ctx *mpcc_pipe; struct pipe_ctx *odm_pipe; @@ -3817,7 +3851,9 @@ static void commit_planes_for_stream(struct dc *dc, * programming has completed (we turn on phantom OTG in order * to complete the plane disable for phantom pipes). */ - dc->hwss.apply_ctx_to_hw(dc, context); + + if (dc->hwss.disable_phantom_streams) + dc->hwss.disable_phantom_streams(dc, context); } if (update_type != UPDATE_TYPE_FAST) @@ -4382,6 +4418,8 @@ bool dc_update_planes_and_stream(struct dc *dc, bool is_plane_addition = 0; bool is_fast_update_only; + dc_exit_ips_for_hw_access(dc); + populate_fast_updates(fast_update, srf_updates, surface_count, stream_update); is_fast_update_only = fast_update_only(dc, fast_update, srf_updates, surface_count, stream_update, stream); @@ -4502,6 +4540,8 @@ void dc_commit_updates_for_stream(struct dc *dc, int i, j; struct dc_fast_update fast_update[MAX_SURFACES] = {0}; + dc_exit_ips_for_hw_access(dc); + populate_fast_updates(fast_update, srf_updates, surface_count, stream_update); stream_status = dc_stream_get_status(stream); context = dc->current_state; @@ -4686,6 +4726,8 @@ void dc_set_power_state( case DC_ACPI_CM_POWER_STATE_D0: dc_state_construct(dc, dc->current_state); + dc_exit_ips_for_hw_access(dc); + dc_z10_restore(dc); dc->hwss.init_hw(dc); @@ -4827,6 +4869,12 @@ void dc_allow_idle_optimizations(struct dc *dc, bool allow) dc->idle_optimizations_allowed = allow; } +void dc_exit_ips_for_hw_access(struct dc *dc) +{ + if (dc->caps.ips_support) + dc_allow_idle_optimizations(dc, false); +} + bool dc_dmub_is_ips_idle_state(struct dc *dc) { uint32_t idle_state = 0; 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 9fbdb09697fd..96ea283bd169 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -822,6 +822,16 @@ static struct rect calculate_odm_slice_in_timing_active(struct pipe_ctx *pipe_ct stream->timing.v_border_bottom + stream->timing.v_border_top; + /* Recout for ODM slices after the first slice need one extra left edge pixel + * for 3-tap chroma subsampling. + */ + if (odm_slice_idx > 0 && + (pipe_ctx->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422 || + pipe_ctx->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420)) { + odm_rec.x -= 1; + odm_rec.width += 1; + } + return odm_rec; } @@ -1438,6 +1448,7 @@ void resource_build_test_pattern_params(struct resource_context *res_ctx, enum controller_dp_test_pattern controller_test_pattern; enum controller_dp_color_space controller_color_space; enum dc_color_depth color_depth = otg_master->stream->timing.display_color_depth; + enum dc_pixel_encoding pixel_encoding = otg_master->stream->timing.pixel_encoding; int h_active = otg_master->stream->timing.h_addressable + otg_master->stream->timing.h_border_left + otg_master->stream->timing.h_border_right; @@ -1469,10 +1480,36 @@ void resource_build_test_pattern_params(struct resource_context *res_ctx, else params->width = last_odm_slice_width; + /* Extra left edge pixel is required for 3-tap chroma subsampling. */ + if (i != 0 && (pixel_encoding == PIXEL_ENCODING_YCBCR422 || + pixel_encoding == PIXEL_ENCODING_YCBCR420)) { + params->offset -= 1; + params->width += 1; + } + offset += odm_slice_width; } } +void resource_build_subsampling_params(struct resource_context *res_ctx, + struct pipe_ctx *otg_master) +{ + struct pipe_ctx *opp_heads[MAX_PIPES]; + int odm_cnt = 1; + int i; + + odm_cnt = resource_get_opp_heads_for_otg_master(otg_master, res_ctx, opp_heads); + + /* For ODM slices after the first slice, extra left edge pixel is required + * for 3-tap chroma subsampling. + */ + if (otg_master->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR422 || + otg_master->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) { + for (i = 0; i < odm_cnt; i++) + opp_heads[i]->stream_res.left_edge_extra_pixel = (i == 0) ? false : true; + } +} + bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; @@ -1834,23 +1871,6 @@ int resource_find_any_free_pipe(struct resource_context *new_res_ctx, 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 && diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c index 88c6436b28b6..180ac47868c2 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c @@ -291,11 +291,14 @@ void dc_state_destruct(struct dc_state *state) dc_stream_release(state->phantom_streams[i]); state->phantom_streams[i] = NULL; } + state->phantom_stream_count = 0; for (i = 0; i < state->phantom_plane_count; i++) { dc_plane_state_release(state->phantom_planes[i]); state->phantom_planes[i] = NULL; } + state->phantom_plane_count = 0; + state->stream_mask = 0; memset(&state->res_ctx, 0, sizeof(state->res_ctx)); memset(&state->pp_display_cfg, 0, sizeof(state->pp_display_cfg)); 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 54670e0b1518..51a970fcb5d0 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -423,6 +423,8 @@ bool dc_stream_add_writeback(struct dc *dc, return false; } + dc_exit_ips_for_hw_access(dc); + wb_info->dwb_params.out_transfer_func = stream->out_transfer_func; dwb = dc->res_pool->dwbc[wb_info->dwb_pipe_inst]; @@ -493,6 +495,8 @@ bool dc_stream_fc_disable_writeback(struct dc *dc, return false; } + dc_exit_ips_for_hw_access(dc); + if (dwb->funcs->set_fc_enable) dwb->funcs->set_fc_enable(dwb, DWB_FRAME_CAPTURE_DISABLE); @@ -542,6 +546,8 @@ bool dc_stream_remove_writeback(struct dc *dc, return false; } + dc_exit_ips_for_hw_access(dc); + /* disable writeback */ if (dc->hwss.disable_writeback) { struct dwbc *dwb = dc->res_pool->dwbc[dwb_pipe_inst]; @@ -557,6 +563,8 @@ bool dc_stream_warmup_writeback(struct dc *dc, int num_dwb, struct dc_writeback_info *wb_info) { + dc_exit_ips_for_hw_access(dc); + if (dc->hwss.mmhubbub_warmup) return dc->hwss.mmhubbub_warmup(dc, num_dwb, wb_info); else @@ -569,6 +577,8 @@ uint32_t dc_stream_get_vblank_counter(const struct dc_stream_state *stream) struct resource_context *res_ctx = &dc->current_state->res_ctx; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { struct timing_generator *tg = res_ctx->pipe_ctx[i].stream_res.tg; @@ -597,6 +607,8 @@ bool dc_stream_send_dp_sdp(const struct dc_stream_state *stream, dc = stream->ctx->dc; res_ctx = &dc->current_state->res_ctx; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[i]; @@ -628,6 +640,8 @@ bool dc_stream_get_scanoutpos(const struct dc_stream_state *stream, struct resource_context *res_ctx = &dc->current_state->res_ctx; + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < MAX_PIPES; i++) { struct timing_generator *tg = res_ctx->pipe_ctx[i].stream_res.tg; @@ -664,6 +678,8 @@ bool dc_stream_dmdata_status_done(struct dc *dc, struct dc_stream_state *stream) if (i == MAX_PIPES) return true; + dc_exit_ips_for_hw_access(dc); + return dc->hwss.dmdata_status_done(pipe); } @@ -698,6 +714,8 @@ bool dc_stream_set_dynamic_metadata(struct dc *dc, pipe_ctx->stream->dmdata_address = attr->address; + dc_exit_ips_for_hw_access(dc); + dc->hwss.program_dmdata_engine(pipe_ctx); if (hubp->funcs->dmdata_set_attributes != NULL && diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c index 19a2c7140ae8..19140fb65787 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c @@ -161,6 +161,8 @@ const struct dc_plane_status *dc_plane_get_status( break; } + dc_exit_ips_for_hw_access(dc); + for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index c9317ea0258e..c789cc2e216d 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -51,7 +51,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.266" +#define DC_VER "3.2.271" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -429,12 +429,12 @@ struct dc_config { bool force_bios_enable_lttpr; uint8_t force_bios_fixed_vs; int sdpif_request_limit_words_per_umc; - bool use_old_fixed_vs_sequence; bool dc_mode_clk_limit_support; bool EnableMinDispClkODM; bool enable_auto_dpm_test_logs; unsigned int disable_ips; unsigned int disable_ips_in_vpb; + bool usb4_bw_alloc_support; }; enum visual_confirm { @@ -987,9 +987,11 @@ struct dc_debug_options { bool psp_disabled_wa; unsigned int ips2_eval_delay_us; unsigned int ips2_entry_delay_us; + bool disable_dmub_reallow_idle; bool disable_timeout; bool disable_extblankadj; unsigned int static_screen_wait_frames; + bool force_chroma_subsampling_1tap; }; struct gpu_info_soc_bounding_box_v1_0; @@ -1068,6 +1070,7 @@ struct dc { } scratch; struct dml2_configuration_options dml2_options; + enum dc_acpi_cm_power_state power_state; }; enum frame_buffer_mode { @@ -2219,11 +2222,9 @@ struct dc_sink_dsc_caps { // 'true' if these are virtual DPCD's DSC caps (immediately upstream of sink in MST topology), // 'false' if they are sink's DSC caps bool is_virtual_dpcd_dsc; -#if defined(CONFIG_DRM_AMD_DC_FP) // 'true' if MST topology supports DSC passthrough for sink // 'false' if MST topology does not support DSC passthrough bool is_dsc_passthrough_supported; -#endif struct dsc_dec_dpcd_caps dsc_dec_caps; }; @@ -2325,6 +2326,7 @@ bool dc_is_plane_eligible_for_idle_optimizations(struct dc *dc, struct dc_plane_ struct dc_cursor_attributes *cursor_attr); void dc_allow_idle_optimizations(struct dc *dc, bool allow); +void dc_exit_ips_for_hw_access(struct dc *dc); bool dc_dmub_is_ips_idle_state(struct dc *dc); /* set min and max memory clock to lowest and highest DPM level, respectively */ 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 2b79a0e5638e..a1477906fe4f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -74,7 +74,10 @@ void dc_dmub_srv_wait_idle(struct dc_dmub_srv *dc_dmub_srv) struct dc_context *dc_ctx = dc_dmub_srv->ctx; enum dmub_status status; - status = dmub_srv_wait_for_idle(dmub, 100000); + do { + status = dmub_srv_wait_for_idle(dmub, 100000); + } while (dc_dmub_srv->ctx->dc->debug.disable_timeout && status != DMUB_STATUS_OK); + if (status != DMUB_STATUS_OK) { DC_ERROR("Error waiting for DMUB idle: status=%d\n", status); dc_dmub_srv_log_diagnostic_data(dc_dmub_srv); @@ -145,7 +148,9 @@ bool dc_dmub_srv_cmd_list_queue_execute(struct dc_dmub_srv *dc_dmub_srv, if (status == DMUB_STATUS_POWER_STATE_D3) return false; - dmub_srv_wait_for_idle(dmub, 100000); + do { + status = dmub_srv_wait_for_idle(dmub, 100000); + } while (dc_dmub_srv->ctx->dc->debug.disable_timeout && status != DMUB_STATUS_OK); /* Requeue the command. */ status = dmub_srv_cmd_queue(dmub, &cmd_list[i]); @@ -186,7 +191,9 @@ bool dc_dmub_srv_wait_for_idle(struct dc_dmub_srv *dc_dmub_srv, // Wait for DMUB to process command if (wait_type != DM_DMUB_WAIT_TYPE_NO_WAIT) { - status = dmub_srv_wait_for_idle(dmub, 100000); + do { + status = dmub_srv_wait_for_idle(dmub, 100000); + } while (dc_dmub_srv->ctx->dc->debug.disable_timeout && status != DMUB_STATUS_OK); if (status != DMUB_STATUS_OK) { DC_LOG_DEBUG("No reply for DMUB command: status=%d\n", status); @@ -780,21 +787,22 @@ static void populate_subvp_cmd_pipe_info(struct dc *dc, } else if (subvp_pipe->next_odm_pipe) { pipe_data->pipe_config.subvp_data.main_split_pipe_index = subvp_pipe->next_odm_pipe->pipe_idx; } else { - pipe_data->pipe_config.subvp_data.main_split_pipe_index = 0; + pipe_data->pipe_config.subvp_data.main_split_pipe_index = 0xF; } // Find phantom pipe index based on phantom stream for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *phantom_pipe = &context->res_ctx.pipe_ctx[j]; - if (phantom_pipe->stream == dc_state_get_paired_subvp_stream(context, subvp_pipe->stream)) { + if (resource_is_pipe_type(phantom_pipe, OTG_MASTER) && + phantom_pipe->stream == dc_state_get_paired_subvp_stream(context, subvp_pipe->stream)) { pipe_data->pipe_config.subvp_data.phantom_pipe_index = phantom_pipe->stream_res.tg->inst; if (phantom_pipe->bottom_pipe) { pipe_data->pipe_config.subvp_data.phantom_split_pipe_index = phantom_pipe->bottom_pipe->plane_res.hubp->inst; } else if (phantom_pipe->next_odm_pipe) { pipe_data->pipe_config.subvp_data.phantom_split_pipe_index = phantom_pipe->next_odm_pipe->plane_res.hubp->inst; } else { - pipe_data->pipe_config.subvp_data.phantom_split_pipe_index = 0; + pipe_data->pipe_config.subvp_data.phantom_split_pipe_index = 0xF; } break; } @@ -1195,6 +1203,9 @@ static void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle) if (dc->debug.dmcub_emulation) return; + if (!dc->ctx->dmub_srv || !dc->ctx->dmub_srv->dmub) + return; + memset(&cmd, 0, sizeof(cmd)); cmd.idle_opt_notify_idle.header.type = DMUB_CMD__IDLE_OPT; cmd.idle_opt_notify_idle.header.sub_type = DMUB_CMD__IDLE_OPT_DCN_NOTIFY_IDLE; @@ -1205,13 +1216,15 @@ static void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle) cmd.idle_opt_notify_idle.cntl_data.driver_idle = allow_idle; if (allow_idle) { + dc_dmub_srv_wait_idle(dc->ctx->dmub_srv); + if (dc->hwss.set_idle_state) dc->hwss.set_idle_state(dc, true); } /* NOTE: This does not use the "wake" interface since this is part of the wake path. */ /* We also do not perform a wait since DMCUB could enter idle after the notification. */ - dm_execute_dmub_cmd(dc->ctx, &cmd, DM_DMUB_WAIT_TYPE_NO_WAIT); + dm_execute_dmub_cmd(dc->ctx, &cmd, allow_idle ? DM_DMUB_WAIT_TYPE_NO_WAIT : DM_DMUB_WAIT_TYPE_WAIT); } static void dc_dmub_srv_exit_low_power_state(const struct dc *dc) @@ -1361,7 +1374,7 @@ bool dc_wake_and_execute_dmub_cmd_list(const struct dc_context *ctx, unsigned in else result = dm_execute_dmub_cmd(ctx, cmd, wait_type); - if (result && reallow_idle) + if (result && reallow_idle && !ctx->dc->debug.disable_dmub_reallow_idle) dc_dmub_srv_apply_idle_power_optimizations(ctx->dc, true); return result; @@ -1410,7 +1423,7 @@ bool dc_wake_and_execute_gpint(const struct dc_context *ctx, enum dmub_gpint_com result = dc_dmub_execute_gpint(ctx, command_code, param, response, wait_type); - if (result && reallow_idle) + if (result && reallow_idle && !ctx->dc->debug.disable_dmub_reallow_idle) dc_dmub_srv_apply_idle_power_optimizations(ctx->dc, true); return result; diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index 811474f4419b..aae2f3a2660d 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -827,9 +827,7 @@ struct dc_dsc_config { uint32_t version_minor; /* DSC minor version. Full version is formed as 1.version_minor. */ bool ycbcr422_simple; /* Tell DSC engine to convert YCbCr 4:2:2 to 'YCbCr 4:2:2 simple'. */ int32_t rc_buffer_size; /* DSC RC buffer block size in bytes */ -#if defined(CONFIG_DRM_AMD_DC_FP) bool is_frl; /* indicate if DSC is applied based on HDMI FRL sink's capability */ -#endif bool is_dp; /* indicate if DSC is applied based on DP's capability */ uint32_t mst_pbn; /* pbn of display on dsc mst hub */ const struct dc_dsc_rc_params_override *rc_params_ovrd; /* DM owned memory. If not NULL, apply custom dsc rc params */ @@ -942,6 +940,7 @@ struct dc_crtc_timing { uint32_t hdmi_vic; uint32_t rid; uint32_t fr_index; + uint32_t frl_uncompressed_video_bandwidth_in_kbps; enum dc_timing_3d_format timing_3d_format; enum dc_color_depth display_color_depth; enum dc_pixel_encoding pixel_encoding; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c index f0458b8f00af..12f3c35b3a34 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.c @@ -239,27 +239,294 @@ static void check_audio_bandwidth_hdmi( } } } +static struct fixed31_32 get_link_symbol_clk_freq_mhz(enum dc_link_rate link_rate) +{ + switch (link_rate) { + case LINK_RATE_LOW: + return dc_fixpt_from_int(162); /* 162 MHz */ + case LINK_RATE_HIGH: + return dc_fixpt_from_int(270); /* 270 MHz */ + case LINK_RATE_HIGH2: + return dc_fixpt_from_int(540); /* 540 MHz */ + case LINK_RATE_HIGH3: + return dc_fixpt_from_int(810); /* 810 MHz */ + case LINK_RATE_UHBR10: + return dc_fixpt_from_fraction(3125, 10); /* 312.5 MHz */ + case LINK_RATE_UHBR13_5: + return dc_fixpt_from_fraction(421875, 1000); /* 421.875 MHz */ + case LINK_RATE_UHBR20: + return dc_fixpt_from_int(625); /* 625 MHz */ + default: + /* Unexpected case, this requires debug if encountered. */ + ASSERT(0); + return dc_fixpt_from_int(0); + } +} + +struct dp_audio_layout_config { + uint8_t layouts_per_sample_denom; + uint8_t symbols_per_layout; + uint8_t max_layouts_per_audio_sdp; +}; + +static void get_audio_layout_config( + uint32_t channel_count, + enum dp_link_encoding encoding, + struct dp_audio_layout_config *output) +{ + /* Assuming L-PCM audio. Current implementation uses max 1 layout per SDP, + * with each layout being the same size (8ch layout). + */ + if (encoding == DP_8b_10b_ENCODING) { + if (channel_count == 2) { + output->layouts_per_sample_denom = 4; + output->symbols_per_layout = 40; + output->max_layouts_per_audio_sdp = 1; + } else if (channel_count == 8 || channel_count == 6) { + output->layouts_per_sample_denom = 1; + output->symbols_per_layout = 40; + output->max_layouts_per_audio_sdp = 1; + } + } else if (encoding == DP_128b_132b_ENCODING) { + if (channel_count == 2) { + output->layouts_per_sample_denom = 4; + output->symbols_per_layout = 10; + output->max_layouts_per_audio_sdp = 1; + } else if (channel_count == 8 || channel_count == 6) { + output->layouts_per_sample_denom = 1; + output->symbols_per_layout = 10; + output->max_layouts_per_audio_sdp = 1; + } + } +} -/*For DP SST, calculate if specified sample rates can fit into a given timing */ -static void check_audio_bandwidth_dpsst( +static uint32_t get_av_stream_map_lane_count( + enum dp_link_encoding encoding, + enum dc_lane_count lane_count, + bool is_mst) +{ + uint32_t av_stream_map_lane_count = 0; + + if (encoding == DP_8b_10b_ENCODING) { + if (!is_mst) + av_stream_map_lane_count = lane_count; + else + av_stream_map_lane_count = 4; + } else if (encoding == DP_128b_132b_ENCODING) { + av_stream_map_lane_count = 4; + } + + ASSERT(av_stream_map_lane_count != 0); + + return av_stream_map_lane_count; +} + +static uint32_t get_audio_sdp_overhead( + enum dp_link_encoding encoding, + enum dc_lane_count lane_count, + bool is_mst) +{ + uint32_t audio_sdp_overhead = 0; + + if (encoding == DP_8b_10b_ENCODING) { + if (is_mst) + audio_sdp_overhead = 16; /* 4 * 2 + 8 */ + else + audio_sdp_overhead = lane_count * 2 + 8; + } else if (encoding == DP_128b_132b_ENCODING) { + audio_sdp_overhead = 10; /* 4 x 2.5 */ + } + + ASSERT(audio_sdp_overhead != 0); + + return audio_sdp_overhead; +} + +static uint32_t calculate_required_audio_bw_in_symbols( const struct audio_crtc_info *crtc_info, + const struct dp_audio_layout_config *layout_config, uint32_t channel_count, - union audio_sample_rates *sample_rates) + uint32_t sample_rate_hz, + uint32_t av_stream_map_lane_count, + uint32_t audio_sdp_overhead) +{ + /* DP spec recommends between 1.05 to 1.1 safety margin to prevent sample under-run */ + struct fixed31_32 audio_sdp_margin = dc_fixpt_from_fraction(110, 100); + struct fixed31_32 horizontal_line_freq_khz = dc_fixpt_from_fraction( + crtc_info->requested_pixel_clock_100Hz, crtc_info->h_total * 10); + struct fixed31_32 samples_per_line; + struct fixed31_32 layouts_per_line; + struct fixed31_32 symbols_per_sdp_max_layout; + struct fixed31_32 remainder; + uint32_t num_sdp_with_max_layouts; + uint32_t required_symbols_per_hblank; + + samples_per_line = dc_fixpt_from_fraction(sample_rate_hz, 1000); + samples_per_line = dc_fixpt_div(samples_per_line, horizontal_line_freq_khz); + layouts_per_line = dc_fixpt_div_int(samples_per_line, layout_config->layouts_per_sample_denom); + + num_sdp_with_max_layouts = dc_fixpt_floor( + dc_fixpt_div_int(layouts_per_line, layout_config->max_layouts_per_audio_sdp)); + symbols_per_sdp_max_layout = dc_fixpt_from_int( + layout_config->max_layouts_per_audio_sdp * layout_config->symbols_per_layout); + symbols_per_sdp_max_layout = dc_fixpt_add_int(symbols_per_sdp_max_layout, audio_sdp_overhead); + symbols_per_sdp_max_layout = dc_fixpt_mul(symbols_per_sdp_max_layout, audio_sdp_margin); + required_symbols_per_hblank = num_sdp_with_max_layouts; + required_symbols_per_hblank *= ((dc_fixpt_ceil(symbols_per_sdp_max_layout) + av_stream_map_lane_count) / + av_stream_map_lane_count) * av_stream_map_lane_count; + + if (num_sdp_with_max_layouts != dc_fixpt_ceil( + dc_fixpt_div_int(layouts_per_line, layout_config->max_layouts_per_audio_sdp))) { + remainder = dc_fixpt_sub_int(layouts_per_line, + num_sdp_with_max_layouts * layout_config->max_layouts_per_audio_sdp); + remainder = dc_fixpt_mul_int(remainder, layout_config->symbols_per_layout); + remainder = dc_fixpt_add_int(remainder, audio_sdp_overhead); + remainder = dc_fixpt_mul(remainder, audio_sdp_margin); + required_symbols_per_hblank += ((dc_fixpt_ceil(remainder) + av_stream_map_lane_count) / + av_stream_map_lane_count) * av_stream_map_lane_count; + } + + return required_symbols_per_hblank; +} + +/* Current calculation only applicable for 8b/10b MST and 128b/132b SST/MST. + */ +static uint32_t calculate_available_hblank_bw_in_symbols( + const struct audio_crtc_info *crtc_info, + const struct audio_dp_link_info *dp_link_info) { - /* do nothing */ + uint64_t hblank = crtc_info->h_total - crtc_info->h_active; + struct fixed31_32 hblank_time_msec = + dc_fixpt_from_fraction(hblank * 10, crtc_info->requested_pixel_clock_100Hz); + struct fixed31_32 lsclkfreq_mhz = + get_link_symbol_clk_freq_mhz(dp_link_info->link_rate); + struct fixed31_32 average_stream_sym_bw_frac; + struct fixed31_32 peak_stream_bw_kbps; + struct fixed31_32 bits_per_pixel; + struct fixed31_32 link_bw_kbps; + struct fixed31_32 available_stream_sym_count; + uint32_t available_hblank_bw = 0; /* in stream symbols */ + + if (crtc_info->dsc_bits_per_pixel) { + bits_per_pixel = dc_fixpt_from_fraction(crtc_info->dsc_bits_per_pixel, 16); + } else { + switch (crtc_info->color_depth) { + case COLOR_DEPTH_666: + bits_per_pixel = dc_fixpt_from_int(6); + break; + case COLOR_DEPTH_888: + bits_per_pixel = dc_fixpt_from_int(8); + break; + case COLOR_DEPTH_101010: + bits_per_pixel = dc_fixpt_from_int(10); + break; + case COLOR_DEPTH_121212: + bits_per_pixel = dc_fixpt_from_int(12); + break; + default: + /* Default to commonly supported color depth. */ + bits_per_pixel = dc_fixpt_from_int(8); + break; + } + + bits_per_pixel = dc_fixpt_mul_int(bits_per_pixel, 3); + + if (crtc_info->pixel_encoding == PIXEL_ENCODING_YCBCR422) { + bits_per_pixel = dc_fixpt_div_int(bits_per_pixel, 3); + bits_per_pixel = dc_fixpt_mul_int(bits_per_pixel, 2); + } else if (crtc_info->pixel_encoding == PIXEL_ENCODING_YCBCR420) { + bits_per_pixel = dc_fixpt_div_int(bits_per_pixel, 2); + } + } + + /* Use simple stream BW calculation because mainlink overhead is + * accounted for separately in the audio BW calculations. + */ + peak_stream_bw_kbps = dc_fixpt_from_fraction(crtc_info->requested_pixel_clock_100Hz, 10); + peak_stream_bw_kbps = dc_fixpt_mul(peak_stream_bw_kbps, bits_per_pixel); + link_bw_kbps = dc_fixpt_from_int(dp_link_info->link_bandwidth_kbps); + average_stream_sym_bw_frac = dc_fixpt_div(peak_stream_bw_kbps, link_bw_kbps); + + available_stream_sym_count = dc_fixpt_mul_int(hblank_time_msec, 1000); + available_stream_sym_count = dc_fixpt_mul(available_stream_sym_count, lsclkfreq_mhz); + available_stream_sym_count = dc_fixpt_mul(available_stream_sym_count, average_stream_sym_bw_frac); + available_hblank_bw = dc_fixpt_floor(available_stream_sym_count); + available_hblank_bw *= dp_link_info->lane_count; + available_hblank_bw -= crtc_info->dsc_num_slices * 4; /* EOC overhead */ + + if (available_hblank_bw < dp_link_info->hblank_min_symbol_width) + available_hblank_bw = dp_link_info->hblank_min_symbol_width; + + if (available_hblank_bw < 12) + available_hblank_bw = 0; + else + available_hblank_bw -= 12; /* Main link overhead */ + + return available_hblank_bw; } -/*For DP MST, calculate if specified sample rates can fit into a given timing */ -static void check_audio_bandwidth_dpmst( +static void check_audio_bandwidth_dp( const struct audio_crtc_info *crtc_info, + const struct audio_dp_link_info *dp_link_info, uint32_t channel_count, union audio_sample_rates *sample_rates) { - /* do nothing */ + struct dp_audio_layout_config layout_config = {0}; + uint32_t available_hblank_bw; + uint32_t av_stream_map_lane_count; + uint32_t audio_sdp_overhead; + + /* TODO: Add validation for SST 8b/10 case */ + if (!dp_link_info->is_mst && dp_link_info->encoding == DP_8b_10b_ENCODING) + return; + + available_hblank_bw = calculate_available_hblank_bw_in_symbols( + crtc_info, dp_link_info); + av_stream_map_lane_count = get_av_stream_map_lane_count( + dp_link_info->encoding, dp_link_info->lane_count, dp_link_info->is_mst); + audio_sdp_overhead = get_audio_sdp_overhead( + dp_link_info->encoding, dp_link_info->lane_count, dp_link_info->is_mst); + get_audio_layout_config( + channel_count, dp_link_info->encoding, &layout_config); + + if (layout_config.max_layouts_per_audio_sdp == 0 || + layout_config.symbols_per_layout == 0 || + layout_config.layouts_per_sample_denom == 0) { + return; + } + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 192000, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_192 = 0; + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 176400, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_176_4 = 0; + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 96000, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_96 = 0; + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 88200, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_88_2 = 0; + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 48000, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_48 = 0; + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 44100, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_44_1 = 0; + if (available_hblank_bw < calculate_required_audio_bw_in_symbols( + crtc_info, &layout_config, channel_count, 32000, + av_stream_map_lane_count, audio_sdp_overhead)) + sample_rates->rate.RATE_32 = 0; } static void check_audio_bandwidth( const struct audio_crtc_info *crtc_info, + const struct audio_dp_link_info *dp_link_info, uint32_t channel_count, enum signal_type signal, union audio_sample_rates *sample_rates) @@ -271,12 +538,9 @@ static void check_audio_bandwidth( break; case SIGNAL_TYPE_EDP: case SIGNAL_TYPE_DISPLAY_PORT: - check_audio_bandwidth_dpsst( - crtc_info, channel_count, sample_rates); - break; case SIGNAL_TYPE_DISPLAY_PORT_MST: - check_audio_bandwidth_dpmst( - crtc_info, channel_count, sample_rates); + check_audio_bandwidth_dp( + crtc_info, dp_link_info, channel_count, sample_rates); break; default: break; @@ -394,7 +658,8 @@ void dce_aud_az_configure( struct audio *audio, enum signal_type signal, const struct audio_crtc_info *crtc_info, - const struct audio_info *audio_info) + const struct audio_info *audio_info, + const struct audio_dp_link_info *dp_link_info) { struct dce_audio *aud = DCE_AUD(audio); @@ -529,6 +794,7 @@ void dce_aud_az_configure( check_audio_bandwidth( crtc_info, + dp_link_info, channel_count, signal, &sample_rates); @@ -588,6 +854,7 @@ void dce_aud_az_configure( check_audio_bandwidth( crtc_info, + dp_link_info, 8, signal, &sample_rate); diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h index dbd2cfed0603..539f881928d1 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_audio.h @@ -170,7 +170,8 @@ void dce_aud_az_disable(struct audio *audio); void dce_aud_az_configure(struct audio *audio, enum signal_type signal, const struct audio_crtc_info *crtc_info, - const struct audio_info *audio_info); + const struct audio_info *audio_info, + const struct audio_dp_link_info *dp_link_info); void dce_aud_wall_dto_setup(struct audio *audio, enum signal_type signal, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c index 38e4797e9476..b010814706fe 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c @@ -258,7 +258,7 @@ static void dmub_replay_residency(struct dmub_replay *dmub, uint8_t panel_inst, *residency = 0; } -/** +/* * Set REPLAY power optimization flags and coasting vtotal. */ static void dmub_replay_set_power_opt_and_coasting_vtotal(struct dmub_replay *dmub, @@ -280,7 +280,7 @@ static void dmub_replay_set_power_opt_and_coasting_vtotal(struct dmub_replay *dm dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT); } -/** +/* * send Replay general cmd to DMUB. */ static void dmub_replay_send_cmd(struct dmub_replay *dmub, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index 3538973bd0c6..b7e57aa27361 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -62,6 +62,26 @@ void cm_helper_program_color_matrices( } +void cm_helper_read_color_matrices(struct dc_context *ctx, + uint16_t *regval, + const struct color_matrices_reg *reg) +{ + uint32_t cur_csc_reg, regval0, regval1; + unsigned int i = 0; + + for (cur_csc_reg = reg->csc_c11_c12; + cur_csc_reg <= reg->csc_c33_c34; cur_csc_reg++) { + REG_GET_2(cur_csc_reg, + csc_c11, ®val0, + csc_c12, ®val1); + + regval[2 * i] = regval0; + regval[(2 * i) + 1] = regval1; + + i++; + } +} + void cm_helper_program_xfer_func( struct dc_context *ctx, const struct pwl_params *params, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h index 0a68b63d6126..decc50b1ac53 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h @@ -114,5 +114,7 @@ bool cm_helper_translate_curve_to_degamma_hw_format( const struct dc_transfer_func *output_tf, struct pwl_params *lut_params); - +void cm_helper_read_color_matrices(struct dc_context *ctx, + uint16_t *regval, + const struct color_matrices_reg *reg); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index ef52e6b6eccf..4e391fd1d71c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -543,7 +543,8 @@ static const struct dpp_funcs dcn10_dpp_funcs = { .dpp_set_hdr_multiplier = dpp1_set_hdr_multiplier, .dpp_program_blnd_lut = NULL, .dpp_program_shaper_lut = NULL, - .dpp_program_3dlut = NULL + .dpp_program_3dlut = NULL, + .dpp_get_gamut_remap = dpp1_cm_get_gamut_remap, }; static struct dpp_caps dcn10_dpp_cap = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h index c9e045666dcc..a039eedc7c24 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.h @@ -1521,4 +1521,7 @@ void dpp1_construct(struct dcn10_dpp *dpp1, const struct dcn_dpp_registers *tf_regs, const struct dcn_dpp_shift *tf_shift, const struct dcn_dpp_mask *tf_mask); + +void dpp1_cm_get_gamut_remap(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust); #endif diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c index 904c2d278998..2f994a3a0b9c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_cm.c @@ -98,7 +98,7 @@ static void program_gamut_remap( if (regval == NULL || select == GAMUT_REMAP_BYPASS) { REG_SET(CM_GAMUT_REMAP_CONTROL, 0, - CM_GAMUT_REMAP_MODE, 0); + CM_GAMUT_REMAP_MODE, 0); return; } switch (select) { @@ -181,6 +181,74 @@ void dpp1_cm_set_gamut_remap( } } +static void read_gamut_remap(struct dcn10_dpp *dpp, + uint16_t *regval, + enum gamut_remap_select *select) +{ + struct color_matrices_reg gam_regs; + uint32_t selection; + + REG_GET(CM_GAMUT_REMAP_CONTROL, + CM_GAMUT_REMAP_MODE, &selection); + + *select = selection; + + gam_regs.shifts.csc_c11 = dpp->tf_shift->CM_GAMUT_REMAP_C11; + gam_regs.masks.csc_c11 = dpp->tf_mask->CM_GAMUT_REMAP_C11; + gam_regs.shifts.csc_c12 = dpp->tf_shift->CM_GAMUT_REMAP_C12; + gam_regs.masks.csc_c12 = dpp->tf_mask->CM_GAMUT_REMAP_C12; + + if (*select == GAMUT_REMAP_COEFF) { + + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_C33_C34); + + cm_helper_read_color_matrices( + dpp->base.ctx, + regval, + &gam_regs); + + } else if (*select == GAMUT_REMAP_COMA_COEFF) { + + gam_regs.csc_c11_c12 = REG(CM_COMA_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_COMA_C33_C34); + + cm_helper_read_color_matrices( + dpp->base.ctx, + regval, + &gam_regs); + + } else if (*select == GAMUT_REMAP_COMB_COEFF) { + + gam_regs.csc_c11_c12 = REG(CM_COMB_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_COMB_C33_C34); + + cm_helper_read_color_matrices( + dpp->base.ctx, + regval, + &gam_regs); + } +} + +void dpp1_cm_get_gamut_remap(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust) +{ + struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base); + uint16_t arr_reg_val[12]; + enum gamut_remap_select select; + + read_gamut_remap(dpp, arr_reg_val, &select); + + if (select == GAMUT_REMAP_BYPASS) { + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; + return; + } + + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW; + convert_hw_matrix(adjust->temperature_matrix, + arr_reg_val, ARRAY_SIZE(arr_reg_val)); +} + static void dpp1_cm_program_color_matrix( struct dcn10_dpp *dpp, const uint16_t *regval) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c index 0dec57679269..48a40dcc7050 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c @@ -23,6 +23,7 @@ * */ +#include "core_types.h" #include "dm_services.h" #include "dcn10_opp.h" #include "reg_helper.h" @@ -160,6 +161,9 @@ static void opp1_set_pixel_encoding( struct dcn10_opp *oppn10, const struct clamping_and_pixel_encoding_params *params) { + bool force_chroma_subsampling_1tap = + oppn10->base.ctx->dc->debug.force_chroma_subsampling_1tap; + switch (params->pixel_encoding) { case PIXEL_ENCODING_RGB: @@ -178,6 +182,9 @@ static void opp1_set_pixel_encoding( default: break; } + + if (force_chroma_subsampling_1tap) + REG_UPDATE(FMT_CONTROL, FMT_SUBSAMPLING_MODE, 0); } /** diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index eaa7032f0f1a..1516c0a48726 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -55,21 +55,23 @@ void dpp20_read_state(struct dpp *dpp_base, REG_GET(DPP_CONTROL, DPP_CLOCK_ENABLE, &s->is_enabled); + + // Degamma LUT (RAM) REG_GET(CM_DGAM_CONTROL, - CM_DGAM_LUT_MODE, &s->dgam_lut_mode); - // BGAM has no ROM, and definition is different, can't reuse same dump - //REG_GET(CM_BLNDGAM_CONTROL, - // CM_BLNDGAM_LUT_MODE, &s->rgam_lut_mode); - REG_GET(CM_GAMUT_REMAP_CONTROL, - CM_GAMUT_REMAP_MODE, &s->gamut_remap_mode); - if (s->gamut_remap_mode) { - s->gamut_remap_c11_c12 = REG_READ(CM_GAMUT_REMAP_C11_C12); - s->gamut_remap_c13_c14 = REG_READ(CM_GAMUT_REMAP_C13_C14); - s->gamut_remap_c21_c22 = REG_READ(CM_GAMUT_REMAP_C21_C22); - s->gamut_remap_c23_c24 = REG_READ(CM_GAMUT_REMAP_C23_C24); - s->gamut_remap_c31_c32 = REG_READ(CM_GAMUT_REMAP_C31_C32); - s->gamut_remap_c33_c34 = REG_READ(CM_GAMUT_REMAP_C33_C34); - } + CM_DGAM_LUT_MODE, &s->dgam_lut_mode); + + // Shaper LUT (RAM), 3D LUT (mode, bit-depth, size) + REG_GET(CM_SHAPER_CONTROL, + CM_SHAPER_LUT_MODE, &s->shaper_lut_mode); + REG_GET_2(CM_3DLUT_READ_WRITE_CONTROL, + CM_3DLUT_CONFIG_STATUS, &s->lut3d_mode, + CM_3DLUT_30BIT_EN, &s->lut3d_bit_depth); + REG_GET(CM_3DLUT_MODE, + CM_3DLUT_SIZE, &s->lut3d_size); + + // Blend/Out Gamma (RAM) + REG_GET(CM_BLNDGAM_LUT_WRITE_EN_MASK, + CM_BLNDGAM_CONFIG_STATUS, &s->rgam_lut_mode); } void dpp2_power_on_obuf( @@ -393,6 +395,7 @@ static struct dpp_funcs dcn20_dpp_funcs = { .set_optional_cursor_attributes = dpp1_cnv_set_optional_cursor_attributes, .dpp_dppclk_control = dpp1_dppclk_control, .dpp_set_hdr_multiplier = dpp2_set_hdr_multiplier, + .dpp_get_gamut_remap = dpp2_cm_get_gamut_remap, }; static struct dpp_caps dcn20_dpp_cap = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h index e735363d0051..672cde46c4b9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.h @@ -775,4 +775,7 @@ bool dpp2_construct(struct dcn20_dpp *dpp2, void dpp2_power_on_obuf( struct dpp *dpp_base, bool power_on); + +void dpp2_cm_get_gamut_remap(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust); #endif /* __DC_HWSS_DCN20_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c index 598caa508d43..58dc69926e8a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp_cm.c @@ -234,6 +234,61 @@ void dpp2_cm_set_gamut_remap( } } +static void read_gamut_remap(struct dcn20_dpp *dpp, + uint16_t *regval, + enum dcn20_gamut_remap_select *select) +{ + struct color_matrices_reg gam_regs; + uint32_t selection; + + IX_REG_GET(CM_TEST_DEBUG_INDEX, CM_TEST_DEBUG_DATA, + CM_TEST_DEBUG_DATA_STATUS_IDX, + CM_TEST_DEBUG_DATA_GAMUT_REMAP_MODE, &selection); + + *select = selection; + + gam_regs.shifts.csc_c11 = dpp->tf_shift->CM_GAMUT_REMAP_C11; + gam_regs.masks.csc_c11 = dpp->tf_mask->CM_GAMUT_REMAP_C11; + gam_regs.shifts.csc_c12 = dpp->tf_shift->CM_GAMUT_REMAP_C12; + gam_regs.masks.csc_c12 = dpp->tf_mask->CM_GAMUT_REMAP_C12; + + if (*select == DCN2_GAMUT_REMAP_COEF_A) { + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_C33_C34); + + cm_helper_read_color_matrices(dpp->base.ctx, + regval, + &gam_regs); + + } else if (*select == DCN2_GAMUT_REMAP_COEF_B) { + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_B_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_B_C33_C34); + + cm_helper_read_color_matrices(dpp->base.ctx, + regval, + &gam_regs); + } +} + +void dpp2_cm_get_gamut_remap(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust) +{ + struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); + uint16_t arr_reg_val[12]; + enum dcn20_gamut_remap_select select; + + read_gamut_remap(dpp, arr_reg_val, &select); + + if (select == DCN2_GAMUT_REMAP_BYPASS) { + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; + return; + } + + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW; + convert_hw_matrix(adjust->temperature_matrix, + arr_reg_val, ARRAY_SIZE(arr_reg_val)); +} + void dpp2_program_input_csc( struct dpp *dpp_base, enum dc_color_space color_space, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c index 5da6e44f284a..16b5ff208d14 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c @@ -542,8 +542,30 @@ static struct mpcc *mpc2_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id) return NULL; } +static void mpc2_read_mpcc_state( + struct mpc *mpc, + int mpcc_inst, + struct mpcc_state *s) +{ + struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); + + REG_GET(MPCC_OPP_ID[mpcc_inst], MPCC_OPP_ID, &s->opp_id); + REG_GET(MPCC_TOP_SEL[mpcc_inst], MPCC_TOP_SEL, &s->dpp_id); + REG_GET(MPCC_BOT_SEL[mpcc_inst], MPCC_BOT_SEL, &s->bot_mpcc_id); + REG_GET_4(MPCC_CONTROL[mpcc_inst], MPCC_MODE, &s->mode, + MPCC_ALPHA_BLND_MODE, &s->alpha_mode, + MPCC_ALPHA_MULTIPLIED_MODE, &s->pre_multiplied_alpha, + MPCC_BLND_ACTIVE_OVERLAP_ONLY, &s->overlap_only); + REG_GET_2(MPCC_STATUS[mpcc_inst], MPCC_IDLE, &s->idle, + MPCC_BUSY, &s->busy); + + /* Gamma block state */ + REG_GET(MPCC_OGAM_LUT_RAM_CONTROL[mpcc_inst], + MPCC_OGAM_CONFIG_STATUS, &s->rgam_mode); +} + static const struct mpc_funcs dcn20_mpc_funcs = { - .read_mpcc_state = mpc1_read_mpcc_state, + .read_mpcc_state = mpc2_read_mpcc_state, .insert_plane = mpc1_insert_plane, .remove_mpcc = mpc1_remove_mpcc, .mpc_init = mpc1_mpc_init, diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dpp.c index a7268027a472..f809a7d21033 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dpp.c @@ -275,6 +275,7 @@ static struct dpp_funcs dcn201_dpp_funcs = { .set_optional_cursor_attributes = dpp1_cnv_set_optional_cursor_attributes, .dpp_dppclk_control = dpp1_dppclk_control, .dpp_set_hdr_multiplier = dpp2_set_hdr_multiplier, + .dpp_get_gamut_remap = dpp2_cm_get_gamut_remap, }; static struct dpp_caps dcn201_dpp_cap = { 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 11f7746f3a65..a3a769aad042 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c @@ -44,12 +44,45 @@ void dpp30_read_state(struct dpp *dpp_base, struct dcn_dpp_state *s) { struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base); + uint32_t gamcor_lut_mode, rgam_lut_mode; REG_GET(DPP_CONTROL, - DPP_CLOCK_ENABLE, &s->is_enabled); + DPP_CLOCK_ENABLE, &s->is_enabled); + + // Pre-degamma (ROM) + REG_GET_2(PRE_DEGAM, + PRE_DEGAM_MODE, &s->pre_dgam_mode, + PRE_DEGAM_SELECT, &s->pre_dgam_select); + + // Gamma Correction (RAM) + REG_GET(CM_GAMCOR_CONTROL, + CM_GAMCOR_MODE_CURRENT, &s->gamcor_mode); + if (s->gamcor_mode) { + REG_GET(CM_GAMCOR_CONTROL, CM_GAMCOR_SELECT_CURRENT, &gamcor_lut_mode); + if (!gamcor_lut_mode) + s->gamcor_mode = LUT_RAM_A; // Otherwise, LUT_RAM_B + } - // TODO: Implement for DCN3 + // Shaper LUT (RAM), 3D LUT (mode, bit-depth, size) + REG_GET(CM_SHAPER_CONTROL, + CM_SHAPER_LUT_MODE, &s->shaper_lut_mode); + REG_GET(CM_3DLUT_MODE, + CM_3DLUT_MODE_CURRENT, &s->lut3d_mode); + REG_GET(CM_3DLUT_READ_WRITE_CONTROL, + CM_3DLUT_30BIT_EN, &s->lut3d_bit_depth); + REG_GET(CM_3DLUT_MODE, + CM_3DLUT_SIZE, &s->lut3d_size); + + // Blend/Out Gamma (RAM) + REG_GET(CM_BLNDGAM_CONTROL, + CM_BLNDGAM_MODE_CURRENT, &s->rgam_lut_mode); + if (s->rgam_lut_mode){ + REG_GET(CM_BLNDGAM_CONTROL, CM_BLNDGAM_SELECT_CURRENT, &rgam_lut_mode); + if (!rgam_lut_mode) + s->rgam_lut_mode = LUT_RAM_A; // Otherwise, LUT_RAM_B + } } + /*program post scaler scs block in dpp CM*/ void dpp3_program_post_csc( struct dpp *dpp_base, @@ -1462,6 +1495,7 @@ static struct dpp_funcs dcn30_dpp_funcs = { .set_optional_cursor_attributes = dpp1_cnv_set_optional_cursor_attributes, .dpp_dppclk_control = dpp1_dppclk_control, .dpp_set_hdr_multiplier = dpp3_set_hdr_multiplier, + .dpp_get_gamut_remap = dpp3_cm_get_gamut_remap, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.h index cea3208e4ab1..2ac8045a87a1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.h @@ -637,4 +637,6 @@ void dpp3_program_cm_dealpha( struct dpp *dpp_base, uint32_t enable, uint32_t additive_blending); +void dpp3_cm_get_gamut_remap(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust); #endif /* __DC_HWSS_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c index e43f77c11c00..54ec144f7b81 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp_cm.c @@ -408,3 +408,57 @@ void dpp3_cm_set_gamut_remap( program_gamut_remap(dpp, arr_reg_val, gamut_mode); } } + +static void read_gamut_remap(struct dcn3_dpp *dpp, + uint16_t *regval, + int *select) +{ + struct color_matrices_reg gam_regs; + uint32_t selection; + + //current coefficient set in use + REG_GET(CM_GAMUT_REMAP_CONTROL, CM_GAMUT_REMAP_MODE_CURRENT, &selection); + + *select = selection; + + gam_regs.shifts.csc_c11 = dpp->tf_shift->CM_GAMUT_REMAP_C11; + gam_regs.masks.csc_c11 = dpp->tf_mask->CM_GAMUT_REMAP_C11; + gam_regs.shifts.csc_c12 = dpp->tf_shift->CM_GAMUT_REMAP_C12; + gam_regs.masks.csc_c12 = dpp->tf_mask->CM_GAMUT_REMAP_C12; + + if (*select == GAMUT_REMAP_COEFF) { + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_C33_C34); + + cm_helper_read_color_matrices(dpp->base.ctx, + regval, + &gam_regs); + + } else if (*select == GAMUT_REMAP_COMA_COEFF) { + gam_regs.csc_c11_c12 = REG(CM_GAMUT_REMAP_B_C11_C12); + gam_regs.csc_c33_c34 = REG(CM_GAMUT_REMAP_B_C33_C34); + + cm_helper_read_color_matrices(dpp->base.ctx, + regval, + &gam_regs); + } +} + +void dpp3_cm_get_gamut_remap(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust) +{ + struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base); + uint16_t arr_reg_val[12]; + int select; + + read_gamut_remap(dpp, arr_reg_val, &select); + + if (select == GAMUT_REMAP_BYPASS) { + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; + return; + } + + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW; + convert_hw_matrix(adjust->temperature_matrix, + arr_reg_val, ARRAY_SIZE(arr_reg_val)); +} diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c index d1500b223858..bf3386cd444d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c @@ -1129,6 +1129,64 @@ void mpc3_set_gamut_remap( } } +static void read_gamut_remap(struct dcn30_mpc *mpc30, + int mpcc_id, + uint16_t *regval, + uint32_t *select) +{ + struct color_matrices_reg gam_regs; + + //current coefficient set in use + REG_GET(MPCC_GAMUT_REMAP_MODE[mpcc_id], MPCC_GAMUT_REMAP_MODE_CURRENT, select); + + gam_regs.shifts.csc_c11 = mpc30->mpc_shift->MPCC_GAMUT_REMAP_C11_A; + gam_regs.masks.csc_c11 = mpc30->mpc_mask->MPCC_GAMUT_REMAP_C11_A; + gam_regs.shifts.csc_c12 = mpc30->mpc_shift->MPCC_GAMUT_REMAP_C12_A; + gam_regs.masks.csc_c12 = mpc30->mpc_mask->MPCC_GAMUT_REMAP_C12_A; + + if (*select == GAMUT_REMAP_COEFF) { + gam_regs.csc_c11_c12 = REG(MPC_GAMUT_REMAP_C11_C12_A[mpcc_id]); + gam_regs.csc_c33_c34 = REG(MPC_GAMUT_REMAP_C33_C34_A[mpcc_id]); + + cm_helper_read_color_matrices( + mpc30->base.ctx, + regval, + &gam_regs); + + } else if (*select == GAMUT_REMAP_COMA_COEFF) { + + gam_regs.csc_c11_c12 = REG(MPC_GAMUT_REMAP_C11_C12_B[mpcc_id]); + gam_regs.csc_c33_c34 = REG(MPC_GAMUT_REMAP_C33_C34_B[mpcc_id]); + + cm_helper_read_color_matrices( + mpc30->base.ctx, + regval, + &gam_regs); + + } + +} + +void mpc3_get_gamut_remap(struct mpc *mpc, + int mpcc_id, + struct mpc_grph_gamut_adjustment *adjust) +{ + struct dcn30_mpc *mpc30 = TO_DCN30_MPC(mpc); + uint16_t arr_reg_val[12]; + int select; + + read_gamut_remap(mpc30, mpcc_id, arr_reg_val, &select); + + if (select == GAMUT_REMAP_BYPASS) { + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_BYPASS; + return; + } + + adjust->gamut_adjust_type = GRAPHICS_GAMUT_ADJUST_TYPE_SW; + convert_hw_matrix(adjust->temperature_matrix, + arr_reg_val, ARRAY_SIZE(arr_reg_val)); +} + bool mpc3_program_3dlut( struct mpc *mpc, const struct tetrahedral_params *params, @@ -1382,8 +1440,54 @@ static void mpc3_set_mpc_mem_lp_mode(struct mpc *mpc) } } +static void mpc3_read_mpcc_state( + struct mpc *mpc, + int mpcc_inst, + struct mpcc_state *s) +{ + struct dcn30_mpc *mpc30 = TO_DCN30_MPC(mpc); + uint32_t rmu_status = 0xf; + + REG_GET(MPCC_OPP_ID[mpcc_inst], MPCC_OPP_ID, &s->opp_id); + REG_GET(MPCC_TOP_SEL[mpcc_inst], MPCC_TOP_SEL, &s->dpp_id); + REG_GET(MPCC_BOT_SEL[mpcc_inst], MPCC_BOT_SEL, &s->bot_mpcc_id); + REG_GET_4(MPCC_CONTROL[mpcc_inst], MPCC_MODE, &s->mode, + MPCC_ALPHA_BLND_MODE, &s->alpha_mode, + MPCC_ALPHA_MULTIPLIED_MODE, &s->pre_multiplied_alpha, + MPCC_BLND_ACTIVE_OVERLAP_ONLY, &s->overlap_only); + REG_GET_2(MPCC_STATUS[mpcc_inst], MPCC_IDLE, &s->idle, + MPCC_BUSY, &s->busy); + + /* Color blocks state */ + REG_GET(MPC_RMU_CONTROL, MPC_RMU0_MUX_STATUS, &rmu_status); + + if (rmu_status == mpcc_inst) { + REG_GET(SHAPER_CONTROL[0], + MPC_RMU_SHAPER_LUT_MODE_CURRENT, &s->shaper_lut_mode); + REG_GET(RMU_3DLUT_MODE[0], + MPC_RMU_3DLUT_MODE_CURRENT, &s->lut3d_mode); + REG_GET(RMU_3DLUT_READ_WRITE_CONTROL[0], + MPC_RMU_3DLUT_30BIT_EN, &s->lut3d_bit_depth); + REG_GET(RMU_3DLUT_MODE[0], + MPC_RMU_3DLUT_SIZE, &s->lut3d_size); + } else { + REG_GET(SHAPER_CONTROL[1], + MPC_RMU_SHAPER_LUT_MODE_CURRENT, &s->shaper_lut_mode); + REG_GET(RMU_3DLUT_MODE[1], + MPC_RMU_3DLUT_MODE_CURRENT, &s->lut3d_mode); + REG_GET(RMU_3DLUT_READ_WRITE_CONTROL[1], + MPC_RMU_3DLUT_30BIT_EN, &s->lut3d_bit_depth); + REG_GET(RMU_3DLUT_MODE[1], + MPC_RMU_3DLUT_SIZE, &s->lut3d_size); + } + + REG_GET_2(MPCC_OGAM_CONTROL[mpcc_inst], + MPCC_OGAM_MODE_CURRENT, &s->rgam_mode, + MPCC_OGAM_SELECT_CURRENT, &s->rgam_lut); +} + static const struct mpc_funcs dcn30_mpc_funcs = { - .read_mpcc_state = mpc1_read_mpcc_state, + .read_mpcc_state = mpc3_read_mpcc_state, .insert_plane = mpc1_insert_plane, .remove_mpcc = mpc1_remove_mpcc, .mpc_init = mpc1_mpc_init, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.h index 5198f2167c7c..9cb96ae95a2f 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.h @@ -1056,6 +1056,10 @@ void mpc3_set_gamut_remap( int mpcc_id, const struct mpc_grph_gamut_adjustment *adjust); +void mpc3_get_gamut_remap(struct mpc *mpc, + int mpcc_id, + struct mpc_grph_gamut_adjustment *adjust); + void mpc3_set_rmu_mux( struct mpc *mpc, int rmu_idx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dpp.c index dcf12a0b031c..681e75c6dbaf 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_dpp.c @@ -133,6 +133,7 @@ static struct dpp_funcs dcn32_dpp_funcs = { .set_optional_cursor_attributes = dpp1_cnv_set_optional_cursor_attributes, .dpp_dppclk_control = dpp1_dppclk_control, .dpp_set_hdr_multiplier = dpp3_set_hdr_multiplier, + .dpp_get_gamut_remap = dpp3_cm_get_gamut_remap, }; diff --git a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h index 4229369c57f4..f4d3f04ec857 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h +++ b/drivers/gpu/drm/amd/display/dc/dm_cp_psp.h @@ -26,6 +26,9 @@ #ifndef DM_CP_PSP_IF__H #define DM_CP_PSP_IF__H +/* + * Interface to CPLIB/PSP to enable ASSR + */ struct dc_link; struct cp_psp_stream_config { diff --git a/drivers/gpu/drm/amd/display/dc/dml/Makefile b/drivers/gpu/drm/amd/display/dc/dml/Makefile index 6042a5a6a44f..59ade76ffb18 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/Makefile +++ b/drivers/gpu/drm/amd/display/dc/dml/Makefile @@ -72,11 +72,11 @@ CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_vba.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn10/dcn10_fpu.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/dcn20_fpu.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_mode_vba_20v2.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn20/display_rq_dlg_calc_20v2.o := $(dml_ccflags) -CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_ccflags) +CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_mode_vba_21.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn21/display_rq_dlg_calc_21.o := $(dml_ccflags) CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_mode_vba_30.o := $(dml_ccflags) $(frame_warn_flag) CFLAGS_$(AMDDALPATH)/dc/dml/dcn30/display_rq_dlg_calc_30.o := $(dml_ccflags) 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 63c48c29ba49..e7f4a2d491cc 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 @@ -4273,7 +4273,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l //Calculate Swath, DET Configuration, DCFCLKDeepSleep // - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { for (k = 0; k < v->NumberOfActivePlanes; ++k) { v->RequiredDPPCLKThisState[k] = v->RequiredDPPCLK[i][j][k]; @@ -4576,7 +4576,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l //Calculate Return BW - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { for (k = 0; k <= v->NumberOfActivePlanes - 1; k++) { if (v->BlendingAndTiming[k] == k) { @@ -4635,7 +4635,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->UrgentOutOfOrderReturnPerChannelVMDataOnly); v->FinalDRAMClockChangeLatency = (v->DRAMClockChangeLatencyOverride > 0 ? v->DRAMClockChangeLatencyOverride : v->DRAMClockChangeLatency); - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { v->DCFCLKState[i][j] = v->DCFCLKPerState[i]; } @@ -4646,7 +4646,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l if (v->ClampMinDCFCLK) { /* Clamp calculated values to actual minimum */ - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { if (v->DCFCLKState[i][j] < mode_lib->soc.min_dcfclk) { v->DCFCLKState[i][j] = mode_lib->soc.min_dcfclk; @@ -4656,7 +4656,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } } - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { v->IdealSDPPortBandwidthPerState[i][j] = dml_min3( v->ReturnBusWidth * v->DCFCLKState[i][j], @@ -4674,7 +4674,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l //Re-ordering Buffer Support Check - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { if ((v->ROBBufferSizeInKByte - v->PixelChunkSizeInKByte) * 1024 / v->ReturnBWPerState[i][j] > (v->RoundTripPingLatencyCycles + 32) / v->DCFCLKState[i][j] + ReorderingBytes / v->ReturnBWPerState[i][j]) { @@ -4692,7 +4692,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l MaxTotalVActiveRDBandwidth = MaxTotalVActiveRDBandwidth + v->ReadBandwidthLuma[k] + v->ReadBandwidthChroma[k]; } - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { v->MaxTotalVerticalActiveAvailableBandwidth[i][j] = dml_min( v->IdealSDPPortBandwidthPerState[i][j] * v->MaxAveragePercentOfIdealSDPPortBWDisplayCanUseInNormalSystemOperation / 100, @@ -4708,7 +4708,7 @@ void dml30_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l //Prefetch Check - for (i = 0; i < mode_lib->soc.num_states; ++i) { + for (i = start_state; i < mode_lib->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { int NextPrefetchModeState = MinPrefetchMode; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c index 3eb3a021ab7d..3f02bb806d42 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c @@ -266,6 +266,17 @@ void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p optimal_uclk_for_dcfclk_sta_targets[i] = bw_params->clk_table.entries[j].memclk_mhz * 16; break; + } else { + /* condition where (dcfclk_sta_targets[i] >= optimal_dcfclk_for_uclk[j]): + * This is required for dcn303 because it just so happens that the memory + * bandwidth is low enough such that all the optimal DCFCLK for each UCLK + * is lower than the smallest DCFCLK STA target. In this case we need to + * populate the optimal UCLK for each DCFCLK STA target to be the max UCLK. + */ + if (j == num_uclk_states - 1) { + optimal_uclk_for_dcfclk_sta_targets[i] = + bw_params->clk_table.entries[j].memclk_mhz * 16; + } } } } 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 dd781a20692e..ba76dd4a2ce2 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 @@ -1288,7 +1288,7 @@ static bool update_pipes_with_split_flags(struct dc *dc, struct dc_state *contex return updated; } -static bool should_allow_odm_power_optimization(struct dc *dc, +static bool should_apply_odm_power_optimization(struct dc *dc, struct dc_state *context, struct vba_vars_st *v, int *split, bool *merge) { @@ -1392,9 +1392,12 @@ static void try_odm_power_optimization_and_revalidate( { int i; unsigned int new_vlevel; + unsigned int cur_policy[MAX_PIPES]; - for (i = 0; i < pipe_cnt; i++) + for (i = 0; i < pipe_cnt; i++) { + cur_policy[i] = pipes[i].pipe.dest.odm_combine_policy; pipes[i].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1; + } new_vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, pipe_cnt); @@ -1403,6 +1406,9 @@ static void try_odm_power_optimization_and_revalidate( memset(merge, 0, MAX_PIPES * sizeof(bool)); *vlevel = dcn20_validate_apply_pipe_split_flags(dc, context, new_vlevel, split, merge); context->bw_ctx.dml.vba.VoltageLevel = *vlevel; + } else { + for (i = 0; i < pipe_cnt; i++) + pipes[i].pipe.dest.odm_combine_policy = cur_policy[i]; } } @@ -1580,7 +1586,7 @@ static void dcn32_full_validate_bw_helper(struct dc *dc, } } - if (should_allow_odm_power_optimization(dc, context, vba, split, merge)) + if (should_apply_odm_power_optimization(dc, context, vba, split, merge)) try_odm_power_optimization_and_revalidate( dc, context, pipes, split, merge, vlevel, *pipe_cnt); @@ -2209,7 +2215,8 @@ bool dcn32_internal_validate_bw(struct dc *dc, int i; pipe_cnt = dc->res_pool->funcs->populate_dml_pipes(dc, context, pipes, fast_validate); - dcn32_update_dml_pipes_odm_policy_based_on_context(dc, context, pipes); + if (!dc->config.enable_windowed_mpo_odm) + dcn32_update_dml_pipes_odm_policy_based_on_context(dc, context, pipes); /* repopulate_pipes = 1 means the pipes were either split or merged. In this case * we have to re-calculate the DET allocation and run through DML once more to diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c index 7ea2bd5374d5..912256006d75 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c @@ -583,9 +583,9 @@ void dcn35_decide_zstate_support(struct dc *dc, struct dc_state *context) plane_count++; } - if (plane_count == 0) { + if (context->stream_count == 0 || plane_count == 0) { support = DCN_ZSTATE_SUPPORT_ALLOW; - } else if (plane_count == 1 && context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) { + } else if (context->stream_count == 1 && context->streams[0]->signal == SIGNAL_TYPE_EDP) { struct dc_link *link = context->streams[0]->sink->link; bool is_pwrseq0 = link && link->link_index == 0; bool is_psr1 = link && link->psr_settings.psr_version == DC_PSR_VERSION_1 && !link->panel_config.psr.disable_psr; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c index 0baf39d64a2d..a52c594e1ba4 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c @@ -141,14 +141,33 @@ static unsigned int find_pipes_assigned_to_plane(struct dml2_context *ctx, { int i; unsigned int num_found = 0; - unsigned int plane_id_assigned_to_pipe; + unsigned int plane_id_assigned_to_pipe = -1; for (i = 0; i < ctx->config.dcn_pipe_count; i++) { - if (state->res_ctx.pipe_ctx[i].plane_state && get_plane_id(ctx, state, state->res_ctx.pipe_ctx[i].plane_state, - state->res_ctx.pipe_ctx[i].stream->stream_id, - ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[state->res_ctx.pipe_ctx[i].pipe_idx], &plane_id_assigned_to_pipe)) { - if (plane_id_assigned_to_pipe == plane_id) - pipes[num_found++] = i; + struct pipe_ctx *pipe = &state->res_ctx.pipe_ctx[i]; + + if (!pipe->plane_state || !pipe->stream) + continue; + + get_plane_id(ctx, state, pipe->plane_state, pipe->stream->stream_id, + ctx->v20.scratch.dml_to_dc_pipe_mapping.dml_pipe_idx_to_plane_index[pipe->pipe_idx], + &plane_id_assigned_to_pipe); + if (plane_id_assigned_to_pipe == plane_id && !pipe->prev_odm_pipe + && (!pipe->top_pipe || pipe->top_pipe->plane_state != pipe->plane_state)) { + while (pipe) { + struct pipe_ctx *mpc_pipe = pipe; + + while (mpc_pipe) { + pipes[num_found++] = mpc_pipe->pipe_idx; + mpc_pipe = mpc_pipe->bottom_pipe; + if (!mpc_pipe) + break; + if (mpc_pipe->plane_state != pipe->plane_state) + mpc_pipe = NULL; + } + pipe = pipe->next_odm_pipe; + } + break; } } @@ -566,8 +585,14 @@ static unsigned int find_pipes_assigned_to_stream(struct dml2_context *ctx, stru unsigned int num_found = 0; for (i = 0; i < ctx->config.dcn_pipe_count; i++) { - if (state->res_ctx.pipe_ctx[i].stream && state->res_ctx.pipe_ctx[i].stream->stream_id == stream_id) { - pipes[num_found++] = i; + struct pipe_ctx *pipe = &state->res_ctx.pipe_ctx[i]; + + if (pipe->stream && pipe->stream->stream_id == stream_id && !pipe->top_pipe && !pipe->prev_odm_pipe) { + while (pipe) { + pipes[num_found++] = pipe->pipe_idx; + pipe = pipe->next_odm_pipe; + } + break; } } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c index 1068b962d1c1..f15d1dbad6a9 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_utils.c @@ -234,7 +234,7 @@ static bool get_plane_id(struct dml2_context *dml2, const struct dc_state *state if (state->streams[i]->stream_id == stream_id) { for (j = 0; j < state->stream_status[i].plane_count; j++) { if (state->stream_status[i].plane_states[j] == plane && - (!is_plane_duplicate || (is_plane_duplicate && (j == plane_index)))) { + (!is_plane_duplicate || (j == plane_index))) { *plane_id = (i << 16) | j; return true; } 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 0df6c55eb326..ac41f9c0a283 100644 --- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c +++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c @@ -137,6 +137,11 @@ uint32_t dc_bandwidth_in_kbps_from_timing( if (link_encoding == DC_LINK_ENCODING_DP_128b_132b) kbps = apply_128b_132b_stream_overhead(timing, kbps); + if (link_encoding == DC_LINK_ENCODING_HDMI_FRL && + timing->vic == 0 && timing->hdmi_vic == 0 && + timing->frl_uncompressed_video_bandwidth_in_kbps != 0) + kbps = timing->frl_uncompressed_video_bandwidth_in_kbps; + return kbps; } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index 2352428bcea3..9d5df4c0da59 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -1291,6 +1291,46 @@ static enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id) } } +static void populate_audio_dp_link_info( + const struct pipe_ctx *pipe_ctx, + struct audio_dp_link_info *dp_link_info) +{ + const struct dc_stream_state *stream = pipe_ctx->stream; + const struct dc_link *link = stream->link; + struct fixed31_32 link_bw_kbps; + + dp_link_info->encoding = link->dc->link_srv->dp_get_encoding_format( + &pipe_ctx->link_config.dp_link_settings); + dp_link_info->is_mst = (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST); + dp_link_info->lane_count = pipe_ctx->link_config.dp_link_settings.lane_count; + dp_link_info->link_rate = pipe_ctx->link_config.dp_link_settings.link_rate; + + link_bw_kbps = dc_fixpt_from_int(dc_link_bandwidth_kbps(link, + &pipe_ctx->link_config.dp_link_settings)); + + /* For audio stream calculations, the video stream should not include FEC or SSC + * in order to get the most pessimistic values. + */ + if (dp_link_info->encoding == DP_8b_10b_ENCODING && + link->dc->link_srv->dp_is_fec_supported(link)) { + link_bw_kbps = dc_fixpt_mul(link_bw_kbps, + dc_fixpt_from_fraction(100, DATA_EFFICIENCY_8b_10b_FEC_EFFICIENCY_x100)); + } else if (dp_link_info->encoding == DP_128b_132b_ENCODING) { + link_bw_kbps = dc_fixpt_mul(link_bw_kbps, + dc_fixpt_from_fraction(10000, 9975)); /* 99.75% SSC overhead*/ + } + + dp_link_info->link_bandwidth_kbps = dc_fixpt_floor(link_bw_kbps); + + /* HW minimum for 128b/132b HBlank is 4 frame symbols. + * TODO: Plumb the actual programmed HBlank min symbol width to here. + */ + if (dp_link_info->encoding == DP_128b_132b_ENCODING) + dp_link_info->hblank_min_symbol_width = 4; + else + dp_link_info->hblank_min_symbol_width = 0; +} + static void build_audio_output( struct dc_state *state, const struct pipe_ctx *pipe_ctx, @@ -1338,6 +1378,15 @@ static void build_audio_output( audio_output->crtc_info.calculated_pixel_clock_100Hz = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk_100hz; + audio_output->crtc_info.pixel_encoding = + stream->timing.pixel_encoding; + + audio_output->crtc_info.dsc_bits_per_pixel = + stream->timing.dsc_cfg.bits_per_pixel; + + audio_output->crtc_info.dsc_num_slices = + stream->timing.dsc_cfg.num_slices_h; + /*for HDMI, audio ACR is with deep color ratio factor*/ if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal) && audio_output->crtc_info.requested_pixel_clock_100Hz == @@ -1371,6 +1420,10 @@ static void build_audio_output( audio_output->pll_info.ss_percentage = pipe_ctx->pll_settings.ss_percentage; + + if (dc_is_dp_signal(pipe_ctx->stream->signal)) { + populate_audio_dp_link_info(pipe_ctx, &audio_output->dp_link_info); + } } static void program_scaler(const struct dc *dc, @@ -1476,7 +1529,7 @@ static enum dc_status dce110_enable_stream_timing( return DC_OK; } -static enum dc_status apply_single_controller_ctx_to_hw( +enum dc_status dce110_apply_single_controller_ctx_to_hw( struct pipe_ctx *pipe_ctx, struct dc_state *context, struct dc *dc) @@ -1507,7 +1560,8 @@ static enum dc_status apply_single_controller_ctx_to_hw( pipe_ctx->stream_res.audio, pipe_ctx->stream->signal, &audio_output.crtc_info, - &pipe_ctx->stream->audio_info); + &pipe_ctx->stream->audio_info, + &audio_output.dp_link_info); } /* make sure no pipes syncd to the pipe being enabled */ @@ -2302,7 +2356,7 @@ enum dc_status dce110_apply_ctx_to_hw( if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe) continue; - status = apply_single_controller_ctx_to_hw( + status = dce110_apply_single_controller_ctx_to_hw( pipe_ctx, context, dc); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h index 08028a1779ae..ed3cc3648e8e 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h @@ -39,6 +39,10 @@ enum dc_status dce110_apply_ctx_to_hw( struct dc *dc, struct dc_state *context); +enum dc_status dce110_apply_single_controller_ctx_to_hw( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); void dce110_enable_stream(struct pipe_ctx *pipe_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index 6dd479e8a348..314798400b16 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -283,33 +283,33 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx) DTN_INFO("\n"); } -void dcn10_log_hw_state(struct dc *dc, - struct dc_log_buffer_ctx *log_ctx) +static void dcn10_log_color_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx) { struct dc_context *dc_ctx = dc->ctx; struct resource_pool *pool = dc->res_pool; int i; - DTN_INFO_BEGIN(); - - dcn10_log_hubbub_state(dc, log_ctx); - - dcn10_log_hubp_states(dc, log_ctx); - - DTN_INFO("DPP: IGAM format IGAM mode DGAM mode RGAM mode" - " GAMUT mode C11 C12 C13 C14 C21 C22 C23 C24 " - "C31 C32 C33 C34\n"); + DTN_INFO("DPP: IGAM format IGAM mode DGAM mode RGAM mode" + " GAMUT adjust " + "C11 C12 C13 C14 " + "C21 C22 C23 C24 " + "C31 C32 C33 C34 \n"); for (i = 0; i < pool->pipe_count; i++) { struct dpp *dpp = pool->dpps[i]; struct dcn_dpp_state s = {0}; dpp->funcs->dpp_read_state(dpp, &s); + dpp->funcs->dpp_get_gamut_remap(dpp, &s.gamut_remap); if (!s.is_enabled) continue; - DTN_INFO("[%2d]: %11xh %-11s %-11s %-11s" - "%8x %08xh %08xh %08xh %08xh %08xh %08xh", + DTN_INFO("[%2d]: %11xh %11s %9s %9s" + " %12s " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld", dpp->inst, s.igam_input_format, (s.igam_lut_mode == 0) ? "BypassFixed" : @@ -329,16 +329,42 @@ void dcn10_log_hw_state(struct dc *dc, ((s.rgam_lut_mode == 3) ? "RAM" : ((s.rgam_lut_mode == 4) ? "RAM" : "Unknown")))), - s.gamut_remap_mode, - s.gamut_remap_c11_c12, - s.gamut_remap_c13_c14, - s.gamut_remap_c21_c22, - s.gamut_remap_c23_c24, - s.gamut_remap_c31_c32, - s.gamut_remap_c33_c34); + (s.gamut_remap.gamut_adjust_type == 0) ? "Bypass" : + ((s.gamut_remap.gamut_adjust_type == 1) ? "HW" : + "SW"), + s.gamut_remap.temperature_matrix[0].value, + s.gamut_remap.temperature_matrix[1].value, + s.gamut_remap.temperature_matrix[2].value, + s.gamut_remap.temperature_matrix[3].value, + s.gamut_remap.temperature_matrix[4].value, + s.gamut_remap.temperature_matrix[5].value, + s.gamut_remap.temperature_matrix[6].value, + s.gamut_remap.temperature_matrix[7].value, + s.gamut_remap.temperature_matrix[8].value, + s.gamut_remap.temperature_matrix[9].value, + s.gamut_remap.temperature_matrix[10].value, + s.gamut_remap.temperature_matrix[11].value); DTN_INFO("\n"); } DTN_INFO("\n"); + DTN_INFO("DPP Color Caps: input_lut_shared:%d icsc:%d" + " dgam_ram:%d dgam_rom: srgb:%d,bt2020:%d,gamma2_2:%d,pq:%d,hlg:%d" + " post_csc:%d gamcor:%d dgam_rom_for_yuv:%d 3d_lut:%d" + " blnd_lut:%d oscs:%d\n\n", + dc->caps.color.dpp.input_lut_shared, + dc->caps.color.dpp.icsc, + dc->caps.color.dpp.dgam_ram, + dc->caps.color.dpp.dgam_rom_caps.srgb, + dc->caps.color.dpp.dgam_rom_caps.bt2020, + dc->caps.color.dpp.dgam_rom_caps.gamma2_2, + dc->caps.color.dpp.dgam_rom_caps.pq, + dc->caps.color.dpp.dgam_rom_caps.hlg, + dc->caps.color.dpp.post_csc, + dc->caps.color.dpp.gamma_corr, + dc->caps.color.dpp.dgam_rom_for_yuv, + dc->caps.color.dpp.hw_3d_lut, + dc->caps.color.dpp.ogam_ram, + dc->caps.color.dpp.ocsc); DTN_INFO("MPCC: OPP DPP MPCCBOT MODE ALPHA_MODE PREMULT OVERLAP_ONLY IDLE\n"); for (i = 0; i < pool->pipe_count; i++) { @@ -352,6 +378,30 @@ void dcn10_log_hw_state(struct dc *dc, s.idle); } DTN_INFO("\n"); + DTN_INFO("MPC Color Caps: gamut_remap:%d, 3dlut:%d, ogam_ram:%d, ocsc:%d\n\n", + dc->caps.color.mpc.gamut_remap, + dc->caps.color.mpc.num_3dluts, + dc->caps.color.mpc.ogam_ram, + dc->caps.color.mpc.ocsc); +} + +void dcn10_log_hw_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx) +{ + struct dc_context *dc_ctx = dc->ctx; + struct resource_pool *pool = dc->res_pool; + int i; + + DTN_INFO_BEGIN(); + + dcn10_log_hubbub_state(dc, log_ctx); + + dcn10_log_hubp_states(dc, log_ctx); + + if (dc->hwss.log_color_state) + dc->hwss.log_color_state(dc, log_ctx); + else + dcn10_log_color_state(dc, log_ctx); DTN_INFO("OTG: v_bs v_be v_ss v_se vpol vmax vmin vmax_sel vmin_sel h_bs h_be h_ss h_se hpol htot vtot underflow blank_en\n"); @@ -1840,6 +1890,9 @@ bool dcn10_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, { struct dpp *dpp = pipe_ctx->plane_res.dpp; + if (!stream) + return false; + if (dpp == NULL) return false; @@ -1862,8 +1915,8 @@ bool dcn10_set_output_transfer_func(struct dc *dc, struct pipe_ctx *pipe_ctx, } else dpp->funcs->dpp_program_regamma_pwl(dpp, NULL, OPP_REGAMMA_BYPASS); - if (stream != NULL && stream->ctx != NULL && - stream->out_transfer_func != NULL) { + if (stream->ctx && + stream->out_transfer_func) { log_tf(stream->ctx, stream->out_transfer_func, dpp->regamma_params.hw_points_num); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index 4853ecac53f9..bc0a21957e33 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -71,6 +71,112 @@ #define FN(reg_name, field_name) \ hws->shifts->field_name, hws->masks->field_name +void dcn20_log_color_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx) +{ + struct dc_context *dc_ctx = dc->ctx; + struct resource_pool *pool = dc->res_pool; + int i; + + DTN_INFO("DPP: DGAM mode SHAPER mode 3DLUT mode 3DLUT bit depth" + " 3DLUT size RGAM mode GAMUT adjust " + "C11 C12 C13 C14 " + "C21 C22 C23 C24 " + "C31 C32 C33 C34 \n"); + + for (i = 0; i < pool->pipe_count; i++) { + struct dpp *dpp = pool->dpps[i]; + struct dcn_dpp_state s = {0}; + + dpp->funcs->dpp_read_state(dpp, &s); + dpp->funcs->dpp_get_gamut_remap(dpp, &s.gamut_remap); + + if (!s.is_enabled) + continue; + + DTN_INFO("[%2d]: %8s %11s %10s %15s %10s %9s %12s " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld", + dpp->inst, + (s.dgam_lut_mode == 0) ? "Bypass" : + ((s.dgam_lut_mode == 1) ? "sRGB" : + ((s.dgam_lut_mode == 2) ? "Ycc" : + ((s.dgam_lut_mode == 3) ? "RAM" : + ((s.dgam_lut_mode == 4) ? "RAM" : + "Unknown")))), + (s.shaper_lut_mode == 1) ? "RAM A" : + ((s.shaper_lut_mode == 2) ? "RAM B" : + "Bypass"), + (s.lut3d_mode == 1) ? "RAM A" : + ((s.lut3d_mode == 2) ? "RAM B" : + "Bypass"), + (s.lut3d_bit_depth <= 0) ? "12-bit" : "10-bit", + (s.lut3d_size == 0) ? "17x17x17" : "9x9x9", + (s.rgam_lut_mode == 1) ? "RAM A" : + ((s.rgam_lut_mode == 1) ? "RAM B" : "Bypass"), + (s.gamut_remap.gamut_adjust_type == 0) ? "Bypass" : + ((s.gamut_remap.gamut_adjust_type == 1) ? "HW" : + "SW"), + s.gamut_remap.temperature_matrix[0].value, + s.gamut_remap.temperature_matrix[1].value, + s.gamut_remap.temperature_matrix[2].value, + s.gamut_remap.temperature_matrix[3].value, + s.gamut_remap.temperature_matrix[4].value, + s.gamut_remap.temperature_matrix[5].value, + s.gamut_remap.temperature_matrix[6].value, + s.gamut_remap.temperature_matrix[7].value, + s.gamut_remap.temperature_matrix[8].value, + s.gamut_remap.temperature_matrix[9].value, + s.gamut_remap.temperature_matrix[10].value, + s.gamut_remap.temperature_matrix[11].value); + DTN_INFO("\n"); + } + DTN_INFO("\n"); + DTN_INFO("DPP Color Caps: input_lut_shared:%d icsc:%d" + " dgam_ram:%d dgam_rom: srgb:%d,bt2020:%d,gamma2_2:%d,pq:%d,hlg:%d" + " post_csc:%d gamcor:%d dgam_rom_for_yuv:%d 3d_lut:%d" + " blnd_lut:%d oscs:%d\n\n", + dc->caps.color.dpp.input_lut_shared, + dc->caps.color.dpp.icsc, + dc->caps.color.dpp.dgam_ram, + dc->caps.color.dpp.dgam_rom_caps.srgb, + dc->caps.color.dpp.dgam_rom_caps.bt2020, + dc->caps.color.dpp.dgam_rom_caps.gamma2_2, + dc->caps.color.dpp.dgam_rom_caps.pq, + dc->caps.color.dpp.dgam_rom_caps.hlg, + dc->caps.color.dpp.post_csc, + dc->caps.color.dpp.gamma_corr, + dc->caps.color.dpp.dgam_rom_for_yuv, + dc->caps.color.dpp.hw_3d_lut, + dc->caps.color.dpp.ogam_ram, + dc->caps.color.dpp.ocsc); + + DTN_INFO("MPCC: OPP DPP MPCCBOT MODE ALPHA_MODE PREMULT OVERLAP_ONLY IDLE" + " OGAM mode\n"); + + for (i = 0; i < pool->pipe_count; i++) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); + if (s.opp_id != 0xf) + DTN_INFO("[%2d]: %2xh %2xh %6xh %4d %10d %7d %12d %4d %9s\n", + i, s.opp_id, s.dpp_id, s.bot_mpcc_id, + s.mode, s.alpha_mode, s.pre_multiplied_alpha, s.overlap_only, + s.idle, + (s.rgam_mode == 1) ? "RAM A" : + ((s.rgam_mode == 2) ? "RAM B" : + "Bypass")); + } + DTN_INFO("\n"); + DTN_INFO("MPC Color Caps: gamut_remap:%d, 3dlut:%d, ogam_ram:%d, ocsc:%d\n\n", + dc->caps.color.mpc.gamut_remap, + dc->caps.color.mpc.num_3dluts, + dc->caps.color.mpc.ogam_ram, + dc->caps.color.mpc.ocsc); +} + + static int find_free_gsl_group(const struct dc *dc) { if (dc->res_pool->gsl_groups.gsl_0 == 0) @@ -1467,7 +1573,8 @@ static void dcn20_detect_pipe_changes(struct dc_state *old_state, * makes this assumption at the moment with how hubp reset is matched to * same index mpcc reset. */ - if (old_pipe->stream_res.opp != new_pipe->stream_res.opp) + if (old_pipe->stream_res.opp != new_pipe->stream_res.opp || + old_pipe->stream_res.left_edge_extra_pixel != new_pipe->stream_res.left_edge_extra_pixel) new_pipe->update_flags.bits.opp_changed = 1; if (old_pipe->stream_res.tg != new_pipe->stream_res.tg) new_pipe->update_flags.bits.tg_changed = 1; @@ -1853,6 +1960,10 @@ static void dcn20_program_pipe( pipe_ctx->stream_res.opp, &pipe_ctx->stream->bit_depth_params, &pipe_ctx->stream->clamping); + + pipe_ctx->stream_res.opp->funcs->opp_program_left_edge_extra_pixel( + pipe_ctx->stream_res.opp, + pipe_ctx->stream_res.left_edge_extra_pixel); } /* Set ABM pipe after other pipe configurations done */ @@ -1958,7 +2069,6 @@ void dcn20_program_front_end_for_ctx( && context->res_ctx.pipe_ctx[i].stream) hws->funcs.blank_pixel_data(dc, &context->res_ctx.pipe_ctx[i], true); - /* Disconnect mpcc */ for (i = 0; i < dc->res_pool->pipe_count; i++) if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable @@ -2561,7 +2671,7 @@ void dcn20_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx) tg->funcs->setup_vertical_interrupt2(tg, start_line); } -static void dcn20_reset_back_end_for_pipe( +void dcn20_reset_back_end_for_pipe( struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h index b94c85340abf..5c874f7b0683 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.h @@ -28,6 +28,8 @@ #include "hw_sequencer_private.h" +void dcn20_log_color_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx); bool dcn20_set_blend_lut( struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state); bool dcn20_set_shaper_3dlut( @@ -84,6 +86,10 @@ enum dc_status dcn20_enable_stream_timing( void dcn20_disable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn20_enable_stream_gating(struct dc *dc, struct pipe_ctx *pipe_ctx); void dcn20_setup_vupdate_interrupt(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_reset_back_end_for_pipe( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context); void dcn20_init_blank( struct dc *dc, struct timing_generator *tg); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c index 8e88dcaf88f5..5c7f380a84f9 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c @@ -206,28 +206,32 @@ void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx) void dcn21_set_pipe(struct pipe_ctx *pipe_ctx) { struct abm *abm = pipe_ctx->stream_res.abm; - uint32_t otg_inst = pipe_ctx->stream_res.tg->inst; + struct timing_generator *tg = pipe_ctx->stream_res.tg; struct panel_cntl *panel_cntl = pipe_ctx->stream->link->panel_cntl; struct dmcu *dmcu = pipe_ctx->stream->ctx->dc->res_pool->dmcu; + uint32_t otg_inst; + + if (!abm && !tg && !panel_cntl) + return; + + otg_inst = tg->inst; if (dmcu) { dce110_set_pipe(pipe_ctx); return; } - if (abm && panel_cntl) { - if (abm->funcs && abm->funcs->set_pipe_ex) { - abm->funcs->set_pipe_ex(abm, + if (abm->funcs && abm->funcs->set_pipe_ex) { + abm->funcs->set_pipe_ex(abm, otg_inst, SET_ABM_PIPE_NORMAL, panel_cntl->inst, panel_cntl->pwrseq_inst); - } else { - dmub_abm_set_pipe(abm, otg_inst, - SET_ABM_PIPE_NORMAL, - panel_cntl->inst, - panel_cntl->pwrseq_inst); - } + } else { + dmub_abm_set_pipe(abm, otg_inst, + SET_ABM_PIPE_NORMAL, + panel_cntl->inst, + panel_cntl->pwrseq_inst); } } @@ -237,34 +241,35 @@ bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx, { struct dc_context *dc = pipe_ctx->stream->ctx; struct abm *abm = pipe_ctx->stream_res.abm; + struct timing_generator *tg = pipe_ctx->stream_res.tg; struct panel_cntl *panel_cntl = pipe_ctx->stream->link->panel_cntl; + uint32_t otg_inst; + + if (!abm && !tg && !panel_cntl) + return false; + + otg_inst = tg->inst; if (dc->dc->res_pool->dmcu) { dce110_set_backlight_level(pipe_ctx, backlight_pwm_u16_16, frame_ramp); return true; } - if (abm != NULL) { - uint32_t otg_inst = pipe_ctx->stream_res.tg->inst; - - if (abm && panel_cntl) { - if (abm->funcs && abm->funcs->set_pipe_ex) { - abm->funcs->set_pipe_ex(abm, - otg_inst, - SET_ABM_PIPE_NORMAL, - panel_cntl->inst, - panel_cntl->pwrseq_inst); - } else { - dmub_abm_set_pipe(abm, - otg_inst, - SET_ABM_PIPE_NORMAL, - panel_cntl->inst, - panel_cntl->pwrseq_inst); - } - } + if (abm->funcs && abm->funcs->set_pipe_ex) { + abm->funcs->set_pipe_ex(abm, + otg_inst, + SET_ABM_PIPE_NORMAL, + panel_cntl->inst, + panel_cntl->pwrseq_inst); + } else { + dmub_abm_set_pipe(abm, + otg_inst, + SET_ABM_PIPE_NORMAL, + panel_cntl->inst, + panel_cntl->pwrseq_inst); } - if (abm && abm->funcs && abm->funcs->set_backlight_level_pwm) + if (abm->funcs && abm->funcs->set_backlight_level_pwm) abm->funcs->set_backlight_level_pwm(abm, backlight_pwm_u16_16, frame_ramp, 0, panel_cntl->inst); else diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c index c34c13e1e0a4..7e6b7f2a6dc9 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c @@ -69,6 +69,155 @@ #define FN(reg_name, field_name) \ hws->shifts->field_name, hws->masks->field_name +void dcn30_log_color_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx) +{ + struct dc_context *dc_ctx = dc->ctx; + struct resource_pool *pool = dc->res_pool; + int i; + + DTN_INFO("DPP: DGAM ROM DGAM ROM type DGAM LUT SHAPER mode" + " 3DLUT mode 3DLUT bit depth 3DLUT size RGAM mode" + " GAMUT adjust " + "C11 C12 C13 C14 " + "C21 C22 C23 C24 " + "C31 C32 C33 C34 \n"); + + for (i = 0; i < pool->pipe_count; i++) { + struct dpp *dpp = pool->dpps[i]; + struct dcn_dpp_state s = {0}; + + dpp->funcs->dpp_read_state(dpp, &s); + dpp->funcs->dpp_get_gamut_remap(dpp, &s.gamut_remap); + + if (!s.is_enabled) + continue; + + DTN_INFO("[%2d]: %7x %13s %8s %11s %10s %15s %10s %9s" + " %12s " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld", + dpp->inst, + s.pre_dgam_mode, + (s.pre_dgam_select == 0) ? "sRGB" : + ((s.pre_dgam_select == 1) ? "Gamma 2.2" : + ((s.pre_dgam_select == 2) ? "Gamma 2.4" : + ((s.pre_dgam_select == 3) ? "Gamma 2.6" : + ((s.pre_dgam_select == 4) ? "BT.709" : + ((s.pre_dgam_select == 5) ? "PQ" : + ((s.pre_dgam_select == 6) ? "HLG" : + "Unknown")))))), + (s.gamcor_mode == 0) ? "Bypass" : + ((s.gamcor_mode == 1) ? "RAM A" : + "RAM B"), + (s.shaper_lut_mode == 1) ? "RAM A" : + ((s.shaper_lut_mode == 2) ? "RAM B" : + "Bypass"), + (s.lut3d_mode == 1) ? "RAM A" : + ((s.lut3d_mode == 2) ? "RAM B" : + "Bypass"), + (s.lut3d_bit_depth <= 0) ? "12-bit" : "10-bit", + (s.lut3d_size == 0) ? "17x17x17" : "9x9x9", + (s.rgam_lut_mode == 0) ? "Bypass" : + ((s.rgam_lut_mode == 1) ? "RAM A" : + "RAM B"), + (s.gamut_remap.gamut_adjust_type == 0) ? "Bypass" : + ((s.gamut_remap.gamut_adjust_type == 1) ? "HW" : + "SW"), + s.gamut_remap.temperature_matrix[0].value, + s.gamut_remap.temperature_matrix[1].value, + s.gamut_remap.temperature_matrix[2].value, + s.gamut_remap.temperature_matrix[3].value, + s.gamut_remap.temperature_matrix[4].value, + s.gamut_remap.temperature_matrix[5].value, + s.gamut_remap.temperature_matrix[6].value, + s.gamut_remap.temperature_matrix[7].value, + s.gamut_remap.temperature_matrix[8].value, + s.gamut_remap.temperature_matrix[9].value, + s.gamut_remap.temperature_matrix[10].value, + s.gamut_remap.temperature_matrix[11].value); + DTN_INFO("\n"); + } + DTN_INFO("\n"); + DTN_INFO("DPP Color Caps: input_lut_shared:%d icsc:%d" + " dgam_ram:%d dgam_rom: srgb:%d,bt2020:%d,gamma2_2:%d,pq:%d,hlg:%d" + " post_csc:%d gamcor:%d dgam_rom_for_yuv:%d 3d_lut:%d" + " blnd_lut:%d oscs:%d\n\n", + dc->caps.color.dpp.input_lut_shared, + dc->caps.color.dpp.icsc, + dc->caps.color.dpp.dgam_ram, + dc->caps.color.dpp.dgam_rom_caps.srgb, + dc->caps.color.dpp.dgam_rom_caps.bt2020, + dc->caps.color.dpp.dgam_rom_caps.gamma2_2, + dc->caps.color.dpp.dgam_rom_caps.pq, + dc->caps.color.dpp.dgam_rom_caps.hlg, + dc->caps.color.dpp.post_csc, + dc->caps.color.dpp.gamma_corr, + dc->caps.color.dpp.dgam_rom_for_yuv, + dc->caps.color.dpp.hw_3d_lut, + dc->caps.color.dpp.ogam_ram, + dc->caps.color.dpp.ocsc); + + DTN_INFO("MPCC: OPP DPP MPCCBOT MODE ALPHA_MODE PREMULT OVERLAP_ONLY IDLE" + " SHAPER mode 3DLUT mode 3DLUT bit-depth 3DLUT size OGAM mode OGAM LUT" + " GAMUT adjust " + "C11 C12 C13 C14 " + "C21 C22 C23 C24 " + "C31 C32 C33 C34 \n"); + + for (i = 0; i < pool->pipe_count; i++) { + struct mpcc_state s = {0}; + + pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s); + mpc3_get_gamut_remap(pool->mpc, i, &s.gamut_remap); + + if (s.opp_id != 0xf) + DTN_INFO("[%2d]: %2xh %2xh %6xh %4d %10d %7d %12d %4d %11s %11s %16s %11s %10s %9s" + " %-12s " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld " + "%010lld %010lld %010lld %010lld\n", + i, s.opp_id, s.dpp_id, s.bot_mpcc_id, + s.mode, s.alpha_mode, s.pre_multiplied_alpha, s.overlap_only, + s.idle, + (s.shaper_lut_mode == 1) ? "RAM A" : + ((s.shaper_lut_mode == 2) ? "RAM B" : + "Bypass"), + (s.lut3d_mode == 1) ? "RAM A" : + ((s.lut3d_mode == 2) ? "RAM B" : + "Bypass"), + (s.lut3d_bit_depth <= 0) ? "12-bit" : "10-bit", + (s.lut3d_size == 0) ? "17x17x17" : "9x9x9", + (s.rgam_mode == 0) ? "Bypass" : + ((s.rgam_mode == 2) ? "RAM" : + "Unknown"), + (s.rgam_mode == 1) ? "B" : "A", + (s.gamut_remap.gamut_adjust_type == 0) ? "Bypass" : + ((s.gamut_remap.gamut_adjust_type == 1) ? "HW" : + "SW"), + s.gamut_remap.temperature_matrix[0].value, + s.gamut_remap.temperature_matrix[1].value, + s.gamut_remap.temperature_matrix[2].value, + s.gamut_remap.temperature_matrix[3].value, + s.gamut_remap.temperature_matrix[4].value, + s.gamut_remap.temperature_matrix[5].value, + s.gamut_remap.temperature_matrix[6].value, + s.gamut_remap.temperature_matrix[7].value, + s.gamut_remap.temperature_matrix[8].value, + s.gamut_remap.temperature_matrix[9].value, + s.gamut_remap.temperature_matrix[10].value, + s.gamut_remap.temperature_matrix[11].value); + + } + DTN_INFO("\n"); + DTN_INFO("MPC Color Caps: gamut_remap:%d, 3dlut:%d, ogam_ram:%d, ocsc:%d\n\n", + dc->caps.color.mpc.gamut_remap, + dc->caps.color.mpc.num_3dluts, + dc->caps.color.mpc.ogam_ram, + dc->caps.color.mpc.ocsc); +} + bool dcn30_set_blend_lut( struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state) { @@ -1015,21 +1164,3 @@ void dcn30_prepare_bandwidth(struct dc *dc, if (!dc->clk_mgr->clks.fw_based_mclk_switching) dc_dmub_srv_p_state_delegate(dc, false, context); } - -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/hwss/dcn30/dcn30_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h index e557e2b98618..638f018a3cb5 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h @@ -52,6 +52,9 @@ bool dcn30_mmhubbub_warmup( unsigned int num_dwb, struct dc_writeback_info *wb_info); +void dcn30_log_color_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx); + bool dcn30_set_blend_lut(struct pipe_ctx *pipe_ctx, const struct dc_plane_state *plane_state); @@ -90,7 +93,4 @@ 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/hwss/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c index 9894caedffed..ef913445a795 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/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 = dcn30_set_static_screen_control, + .set_static_screen_control = dcn10_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/hwss/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c index 7423880fabb6..a760f0c6fe98 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c @@ -98,10 +98,8 @@ static void enable_memory_low_power(struct dc *dc) for (i = 0; i < dc->res_pool->stream_enc_count; i++) if (dc->res_pool->stream_enc[i]->vpg) dc->res_pool->stream_enc[i]->vpg->funcs->vpg_powerdown(dc->res_pool->stream_enc[i]->vpg); -#if defined(CONFIG_DRM_AMD_DC_FP) for (i = 0; i < dc->res_pool->hpo_dp_stream_enc_count; i++) dc->res_pool->hpo_dp_stream_enc[i]->vpg->funcs->vpg_powerdown(dc->res_pool->hpo_dp_stream_enc[i]->vpg); -#endif } } @@ -617,3 +615,21 @@ void dcn31_setup_hpo_hw_control(const struct dce_hwseq *hws, bool enable) if (hws->ctx->dc->debug.hpo_optimization) REG_UPDATE(HPO_TOP_HW_CONTROL, HPO_IO_EN, !!enable); } + +void dcn31_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/hwss/dcn31/dcn31_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.h index edfc01d6ad73..b8bc939da155 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.h @@ -56,4 +56,8 @@ bool dcn31_is_abm_supported(struct dc *dc, void dcn31_init_pipes(struct dc *dc, struct dc_state *context); void dcn31_setup_hpo_hw_control(const struct dce_hwseq *hws, bool enable); +void dcn31_set_static_screen_control(struct pipe_ctx **pipe_ctx, + int num_pipes, const struct dc_static_screen_params *params); + + #endif /* __DC_HWSS_DCN31_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c index 669f524bd064..c06cc2c5da92 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/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 = dcn30_set_static_screen_control, + .set_static_screen_control = dcn31_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/hwss/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c index ccb7e317e86a..542ce3b7f9e4 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/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 = dcn30_set_static_screen_control, + .set_static_screen_control = dcn31_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/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index 6c9299c7683d..aa36d7a56ca8 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -1474,9 +1474,44 @@ void dcn32_update_dsc_pg(struct dc *dc, } } +void dcn32_disable_phantom_streams(struct dc *dc, struct dc_state *context) +{ + struct dce_hwseq *hws = dc->hwseq; + int i; + + for (i = dc->res_pool->pipe_count - 1; i >= 0 ; i--) { + struct pipe_ctx *pipe_ctx_old = + &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx_old->stream) + continue; + + if (dc_state_get_pipe_subvp_type(dc->current_state, pipe_ctx_old) != SUBVP_PHANTOM) + continue; + + if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) + continue; + + if (!pipe_ctx->stream || pipe_need_reprogram(pipe_ctx_old, pipe_ctx) || + (pipe_ctx->stream && dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM)) { + struct clock_source *old_clk = pipe_ctx_old->clock_source; + + if (hws->funcs.reset_back_end_for_pipe) + hws->funcs.reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state); + if (hws->funcs.enable_stream_gating) + hws->funcs.enable_stream_gating(dc, pipe_ctx_old); + if (old_clk) + old_clk->funcs->cs_power_down(old_clk); + } + } +} + void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context) { unsigned int i; + enum dc_status status = DC_OK; + struct dce_hwseq *hws = dc->hwseq; for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe = &context->res_ctx.pipe_ctx[i]; @@ -1497,16 +1532,39 @@ void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context) } } for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i]; - - if (new_pipe->stream && dc_state_get_pipe_subvp_type(context, new_pipe) == SUBVP_PHANTOM) { - // If old context or new context has phantom pipes, apply - // the phantom timings now. We can't change the phantom - // pipe configuration safely without driver acquiring - // the DMCUB lock first. - dc->hwss.apply_ctx_to_hw(dc, context); - break; + struct pipe_ctx *pipe_ctx_old = + &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (pipe_ctx->stream == NULL) + continue; + + if (dc_state_get_pipe_subvp_type(context, pipe_ctx) != SUBVP_PHANTOM) + continue; + + if (pipe_ctx->stream == pipe_ctx_old->stream && + pipe_ctx->stream->link->link_state_valid) { + continue; } + + if (pipe_ctx_old->stream && !pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) + continue; + + if (pipe_ctx->top_pipe || pipe_ctx->prev_odm_pipe) + continue; + + if (hws->funcs.apply_single_controller_ctx_to_hw) + status = hws->funcs.apply_single_controller_ctx_to_hw( + pipe_ctx, + context, + dc); + + ASSERT(status == DC_OK); + +#ifdef CONFIG_DRM_AMD_DC_FP + if (hws->funcs.resync_fifo_dccg_dio) + hws->funcs.resync_fifo_dccg_dio(hws, dc, context); +#endif } } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h index cecf7f0f5671..069e20bc87c0 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h @@ -111,6 +111,8 @@ void dcn32_update_dsc_pg(struct dc *dc, void dcn32_enable_phantom_streams(struct dc *dc, struct dc_state *context); +void dcn32_disable_phantom_streams(struct dc *dc, struct dc_state *context); + void dcn32_init_blank( struct dc *dc, struct timing_generator *tg); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c index 427cfc8c24a4..2b073123d3ed 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/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 = dcn30_set_static_screen_control, + .set_static_screen_control = dcn31_set_static_screen_control, .setup_stereo = dcn10_setup_stereo, .set_avmute = dcn30_set_avmute, .log_hw_state = dcn10_log_hw_state, @@ -109,6 +109,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = { .get_dcc_en_bits = dcn10_get_dcc_en_bits, .commit_subvp_config = dcn32_commit_subvp_config, .enable_phantom_streams = dcn32_enable_phantom_streams, + .disable_phantom_streams = dcn32_disable_phantom_streams, .subvp_pipe_control_lock = dcn32_subvp_pipe_control_lock, .update_visual_confirm_color = dcn10_update_visual_confirm_color, .subvp_pipe_control_lock_fast = dcn32_subvp_pipe_control_lock_fast, @@ -159,6 +160,8 @@ static const struct hwseq_private_funcs dcn32_private_funcs = { .set_pixels_per_cycle = dcn32_set_pixels_per_cycle, .resync_fifo_dccg_dio = dcn32_resync_fifo_dccg_dio, .is_dp_dig_pixel_rate_div_policy = dcn32_is_dp_dig_pixel_rate_div_policy, + .apply_single_controller_ctx_to_hw = dce110_apply_single_controller_ctx_to_hw, + .reset_back_end_for_pipe = dcn20_reset_back_end_for_pipe, }; void dcn32_hw_sequencer_init_functions(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c index 8b6c49622f3b..4b92df23ff0d 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c @@ -1342,8 +1342,8 @@ void dcn35_set_drr(struct pipe_ctx **pipe_ctx, { int i = 0; struct drr_params params = {0}; - // DRR set trigger event mapped to OTG_TRIG_A (bit 11) for manual control flow - unsigned int event_triggers = 0x800; + // DRR set trigger event mapped to OTG_TRIG_A + unsigned int event_triggers = 0x2;//Bit[1]: OTG_TRIG_A // Note DRR trigger events are generated regardless of whether num frames met. unsigned int num_frames = 2; @@ -1377,3 +1377,20 @@ void dcn35_set_drr(struct pipe_ctx **pipe_ctx, } } } +void dcn35_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 |= 0x200;/*bit 9 : 10 0000 0000*/ + if (params->triggers.cursor_update) + triggers |= 0x8;/*bit3*/ + 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/hwss/dcn35/dcn35_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h index fd66316e33de..c354efa6c1b2 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.h @@ -90,4 +90,7 @@ uint32_t dcn35_get_idle_state(const struct dc *dc); void dcn35_set_drr(struct pipe_ctx **pipe_ctx, int num_pipes, struct dc_crtc_timing_adjust adjust); +void dcn35_set_static_screen_control(struct pipe_ctx **pipe_ctx, + int num_pipes, const struct dc_static_screen_params *params); + #endif /* __DC_HWSS_DCN35_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c index a630aa77dcec..a93073055e7b 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c @@ -70,7 +70,7 @@ static const struct hw_sequencer_funcs dcn35_funcs = { .update_bandwidth = dcn20_update_bandwidth, .set_drr = dcn35_set_drr, .get_position = dcn10_get_position, - .set_static_screen_control = dcn30_set_static_screen_control, + .set_static_screen_control = dcn35_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/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c index 143d3fc0221c..ab17fa1c64e8 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c @@ -69,7 +69,7 @@ static const struct hw_sequencer_funcs dcn351_funcs = { .update_bandwidth = dcn20_update_bandwidth, .set_drr = dcn10_set_drr, .get_position = dcn10_get_position, - .set_static_screen_control = dcn30_set_static_screen_control, + .set_static_screen_control = dcn35_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/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h index a54399383318..f89f205e42a1 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h @@ -339,6 +339,8 @@ struct hw_sequencer_funcs { /* HW State Logging Related */ void (*log_hw_state)(struct dc *dc, struct dc_log_buffer_ctx *log_ctx); + void (*log_color_state)(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx); void (*get_hw_state)(struct dc *dc, char *pBuf, unsigned int bufSize, unsigned int mask); void (*clear_status_bits)(struct dc *dc, unsigned int mask); @@ -379,6 +381,7 @@ struct hw_sequencer_funcs { struct dc_cursor_attributes *cursor_attr); void (*commit_subvp_config)(struct dc *dc, struct dc_state *context); void (*enable_phantom_streams)(struct dc *dc, struct dc_state *context); + void (*disable_phantom_streams)(struct dc *dc, struct dc_state *context); void (*subvp_pipe_control_lock)(struct dc *dc, struct dc_state *context, bool lock, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h index 6137cf09aa54..554cfab5ab24 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h @@ -155,7 +155,6 @@ struct hwseq_private_funcs { void (*setup_hpo_hw_control)(const struct dce_hwseq *hws, bool enable); void (*enable_plane)(struct dc *dc, struct pipe_ctx *pipe_ctx, struct dc_state *context); -#ifdef CONFIG_DRM_AMD_DC_FP void (*program_mall_pipe_config)(struct dc *dc, struct dc_state *context); void (*update_force_pstate)(struct dc *dc, struct dc_state *context); void (*update_mall_sel)(struct dc *dc, struct dc_state *context); @@ -165,8 +164,14 @@ struct hwseq_private_funcs { void (*set_pixels_per_cycle)(struct pipe_ctx *pipe_ctx); void (*resync_fifo_dccg_dio)(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context); + enum dc_status (*apply_single_controller_ctx_to_hw)( + struct pipe_ctx *pipe_ctx, + struct dc_state *context, + struct dc *dc); bool (*is_dp_dig_pixel_rate_div_policy)(struct pipe_ctx *pipe_ctx); -#endif + void (*reset_back_end_for_pipe)(struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context); }; struct dce_hwseq { 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 3a6bf77a6873..ebb659c327e0 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -333,6 +333,8 @@ struct stream_resource { uint8_t gsl_group; struct test_pattern_params test_pattern_params; + + bool left_edge_extra_pixel; }; struct plane_resource { diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/audio.h b/drivers/gpu/drm/amd/display/dc/inc/hw/audio.h index 6ed1fb8c9300..b6203253111c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/audio.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/audio.h @@ -43,7 +43,8 @@ struct audio_funcs { void (*az_configure)(struct audio *audio, enum signal_type signal, const struct audio_crtc_info *crtc_info, - const struct audio_info *audio_info); + const struct audio_info *audio_info, + const struct audio_dp_link_info *dp_link_info); void (*wall_dto_setup)(struct audio *audio, enum signal_type signal, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h index 6f4c97543c14..f4d4a68c91dc 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr_internal.h @@ -356,6 +356,7 @@ struct clk_mgr_internal { long long wm_range_table_addr; bool dpm_present; + bool pme_trigger_pending; }; struct clk_mgr_internal_funcs { @@ -393,6 +394,11 @@ static inline int khz_to_mhz_ceil(int khz) return (khz + 999) / 1000; } +static inline int khz_to_mhz_floor(int khz) +{ + return khz / 1000; +} + int clk_mgr_helper_get_active_display_cnt( struct dc *dc, struct dc_state *context); 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 901891316dfb..2ae7484d18af 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h @@ -26,6 +26,12 @@ #ifndef __DAL_DCHUBBUB_H__ #define __DAL_DCHUBBUB_H__ +/** + * DOC: overview + * + * There is only one common DCHUBBUB. It contains the common request and return + * blocks for the Data Fabric Interface that are not clock/power gated. + */ enum dcc_control { dcc_control__256_256_xxx, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h index f4aa76e02518..0f24afbf4388 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h @@ -27,6 +27,31 @@ #ifndef __DAL_DPP_H__ #define __DAL_DPP_H__ +/** + * DOC: overview + * + * The DPP (Display Pipe and Plane) block is the unified display data + * processing engine in DCN for processing graphic or video data on per DPP + * rectangle base. This rectangle can be a part of SLS (Single Large Surface), + * or a layer to be blended with other DPP, or a rectangle associated with a + * display tile. + * + * It provides various functions including: + * - graphic color keyer + * - graphic cursor compositing + * - graphic or video image source to destination scaling + * - image sharping + * - video format conversion from 4:2:0 or 4:2:2 to 4:4:4 + * - Color Space Conversion + * - Host LUT gamma adjustment + * - Color Gamut Remap + * - brightness and contrast adjustment. + * + * DPP pipe consists of Converter and Cursor (CNVC), Scaler (DSCL), Color + * Management (CM), Output Buffer (OBUF) and Digital Bypass (DPB) module + * connected in a video/graphics pipeline. + */ + #include "transform.h" #include "cursor_reg_cache.h" @@ -141,6 +166,7 @@ struct dcn_dpp_state { uint32_t igam_input_format; uint32_t dgam_lut_mode; uint32_t rgam_lut_mode; + // gamut_remap data for dcn10_get_cm_states() uint32_t gamut_remap_mode; uint32_t gamut_remap_c11_c12; uint32_t gamut_remap_c13_c14; @@ -148,6 +174,16 @@ struct dcn_dpp_state { uint32_t gamut_remap_c23_c24; uint32_t gamut_remap_c31_c32; uint32_t gamut_remap_c33_c34; + // gamut_remap data for dcn*_log_color_state() + struct dpp_grph_csc_adjustment gamut_remap; + uint32_t shaper_lut_mode; + uint32_t lut3d_mode; + uint32_t lut3d_bit_depth; + uint32_t lut3d_size; + uint32_t blnd_lut_mode; + uint32_t pre_dgam_mode; + uint32_t pre_dgam_select; + uint32_t gamcor_mode; }; struct CM_bias_params { @@ -290,6 +326,9 @@ struct dpp_funcs { void (*dpp_cnv_set_alpha_keyer)( struct dpp *dpp_base, struct cnv_color_keyer_params *color_keyer); + + void (*dpp_get_gamut_remap)(struct dpp *dpp_base, + struct dpp_grph_csc_adjustment *adjust); }; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 7f3f9b69e903..72610cd7eae0 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -26,13 +26,24 @@ #ifndef __DAL_HUBP_H__ #define __DAL_HUBP_H__ +/** + * DOC: overview + * + * Display Controller Hub (DCHUB) is the gateway between the Scalable Data Port + * (SDP) and DCN. This component has multiple features, such as memory + * arbitration, rotation, and cursor manipulation. + * + * There is one HUBP allocated per pipe, which fetches data and converts + * different pixel formats (i.e. ARGB8888, NV12, etc) into linear, interleaved + * and fixed-depth streams of pixel data. + */ + #include "mem_input.h" #include "cursor_reg_cache.h" #define OPP_ID_INVALID 0xf #define MAX_TTU 0xffffff - enum cursor_pitch { CURSOR_PITCH_64_PIXELS = 0, CURSOR_PITCH_128_PIXELS, @@ -146,9 +157,7 @@ struct hubp_funcs { void (*set_blank)(struct hubp *hubp, bool blank); void (*set_blank_regs)(struct hubp *hubp, bool blank); -#ifdef CONFIG_DRM_AMD_DC_FP void (*phantom_hubp_post_enable)(struct hubp *hubp); -#endif void (*set_hubp_blank_en)(struct hubp *hubp, bool blank); void (*set_cursor_attributes)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h index 61a2406dcc53..ba9b942ce09f 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h @@ -23,13 +23,28 @@ */ /** - * DOC: mpc-overview + * DOC: overview * - * Multiple Pipe/Plane Combined (MPC) is a component in the hardware pipeline + * Multiple Pipe/Plane Combiner (MPC) is a component in the hardware pipeline * that performs blending of multiple planes, using global and per-pixel alpha. * It also performs post-blending color correction operations according to the * hardware capabilities, such as color transformation matrix and gamma 1D and * 3D LUT. + * + * MPC receives output from all DPP pipes and combines them to multiple outputs + * supporting "M MPC inputs -> N MPC outputs" flexible composition + * architecture. It features: + * + * - Programmable blending structure to allow software controlled blending and + * cascading; + * - Programmable window location of each DPP in active region of display; + * - Combining multiple DPP pipes in one active region when a single DPP pipe + * cannot process very large surface; + * - Combining multiple DPP from different SLS with blending; + * - Stereo formats from single DPP in top-bottom or side-by-side modes; + * - Stereo formats from 2 DPPs; + * - Alpha blending of multiple layers from different DPP pipes; + * - Programmable background color; */ #ifndef __DC_MPCC_H__ @@ -83,34 +98,66 @@ enum mpcc_alpha_blend_mode { /** * struct mpcc_blnd_cfg - MPCC blending configuration - * - * @black_color: background color - * @alpha_mode: alpha blend mode (MPCC_ALPHA_BLND_MODE) - * @pre_multiplied_alpha: whether pixel color values were pre-multiplied by the - * alpha channel (MPCC_ALPHA_MULTIPLIED_MODE) - * @global_gain: used when blend mode considers both pixel alpha and plane - * alpha value and assumes the global alpha value. - * @global_alpha: plane alpha value - * @overlap_only: whether overlapping of different planes is allowed - * @bottom_gain_mode: blend mode for bottom gain setting - * @background_color_bpc: background color for bpc - * @top_gain: top gain setting - * @bottom_inside_gain: blend mode for bottom inside - * @bottom_outside_gain: blend mode for bottom outside */ struct mpcc_blnd_cfg { - struct tg_color black_color; /* background color */ - enum mpcc_alpha_blend_mode alpha_mode; /* alpha blend mode */ - bool pre_multiplied_alpha; /* alpha pre-multiplied mode flag */ + /** + * @black_color: background color. + */ + struct tg_color black_color; + + /** + * @alpha_mode: alpha blend mode (MPCC_ALPHA_BLND_MODE). + */ + enum mpcc_alpha_blend_mode alpha_mode; + + /*** + * @@pre_multiplied_alpha: + * + * Whether pixel color values were pre-multiplied by the alpha channel + * (MPCC_ALPHA_MULTIPLIED_MODE). + */ + bool pre_multiplied_alpha; + + /** + * @global_gain: Used when blend mode considers both pixel alpha and plane. + */ int global_gain; + + /** + * @global_alpha: Plane alpha value. + */ int global_alpha; + + /** + * @@overlap_only: Whether overlapping of different planes is allowed. + */ bool overlap_only; /* MPCC top/bottom gain settings */ + + /** + * @bottom_gain_mode: Blend mode for bottom gain setting. + */ int bottom_gain_mode; + + /** + * @background_color_bpc: Background color for bpc. + */ int background_color_bpc; + + /** + * @top_gain: Top gain setting. + */ int top_gain; + + /** + * @bottom_inside_gain: Blend mode for bottom inside. + */ int bottom_inside_gain; + + /** + * @bottom_outside_gain: Blend mode for bottom outside. + */ int bottom_outside_gain; }; @@ -150,34 +197,58 @@ struct mpc_dwb_flow_control { /** * struct mpcc - MPCC connection and blending configuration for a single MPCC instance. - * @mpcc_id: MPCC physical instance - * @dpp_id: DPP input to this MPCC - * @mpcc_bot: pointer to bottom layer MPCC. NULL when not connected. - * @blnd_cfg: the blending configuration for this MPCC - * @sm_cfg: stereo mix setting for this MPCC - * @shared_bottom: if MPCC output to both OPP and DWB endpoints, true. Otherwise, false. * * This struct is used as a node in an MPC tree. */ struct mpcc { - int mpcc_id; /* MPCC physical instance */ - int dpp_id; /* DPP input to this MPCC */ - struct mpcc *mpcc_bot; /* pointer to bottom layer MPCC. NULL when not connected */ - struct mpcc_blnd_cfg blnd_cfg; /* The blending configuration for this MPCC */ - struct mpcc_sm_cfg sm_cfg; /* stereo mix setting for this MPCC */ - bool shared_bottom; /* TRUE if MPCC output to both OPP and DWB endpoints, else FALSE */ + /** + * @mpcc_id: MPCC physical instance. + */ + int mpcc_id; + + /** + * @dpp_id: DPP input to this MPCC + */ + int dpp_id; + + /** + * @mpcc_bot: Pointer to bottom layer MPCC. NULL when not connected. + */ + struct mpcc *mpcc_bot; + + /** + * @blnd_cfg: The blending configuration for this MPCC. + */ + struct mpcc_blnd_cfg blnd_cfg; + + /** + * @sm_cfg: stereo mix setting for this MPCC + */ + struct mpcc_sm_cfg sm_cfg; + + /** + * @shared_bottom: + * + * If MPCC output to both OPP and DWB endpoints, true. Otherwise, false. + */ + bool shared_bottom; }; /** * struct mpc_tree - MPC tree represents all MPCC connections for a pipe. * - * @opp_id: the OPP instance that owns this MPC tree - * @opp_list: the top MPCC layer of the MPC tree that outputs to OPP endpoint * */ struct mpc_tree { - int opp_id; /* The OPP instance that owns this MPC tree */ - struct mpcc *opp_list; /* The top MPCC layer of the MPC tree that outputs to OPP endpoint */ + /** + * @opp_id: The OPP instance that owns this MPC tree. + */ + int opp_id; + + /** + * @opp_list: the top MPCC layer of the MPC tree that outputs to OPP endpoint + */ + struct mpcc *opp_list; }; struct mpc { @@ -199,6 +270,13 @@ struct mpcc_state { uint32_t overlap_only; uint32_t idle; uint32_t busy; + uint32_t shaper_lut_mode; + uint32_t lut3d_mode; + uint32_t lut3d_bit_depth; + uint32_t lut3d_size; + uint32_t rgam_mode; + uint32_t rgam_lut; + struct mpc_grph_gamut_adjustment gamut_remap; }; /** @@ -217,16 +295,20 @@ struct mpc_funcs { * Only used for planes that are part of blending chain for OPP output * * Parameters: - * [in/out] mpc - MPC context. - * [in/out] tree - MPC tree structure that plane will be added to. - * [in] blnd_cfg - MPCC blending configuration for the new blending layer. - * [in] sm_cfg - MPCC stereo mix configuration for the new blending layer. - * stereo mix must disable for the very bottom layer of the tree config. - * [in] insert_above_mpcc - Insert new plane above this MPCC. If NULL, insert as bottom plane. - * [in] dpp_id - DPP instance for the plane to be added. - * [in] mpcc_id - The MPCC physical instance to use for blending. - * - * Return: struct mpcc* - MPCC that was added. + * + * - [in/out] mpc - MPC context. + * - [in/out] tree - MPC tree structure that plane will be added to. + * - [in] blnd_cfg - MPCC blending configuration for the new blending layer. + * - [in] sm_cfg - MPCC stereo mix configuration for the new blending layer. + * stereo mix must disable for the very bottom layer of the tree config. + * - [in] insert_above_mpcc - Insert new plane above this MPCC. + * If NULL, insert as bottom plane. + * - [in] dpp_id - DPP instance for the plane to be added. + * - [in] mpcc_id - The MPCC physical instance to use for blending. + * + * Return: + * + * struct mpcc* - MPCC that was added. */ struct mpcc* (*insert_plane)( struct mpc *mpc, @@ -243,11 +325,14 @@ struct mpc_funcs { * Remove a specified MPCC from the MPC tree. * * Parameters: - * [in/out] mpc - MPC context. - * [in/out] tree - MPC tree structure that plane will be removed from. - * [in/out] mpcc - MPCC to be removed from tree. * - * Return: void + * - [in/out] mpc - MPC context. + * - [in/out] tree - MPC tree structure that plane will be removed from. + * - [in/out] mpcc - MPCC to be removed from tree. + * + * Return: + * + * void */ void (*remove_mpcc)( struct mpc *mpc, @@ -260,9 +345,12 @@ struct mpc_funcs { * Reset the MPCC HW status by disconnecting all muxes. * * Parameters: - * [in/out] mpc - MPC context. * - * Return: void + * - [in/out] mpc - MPC context. + * + * Return: + * + * void */ void (*mpc_init)(struct mpc *mpc); void (*mpc_init_single_inst)( @@ -275,11 +363,14 @@ struct mpc_funcs { * Update the blending configuration for a specified MPCC. * * Parameters: - * [in/out] mpc - MPC context. - * [in] blnd_cfg - MPCC blending configuration. - * [in] mpcc_id - The MPCC physical instance. * - * Return: void + * - [in/out] mpc - MPC context. + * - [in] blnd_cfg - MPCC blending configuration. + * - [in] mpcc_id - The MPCC physical instance. + * + * Return: + * + * void */ void (*update_blending)( struct mpc *mpc, @@ -289,15 +380,18 @@ struct mpc_funcs { /** * @cursor_lock: * - * Lock cursor updates for the specified OPP. - * OPP defines the set of MPCC that are locked together for cursor. + * Lock cursor updates for the specified OPP. OPP defines the set of + * MPCC that are locked together for cursor. * * Parameters: - * [in] mpc - MPC context. - * [in] opp_id - The OPP to lock cursor updates on - * [in] lock - lock/unlock the OPP * - * Return: void + * - [in] mpc - MPC context. + * - [in] opp_id - The OPP to lock cursor updates on + * - [in] lock - lock/unlock the OPP + * + * Return: + * + * void */ void (*cursor_lock)( struct mpc *mpc, @@ -307,20 +401,25 @@ struct mpc_funcs { /** * @insert_plane_to_secondary: * - * Add DPP into secondary MPC tree based on specified blending position. - * Only used for planes that are part of blending chain for DWB output + * Add DPP into secondary MPC tree based on specified blending + * position. Only used for planes that are part of blending chain for + * DWB output * * Parameters: - * [in/out] mpc - MPC context. - * [in/out] tree - MPC tree structure that plane will be added to. - * [in] blnd_cfg - MPCC blending configuration for the new blending layer. - * [in] sm_cfg - MPCC stereo mix configuration for the new blending layer. - * stereo mix must disable for the very bottom layer of the tree config. - * [in] insert_above_mpcc - Insert new plane above this MPCC. If NULL, insert as bottom plane. - * [in] dpp_id - DPP instance for the plane to be added. - * [in] mpcc_id - The MPCC physical instance to use for blending. - * - * Return: struct mpcc* - MPCC that was added. + * + * - [in/out] mpc - MPC context. + * - [in/out] tree - MPC tree structure that plane will be added to. + * - [in] blnd_cfg - MPCC blending configuration for the new blending layer. + * - [in] sm_cfg - MPCC stereo mix configuration for the new blending layer. + * stereo mix must disable for the very bottom layer of the tree config. + * - [in] insert_above_mpcc - Insert new plane above this MPCC. If + * NULL, insert as bottom plane. + * - [in] dpp_id - DPP instance for the plane to be added. + * - [in] mpcc_id - The MPCC physical instance to use for blending. + * + * Return: + * + * struct mpcc* - MPCC that was added. */ struct mpcc* (*insert_plane_to_secondary)( struct mpc *mpc, @@ -337,10 +436,14 @@ struct mpc_funcs { * Remove a specified DPP from the 'secondary' MPC tree. * * Parameters: - * [in/out] mpc - MPC context. - * [in/out] tree - MPC tree structure that plane will be removed from. - * [in] mpcc - MPCC to be removed from tree. - * Return: void + * + * - [in/out] mpc - MPC context. + * - [in/out] tree - MPC tree structure that plane will be removed from. + * - [in] mpcc - MPCC to be removed from tree. + * + * Return: + * + * void */ void (*remove_mpcc_from_secondary)( struct mpc *mpc, diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h index 7617fabbd16e..aee5372e292c 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/opp.h @@ -23,6 +23,22 @@ * */ +/** + * DOC: overview + * + * The Output Plane Processor (OPP) block groups have functions that format + * pixel streams such that they are suitable for display at the display device. + * The key functions contained in the OPP are: + * + * - Adaptive Backlight Modulation (ABM) + * - Formatter (FMT) which provide pixel-by-pixel operations for format the + * incoming pixel stream. + * - Output Buffer that provide pixel replication, and overlapping. + * - Interface between MPC and OPTC. + * - Clock and reset generation. + * - CRC generation. + */ + #ifndef __DAL_OPP_H__ #define __DAL_OPP_H__ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index 9a00a99317b2..d98d72f35be5 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -182,9 +182,7 @@ struct timing_generator_funcs { bool (*enable_crtc)(struct timing_generator *tg); bool (*disable_crtc)(struct timing_generator *tg); -#ifdef CONFIG_DRM_AMD_DC_FP void (*phantom_crtc_post_enable)(struct timing_generator *tg); -#endif void (*disable_phantom_crtc)(struct timing_generator *tg); bool (*immediate_disable_crtc)(struct timing_generator *tg); bool (*is_counter_moving)(struct timing_generator *tg); diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index c958ef37b78a..b14d52e52fa2 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -107,6 +107,10 @@ void resource_build_test_pattern_params( struct resource_context *res_ctx, struct pipe_ctx *pipe_ctx); +void resource_build_subsampling_params( + struct resource_context *res_ctx, + struct pipe_ctx *pipe_ctx); + bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx); enum dc_status resource_build_scaling_params_for_context( @@ -427,22 +431,18 @@ struct pipe_ctx *resource_get_primary_dpp_pipe(const struct pipe_ctx *dpp_pipe); int resource_get_mpc_slice_index(const struct pipe_ctx *dpp_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. + * Get the number of MPC slices associated with the pipe. + * The function returns 0 if the pipe is not associated with an MPC combine + * pipe topology. */ -int resource_get_mpc_slice_count(const struct pipe_ctx *opp_head); +int resource_get_mpc_slice_count(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. + * Get the number of ODM slices associated with the pipe. + * The function returns 0 if the pipe is not associated with an ODM combine + * pipe topology. */ -int resource_get_odm_slice_count(const struct pipe_ctx *otg_master); +int resource_get_odm_slice_count(const struct pipe_ctx *pipe); /* Get the ODM slice index counting from 0 from left most slice */ int resource_get_odm_slice_index(const struct pipe_ctx *opp_head); 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 f4633d3cf9b9..a1f72fe378ee 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 @@ -22,6 +22,16 @@ * Authors: AMD * */ + +/** + * DOC: overview + * + * Display Input Output (DIO), is the display input and output unit in DCN. It + * includes output encoders to support different display output, like + * DisplayPort, HDMI, DVI interface, and others. It also includes the control + * and status channels for these interfaces. + */ + #ifndef __LINK_HWSS_DIO_H__ #define __LINK_HWSS_DIO_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 24153b0df503..b8c4a04dd175 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c @@ -41,6 +41,7 @@ #include "protocols/link_dp_dpia.h" #include "protocols/link_dp_phy.h" #include "protocols/link_dp_training.h" +#include "protocols/link_dp_dpia_bw.h" #include "accessories/link_dp_trace.h" #include "link_enc_cfg.h" @@ -991,6 +992,23 @@ static bool detect_link_and_local_sink(struct dc_link *link, if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && link->reported_link_cap.link_rate > LINK_RATE_HIGH3) link->reported_link_cap.link_rate = LINK_RATE_HIGH3; + + /* + * If this is DP over USB4 link then we need to: + * - Enable BW ALLOC support on DPtx if applicable + */ + if (dc->config.usb4_bw_alloc_support) { + if (link_dp_dpia_set_dptx_usb4_bw_alloc_support(link)) { + /* update with non reduced link cap if bw allocation mode is supported */ + if (link->dpia_bw_alloc_config.nrd_max_link_rate && + link->dpia_bw_alloc_config.nrd_max_lane_count) { + link->reported_link_cap.link_rate = + link->dpia_bw_alloc_config.nrd_max_link_rate; + link->reported_link_cap.lane_count = + link->dpia_bw_alloc_config.nrd_max_lane_count; + } + } + } break; } 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 3cbfbf8d107e..a72de44a5747 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c @@ -2197,6 +2197,64 @@ static enum dc_status enable_link( static bool allocate_usb4_bandwidth_for_stream(struct dc_stream_state *stream, int bw) { + struct dc_link *link = stream->sink->link; + int req_bw = bw; + + DC_LOGGER_INIT(link->ctx->logger); + + if (!link->dpia_bw_alloc_config.bw_alloc_enabled) + return false; + + if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + int sink_index = 0; + int i = 0; + + for (i = 0; i < link->sink_count; i++) { + if (link->remote_sinks[i] == NULL) + continue; + + if (stream->sink->sink_id != link->remote_sinks[i]->sink_id) + req_bw += link->dpia_bw_alloc_config.remote_sink_req_bw[i]; + else + sink_index = i; + } + + link->dpia_bw_alloc_config.remote_sink_req_bw[sink_index] = bw; + } + + /* get dp overhead for dp tunneling */ + link->dpia_bw_alloc_config.dp_overhead = link_dp_dpia_get_dp_overhead_in_dp_tunneling(link); + req_bw += link->dpia_bw_alloc_config.dp_overhead; + + if (link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, req_bw)) { + if (req_bw <= link->dpia_bw_alloc_config.allocated_bw) { + DC_LOG_DEBUG("%s, Success in allocate bw for link(%d), allocated_bw(%d), dp_overhead(%d)\n", + __func__, link->link_index, link->dpia_bw_alloc_config.allocated_bw, + link->dpia_bw_alloc_config.dp_overhead); + } else { + // Cannot get the required bandwidth. + DC_LOG_ERROR("%s, Failed to allocate bw for link(%d), allocated_bw(%d), dp_overhead(%d)\n", + __func__, link->link_index, link->dpia_bw_alloc_config.allocated_bw, + link->dpia_bw_alloc_config.dp_overhead); + return false; + } + } else { + DC_LOG_DEBUG("%s, usb4 request bw timeout\n", __func__); + return false; + } + + if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) { + int i = 0; + + for (i = 0; i < link->sink_count; i++) { + if (link->remote_sinks[i] == NULL) + continue; + DC_LOG_DEBUG("%s, remote_sink=%s, request_bw=%d\n", __func__, + (const char *)(&link->remote_sinks[i]->edid_caps.display_name[0]), + link->dpia_bw_alloc_config.remote_sink_req_bw[i]); + } + } + return true; } 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 8fe66c367850..1c038e2a527b 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c @@ -125,11 +125,9 @@ static bool dp_active_dongle_validate_timing( if (dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps > 0) { // DP to HDMI FRL converter struct dc_crtc_timing outputTiming = *timing; -#if defined(CONFIG_DRM_AMD_DC_FP) if (timing->flags.DSC && !timing->dsc_cfg.is_frl) /* 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, DC_LINK_ENCODING_HDMI_FRL) > dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps) return false; 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 5a0b04518956..e06d3c2d8910 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 @@ -1505,10 +1505,7 @@ enum link_training_result dp_perform_link_training( * Non-LT AUX transactions inside training mode. */ if ((link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && encoding == DP_8b_10b_ENCODING) - if (link->dc->config.use_old_fixed_vs_sequence) - status = dp_perform_fixed_vs_pe_training_sequence_legacy(link, link_res, <_settings); - else - status = dp_perform_fixed_vs_pe_training_sequence(link, link_res, <_settings); + status = dp_perform_fixed_vs_pe_training_sequence(link, link_res, <_settings); else if (encoding == DP_8b_10b_ENCODING) status = dp_perform_8b_10b_link_training(link, link_res, <_settings); else if (encoding == DP_128b_132b_ENCODING) 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 7087cdc9e977..b5cf75975fff 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 @@ -186,356 +186,6 @@ static enum link_training_result perform_fixed_vs_pe_nontransparent_training_seq return status; } - -enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( - struct dc_link *link, - const struct link_resource *link_res, - struct link_training_settings *lt_settings) -{ - const uint8_t vendor_lttpr_write_data_reset[4] = {0x1, 0x50, 0x63, 0xFF}; - const uint8_t offset = dp_parse_lttpr_repeater_count( - link->dpcd_caps.lttpr_caps.phy_repeater_cnt); - const uint8_t vendor_lttpr_write_data_intercept_en[4] = {0x1, 0x55, 0x63, 0x0}; - const uint8_t vendor_lttpr_write_data_intercept_dis[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}; - 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}; - const uint8_t vendor_lttpr_write_data_dpmf[4] = {0x1, 0x6, 0x70, 0x87}; - enum link_training_result status = LINK_TRAINING_SUCCESS; - uint8_t lane = 0; - union down_spread_ctrl downspread = {0}; - union lane_count_set lane_count_set = {0}; - uint8_t toggle_rate; - uint8_t rate; - - /* Only 8b/10b is supported */ - ASSERT(link_dp_get_encoding_format(<_settings->link_settings) == - DP_8b_10b_ENCODING); - - if (lt_settings->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) { - status = perform_fixed_vs_pe_nontransparent_training_sequence(link, link_res, lt_settings); - return status; - } - - if (offset != 0xFF) { - 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 */ - } else if (offset > 2) { - pre_disable_intercept_delay_ms = link->dc->debug.fixed_vs_aux_delay_config_wa * 2; - } - } - - /* Vendor specific: Reset lane settings */ - 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 */ - 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. */ - - downspread.raw = (uint8_t)(lt_settings->link_settings.link_spread); - - lane_count_set.bits.LANE_COUNT_SET = - lt_settings->link_settings.lane_count; - - lane_count_set.bits.ENHANCED_FRAMING = lt_settings->enhanced_framing; - lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = 0; - - - if (lt_settings->pattern_for_eq < DP_TRAINING_PATTERN_SEQUENCE_4) { - lane_count_set.bits.POST_LT_ADJ_REQ_GRANTED = - link->dpcd_caps.max_ln_count.bits.POST_LT_ADJ_REQ_SUPPORTED; - } - - core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL, - &downspread.raw, sizeof(downspread)); - - core_link_write_dpcd(link, DP_LANE_COUNT_SET, - &lane_count_set.raw, 1); - - rate = get_dpcd_link_rate(<_settings->link_settings); - - /* Vendor specific: Toggle link rate */ - toggle_rate = (rate == 0x6) ? 0xA : 0x6; - - if (link->vendor_specific_lttpr_link_rate_wa == rate || link->vendor_specific_lttpr_link_rate_wa == 0) { - core_link_write_dpcd( - link, - DP_LINK_BW_SET, - &toggle_rate, - 1); - } - - link->vendor_specific_lttpr_link_rate_wa = rate; - - core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1); - - DC_LOG_HW_LINK_TRAINING("%s\n %x rate = %x\n %x lane = %x framing = %x\n %x spread = %x\n", - __func__, - DP_LINK_BW_SET, - lt_settings->link_settings.link_rate, - DP_LANE_COUNT_SET, - lt_settings->link_settings.lane_count, - lt_settings->enhanced_framing, - DP_DOWNSPREAD_CTRL, - lt_settings->link_settings.link_spread); - - link_configure_fixed_vs_pe_retimer(link->ddc, - &vendor_lttpr_write_data_dpmf[0], - sizeof(vendor_lttpr_write_data_dpmf)); - - 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 */ - if (status == LINK_TRAINING_SUCCESS) { - const uint8_t max_vendor_dpcd_retries = 10; - uint32_t retries_cr; - uint32_t retry_count; - uint32_t wait_time_microsec; - enum dc_lane_count lane_count = lt_settings->link_settings.lane_count; - 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}; - uint8_t i = 0; - - retries_cr = 0; - retry_count = 0; - - memset(&dpcd_lane_status, '\0', sizeof(dpcd_lane_status)); - memset(&dpcd_lane_status_updated, '\0', - sizeof(dpcd_lane_status_updated)); - - while ((retries_cr < LINK_TRAINING_MAX_RETRY_COUNT) && - (retry_count < LINK_TRAINING_MAX_CR_RETRY)) { - - - /* 1. call HWSS to set lane settings */ - dp_set_hw_lane_settings( - link, - link_res, - lt_settings, - 0); - - /* 2. update DPCD of the receiver */ - if (!retry_count) { - /* EPR #361076 - write as a 5-byte burst, - * but only for the 1-st iteration. - */ - dpcd_set_lt_pattern_and_lane_settings( - link, - lt_settings, - lt_settings->pattern_for_cr, - 0); - /* Vendor specific: Disable intercept */ - for (i = 0; i < max_vendor_dpcd_retries; i++) { - if (pre_disable_intercept_delay_ms != 0) - msleep(pre_disable_intercept_delay_ms); - if (link_configure_fixed_vs_pe_retimer(link->ddc, - &vendor_lttpr_write_data_intercept_dis[0], - sizeof(vendor_lttpr_write_data_intercept_dis))) - break; - - link_configure_fixed_vs_pe_retimer(link->ddc, - &vendor_lttpr_write_data_intercept_en[0], - sizeof(vendor_lttpr_write_data_intercept_en)); - } - } else { - vendor_lttpr_write_data_vs[3] = 0; - vendor_lttpr_write_data_pe[3] = 0; - - for (lane = 0; lane < lane_count; lane++) { - vendor_lttpr_write_data_vs[3] |= - lt_settings->dpcd_lane_settings[lane].bits.VOLTAGE_SWING_SET << (2 * lane); - vendor_lttpr_write_data_pe[3] |= - lt_settings->dpcd_lane_settings[lane].bits.PRE_EMPHASIS_SET << (2 * lane); - } - - /* Vendor specific: Update VS and PE to DPRX requested value */ - 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, - lt_settings, - 0); - } - - /* 3. wait receiver to lock-on*/ - wait_time_microsec = lt_settings->cr_pattern_time; - - dp_wait_for_training_aux_rd_interval( - link, - wait_time_microsec); - - /* 4. Read lane status and requested drive - * settings as set by the sink - */ - dp_get_lane_status_and_lane_adjust( - link, - lt_settings, - dpcd_lane_status, - &dpcd_lane_status_updated, - dpcd_lane_adjust, - 0); - - /* 5. check CR done*/ - if (dp_is_cr_done(lane_count, dpcd_lane_status)) { - status = LINK_TRAINING_SUCCESS; - break; - } - - /* 6. max VS reached*/ - if (dp_is_max_vs_reached(lt_settings)) - break; - - /* 7. same lane settings */ - /* Note: settings are the same for all lanes, - * so comparing first lane is sufficient - */ - if (lt_settings->dpcd_lane_settings[0].bits.VOLTAGE_SWING_SET == - dpcd_lane_adjust[0].bits.VOLTAGE_SWING_LANE) - retries_cr++; - else - retries_cr = 0; - - /* 8. update VS/PE/PC2 in lt_settings*/ - dp_decide_lane_settings(lt_settings, dpcd_lane_adjust, - lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings); - retry_count++; - } - - if (retry_count >= LINK_TRAINING_MAX_CR_RETRY) { - ASSERT(0); - DC_LOG_ERROR("%s: Link Training Error, could not get CR after %d tries. Possibly voltage swing issue", - __func__, - LINK_TRAINING_MAX_CR_RETRY); - - } - - status = dp_get_cr_failure(lane_count, dpcd_lane_status); - } - - /* Perform Channel EQ Sequence */ - if (status == LINK_TRAINING_SUCCESS) { - enum dc_dp_training_pattern tr_pattern; - uint32_t retries_ch_eq; - uint32_t wait_time_microsec; - enum dc_lane_count lane_count = lt_settings->link_settings.lane_count; - union lane_align_status_updated dpcd_lane_status_updated = {0}; - union lane_status dpcd_lane_status[LANE_COUNT_DP_MAX] = {0}; - union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX] = {0}; - - /* Note: also check that TPS4 is a supported feature*/ - tr_pattern = lt_settings->pattern_for_eq; - - dp_set_hw_training_pattern(link, link_res, tr_pattern, 0); - - status = LINK_TRAINING_EQ_FAIL_EQ; - - for (retries_ch_eq = 0; retries_ch_eq <= LINK_TRAINING_MAX_RETRY_COUNT; - retries_ch_eq++) { - - dp_set_hw_lane_settings(link, link_res, lt_settings, 0); - - vendor_lttpr_write_data_vs[3] = 0; - vendor_lttpr_write_data_pe[3] = 0; - - for (lane = 0; lane < lane_count; lane++) { - vendor_lttpr_write_data_vs[3] |= - lt_settings->dpcd_lane_settings[lane].bits.VOLTAGE_SWING_SET << (2 * lane); - vendor_lttpr_write_data_pe[3] |= - lt_settings->dpcd_lane_settings[lane].bits.PRE_EMPHASIS_SET << (2 * lane); - } - - /* Vendor specific: Update VS and PE to DPRX requested value */ - 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) - /* EPR #361076 - write as a 5-byte burst, - * but only for the 1-st iteration - */ - - dpcd_set_lt_pattern_and_lane_settings( - link, - lt_settings, - tr_pattern, 0); - else - dpcd_set_lane_settings(link, lt_settings, 0); - - /* 3. wait for receiver to lock-on*/ - wait_time_microsec = lt_settings->eq_pattern_time; - - dp_wait_for_training_aux_rd_interval( - link, - wait_time_microsec); - - /* 4. Read lane status and requested - * drive settings as set by the sink - */ - dp_get_lane_status_and_lane_adjust( - link, - lt_settings, - dpcd_lane_status, - &dpcd_lane_status_updated, - dpcd_lane_adjust, - 0); - - /* 5. check CR done*/ - if (!dp_is_cr_done(lane_count, dpcd_lane_status)) { - status = LINK_TRAINING_EQ_FAIL_CR; - break; - } - - /* 6. check CHEQ done*/ - if (dp_is_ch_eq_done(lane_count, dpcd_lane_status) && - dp_is_symbol_locked(lane_count, dpcd_lane_status) && - dp_is_interlane_aligned(dpcd_lane_status_updated)) { - status = LINK_TRAINING_SUCCESS; - break; - } - - /* 7. update VS/PE/PC2 in lt_settings*/ - dp_decide_lane_settings(lt_settings, dpcd_lane_adjust, - lt_settings->hw_lane_settings, lt_settings->dpcd_lane_settings); - } - } - - return status; -} - enum link_training_result dp_perform_fixed_vs_pe_training_sequence( struct dc_link *link, const struct link_resource *link_res, @@ -620,18 +270,20 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( rate = get_dpcd_link_rate(<_settings->link_settings); - /* Vendor specific: Toggle link rate */ - toggle_rate = (rate == 0x6) ? 0xA : 0x6; + if (!link->dpcd_caps.lttpr_caps.main_link_channel_coding.bits.DP_128b_132b_SUPPORTED) { + /* Vendor specific: Toggle link rate */ + toggle_rate = (rate == 0x6) ? 0xA : 0x6; - if (link->vendor_specific_lttpr_link_rate_wa == rate || link->vendor_specific_lttpr_link_rate_wa == 0) { - core_link_write_dpcd( - link, - DP_LINK_BW_SET, - &toggle_rate, - 1); - } + if (link->vendor_specific_lttpr_link_rate_wa == rate || link->vendor_specific_lttpr_link_rate_wa == 0) { + core_link_write_dpcd( + link, + DP_LINK_BW_SET, + &toggle_rate, + 1); + } - link->vendor_specific_lttpr_link_rate_wa = rate; + link->vendor_specific_lttpr_link_rate_wa = rate; + } core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1); diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h index c0d6ea329504..e61970e27661 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.h @@ -28,11 +28,6 @@ #define __DC_LINK_DP_FIXED_VS_PE_RETIMER_H__ #include "link_dp_training.h" -enum link_training_result dp_perform_fixed_vs_pe_training_sequence_legacy( - struct dc_link *link, - const struct link_resource *link_res, - struct link_training_settings *lt_settings); - enum link_training_result dp_perform_fixed_vs_pe_training_sequence( struct dc_link *link, const struct link_resource *link_res, 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 046d3e205415..443215b96308 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 @@ -287,7 +287,7 @@ bool set_default_brightness_aux(struct dc_link *link) if (link && link->dpcd_sink_ext_caps.bits.oled == 1) { if (!read_default_bl_aux(link, &default_backlight)) default_backlight = 150000; - // if < 1 nits or > 5000, it might be wrong readback + // if > 5000, it might be wrong readback. 0 nits is a valid default value for OLED panel. if (default_backlight < 1000 || default_backlight > 5000000) default_backlight = 150000; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c index 37a64186f324..ecc477ef8e3b 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c @@ -2169,6 +2169,17 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params optimal_uclk_for_dcfclk_sta_targets[i] = bw_params->clk_table.entries[j].memclk_mhz * 16; break; + } else { + /* condition where (dcfclk_sta_targets[i] >= optimal_dcfclk_for_uclk[j]): + * If it just so happens that the memory bandwidth is low enough such that + * all the optimal DCFCLK for each UCLK is lower than the smallest DCFCLK STA + * target, we need to populate the optimal UCLK for each DCFCLK STA target to + * be the max UCLK. + */ + if (j == num_uclk_states - 1) { + optimal_uclk_for_dcfclk_sta_targets[i] = + bw_params->clk_table.entries[j].memclk_mhz * 16; + } } } } diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c index 511ff6b5b985..7538b548c572 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c @@ -999,7 +999,7 @@ static struct stream_encoder *dcn301_stream_encoder_create(enum engine_id eng_id vpg = dcn301_vpg_create(ctx, vpg_inst); afmt = dcn301_afmt_create(ctx, afmt_inst); - if (!enc1 || !vpg || !afmt) { + if (!enc1 || !vpg || !afmt || eng_id >= ARRAY_SIZE(stream_enc_regs)) { kfree(enc1); kfree(vpg); kfree(afmt); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c index 31035fc3d868..04d142f97474 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c @@ -1941,8 +1941,6 @@ static bool dcn31_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; - dc->config.use_old_fixed_vs_sequence = true; - /* Use pipe context based otg sync logic */ dc->config.use_pipe_ctx_sync_logic = true; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index c4d71e7f18af..6f10052caeef 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1829,7 +1829,21 @@ int dcn32_populate_dml_pipes_from_context( dcn32_zero_pipe_dcc_fraction(pipes, pipe_cnt); DC_FP_END(); pipes[pipe_cnt].pipe.dest.vfront_porch = timing->v_front_porch; - pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal; + if (dc->config.enable_windowed_mpo_odm && + dc->debug.enable_single_display_2to1_odm_policy) { + switch (resource_get_odm_slice_count(pipe)) { + case 2: + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_2to1; + break; + case 4: + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_4to1; + break; + default: + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal; + } + } else { + pipes[pipe_cnt].pipe.dest.odm_combine_policy = dm_odm_combine_policy_dal; + } pipes[pipe_cnt].pipe.src.gpuvm_min_page_size_kbytes = 256; // according to spreadsheet pipes[pipe_cnt].pipe.src.unbounded_req_mode = false; pipes[pipe_cnt].pipe.scale_ratio_depth.lb_depth = dm_lb_19; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c index 74412e5f03fe..6f832bf278cf 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c @@ -1760,6 +1760,7 @@ static bool dcn321_resource_construct( dc->caps.color.mpc.ocsc = 1; dc->config.dc_mode_clk_limit_support = true; + dc->config.enable_windowed_mpo_odm = false; /* read VBIOS LTTPR caps */ { if (ctx->dc_bios->funcs->get_lttpr_caps) { diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c index 761ec9891875..e534e87cc85b 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c @@ -701,7 +701,7 @@ static const struct dc_plane_cap plane_cap = { // 6:1 downscaling ratio: 1000/6 = 166.666 .max_downscale_factor = { - .argb8888 = 167, + .argb8888 = 250, .nv12 = 167, .fp16 = 167 }, @@ -764,6 +764,7 @@ static const struct dc_debug_options debug_defaults_drv = { }, .seamless_boot_odm_combine = DML_FAIL_SOURCE_PIXEL_FORMAT, .enable_z9_disable_interface = true, /* Allow support for the PMFW interface for disable Z9*/ + .minimum_z8_residency_time = 2100, .using_dml2 = true, .support_eDP1_5 = true, .enable_hpo_pg_support = false, @@ -780,8 +781,9 @@ static const struct dc_debug_options debug_defaults_drv = { .disable_z10 = false, .ignore_pg = true, .psp_disabled_wa = true, - .ips2_eval_delay_us = 200, - .ips2_entry_delay_us = 400, + .ips2_eval_delay_us = 1650, + .ips2_entry_delay_us = 800, + .disable_dmub_reallow_idle = true, .static_screen_wait_frames = 2, }; @@ -2130,6 +2132,7 @@ static bool dcn35_resource_construct( dc->dml2_options.dcn_pipe_count = pool->base.pipe_count; dc->dml2_options.use_native_pstate_optimization = true; dc->dml2_options.use_native_soc_bb_construction = true; + dc->dml2_options.minimize_dispclk_using_odm = false; if (dc->config.EnableMinDispClkODM) dc->dml2_options.minimize_dispclk_using_odm = true; dc->dml2_options.enable_windowed_mpo_odm = dc->config.enable_windowed_mpo_odm; diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index c78c9224ab60..ae30fe2b6d0d 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -78,6 +78,12 @@ struct dmub_srv_dcn31_regs; struct dmcub_trace_buf_entry; +/* enum dmub_window_memory_type - memory location type specification for windows */ +enum dmub_window_memory_type { + DMUB_WINDOW_MEMORY_TYPE_FB = 0, + DMUB_WINDOW_MEMORY_TYPE_GART +}; + /* enum dmub_status - return code for dmcub functions */ enum dmub_status { DMUB_STATUS_OK = 0, @@ -203,7 +209,7 @@ struct dmub_srv_region_params { uint32_t vbios_size; const uint8_t *fw_inst_const; const uint8_t *fw_bss_data; - bool is_mailbox_in_inbox; + const enum dmub_window_memory_type *window_memory_type; }; /** @@ -223,7 +229,7 @@ struct dmub_srv_region_params { */ struct dmub_srv_region_info { uint32_t fb_size; - uint32_t inbox_size; + uint32_t gart_size; uint8_t num_regions; struct dmub_region regions[DMUB_WINDOW_TOTAL]; }; @@ -239,9 +245,10 @@ struct dmub_srv_region_info { struct dmub_srv_memory_params { const struct dmub_srv_region_info *region_info; void *cpu_fb_addr; - void *cpu_inbox_addr; + void *cpu_gart_addr; uint64_t gpu_fb_addr; - uint64_t gpu_inbox_addr; + uint64_t gpu_gart_addr; + const enum dmub_window_memory_type *window_memory_type; }; /** @@ -443,7 +450,6 @@ struct dmub_srv_create_params { struct dmub_srv_base_funcs funcs; struct dmub_srv_hw_funcs *hw_funcs; void *user_ctx; - struct dc_context *dc_ctx; enum dmub_asic asic; uint32_t fw_version; bool is_virtual; 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 e699731ee68e..59b96136871e 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -26,15 +26,6 @@ #ifndef DMUB_CMD_H #define DMUB_CMD_H -#if defined(_TEST_HARNESS) || defined(FPGA_USB4) -#include "dmub_fw_types.h" -#include "include_legacy/atomfirmware.h" - -#if defined(_TEST_HARNESS) -#include <string.h> -#endif -#else - #include <asm/byteorder.h> #include <linux/types.h> #include <linux/string.h> @@ -42,8 +33,6 @@ #include "atomfirmware.h" -#endif // defined(_TEST_HARNESS) || defined(FPGA_USB4) - //<DMUB_TYPES>================================================================== /* Basic type definitions. */ @@ -403,15 +392,16 @@ union replay_debug_flags { /** * 0x400 (bit 10) - * @force_disable_ips1: Force disable IPS1 state + * @enable_ips_visual_confirm: Enable IPS visual confirm when entering IPS + * If we enter IPS2, the Visual confirm bar will change to yellow */ - uint32_t force_disable_ips1 : 1; + uint32_t enable_ips_visual_confirm : 1; /** * 0x800 (bit 11) - * @force_disable_ips2: Force disable IPS2 state + * @enable_ips_residency_profiling: Enable IPS residency profiling */ - uint32_t force_disable_ips2 : 1; + uint32_t enable_ips_residency_profiling : 1; uint32_t reserved : 20; } bitfields; @@ -1270,11 +1260,11 @@ struct dmub_cmd_PLAT_54186_wa { uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_HIGH_C; /**< reg value */ uint32_t DCSURF_PRIMARY_SURFACE_ADDRESS_C; /**< reg value */ struct { - uint8_t hubp_inst : 4; /**< HUBP instance */ - uint8_t tmz_surface : 1; /**< TMZ enable or disable */ - uint8_t immediate :1; /**< Immediate flip */ - uint8_t vmid : 4; /**< VMID */ - uint8_t grph_stereo : 1; /**< 1 if stereo */ + uint32_t hubp_inst : 4; /**< HUBP instance */ + uint32_t tmz_surface : 1; /**< TMZ enable or disable */ + uint32_t immediate :1; /**< Immediate flip */ + uint32_t vmid : 4; /**< VMID */ + uint32_t grph_stereo : 1; /**< 1 if stereo */ uint32_t reserved : 21; /**< Reserved */ } flip_params; /**< Pageflip parameters */ uint32_t reserved[9]; /**< Reserved bits */ diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c index 2daa1e0c8061..305463b8f110 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c @@ -32,8 +32,6 @@ #include "dcn/dcn_3_2_0_offset.h" #include "dcn/dcn_3_2_0_sh_mask.h" -#define DCN_BASE__INST0_SEG2 0x000034C0 - #define BASE_INNER(seg) ctx->dcn_reg_offsets[seg] #define CTX dmub #define REGS dmub->regs_dcn32 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 9ad738805320..569c2a27a042 100644 --- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv.c @@ -417,58 +417,44 @@ void dmub_srv_destroy(struct dmub_srv *dmub) dmub_memset(dmub, 0, sizeof(*dmub)); } +static uint32_t dmub_srv_calc_regions_for_memory_type(const struct dmub_srv_region_params *params, + struct dmub_srv_region_info *out, + const uint32_t *window_sizes, + enum dmub_window_memory_type memory_type) +{ + uint32_t i, top = 0; + + for (i = 0; i < DMUB_WINDOW_TOTAL; ++i) { + if (params->window_memory_type[i] == memory_type) { + struct dmub_region *region = &out->regions[i]; + + region->base = dmub_align(top, 256); + region->top = region->base + dmub_align(window_sizes[i], 64); + top = region->top; + } + } + + return dmub_align(top, 4096); +} + enum dmub_status -dmub_srv_calc_region_info(struct dmub_srv *dmub, - const struct dmub_srv_region_params *params, - struct dmub_srv_region_info *out) + dmub_srv_calc_region_info(struct dmub_srv *dmub, + const struct dmub_srv_region_params *params, + struct dmub_srv_region_info *out) { - struct dmub_region *inst = &out->regions[DMUB_WINDOW_0_INST_CONST]; - struct dmub_region *stack = &out->regions[DMUB_WINDOW_1_STACK]; - struct dmub_region *data = &out->regions[DMUB_WINDOW_2_BSS_DATA]; - struct dmub_region *bios = &out->regions[DMUB_WINDOW_3_VBIOS]; - struct dmub_region *mail = &out->regions[DMUB_WINDOW_4_MAILBOX]; - struct dmub_region *trace_buff = &out->regions[DMUB_WINDOW_5_TRACEBUFF]; - struct dmub_region *fw_state = &out->regions[DMUB_WINDOW_6_FW_STATE]; - struct dmub_region *scratch_mem = &out->regions[DMUB_WINDOW_7_SCRATCH_MEM]; const struct dmub_fw_meta_info *fw_info; uint32_t fw_state_size = DMUB_FW_STATE_SIZE; uint32_t trace_buffer_size = DMUB_TRACE_BUFFER_SIZE; - uint32_t scratch_mem_size = DMUB_SCRATCH_MEM_SIZE; - uint32_t previous_top = 0; + uint32_t window_sizes[DMUB_WINDOW_TOTAL] = { 0 }; + if (!dmub->sw_init) return DMUB_STATUS_INVALID; memset(out, 0, sizeof(*out)); + memset(window_sizes, 0, sizeof(window_sizes)); out->num_regions = DMUB_NUM_WINDOWS; - inst->base = 0x0; - inst->top = inst->base + params->inst_const_size; - - data->base = dmub_align(inst->top, 256); - data->top = data->base + params->bss_data_size; - - /* - * All cache windows below should be aligned to the size - * of the DMCUB cache line, 64 bytes. - */ - - stack->base = dmub_align(data->top, 256); - stack->top = stack->base + DMUB_STACK_SIZE + DMUB_CONTEXT_SIZE; - - bios->base = dmub_align(stack->top, 256); - bios->top = bios->base + params->vbios_size; - - if (params->is_mailbox_in_inbox) { - mail->base = 0; - mail->top = mail->base + DMUB_MAILBOX_SIZE; - previous_top = bios->top; - } else { - mail->base = dmub_align(bios->top, 256); - mail->top = mail->base + DMUB_MAILBOX_SIZE; - previous_top = mail->top; - } - fw_info = dmub_get_fw_meta_info(params); if (fw_info) { @@ -486,19 +472,20 @@ dmub_srv_calc_region_info(struct dmub_srv *dmub, dmub->fw_version = fw_info->fw_version; } - trace_buff->base = dmub_align(previous_top, 256); - trace_buff->top = trace_buff->base + dmub_align(trace_buffer_size, 64); - - fw_state->base = dmub_align(trace_buff->top, 256); - fw_state->top = fw_state->base + dmub_align(fw_state_size, 64); + window_sizes[DMUB_WINDOW_0_INST_CONST] = params->inst_const_size; + window_sizes[DMUB_WINDOW_1_STACK] = DMUB_STACK_SIZE + DMUB_CONTEXT_SIZE; + window_sizes[DMUB_WINDOW_2_BSS_DATA] = params->bss_data_size; + window_sizes[DMUB_WINDOW_3_VBIOS] = params->vbios_size; + window_sizes[DMUB_WINDOW_4_MAILBOX] = DMUB_MAILBOX_SIZE; + window_sizes[DMUB_WINDOW_5_TRACEBUFF] = trace_buffer_size; + window_sizes[DMUB_WINDOW_6_FW_STATE] = fw_state_size; + window_sizes[DMUB_WINDOW_7_SCRATCH_MEM] = DMUB_SCRATCH_MEM_SIZE; - scratch_mem->base = dmub_align(fw_state->top, 256); - scratch_mem->top = scratch_mem->base + dmub_align(scratch_mem_size, 64); + out->fb_size = + dmub_srv_calc_regions_for_memory_type(params, out, window_sizes, DMUB_WINDOW_MEMORY_TYPE_FB); - out->fb_size = dmub_align(scratch_mem->top, 4096); - - if (params->is_mailbox_in_inbox) - out->inbox_size = dmub_align(mail->top, 4096); + out->gart_size = + dmub_srv_calc_regions_for_memory_type(params, out, window_sizes, DMUB_WINDOW_MEMORY_TYPE_GART); return DMUB_STATUS_OK; } @@ -507,8 +494,6 @@ enum dmub_status dmub_srv_calc_mem_info(struct dmub_srv *dmub, const struct dmub_srv_memory_params *params, struct dmub_srv_fb_info *out) { - uint8_t *cpu_base; - uint64_t gpu_base; uint32_t i; if (!dmub->sw_init) @@ -519,19 +504,16 @@ enum dmub_status dmub_srv_calc_mem_info(struct dmub_srv *dmub, if (params->region_info->num_regions != DMUB_NUM_WINDOWS) return DMUB_STATUS_INVALID; - cpu_base = (uint8_t *)params->cpu_fb_addr; - gpu_base = params->gpu_fb_addr; - for (i = 0; i < DMUB_NUM_WINDOWS; ++i) { const struct dmub_region *reg = ¶ms->region_info->regions[i]; - out->fb[i].cpu_addr = cpu_base + reg->base; - out->fb[i].gpu_addr = gpu_base + reg->base; - - if (i == DMUB_WINDOW_4_MAILBOX && params->cpu_inbox_addr != 0) { - out->fb[i].cpu_addr = (uint8_t *)params->cpu_inbox_addr + reg->base; - out->fb[i].gpu_addr = params->gpu_inbox_addr + reg->base; + if (params->window_memory_type[i] == DMUB_WINDOW_MEMORY_TYPE_GART) { + out->fb[i].cpu_addr = (uint8_t *)params->cpu_gart_addr + reg->base; + out->fb[i].gpu_addr = params->gpu_gart_addr + reg->base; + } else { + out->fb[i].cpu_addr = (uint8_t *)params->cpu_fb_addr + reg->base; + out->fb[i].gpu_addr = params->gpu_fb_addr + reg->base; } out->fb[i].size = reg->top - reg->base; @@ -809,11 +791,20 @@ enum dmub_status dmub_srv_cmd_execute(struct dmub_srv *dmub) bool dmub_srv_is_hw_pwr_up(struct dmub_srv *dmub) { + union dmub_fw_boot_status status; + if (!dmub->hw_funcs.is_hw_powered_up) return true; - return dmub->hw_funcs.is_hw_powered_up(dmub) && - dmub->hw_funcs.is_hw_init(dmub); + if (!dmub->hw_funcs.is_hw_powered_up(dmub)) + return false; + + if (!dmub->hw_funcs.is_hw_init(dmub)) + return false; + + status = dmub->hw_funcs.get_fw_status(dmub); + + return status.bits.dal_fw && status.bits.mailbox_rdy; } enum dmub_status dmub_srv_wait_for_hw_pwr_up(struct dmub_srv *dmub, diff --git a/drivers/gpu/drm/amd/display/include/audio_types.h b/drivers/gpu/drm/amd/display/include/audio_types.h index 915a031a43cb..e4a26143f14c 100644 --- a/drivers/gpu/drm/amd/display/include/audio_types.h +++ b/drivers/gpu/drm/amd/display/include/audio_types.h @@ -27,11 +27,21 @@ #define __AUDIO_TYPES_H__ #include "signal_types.h" +#include "fixed31_32.h" +#include "dc_dp_types.h" #define AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS 20 #define MAX_HW_AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS 18 #define MULTI_CHANNEL_SPLIT_NO_ASSO_INFO 0xFFFFFFFF +struct audio_dp_link_info { + uint32_t link_bandwidth_kbps; + uint32_t hblank_min_symbol_width; + enum dp_link_encoding encoding; + enum dc_link_rate link_rate; + enum dc_lane_count lane_count; + bool is_mst; +}; struct audio_crtc_info { uint32_t h_total; @@ -42,7 +52,10 @@ struct audio_crtc_info { uint32_t calculated_pixel_clock_100Hz; /* in 100Hz */ uint32_t refresh_rate; enum dc_color_depth color_depth; + enum dc_pixel_encoding pixel_encoding; bool interlaced; + uint32_t dsc_bits_per_pixel; + uint32_t dsc_num_slices; }; struct azalia_clock_info { uint32_t pixel_clock_in_10khz; @@ -95,6 +108,8 @@ struct audio_output { enum signal_type signal; /* video timing */ struct audio_crtc_info crtc_info; + /* DP link info */ + struct audio_dp_link_info dp_link_info; /* PLL for audio */ struct audio_pll_info pll_info; }; diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index df2c7ffe190f..a89d93154ddb 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -244,6 +244,7 @@ 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 { diff --git a/drivers/gpu/drm/amd/include/arct_ip_offset.h b/drivers/gpu/drm/amd/include/arct_ip_offset.h index af1c46991429..7dd876f7df74 100644 --- a/drivers/gpu/drm/amd/include/arct_ip_offset.h +++ b/drivers/gpu/drm/amd/include/arct_ip_offset.h @@ -25,13 +25,11 @@ #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/include/asic_reg/dcn/dcn_3_1_6_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_offset.h index 222fa8d13269..a05bf8e4f58d 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_offset.h @@ -626,6 +626,8 @@ #define regDTBCLK_DTO2_MODULO_BASE_IDX 2 #define regDTBCLK_DTO3_MODULO 0x0022 #define regDTBCLK_DTO3_MODULO_BASE_IDX 2 +#define regHDMICHARCLK0_CLOCK_CNTL 0x004a +#define regHDMICHARCLK0_CLOCK_CNTL_BASE_IDX 2 #define regPHYASYMCLK_CLOCK_CNTL 0x0052 #define regPHYASYMCLK_CLOCK_CNTL_BASE_IDX 2 #define regPHYBSYMCLK_CLOCK_CNTL 0x0053 @@ -638,6 +640,8 @@ #define regPHYESYMCLK_CLOCK_CNTL_BASE_IDX 2 #define regPHYFSYMCLK_CLOCK_CNTL 0x0057 #define regPHYFSYMCLK_CLOCK_CNTL_BASE_IDX 2 +#define regHDMISTREAMCLK_CNTL 0x0059 +#define regHDMISTREAMCLK_CNTL_BASE_IDX 2 #define regDCCG_GATE_DISABLE_CNTL3 0x005a #define regDCCG_GATE_DISABLE_CNTL3_BASE_IDX 2 #define regHDMISTREAMCLK0_DTO_PARAM 0x005b diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_sh_mask.h index 8ddb03a1dc39..df84941bbe5b 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_1_6_sh_mask.h @@ -1933,6 +1933,11 @@ //DTBCLK_DTO3_MODULO #define DTBCLK_DTO3_MODULO__DTBCLK_DTO3_MODULO__SHIFT 0x0 #define DTBCLK_DTO3_MODULO__DTBCLK_DTO3_MODULO_MASK 0xFFFFFFFFL +//HDMICHARCLK0_CLOCK_CNTL +#define HDMICHARCLK0_CLOCK_CNTL__HDMICHARCLK0_EN__SHIFT 0x0 +#define HDMICHARCLK0_CLOCK_CNTL__HDMICHARCLK0_SRC_SEL__SHIFT 0x4 +#define HDMICHARCLK0_CLOCK_CNTL__HDMICHARCLK0_EN_MASK 0x00000001L +#define HDMICHARCLK0_CLOCK_CNTL__HDMICHARCLK0_SRC_SEL_MASK 0x00000070L //PHYASYMCLK_CLOCK_CNTL #define PHYASYMCLK_CLOCK_CNTL__PHYASYMCLK_FORCE_EN__SHIFT 0x0 #define PHYASYMCLK_CLOCK_CNTL__PHYASYMCLK_FORCE_SRC_SEL__SHIFT 0x4 @@ -1967,6 +1972,11 @@ #define PHYFSYMCLK_CLOCK_CNTL__PHYFSYMCLK_FORCE_SRC_SEL__SHIFT 0x4 #define PHYFSYMCLK_CLOCK_CNTL__PHYFSYMCLK_FORCE_EN_MASK 0x00000001L #define PHYFSYMCLK_CLOCK_CNTL__PHYFSYMCLK_FORCE_SRC_SEL_MASK 0x00000030L +//HDMISTREAMCLK_CNTL +#define HDMISTREAMCLK_CNTL__HDMISTREAMCLK0_SRC_SEL__SHIFT 0x0 +#define HDMISTREAMCLK_CNTL__HDMISTREAMCLK0_DTO_FORCE_DIS__SHIFT 0x10 +#define HDMISTREAMCLK_CNTL__HDMISTREAMCLK0_SRC_SEL_MASK 0x00000003L +#define HDMISTREAMCLK_CNTL__HDMISTREAMCLK0_DTO_FORCE_DIS_MASK 0x00010000L //DCCG_GATE_DISABLE_CNTL3 #define DCCG_GATE_DISABLE_CNTL3__HDMISTREAMCLK0_GATE_DISABLE__SHIFT 0x0 #define DCCG_GATE_DISABLE_CNTL3__HDMISTREAMCLK1_GATE_DISABLE__SHIFT 0x1 diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_offset.h index 7cf0a625277b..33b5d9be06b1 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_offset.h @@ -4802,6 +4802,10 @@ #define regCM0_CM_DEALPHA_BASE_IDX 2 #define regCM0_CM_COEF_FORMAT 0x0d8c #define regCM0_CM_COEF_FORMAT_BASE_IDX 2 +#define regCM0_CM_TEST_DEBUG_INDEX 0x0d8d +#define regCM0_CM_TEST_DEBUG_INDEX_BASE_IDX 2 +#define regCM0_CM_TEST_DEBUG_DATA 0x0d8e +#define regCM0_CM_TEST_DEBUG_DATA_BASE_IDX 2 // addressBlock: dce_dc_dpp0_dispdec_dpp_dcperfmon_dc_perfmon_dispdec @@ -5210,6 +5214,10 @@ #define regCM1_CM_DEALPHA_BASE_IDX 2 #define regCM1_CM_COEF_FORMAT 0x0ef7 #define regCM1_CM_COEF_FORMAT_BASE_IDX 2 +#define regCM1_CM_TEST_DEBUG_INDEX 0x0ef8 +#define regCM1_CM_TEST_DEBUG_INDEX_BASE_IDX 2 +#define regCM1_CM_TEST_DEBUG_DATA 0x0ef9 +#define regCM1_CM_TEST_DEBUG_DATA_BASE_IDX 2 // addressBlock: dce_dc_dpp1_dispdec_dpp_dcperfmon_dc_perfmon_dispdec @@ -5618,6 +5626,10 @@ #define regCM2_CM_DEALPHA_BASE_IDX 2 #define regCM2_CM_COEF_FORMAT 0x1062 #define regCM2_CM_COEF_FORMAT_BASE_IDX 2 +#define regCM2_CM_TEST_DEBUG_INDEX 0x1063 +#define regCM2_CM_TEST_DEBUG_INDEX_BASE_IDX 2 +#define regCM2_CM_TEST_DEBUG_DATA 0x1064 +#define regCM2_CM_TEST_DEBUG_DATA_BASE_IDX 2 // addressBlock: dce_dc_dpp2_dispdec_dpp_dcperfmon_dc_perfmon_dispdec @@ -6026,6 +6038,10 @@ #define regCM3_CM_DEALPHA_BASE_IDX 2 #define regCM3_CM_COEF_FORMAT 0x11cd #define regCM3_CM_COEF_FORMAT_BASE_IDX 2 +#define regCM3_CM_TEST_DEBUG_INDEX 0x11ce +#define regCM3_CM_TEST_DEBUG_INDEX_BASE_IDX 2 +#define regCM3_CM_TEST_DEBUG_DATA 0x11cf +#define regCM3_CM_TEST_DEBUG_DATA_BASE_IDX 2 // addressBlock: dce_dc_dpp3_dispdec_dpp_dcperfmon_dc_perfmon_dispdec @@ -10568,6 +10584,8 @@ #define regDSCC0_DSCC_RATE_CONTROL_BUFFER2_MAX_FULLNESS_LEVEL_BASE_IDX 2 #define regDSCC0_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL 0x3035 #define regDSCC0_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL_BASE_IDX 2 +#define regDSCC0_DSCC_TEST_DEBUG_BUS_ROTATE 0x303a +#define regDSCC0_DSCC_TEST_DEBUG_BUS_ROTATE_BASE_IDX 2 // addressBlock: dce_dc_dsc0_dispdec_dsc_dcperfmon_dc_perfmon_dispdec @@ -10697,6 +10715,8 @@ #define regDSCC1_DSCC_RATE_CONTROL_BUFFER2_MAX_FULLNESS_LEVEL_BASE_IDX 2 #define regDSCC1_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL 0x3091 #define regDSCC1_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL_BASE_IDX 2 +#define regDSCC1_DSCC_TEST_DEBUG_BUS_ROTATE 0x3096 +#define regDSCC1_DSCC_TEST_DEBUG_BUS_ROTATE_BASE_IDX 2 // addressBlock: dce_dc_dsc1_dispdec_dsc_dcperfmon_dc_perfmon_dispdec @@ -10827,6 +10847,8 @@ #define regDSCC2_DSCC_RATE_CONTROL_BUFFER2_MAX_FULLNESS_LEVEL_BASE_IDX 2 #define regDSCC2_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL 0x30ed #define regDSCC2_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL_BASE_IDX 2 +#define regDSCC2_DSCC_TEST_DEBUG_BUS_ROTATE 0x30f2 +#define regDSCC2_DSCC_TEST_DEBUG_BUS_ROTATE_BASE_IDX 2 // addressBlock: dce_dc_dsc2_dispdec_dsc_dcperfmon_dc_perfmon_dispdec @@ -10957,6 +10979,8 @@ #define regDSCC3_DSCC_RATE_CONTROL_BUFFER2_MAX_FULLNESS_LEVEL_BASE_IDX 2 #define regDSCC3_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL 0x3149 #define regDSCC3_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL_BASE_IDX 2 +#define regDSCC3_DSCC_TEST_DEBUG_BUS_ROTATE 0x314e +#define regDSCC3_DSCC_TEST_DEBUG_BUS_ROTATE_BASE_IDX 2 // addressBlock: dce_dc_dsc3_dispdec_dsc_dcperfmon_dc_perfmon_dispdec diff --git a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_sh_mask.h index fca72e2ec929..ff77b71167eb 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_sh_mask.h +++ b/drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_5_0_sh_mask.h @@ -16556,6 +16556,13 @@ #define CM0_CM_COEF_FORMAT__CM_BIAS_FORMAT_MASK 0x00000001L #define CM0_CM_COEF_FORMAT__CM_POST_CSC_COEF_FORMAT_MASK 0x00000010L #define CM0_CM_COEF_FORMAT__CM_GAMUT_REMAP_COEF_FORMAT_MASK 0x00000100L + +//CM0_CM_TEST_DEBUG_INDEX +#define CM0_CM_TEST_DEBUG_INDEX__CM_TEST_DEBUG_INDEX__SHIFT 0x0 +#define CM0_CM_TEST_DEBUG_INDEX__CM_TEST_DEBUG_WRITE_EN__SHIFT 0x8 +#define CM0_CM_TEST_DEBUG_INDEX__CM_TEST_DEBUG_INDEX_MASK 0x000000FFL +#define CM0_CM_TEST_DEBUG_INDEX__CM_TEST_DEBUG_WRITE_EN_MASK 0x00000100L + #define DC_PERFMON10_PERFCOUNTER_CNTL__PERFCOUNTER_EVENT_SEL__SHIFT 0x0 #define DC_PERFMON10_PERFCOUNTER_CNTL__PERFCOUNTER_CVALUE_SEL__SHIFT 0x9 #define DC_PERFMON10_PERFCOUNTER_CNTL__PERFCOUNTER_INC_MODE__SHIFT 0xc @@ -27176,6 +27183,23 @@ #define DIG0_AFMT_CNTL__AFMT_AUDIO_CLOCK_ON__SHIFT 0x8 #define DIG0_AFMT_CNTL__AFMT_AUDIO_CLOCK_EN_MASK 0x00000001L #define DIG0_AFMT_CNTL__AFMT_AUDIO_CLOCK_ON_MASK 0x00000100L + +//DIG0_DIG_BE_CLK_CNTL +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_MODE__SHIFT 0x0 +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_CLK_EN__SHIFT 0x4 +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SOFT_RESET__SHIFT 0x5 +#define DIG0_DIG_BE_CLK_CNTL__HDCP_SOFT_RESET__SHIFT 0x6 +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SYMCLK_G_CLOCK_ON__SHIFT 0xb +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SYMCLK_G_HDCP_CLOCK_ON__SHIFT 0xc +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SYMCLK_G_TMDS_CLOCK_ON__SHIFT 0xd +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_MODE_MASK 0x00000007L +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_CLK_EN_MASK 0x00000010L +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SOFT_RESET_MASK 0x00000020L +#define DIG0_DIG_BE_CLK_CNTL__HDCP_SOFT_RESET_MASK 0x00000040L +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SYMCLK_G_CLOCK_ON_MASK 0x00000800L +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SYMCLK_G_HDCP_CLOCK_ON_MASK 0x00001000L +#define DIG0_DIG_BE_CLK_CNTL__DIG_BE_SYMCLK_G_TMDS_CLOCK_ON_MASK 0x00002000L + #define DIG0_DIG_BE_CNTL__DIG_DUAL_LINK_ENABLE__SHIFT 0x0 #define DIG0_DIG_BE_CNTL__DIG_SWAP__SHIFT 0x1 #define DIG0_DIG_BE_CNTL__DIG_RB_SWITCH_EN__SHIFT 0x2 @@ -36716,6 +36740,17 @@ #define DSCC0_DSCC_RATE_CONTROL_BUFFER2_MAX_FULLNESS_LEVEL__DSCC_RATE_CONTROL_BUFFER2_MAX_FULLNESS_LEVEL_MASK 0x0003FFFFL #define DSCC0_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL__DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL__SHIFT 0x0 #define DSCC0_DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL__DSCC_RATE_CONTROL_BUFFER3_MAX_FULLNESS_LEVEL_MASK 0x0003FFFFL + +//DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS0_ROTATE__SHIFT 0x0 +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS1_ROTATE__SHIFT 0x8 +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS2_ROTATE__SHIFT 0x10 +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS3_ROTATE__SHIFT 0x18 +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS0_ROTATE_MASK 0x0000001FL +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS1_ROTATE_MASK 0x00001F00L +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS2_ROTATE_MASK 0x001F0000L +#define DSCC0_DSCC_TEST_DEBUG_BUS_ROTATE__DSCC_TEST_DEBUG_BUS3_ROTATE_MASK 0x1F000000L + #define DC_PERFMON17_PERFCOUNTER_CNTL__PERFCOUNTER_EVENT_SEL__SHIFT 0x0 #define DC_PERFMON17_PERFCOUNTER_CNTL__PERFCOUNTER_CVALUE_SEL__SHIFT 0x9 #define DC_PERFMON17_PERFCOUNTER_CNTL__PERFCOUNTER_INC_MODE__SHIFT 0xc @@ -38488,6 +38523,18 @@ #define DWB_OGAM_LUT_INDEX__DWB_OGAM_LUT_INDEX_MASK 0x000001FFL #define DWB_OGAM_LUT_DATA__DWB_OGAM_LUT_DATA__SHIFT 0x0 #define DWB_OGAM_LUT_DATA__DWB_OGAM_LUT_DATA_MASK 0x0003FFFFL +//DWB_OGAM_LUT_CONTROL +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_WRITE_COLOR_MASK__SHIFT 0x0 +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_READ_COLOR_SEL__SHIFT 0x4 +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_READ_DBG__SHIFT 0x8 +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_HOST_SEL__SHIFT 0xc +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_CONFIG_MODE__SHIFT 0x10 +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_WRITE_COLOR_MASK_MASK 0x00000007L +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_READ_COLOR_SEL_MASK 0x00000030L +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_READ_DBG_MASK 0x00000100L +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_HOST_SEL_MASK 0x00001000L +#define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_CONFIG_MODE_MASK 0x00010000L + #define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_WRITE_COLOR_MASK__SHIFT 0x0 #define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_READ_COLOR_SEL__SHIFT 0x4 #define DWB_OGAM_LUT_CONTROL__DWB_OGAM_LUT_HOST_SEL__SHIFT 0xc @@ -52008,6 +52055,14 @@ #define DIO_CLK_CNTL__SYMCLK_R_GATE_DIS__SHIFT 0x10 #define DIO_CLK_CNTL__SYMCLK_G_GATE_DIS__SHIFT 0x11 #define DIO_CLK_CNTL__DIO_FGCG_REP_DIS__SHIFT 0x14 +#define DIO_CLK_CNTL__DISPCLK_G_HDCP_GATE_DIS__SHIFT 0x15 +#define DIO_CLK_CNTL__SYMCLKA_G_HDCP_GATE_DIS__SHIFT 0x16 +#define DIO_CLK_CNTL__SYMCLKB_G_HDCP_GATE_DIS__SHIFT 0x17 +#define DIO_CLK_CNTL__SYMCLKC_G_HDCP_GATE_DIS__SHIFT 0x18 +#define DIO_CLK_CNTL__SYMCLKD_G_HDCP_GATE_DIS__SHIFT 0x19 +#define DIO_CLK_CNTL__SYMCLKE_G_HDCP_GATE_DIS__SHIFT 0x1a +#define DIO_CLK_CNTL__SYMCLKF_G_HDCP_GATE_DIS__SHIFT 0x1b +#define DIO_CLK_CNTL__SYMCLKG_G_HDCP_GATE_DIS__SHIFT 0x1c #define DIO_CLK_CNTL__DIO_TEST_CLK_SEL_MASK 0x0000007FL #define DIO_CLK_CNTL__DISPCLK_R_GATE_DIS_MASK 0x00000200L #define DIO_CLK_CNTL__DISPCLK_G_GATE_DIS_MASK 0x00000400L @@ -52019,6 +52074,16 @@ #define DIO_CLK_CNTL__SYMCLK_R_GATE_DIS_MASK 0x00010000L #define DIO_CLK_CNTL__SYMCLK_G_GATE_DIS_MASK 0x00020000L #define DIO_CLK_CNTL__DIO_FGCG_REP_DIS_MASK 0x00100000L + +#define DIO_CLK_CNTL__DISPCLK_G_HDCP_GATE_DIS_MASK 0x00200000L +#define DIO_CLK_CNTL__SYMCLKA_G_HDCP_GATE_DIS_MASK 0x00400000L +#define DIO_CLK_CNTL__SYMCLKB_G_HDCP_GATE_DIS_MASK 0x00800000L +#define DIO_CLK_CNTL__SYMCLKC_G_HDCP_GATE_DIS_MASK 0x01000000L +#define DIO_CLK_CNTL__SYMCLKD_G_HDCP_GATE_DIS_MASK 0x02000000L +#define DIO_CLK_CNTL__SYMCLKE_G_HDCP_GATE_DIS_MASK 0x04000000L +#define DIO_CLK_CNTL__SYMCLKF_G_HDCP_GATE_DIS_MASK 0x08000000L +#define DIO_CLK_CNTL__SYMCLKG_G_HDCP_GATE_DIS_MASK 0x10000000L + #define DIO_PSP_INTERRUPT_STATUS__DIO_PSP_INTERRUPT_STATUS__SHIFT 0x0 #define DIO_PSP_INTERRUPT_STATUS__DIO_PSP_INTERRUPT_MESSAGE__SHIFT 0x1 #define DIO_PSP_INTERRUPT_STATUS__DIO_PSP_INTERRUPT_STATUS_MASK 0x00000001L diff --git a/drivers/gpu/drm/amd/include/atom-bits.h b/drivers/gpu/drm/amd/include/atom-bits.h index e8fae5c77514..2bfd6d0ff050 100644 --- a/drivers/gpu/drm/amd/include/atom-bits.h +++ b/drivers/gpu/drm/amd/include/atom-bits.h @@ -33,7 +33,7 @@ static inline uint8_t get_u8(void *bios, int ptr) #define CU8(ptr) get_u8(ctx->bios, (ptr)) static inline uint16_t get_u16(void *bios, int ptr) { - return get_u8(bios ,ptr)|(((uint16_t)get_u8(bios, ptr+1))<<8); + return get_u8(bios, ptr)|(((uint16_t)get_u8(bios, ptr+1))<<8); } #define U16(ptr) get_u16(ctx->ctx->bios, (ptr)) #define CU16(ptr) get_u16(ctx->bios, (ptr)) diff --git a/drivers/gpu/drm/amd/include/beige_goby_ip_offset.h b/drivers/gpu/drm/amd/include/beige_goby_ip_offset.h index 26044cb285d2..48542ea6882a 100644 --- a/drivers/gpu/drm/amd/include/beige_goby_ip_offset.h +++ b/drivers/gpu/drm/amd/include/beige_goby_ip_offset.h @@ -26,13 +26,11 @@ #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]; }; diff --git a/drivers/gpu/drm/amd/include/cgs_common.h b/drivers/gpu/drm/amd/include/cgs_common.h index 60a6536ff656..f40b6a03fe63 100644 --- a/drivers/gpu/drm/amd/include/cgs_common.h +++ b/drivers/gpu/drm/amd/include/cgs_common.h @@ -149,27 +149,26 @@ struct cgs_ops { struct cgs_os_ops; /* To be define in OS-specific CGS header */ -struct cgs_device -{ +struct cgs_device { const struct cgs_ops *ops; /* to be embedded at the start of driver private structure */ }; /* Convenience macros that make CGS indirect function calls look like * normal function calls */ -#define CGS_CALL(func,dev,...) \ +#define CGS_CALL(func, dev, ...) \ (((struct cgs_device *)dev)->ops->func(dev, ##__VA_ARGS__)) -#define CGS_OS_CALL(func,dev,...) \ +#define CGS_OS_CALL(func, dev, ...) \ (((struct cgs_device *)dev)->os_ops->func(dev, ##__VA_ARGS__)) -#define cgs_read_register(dev,offset) \ - CGS_CALL(read_register,dev,offset) -#define cgs_write_register(dev,offset,value) \ - CGS_CALL(write_register,dev,offset,value) -#define cgs_read_ind_register(dev,space,index) \ - CGS_CALL(read_ind_register,dev,space,index) -#define cgs_write_ind_register(dev,space,index,value) \ - CGS_CALL(write_ind_register,dev,space,index,value) +#define cgs_read_register(dev, offset) \ + CGS_CALL(read_register, dev, offset) +#define cgs_write_register(dev, offset, value) \ + CGS_CALL(write_register, dev, offset, value) +#define cgs_read_ind_register(dev, space, index) \ + CGS_CALL(read_ind_register, dev, space, index) +#define cgs_write_ind_register(dev, space, index, value) \ + CGS_CALL(write_ind_register, dev, space, index, value) #define cgs_get_firmware_info(dev, type, info) \ CGS_CALL(get_firmware_info, dev, type, info) diff --git a/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h b/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h index ce79e5de8ce3..1a73296a9a74 100644 --- a/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h +++ b/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h @@ -25,13 +25,11 @@ #define MAX_SEGMENT 5 -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/include/dimgrey_cavefish_ip_offset.h b/drivers/gpu/drm/amd/include/dimgrey_cavefish_ip_offset.h index f84996a73de9..53cb4296df88 100644 --- a/drivers/gpu/drm/amd/include/dimgrey_cavefish_ip_offset.h +++ b/drivers/gpu/drm/amd/include/dimgrey_cavefish_ip_offset.h @@ -25,13 +25,11 @@ #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]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/include/dm_pp_interface.h b/drivers/gpu/drm/amd/include/dm_pp_interface.h index 1d93a0c574c9..acd1cef61b7c 100644 --- a/drivers/gpu/drm/amd/include/dm_pp_interface.h +++ b/drivers/gpu/drm/amd/include/dm_pp_interface.h @@ -27,7 +27,7 @@ #define PP_MAX_CLOCK_LEVELS 16 -enum amd_pp_display_config_type{ +enum amd_pp_display_config_type { AMD_PP_DisplayConfigType_None = 0, AMD_PP_DisplayConfigType_DP54 , AMD_PP_DisplayConfigType_DP432 , @@ -36,8 +36,8 @@ enum amd_pp_display_config_type{ AMD_PP_DisplayConfigType_DP243, AMD_PP_DisplayConfigType_DP216, AMD_PP_DisplayConfigType_DP162, - AMD_PP_DisplayConfigType_HDMI6G , - AMD_PP_DisplayConfigType_HDMI297 , + AMD_PP_DisplayConfigType_HDMI6G, + AMD_PP_DisplayConfigType_HDMI297, AMD_PP_DisplayConfigType_HDMI162, AMD_PP_DisplayConfigType_LVDS, AMD_PP_DisplayConfigType_DVI, @@ -45,8 +45,7 @@ enum amd_pp_display_config_type{ AMD_PP_DisplayConfigType_VGA }; -struct single_display_configuration -{ +struct single_display_configuration { uint32_t controller_index; uint32_t controller_id; uint32_t signal_type; diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index edcb85560ced..32054ecf0b87 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -244,8 +244,7 @@ enum pp_df_cstate { * @PP_PWR_LIMIT_DEFAULT: Default Power Limit * @PP_PWR_LIMIT_MAX: Maximum Power Limit */ -enum pp_power_limit_level -{ +enum pp_power_limit_level { PP_PWR_LIMIT_MIN = -1, PP_PWR_LIMIT_CURRENT, PP_PWR_LIMIT_DEFAULT, @@ -260,8 +259,7 @@ enum pp_power_limit_level * @PP_PWR_TYPE_FAST: manages the ~10 ms moving average of APU power, * where supported. */ -enum pp_power_type -{ +enum pp_power_type { PP_PWR_TYPE_SUSTAINED, PP_PWR_TYPE_FAST, }; diff --git a/drivers/gpu/drm/amd/include/navi12_ip_offset.h b/drivers/gpu/drm/amd/include/navi12_ip_offset.h index d8fc00478b6a..e94d80ec8d92 100644 --- a/drivers/gpu/drm/amd/include/navi12_ip_offset.h +++ b/drivers/gpu/drm/amd/include/navi12_ip_offset.h @@ -25,13 +25,11 @@ #define MAX_SEGMENT 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]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/include/navi14_ip_offset.h b/drivers/gpu/drm/amd/include/navi14_ip_offset.h index c39ef651adc6..508011288dea 100644 --- a/drivers/gpu/drm/amd/include/navi14_ip_offset.h +++ b/drivers/gpu/drm/amd/include/navi14_ip_offset.h @@ -25,13 +25,11 @@ #define MAX_SEGMENT 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]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/include/pptable.h b/drivers/gpu/drm/amd/include/pptable.h index 5aac8d545bdc..2e8e6c9875f6 100644 --- a/drivers/gpu/drm/amd/include/pptable.h +++ b/drivers/gpu/drm/amd/include/pptable.h @@ -491,7 +491,7 @@ typedef struct _ClockInfoArray{ //sizeof(ATOM_PPLIB_CLOCK_INFO) UCHAR ucEntrySize; - UCHAR clockInfo[1]; + UCHAR clockInfo[]; }ClockInfoArray; typedef struct _NonClockInfoArray{ @@ -501,7 +501,7 @@ typedef struct _NonClockInfoArray{ //sizeof(ATOM_PPLIB_NONCLOCK_INFO) UCHAR ucEntrySize; - ATOM_PPLIB_NONCLOCK_INFO nonClockInfo[1]; + ATOM_PPLIB_NONCLOCK_INFO nonClockInfo[]; }NonClockInfoArray; typedef struct _ATOM_PPLIB_Clock_Voltage_Dependency_Record @@ -658,7 +658,7 @@ typedef struct _ATOM_PPLIB_SAMClk_Voltage_Limit_Record typedef struct _ATOM_PPLIB_SAMClk_Voltage_Limit_Table{ UCHAR numEntries; - ATOM_PPLIB_SAMClk_Voltage_Limit_Record entries[1]; + ATOM_PPLIB_SAMClk_Voltage_Limit_Record entries[]; }ATOM_PPLIB_SAMClk_Voltage_Limit_Table; typedef struct _ATOM_PPLIB_SAMU_Table diff --git a/drivers/gpu/drm/amd/include/renoir_ip_offset.h b/drivers/gpu/drm/amd/include/renoir_ip_offset.h index 7dff85c81e5a..fa023cfdf72d 100644 --- a/drivers/gpu/drm/amd/include/renoir_ip_offset.h +++ b/drivers/gpu/drm/amd/include/renoir_ip_offset.h @@ -25,13 +25,11 @@ #define MAX_SEGMENT 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]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/include/sienna_cichlid_ip_offset.h b/drivers/gpu/drm/amd/include/sienna_cichlid_ip_offset.h index b07bc2dd895d..054790470800 100644 --- a/drivers/gpu/drm/amd/include/sienna_cichlid_ip_offset.h +++ b/drivers/gpu/drm/amd/include/sienna_cichlid_ip_offset.h @@ -25,13 +25,11 @@ #define MAX_SEGMENT 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]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/include/v10_structs.h b/drivers/gpu/drm/amd/include/v10_structs.h index c0e98a98a641..58002a83d1df 100644 --- a/drivers/gpu/drm/amd/include/v10_structs.h +++ b/drivers/gpu/drm/amd/include/v10_structs.h @@ -24,8 +24,7 @@ #ifndef V10_STRUCTS_H_ #define V10_STRUCTS_H_ -struct v10_gfx_mqd -{ +struct v10_gfx_mqd { uint32_t reserved_0; // offset: 0 (0x0) uint32_t reserved_1; // offset: 1 (0x1) uint32_t reserved_2; // offset: 2 (0x2) diff --git a/drivers/gpu/drm/amd/include/vangogh_ip_offset.h b/drivers/gpu/drm/amd/include/vangogh_ip_offset.h index 691073ed780e..695d7d04dfa6 100644 --- a/drivers/gpu/drm/amd/include/vangogh_ip_offset.h +++ b/drivers/gpu/drm/amd/include/vangogh_ip_offset.h @@ -28,13 +28,11 @@ #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]; } __maybe_unused; diff --git a/drivers/gpu/drm/amd/include/vega10_ip_offset.h b/drivers/gpu/drm/amd/include/vega10_ip_offset.h index 3a22a5d16919..1e1ca69f21f7 100644 --- a/drivers/gpu/drm/amd/include/vega10_ip_offset.h +++ b/drivers/gpu/drm/amd/include/vega10_ip_offset.h @@ -24,13 +24,11 @@ #define MAX_INSTANCE 5 #define MAX_SEGMENT 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]; }; diff --git a/drivers/gpu/drm/amd/include/vega20_ip_offset.h b/drivers/gpu/drm/amd/include/vega20_ip_offset.h index 1deb68f3d334..92cf2d9e767f 100644 --- a/drivers/gpu/drm/amd/include/vega20_ip_offset.h +++ b/drivers/gpu/drm/amd/include/vega20_ip_offset.h @@ -25,139 +25,137 @@ #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]; } __maybe_unused; -static const struct IP_BASE ATHUB_BASE ={ { { { 0x00000C20, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE ATHUB_BASE = { { { { 0x00000C20, 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 CLK_BASE = { { { { 0x00016C00, 0x00016E00, 0x00017000, 0x00017200, 0x0001B000, 0x0001B200 } }, { { 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 CLK_BASE ={ { { { 0x00016C00, 0x00016E00, 0x00017000, 0x00017200, 0x0001B000, 0x0001B200 } }, +static const struct IP_BASE DCE_BASE = { { { { 0x00000012, 0x000000C0, 0x000034C0, 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 DF_BASE = { { { { 0x00007000, 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 FUSE_BASE = { { { { 0x00017400, 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 DCE_BASE ={ { { { 0x00000012, 0x000000C0, 0x000034C0, 0, 0, 0 } }, +static const struct IP_BASE GC_BASE = { { { { 0x00002000, 0x0000A000, 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 DF_BASE ={ { { { 0x00007000, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE HDP_BASE = { { { { 0x00000F20, 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 FUSE_BASE ={ { { { 0x00017400, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE MMHUB_BASE = { { { { 0x0001A000, 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 GC_BASE ={ { { { 0x00002000, 0x0000A000, 0, 0, 0, 0 } }, +static const struct IP_BASE MP0_BASE = { { { { 0x00016000, 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 HDP_BASE ={ { { { 0x00000F20, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE MP1_BASE = { { { { 0x00016000, 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 MMHUB_BASE ={ { { { 0x0001A000, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE NBIO_BASE = { { { { 0x00000000, 0x00000014, 0x00000D20, 0x00010400, 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 MP0_BASE ={ { { { 0x00016000, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE OSSSYS_BASE = { { { { 0x000010A0, 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 MP1_BASE ={ { { { 0x00016000, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE SDMA0_BASE = { { { { 0x00001260, 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, 0, 0 } }, +static const struct IP_BASE SDMA1_BASE = { { { { 0x00001860, 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 OSSSYS_BASE ={ { { { 0x000010A0, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE SMUIO_BASE = { { { { 0x00016800, 0x00016A00, 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 SDMA0_BASE ={ { { { 0x00001260, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE THM_BASE = { { { { 0x00016600, 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 SDMA1_BASE ={ { { { 0x00001860, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE UMC_BASE = { { { { 0x00014000, 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 SMUIO_BASE ={ { { { 0x00016800, 0x00016A00, 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 THM_BASE ={ { { { 0x00016600, 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 UMC_BASE ={ { { { 0x00014000, 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 UVD_BASE ={ { { { 0x00007800, 0x00007E00, 0, 0, 0, 0 } }, +static const struct IP_BASE UVD_BASE = { { { { 0x00007800, 0x00007E00, 0, 0, 0, 0 } }, { { 0, 0x00009000, 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 } } } }; /* Adjust VCE_BASE to make vce_4_1 use vce_4_0 offset header files*/ -static const struct IP_BASE VCE_BASE ={ { { { 0x00007E00/* 0x00008800 */, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE VCE_BASE = { { { { 0x00007E00/* 0x00008800 */, 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 XDMA_BASE ={ { { { 0x00003400, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE XDMA_BASE = { { { { 0x00003400, 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 RSMU_BASE ={ { { { 0x00012000, 0, 0, 0, 0, 0 } }, +static const struct IP_BASE RSMU_BASE = { { { { 0x00012000, 0, 0, 0, 0, 0 } }, { { 0, 0, 0, 0, 0, 0 } }, { { 0, 0, 0, 0, 0, 0 } }, { { 0, 0, 0, 0, 0, 0 } }, diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c index f503e61faa60..b1b4c09c3467 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c @@ -226,7 +226,7 @@ int atomctrl_set_engine_dram_timings_rv770( return amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, DynamicMemorySettings), - (uint32_t *)&engine_clock_parameters); + (uint32_t *)&engine_clock_parameters, sizeof(engine_clock_parameters)); } /* @@ -297,7 +297,7 @@ int atomctrl_get_memory_pll_dividers_si( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryClockParam), - (uint32_t *)&mpll_parameters); + (uint32_t *)&mpll_parameters, sizeof(mpll_parameters)); if (0 == result) { mpll_param->mpll_fb_divider.clk_frac = @@ -345,7 +345,7 @@ int atomctrl_get_memory_pll_dividers_vi(struct pp_hwmgr *hwmgr, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryClockParam), - (uint32_t *)&mpll_parameters); + (uint32_t *)&mpll_parameters, sizeof(mpll_parameters)); if (!result) mpll_param->mpll_post_divider = @@ -366,7 +366,7 @@ int atomctrl_get_memory_pll_dividers_ai(struct pp_hwmgr *hwmgr, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryClockParam), - (uint32_t *)&mpll_parameters); + (uint32_t *)&mpll_parameters, sizeof(mpll_parameters)); /* VEGAM's mpll takes sometime to finish computing */ udelay(10); @@ -396,7 +396,7 @@ int atomctrl_get_engine_pll_dividers_kong(struct pp_hwmgr *hwmgr, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL), - (uint32_t *)&pll_parameters); + (uint32_t *)&pll_parameters, sizeof(pll_parameters)); if (0 == result) { dividers->pll_post_divider = pll_parameters.ucPostDiv; @@ -420,7 +420,7 @@ int atomctrl_get_engine_pll_dividers_vi( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL), - (uint32_t *)&pll_patameters); + (uint32_t *)&pll_patameters, sizeof(pll_patameters)); if (0 == result) { dividers->pll_post_divider = @@ -457,7 +457,7 @@ int atomctrl_get_engine_pll_dividers_ai(struct pp_hwmgr *hwmgr, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL), - (uint32_t *)&pll_patameters); + (uint32_t *)&pll_patameters, sizeof(pll_patameters)); if (0 == result) { dividers->usSclk_fcw_frac = le16_to_cpu(pll_patameters.usSclk_fcw_frac); @@ -490,7 +490,7 @@ int atomctrl_get_dfs_pll_dividers_vi( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ComputeMemoryEnginePLL), - (uint32_t *)&pll_patameters); + (uint32_t *)&pll_patameters, sizeof(pll_patameters)); if (0 == result) { dividers->pll_post_divider = @@ -773,7 +773,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -794,7 +794,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -814,7 +814,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -835,7 +835,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -857,7 +857,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -878,7 +878,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -909,7 +909,7 @@ int atomctrl_calculate_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues); + (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); if (result) return result; @@ -1134,7 +1134,7 @@ int atomctrl_get_voltage_evv_on_sclk( result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, GetVoltageInfo), - (uint32_t *)&get_voltage_info_param_space); + (uint32_t *)&get_voltage_info_param_space, sizeof(get_voltage_info_param_space)); *voltage = result ? 0 : le16_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_2 *) @@ -1179,7 +1179,7 @@ int atomctrl_get_voltage_evv(struct pp_hwmgr *hwmgr, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, GetVoltageInfo), - (uint32_t *)&get_voltage_info_param_space); + (uint32_t *)&get_voltage_info_param_space, sizeof(get_voltage_info_param_space)); if (0 != result) return result; @@ -1359,7 +1359,7 @@ int atomctrl_read_efuse(struct pp_hwmgr *hwmgr, uint16_t start_index, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&efuse_param); + (uint32_t *)&efuse_param, sizeof(efuse_param)); *efuse = result ? 0 : le32_to_cpu(efuse_param.ulEfuseValue) & mask; return result; @@ -1380,7 +1380,7 @@ int atomctrl_set_ac_timing_ai(struct pp_hwmgr *hwmgr, uint32_t memory_clock, result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, DynamicMemorySettings), - (uint32_t *)&memory_clock_parameters); + (uint32_t *)&memory_clock_parameters, sizeof(memory_clock_parameters)); return result; } @@ -1399,7 +1399,7 @@ int atomctrl_get_voltage_evv_on_sclk_ai(struct pp_hwmgr *hwmgr, uint8_t voltage_ result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, GetVoltageInfo), - (uint32_t *)&get_voltage_info_param_space); + (uint32_t *)&get_voltage_info_param_space, sizeof(get_voltage_info_param_space)); *voltage = result ? 0 : le32_to_cpu(((GET_EVV_VOLTAGE_INFO_OUTPUT_PARAMETER_V1_3 *)(&get_voltage_info_param_space))->ulVoltageLevel); @@ -1526,7 +1526,7 @@ int atomctrl_get_leakage_id_from_efuse(struct pp_hwmgr *hwmgr, uint16_t *virtual result = amdgpu_atom_execute_table(adev->mode_info.atom_context, GetIndexIntoMasterTable(COMMAND, SetVoltage), - (uint32_t *)voltage_parameters); + (uint32_t *)voltage_parameters, sizeof(*voltage_parameters)); *virtual_voltage_id = voltage_parameters->usVoltageLevel; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c index a47a47238e2b..82d540334318 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomfwctrl.c @@ -258,7 +258,7 @@ int pp_atomfwctrl_get_gpu_pll_dividers_vega10(struct pp_hwmgr *hwmgr, idx = GetIndexIntoMasterCmdTable(computegpuclockparam); if (amdgpu_atom_execute_table( - adev->mode_info.atom_context, idx, (uint32_t *)&pll_parameters)) + adev->mode_info.atom_context, idx, (uint32_t *)&pll_parameters, sizeof(pll_parameters))) return -EINVAL; pll_output = (struct compute_gpu_clock_output_parameter_v1_8 *) @@ -505,7 +505,7 @@ int pp_atomfwctrl_get_clk_information_by_clkid(struct pp_hwmgr *hwmgr, ix = GetIndexIntoMasterCmdTable(getsmuclockinfo); if (amdgpu_atom_execute_table( - adev->mode_info.atom_context, ix, (uint32_t *)¶meters)) + adev->mode_info.atom_context, ix, (uint32_t *)¶meters, sizeof(parameters))) return -EINVAL; output = (struct atom_get_smu_clock_info_output_parameters_v3_1 *)¶meters; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c index c7bfa68bf00f..f6545093bfc1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c @@ -514,7 +514,7 @@ static int smu_v11_0_atom_get_smu_clockinfo(struct amdgpu_device *adev, getsmuclockinfo); ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index, - (uint32_t *)&input); + (uint32_t *)&input, sizeof(input)); if (ret) return -EINVAL; @@ -1432,24 +1432,24 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev, dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n"); orderly_poweroff(true); } else if (client_id == SOC15_IH_CLIENTID_MP1) { - if (src_id == 0xfe) { + if (src_id == SMU_IH_INTERRUPT_ID_TO_DRIVER) { /* ACK SMUToHost interrupt */ data = RREG32_SOC15(MP1, 0, mmMP1_SMN_IH_SW_INT_CTRL); data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1); WREG32_SOC15(MP1, 0, mmMP1_SMN_IH_SW_INT_CTRL, data); switch (ctxid) { - case 0x3: + case SMU_IH_INTERRUPT_CONTEXT_ID_AC: dev_dbg(adev->dev, "Switched to AC mode!\n"); schedule_work(&smu->interrupt_work); adev->pm.ac_power = true; break; - case 0x4: + case SMU_IH_INTERRUPT_CONTEXT_ID_DC: dev_dbg(adev->dev, "Switched to DC mode!\n"); schedule_work(&smu->interrupt_work); adev->pm.ac_power = false; break; - case 0x7: + case SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING: /* * Increment the throttle interrupt counter */ @@ -1462,6 +1462,10 @@ static int smu_v11_0_irq_process(struct amdgpu_device *adev, schedule_work(&smu->throttling_logging_work); break; + default: + dev_dbg(adev->dev, "Unhandled context id %d from client:%d!\n", + ctxid, client_id); + break; } } } @@ -1504,7 +1508,7 @@ int smu_v11_0_register_irq_handler(struct smu_context *smu) return ret; ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, - 0xfe, + SMU_IH_INTERRUPT_ID_TO_DRIVER, irq_src); if (ret) return ret; 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 5e408a195860..ed15f5a0fd11 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 @@ -301,7 +301,7 @@ static int smu_v12_0_atom_get_smu_clockinfo(struct amdgpu_device *adev, getsmuclockinfo); ret = amdgpu_atom_execute_table(adev->mode_info.atom_context, index, - (uint32_t *)&input); + (uint32_t *)&input, sizeof(input)); if (ret) return -EINVAL; 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 c486182ff275..48170bb5112e 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 @@ -1369,24 +1369,24 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, dev_emerg(adev->dev, "ERROR: System is going to shutdown due to GPU HW CTF!\n"); orderly_poweroff(true); } else if (client_id == SOC15_IH_CLIENTID_MP1) { - if (src_id == 0xfe) { + if (src_id == SMU_IH_INTERRUPT_ID_TO_DRIVER) { /* ACK SMUToHost interrupt */ data = RREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL); data = REG_SET_FIELD(data, MP1_SMN_IH_SW_INT_CTRL, INT_ACK, 1); WREG32_SOC15(MP1, 0, regMP1_SMN_IH_SW_INT_CTRL, data); switch (ctxid) { - case 0x3: + case SMU_IH_INTERRUPT_CONTEXT_ID_AC: dev_dbg(adev->dev, "Switched to AC mode!\n"); smu_v13_0_ack_ac_dc_interrupt(smu); adev->pm.ac_power = true; break; - case 0x4: + case SMU_IH_INTERRUPT_CONTEXT_ID_DC: dev_dbg(adev->dev, "Switched to DC mode!\n"); smu_v13_0_ack_ac_dc_interrupt(smu); adev->pm.ac_power = false; break; - case 0x7: + case SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING: /* * Increment the throttle interrupt counter */ @@ -1399,7 +1399,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, schedule_work(&smu->throttling_logging_work); break; - case 0x8: + case SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL: high = smu->thermal_range.software_shutdown_temp + smu->thermal_range.software_shutdown_temp_offset; high = min_t(typeof(high), @@ -1416,7 +1416,7 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, data = data & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK); WREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL, data); break; - case 0x9: + case SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY: high = min_t(typeof(high), SMU_THERMAL_MAXIMUM_ALERT_TEMP, smu->thermal_range.software_shutdown_temp); @@ -1429,6 +1429,10 @@ static int smu_v13_0_irq_process(struct amdgpu_device *adev, data = data & (~THM_THERMAL_INT_CTRL__THERM_TRIGGER_MASK_MASK); WREG32_SOC15(THM, 0, regTHM_THERMAL_INT_CTRL, data); break; + default: + dev_dbg(adev->dev, "Unhandled context id %d from client:%d!\n", + ctxid, client_id); + break; } } } @@ -1473,7 +1477,7 @@ int smu_v13_0_register_irq_handler(struct smu_context *smu) return ret; ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, - 0xfe, + SMU_IH_INTERRUPT_ID_TO_DRIVER, irq_src); if (ret) return ret; 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 7e1941cf1796..1b96158b63bf 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 @@ -45,6 +45,7 @@ #include <linux/pci.h> #include "amdgpu_ras.h" #include "amdgpu_mca.h" +#include "amdgpu_aca.h" #include "smu_cmn.h" #include "mp/mp_13_0_6_offset.h" #include "mp/mp_13_0_6_sh_mask.h" @@ -1438,7 +1439,10 @@ static int smu_v13_0_6_irq_process(struct amdgpu_device *adev, entry->src_data[1]); schedule_work(&smu->throttling_logging_work); } - + break; + default: + dev_dbg(adev->dev, "Unhandled context id %d from client:%d!\n", + ctxid, client_id); break; } } @@ -2547,18 +2551,22 @@ static int mca_umc_mca_get_err_count(const struct mca_ras_info *mca_ras, struct enum amdgpu_mca_error_type type, struct mca_bank_entry *entry, uint32_t *count) { uint64_t status0; + uint32_t ext_error_code; + uint32_t odecc_err_cnt; status0 = entry->regs[MCA_REG_IDX_STATUS]; + ext_error_code = MCA_REG__STATUS__ERRORCODEEXT(status0); + odecc_err_cnt = MCA_REG__MISC0__ERRCNT(entry->regs[MCA_REG_IDX_MISC0]); if (!REG_GET_FIELD(status0, MCMP1_STATUST0, Val)) { *count = 0; return 0; } - if (type == AMDGPU_MCA_ERROR_TYPE_UE && umc_v12_0_is_uncorrectable_error(adev, status0)) - *count = 1; - else if (type == AMDGPU_MCA_ERROR_TYPE_CE && umc_v12_0_is_correctable_error(adev, status0)) - *count = 1; + if (umc_v12_0_is_deferred_error(adev, status0) || + umc_v12_0_is_uncorrectable_error(adev, status0) || + umc_v12_0_is_correctable_error(adev, status0)) + *count = (ext_error_code == 0) ? odecc_err_cnt : 1; return 0; } @@ -2857,6 +2865,143 @@ static const struct amdgpu_mca_smu_funcs smu_v13_0_6_mca_smu_funcs = { .mca_get_valid_mca_count = mca_smu_get_valid_mca_count, }; +static int aca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + + return smu_v13_0_6_mca_set_debug_mode(smu, enable); +} + +static int smu_v13_0_6_get_valid_aca_count(struct smu_context *smu, enum aca_error_type type, u32 *count) +{ + uint32_t msg; + int ret; + + if (!count) + return -EINVAL; + + switch (type) { + case ACA_ERROR_TYPE_UE: + msg = SMU_MSG_QueryValidMcaCount; + break; + case ACA_ERROR_TYPE_CE: + msg = SMU_MSG_QueryValidMcaCeCount; + break; + default: + return -EINVAL; + } + + ret = smu_cmn_send_smc_msg(smu, msg, count); + if (ret) { + *count = 0; + return ret; + } + + return 0; +} + +static int aca_smu_get_valid_aca_count(struct amdgpu_device *adev, + enum aca_error_type type, u32 *count) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + int ret; + + switch (type) { + case ACA_ERROR_TYPE_UE: + case ACA_ERROR_TYPE_CE: + ret = smu_v13_0_6_get_valid_aca_count(smu, type, count); + break; + default: + ret = -EINVAL; + break; + } + + return ret; +} + +static int __smu_v13_0_6_aca_bank_dump(struct smu_context *smu, enum aca_error_type type, + int idx, int offset, u32 *val) +{ + uint32_t msg, param; + + switch (type) { + case ACA_ERROR_TYPE_UE: + msg = SMU_MSG_McaBankDumpDW; + break; + case ACA_ERROR_TYPE_CE: + msg = SMU_MSG_McaBankCeDumpDW; + break; + default: + return -EINVAL; + } + + param = ((idx & 0xffff) << 16) | (offset & 0xfffc); + + return smu_cmn_send_smc_msg_with_param(smu, msg, param, (uint32_t *)val); +} + +static int smu_v13_0_6_aca_bank_dump(struct smu_context *smu, enum aca_error_type type, + int idx, int offset, u32 *val, int count) +{ + int ret, i; + + if (!val) + return -EINVAL; + + for (i = 0; i < count; i++) { + ret = __smu_v13_0_6_aca_bank_dump(smu, type, idx, offset + (i << 2), &val[i]); + if (ret) + return ret; + } + + return 0; +} + +static int aca_bank_read_reg(struct amdgpu_device *adev, enum aca_error_type type, + int idx, int reg_idx, u64 *val) +{ + struct smu_context *smu = adev->powerplay.pp_handle; + u32 data[2] = {0, 0}; + int ret; + + if (!val || reg_idx >= ACA_REG_IDX_COUNT) + return -EINVAL; + + ret = smu_v13_0_6_aca_bank_dump(smu, type, idx, reg_idx * 8, data, ARRAY_SIZE(data)); + if (ret) + return ret; + + *val = (u64)data[1] << 32 | data[0]; + + dev_dbg(adev->dev, "mca read bank reg: type:%s, index: %d, reg_idx: %d, val: 0x%016llx\n", + type == ACA_ERROR_TYPE_UE ? "UE" : "CE", idx, reg_idx, *val); + + return 0; +} + +static int aca_smu_get_valid_aca_bank(struct amdgpu_device *adev, + enum aca_error_type type, int idx, struct aca_bank *bank) +{ + int i, ret, count; + + count = min_t(int, 16, ARRAY_SIZE(bank->regs)); + for (i = 0; i < count; i++) { + ret = aca_bank_read_reg(adev, type, idx, i, &bank->regs[i]); + if (ret) + return ret; + } + + return 0; +} + +static const struct aca_smu_funcs smu_v13_0_6_aca_smu_funcs = { + .max_ue_bank_count = 12, + .max_ce_bank_count = 12, + .set_debug_mode = aca_smu_set_debug_mode, + .get_valid_aca_count = aca_smu_get_valid_aca_count, + .get_valid_aca_bank = aca_smu_get_valid_aca_bank, +}; + static int smu_v13_0_6_select_xgmi_plpd_policy(struct smu_context *smu, enum pp_xgmi_plpd_mode mode) { @@ -2895,13 +3040,6 @@ static int smu_v13_0_6_select_xgmi_plpd_policy(struct smu_context *smu, return ret; } -static ssize_t smu_v13_0_6_get_ecc_info(struct smu_context *smu, - void *table) -{ - /* Support ecc info by default */ - return 0; -} - static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { /* init dpm */ .get_allowed_feature_mask = smu_v13_0_6_get_allowed_feature_mask, @@ -2956,7 +3094,6 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = { .i2c_init = smu_v13_0_6_i2c_control_init, .i2c_fini = smu_v13_0_6_i2c_control_fini, .send_hbm_bad_pages_num = smu_v13_0_6_smu_send_hbm_bad_page_num, - .get_ecc_info = smu_v13_0_6_get_ecc_info, }; void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu) @@ -2969,4 +3106,5 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu) smu->smc_driver_if_version = SMU13_0_6_DRIVER_IF_VERSION; smu_v13_0_set_smu_mailbox_registers(smu); amdgpu_mca_smu_init_funcs(smu->adev, &smu_v13_0_6_mca_smu_funcs); + amdgpu_aca_set_smu_funcs(smu->adev, &smu_v13_0_6_aca_smu_funcs); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c index 4894f7ee737b..2aa7e9945a0b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c @@ -892,7 +892,7 @@ int smu_v14_0_register_irq_handler(struct smu_context *smu) // TODO: THM related ret = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_MP1, - 0xfe, + SMU_IH_INTERRUPT_ID_TO_DRIVER, irq_src); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 00cd615bbcdc..b8dbd4e25348 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -378,8 +378,15 @@ int smu_cmn_send_smc_msg_with_param(struct smu_context *smu, res = __smu_cmn_reg2errno(smu, reg); if (res != 0) __smu_cmn_reg_print_error(smu, reg, index, param, msg); - if (read_arg) + if (read_arg) { smu_cmn_read_arg(smu, read_arg); + dev_dbg(adev->dev, "smu send message: %s(%d) param: 0x%08x, resp: 0x%08x,\ + readval: 0x%08x\n", + smu_get_message_name(smu, msg), index, param, reg, *read_arg); + } else { + dev_dbg(adev->dev, "smu send message: %s(%d) param: 0x%08x, resp: 0x%08x\n", + smu_get_message_name(smu, msg), index, param, reg); + } Out: if (unlikely(adev->pm.smu_debug_mask & SMU_DEBUG_HALT_ON_ERROR) && res) { amdgpu_device_halt(adev); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h index cc590e27d88a..81bfce1406e5 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h @@ -30,6 +30,16 @@ #define FDO_PWM_MODE_STATIC 1 #define FDO_PWM_MODE_STATIC_RPM 5 +#define SMU_IH_INTERRUPT_ID_TO_DRIVER 0xFE +#define SMU_IH_INTERRUPT_CONTEXT_ID_BACO 0x2 +#define SMU_IH_INTERRUPT_CONTEXT_ID_AC 0x3 +#define SMU_IH_INTERRUPT_CONTEXT_ID_DC 0x4 +#define SMU_IH_INTERRUPT_CONTEXT_ID_AUDIO_D0 0x5 +#define SMU_IH_INTERRUPT_CONTEXT_ID_AUDIO_D3 0x6 +#define SMU_IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING 0x7 +#define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL 0x8 +#define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY 0x9 + extern const int link_speed[]; /* Helper to Convert from PCIE Gen 1/2/3/4/5/6 to 0.1 GT/s speed units */ diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c index 91f2bc405cba..0279c8aabdd1 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.c +++ b/drivers/gpu/drm/i915/display/i9xx_plane.c @@ -1060,3 +1060,33 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc, plane_config->fb = intel_fb; } + +bool i9xx_fixup_initial_plane_config(struct intel_crtc *crtc, + const struct intel_initial_plane_config *plane_config) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + struct intel_plane *plane = to_intel_plane(crtc->base.primary); + const struct intel_plane_state *plane_state = + to_intel_plane_state(plane->base.state); + enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; + u32 base; + + if (!plane_state->uapi.visible) + return false; + + base = intel_plane_ggtt_offset(plane_state); + + /* + * We may have moved the surface to a different + * part of ggtt, make the plane aware of that. + */ + if (plane_config->base == base) + return false; + + if (DISPLAY_VER(dev_priv) >= 4) + intel_de_write(dev_priv, DSPSURF(i9xx_plane), base); + else + intel_de_write(dev_priv, DSPADDR(i9xx_plane), base); + + return true; +} diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.h b/drivers/gpu/drm/i915/display/i9xx_plane.h index b3d724a144cb..0ca12d1e6839 100644 --- a/drivers/gpu/drm/i915/display/i9xx_plane.h +++ b/drivers/gpu/drm/i915/display/i9xx_plane.h @@ -26,6 +26,8 @@ intel_primary_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe); void i9xx_get_initial_plane_config(struct intel_crtc *crtc, struct intel_initial_plane_config *plane_config); +bool i9xx_fixup_initial_plane_config(struct intel_crtc *crtc, + const struct intel_initial_plane_config *plane_config); #else static inline unsigned int i965_plane_max_stride(struct intel_plane *plane, u32 pixel_format, u64 modifier, @@ -46,6 +48,11 @@ static inline void i9xx_get_initial_plane_config(struct intel_crtc *crtc, struct intel_initial_plane_config *plane_config) { } +static inline bool i9xx_fixup_initial_plane_config(struct intel_crtc *crtc, + const struct intel_initial_plane_config *plane_config) +{ + return false; +} #endif #endif diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c index 06c2455bdd78..76d77d5a0409 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c @@ -217,6 +217,9 @@ intel_plane_relative_data_rate(const struct intel_crtc_state *crtc_state, int width, height; unsigned int rel_data_rate; + if (plane->id == PLANE_CURSOR) + return 0; + if (!plane_state->uapi.visible) return 0; @@ -244,9 +247,6 @@ intel_plane_relative_data_rate(const struct intel_crtc_state *crtc_state, rel_data_rate = width * height * fb->format->cpp[color_plane]; - if (plane->id == PLANE_CURSOR) - return rel_data_rate; - return intel_adjusted_rate(&plane_state->uapi.src, &plane_state->uapi.dst, rel_data_rate); diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c index 3f3cd944a1c5..1946d7fb3c2e 100644 --- a/drivers/gpu/drm/i915/display/intel_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_backlight.c @@ -1465,7 +1465,7 @@ static bool cnp_backlight_controller_is_valid(struct drm_i915_private *i915, int if (controller == 1 && INTEL_PCH_TYPE(i915) >= PCH_ICP && - INTEL_PCH_TYPE(i915) < PCH_MTP) + INTEL_PCH_TYPE(i915) <= PCH_ADP) return intel_de_read(i915, SOUTH_CHICKEN1) & ICP_SECOND_PPS_IO_SELECT; return true; diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c index aa169b0055e9..5f04e495fd27 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.c +++ b/drivers/gpu/drm/i915/display/intel_bios.c @@ -2204,8 +2204,7 @@ static u8 map_ddc_pin(struct drm_i915_private *i915, u8 vbt_pin) if (IS_DGFX(i915)) return vbt_pin; - if (INTEL_PCH_TYPE(i915) >= PCH_LNL || HAS_PCH_MTP(i915) || - IS_ALDERLAKE_P(i915)) { + if (INTEL_PCH_TYPE(i915) >= PCH_MTL || IS_ALDERLAKE_P(i915)) { ddc_pin_map = adlp_ddc_pin_map; n_entries = ARRAY_SIZE(adlp_ddc_pin_map); } else if (IS_ALDERLAKE_S(i915)) { @@ -3074,7 +3073,7 @@ err_unmap_oprom: */ void intel_bios_init(struct drm_i915_private *i915) { - const struct vbt_header *vbt = i915->display.opregion.vbt; + const struct vbt_header *vbt; struct vbt_header *oprom_vbt = NULL; const struct bdb_header *bdb; @@ -3089,6 +3088,8 @@ void intel_bios_init(struct drm_i915_private *i915) init_vbt_defaults(i915); + vbt = intel_opregion_get_vbt(i915, NULL); + /* * If the OpRegion does not have VBT, look in SPI flash through MMIO or * PCI mapping @@ -3306,7 +3307,7 @@ bool intel_bios_is_lvds_present(struct drm_i915_private *i915, u8 *i2c_pin) * additional data. Trust that if the VBT was written into * the OpRegion then they have validated the LVDS's existence. */ - if (i915->display.opregion.vbt) + if (intel_opregion_get_vbt(i915, NULL)) return true; } @@ -3657,3 +3658,30 @@ void intel_bios_for_each_encoder(struct drm_i915_private *i915, list_for_each_entry(devdata, &i915->display.vbt.display_devices, node) func(i915, devdata); } + +static int intel_bios_vbt_show(struct seq_file *m, void *unused) +{ + struct drm_i915_private *i915 = m->private; + const void *vbt; + size_t vbt_size; + + /* + * FIXME: VBT might originate from other places than opregion, and then + * this would be incorrect. + */ + vbt = intel_opregion_get_vbt(i915, &vbt_size); + if (vbt) + seq_write(m, vbt, vbt_size); + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(intel_bios_vbt); + +void intel_bios_debugfs_register(struct drm_i915_private *i915) +{ + struct drm_minor *minor = i915->drm.primary; + + debugfs_create_file("i915_vbt", 0444, minor->debugfs_root, + i915, &intel_bios_vbt_fops); +} diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h index 49e24b7cf675..41bfb009d4b0 100644 --- a/drivers/gpu/drm/i915/display/intel_bios.h +++ b/drivers/gpu/drm/i915/display/intel_bios.h @@ -246,13 +246,10 @@ bool intel_bios_is_valid_vbt(const void *buf, size_t size); bool intel_bios_is_tv_present(struct drm_i915_private *dev_priv); bool intel_bios_is_lvds_present(struct drm_i915_private *dev_priv, u8 *i2c_pin); bool intel_bios_is_port_present(struct drm_i915_private *dev_priv, enum port port); -bool intel_bios_is_port_edp(struct drm_i915_private *dev_priv, enum port port); bool intel_bios_is_dsi_present(struct drm_i915_private *dev_priv, enum port *port); bool intel_bios_get_dsc_params(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, int dsc_max_bpc); -bool intel_bios_port_supports_typec_usb(struct drm_i915_private *i915, enum port port); -bool intel_bios_port_supports_tbt(struct drm_i915_private *i915, enum port port); const struct intel_bios_encoder_data * intel_bios_encoder_data_lookup(struct drm_i915_private *i915, enum port port); @@ -283,4 +280,6 @@ 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)); +void intel_bios_debugfs_register(struct drm_i915_private *i915); + #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 c985ebb6831a..26200ee3e23f 100644 --- a/drivers/gpu/drm/i915/display/intel_cdclk.c +++ b/drivers/gpu/drm/i915/display/intel_cdclk.c @@ -1227,183 +1227,182 @@ struct intel_cdclk_vals { u32 cdclk; u16 refclk; u16 waveform; - u8 divider; /* CD2X divider * 2 */ u8 ratio; }; static const struct intel_cdclk_vals bxt_cdclk_table[] = { - { .refclk = 19200, .cdclk = 144000, .divider = 8, .ratio = 60 }, - { .refclk = 19200, .cdclk = 288000, .divider = 4, .ratio = 60 }, - { .refclk = 19200, .cdclk = 384000, .divider = 3, .ratio = 60 }, - { .refclk = 19200, .cdclk = 576000, .divider = 2, .ratio = 60 }, - { .refclk = 19200, .cdclk = 624000, .divider = 2, .ratio = 65 }, + { .refclk = 19200, .cdclk = 144000, .ratio = 60 }, + { .refclk = 19200, .cdclk = 288000, .ratio = 60 }, + { .refclk = 19200, .cdclk = 384000, .ratio = 60 }, + { .refclk = 19200, .cdclk = 576000, .ratio = 60 }, + { .refclk = 19200, .cdclk = 624000, .ratio = 65 }, {} }; static const struct intel_cdclk_vals glk_cdclk_table[] = { - { .refclk = 19200, .cdclk = 79200, .divider = 8, .ratio = 33 }, - { .refclk = 19200, .cdclk = 158400, .divider = 4, .ratio = 33 }, - { .refclk = 19200, .cdclk = 316800, .divider = 2, .ratio = 33 }, + { .refclk = 19200, .cdclk = 79200, .ratio = 33 }, + { .refclk = 19200, .cdclk = 158400, .ratio = 33 }, + { .refclk = 19200, .cdclk = 316800, .ratio = 33 }, {} }; static const struct intel_cdclk_vals icl_cdclk_table[] = { - { .refclk = 19200, .cdclk = 172800, .divider = 2, .ratio = 18 }, - { .refclk = 19200, .cdclk = 192000, .divider = 2, .ratio = 20 }, - { .refclk = 19200, .cdclk = 307200, .divider = 2, .ratio = 32 }, - { .refclk = 19200, .cdclk = 326400, .divider = 4, .ratio = 68 }, - { .refclk = 19200, .cdclk = 556800, .divider = 2, .ratio = 58 }, - { .refclk = 19200, .cdclk = 652800, .divider = 2, .ratio = 68 }, - - { .refclk = 24000, .cdclk = 180000, .divider = 2, .ratio = 15 }, - { .refclk = 24000, .cdclk = 192000, .divider = 2, .ratio = 16 }, - { .refclk = 24000, .cdclk = 312000, .divider = 2, .ratio = 26 }, - { .refclk = 24000, .cdclk = 324000, .divider = 4, .ratio = 54 }, - { .refclk = 24000, .cdclk = 552000, .divider = 2, .ratio = 46 }, - { .refclk = 24000, .cdclk = 648000, .divider = 2, .ratio = 54 }, - - { .refclk = 38400, .cdclk = 172800, .divider = 2, .ratio = 9 }, - { .refclk = 38400, .cdclk = 192000, .divider = 2, .ratio = 10 }, - { .refclk = 38400, .cdclk = 307200, .divider = 2, .ratio = 16 }, - { .refclk = 38400, .cdclk = 326400, .divider = 4, .ratio = 34 }, - { .refclk = 38400, .cdclk = 556800, .divider = 2, .ratio = 29 }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34 }, + { .refclk = 19200, .cdclk = 172800, .ratio = 18 }, + { .refclk = 19200, .cdclk = 192000, .ratio = 20 }, + { .refclk = 19200, .cdclk = 307200, .ratio = 32 }, + { .refclk = 19200, .cdclk = 326400, .ratio = 68 }, + { .refclk = 19200, .cdclk = 556800, .ratio = 58 }, + { .refclk = 19200, .cdclk = 652800, .ratio = 68 }, + + { .refclk = 24000, .cdclk = 180000, .ratio = 15 }, + { .refclk = 24000, .cdclk = 192000, .ratio = 16 }, + { .refclk = 24000, .cdclk = 312000, .ratio = 26 }, + { .refclk = 24000, .cdclk = 324000, .ratio = 54 }, + { .refclk = 24000, .cdclk = 552000, .ratio = 46 }, + { .refclk = 24000, .cdclk = 648000, .ratio = 54 }, + + { .refclk = 38400, .cdclk = 172800, .ratio = 9 }, + { .refclk = 38400, .cdclk = 192000, .ratio = 10 }, + { .refclk = 38400, .cdclk = 307200, .ratio = 16 }, + { .refclk = 38400, .cdclk = 326400, .ratio = 34 }, + { .refclk = 38400, .cdclk = 556800, .ratio = 29 }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34 }, {} }; static const struct intel_cdclk_vals rkl_cdclk_table[] = { - { .refclk = 19200, .cdclk = 172800, .divider = 4, .ratio = 36 }, - { .refclk = 19200, .cdclk = 192000, .divider = 4, .ratio = 40 }, - { .refclk = 19200, .cdclk = 307200, .divider = 4, .ratio = 64 }, - { .refclk = 19200, .cdclk = 326400, .divider = 8, .ratio = 136 }, - { .refclk = 19200, .cdclk = 556800, .divider = 4, .ratio = 116 }, - { .refclk = 19200, .cdclk = 652800, .divider = 4, .ratio = 136 }, - - { .refclk = 24000, .cdclk = 180000, .divider = 4, .ratio = 30 }, - { .refclk = 24000, .cdclk = 192000, .divider = 4, .ratio = 32 }, - { .refclk = 24000, .cdclk = 312000, .divider = 4, .ratio = 52 }, - { .refclk = 24000, .cdclk = 324000, .divider = 8, .ratio = 108 }, - { .refclk = 24000, .cdclk = 552000, .divider = 4, .ratio = 92 }, - { .refclk = 24000, .cdclk = 648000, .divider = 4, .ratio = 108 }, - - { .refclk = 38400, .cdclk = 172800, .divider = 4, .ratio = 18 }, - { .refclk = 38400, .cdclk = 192000, .divider = 4, .ratio = 20 }, - { .refclk = 38400, .cdclk = 307200, .divider = 4, .ratio = 32 }, - { .refclk = 38400, .cdclk = 326400, .divider = 8, .ratio = 68 }, - { .refclk = 38400, .cdclk = 556800, .divider = 4, .ratio = 58 }, - { .refclk = 38400, .cdclk = 652800, .divider = 4, .ratio = 68 }, + { .refclk = 19200, .cdclk = 172800, .ratio = 36 }, + { .refclk = 19200, .cdclk = 192000, .ratio = 40 }, + { .refclk = 19200, .cdclk = 307200, .ratio = 64 }, + { .refclk = 19200, .cdclk = 326400, .ratio = 136 }, + { .refclk = 19200, .cdclk = 556800, .ratio = 116 }, + { .refclk = 19200, .cdclk = 652800, .ratio = 136 }, + + { .refclk = 24000, .cdclk = 180000, .ratio = 30 }, + { .refclk = 24000, .cdclk = 192000, .ratio = 32 }, + { .refclk = 24000, .cdclk = 312000, .ratio = 52 }, + { .refclk = 24000, .cdclk = 324000, .ratio = 108 }, + { .refclk = 24000, .cdclk = 552000, .ratio = 92 }, + { .refclk = 24000, .cdclk = 648000, .ratio = 108 }, + + { .refclk = 38400, .cdclk = 172800, .ratio = 18 }, + { .refclk = 38400, .cdclk = 192000, .ratio = 20 }, + { .refclk = 38400, .cdclk = 307200, .ratio = 32 }, + { .refclk = 38400, .cdclk = 326400, .ratio = 68 }, + { .refclk = 38400, .cdclk = 556800, .ratio = 58 }, + { .refclk = 38400, .cdclk = 652800, .ratio = 68 }, {} }; static const struct intel_cdclk_vals adlp_a_step_cdclk_table[] = { - { .refclk = 19200, .cdclk = 307200, .divider = 2, .ratio = 32 }, - { .refclk = 19200, .cdclk = 556800, .divider = 2, .ratio = 58 }, - { .refclk = 19200, .cdclk = 652800, .divider = 2, .ratio = 68 }, + { .refclk = 19200, .cdclk = 307200, .ratio = 32 }, + { .refclk = 19200, .cdclk = 556800, .ratio = 58 }, + { .refclk = 19200, .cdclk = 652800, .ratio = 68 }, - { .refclk = 24000, .cdclk = 312000, .divider = 2, .ratio = 26 }, - { .refclk = 24000, .cdclk = 552000, .divider = 2, .ratio = 46 }, - { .refclk = 24400, .cdclk = 648000, .divider = 2, .ratio = 54 }, + { .refclk = 24000, .cdclk = 312000, .ratio = 26 }, + { .refclk = 24000, .cdclk = 552000, .ratio = 46 }, + { .refclk = 24400, .cdclk = 648000, .ratio = 54 }, - { .refclk = 38400, .cdclk = 307200, .divider = 2, .ratio = 16 }, - { .refclk = 38400, .cdclk = 556800, .divider = 2, .ratio = 29 }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34 }, + { .refclk = 38400, .cdclk = 307200, .ratio = 16 }, + { .refclk = 38400, .cdclk = 556800, .ratio = 29 }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34 }, {} }; static const struct intel_cdclk_vals adlp_cdclk_table[] = { - { .refclk = 19200, .cdclk = 172800, .divider = 3, .ratio = 27 }, - { .refclk = 19200, .cdclk = 192000, .divider = 2, .ratio = 20 }, - { .refclk = 19200, .cdclk = 307200, .divider = 2, .ratio = 32 }, - { .refclk = 19200, .cdclk = 556800, .divider = 2, .ratio = 58 }, - { .refclk = 19200, .cdclk = 652800, .divider = 2, .ratio = 68 }, - - { .refclk = 24000, .cdclk = 176000, .divider = 3, .ratio = 22 }, - { .refclk = 24000, .cdclk = 192000, .divider = 2, .ratio = 16 }, - { .refclk = 24000, .cdclk = 312000, .divider = 2, .ratio = 26 }, - { .refclk = 24000, .cdclk = 552000, .divider = 2, .ratio = 46 }, - { .refclk = 24000, .cdclk = 648000, .divider = 2, .ratio = 54 }, - - { .refclk = 38400, .cdclk = 179200, .divider = 3, .ratio = 14 }, - { .refclk = 38400, .cdclk = 192000, .divider = 2, .ratio = 10 }, - { .refclk = 38400, .cdclk = 307200, .divider = 2, .ratio = 16 }, - { .refclk = 38400, .cdclk = 556800, .divider = 2, .ratio = 29 }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34 }, + { .refclk = 19200, .cdclk = 172800, .ratio = 27 }, + { .refclk = 19200, .cdclk = 192000, .ratio = 20 }, + { .refclk = 19200, .cdclk = 307200, .ratio = 32 }, + { .refclk = 19200, .cdclk = 556800, .ratio = 58 }, + { .refclk = 19200, .cdclk = 652800, .ratio = 68 }, + + { .refclk = 24000, .cdclk = 176000, .ratio = 22 }, + { .refclk = 24000, .cdclk = 192000, .ratio = 16 }, + { .refclk = 24000, .cdclk = 312000, .ratio = 26 }, + { .refclk = 24000, .cdclk = 552000, .ratio = 46 }, + { .refclk = 24000, .cdclk = 648000, .ratio = 54 }, + + { .refclk = 38400, .cdclk = 179200, .ratio = 14 }, + { .refclk = 38400, .cdclk = 192000, .ratio = 10 }, + { .refclk = 38400, .cdclk = 307200, .ratio = 16 }, + { .refclk = 38400, .cdclk = 556800, .ratio = 29 }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34 }, {} }; static const struct intel_cdclk_vals rplu_cdclk_table[] = { - { .refclk = 19200, .cdclk = 172800, .divider = 3, .ratio = 27 }, - { .refclk = 19200, .cdclk = 192000, .divider = 2, .ratio = 20 }, - { .refclk = 19200, .cdclk = 307200, .divider = 2, .ratio = 32 }, - { .refclk = 19200, .cdclk = 480000, .divider = 2, .ratio = 50 }, - { .refclk = 19200, .cdclk = 556800, .divider = 2, .ratio = 58 }, - { .refclk = 19200, .cdclk = 652800, .divider = 2, .ratio = 68 }, - - { .refclk = 24000, .cdclk = 176000, .divider = 3, .ratio = 22 }, - { .refclk = 24000, .cdclk = 192000, .divider = 2, .ratio = 16 }, - { .refclk = 24000, .cdclk = 312000, .divider = 2, .ratio = 26 }, - { .refclk = 24000, .cdclk = 480000, .divider = 2, .ratio = 40 }, - { .refclk = 24000, .cdclk = 552000, .divider = 2, .ratio = 46 }, - { .refclk = 24000, .cdclk = 648000, .divider = 2, .ratio = 54 }, - - { .refclk = 38400, .cdclk = 179200, .divider = 3, .ratio = 14 }, - { .refclk = 38400, .cdclk = 192000, .divider = 2, .ratio = 10 }, - { .refclk = 38400, .cdclk = 307200, .divider = 2, .ratio = 16 }, - { .refclk = 38400, .cdclk = 480000, .divider = 2, .ratio = 25 }, - { .refclk = 38400, .cdclk = 556800, .divider = 2, .ratio = 29 }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34 }, + { .refclk = 19200, .cdclk = 172800, .ratio = 27 }, + { .refclk = 19200, .cdclk = 192000, .ratio = 20 }, + { .refclk = 19200, .cdclk = 307200, .ratio = 32 }, + { .refclk = 19200, .cdclk = 480000, .ratio = 50 }, + { .refclk = 19200, .cdclk = 556800, .ratio = 58 }, + { .refclk = 19200, .cdclk = 652800, .ratio = 68 }, + + { .refclk = 24000, .cdclk = 176000, .ratio = 22 }, + { .refclk = 24000, .cdclk = 192000, .ratio = 16 }, + { .refclk = 24000, .cdclk = 312000, .ratio = 26 }, + { .refclk = 24000, .cdclk = 480000, .ratio = 40 }, + { .refclk = 24000, .cdclk = 552000, .ratio = 46 }, + { .refclk = 24000, .cdclk = 648000, .ratio = 54 }, + + { .refclk = 38400, .cdclk = 179200, .ratio = 14 }, + { .refclk = 38400, .cdclk = 192000, .ratio = 10 }, + { .refclk = 38400, .cdclk = 307200, .ratio = 16 }, + { .refclk = 38400, .cdclk = 480000, .ratio = 25 }, + { .refclk = 38400, .cdclk = 556800, .ratio = 29 }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34 }, {} }; static const struct intel_cdclk_vals dg2_cdclk_table[] = { - { .refclk = 38400, .cdclk = 163200, .divider = 2, .ratio = 34, .waveform = 0x8888 }, - { .refclk = 38400, .cdclk = 204000, .divider = 2, .ratio = 34, .waveform = 0x9248 }, - { .refclk = 38400, .cdclk = 244800, .divider = 2, .ratio = 34, .waveform = 0xa4a4 }, - { .refclk = 38400, .cdclk = 285600, .divider = 2, .ratio = 34, .waveform = 0xa54a }, - { .refclk = 38400, .cdclk = 326400, .divider = 2, .ratio = 34, .waveform = 0xaaaa }, - { .refclk = 38400, .cdclk = 367200, .divider = 2, .ratio = 34, .waveform = 0xad5a }, - { .refclk = 38400, .cdclk = 408000, .divider = 2, .ratio = 34, .waveform = 0xb6b6 }, - { .refclk = 38400, .cdclk = 448800, .divider = 2, .ratio = 34, .waveform = 0xdbb6 }, - { .refclk = 38400, .cdclk = 489600, .divider = 2, .ratio = 34, .waveform = 0xeeee }, - { .refclk = 38400, .cdclk = 530400, .divider = 2, .ratio = 34, .waveform = 0xf7de }, - { .refclk = 38400, .cdclk = 571200, .divider = 2, .ratio = 34, .waveform = 0xfefe }, - { .refclk = 38400, .cdclk = 612000, .divider = 2, .ratio = 34, .waveform = 0xfffe }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34, .waveform = 0xffff }, + { .refclk = 38400, .cdclk = 163200, .ratio = 34, .waveform = 0x8888 }, + { .refclk = 38400, .cdclk = 204000, .ratio = 34, .waveform = 0x9248 }, + { .refclk = 38400, .cdclk = 244800, .ratio = 34, .waveform = 0xa4a4 }, + { .refclk = 38400, .cdclk = 285600, .ratio = 34, .waveform = 0xa54a }, + { .refclk = 38400, .cdclk = 326400, .ratio = 34, .waveform = 0xaaaa }, + { .refclk = 38400, .cdclk = 367200, .ratio = 34, .waveform = 0xad5a }, + { .refclk = 38400, .cdclk = 408000, .ratio = 34, .waveform = 0xb6b6 }, + { .refclk = 38400, .cdclk = 448800, .ratio = 34, .waveform = 0xdbb6 }, + { .refclk = 38400, .cdclk = 489600, .ratio = 34, .waveform = 0xeeee }, + { .refclk = 38400, .cdclk = 530400, .ratio = 34, .waveform = 0xf7de }, + { .refclk = 38400, .cdclk = 571200, .ratio = 34, .waveform = 0xfefe }, + { .refclk = 38400, .cdclk = 612000, .ratio = 34, .waveform = 0xfffe }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34, .waveform = 0xffff }, {} }; static const struct intel_cdclk_vals mtl_cdclk_table[] = { - { .refclk = 38400, .cdclk = 172800, .divider = 2, .ratio = 16, .waveform = 0xad5a }, - { .refclk = 38400, .cdclk = 192000, .divider = 2, .ratio = 16, .waveform = 0xb6b6 }, - { .refclk = 38400, .cdclk = 307200, .divider = 2, .ratio = 16, .waveform = 0x0000 }, - { .refclk = 38400, .cdclk = 480000, .divider = 2, .ratio = 25, .waveform = 0x0000 }, - { .refclk = 38400, .cdclk = 556800, .divider = 2, .ratio = 29, .waveform = 0x0000 }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34, .waveform = 0x0000 }, + { .refclk = 38400, .cdclk = 172800, .ratio = 16, .waveform = 0xad5a }, + { .refclk = 38400, .cdclk = 192000, .ratio = 16, .waveform = 0xb6b6 }, + { .refclk = 38400, .cdclk = 307200, .ratio = 16, .waveform = 0x0000 }, + { .refclk = 38400, .cdclk = 480000, .ratio = 25, .waveform = 0x0000 }, + { .refclk = 38400, .cdclk = 556800, .ratio = 29, .waveform = 0x0000 }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34, .waveform = 0x0000 }, {} }; static const struct intel_cdclk_vals lnl_cdclk_table[] = { - { .refclk = 38400, .cdclk = 153600, .divider = 2, .ratio = 16, .waveform = 0xaaaa }, - { .refclk = 38400, .cdclk = 172800, .divider = 2, .ratio = 16, .waveform = 0xad5a }, - { .refclk = 38400, .cdclk = 192000, .divider = 2, .ratio = 16, .waveform = 0xb6b6 }, - { .refclk = 38400, .cdclk = 211200, .divider = 2, .ratio = 16, .waveform = 0xdbb6 }, - { .refclk = 38400, .cdclk = 230400, .divider = 2, .ratio = 16, .waveform = 0xeeee }, - { .refclk = 38400, .cdclk = 249600, .divider = 2, .ratio = 16, .waveform = 0xf7de }, - { .refclk = 38400, .cdclk = 268800, .divider = 2, .ratio = 16, .waveform = 0xfefe }, - { .refclk = 38400, .cdclk = 288000, .divider = 2, .ratio = 16, .waveform = 0xfffe }, - { .refclk = 38400, .cdclk = 307200, .divider = 2, .ratio = 16, .waveform = 0xffff }, - { .refclk = 38400, .cdclk = 330000, .divider = 2, .ratio = 25, .waveform = 0xdbb6 }, - { .refclk = 38400, .cdclk = 360000, .divider = 2, .ratio = 25, .waveform = 0xeeee }, - { .refclk = 38400, .cdclk = 390000, .divider = 2, .ratio = 25, .waveform = 0xf7de }, - { .refclk = 38400, .cdclk = 420000, .divider = 2, .ratio = 25, .waveform = 0xfefe }, - { .refclk = 38400, .cdclk = 450000, .divider = 2, .ratio = 25, .waveform = 0xfffe }, - { .refclk = 38400, .cdclk = 480000, .divider = 2, .ratio = 25, .waveform = 0xffff }, - { .refclk = 38400, .cdclk = 487200, .divider = 2, .ratio = 29, .waveform = 0xfefe }, - { .refclk = 38400, .cdclk = 522000, .divider = 2, .ratio = 29, .waveform = 0xfffe }, - { .refclk = 38400, .cdclk = 556800, .divider = 2, .ratio = 29, .waveform = 0xffff }, - { .refclk = 38400, .cdclk = 571200, .divider = 2, .ratio = 34, .waveform = 0xfefe }, - { .refclk = 38400, .cdclk = 612000, .divider = 2, .ratio = 34, .waveform = 0xfffe }, - { .refclk = 38400, .cdclk = 652800, .divider = 2, .ratio = 34, .waveform = 0xffff }, + { .refclk = 38400, .cdclk = 153600, .ratio = 16, .waveform = 0xaaaa }, + { .refclk = 38400, .cdclk = 172800, .ratio = 16, .waveform = 0xad5a }, + { .refclk = 38400, .cdclk = 192000, .ratio = 16, .waveform = 0xb6b6 }, + { .refclk = 38400, .cdclk = 211200, .ratio = 16, .waveform = 0xdbb6 }, + { .refclk = 38400, .cdclk = 230400, .ratio = 16, .waveform = 0xeeee }, + { .refclk = 38400, .cdclk = 249600, .ratio = 16, .waveform = 0xf7de }, + { .refclk = 38400, .cdclk = 268800, .ratio = 16, .waveform = 0xfefe }, + { .refclk = 38400, .cdclk = 288000, .ratio = 16, .waveform = 0xfffe }, + { .refclk = 38400, .cdclk = 307200, .ratio = 16, .waveform = 0xffff }, + { .refclk = 38400, .cdclk = 330000, .ratio = 25, .waveform = 0xdbb6 }, + { .refclk = 38400, .cdclk = 360000, .ratio = 25, .waveform = 0xeeee }, + { .refclk = 38400, .cdclk = 390000, .ratio = 25, .waveform = 0xf7de }, + { .refclk = 38400, .cdclk = 420000, .ratio = 25, .waveform = 0xfefe }, + { .refclk = 38400, .cdclk = 450000, .ratio = 25, .waveform = 0xfffe }, + { .refclk = 38400, .cdclk = 480000, .ratio = 25, .waveform = 0xffff }, + { .refclk = 38400, .cdclk = 487200, .ratio = 29, .waveform = 0xfefe }, + { .refclk = 38400, .cdclk = 522000, .ratio = 29, .waveform = 0xfffe }, + { .refclk = 38400, .cdclk = 556800, .ratio = 29, .waveform = 0xffff }, + { .refclk = 38400, .cdclk = 571200, .ratio = 34, .waveform = 0xfefe }, + { .refclk = 38400, .cdclk = 612000, .ratio = 34, .waveform = 0xfffe }, + { .refclk = 38400, .cdclk = 652800, .ratio = 34, .waveform = 0xffff }, {} }; @@ -1901,15 +1900,47 @@ static bool pll_enable_wa_needed(struct drm_i915_private *dev_priv) dev_priv->display.cdclk.hw.vco > 0; } +static u32 bxt_cdclk_ctl(struct drm_i915_private *i915, + const struct intel_cdclk_config *cdclk_config, + enum pipe pipe) +{ + int cdclk = cdclk_config->cdclk; + int vco = cdclk_config->vco; + int unsquashed_cdclk; + u16 waveform; + u32 val; + + waveform = cdclk_squash_waveform(i915, cdclk); + + unsquashed_cdclk = DIV_ROUND_CLOSEST(cdclk * cdclk_squash_len, + cdclk_squash_divider(waveform)); + + val = bxt_cdclk_cd2x_div_sel(i915, unsquashed_cdclk, vco) | + bxt_cdclk_cd2x_pipe(i915, pipe); + + /* + * Disable SSA Precharge when CD clock frequency < 500 MHz, + * enable otherwise. + */ + if ((IS_GEMINILAKE(i915) || IS_BROXTON(i915)) && + cdclk >= 500000) + val |= BXT_CDCLK_SSA_PRECHARGE_ENABLE; + + if (DISPLAY_VER(i915) >= 20) + val |= MDCLK_SOURCE_SEL_CDCLK_PLL; + else + val |= skl_cdclk_decimal(cdclk); + + return val; +} + static void _bxt_set_cdclk(struct drm_i915_private *dev_priv, const struct intel_cdclk_config *cdclk_config, enum pipe pipe) { int cdclk = cdclk_config->cdclk; int vco = cdclk_config->vco; - int unsquashed_cdclk; u16 waveform; - u32 val; if (HAS_CDCLK_CRAWL(dev_priv) && dev_priv->display.cdclk.hw.vco > 0 && vco > 0 && !cdclk_pll_is_unknown(dev_priv->display.cdclk.hw.vco)) { @@ -1926,29 +1957,10 @@ static void _bxt_set_cdclk(struct drm_i915_private *dev_priv, waveform = cdclk_squash_waveform(dev_priv, cdclk); - unsquashed_cdclk = DIV_ROUND_CLOSEST(cdclk * cdclk_squash_len, - cdclk_squash_divider(waveform)); - if (HAS_CDCLK_SQUASH(dev_priv)) dg2_cdclk_squash_program(dev_priv, waveform); - val = bxt_cdclk_cd2x_div_sel(dev_priv, unsquashed_cdclk, vco) | - bxt_cdclk_cd2x_pipe(dev_priv, pipe); - - /* - * Disable SSA Precharge when CD clock frequency < 500 MHz, - * enable otherwise. - */ - if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) && - cdclk >= 500000) - val |= BXT_CDCLK_SSA_PRECHARGE_ENABLE; - - if (DISPLAY_VER(dev_priv) >= 20) - val |= MDCLK_SOURCE_SEL_CDCLK_PLL; - else - val |= skl_cdclk_decimal(cdclk); - - intel_de_write(dev_priv, CDCLK_CTL, val); + intel_de_write(dev_priv, CDCLK_CTL, bxt_cdclk_ctl(dev_priv, cdclk_config, pipe)); if (pipe != INVALID_PIPE) intel_crtc_wait_for_next_vblank(intel_crtc_for_pipe(dev_priv, pipe)); @@ -2039,7 +2051,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, static void bxt_sanitize_cdclk(struct drm_i915_private *dev_priv) { u32 cdctl, expected; - int cdclk, clock, vco; + int cdclk, vco; intel_update_cdclk(dev_priv); intel_cdclk_dump_config(dev_priv, &dev_priv->display.cdclk.hw, "Current CDCLK"); @@ -2048,20 +2060,6 @@ static void bxt_sanitize_cdclk(struct drm_i915_private *dev_priv) dev_priv->display.cdclk.hw.cdclk == dev_priv->display.cdclk.hw.bypass) goto sanitize; - /* DPLL okay; verify the cdclock - * - * Some BIOS versions leave an incorrect decimal frequency value and - * set reserved MBZ bits in CDCLK_CTL at least during exiting from S4, - * so sanitize this register. - */ - cdctl = intel_de_read(dev_priv, CDCLK_CTL); - /* - * Let's ignore the pipe field, since BIOS could have configured the - * dividers both synching to an active pipe, or asynchronously - * (PIPE_NONE). - */ - cdctl &= ~bxt_cdclk_cd2x_pipe(dev_priv, INVALID_PIPE); - /* Make sure this is a legal cdclk value for the platform */ cdclk = bxt_calc_cdclk(dev_priv, dev_priv->display.cdclk.hw.cdclk); if (cdclk != dev_priv->display.cdclk.hw.cdclk) @@ -2072,24 +2070,21 @@ static void bxt_sanitize_cdclk(struct drm_i915_private *dev_priv) if (vco != dev_priv->display.cdclk.hw.vco) goto sanitize; - expected = skl_cdclk_decimal(cdclk); - - /* Figure out what CD2X divider we should be using for this cdclk */ - if (HAS_CDCLK_SQUASH(dev_priv)) - clock = dev_priv->display.cdclk.hw.vco / 2; - else - clock = dev_priv->display.cdclk.hw.cdclk; - - expected |= bxt_cdclk_cd2x_div_sel(dev_priv, clock, - dev_priv->display.cdclk.hw.vco); + /* + * Some BIOS versions leave an incorrect decimal frequency value and + * set reserved MBZ bits in CDCLK_CTL at least during exiting from S4, + * so sanitize this register. + */ + cdctl = intel_de_read(dev_priv, CDCLK_CTL); + expected = bxt_cdclk_ctl(dev_priv, &dev_priv->display.cdclk.hw, INVALID_PIPE); /* - * Disable SSA Precharge when CD clock frequency < 500 MHz, - * enable otherwise. + * Let's ignore the pipe field, since BIOS could have configured the + * dividers both synching to an active pipe, or asynchronously + * (PIPE_NONE). */ - if ((IS_GEMINILAKE(dev_priv) || IS_BROXTON(dev_priv)) && - dev_priv->display.cdclk.hw.cdclk >= 500000) - expected |= BXT_CDCLK_SSA_PRECHARGE_ENABLE; + cdctl &= ~bxt_cdclk_cd2x_pipe(dev_priv, INVALID_PIPE); + expected &= ~bxt_cdclk_cd2x_pipe(dev_priv, INVALID_PIPE); if (cdctl == expected) /* All well; nothing to sanitize */ @@ -3467,15 +3462,15 @@ u32 intel_read_rawclk(struct drm_i915_private *dev_priv) { u32 freq; - if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1) - freq = dg1_rawclk(dev_priv); - else if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTP) + if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTL) /* * MTL always uses a 38.4 MHz rawclk. The bspec tells us * "RAWCLK_FREQ defaults to the values for 38.4 and does * not need to be programmed." */ freq = 38400; + else if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1) + freq = dg1_rawclk(dev_priv); else if (INTEL_PCH_TYPE(dev_priv) >= PCH_CNP) freq = cnp_rawclk(dev_priv); else if (HAS_PCH_SPLIT(dev_priv)) diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c index abaacea5c2cc..b9733a73e21d 100644 --- a/drivers/gpu/drm/i915/display/intel_crt.c +++ b/drivers/gpu/drm/i915/display/intel_crt.c @@ -42,6 +42,7 @@ #include "intel_ddi.h" #include "intel_ddi_buf_trans.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_fdi.h" #include "intel_fdi_regs.h" @@ -846,6 +847,9 @@ intel_crt_detect(struct drm_connector *connector, if (!intel_display_device_enabled(dev_priv)) return connector_status_disconnected; + if (!intel_display_driver_check_access(dev_priv)) + return connector->status; + if (dev_priv->display.params.load_detect_test) { wakeref = intel_display_power_get(dev_priv, intel_encoder->power_domain); @@ -1069,6 +1073,7 @@ void intel_crt_init(struct drm_i915_private *dev_priv) } else { intel_connector->polled = DRM_CONNECTOR_POLL_CONNECT; } + intel_connector->base.polled = intel_connector->polled; if (HAS_DDI(dev_priv)) { assert_port_valid(dev_priv, PORT_E); diff --git a/drivers/gpu/drm/i915/display/intel_crtc.c b/drivers/gpu/drm/i915/display/intel_crtc.c index 8a84a31c7b48..25593f6aae7d 100644 --- a/drivers/gpu/drm/i915/display/intel_crtc.c +++ b/drivers/gpu/drm/i915/display/intel_crtc.c @@ -461,70 +461,6 @@ int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode, 1000 * adjusted_mode->crtc_htotal); } -static int intel_mode_vblank_start(const struct drm_display_mode *mode) -{ - int vblank_start = mode->crtc_vblank_start; - - if (mode->flags & DRM_MODE_FLAG_INTERLACE) - vblank_start = DIV_ROUND_UP(vblank_start, 2); - - return vblank_start; -} - -static void intel_crtc_vblank_evade_scanlines(struct intel_atomic_state *state, - struct intel_crtc *crtc, - int *min, int *max, int *vblank_start) -{ - const struct intel_crtc_state *old_crtc_state = - intel_atomic_get_old_crtc_state(state, crtc); - const struct intel_crtc_state *new_crtc_state = - intel_atomic_get_new_crtc_state(state, crtc); - const struct intel_crtc_state *crtc_state; - const struct drm_display_mode *adjusted_mode; - - /* - * During fastsets/etc. the transcoder is still - * running with the old timings at this point. - * - * TODO: maybe just use the active timings here? - */ - if (intel_crtc_needs_modeset(new_crtc_state)) - crtc_state = new_crtc_state; - else - crtc_state = old_crtc_state; - - adjusted_mode = &crtc_state->hw.adjusted_mode; - - if (crtc->mode_flags & I915_MODE_FLAG_VRR) { - /* timing changes should happen with VRR disabled */ - drm_WARN_ON(state->base.dev, intel_crtc_needs_modeset(new_crtc_state) || - new_crtc_state->update_m_n || new_crtc_state->update_lrr); - - if (intel_vrr_is_push_sent(crtc_state)) - *vblank_start = intel_vrr_vmin_vblank_start(crtc_state); - else - *vblank_start = intel_vrr_vmax_vblank_start(crtc_state); - } else { - *vblank_start = intel_mode_vblank_start(adjusted_mode); - } - - /* FIXME needs to be calibrated sensibly */ - *min = *vblank_start - intel_usecs_to_scanlines(adjusted_mode, - VBLANK_EVASION_TIME_US); - *max = *vblank_start - 1; - - /* - * M/N and TRANS_VTOTAL are double buffered on the transcoder's - * undelayed vblank, so with seamless M/N and LRR we must evade - * both vblanks. - * - * DSB execution waits for the transcoder's undelayed vblank, - * hence we must kick off the commit before that. - */ - if (new_crtc_state->dsb || new_crtc_state->update_m_n || new_crtc_state->update_lrr) - *min -= adjusted_mode->crtc_vblank_start - adjusted_mode->crtc_vdisplay; -} - /** * intel_pipe_update_start() - start update of a set of display registers * @state: the atomic state @@ -542,14 +478,12 @@ void intel_pipe_update_start(struct intel_atomic_state *state, struct intel_crtc *crtc) { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); struct intel_crtc_state *new_crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - long timeout = msecs_to_jiffies_timeout(1); - int scanline, min, max, vblank_start; - wait_queue_head_t *wq = drm_crtc_vblank_waitqueue(&crtc->base); - bool need_vlv_dsi_wa = (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) && - intel_crtc_has_type(new_crtc_state, INTEL_OUTPUT_DSI); - DEFINE_WAIT(wait); + struct intel_vblank_evade_ctx evade; + int scanline; intel_psr_lock(new_crtc_state); @@ -566,9 +500,7 @@ void intel_pipe_update_start(struct intel_atomic_state *state, if (intel_crtc_needs_vblank_work(new_crtc_state)) intel_crtc_vblank_work_init(new_crtc_state); - intel_crtc_vblank_evade_scanlines(state, crtc, &min, &max, &vblank_start); - if (min <= 0 || max <= 0) - goto irq_disable; + intel_vblank_evade_init(old_crtc_state, new_crtc_state, &evade); if (drm_WARN_ON(&dev_priv->drm, drm_crtc_vblank_get(&crtc->base))) goto irq_disable; @@ -582,58 +514,14 @@ void intel_pipe_update_start(struct intel_atomic_state *state, local_irq_disable(); - crtc->debug.min_vbl = min; - crtc->debug.max_vbl = max; + crtc->debug.min_vbl = evade.min; + crtc->debug.max_vbl = evade.max; trace_intel_pipe_update_start(crtc); - for (;;) { - /* - * prepare_to_wait() has a memory barrier, which guarantees - * other CPUs can see the task state update by the time we - * read the scanline. - */ - prepare_to_wait(wq, &wait, TASK_UNINTERRUPTIBLE); - - scanline = intel_get_crtc_scanline(crtc); - if (scanline < min || scanline > max) - break; - - if (!timeout) { - drm_err(&dev_priv->drm, - "Potential atomic update failure on pipe %c\n", - pipe_name(crtc->pipe)); - break; - } - - local_irq_enable(); - - timeout = schedule_timeout(timeout); - - local_irq_disable(); - } - - finish_wait(wq, &wait); + scanline = intel_vblank_evade(&evade); drm_crtc_vblank_put(&crtc->base); - /* - * On VLV/CHV DSI the scanline counter would appear to - * increment approx. 1/3 of a scanline before start of vblank. - * The registers still get latched at start of vblank however. - * This means we must not write any registers on the first - * line of vblank (since not the whole line is actually in - * vblank). And unfortunately we can't use the interrupt to - * wait here since it will fire too soon. We could use the - * frame start interrupt instead since it will fire after the - * critical scanline, but that would require more changes - * in the interrupt code. So for now we'll just do the nasty - * thing and poll for the bad scanline to pass us by. - * - * FIXME figure out if BXT+ DSI suffers from this as well - */ - while (need_vlv_dsi_wa && scanline == vblank_start) - scanline = intel_get_crtc_scanline(crtc); - crtc->debug.scanline_start = scanline; crtc->debug.start_vbl_time = ktime_get(); crtc->debug.start_vbl_count = intel_crtc_get_vblank_counter(crtc); diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c index 926e2de00eb5..f8b33999d43f 100644 --- a/drivers/gpu/drm/i915/display/intel_cursor.c +++ b/drivers/gpu/drm/i915/display/intel_cursor.c @@ -22,6 +22,7 @@ #include "intel_frontbuffer.h" #include "intel_psr.h" #include "intel_psr_regs.h" +#include "intel_vblank.h" #include "skl_watermark.h" #include "gem/i915_gem_object.h" @@ -47,12 +48,23 @@ static u32 intel_cursor_base(const struct intel_plane_state *plane_state) return base + plane_state->view.color_plane[0].offset; } -static u32 intel_cursor_position(const struct intel_plane_state *plane_state) +static u32 intel_cursor_position(const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state, + bool early_tpt) { int x = plane_state->uapi.dst.x1; int y = plane_state->uapi.dst.y1; u32 pos = 0; + /* + * Formula from Bspec: + * MAX(-1 * <Cursor vertical size from CUR_CTL base on cursor mode + * select setting> + 1, CUR_POS Y Position - Update region Y position + */ + if (early_tpt) + y = max(-1 * drm_rect_height(&plane_state->uapi.dst) + 1, + y - crtc_state->psr2_su_area.y1); + if (x < 0) { pos |= CURSOR_POS_X_SIGN; x = -x; @@ -274,7 +286,7 @@ static void i845_cursor_update_arm(struct intel_plane *plane, size = CURSOR_HEIGHT(height) | CURSOR_WIDTH(width); base = intel_cursor_base(plane_state); - pos = intel_cursor_position(plane_state); + pos = intel_cursor_position(crtc_state, plane_state, false); } /* On these chipsets we can only modify the base/size/stride @@ -503,17 +515,24 @@ static void i9xx_cursor_update_sel_fetch_arm(struct intel_plane *plane, const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state) { - struct drm_i915_private *i915 = to_i915(plane->base.dev); + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum pipe pipe = plane->pipe; if (!crtc_state->enable_psr2_sel_fetch) return; - if (drm_rect_height(&plane_state->psr2_sel_fetch_area) > 0) - intel_de_write_fw(i915, PLANE_SEL_FETCH_CTL(pipe, plane->id), + if (drm_rect_height(&plane_state->psr2_sel_fetch_area) > 0) { + if (crtc_state->enable_psr2_su_region_et) { + u32 val = intel_cursor_position(crtc_state, plane_state, + true); + intel_de_write_fw(dev_priv, CURPOS_ERLY_TPT(pipe), val); + } + + intel_de_write_fw(dev_priv, PLANE_SEL_FETCH_CTL(pipe, plane->id), plane_state->ctl); - else + } else { i9xx_cursor_disable_sel_fetch_arm(plane, crtc_state); + } } /* TODO: split into noarm+arm pair */ @@ -536,7 +555,7 @@ static void i9xx_cursor_update_arm(struct intel_plane *plane, fbc_ctl = CUR_FBC_EN | CUR_FBC_HEIGHT(height - 1); base = intel_cursor_base(plane_state); - pos = intel_cursor_position(plane_state); + pos = intel_cursor_position(crtc_state, plane_state, false); } /* @@ -647,12 +666,14 @@ intel_legacy_cursor_update(struct drm_plane *_plane, { struct intel_plane *plane = to_intel_plane(_plane); struct intel_crtc *crtc = to_intel_crtc(_crtc); + struct drm_i915_private *i915 = to_i915(plane->base.dev); struct intel_plane_state *old_plane_state = to_intel_plane_state(plane->base.state); struct intel_plane_state *new_plane_state; struct intel_crtc_state *crtc_state = to_intel_crtc_state(crtc->base.state); struct intel_crtc_state *new_crtc_state; + struct intel_vblank_evade_ctx evade; int ret; /* @@ -745,13 +766,25 @@ intel_legacy_cursor_update(struct drm_plane *_plane, */ crtc_state->active_planes = new_crtc_state->active_planes; - /* - * Technically we should do a vblank evasion here to make - * sure all the cursor registers update on the same frame. - * For now just make sure the register writes happen as - * quickly as possible to minimize the race window. - */ - local_irq_disable(); + intel_vblank_evade_init(crtc_state, crtc_state, &evade); + + intel_psr_lock(crtc_state); + + if (!drm_WARN_ON(&i915->drm, drm_crtc_vblank_get(&crtc->base))) { + /* + * TODO: maybe check if we're still in PSR + * and skip the vblank evasion entirely? + */ + intel_psr_wait_for_idle_locked(crtc_state); + + local_irq_disable(); + + intel_vblank_evade(&evade); + + drm_crtc_vblank_put(&crtc->base); + } else { + local_irq_disable(); + } if (new_plane_state->uapi.visible) { intel_plane_update_noarm(plane, crtc_state, new_plane_state); @@ -762,6 +795,8 @@ intel_legacy_cursor_update(struct drm_plane *_plane, local_irq_enable(); + intel_psr_unlock(crtc_state); + intel_plane_unpin_fb(old_plane_state); out_free: diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c index 6b25e195232f..288a00e083c8 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c @@ -78,7 +78,7 @@ static void intel_cx0_program_msgbus_timer(struct intel_encoder *encoder) for_each_cx0_lane_in_mask(INTEL_CX0_BOTH_LANES, lane) intel_de_rmw(i915, - XELPDP_PORT_MSGBUS_TIMER(encoder->port, lane), + XELPDP_PORT_MSGBUS_TIMER(i915, encoder->port, lane), XELPDP_PORT_MSGBUS_TIMER_VAL_MASK, XELPDP_PORT_MSGBUS_TIMER_VAL); } @@ -117,7 +117,7 @@ static void intel_cx0_phy_transaction_end(struct intel_encoder *encoder, intel_w static void intel_clear_response_ready_flag(struct drm_i915_private *i915, enum port port, int lane) { - intel_de_rmw(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane), + intel_de_rmw(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(i915, port, lane), 0, XELPDP_PORT_P2M_RESPONSE_READY | XELPDP_PORT_P2M_ERROR_SET); } @@ -125,10 +125,10 @@ static void intel_cx0_bus_reset(struct drm_i915_private *i915, enum port port, i { enum phy phy = intel_port_to_phy(i915, port); - intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_RESET); - if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_RESET, XELPDP_MSGBUS_TIMEOUT_SLOW)) { drm_err_once(&i915->drm, "Failed to bring PHY %c to idle.\n", phy_name(phy)); @@ -144,7 +144,7 @@ static int intel_cx0_wait_for_ack(struct drm_i915_private *i915, enum port port, enum phy phy = intel_port_to_phy(i915, port); if (__intel_de_wait_for_register(i915, - XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane), + XELPDP_PORT_P2M_MSGBUS_STATUS(i915, port, lane), XELPDP_PORT_P2M_RESPONSE_READY, XELPDP_PORT_P2M_RESPONSE_READY, XELPDP_MSGBUS_TIMEOUT_FAST_US, @@ -152,7 +152,7 @@ static int intel_cx0_wait_for_ack(struct drm_i915_private *i915, enum port port, drm_dbg_kms(&i915->drm, "PHY %c Timeout waiting for message ACK. Status: 0x%x\n", phy_name(phy), *val); - if (!(intel_de_read(i915, XELPDP_PORT_MSGBUS_TIMER(port, lane)) & + if (!(intel_de_read(i915, XELPDP_PORT_MSGBUS_TIMER(i915, port, lane)) & XELPDP_PORT_MSGBUS_TIMER_TIMED_OUT)) drm_dbg_kms(&i915->drm, "PHY %c Hardware did not detect a timeout\n", @@ -186,7 +186,7 @@ static int __intel_cx0_read_once(struct drm_i915_private *i915, enum port port, int ack; u32 val; - if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_PENDING, XELPDP_MSGBUS_TIMEOUT_SLOW)) { drm_dbg_kms(&i915->drm, @@ -195,7 +195,7 @@ static int __intel_cx0_read_once(struct drm_i915_private *i915, enum port port, return -ETIMEDOUT; } - intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_PENDING | XELPDP_PORT_M2P_COMMAND_READ | XELPDP_PORT_M2P_ADDRESS(addr)); @@ -253,7 +253,7 @@ static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port, int ack; u32 val; - if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_PENDING, XELPDP_MSGBUS_TIMEOUT_SLOW)) { drm_dbg_kms(&i915->drm, @@ -262,14 +262,14 @@ static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port, return -ETIMEDOUT; } - intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + intel_de_write(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_PENDING | (committed ? XELPDP_PORT_M2P_COMMAND_WRITE_COMMITTED : XELPDP_PORT_M2P_COMMAND_WRITE_UNCOMMITTED) | XELPDP_PORT_M2P_DATA(data) | XELPDP_PORT_M2P_ADDRESS(addr)); - if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_PENDING, XELPDP_MSGBUS_TIMEOUT_SLOW)) { drm_dbg_kms(&i915->drm, @@ -282,7 +282,7 @@ static int __intel_cx0_write_once(struct drm_i915_private *i915, enum port port, 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)) & + } else if ((intel_de_read(i915, XELPDP_PORT_P2M_MSGBUS_STATUS(i915, port, lane)) & XELPDP_PORT_P2M_ERROR_SET)) { drm_dbg_kms(&i915->drm, "PHY %c Error occurred during write command.\n", phy_name(phy)); @@ -2096,13 +2096,54 @@ int intel_cx0pll_calc_state(struct intel_crtc_state *crtc_state, return intel_c20pll_calc_state(crtc_state, encoder); } -static bool intel_c20_use_mplla(u32 clock) +static bool intel_c20phy_use_mpllb(const struct intel_c20pll_state *state) { - /* 10G and 20G rates use MPLLA */ - if (clock == 1000000 || clock == 2000000) - return true; + return state->tx[0] & C20_PHY_USE_MPLLB; +} - return false; +static int intel_c20pll_calc_port_clock(struct intel_encoder *encoder, + const struct intel_c20pll_state *pll_state) +{ + unsigned int frac, frac_en, frac_quot, frac_rem, frac_den; + unsigned int multiplier, refclk = 38400; + unsigned int tx_clk_div; + unsigned int ref_clk_mpllb_div; + unsigned int fb_clk_div4_en; + unsigned int ref, vco; + unsigned int tx_rate_mult; + unsigned int tx_rate = REG_FIELD_GET(C20_PHY_TX_RATE, pll_state->tx[0]); + + if (intel_c20phy_use_mpllb(pll_state)) { + tx_rate_mult = 1; + frac_en = REG_FIELD_GET(C20_MPLLB_FRACEN, pll_state->mpllb[6]); + frac_quot = pll_state->mpllb[8]; + frac_rem = pll_state->mpllb[9]; + frac_den = pll_state->mpllb[7]; + multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mpllb[0]); + tx_clk_div = REG_FIELD_GET(C20_MPLLB_TX_CLK_DIV_MASK, pll_state->mpllb[0]); + ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mpllb[6]); + fb_clk_div4_en = 0; + } else { + tx_rate_mult = 2; + frac_en = REG_FIELD_GET(C20_MPLLA_FRACEN, pll_state->mplla[6]); + frac_quot = pll_state->mplla[8]; + frac_rem = pll_state->mplla[9]; + frac_den = pll_state->mplla[7]; + multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mplla[0]); + tx_clk_div = REG_FIELD_GET(C20_MPLLA_TX_CLK_DIV_MASK, pll_state->mplla[1]); + ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mplla[6]); + fb_clk_div4_en = REG_FIELD_GET(C20_FB_CLK_DIV4_EN, pll_state->mplla[0]); + } + + if (frac_en) + frac = frac_quot + DIV_ROUND_CLOSEST(frac_rem, frac_den); + else + frac = 0; + + ref = DIV_ROUND_CLOSEST(refclk * (1 << (1 + fb_clk_div4_en)), 1 << ref_clk_mpllb_div); + vco = DIV_ROUND_CLOSEST_ULL(mul_u32_u32(ref, (multiplier << (17 - 2)) + frac) >> 17, 10); + + return vco << tx_rate_mult >> tx_clk_div >> tx_rate; } static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder, @@ -2138,7 +2179,7 @@ static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder, PHY_C20_A_CMN_CNTX_CFG(i)); } - if (pll_state->tx[0] & C20_PHY_USE_MPLLB) { + if (intel_c20phy_use_mpllb(pll_state)) { /* MPLLB configuration */ for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) { if (cntx) @@ -2160,6 +2201,8 @@ static void intel_c20pll_readout_hw_state(struct intel_encoder *encoder, } } + pll_state->clock = intel_c20pll_calc_port_clock(encoder, pll_state); + intel_cx0_phy_transaction_end(encoder, wakeref); } @@ -2174,12 +2217,12 @@ void intel_c20pll_dump_hw_state(struct drm_i915_private *i915, drm_dbg_kms(&i915->drm, "cmn[0] = 0x%.4x, cmn[1] = 0x%.4x, cmn[2] = 0x%.4x, cmn[3] = 0x%.4x\n", hw_state->cmn[0], hw_state->cmn[1], hw_state->cmn[2], hw_state->cmn[3]); - if (intel_c20_use_mplla(hw_state->clock)) { - for (i = 0; i < ARRAY_SIZE(hw_state->mplla); i++) - drm_dbg_kms(&i915->drm, "mplla[%d] = 0x%.4x\n", i, hw_state->mplla[i]); - } else { + if (intel_c20phy_use_mpllb(hw_state)) { for (i = 0; i < ARRAY_SIZE(hw_state->mpllb); i++) drm_dbg_kms(&i915->drm, "mpllb[%d] = 0x%.4x\n", i, hw_state->mpllb[i]); + } else { + for (i = 0; i < ARRAY_SIZE(hw_state->mplla); i++) + drm_dbg_kms(&i915->drm, "mplla[%d] = 0x%.4x\n", i, hw_state->mplla[i]); } } @@ -2326,27 +2369,27 @@ static void intel_c20_pll_program(struct drm_i915_private *i915, } /* 3.3 mpllb or mplla configuration */ - if (intel_c20_use_mplla(clock)) { - for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) { + if (intel_c20phy_use_mpllb(pll_state)) { + for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) { if (cntx) intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, - PHY_C20_A_MPLLA_CNTX_CFG(i), - pll_state->mplla[i]); + PHY_C20_A_MPLLB_CNTX_CFG(i), + pll_state->mpllb[i]); else intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, - PHY_C20_B_MPLLA_CNTX_CFG(i), - pll_state->mplla[i]); + PHY_C20_B_MPLLB_CNTX_CFG(i), + pll_state->mpllb[i]); } } else { - for (i = 0; i < ARRAY_SIZE(pll_state->mpllb); i++) { + for (i = 0; i < ARRAY_SIZE(pll_state->mplla); i++) { if (cntx) intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, - PHY_C20_A_MPLLB_CNTX_CFG(i), - pll_state->mpllb[i]); + PHY_C20_A_MPLLA_CNTX_CFG(i), + pll_state->mplla[i]); else intel_c20_sram_write(i915, encoder->port, INTEL_CX0_LANE0, - PHY_C20_B_MPLLB_CNTX_CFG(i), - pll_state->mpllb[i]); + PHY_C20_B_MPLLA_CNTX_CFG(i), + pll_state->mplla[i]); } } @@ -2408,51 +2451,6 @@ static int intel_c10pll_calc_port_clock(struct intel_encoder *encoder, return tmpclk; } -static int intel_c20pll_calc_port_clock(struct intel_encoder *encoder, - const struct intel_c20pll_state *pll_state) -{ - unsigned int frac, frac_en, frac_quot, frac_rem, frac_den; - unsigned int multiplier, refclk = 38400; - unsigned int tx_clk_div; - unsigned int ref_clk_mpllb_div; - unsigned int fb_clk_div4_en; - unsigned int ref, vco; - unsigned int tx_rate_mult; - unsigned int tx_rate = REG_FIELD_GET(C20_PHY_TX_RATE, pll_state->tx[0]); - - if (pll_state->tx[0] & C20_PHY_USE_MPLLB) { - tx_rate_mult = 1; - frac_en = REG_FIELD_GET(C20_MPLLB_FRACEN, pll_state->mpllb[6]); - frac_quot = pll_state->mpllb[8]; - frac_rem = pll_state->mpllb[9]; - frac_den = pll_state->mpllb[7]; - multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mpllb[0]); - tx_clk_div = REG_FIELD_GET(C20_MPLLB_TX_CLK_DIV_MASK, pll_state->mpllb[0]); - ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mpllb[6]); - fb_clk_div4_en = 0; - } else { - tx_rate_mult = 2; - frac_en = REG_FIELD_GET(C20_MPLLA_FRACEN, pll_state->mplla[6]); - frac_quot = pll_state->mplla[8]; - frac_rem = pll_state->mplla[9]; - frac_den = pll_state->mplla[7]; - multiplier = REG_FIELD_GET(C20_MULTIPLIER_MASK, pll_state->mplla[0]); - tx_clk_div = REG_FIELD_GET(C20_MPLLA_TX_CLK_DIV_MASK, pll_state->mplla[1]); - ref_clk_mpllb_div = REG_FIELD_GET(C20_REF_CLK_MPLLB_DIV_MASK, pll_state->mplla[6]); - fb_clk_div4_en = REG_FIELD_GET(C20_FB_CLK_DIV4_EN, pll_state->mplla[0]); - } - - if (frac_en) - frac = frac_quot + DIV_ROUND_CLOSEST(frac_rem, frac_den); - else - frac = 0; - - ref = DIV_ROUND_CLOSEST(refclk * (1 << (1 + fb_clk_div4_en)), 1 << ref_clk_mpllb_div); - vco = DIV_ROUND_CLOSEST_ULL(mul_u32_u32(ref, (multiplier << (17 - 2)) + frac) >> 17, 10); - - return vco << tx_rate_mult >> tx_clk_div >> tx_rate; -} - static void intel_program_port_clock_ctl(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, bool lane_reversal) @@ -2460,7 +2458,8 @@ static void intel_program_port_clock_ctl(struct intel_encoder *encoder, struct drm_i915_private *i915 = to_i915(encoder->base.dev); u32 val = 0; - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(encoder->port), XELPDP_PORT_REVERSAL, + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(i915, encoder->port), + XELPDP_PORT_REVERSAL, lane_reversal ? XELPDP_PORT_REVERSAL : 0); if (lane_reversal) @@ -2481,7 +2480,7 @@ static void intel_program_port_clock_ctl(struct intel_encoder *encoder, else val |= crtc_state->cx0pll_state.ssc_enabled ? XELPDP_SSC_ENABLE_PLLB : 0; - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_LANE1_PHY_CLOCK_SELECT | XELPDP_FORWARD_CLOCK_UNGATE | XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_SSC_ENABLE_PLLA | XELPDP_SSC_ENABLE_PLLB, val); @@ -2514,15 +2513,16 @@ static void intel_cx0_powerdown_change_sequence(struct drm_i915_private *i915, u8 lane_mask, u8 state) { enum phy phy = intel_port_to_phy(i915, port); + i915_reg_t buf_ctl2_reg = XELPDP_PORT_BUF_CTL2(i915, port); int lane; - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + intel_de_rmw(i915, buf_ctl2_reg, intel_cx0_get_powerdown_state(INTEL_CX0_BOTH_LANES, XELPDP_LANE_POWERDOWN_NEW_STATE_MASK), intel_cx0_get_powerdown_state(lane_mask, state)); /* Wait for pending transactions.*/ for_each_cx0_lane_in_mask(lane_mask, lane) - if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(port, lane), + if (intel_de_wait_for_clear(i915, XELPDP_PORT_M2P_MSGBUS_CTL(i915, port, lane), XELPDP_PORT_M2P_TRANSACTION_PENDING, XELPDP_MSGBUS_TIMEOUT_SLOW)) { drm_dbg_kms(&i915->drm, @@ -2531,12 +2531,12 @@ static void intel_cx0_powerdown_change_sequence(struct drm_i915_private *i915, intel_cx0_bus_reset(i915, port, lane); } - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + intel_de_rmw(i915, buf_ctl2_reg, intel_cx0_get_powerdown_update(INTEL_CX0_BOTH_LANES), intel_cx0_get_powerdown_update(lane_mask)); /* Update Timeout Value */ - if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port), + if (__intel_de_wait_for_register(i915, buf_ctl2_reg, intel_cx0_get_powerdown_update(lane_mask), 0, XELPDP_PORT_POWERDOWN_UPDATE_TIMEOUT_US, 0, NULL)) drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n", @@ -2545,10 +2545,10 @@ static void intel_cx0_powerdown_change_sequence(struct drm_i915_private *i915, static void intel_cx0_setup_powerdown(struct drm_i915_private *i915, enum port port) { - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(i915, port), XELPDP_POWER_STATE_READY_MASK, XELPDP_POWER_STATE_READY(CX0_P2_STATE_READY)); - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL3(port), + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL3(i915, port), XELPDP_POWER_STATE_ACTIVE_MASK | XELPDP_PLL_LANE_STAGGERING_DELAY_MASK, XELPDP_POWER_STATE_ACTIVE(CX0_P0_STATE_ACTIVE) | @@ -2593,27 +2593,27 @@ static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, XELPDP_LANE_PHY_CURRENT_STATUS(1)) : XELPDP_LANE_PHY_CURRENT_STATUS(0); - if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL1(port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL1(i915, port), XELPDP_PORT_BUF_SOC_PHY_READY, XELPDP_PORT_BUF_SOC_PHY_READY, XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US, 0, NULL)) drm_warn(&i915->drm, "PHY %c failed to bring out of SOC reset after %dus.\n", phy_name(phy), XELPDP_PORT_BUF_SOC_READY_TIMEOUT_US); - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), lane_pipe_reset, + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(i915, port), lane_pipe_reset, lane_pipe_reset); - if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_BUF_CTL2(i915, port), lane_phy_current_status, lane_phy_current_status, XELPDP_PORT_RESET_START_TIMEOUT_US, 0, NULL)) drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dus.\n", phy_name(phy), XELPDP_PORT_RESET_START_TIMEOUT_US); - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, port), intel_cx0_get_pclk_refclk_request(owned_lane_mask), intel_cx0_get_pclk_refclk_request(lane_mask)); - if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(i915, port), intel_cx0_get_pclk_refclk_ack(owned_lane_mask), intel_cx0_get_pclk_refclk_ack(lane_mask), XELPDP_REFCLK_ENABLE_TIMEOUT_US, 0, NULL)) @@ -2624,9 +2624,10 @@ static void intel_cx0_phy_lane_reset(struct drm_i915_private *i915, CX0_P2_STATE_RESET); intel_cx0_setup_powerdown(i915, port); - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(port), lane_pipe_reset, 0); + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL2(i915, port), lane_pipe_reset, 0); - if (intel_de_wait_for_clear(i915, XELPDP_PORT_BUF_CTL2(port), lane_phy_current_status, + if (intel_de_wait_for_clear(i915, XELPDP_PORT_BUF_CTL2(i915, port), + lane_phy_current_status, XELPDP_PORT_RESET_END_TIMEOUT)) drm_warn(&i915->drm, "PHY %c failed to bring out of Lane reset after %dms.\n", phy_name(phy), XELPDP_PORT_RESET_END_TIMEOUT); @@ -2761,12 +2762,12 @@ static void intel_cx0pll_enable(struct intel_encoder *encoder, * 9. Set PORT_CLOCK_CTL register PCLK PLL Request * LN<Lane for maxPCLK> to "1" to enable PLL. */ - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), intel_cx0_get_pclk_pll_request(INTEL_CX0_BOTH_LANES), intel_cx0_get_pclk_pll_request(maxpclk_lane)); /* 10. Poll on PORT_CLOCK_CTL PCLK PLL Ack LN<Lane for maxPCLK> == "1". */ - if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), intel_cx0_get_pclk_pll_ack(INTEL_CX0_BOTH_LANES), intel_cx0_get_pclk_pll_ack(maxpclk_lane), XELPDP_PCLK_PLL_ENABLE_TIMEOUT_US, 0, NULL)) @@ -2786,7 +2787,7 @@ int intel_mtl_tbt_calc_port_clock(struct intel_encoder *encoder) { struct drm_i915_private *i915 = to_i915(encoder->base.dev); u32 clock; - u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port)); + u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port)); clock = REG_FIELD_GET(XELPDP_DDI_CLOCK_SELECT_MASK, val); @@ -2839,11 +2840,11 @@ static void intel_mtl_tbt_pll_enable(struct intel_encoder *encoder, */ val |= XELPDP_DDI_CLOCK_SELECT(intel_mtl_tbt_clock_select(i915, crtc_state->port_clock)); val |= XELPDP_FORWARD_CLOCK_UNGATE; - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_FORWARD_CLOCK_UNGATE, val); /* 2. Read back PORT_CLOCK_CTL REGISTER */ - val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port)); + val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port)); /* * 3. Follow the Display Voltage Frequency Switching - Sequence @@ -2854,10 +2855,10 @@ static void intel_mtl_tbt_pll_enable(struct intel_encoder *encoder, * 4. Set PORT_CLOCK_CTL register TBT CLOCK Request to "1" to enable PLL. */ val |= XELPDP_TBT_CLOCK_REQUEST; - intel_de_write(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), val); + intel_de_write(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), val); /* 5. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "1". */ - if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_TBT_CLOCK_ACK, XELPDP_TBT_CLOCK_ACK, 100, 0, NULL)) @@ -2909,7 +2910,7 @@ static void intel_cx0pll_disable(struct intel_encoder *encoder) * 3. Set PORT_CLOCK_CTL register PCLK PLL Request LN<Lane for maxPCLK> * to "0" to disable PLL. */ - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), intel_cx0_get_pclk_pll_request(INTEL_CX0_BOTH_LANES) | intel_cx0_get_pclk_refclk_request(INTEL_CX0_BOTH_LANES), 0); @@ -2919,7 +2920,7 @@ static void intel_cx0pll_disable(struct intel_encoder *encoder) /* * 5. Poll on PORT_CLOCK_CTL PCLK PLL Ack LN<Lane for maxPCLK**> == "0". */ - if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), intel_cx0_get_pclk_pll_ack(INTEL_CX0_BOTH_LANES) | intel_cx0_get_pclk_refclk_ack(INTEL_CX0_BOTH_LANES), 0, XELPDP_PCLK_PLL_DISABLE_TIMEOUT_US, 0, NULL)) @@ -2932,9 +2933,9 @@ static void intel_cx0pll_disable(struct intel_encoder *encoder) */ /* 7. Program PORT_CLOCK_CTL register to disable and gate clocks. */ - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_DDI_CLOCK_SELECT_MASK, 0); - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_FORWARD_CLOCK_UNGATE, 0); intel_cx0_phy_transaction_end(encoder, wakeref); @@ -2953,11 +2954,11 @@ static void intel_mtl_tbt_pll_disable(struct intel_encoder *encoder) /* * 2. Set PORT_CLOCK_CTL register TBT CLOCK Request to "0" to disable PLL. */ - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_TBT_CLOCK_REQUEST, 0); /* 3. Poll on PORT_CLOCK_CTL TBT CLOCK Ack == "0". */ - if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + if (__intel_de_wait_for_register(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_TBT_CLOCK_ACK, 0, 10, 0, NULL)) drm_warn(&i915->drm, "[ENCODER:%d:%s][%c] PHY PLL not unlocked after 10us.\n", encoder->base.base.id, encoder->base.name, phy_name(phy)); @@ -2970,7 +2971,7 @@ static void intel_mtl_tbt_pll_disable(struct intel_encoder *encoder) /* * 5. Program PORT CLOCK CTRL register to disable and gate clocks */ - intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port), XELPDP_DDI_CLOCK_SELECT_MASK | XELPDP_FORWARD_CLOCK_UNGATE, 0); @@ -2997,7 +2998,7 @@ intel_mtl_port_pll_type(struct intel_encoder *encoder, * TODO: Determine the PLL type from the SW state, once MTL PLL * handling is done via the standard shared DPLL framework. */ - u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(encoder->port)); + u32 val = intel_de_read(i915, XELPDP_PORT_CLOCK_CTL(i915, encoder->port)); u32 clock = REG_FIELD_GET(XELPDP_DDI_CLOCK_SELECT_MASK, val); if (clock == XELPDP_DDI_CLOCK_SELECT_MAXPCLK || @@ -3016,6 +3017,9 @@ static void intel_c10pll_state_verify(const struct intel_crtc_state *state, const struct intel_c10pll_state *mpllb_sw_state = &state->cx0pll_state.c10; int i; + if (intel_crtc_needs_fastset(state)) + return; + for (i = 0; i < ARRAY_SIZE(mpllb_sw_state->pll); i++) { u8 expected = mpllb_sw_state->pll[i]; @@ -3067,10 +3071,15 @@ static void intel_c20pll_state_verify(const struct intel_crtc_state *state, { struct drm_i915_private *i915 = to_i915(crtc->base.dev); const struct intel_c20pll_state *mpll_sw_state = &state->cx0pll_state.c20; - bool sw_use_mpllb = mpll_sw_state->tx[0] & C20_PHY_USE_MPLLB; - bool hw_use_mpllb = mpll_hw_state->tx[0] & C20_PHY_USE_MPLLB; + bool sw_use_mpllb = intel_c20phy_use_mpllb(mpll_sw_state); + bool hw_use_mpllb = intel_c20phy_use_mpllb(mpll_hw_state); int i; + I915_STATE_WARN(i915, mpll_hw_state->clock != mpll_sw_state->clock, + "[CRTC:%d:%s] mismatch in C20: Register CLOCK (expected %d, found %d)", + crtc->base.base.id, crtc->base.name, + mpll_sw_state->clock, mpll_hw_state->clock); + I915_STATE_WARN(i915, sw_use_mpllb != hw_use_mpllb, "[CRTC:%d:%s] mismatch in C20: Register MPLLB selection (expected %d, found %d)", crtc->base.base.id, crtc->base.name, diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h index adf8f4ce0d49..bdd0c8c4ef97 100644 --- a/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h +++ b/drivers/gpu/drm/i915/display/intel_cx0_phy_regs.h @@ -7,16 +7,39 @@ #define __INTEL_CX0_PHY_REGS_H__ #include "i915_reg_defs.h" +#include "intel_display_limits.h" + +/* + * Wrapper macro to convert from port number to the index used in some of the + * registers. For Display version 20 and above it converts the port number to a + * single range, starting with the TC offsets. When used together with + * _PICK_EVEN_2RANGES(idx, PORT_TC1, ...), this single range will be the second + * range. Example: + * + * PORT_TC1 -> PORT_TC1 + * PORT_TC2 -> PORT_TC2 + * PORT_TC3 -> PORT_TC3 + * PORT_TC4 -> PORT_TC4 + * PORT_A -> PORT_TC4 + 1 + * PORT_B -> PORT_TC4 + 2 + * ... + */ +#define __xe2lpd_port_idx(port) \ + (port >= PORT_TC1 ? port : PORT_TC4 + 1 + port - PORT_A) #define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A 0x64040 #define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B 0x64140 #define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1 0x16F240 #define _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2 0x16F440 -#define XELPDP_PORT_M2P_MSGBUS_CTL(port, lane) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ +#define _XELPDP_PORT_M2P_MSGBUS_CTL(idx, lane) _MMIO(_PICK_EVEN_2RANGES(idx, PORT_TC1, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2) + (lane) * 4) +#define XELPDP_PORT_M2P_MSGBUS_CTL(i915__, port, lane) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_M2P_MSGBUS_CTL(__xe2lpd_port_idx(port), lane) : \ + _XELPDP_PORT_M2P_MSGBUS_CTL(port, lane)) #define XELPDP_PORT_M2P_TRANSACTION_PENDING REG_BIT(31) #define XELPDP_PORT_M2P_COMMAND_TYPE_MASK REG_GENMASK(30, 27) #define XELPDP_PORT_M2P_COMMAND_WRITE_UNCOMMITTED REG_FIELD_PREP(XELPDP_PORT_M2P_COMMAND_TYPE_MASK, 0x1) @@ -27,11 +50,16 @@ #define XELPDP_PORT_M2P_TRANSACTION_RESET REG_BIT(15) #define XELPDP_PORT_M2P_ADDRESS_MASK REG_GENMASK(11, 0) #define XELPDP_PORT_M2P_ADDRESS(val) REG_FIELD_PREP(XELPDP_PORT_M2P_ADDRESS_MASK, val) -#define XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ + +#define _XELPDP_PORT_P2M_MSGBUS_STATUS(idx, lane) _MMIO(_PICK_EVEN_2RANGES(idx, PORT_TC1, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_A, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_B, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC1, \ _XELPDP_PORT_M2P_MSGBUS_CTL_LN0_USBC2) + (lane) * 4 + 8) +#define XELPDP_PORT_P2M_MSGBUS_STATUS(i915__, port, lane) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_P2M_MSGBUS_STATUS(__xe2lpd_port_idx(port), lane) : \ + _XELPDP_PORT_P2M_MSGBUS_STATUS(port, lane)) #define XELPDP_PORT_P2M_RESPONSE_READY REG_BIT(31) #define XELPDP_PORT_P2M_COMMAND_TYPE_MASK REG_GENMASK(30, 27) #define XELPDP_PORT_P2M_COMMAND_READ_ACK 0x4 @@ -54,11 +82,15 @@ #define _XELPDP_PORT_BUF_CTL1_LN0_B 0x64104 #define _XELPDP_PORT_BUF_CTL1_LN0_USBC1 0x16F200 #define _XELPDP_PORT_BUF_CTL1_LN0_USBC2 0x16F400 -#define XELPDP_PORT_BUF_CTL1(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ +#define _XELPDP_PORT_BUF_CTL1(idx) _MMIO(_PICK_EVEN_2RANGES(idx, PORT_TC1, \ _XELPDP_PORT_BUF_CTL1_LN0_A, \ _XELPDP_PORT_BUF_CTL1_LN0_B, \ _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \ _XELPDP_PORT_BUF_CTL1_LN0_USBC2)) +#define XELPDP_PORT_BUF_CTL1(i915__, port) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_BUF_CTL1(__xe2lpd_port_idx(port)) : \ + _XELPDP_PORT_BUF_CTL1(port)) #define XELPDP_PORT_BUF_D2D_LINK_ENABLE REG_BIT(29) #define XELPDP_PORT_BUF_D2D_LINK_STATE REG_BIT(28) #define XELPDP_PORT_BUF_SOC_PHY_READY REG_BIT(24) @@ -75,12 +107,15 @@ #define XELPDP_PORT_WIDTH_MASK REG_GENMASK(3, 1) #define XELPDP_PORT_WIDTH(val) REG_FIELD_PREP(XELPDP_PORT_WIDTH_MASK, val) -#define XELPDP_PORT_BUF_CTL2(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ +#define _XELPDP_PORT_BUF_CTL2(idx) _MMIO(_PICK_EVEN_2RANGES(idx, PORT_TC1, \ _XELPDP_PORT_BUF_CTL1_LN0_A, \ _XELPDP_PORT_BUF_CTL1_LN0_B, \ _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \ _XELPDP_PORT_BUF_CTL1_LN0_USBC2) + 4) - +#define XELPDP_PORT_BUF_CTL2(i915__, port) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_BUF_CTL2(__xe2lpd_port_idx(port)) : \ + _XELPDP_PORT_BUF_CTL2(port)) #define XELPDP_LANE_PIPE_RESET(lane) _PICK(lane, REG_BIT(31), REG_BIT(30)) #define XELPDP_LANE_PHY_CURRENT_STATUS(lane) _PICK(lane, REG_BIT(29), REG_BIT(28)) #define XELPDP_LANE_POWERDOWN_UPDATE(lane) _PICK(lane, REG_BIT(25), REG_BIT(24)) @@ -95,11 +130,15 @@ #define XELPDP_POWER_STATE_READY_MASK REG_GENMASK(7, 4) #define XELPDP_POWER_STATE_READY(val) REG_FIELD_PREP(XELPDP_POWER_STATE_READY_MASK, val) -#define XELPDP_PORT_BUF_CTL3(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ +#define _XELPDP_PORT_BUF_CTL3(idx) _MMIO(_PICK_EVEN_2RANGES(idx, PORT_TC1, \ _XELPDP_PORT_BUF_CTL1_LN0_A, \ _XELPDP_PORT_BUF_CTL1_LN0_B, \ _XELPDP_PORT_BUF_CTL1_LN0_USBC1, \ _XELPDP_PORT_BUF_CTL1_LN0_USBC2) + 8) +#define XELPDP_PORT_BUF_CTL3(i915__, port) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_BUF_CTL3(__xe2lpd_port_idx(port)) : \ + _XELPDP_PORT_BUF_CTL3(port)) #define XELPDP_PLL_LANE_STAGGERING_DELAY_MASK REG_GENMASK(15, 8) #define XELPDP_PLL_LANE_STAGGERING_DELAY(val) REG_FIELD_PREP(XELPDP_PLL_LANE_STAGGERING_DELAY_MASK, val) #define XELPDP_POWER_STATE_ACTIVE_MASK REG_GENMASK(3, 0) @@ -114,11 +153,15 @@ #define _XELPDP_PORT_MSGBUS_TIMER_LN0_B 0x641d8 #define _XELPDP_PORT_MSGBUS_TIMER_LN0_USBC1 0x16f258 #define _XELPDP_PORT_MSGBUS_TIMER_LN0_USBC2 0x16f458 -#define XELPDP_PORT_MSGBUS_TIMER(port, lane) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ +#define _XELPDP_PORT_MSGBUS_TIMER(port, lane) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ _XELPDP_PORT_MSGBUS_TIMER_LN0_A, \ _XELPDP_PORT_MSGBUS_TIMER_LN0_B, \ _XELPDP_PORT_MSGBUS_TIMER_LN0_USBC1, \ _XELPDP_PORT_MSGBUS_TIMER_LN0_USBC2) + (lane) * 4) +#define XELPDP_PORT_MSGBUS_TIMER(i915__, port, lane) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_MSGBUS_TIMER(__xe2lpd_port_idx(port), lane) : \ + _XELPDP_PORT_MSGBUS_TIMER(port, lane)) #define XELPDP_PORT_MSGBUS_TIMER_TIMED_OUT REG_BIT(31) #define XELPDP_PORT_MSGBUS_TIMER_VAL_MASK REG_GENMASK(23, 0) #define XELPDP_PORT_MSGBUS_TIMER_VAL REG_FIELD_PREP(XELPDP_PORT_MSGBUS_TIMER_VAL_MASK, 0xa000) @@ -127,11 +170,15 @@ #define _XELPDP_PORT_CLOCK_CTL_B 0x641E0 #define _XELPDP_PORT_CLOCK_CTL_USBC1 0x16F260 #define _XELPDP_PORT_CLOCK_CTL_USBC2 0x16F460 -#define XELPDP_PORT_CLOCK_CTL(port) _MMIO(_PICK_EVEN_2RANGES(port, PORT_TC1, \ +#define _XELPDP_PORT_CLOCK_CTL(idx) _MMIO(_PICK_EVEN_2RANGES(idx, PORT_TC1, \ _XELPDP_PORT_CLOCK_CTL_A, \ _XELPDP_PORT_CLOCK_CTL_B, \ _XELPDP_PORT_CLOCK_CTL_USBC1, \ _XELPDP_PORT_CLOCK_CTL_USBC2)) +#define XELPDP_PORT_CLOCK_CTL(i915__, port) \ + (DISPLAY_VER(i915__) >= 20 ? \ + _XELPDP_PORT_CLOCK_CTL(__xe2lpd_port_idx(port)) : \ + _XELPDP_PORT_CLOCK_CTL(port)) #define XELPDP_LANE_PCLK_PLL_REQUEST(lane) REG_BIT(31 - ((lane) * 4)) #define XELPDP_LANE_PCLK_PLL_ACK(lane) REG_BIT(30 - ((lane) * 4)) #define XELPDP_LANE_PCLK_REFCLK_REQUEST(lane) REG_BIT(29 - ((lane) * 4)) diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c index 12a29363e5df..bea441590204 100644 --- a/drivers/gpu/drm/i915/display/intel_ddi.c +++ b/drivers/gpu/drm/i915/display/intel_ddi.c @@ -178,7 +178,7 @@ static void mtl_wait_ddi_buf_idle(struct drm_i915_private *i915, enum port port) int ret; /* FIXME: find out why Bspec's 100us timeout is too short */ - ret = wait_for_us((intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & + ret = wait_for_us((intel_de_read(i915, XELPDP_PORT_BUF_CTL1(i915, port)) & XELPDP_PORT_BUF_PHY_IDLE), 10000); if (ret) drm_err(&i915->drm, "Timeout waiting for DDI BUF %c to get idle\n", @@ -226,7 +226,9 @@ static void intel_wait_ddi_buf_active(struct drm_i915_private *dev_priv, } if (DISPLAY_VER(dev_priv) >= 14) - ret = _wait_for(!(intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_PORT_BUF_PHY_IDLE), + ret = _wait_for(!(intel_de_read(dev_priv, + XELPDP_PORT_BUF_CTL1(dev_priv, port)) & + XELPDP_PORT_BUF_PHY_IDLE), timeout_us, 10, 10); else ret = _wait_for(!(intel_de_read(dev_priv, DDI_BUF_CTL(port)) & DDI_BUF_IS_IDLE), @@ -2429,13 +2431,22 @@ mtl_ddi_enable_d2d(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum port port = encoder->port; + i915_reg_t reg; + u32 set_bits, wait_bits; - intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), 0, - XELPDP_PORT_BUF_D2D_LINK_ENABLE); + if (DISPLAY_VER(dev_priv) >= 20) { + reg = DDI_BUF_CTL(port); + set_bits = XE2LPD_DDI_BUF_D2D_LINK_ENABLE; + wait_bits = XE2LPD_DDI_BUF_D2D_LINK_STATE; + } else { + reg = XELPDP_PORT_BUF_CTL1(dev_priv, port); + set_bits = XELPDP_PORT_BUF_D2D_LINK_ENABLE; + wait_bits = XELPDP_PORT_BUF_D2D_LINK_STATE; + } - if (wait_for_us((intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & - XELPDP_PORT_BUF_D2D_LINK_STATE), 100)) { - drm_err(&dev_priv->drm, "Timeout waiting for D2D Link enable for PORT_BUF_CTL %c\n", + intel_de_rmw(dev_priv, reg, 0, set_bits); + if (wait_for_us(intel_de_read(dev_priv, reg) & wait_bits, 100)) { + drm_err(&dev_priv->drm, "Timeout waiting for D2D Link enable for DDI/PORT_BUF_CTL %c\n", port_name(port)); } } @@ -2448,7 +2459,7 @@ static void mtl_port_buf_ctl_program(struct intel_encoder *encoder, enum port port = encoder->port; u32 val; - val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)); + val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(i915, port)); val &= ~XELPDP_PORT_WIDTH_MASK; val |= XELPDP_PORT_WIDTH(mtl_get_port_width(crtc_state->lane_count)); @@ -2461,7 +2472,7 @@ static void mtl_port_buf_ctl_program(struct intel_encoder *encoder, if (dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL) val |= XELPDP_PORT_REVERSAL; - intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val); + intel_de_write(i915, XELPDP_PORT_BUF_CTL1(i915, port), val); } static void mtl_port_buf_ctl_io_selection(struct intel_encoder *encoder) @@ -2472,7 +2483,7 @@ static void mtl_port_buf_ctl_io_selection(struct intel_encoder *encoder) val = intel_tc_port_in_tbt_alt_mode(dig_port) ? XELPDP_PORT_BUF_IO_SELECT_TBT : 0; - intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(encoder->port), + intel_de_rmw(i915, XELPDP_PORT_BUF_CTL1(i915, encoder->port), XELPDP_PORT_BUF_IO_SELECT_TBT, val); } @@ -2898,13 +2909,22 @@ mtl_ddi_disable_d2d_link(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum port port = encoder->port; + i915_reg_t reg; + u32 clr_bits, wait_bits; - intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), - XELPDP_PORT_BUF_D2D_LINK_ENABLE, 0); + if (DISPLAY_VER(dev_priv) >= 20) { + reg = DDI_BUF_CTL(port); + clr_bits = XE2LPD_DDI_BUF_D2D_LINK_ENABLE; + wait_bits = XE2LPD_DDI_BUF_D2D_LINK_STATE; + } else { + reg = XELPDP_PORT_BUF_CTL1(dev_priv, port); + clr_bits = XELPDP_PORT_BUF_D2D_LINK_ENABLE; + wait_bits = XELPDP_PORT_BUF_D2D_LINK_STATE; + } - if (wait_for_us(!(intel_de_read(dev_priv, XELPDP_PORT_BUF_CTL1(port)) & - XELPDP_PORT_BUF_D2D_LINK_STATE), 100)) - drm_err(&dev_priv->drm, "Timeout waiting for D2D Link disable for PORT_BUF_CTL %c\n", + intel_de_rmw(dev_priv, reg, clr_bits, 0); + if (wait_for_us(!(intel_de_read(dev_priv, reg) & wait_bits), 100)) + drm_err(&dev_priv->drm, "Timeout waiting for D2D Link disable for DDI/PORT_BUF_CTL %c\n", port_name(port)); } @@ -3038,7 +3058,7 @@ static void intel_ddi_post_disable_dp(struct intel_atomic_state *state, /* De-select Thunderbolt */ if (DISPLAY_VER(dev_priv) >= 14) - intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(encoder->port), + intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(dev_priv, encoder->port), XELPDP_PORT_BUF_IO_SELECT_TBT, 0); } @@ -3319,10 +3339,13 @@ static void intel_enable_ddi_hdmi(struct intel_atomic_state *state, if (dig_port->saved_port_bits & DDI_BUF_PORT_REVERSAL) port_buf |= XELPDP_PORT_REVERSAL; - intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(port), + intel_de_rmw(dev_priv, XELPDP_PORT_BUF_CTL1(dev_priv, port), XELPDP_PORT_WIDTH_MASK | XELPDP_PORT_REVERSAL, port_buf); buf_ctl |= DDI_PORT_WIDTH(lane_count); + + if (DISPLAY_VER(dev_priv) >= 20) + buf_ctl |= XE2LPD_DDI_BUF_D2D_LINK_ENABLE; } else if (IS_ALDERLAKE_P(dev_priv) && intel_phy_is_tc(dev_priv, phy)) { drm_WARN_ON(&dev_priv->drm, !intel_tc_port_in_legacy_mode(dig_port)); buf_ctl |= DDI_BUF_CTL_TC_PHY_OWNERSHIP; @@ -3543,6 +3566,9 @@ static void mtl_ddi_prepare_link_retrain(struct intel_dp *intel_dp, /* 6.i Configure and enable DDI_CTL_DE to start sending valid data to port slice */ intel_dp->DP |= DDI_BUF_CTL_ENABLE; + if (DISPLAY_VER(dev_priv) >= 20) + intel_dp->DP |= XE2LPD_DDI_BUF_D2D_LINK_ENABLE; + intel_de_write(dev_priv, DDI_BUF_CTL(port), intel_dp->DP); intel_de_posting_read(dev_priv, DDI_BUF_CTL(port)); @@ -3941,11 +3967,11 @@ static void intel_ddi_get_config(struct intel_encoder *encoder, if (DISPLAY_VER(dev_priv) >= 8) bdw_get_trans_port_sync_config(pipe_config); + intel_psr_get_config(encoder, pipe_config); + intel_read_dp_sdp(encoder, pipe_config, HDMI_PACKET_TYPE_GAMUT_METADATA); intel_read_dp_sdp(encoder, pipe_config, DP_SDP_VSC); - intel_psr_get_config(encoder, pipe_config); - intel_audio_codec_get_config(encoder, pipe_config); } @@ -5117,6 +5143,9 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, encoder->suspend_complete = intel_ddi_tc_encoder_suspend_complete; encoder->shutdown_complete = intel_ddi_tc_encoder_shutdown_complete; + dig_port->lock = intel_tc_port_lock; + dig_port->unlock = intel_tc_port_unlock; + if (intel_tc_port_init(dig_port, is_legacy) < 0) goto err; } diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 57f78492f6ea..7db0655d8c9e 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -104,6 +104,7 @@ #include "intel_pmdemand.h" #include "intel_pps.h" #include "intel_psr.h" +#include "intel_psr_regs.h" #include "intel_sdvo.h" #include "intel_snps_phy.h" #include "intel_tc.h" @@ -2706,6 +2707,15 @@ static void intel_set_pipe_src_size(const struct intel_crtc_state *crtc_state) */ intel_de_write(dev_priv, PIPESRC(pipe), PIPESRC_WIDTH(width - 1) | PIPESRC_HEIGHT(height - 1)); + + if (!crtc_state->enable_psr2_su_region_et) + return; + + width = drm_rect_width(&crtc_state->psr2_su_area); + height = drm_rect_height(&crtc_state->psr2_su_area); + + intel_de_write(dev_priv, PIPE_SRCSZ_ERLY_TPT(pipe), + PIPESRC_WIDTH(width - 1) | PIPESRC_HEIGHT(height - 1)); } static bool intel_pipe_is_interlaced(const struct intel_crtc_state *crtc_state) @@ -4764,7 +4774,11 @@ static bool intel_compare_dp_vsc_sdp(const struct drm_dp_vsc_sdp *a, const struct drm_dp_vsc_sdp *b) { - return memcmp(a, b, sizeof(*a)) == 0; + return a->pixelformat == b->pixelformat && + a->colorimetry == b->colorimetry && + a->bpc == b->bpc && + a->dynamic_range == b->dynamic_range && + a->content_type == b->content_type; } static bool @@ -5044,8 +5058,7 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, } while (0) #define PIPE_CONF_CHECK_DP_VSC_SDP(name) do { \ - if (!current_config->has_psr && !pipe_config->has_psr && \ - !intel_compare_dp_vsc_sdp(¤t_config->infoframes.name, \ + if (!intel_compare_dp_vsc_sdp(¤t_config->infoframes.name, \ &pipe_config->infoframes.name)) { \ pipe_config_dp_vsc_sdp_mismatch(dev_priv, fastset, __stringify(name), \ ¤t_config->infoframes.name, \ @@ -5198,13 +5211,6 @@ intel_pipe_config_compare(const struct intel_crtc_state *current_config, PIPE_CONF_CHECK_CSC(csc); PIPE_CONF_CHECK_CSC(output_csc); - - if (current_config->active_planes) { - PIPE_CONF_CHECK_BOOL(has_psr); - PIPE_CONF_CHECK_BOOL(has_psr2); - PIPE_CONF_CHECK_BOOL(enable_psr2_sel_fetch); - PIPE_CONF_CHECK_I(dc3co_exitline); - } } PIPE_CONF_CHECK_BOOL(double_wide); @@ -6306,6 +6312,9 @@ int intel_atomic_check(struct drm_device *dev, int ret, i; bool any_ms = false; + if (!intel_display_driver_check_access(dev_priv)) + return -ENODEV; + for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { /* @@ -7067,6 +7076,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) drm_atomic_helper_wait_for_dependencies(&state->base); drm_dp_mst_atomic_wait_for_dependencies(&state->base); + intel_atomic_global_state_wait_for_dependencies(state); /* * During full modesets we write a lot of registers, wait @@ -7243,6 +7253,7 @@ static void intel_atomic_commit_tail(struct intel_atomic_state *state) intel_pmdemand_post_plane_update(state); drm_atomic_helper_commit_hw_done(&state->base); + intel_atomic_global_state_commit_done(state); if (state->modeset) { /* As one of the primary mmio accessors, KMS has a high @@ -7293,6 +7304,38 @@ static void intel_atomic_track_fbs(struct intel_atomic_state *state) plane->frontbuffer_bit); } +static int intel_atomic_setup_commit(struct intel_atomic_state *state, bool nonblock) +{ + int ret; + + ret = drm_atomic_helper_setup_commit(&state->base, nonblock); + if (ret) + return ret; + + ret = intel_atomic_global_state_setup_commit(state); + if (ret) + return ret; + + return 0; +} + +static int intel_atomic_swap_state(struct intel_atomic_state *state) +{ + int ret; + + ret = drm_atomic_helper_swap_state(&state->base, true); + if (ret) + return ret; + + intel_atomic_swap_global_state(state); + + intel_shared_dpll_swap_state(state); + + intel_atomic_track_fbs(state); + + return 0; +} + int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state, bool nonblock) { @@ -7338,11 +7381,9 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state, return ret; } - ret = drm_atomic_helper_setup_commit(&state->base, nonblock); - if (!ret) - ret = drm_atomic_helper_swap_state(&state->base, true); + ret = intel_atomic_setup_commit(state, nonblock); if (!ret) - intel_atomic_swap_global_state(state); + ret = intel_atomic_swap_state(state); if (ret) { struct intel_crtc_state *new_crtc_state; @@ -7356,8 +7397,6 @@ int intel_atomic_commit(struct drm_device *dev, struct drm_atomic_state *_state, intel_runtime_pm_put(&dev_priv->runtime_pm, state->wakeref); return ret; } - intel_shared_dpll_swap_state(state); - intel_atomic_track_fbs(state); drm_atomic_state_get(&state->base); INIT_WORK(&state->base.commit_work, intel_atomic_commit_work); @@ -7810,6 +7849,7 @@ static const struct intel_display_funcs skl_display_funcs = { .crtc_disable = hsw_crtc_disable, .commit_modeset_enables = skl_commit_modeset_enables, .get_initial_plane_config = skl_get_initial_plane_config, + .fixup_initial_plane_config = skl_fixup_initial_plane_config, }; static const struct intel_display_funcs ddi_display_funcs = { @@ -7818,6 +7858,7 @@ static const struct intel_display_funcs ddi_display_funcs = { .crtc_disable = hsw_crtc_disable, .commit_modeset_enables = intel_commit_modeset_enables, .get_initial_plane_config = i9xx_get_initial_plane_config, + .fixup_initial_plane_config = i9xx_fixup_initial_plane_config, }; static const struct intel_display_funcs pch_split_display_funcs = { @@ -7826,6 +7867,7 @@ static const struct intel_display_funcs pch_split_display_funcs = { .crtc_disable = ilk_crtc_disable, .commit_modeset_enables = intel_commit_modeset_enables, .get_initial_plane_config = i9xx_get_initial_plane_config, + .fixup_initial_plane_config = i9xx_fixup_initial_plane_config, }; static const struct intel_display_funcs vlv_display_funcs = { @@ -7834,6 +7876,7 @@ static const struct intel_display_funcs vlv_display_funcs = { .crtc_disable = i9xx_crtc_disable, .commit_modeset_enables = intel_commit_modeset_enables, .get_initial_plane_config = i9xx_get_initial_plane_config, + .fixup_initial_plane_config = i9xx_fixup_initial_plane_config, }; static const struct intel_display_funcs i9xx_display_funcs = { @@ -7842,6 +7885,7 @@ static const struct intel_display_funcs i9xx_display_funcs = { .crtc_disable = i9xx_crtc_disable, .commit_modeset_enables = intel_commit_modeset_enables, .get_initial_plane_config = i9xx_get_initial_plane_config, + .fixup_initial_plane_config = i9xx_fixup_initial_plane_config, }; /** diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h index 47297ed85822..fdeaac994e17 100644 --- a/drivers/gpu/drm/i915/display/intel_display_core.h +++ b/drivers/gpu/drm/i915/display/intel_display_core.h @@ -28,6 +28,8 @@ #include "intel_opregion.h" #include "intel_wm_types.h" +struct task_struct; + struct drm_i915_private; struct drm_property; struct drm_property_blob; @@ -47,6 +49,7 @@ struct intel_fbdev; struct intel_fdi_funcs; struct intel_hotplug_funcs; struct intel_initial_plane_config; +struct intel_opregion; struct intel_overlay; /* Amount of SAGV/QGV points, BSpec precisely defines this */ @@ -64,6 +67,8 @@ struct intel_display_funcs { struct intel_crtc_state *); void (*get_initial_plane_config)(struct intel_crtc *, struct intel_initial_plane_config *); + bool (*fixup_initial_plane_config)(struct intel_crtc *crtc, + const struct intel_initial_plane_config *plane_config); void (*crtc_enable)(struct intel_atomic_state *state, struct intel_crtc *crtc); void (*crtc_disable)(struct intel_atomic_state *state, @@ -172,6 +177,12 @@ struct intel_hotplug { struct work_struct poll_init_work; bool poll_enabled; + /* + * Queuing of hotplug_work, reenable_work and poll_init_work is + * enabled. Protected by drm_i915_private::irq_lock. + */ + bool detection_work_enabled; + unsigned int hpd_storm_threshold; /* Whether or not to count short HPD IRQs in HPD storms */ u8 hpd_short_storm_enabled; @@ -299,6 +310,11 @@ struct intel_display { } funcs; struct { + bool any_task_allowed; + struct task_struct *allowed_task; + } access; + + struct { /* backlight registers and fields in struct intel_panel */ struct mutex lock; } backlight; @@ -513,7 +529,7 @@ struct intel_display { struct intel_fbc *fbc[I915_MAX_FBCS]; struct intel_frontbuffer_tracking fb_tracking; struct intel_hotplug hotplug; - struct intel_opregion opregion; + struct intel_opregion *opregion; struct intel_overlay *overlay; struct intel_display_params params; struct intel_vbt_data vbt; diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c index d951edb36687..6f2d13c8ccf7 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c @@ -86,28 +86,6 @@ static int i915_sr_status(struct seq_file *m, void *unused) return 0; } -static int i915_opregion(struct seq_file *m, void *unused) -{ - struct drm_i915_private *i915 = node_to_i915(m->private); - struct intel_opregion *opregion = &i915->display.opregion; - - if (opregion->header) - seq_write(m, opregion->header, OPREGION_SIZE); - - return 0; -} - -static int i915_vbt(struct seq_file *m, void *unused) -{ - struct drm_i915_private *i915 = node_to_i915(m->private); - struct intel_opregion *opregion = &i915->display.opregion; - - if (opregion->vbt) - seq_write(m, opregion->vbt, opregion->vbt_size); - - return 0; -} - static int i915_gem_framebuffer_info(struct seq_file *m, void *data) { struct drm_i915_private *dev_priv = node_to_i915(m->private); @@ -1066,8 +1044,6 @@ static const struct file_operations i915_fifo_underrun_reset_ops = { static const struct drm_info_list intel_display_debugfs_list[] = { {"i915_frontbuffer_tracking", i915_frontbuffer_tracking, 0}, {"i915_sr_status", i915_sr_status, 0}, - {"i915_opregion", i915_opregion, 0}, - {"i915_vbt", i915_vbt, 0}, {"i915_gem_framebuffer", i915_gem_framebuffer_info, 0}, {"i915_power_domain_info", i915_power_domain_info, 0}, {"i915_display_info", i915_display_info, 0}, @@ -1105,10 +1081,12 @@ void intel_display_debugfs_register(struct drm_i915_private *i915) ARRAY_SIZE(intel_display_debugfs_list), minor->debugfs_root, minor); + intel_bios_debugfs_register(i915); intel_cdclk_debugfs_register(i915); intel_dmc_debugfs_register(i915); intel_fbc_debugfs_register(i915); intel_hpd_debugfs_register(i915); + intel_opregion_debugfs_register(i915); intel_psr_debugfs_register(i915); intel_wm_debugfs_register(i915); intel_display_debugfs_params(i915); diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c b/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c index b7e68eb62452..f35718748555 100644 --- a/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c @@ -3,6 +3,7 @@ * Copyright © 2023 Intel Corporation */ +#include <linux/debugfs.h> #include <linux/kernel.h> #include <drm/drm_drv.h> diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c index 0b522c6a8d6f..c02d79b50006 100644 --- a/drivers/gpu/drm/i915/display/intel_display_device.c +++ b/drivers/gpu/drm/i915/display/intel_display_device.c @@ -1012,7 +1012,7 @@ static void __intel_display_device_info_runtime_init(struct drm_i915_private *i9 goto display_fused_off; } - if (IS_GRAPHICS_VER(i915, 7, 8) && HAS_PCH_SPLIT(i915)) { + if (IS_DISPLAY_VER(i915, 7, 8) && HAS_PCH_SPLIT(i915)) { u32 fuse_strap = intel_de_read(i915, FUSE_STRAP); u32 sfuse_strap = intel_de_read(i915, SFUSE_STRAP); diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c index c874756d8b13..4f7ba7eb03d2 100644 --- a/drivers/gpu/drm/i915/display/intel_display_driver.c +++ b/drivers/gpu/drm/i915/display/intel_display_driver.c @@ -45,6 +45,7 @@ #include "intel_hdcp.h" #include "intel_hotplug.h" #include "intel_hti.h" +#include "intel_modeset_lock.h" #include "intel_modeset_setup.h" #include "intel_opregion.h" #include "intel_overlay.h" @@ -276,12 +277,144 @@ cleanup_bios: return ret; } +static void set_display_access(struct drm_i915_private *i915, + bool any_task_allowed, + struct task_struct *allowed_task) +{ + struct drm_modeset_acquire_ctx ctx; + int err; + + intel_modeset_lock_ctx_retry(&ctx, NULL, 0, err) { + err = drm_modeset_lock_all_ctx(&i915->drm, &ctx); + if (err) + continue; + + i915->display.access.any_task_allowed = any_task_allowed; + i915->display.access.allowed_task = allowed_task; + } + + drm_WARN_ON(&i915->drm, err); +} + +/** + * intel_display_driver_enable_user_access - Enable display HW access for all threads + * @i915: i915 device instance + * + * Enable the display HW access for all threads. Examples for such accesses + * are modeset commits and connector probing. + * + * This function should be called during driver loading and system resume once + * all the HW initialization steps are done. + */ +void intel_display_driver_enable_user_access(struct drm_i915_private *i915) +{ + set_display_access(i915, true, NULL); + + intel_hpd_enable_detection_work(i915); +} + +/** + * intel_display_driver_disable_user_access - Disable display HW access for user threads + * @i915: i915 device instance + * + * Disable the display HW access for user threads. Examples for such accesses + * are modeset commits and connector probing. For the current thread the + * access is still enabled, which should only perform HW init/deinit + * programming (as the initial modeset during driver loading or the disabling + * modeset during driver unloading and system suspend/shutdown). This function + * should be followed by calling either intel_display_driver_enable_user_access() + * after completing the HW init programming or + * intel_display_driver_suspend_access() after completing the HW deinit + * programming. + * + * This function should be called during driver loading/unloading and system + * suspend/shutdown before starting the HW init/deinit programming. + */ +void intel_display_driver_disable_user_access(struct drm_i915_private *i915) +{ + intel_hpd_disable_detection_work(i915); + + set_display_access(i915, false, current); +} + +/** + * intel_display_driver_suspend_access - Suspend display HW access for all threads + * @i915: i915 device instance + * + * Disable the display HW access for all threads. Examples for such accesses + * are modeset commits and connector probing. This call should be either + * followed by calling intel_display_driver_resume_access(), or the driver + * should be unloaded/shutdown. + * + * This function should be called during driver unloading and system + * suspend/shutdown after completing the HW deinit programming. + */ +void intel_display_driver_suspend_access(struct drm_i915_private *i915) +{ + set_display_access(i915, false, NULL); +} + +/** + * intel_display_driver_resume_access - Resume display HW access for the resume thread + * @i915: i915 device instance + * + * Enable the display HW access for the current resume thread, keeping the + * access disabled for all other (user) threads. Examples for such accesses + * are modeset commits and connector probing. The resume thread should only + * perform HW init programming (as the restoring modeset). This function + * should be followed by calling intel_display_driver_enable_user_access(), + * after completing the HW init programming steps. + * + * This function should be called during system resume before starting the HW + * init steps. + */ +void intel_display_driver_resume_access(struct drm_i915_private *i915) +{ + set_display_access(i915, false, current); +} + +/** + * intel_display_driver_check_access - Check if the current thread has disaplay HW access + * @i915: i915 device instance + * + * Check whether the current thread has display HW access, print a debug + * message if it doesn't. Such accesses are modeset commits and connector + * probing. If the function returns %false any HW access should be prevented. + * + * Returns %true if the current thread has display HW access, %false + * otherwise. + */ +bool intel_display_driver_check_access(struct drm_i915_private *i915) +{ + char comm[TASK_COMM_LEN]; + char current_task[TASK_COMM_LEN + 16]; + char allowed_task[TASK_COMM_LEN + 16] = "none"; + + if (i915->display.access.any_task_allowed || + i915->display.access.allowed_task == current) + return true; + + snprintf(current_task, sizeof(current_task), "%s[%d]", + get_task_comm(comm, current), + task_pid_vnr(current)); + + if (i915->display.access.allowed_task) + snprintf(allowed_task, sizeof(allowed_task), "%s[%d]", + get_task_comm(comm, i915->display.access.allowed_task), + task_pid_vnr(i915->display.access.allowed_task)); + + drm_dbg_kms(&i915->drm, + "Reject display access from task %s (allowed to %s)\n", + current_task, allowed_task); + + return false; +} + /* part #2: call after irq install, but before gem init */ int intel_display_driver_probe_nogem(struct drm_i915_private *i915) { struct drm_device *dev = &i915->drm; enum pipe pipe; - struct intel_crtc *crtc; int ret; if (!HAS_DISPLAY(i915)) @@ -315,8 +448,6 @@ int intel_display_driver_probe_nogem(struct drm_i915_private *i915) intel_display_driver_init_hw(i915); intel_dpll_update_ref_clks(i915); - intel_hdcp_component_init(i915); - if (i915->display.cdclk.max_cdclk_freq == 0) intel_update_max_cdclk(i915); @@ -326,16 +457,14 @@ int intel_display_driver_probe_nogem(struct drm_i915_private *i915) intel_vga_disable(i915); intel_setup_outputs(i915); + intel_display_driver_disable_user_access(i915); + drm_modeset_lock_all(dev); intel_modeset_setup_hw_state(i915, dev->mode_config.acquire_ctx); intel_acpi_assign_connector_fwnodes(i915); drm_modeset_unlock_all(dev); - for_each_intel_crtc(dev, crtc) { - if (!to_intel_crtc_state(crtc->base.state)->uapi.active) - continue; - intel_crtc_initial_plane_config(crtc); - } + intel_initial_plane_config(i915); /* * Make sure hardware watermarks really match the state we read out. @@ -357,6 +486,13 @@ int intel_display_driver_probe(struct drm_i915_private *i915) return 0; /* + * This will bind stuff into ggtt, so it needs to be done after + * the BIOS fb takeover and whatever else magic ggtt reservations + * happen during gem/ggtt init. + */ + intel_hdcp_component_init(i915); + + /* * Force all active planes to recompute their states. So that on * mode_setcrtc after probe, all the intel_plane_state variables * are already calculated and there is no assert_plane warnings @@ -374,7 +510,6 @@ int intel_display_driver_probe(struct drm_i915_private *i915) /* Only enable hotplug handling once the fbdev is fully set up. */ intel_hpd_init(i915); - intel_hpd_poll_disable(i915); skl_watermark_ipc_init(i915); @@ -395,6 +530,8 @@ void intel_display_driver_register(struct drm_i915_private *i915) intel_audio_init(i915); + intel_display_driver_enable_user_access(i915); + intel_display_debugfs_register(i915); /* @@ -413,6 +550,7 @@ void intel_display_driver_register(struct drm_i915_private *i915) * fbdev->async_cookie. */ drm_kms_helper_poll_init(&i915->drm); + intel_hpd_poll_disable(i915); intel_display_device_info_print(DISPLAY_INFO(i915), DISPLAY_RUNTIME_INFO(i915), &p); @@ -441,6 +579,8 @@ void intel_display_driver_remove_noirq(struct drm_i915_private *i915) if (!HAS_DISPLAY(i915)) return; + intel_display_driver_suspend_access(i915); + /* * Due to the hpd irq storm handling the hotplug work can re-arm the * poll handlers. Hence disable polling after hpd handling is shut down. @@ -487,14 +627,17 @@ void intel_display_driver_unregister(struct drm_i915_private *i915) return; intel_fbdev_unregister(i915); - intel_audio_deinit(i915); - /* * After flushing the fbdev (incl. a late async config which * will have delayed queuing of a hotplug event), then flush * the hotplug events. */ drm_kms_helper_poll_fini(&i915->drm); + + intel_display_driver_disable_user_access(i915); + + intel_audio_deinit(i915); + drm_atomic_helper_shutdown(&i915->drm); acpi_video_unregister(); diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.h b/drivers/gpu/drm/i915/display/intel_display_driver.h index c276a58ee329..42cc4af6d3fd 100644 --- a/drivers/gpu/drm/i915/display/intel_display_driver.h +++ b/drivers/gpu/drm/i915/display/intel_display_driver.h @@ -32,5 +32,11 @@ int __intel_display_driver_resume(struct drm_i915_private *i915, struct drm_atomic_state *state, struct drm_modeset_acquire_ctx *ctx); +void intel_display_driver_enable_user_access(struct drm_i915_private *i915); +void intel_display_driver_disable_user_access(struct drm_i915_private *i915); +void intel_display_driver_suspend_access(struct drm_i915_private *i915); +void intel_display_driver_resume_access(struct drm_i915_private *i915); +bool intel_display_driver_check_access(struct drm_i915_private *i915); + #endif /* __INTEL_DISPLAY_DRIVER_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c index a7d8f3fc98de..f846c5b108b5 100644 --- a/drivers/gpu/drm/i915/display/intel_display_irq.c +++ b/drivers/gpu/drm/i915/display/intel_display_irq.c @@ -266,12 +266,12 @@ void i915_disable_pipestat(struct drm_i915_private *dev_priv, intel_uncore_posting_read(&dev_priv->uncore, reg); } -static bool i915_has_asle(struct drm_i915_private *dev_priv) +static bool i915_has_asle(struct drm_i915_private *i915) { - if (!dev_priv->display.opregion.asle) + if (!IS_PINEVIEW(i915) && !IS_MOBILE(i915)) return false; - return IS_PINEVIEW(dev_priv) || IS_MOBILE(dev_priv); + return intel_opregion_asle_present(i915); } /** @@ -986,7 +986,7 @@ static void gen8_read_and_ack_pch_irqs(struct drm_i915_private *i915, u32 *pch_i * their flags both in the PICA and SDE IIR. */ if (*pch_iir & SDE_PICAINTERRUPT) { - drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTP); + drm_WARN_ON(&i915->drm, INTEL_PCH_TYPE(i915) < PCH_MTL); pica_ier = intel_de_rmw(i915, PICAINTERRUPT_IER, ~0, 0); *pica_iir = intel_de_read(i915, PICAINTERRUPT_IIR); @@ -1587,7 +1587,7 @@ 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) { + if (DISPLAY_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 | diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 3fdd8a517983..01eb6e4e6049 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -780,6 +780,8 @@ struct intel_plane_state { struct intel_initial_plane_config { struct intel_framebuffer *fb; + struct intel_memory_region *mem; + resource_size_t phys_base; struct i915_vma *vma; unsigned int tiling; int size; @@ -1213,12 +1215,12 @@ struct intel_crtc_state { bool has_psr; bool has_psr2; bool enable_psr2_sel_fetch; + bool enable_psr2_su_region_et; bool req_psr2_sdp_prior_scanline; bool has_panel_replay; bool wm_level_disabled; u32 dc3co_exitline; u16 su_y_granularity; - struct drm_dp_vsc_sdp psr_vsc; /* * Frequence the dpll for the port should run at. Differs from the @@ -1402,6 +1404,8 @@ struct intel_crtc_state { u32 psr2_man_track_ctl; + struct drm_rect psr2_su_area; + /* Variable Refresh Rate state */ struct { bool enable, in_range; @@ -1682,13 +1686,14 @@ struct intel_psr { /* Mutex for PSR state of the transcoder */ struct mutex lock; -#define I915_PSR_DEBUG_MODE_MASK 0x0f -#define I915_PSR_DEBUG_DEFAULT 0x00 -#define I915_PSR_DEBUG_DISABLE 0x01 -#define I915_PSR_DEBUG_ENABLE 0x02 -#define I915_PSR_DEBUG_FORCE_PSR1 0x03 -#define I915_PSR_DEBUG_ENABLE_SEL_FETCH 0x4 -#define I915_PSR_DEBUG_IRQ 0x10 +#define I915_PSR_DEBUG_MODE_MASK 0x0f +#define I915_PSR_DEBUG_DEFAULT 0x00 +#define I915_PSR_DEBUG_DISABLE 0x01 +#define I915_PSR_DEBUG_ENABLE 0x02 +#define I915_PSR_DEBUG_FORCE_PSR1 0x03 +#define I915_PSR_DEBUG_ENABLE_SEL_FETCH 0x4 +#define I915_PSR_DEBUG_IRQ 0x10 +#define I915_PSR_DEBUG_SU_REGION_ET_DISABLE 0x20 u32 debug; bool sink_support; @@ -1702,14 +1707,20 @@ struct intel_psr { unsigned int busy_frontbuffer_bits; bool sink_psr2_support; bool link_standby; - bool colorimetry_support; bool psr2_enabled; bool psr2_sel_fetch_enabled; bool psr2_sel_fetch_cff_enabled; bool req_psr2_sdp_prior_scanline; u8 sink_sync_latency; - u8 io_wake_lines; - u8 fast_wake_lines; + + struct { + u8 io_wake_lines; + u8 fast_wake_lines; + + /* LNL and beyond */ + u8 check_entry_lines; + } alpm_parameters; + ktime_t last_entry_attempt; ktime_t last_exit; bool sink_not_reliable; @@ -1833,6 +1844,8 @@ struct intel_dp { /* When we last wrote the OUI for eDP */ unsigned long last_oui_write; + + bool colorimetry_support; }; enum lspcon_vendor { @@ -1890,6 +1903,9 @@ struct intel_digital_port { u32 (*infoframes_enabled)(struct intel_encoder *encoder, const struct intel_crtc_state *pipe_config); bool (*connected)(struct intel_encoder *encoder); + + void (*lock)(struct intel_digital_port *dig_port); + void (*unlock)(struct intel_digital_port *dig_port); }; struct intel_dp_mst_encoder { diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c index b70502586ab9..835781624482 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.c +++ b/drivers/gpu/drm/i915/display/intel_dmc.c @@ -1158,7 +1158,7 @@ static int intel_dmc_debugfs_status_show(struct seq_file *m, void *unused) str_yes_no(intel_dmc_has_payload(i915))); seq_printf(m, "path: %s\n", dmc ? dmc->fw_path : "N/A"); seq_printf(m, "Pipe A fw needed: %s\n", - str_yes_no(GRAPHICS_VER(i915) >= 12)); + str_yes_no(DISPLAY_VER(i915) >= 12)); seq_printf(m, "Pipe A fw loaded: %s\n", str_yes_no(has_dmc_id_fw(i915, DMC_FW_PIPEA))); seq_printf(m, "Pipe B fw needed: %s\n", diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index f5ef95da5534..ab415f41924d 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -56,6 +56,7 @@ #include "intel_cx0_phy.h" #include "intel_ddi.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dp_aux.h" @@ -2616,58 +2617,38 @@ static void intel_dp_compute_vsc_sdp(struct intel_dp *intel_dp, struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) { - struct drm_dp_vsc_sdp *vsc = &crtc_state->infoframes.vsc; + struct drm_dp_vsc_sdp *vsc; - /* When a crtc state has PSR, VSC SDP will be handled by PSR routine */ - if (crtc_state->has_psr) + if ((!intel_dp->colorimetry_support || + !intel_dp_needs_vsc_sdp(crtc_state, conn_state)) && + !crtc_state->has_psr) return; - if (!intel_dp_needs_vsc_sdp(crtc_state, conn_state)) - return; + vsc = &crtc_state->infoframes.vsc; crtc_state->infoframes.enable |= intel_hdmi_infoframe_enable(DP_SDP_VSC); vsc->sdp_type = DP_SDP_VSC; - intel_dp_compute_vsc_colorimetry(crtc_state, conn_state, - &crtc_state->infoframes.vsc); -} -void intel_dp_compute_psr_vsc_sdp(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state, - struct drm_dp_vsc_sdp *vsc) -{ - vsc->sdp_type = DP_SDP_VSC; - - if (crtc_state->has_psr2) { - if (intel_dp->psr.colorimetry_support && - intel_dp_needs_vsc_sdp(crtc_state, conn_state)) { - /* [PSR2, +Colorimetry] */ - intel_dp_compute_vsc_colorimetry(crtc_state, conn_state, - vsc); - } else { - /* - * [PSR2, -Colorimetry] - * Prepare VSC Header for SU as per eDP 1.4 spec, Table 6-11 - * 3D stereo + PSR/PSR2 + Y-coordinate. - */ - vsc->revision = 0x4; - vsc->length = 0xe; - } + /* Needs colorimetry */ + if (intel_dp_needs_vsc_sdp(crtc_state, conn_state)) { + intel_dp_compute_vsc_colorimetry(crtc_state, conn_state, + vsc); + } else if (crtc_state->has_psr2) { + /* + * [PSR2 without colorimetry] + * Prepare VSC Header for SU as per eDP 1.4 spec, Table 6-11 + * 3D stereo + PSR/PSR2 + Y-coordinate. + */ + vsc->revision = 0x4; + vsc->length = 0xe; } else if (crtc_state->has_panel_replay) { - if (intel_dp->psr.colorimetry_support && - intel_dp_needs_vsc_sdp(crtc_state, conn_state)) { - /* [Panel Replay with colorimetry info] */ - intel_dp_compute_vsc_colorimetry(crtc_state, conn_state, - vsc); - } else { - /* - * [Panel Replay without colorimetry info] - * Prepare VSC Header for SU as per DP 2.0 spec, Table 2-223 - * VSC SDP supporting 3D stereo + Panel Replay. - */ - vsc->revision = 0x6; - vsc->length = 0x10; - } + /* + * [Panel Replay without colorimetry info] + * Prepare VSC Header for SU as per DP 2.0 spec, Table 2-223 + * VSC SDP supporting 3D stereo + Panel Replay. + */ + vsc->revision = 0x6; + vsc->length = 0x10; } else { /* * [PSR1] @@ -3345,13 +3326,6 @@ bool intel_dp_initial_fastset_check(struct intel_encoder *encoder, fastset = false; } - if (CAN_PSR(intel_dp)) { - drm_dbg_kms(&i915->drm, "[ENCODER:%d:%s] Forcing full modeset to compute PSR state\n", - encoder->base.base.id, encoder->base.name); - crtc_state->uapi.mode_changed = true; - fastset = false; - } - return fastset; } @@ -4288,24 +4262,6 @@ static void intel_write_dp_sdp(struct intel_encoder *encoder, dig_port->write_infoframe(encoder, crtc_state, type, &sdp, len); } -void intel_write_dp_vsc_sdp(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - const struct drm_dp_vsc_sdp *vsc) -{ - struct intel_digital_port *dig_port = enc_to_dig_port(encoder); - struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); - struct dp_sdp sdp = {}; - ssize_t len; - - len = intel_dp_vsc_sdp_pack(vsc, &sdp, sizeof(sdp)); - - if (drm_WARN_ON(&dev_priv->drm, len < 0)) - return; - - dig_port->write_infoframe(encoder, crtc_state, DP_SDP_VSC, - &sdp, len); -} - void intel_dp_set_infoframes(struct intel_encoder *encoder, bool enable, const struct intel_crtc_state *crtc_state, @@ -4332,9 +4288,7 @@ void intel_dp_set_infoframes(struct intel_encoder *encoder, if (!enable) return; - /* When PSR is enabled, VSC SDP is handled by PSR routine */ - if (!crtc_state->has_psr) - intel_write_dp_sdp(encoder, crtc_state, DP_SDP_VSC); + intel_write_dp_sdp(encoder, crtc_state, DP_SDP_VSC); intel_write_dp_sdp(encoder, crtc_state, HDMI_PACKET_TYPE_GAMUT_METADATA); } @@ -4465,10 +4419,6 @@ static void intel_read_dp_vsc_sdp(struct intel_encoder *encoder, struct dp_sdp sdp = {}; int ret; - /* When PSR is enabled, VSC SDP is handled by PSR routine */ - if (crtc_state->has_psr) - return; - if ((crtc_state->infoframes.enable & intel_hdmi_infoframe_enable(type)) == 0) return; @@ -4679,31 +4629,36 @@ static void intel_dp_phy_pattern_update(struct intel_dp *intel_dp, struct drm_dp_phy_test_params *data = &intel_dp->compliance.test_data.phytest; struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); + struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base; enum pipe pipe = crtc->pipe; u32 pattern_val; switch (data->phy_pattern) { - case DP_PHY_TEST_PATTERN_NONE: + case DP_LINK_QUAL_PATTERN_DISABLE: drm_dbg_kms(&dev_priv->drm, "Disable Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), 0x0); + if (DISPLAY_VER(dev_priv) >= 10) + intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), + DP_TP_CTL_TRAIN_PAT4_SEL_MASK | DP_TP_CTL_LINK_TRAIN_MASK, + DP_TP_CTL_LINK_TRAIN_NORMAL); break; - case DP_PHY_TEST_PATTERN_D10_2: + case DP_LINK_QUAL_PATTERN_D10_2: drm_dbg_kms(&dev_priv->drm, "Set D10.2 Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_D10_2); break; - case DP_PHY_TEST_PATTERN_ERROR_COUNT: + case DP_LINK_QUAL_PATTERN_ERROR_RATE: drm_dbg_kms(&dev_priv->drm, "Set Error Count Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_SCRAMBLED_0); break; - case DP_PHY_TEST_PATTERN_PRBS7: + case DP_LINK_QUAL_PATTERN_PRBS7: drm_dbg_kms(&dev_priv->drm, "Set PRBS7 Phy Test Pattern\n"); intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_PRBS7); break; - case DP_PHY_TEST_PATTERN_80BIT_CUSTOM: + case DP_LINK_QUAL_PATTERN_80BIT_CUSTOM: /* * FIXME: Ideally pattern should come from DPCD 0x250. As * current firmware of DPR-100 could not set it, so hardcoding @@ -4721,7 +4676,7 @@ static void intel_dp_phy_pattern_update(struct intel_dp *intel_dp, DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_CUSTOM80); break; - case DP_PHY_TEST_PATTERN_CP2520: + case DP_LINK_QUAL_PATTERN_CP2520_PAT_1: /* * FIXME: Ideally pattern should come from DPCD 0x24A. As * current firmware of DPR-100 could not set it, so hardcoding @@ -4733,8 +4688,19 @@ static void intel_dp_phy_pattern_update(struct intel_dp *intel_dp, DDI_DP_COMP_CTL_ENABLE | DDI_DP_COMP_CTL_HBR2 | pattern_val); break; + case DP_LINK_QUAL_PATTERN_CP2520_PAT_3: + if (DISPLAY_VER(dev_priv) < 10) { + drm_warn(&dev_priv->drm, "Platform does not support TPS4\n"); + break; + } + drm_dbg_kms(&dev_priv->drm, "Set TPS4 compliance Phy Test Pattern\n"); + intel_de_write(dev_priv, DDI_DP_COMP_CTL(pipe), 0x0); + intel_de_rmw(dev_priv, dp_tp_ctl_reg(encoder, crtc_state), + DP_TP_CTL_TRAIN_PAT4_SEL_MASK | DP_TP_CTL_LINK_TRAIN_MASK, + DP_TP_CTL_TRAIN_PAT4_SEL_TP4A | DP_TP_CTL_LINK_TRAIN_PAT4); + break; default: - WARN(1, "Invalid Phy Test Pattern\n"); + drm_warn(&dev_priv->drm, "Invalid Phy Test Pattern\n"); } } @@ -5453,8 +5419,24 @@ edp_detect(struct intel_dp *intel_dp) return connector_status_connected; } +void intel_digital_port_lock(struct intel_encoder *encoder) +{ + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + + if (dig_port->lock) + dig_port->lock(dig_port); +} + +void intel_digital_port_unlock(struct intel_encoder *encoder) +{ + struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + + if (dig_port->unlock) + dig_port->unlock(dig_port); +} + /* - * intel_digital_port_connected - is the specified port connected? + * intel_digital_port_connected_locked - is the specified port connected? * @encoder: intel_encoder * * In cases where there's a connector physically connected but it can't be used @@ -5462,21 +5444,44 @@ edp_detect(struct intel_dp *intel_dp) * pretty much treat the port as disconnected. This is relevant for type-C * (starting on ICL) where there's ownership involved. * + * The caller must hold the lock acquired by calling intel_digital_port_lock() + * when calling this function. + * * Return %true if port is connected, %false otherwise. */ -bool intel_digital_port_connected(struct intel_encoder *encoder) +bool intel_digital_port_connected_locked(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_digital_port *dig_port = enc_to_dig_port(encoder); + bool is_glitch_free = intel_tc_port_handles_hpd_glitches(dig_port); bool is_connected = false; intel_wakeref_t wakeref; - with_intel_display_power(dev_priv, POWER_DOMAIN_DISPLAY_CORE, wakeref) - is_connected = dig_port->connected(encoder); + with_intel_display_power(dev_priv, POWER_DOMAIN_DISPLAY_CORE, wakeref) { + unsigned long wait_expires = jiffies + msecs_to_jiffies_timeout(4); + + do { + is_connected = dig_port->connected(encoder); + if (is_connected || is_glitch_free) + break; + usleep_range(10, 30); + } while (time_before(jiffies, wait_expires)); + } return is_connected; } +bool intel_digital_port_connected(struct intel_encoder *encoder) +{ + bool ret; + + intel_digital_port_lock(encoder); + ret = intel_digital_port_connected_locked(encoder); + intel_digital_port_unlock(encoder); + + return ret; +} + static const struct drm_edid * intel_dp_get_edid(struct intel_dp *intel_dp) { @@ -5670,6 +5675,9 @@ intel_dp_detect(struct drm_connector *connector, if (!intel_display_device_enabled(dev_priv)) return connector_status_disconnected; + if (!intel_display_driver_check_access(dev_priv)) + return connector->status; + /* Can't disconnect eDP */ if (intel_dp_is_edp(intel_dp)) status = edp_detect(intel_dp); @@ -5770,6 +5778,10 @@ intel_dp_force(struct drm_connector *connector) drm_dbg_kms(&dev_priv->drm, "[CONNECTOR:%d:%s]\n", connector->base.id, connector->name); + + if (!intel_display_driver_check_access(dev_priv)) + return; + intel_dp_unset_edid(intel_dp); if (connector->status != connector_status_connected) @@ -6054,7 +6066,7 @@ static void intel_dp_oob_hotplug_event(struct drm_connector *connector, spin_unlock_irq(&i915->irq_lock); if (need_work) - queue_delayed_work(i915->unordered_wq, &i915->display.hotplug.hotplug_work, 0); + intel_hpd_schedule_detection(i915); } static const struct drm_connector_funcs intel_dp_connector_funcs = { @@ -6497,6 +6509,7 @@ intel_dp_init_connector(struct intel_digital_port *dig_port, connector->interlace_allowed = true; intel_connector->polled = DRM_CONNECTOR_POLL_HPD; + intel_connector->base.polled = intel_connector->polled; intel_connector_attach_encoder(intel_connector, intel_encoder); @@ -6527,6 +6540,9 @@ intel_dp_init_connector(struct intel_digital_port *dig_port, "HDCP init failed, skipping.\n"); } + intel_dp->colorimetry_support = + intel_dp_get_colorimetry_status(intel_dp); + intel_dp->frl.is_trained = false; intel_dp->frl.trained_rate_gbps = 0; diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h index 05db46b111f2..530cc97bc42f 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.h +++ b/drivers/gpu/drm/i915/display/intel_dp.h @@ -109,20 +109,16 @@ int intel_dp_max_data_rate(int max_link_rate, int max_lanes); bool intel_dp_can_bigjoiner(struct intel_dp *intel_dp); bool intel_dp_needs_vsc_sdp(const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); -void intel_dp_compute_psr_vsc_sdp(struct intel_dp *intel_dp, - const struct intel_crtc_state *crtc_state, - const struct drm_connector_state *conn_state, - struct drm_dp_vsc_sdp *vsc); -void intel_write_dp_vsc_sdp(struct intel_encoder *encoder, - const struct intel_crtc_state *crtc_state, - const struct drm_dp_vsc_sdp *vsc); void intel_dp_set_infoframes(struct intel_encoder *encoder, bool enable, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state); void intel_read_dp_sdp(struct intel_encoder *encoder, struct intel_crtc_state *crtc_state, unsigned int type); +void intel_digital_port_lock(struct intel_encoder *encoder); +void intel_digital_port_unlock(struct intel_encoder *encoder); bool intel_digital_port_connected(struct intel_encoder *encoder); +bool intel_digital_port_connected_locked(struct intel_encoder *encoder); int intel_dp_dsc_compute_max_bpp(const struct intel_connector *connector, u8 dsc_max_bpc); u16 intel_dp_dsc_get_max_compressed_bpp(struct drm_i915_private *i915, diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c index 2e2af71bcd5a..4f4a0e3b3114 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c @@ -9,6 +9,7 @@ #include "intel_bios.h" #include "intel_de.h" #include "intel_display_types.h" +#include "intel_dp.h" #include "intel_dp_aux.h" #include "intel_dp_aux_regs.h" #include "intel_pps.h" @@ -228,9 +229,8 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp, u32 aux_send_ctl_flags) { struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct intel_encoder *encoder = &dig_port->base; struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); - enum phy phy = intel_port_to_phy(i915, dig_port->base.port); - bool is_tc_port = intel_phy_is_tc(i915, phy); i915_reg_t ch_ctl, ch_data[5]; u32 aux_clock_divider; enum intel_display_power_domain aux_domain; @@ -245,18 +245,16 @@ intel_dp_aux_xfer(struct intel_dp *intel_dp, for (i = 0; i < ARRAY_SIZE(ch_data); i++) ch_data[i] = intel_dp->aux_ch_data_reg(intel_dp, i); - if (is_tc_port) { - intel_tc_port_lock(dig_port); - /* - * Abort transfers on a disconnected port as required by - * DP 1.4a link CTS 4.2.1.5, also avoiding the long AUX - * timeouts that would otherwise happen. - * TODO: abort the transfer on non-TC ports as well. - */ - if (!intel_tc_port_connected_locked(&dig_port->base)) { - ret = -ENXIO; - goto out_unlock; - } + intel_digital_port_lock(encoder); + /* + * Abort transfers on a disconnected port as required by + * DP 1.4a link CTS 4.2.1.5, also avoiding the long AUX + * timeouts that would otherwise happen. + */ + if (!intel_dp_is_edp(intel_dp) && + !intel_digital_port_connected_locked(&dig_port->base)) { + ret = -ENXIO; + goto out_unlock; } aux_domain = intel_aux_power_domain(dig_port); @@ -423,8 +421,7 @@ out: intel_pps_unlock(intel_dp, pps_wakeref); intel_display_power_put_async(i915, aux_domain, aux_wakeref); out_unlock: - if (is_tc_port) - intel_tc_port_unlock(dig_port); + intel_digital_port_unlock(encoder); return ret; } diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c index 8a9432335030..5fa25a5a36b5 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c @@ -37,6 +37,7 @@ #include "intel_crtc.h" #include "intel_ddi.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_dp_hdcp.h" @@ -1410,6 +1411,9 @@ intel_dp_mst_detect(struct drm_connector *connector, if (drm_connector_is_unregistered(connector)) return connector_status_disconnected; + if (!intel_display_driver_check_access(i915)) + return connector->status; + return drm_dp_mst_detect_port(connector, ctx, &intel_dp->mst_mgr, intel_connector->port); } diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c index ef57dad1a9cb..e7e0a4cf9f93 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c @@ -1263,11 +1263,11 @@ static const struct dpll_info hsw_plls[] = { { .name = "WRPLL 2", .funcs = &hsw_ddi_wrpll_funcs, .id = DPLL_ID_WRPLL2, }, { .name = "SPLL", .funcs = &hsw_ddi_spll_funcs, .id = DPLL_ID_SPLL, }, { .name = "LCPLL 810", .funcs = &hsw_ddi_lcpll_funcs, .id = DPLL_ID_LCPLL_810, - .flags = INTEL_DPLL_ALWAYS_ON, }, + .always_on = true, }, { .name = "LCPLL 1350", .funcs = &hsw_ddi_lcpll_funcs, .id = DPLL_ID_LCPLL_1350, - .flags = INTEL_DPLL_ALWAYS_ON, }, + .always_on = true, }, { .name = "LCPLL 2700", .funcs = &hsw_ddi_lcpll_funcs, .id = DPLL_ID_LCPLL_2700, - .flags = INTEL_DPLL_ALWAYS_ON, }, + .always_on = true, }, {} }; @@ -1945,7 +1945,7 @@ static const struct intel_shared_dpll_funcs skl_ddi_dpll0_funcs = { static const struct dpll_info skl_plls[] = { { .name = "DPLL 0", .funcs = &skl_ddi_dpll0_funcs, .id = DPLL_ID_SKL_DPLL0, - .flags = INTEL_DPLL_ALWAYS_ON, }, + .always_on = true, }, { .name = "DPLL 1", .funcs = &skl_ddi_pll_funcs, .id = DPLL_ID_SKL_DPLL1, }, { .name = "DPLL 2", .funcs = &skl_ddi_pll_funcs, .id = DPLL_ID_SKL_DPLL2, }, { .name = "DPLL 3", .funcs = &skl_ddi_pll_funcs, .id = DPLL_ID_SKL_DPLL3, }, @@ -3308,6 +3308,8 @@ static int icl_compute_tc_phy_dplls(struct intel_atomic_state *state, struct drm_i915_private *i915 = to_i915(state->base.dev); struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); + const struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); struct icl_port_dpll *port_dpll = &crtc_state->icl_port_dplls[ICL_PORT_DPLL_DEFAULT]; struct skl_wrpll_params pll_params = {}; @@ -3326,7 +3328,11 @@ static int icl_compute_tc_phy_dplls(struct intel_atomic_state *state, return ret; /* this is mainly for the fastset check */ - icl_set_active_port_dpll(crtc_state, ICL_PORT_DPLL_MG_PHY); + if (old_crtc_state->shared_dpll && + old_crtc_state->shared_dpll->info->id == DPLL_ID_ICL_TBTPLL) + icl_set_active_port_dpll(crtc_state, ICL_PORT_DPLL_DEFAULT); + else + icl_set_active_port_dpll(crtc_state, ICL_PORT_DPLL_MG_PHY); crtc_state->port_clock = icl_ddi_mg_pll_get_freq(i915, NULL, &port_dpll->hw_state); @@ -4023,7 +4029,8 @@ static const struct intel_shared_dpll_funcs mg_pll_funcs = { static const struct dpll_info icl_plls[] = { { .name = "DPLL 0", .funcs = &combo_pll_funcs, .id = DPLL_ID_ICL_DPLL0, }, { .name = "DPLL 1", .funcs = &combo_pll_funcs, .id = DPLL_ID_ICL_DPLL1, }, - { .name = "TBT PLL", .funcs = &tbt_pll_funcs, .id = DPLL_ID_ICL_TBTPLL, }, + { .name = "TBT PLL", .funcs = &tbt_pll_funcs, .id = DPLL_ID_ICL_TBTPLL, + .is_alt_port_dpll = true, }, { .name = "MG PLL 1", .funcs = &mg_pll_funcs, .id = DPLL_ID_ICL_MGPLL1, }, { .name = "MG PLL 2", .funcs = &mg_pll_funcs, .id = DPLL_ID_ICL_MGPLL2, }, { .name = "MG PLL 3", .funcs = &mg_pll_funcs, .id = DPLL_ID_ICL_MGPLL3, }, @@ -4068,7 +4075,8 @@ static const struct intel_shared_dpll_funcs dkl_pll_funcs = { static const struct dpll_info tgl_plls[] = { { .name = "DPLL 0", .funcs = &combo_pll_funcs, .id = DPLL_ID_ICL_DPLL0, }, { .name = "DPLL 1", .funcs = &combo_pll_funcs, .id = DPLL_ID_ICL_DPLL1, }, - { .name = "TBT PLL", .funcs = &tbt_pll_funcs, .id = DPLL_ID_ICL_TBTPLL, }, + { .name = "TBT PLL", .funcs = &tbt_pll_funcs, .id = DPLL_ID_ICL_TBTPLL, + .is_alt_port_dpll = true, }, { .name = "TC PLL 1", .funcs = &dkl_pll_funcs, .id = DPLL_ID_ICL_MGPLL1, }, { .name = "TC PLL 2", .funcs = &dkl_pll_funcs, .id = DPLL_ID_ICL_MGPLL2, }, { .name = "TC PLL 3", .funcs = &dkl_pll_funcs, .id = DPLL_ID_ICL_MGPLL3, }, @@ -4141,7 +4149,8 @@ static const struct intel_dpll_mgr adls_pll_mgr = { static const struct dpll_info adlp_plls[] = { { .name = "DPLL 0", .funcs = &combo_pll_funcs, .id = DPLL_ID_ICL_DPLL0, }, { .name = "DPLL 1", .funcs = &combo_pll_funcs, .id = DPLL_ID_ICL_DPLL1, }, - { .name = "TBT PLL", .funcs = &tbt_pll_funcs, .id = DPLL_ID_ICL_TBTPLL, }, + { .name = "TBT PLL", .funcs = &tbt_pll_funcs, .id = DPLL_ID_ICL_TBTPLL, + .is_alt_port_dpll = true, }, { .name = "TC PLL 1", .funcs = &dkl_pll_funcs, .id = DPLL_ID_ICL_MGPLL1, }, { .name = "TC PLL 2", .funcs = &dkl_pll_funcs, .id = DPLL_ID_ICL_MGPLL2, }, { .name = "TC PLL 3", .funcs = &dkl_pll_funcs, .id = DPLL_ID_ICL_MGPLL3, }, @@ -4465,31 +4474,29 @@ verify_single_dpll_state(struct drm_i915_private *i915, struct intel_crtc *crtc, const struct intel_crtc_state *new_crtc_state) { - struct intel_dpll_hw_state dpll_hw_state; + struct intel_dpll_hw_state dpll_hw_state = {}; u8 pipe_mask; bool active; - memset(&dpll_hw_state, 0, sizeof(dpll_hw_state)); - - drm_dbg_kms(&i915->drm, "%s\n", pll->info->name); - active = intel_dpll_get_hw_state(i915, pll, &dpll_hw_state); - if (!(pll->info->flags & INTEL_DPLL_ALWAYS_ON)) { + if (!pll->info->always_on) { I915_STATE_WARN(i915, !pll->on && pll->active_mask, - "pll in active use but not on in sw tracking\n"); + "%s: pll in active use but not on in sw tracking\n", + pll->info->name); I915_STATE_WARN(i915, pll->on && !pll->active_mask, - "pll is on but not used by any active pipe\n"); + "%s: pll is on but not used by any active pipe\n", + pll->info->name); I915_STATE_WARN(i915, pll->on != active, - "pll on state mismatch (expected %i, found %i)\n", - pll->on, active); + "%s: pll on state mismatch (expected %i, found %i)\n", + pll->info->name, pll->on, active); } if (!crtc) { I915_STATE_WARN(i915, pll->active_mask & ~pll->state.pipe_mask, - "more active pll users than references: 0x%x vs 0x%x\n", - pll->active_mask, pll->state.pipe_mask); + "%s: more active pll users than references: 0x%x vs 0x%x\n", + pll->info->name, pll->active_mask, pll->state.pipe_mask); return; } @@ -4498,21 +4505,29 @@ verify_single_dpll_state(struct drm_i915_private *i915, if (new_crtc_state->hw.active) I915_STATE_WARN(i915, !(pll->active_mask & pipe_mask), - "pll active mismatch (expected pipe %c in active mask 0x%x)\n", - pipe_name(crtc->pipe), pll->active_mask); + "%s: pll active mismatch (expected pipe %c in active mask 0x%x)\n", + pll->info->name, pipe_name(crtc->pipe), pll->active_mask); else I915_STATE_WARN(i915, pll->active_mask & pipe_mask, - "pll active mismatch (didn't expect pipe %c in active mask 0x%x)\n", - pipe_name(crtc->pipe), pll->active_mask); + "%s: pll active mismatch (didn't expect pipe %c in active mask 0x%x)\n", + pll->info->name, pipe_name(crtc->pipe), pll->active_mask); I915_STATE_WARN(i915, !(pll->state.pipe_mask & pipe_mask), - "pll enabled crtcs mismatch (expected 0x%x in 0x%x)\n", - pipe_mask, pll->state.pipe_mask); + "%s: pll enabled crtcs mismatch (expected 0x%x in 0x%x)\n", + pll->info->name, pipe_mask, pll->state.pipe_mask); I915_STATE_WARN(i915, pll->on && memcmp(&pll->state.hw_state, &dpll_hw_state, sizeof(dpll_hw_state)), - "pll hw state mismatch\n"); + "%s: pll hw state mismatch\n", + pll->info->name); +} + +static bool has_alt_port_dpll(const struct intel_shared_dpll *old_pll, + const struct intel_shared_dpll *new_pll) +{ + return old_pll && new_pll && old_pll != new_pll && + (old_pll->info->is_alt_port_dpll || new_pll->info->is_alt_port_dpll); } void intel_shared_dpll_state_verify(struct intel_atomic_state *state, @@ -4534,11 +4549,15 @@ void intel_shared_dpll_state_verify(struct intel_atomic_state *state, struct intel_shared_dpll *pll = old_crtc_state->shared_dpll; I915_STATE_WARN(i915, pll->active_mask & pipe_mask, - "pll active mismatch (didn't expect pipe %c in active mask (0x%x))\n", - pipe_name(crtc->pipe), pll->active_mask); - I915_STATE_WARN(i915, pll->state.pipe_mask & pipe_mask, - "pll enabled crtcs mismatch (found pipe %c in enabled mask (0x%x))\n", - pipe_name(crtc->pipe), pll->state.pipe_mask); + "%s: pll active mismatch (didn't expect pipe %c in active mask (0x%x))\n", + pll->info->name, pipe_name(crtc->pipe), pll->active_mask); + + /* TC ports have both MG/TC and TBT PLL referenced simultaneously */ + I915_STATE_WARN(i915, !has_alt_port_dpll(old_crtc_state->shared_dpll, + new_crtc_state->shared_dpll) && + pll->state.pipe_mask & pipe_mask, + "%s: pll enabled crtcs mismatch (found pipe %c in enabled mask (0x%x))\n", + pll->info->name, pipe_name(crtc->pipe), pll->state.pipe_mask); } } diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h index 2e7ea0d8d3ff..616afe861b46 100644 --- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.h +++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.h @@ -276,15 +276,21 @@ struct dpll_info { */ enum intel_display_power_domain power_domain; -#define INTEL_DPLL_ALWAYS_ON (1 << 0) /** - * @flags: + * @always_on: * - * INTEL_DPLL_ALWAYS_ON - * Inform the state checker that the DPLL is kept enabled even if - * not in use by any CRTC. + * Inform the state checker that the DPLL is kept enabled even if + * not in use by any CRTC. */ - u32 flags; + bool always_on; + + /** + * @is_alt_port_dpll: + * + * Inform the state checker that the DPLL can be used as a fallback + * (for TC->TBT fallback). + */ + bool is_alt_port_dpll; }; /** diff --git a/drivers/gpu/drm/i915/display/intel_dsb.c b/drivers/gpu/drm/i915/display/intel_dsb.c index 482c28b5c2de..a6c7122fd671 100644 --- a/drivers/gpu/drm/i915/display/intel_dsb.c +++ b/drivers/gpu/drm/i915/display/intel_dsb.c @@ -453,6 +453,10 @@ struct intel_dsb *intel_dsb_prepare(const struct intel_crtc_state *crtc_state, if (!HAS_DSB(i915)) return NULL; + /* TODO: DSB is broken in Xe KMD, so disabling it until fixed */ + if (!IS_ENABLED(I915)) + return NULL; + dsb = kzalloc(sizeof(*dsb), GFP_KERNEL); if (!dsb) goto out; diff --git a/drivers/gpu/drm/i915/display/intel_dvo.c b/drivers/gpu/drm/i915/display/intel_dvo.c index 9111e9d46486..8ca9ae4798a8 100644 --- a/drivers/gpu/drm/i915/display/intel_dvo.c +++ b/drivers/gpu/drm/i915/display/intel_dvo.c @@ -35,6 +35,7 @@ #include "i915_reg.h" #include "intel_connector.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_dvo.h" #include "intel_dvo_dev.h" @@ -328,6 +329,9 @@ intel_dvo_detect(struct drm_connector *_connector, bool force) if (!intel_display_device_enabled(i915)) return connector_status_disconnected; + if (!intel_display_driver_check_access(i915)) + return connector->base.status; + return intel_dvo->dev.dev_ops->detect(&intel_dvo->dev); } @@ -536,6 +540,7 @@ void intel_dvo_init(struct drm_i915_private *i915) if (intel_dvo->dev.type == INTEL_DVO_CHIP_TMDS) connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + connector->base.polled = connector->polled; drm_connector_init_with_ddc(&i915->drm, &connector->base, &intel_dvo_connector_funcs, diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c index f17a1afb4929..b453fcbd67da 100644 --- a/drivers/gpu/drm/i915/display/intel_fbc.c +++ b/drivers/gpu/drm/i915/display/intel_fbc.c @@ -1087,18 +1087,7 @@ static bool i8xx_fbc_tiling_valid(const struct intel_plane_state *plane_state) static bool skl_fbc_tiling_valid(const struct intel_plane_state *plane_state) { - const struct drm_framebuffer *fb = plane_state->hw.fb; - - switch (fb->modifier) { - case DRM_FORMAT_MOD_LINEAR: - case I915_FORMAT_MOD_Y_TILED: - case I915_FORMAT_MOD_Yf_TILED: - case I915_FORMAT_MOD_4_TILED: - case I915_FORMAT_MOD_X_TILED: - return true; - default: - return false; - } + return true; } static bool tiling_is_valid(const struct intel_plane_state *plane_state) diff --git a/drivers/gpu/drm/i915/display/intel_fbdev_fb.c b/drivers/gpu/drm/i915/display/intel_fbdev_fb.c index 717c3a3237c4..0665f943f65f 100644 --- a/drivers/gpu/drm/i915/display/intel_fbdev_fb.c +++ b/drivers/gpu/drm/i915/display/intel_fbdev_fb.c @@ -78,8 +78,9 @@ int intel_fbdev_fb_fill_info(struct drm_i915_private *i915, struct fb_info *info /* Use fbdev's framebuffer from lmem for discrete */ info->fix.smem_start = - (unsigned long)(mem->io_start + - i915_gem_object_get_dma_address(obj, 0)); + (unsigned long)(mem->io.start + + i915_gem_object_get_dma_address(obj, 0) - + mem->region.start); info->fix.smem_len = obj->base.size; } else { struct i915_ggtt *ggtt = to_gt(i915)->ggtt; diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c index e8e8be54143b..cbcd1e91b7be 100644 --- a/drivers/gpu/drm/i915/display/intel_global_state.c +++ b/drivers/gpu/drm/i915/display/intel_global_state.c @@ -10,12 +10,55 @@ #include "intel_display_types.h" #include "intel_global_state.h" +struct intel_global_commit { + struct kref ref; + struct completion done; +}; + +static struct intel_global_commit *commit_new(void) +{ + struct intel_global_commit *commit; + + commit = kzalloc(sizeof(*commit), GFP_KERNEL); + if (!commit) + return NULL; + + init_completion(&commit->done); + kref_init(&commit->ref); + + return commit; +} + +static void __commit_free(struct kref *kref) +{ + struct intel_global_commit *commit = + container_of(kref, typeof(*commit), ref); + + kfree(commit); +} + +static struct intel_global_commit *commit_get(struct intel_global_commit *commit) +{ + if (commit) + kref_get(&commit->ref); + + return commit; +} + +static void commit_put(struct intel_global_commit *commit) +{ + if (commit) + kref_put(&commit->ref, __commit_free); +} + static void __intel_atomic_global_state_free(struct kref *kref) { struct intel_global_state *obj_state = container_of(kref, struct intel_global_state, ref); struct intel_global_obj *obj = obj_state->obj; + commit_put(obj_state->commit); + obj->funcs->atomic_destroy_state(obj, obj_state); } @@ -127,6 +170,8 @@ intel_atomic_get_global_obj_state(struct intel_atomic_state *state, obj_state->obj = obj; obj_state->changed = false; + obj_state->serialized = false; + obj_state->commit = NULL; kref_init(&obj_state->ref); @@ -239,19 +284,13 @@ int intel_atomic_lock_global_state(struct intel_global_state *obj_state) int intel_atomic_serialize_global_state(struct intel_global_state *obj_state) { - struct intel_atomic_state *state = obj_state->state; - struct drm_i915_private *dev_priv = to_i915(state->base.dev); - struct intel_crtc *crtc; + int ret; - for_each_intel_crtc(&dev_priv->drm, crtc) { - struct intel_crtc_state *crtc_state; + ret = intel_atomic_lock_global_state(obj_state); + if (ret) + return ret; - crtc_state = intel_atomic_get_crtc_state(&state->base, crtc); - if (IS_ERR(crtc_state)) - return PTR_ERR(crtc_state); - } - - obj_state->changed = true; + obj_state->serialized = true; return 0; } @@ -267,3 +306,79 @@ intel_atomic_global_state_is_serialized(struct intel_atomic_state *state) return false; return true; } + +int +intel_atomic_global_state_setup_commit(struct intel_atomic_state *state) +{ + const struct intel_global_state *old_obj_state; + struct intel_global_state *new_obj_state; + struct intel_global_obj *obj; + int i; + + for_each_oldnew_global_obj_in_state(state, obj, old_obj_state, + new_obj_state, i) { + struct intel_global_commit *commit = NULL; + + if (new_obj_state->serialized) { + /* + * New commit which is going to be completed + * after the hardware reprogramming is done. + */ + commit = commit_new(); + if (!commit) + return -ENOMEM; + } else if (new_obj_state->changed) { + /* + * We're going to swap to this state, so carry the + * previous commit along, in case it's not yet done. + */ + commit = commit_get(old_obj_state->commit); + } + + new_obj_state->commit = commit; + } + + return 0; +} + +int +intel_atomic_global_state_wait_for_dependencies(struct intel_atomic_state *state) +{ + struct drm_i915_private *i915 = to_i915(state->base.dev); + const struct intel_global_state *old_obj_state; + struct intel_global_obj *obj; + int i; + + for_each_old_global_obj_in_state(state, obj, old_obj_state, i) { + struct intel_global_commit *commit = old_obj_state->commit; + long ret; + + if (!commit) + continue; + + ret = wait_for_completion_timeout(&commit->done, 10 * HZ); + if (ret == 0) { + drm_err(&i915->drm, "global state timed out\n"); + return -ETIMEDOUT; + } + } + + return 0; +} + +void +intel_atomic_global_state_commit_done(struct intel_atomic_state *state) +{ + const struct intel_global_state *new_obj_state; + struct intel_global_obj *obj; + int i; + + for_each_new_global_obj_in_state(state, obj, new_obj_state, i) { + struct intel_global_commit *commit = new_obj_state->commit; + + if (!new_obj_state->serialized) + continue; + + complete_all(&commit->done); + } +} diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h index 5477de8f0b30..5c8545d7a76a 100644 --- a/drivers/gpu/drm/i915/display/intel_global_state.h +++ b/drivers/gpu/drm/i915/display/intel_global_state.h @@ -54,11 +54,14 @@ struct intel_global_obj { (__i)++) \ for_each_if(obj) +struct intel_global_commit; + struct intel_global_state { struct intel_global_obj *obj; struct intel_atomic_state *state; + struct intel_global_commit *commit; struct kref ref; - bool changed; + bool changed, serialized; }; struct __intel_global_objs_state { @@ -87,6 +90,10 @@ void intel_atomic_clear_global_state(struct intel_atomic_state *state); int intel_atomic_lock_global_state(struct intel_global_state *obj_state); int intel_atomic_serialize_global_state(struct intel_global_state *obj_state); +int intel_atomic_global_state_setup_commit(struct intel_atomic_state *state); +void intel_atomic_global_state_commit_done(struct intel_atomic_state *state); +int intel_atomic_global_state_wait_for_dependencies(struct intel_atomic_state *state); + bool intel_atomic_global_state_is_serialized(struct intel_atomic_state *state); #endif diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c index e9e4dcf345f9..d3e03ed5b79c 100644 --- a/drivers/gpu/drm/i915/display/intel_gmbus.c +++ b/drivers/gpu/drm/i915/display/intel_gmbus.c @@ -155,7 +155,7 @@ static const struct gmbus_pin *get_gmbus_pin(struct drm_i915_private *i915, const struct gmbus_pin *pins; size_t size; - if (INTEL_PCH_TYPE(i915) >= PCH_LNL) { + if (INTEL_PCH_TYPE(i915) >= PCH_MTL) { pins = gmbus_pins_mtp; size = ARRAY_SIZE(gmbus_pins_mtp); } else if (INTEL_PCH_TYPE(i915) >= PCH_DG2) { @@ -164,9 +164,6 @@ static const struct gmbus_pin *get_gmbus_pin(struct drm_i915_private *i915, } else if (INTEL_PCH_TYPE(i915) >= PCH_DG1) { pins = gmbus_pins_dg1; size = ARRAY_SIZE(gmbus_pins_dg1); - } else if (INTEL_PCH_TYPE(i915) >= PCH_MTP) { - pins = gmbus_pins_mtp; - size = ARRAY_SIZE(gmbus_pins_mtp); } else if (INTEL_PCH_TYPE(i915) >= PCH_ICP) { pins = gmbus_pins_icp; size = ARRAY_SIZE(gmbus_pins_icp); diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c index 39b3f7c0c77c..c3e692e7f790 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp.c @@ -347,7 +347,7 @@ u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *i915, default: drm_err(&i915->drm, "Unknown transcoder %d\n", cpu_transcoder); - return -EINVAL; + return 0; } } @@ -364,7 +364,7 @@ u32 intel_hdcp_get_repeater_ctl(struct drm_i915_private *i915, return HDCP_DDIE_REP_PRESENT | HDCP_DDIE_SHA1_M0; default: drm_err(&i915->drm, "Unknown port %d\n", port); - return -EINVAL; + return 0; } } @@ -853,8 +853,8 @@ static int intel_hdcp_auth(struct intel_connector *connector) if (shim->stream_encryption) { ret = shim->stream_encryption(connector, true); if (ret) { - drm_err(&i915->drm, "[%s:%d] Failed to enable HDCP 1.4 stream enc\n", - connector->base.name, connector->base.base.id); + drm_err(&i915->drm, "[CONNECTOR:%d:%s] Failed to enable HDCP 1.4 stream enc\n", + connector->base.base.id, connector->base.name); return ret; } drm_dbg_kms(&i915->drm, "HDCP 1.4 transcoder: %s stream encrypted\n", @@ -878,14 +878,14 @@ static int _intel_hdcp_disable(struct intel_connector *connector) u32 repeater_ctl; int ret; - drm_dbg_kms(&i915->drm, "[%s:%d] HDCP is being disabled...\n", - connector->base.name, connector->base.base.id); + drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] HDCP is being disabled...\n", + connector->base.base.id, connector->base.name); if (hdcp->shim->stream_encryption) { ret = hdcp->shim->stream_encryption(connector, false); if (ret) { - drm_err(&i915->drm, "[%s:%d] Failed to disable HDCP 1.4 stream enc\n", - connector->base.name, connector->base.base.id); + drm_err(&i915->drm, "[CONNECTOR:%d:%s] Failed to disable HDCP 1.4 stream enc\n", + connector->base.base.id, connector->base.name); return ret; } drm_dbg_kms(&i915->drm, "HDCP 1.4 transcoder: %s stream encryption disabled\n", @@ -929,8 +929,8 @@ static int intel_hdcp1_enable(struct intel_connector *connector) struct intel_hdcp *hdcp = &connector->hdcp; int i, ret, tries = 3; - drm_dbg_kms(&i915->drm, "[%s:%d] HDCP is being enabled...\n", - connector->base.name, connector->base.base.id); + drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] HDCP is being enabled...\n", + connector->base.base.id, connector->base.name); if (!hdcp_key_loadable(i915)) { drm_err(&i915->drm, "HDCP key Load is not possible\n"); @@ -1027,8 +1027,8 @@ static int intel_hdcp_check_link(struct intel_connector *connector) if (drm_WARN_ON(&i915->drm, !intel_hdcp_in_use(i915, cpu_transcoder, port))) { drm_err(&i915->drm, - "%s:%d HDCP link stopped encryption,%x\n", - connector->base.name, connector->base.base.id, + "[CONNECTOR:%d:%s] HDCP link stopped encryption,%x\n", + connector->base.base.id, connector->base.name, intel_de_read(i915, HDCP_STATUS(i915, cpu_transcoder, port))); ret = -ENXIO; intel_hdcp_update_value(connector, @@ -1046,8 +1046,8 @@ static int intel_hdcp_check_link(struct intel_connector *connector) } drm_dbg_kms(&i915->drm, - "[%s:%d] HDCP link failed, retrying authentication\n", - connector->base.name, connector->base.base.id); + "[CONNECTOR:%d:%s] HDCP link failed, retrying authentication\n", + connector->base.base.id, connector->base.name); ret = _intel_hdcp_disable(connector); if (ret) { @@ -1633,6 +1633,12 @@ int hdcp2_authenticate_repeater_topology(struct intel_connector *connector) !HDCP_2_2_HDCP1_DEVICE_CONNECTED(rx_info[1]) && !HDCP_2_2_HDCP_2_0_REP_CONNECTED(rx_info[1]); + if (!dig_port->hdcp_mst_type1_capable && hdcp->content_type) { + drm_dbg_kms(&i915->drm, + "HDCP1.x or 2.0 Legacy Device Downstream\n"); + return -EINVAL; + } + /* Converting and Storing the seq_num_v to local variable as DWORD */ seq_num_v = drm_hdcp_be24_to_cpu((const u8 *)msgs.recvid_list.seq_num_v); @@ -1731,8 +1737,8 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector) if (!(intel_de_read(i915, HDCP2_STATUS(i915, cpu_transcoder, port)) & LINK_ENCRYPTION_STATUS)) { - drm_err(&i915->drm, "[%s:%d] HDCP 2.2 Link is not encrypted\n", - connector->base.name, connector->base.base.id); + drm_err(&i915->drm, "[CONNECTOR:%d:%s] HDCP 2.2 Link is not encrypted\n", + connector->base.base.id, connector->base.name); ret = -EPERM; goto link_recover; } @@ -1740,8 +1746,8 @@ static int hdcp2_enable_stream_encryption(struct intel_connector *connector) if (hdcp->shim->stream_2_2_encryption) { ret = hdcp->shim->stream_2_2_encryption(connector, true); if (ret) { - drm_err(&i915->drm, "[%s:%d] Failed to enable HDCP 2.2 stream enc\n", - connector->base.name, connector->base.base.id); + drm_err(&i915->drm, "[CONNECTOR:%d:%s] Failed to enable HDCP 2.2 stream enc\n", + connector->base.base.id, connector->base.name); return ret; } drm_dbg_kms(&i915->drm, "HDCP 2.2 transcoder: %s stream encrypted\n", @@ -1925,8 +1931,8 @@ static int _intel_hdcp2_enable(struct intel_connector *connector) struct intel_hdcp *hdcp = &connector->hdcp; int ret; - drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 is being enabled. Type: %d\n", - connector->base.name, connector->base.base.id, + drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] HDCP2.2 is being enabled. Type: %d\n", + connector->base.base.id, connector->base.name, hdcp->content_type); ret = hdcp2_authenticate_and_encrypt(connector); @@ -1936,8 +1942,8 @@ static int _intel_hdcp2_enable(struct intel_connector *connector) return ret; } - drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 is enabled. Type %d\n", - connector->base.name, connector->base.base.id, + drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] HDCP2.2 is enabled. Type %d\n", + connector->base.base.id, connector->base.name, hdcp->content_type); hdcp->hdcp2_encrypted = true; @@ -1953,14 +1959,14 @@ _intel_hdcp2_disable(struct intel_connector *connector, bool hdcp2_link_recovery struct intel_hdcp *hdcp = &connector->hdcp; int ret; - drm_dbg_kms(&i915->drm, "[%s:%d] HDCP2.2 is being Disabled\n", - connector->base.name, connector->base.base.id); + drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s] HDCP2.2 is being Disabled\n", + connector->base.base.id, connector->base.name); if (hdcp->shim->stream_2_2_encryption) { ret = hdcp->shim->stream_2_2_encryption(connector, false); if (ret) { - drm_err(&i915->drm, "[%s:%d] Failed to disable HDCP 2.2 stream enc\n", - connector->base.name, connector->base.base.id); + drm_err(&i915->drm, "[CONNECTOR:%d:%s] Failed to disable HDCP 2.2 stream enc\n", + connector->base.base.id, connector->base.name); return ret; } drm_dbg_kms(&i915->drm, "HDCP 2.2 transcoder: %s stream encryption disabled\n", @@ -2040,20 +2046,20 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) goto out; } drm_dbg_kms(&i915->drm, - "[%s:%d] Repeater topology auth failed.(%d)\n", - connector->base.name, connector->base.base.id, + "[CONNECTOR:%d:%s] Repeater topology auth failed.(%d)\n", + connector->base.base.id, connector->base.name, ret); } else { drm_dbg_kms(&i915->drm, - "[%s:%d] HDCP2.2 link failed, retrying auth\n", - connector->base.name, connector->base.base.id); + "[CONNECTOR:%d:%s] HDCP2.2 link failed, retrying auth\n", + connector->base.base.id, connector->base.name); } ret = _intel_hdcp2_disable(connector, true); if (ret) { drm_err(&i915->drm, - "[%s:%d] Failed to disable hdcp2.2 (%d)\n", - connector->base.name, connector->base.base.id, ret); + "[CONNECTOR:%d:%s] Failed to disable hdcp2.2 (%d)\n", + connector->base.base.id, connector->base.name, ret); intel_hdcp_update_value(connector, DRM_MODE_CONTENT_PROTECTION_DESIRED, true); goto out; @@ -2062,8 +2068,8 @@ static int intel_hdcp2_check_link(struct intel_connector *connector) ret = _intel_hdcp2_enable(connector); if (ret) { drm_dbg_kms(&i915->drm, - "[%s:%d] Failed to enable hdcp2.2 (%d)\n", - connector->base.name, connector->base.base.id, + "[CONNECTOR:%d:%s] Failed to enable hdcp2.2 (%d)\n", + connector->base.base.id, connector->base.name, ret); intel_hdcp_update_value(connector, DRM_MODE_CONTENT_PROTECTION_DESIRED, @@ -2341,8 +2347,8 @@ static int _intel_hdcp_enable(struct intel_atomic_state *state, return -ENOENT; if (!connector->encoder) { - drm_err(&i915->drm, "[%s:%d] encoder is not initialized\n", - connector->base.name, connector->base.base.id); + drm_err(&i915->drm, "[CONNECTOR:%d:%s] encoder is not initialized\n", + connector->base.base.id, connector->base.name); return -ENODEV; } diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c index 18117b789b16..302bff75b06c 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c +++ b/drivers/gpu/drm/i915/display/intel_hdcp_gsc.c @@ -65,7 +65,7 @@ static int intel_hdcp_gsc_initialize_message(struct drm_i915_private *i915, goto out_unmap; } - err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL); + err = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); if (err) goto out_unmap; diff --git a/drivers/gpu/drm/i915/display/intel_hdcp_regs.h b/drivers/gpu/drm/i915/display/intel_hdcp_regs.h index 8023c85c7fa0..a568a457e532 100644 --- a/drivers/gpu/drm/i915/display/intel_hdcp_regs.h +++ b/drivers/gpu/drm/i915/display/intel_hdcp_regs.h @@ -8,6 +8,8 @@ #include "intel_display_reg_defs.h" +#define TRANS_HDCP(__i915) (DISPLAY_VER(__i915) >= 12) + /* HDCP Key Registers */ #define HDCP_KEY_CONF _MMIO(0x66c00) #define HDCP_AKSV_SEND_TRIGGER REG_BIT(31) @@ -82,7 +84,7 @@ #define TRANS_HDCP_CONF(trans) _MMIO_TRANS(trans, _TRANSA_HDCP_CONF, \ _TRANSB_HDCP_CONF) #define HDCP_CONF(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_CONF(trans) : \ PORT_HDCP_CONF(port)) @@ -95,7 +97,7 @@ _TRANSA_HDCP_ANINIT, \ _TRANSB_HDCP_ANINIT) #define HDCP_ANINIT(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_ANINIT(trans) : \ PORT_HDCP_ANINIT(port)) @@ -105,7 +107,7 @@ #define TRANS_HDCP_ANLO(trans) _MMIO_TRANS(trans, _TRANSA_HDCP_ANLO, \ _TRANSB_HDCP_ANLO) #define HDCP_ANLO(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_ANLO(trans) : \ PORT_HDCP_ANLO(port)) @@ -115,7 +117,7 @@ #define TRANS_HDCP_ANHI(trans) _MMIO_TRANS(trans, _TRANSA_HDCP_ANHI, \ _TRANSB_HDCP_ANHI) #define HDCP_ANHI(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_ANHI(trans) : \ PORT_HDCP_ANHI(port)) @@ -126,7 +128,7 @@ _TRANSA_HDCP_BKSVLO, \ _TRANSB_HDCP_BKSVLO) #define HDCP_BKSVLO(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_BKSVLO(trans) : \ PORT_HDCP_BKSVLO(port)) @@ -137,7 +139,7 @@ _TRANSA_HDCP_BKSVHI, \ _TRANSB_HDCP_BKSVHI) #define HDCP_BKSVHI(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_BKSVHI(trans) : \ PORT_HDCP_BKSVHI(port)) @@ -148,7 +150,7 @@ _TRANSA_HDCP_RPRIME, \ _TRANSB_HDCP_RPRIME) #define HDCP_RPRIME(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_RPRIME(trans) : \ PORT_HDCP_RPRIME(port)) @@ -159,7 +161,7 @@ _TRANSA_HDCP_STATUS, \ _TRANSB_HDCP_STATUS) #define HDCP_STATUS(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP_STATUS(trans) : \ PORT_HDCP_STATUS(port)) @@ -200,7 +202,7 @@ #define AUTH_FORCE_CLR_INPUTCTR REG_BIT(19) #define AUTH_CLR_KEYS REG_BIT(18) #define HDCP2_AUTH(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP2_AUTH(trans) : \ PORT_HDCP2_AUTH(port)) @@ -211,7 +213,7 @@ _TRANSB_HDCP2_CTL) #define CTL_LINK_ENCRYPTION_REQ REG_BIT(31) #define HDCP2_CTL(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP2_CTL(trans) : \ PORT_HDCP2_CTL(port)) @@ -225,7 +227,7 @@ #define LINK_AUTH_STATUS REG_BIT(21) #define LINK_ENCRYPTION_STATUS REG_BIT(20) #define HDCP2_STATUS(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP2_STATUS(trans) : \ PORT_HDCP2_STATUS(port)) @@ -247,7 +249,7 @@ #define STREAM_ENCRYPTION_STATUS REG_BIT(31) #define STREAM_TYPE_STATUS REG_BIT(30) #define HDCP2_STREAM_STATUS(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP2_STREAM_STATUS(trans) : \ PIPE_HDCP2_STREAM_STATUS(pipe)) @@ -263,7 +265,7 @@ _TRANSB_HDCP2_AUTH_STREAM) #define AUTH_STREAM_TYPE REG_BIT(31) #define HDCP2_AUTH_STREAM(dev_priv, trans, port) \ - (GRAPHICS_VER(dev_priv) >= 12 ? \ + (TRANS_HDCP(dev_priv) ? \ TRANS_HDCP2_AUTH_STREAM(trans) : \ PORT_HDCP2_AUTH_STREAM(port)) diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c index 39e4f5f7c817..7020e5806109 100644 --- a/drivers/gpu/drm/i915/display/intel_hdmi.c +++ b/drivers/gpu/drm/i915/display/intel_hdmi.c @@ -49,6 +49,7 @@ #include "intel_cx0_phy.h" #include "intel_ddi.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_dp.h" #include "intel_gmbus.h" @@ -523,10 +524,12 @@ void hsw_write_infoframe(struct intel_encoder *encoder, 0); /* Wa_14013475917 */ - if (IS_DISPLAY_VER(dev_priv, 13, 14) && crtc_state->has_psr && type == DP_SDP_VSC) - return; + if (!(IS_DISPLAY_VER(dev_priv, 13, 14) && crtc_state->has_psr && type == DP_SDP_VSC)) + val |= hsw_infoframe_enable(type); + + if (type == DP_SDP_VSC) + val |= VSC_DIP_HW_DATA_SW_HEA; - val |= hsw_infoframe_enable(type); intel_de_write(dev_priv, ctl_reg, val); intel_de_posting_read(dev_priv, ctl_reg); } @@ -2503,6 +2506,9 @@ intel_hdmi_detect(struct drm_connector *connector, bool force) if (!intel_display_device_enabled(dev_priv)) return connector_status_disconnected; + if (!intel_display_driver_check_access(dev_priv)) + return connector->status; + wakeref = intel_display_power_get(dev_priv, POWER_DOMAIN_GMBUS); if (DISPLAY_VER(dev_priv) >= 11 && @@ -2531,6 +2537,9 @@ intel_hdmi_force(struct drm_connector *connector) drm_dbg_kms(&i915->drm, "[CONNECTOR:%d:%s]\n", connector->base.id, connector->name); + if (!intel_display_driver_check_access(i915)) + return; + intel_hdmi_unset_edid(connector); if (connector->status != connector_status_connected) @@ -3015,6 +3024,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *dig_port, connector->ycbcr_420_allowed = true; intel_connector->polled = DRM_CONNECTOR_POLL_HPD; + intel_connector->base.polled = intel_connector->polled; if (HAS_DDI(dev_priv)) intel_connector->get_hw_state = intel_ddi_connector_get_hw_state; diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c index 0c0700c6ec66..d9ec349f3c8c 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug.c @@ -177,6 +177,46 @@ static bool intel_hpd_irq_storm_detect(struct drm_i915_private *dev_priv, return storm; } +static bool detection_work_enabled(struct drm_i915_private *i915) +{ + lockdep_assert_held(&i915->irq_lock); + + return i915->display.hotplug.detection_work_enabled; +} + +static bool +mod_delayed_detection_work(struct drm_i915_private *i915, struct delayed_work *work, int delay) +{ + lockdep_assert_held(&i915->irq_lock); + + if (!detection_work_enabled(i915)) + return false; + + return mod_delayed_work(i915->unordered_wq, work, delay); +} + +static bool +queue_delayed_detection_work(struct drm_i915_private *i915, struct delayed_work *work, int delay) +{ + lockdep_assert_held(&i915->irq_lock); + + if (!detection_work_enabled(i915)) + return false; + + return queue_delayed_work(i915->unordered_wq, work, delay); +} + +static bool +queue_detection_work(struct drm_i915_private *i915, struct work_struct *work) +{ + lockdep_assert_held(&i915->irq_lock); + + if (!detection_work_enabled(i915)) + return false; + + return queue_work(i915->unordered_wq, work); +} + static void intel_hpd_irq_storm_switch_to_polling(struct drm_i915_private *dev_priv) { @@ -213,9 +253,9 @@ 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_reschedule(&dev_priv->drm); - mod_delayed_work(dev_priv->unordered_wq, - &dev_priv->display.hotplug.reenable_work, - msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); + mod_delayed_detection_work(dev_priv, + &dev_priv->display.hotplug.reenable_work, + msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); } } @@ -348,9 +388,9 @@ static void i915_digport_work_func(struct work_struct *work) if (old_bits) { spin_lock_irq(&dev_priv->irq_lock); dev_priv->display.hotplug.event_bits |= old_bits; + queue_delayed_detection_work(dev_priv, + &dev_priv->display.hotplug.hotplug_work, 0); spin_unlock_irq(&dev_priv->irq_lock); - queue_delayed_work(dev_priv->unordered_wq, - &dev_priv->display.hotplug.hotplug_work, 0); } } @@ -467,11 +507,11 @@ static void i915_hotplug_work_func(struct work_struct *work) if (retry) { spin_lock_irq(&dev_priv->irq_lock); dev_priv->display.hotplug.retry_bits |= retry; - spin_unlock_irq(&dev_priv->irq_lock); - mod_delayed_work(dev_priv->unordered_wq, - &dev_priv->display.hotplug.hotplug_work, - msecs_to_jiffies(HPD_RETRY_DELAY)); + mod_delayed_detection_work(dev_priv, + &dev_priv->display.hotplug.hotplug_work, + msecs_to_jiffies(HPD_RETRY_DELAY)); + spin_unlock_irq(&dev_priv->irq_lock); } } @@ -590,7 +630,6 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, */ if (storm_detected) intel_hpd_irq_setup(dev_priv); - spin_unlock(&dev_priv->irq_lock); /* * Our hotplug handler can grab modeset locks (by calling down into the @@ -601,8 +640,10 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, if (queue_dig) queue_work(dev_priv->display.hotplug.dp_wq, &dev_priv->display.hotplug.dig_port_work); if (queue_hp) - queue_delayed_work(dev_priv->unordered_wq, - &dev_priv->display.hotplug.hotplug_work, 0); + queue_delayed_detection_work(dev_priv, + &dev_priv->display.hotplug.hotplug_work, 0); + + spin_unlock(&dev_priv->irq_lock); } /** @@ -710,6 +751,8 @@ static void i915_hpd_poll_init_work(struct work_struct *work) cancel_work(&dev_priv->display.hotplug.poll_init_work); } + spin_lock_irq(&dev_priv->irq_lock); + drm_connector_list_iter_begin(&dev_priv->drm, &conn_iter); for_each_intel_connector_iter(connector, &conn_iter) { enum hpd_pin pin; @@ -718,6 +761,9 @@ static void i915_hpd_poll_init_work(struct work_struct *work) if (pin == HPD_NONE) continue; + if (dev_priv->display.hotplug.stats[pin].state == HPD_DISABLED) + continue; + connector->base.polled = connector->polled; if (enabled && connector->base.polled == DRM_CONNECTOR_POLL_HPD) @@ -726,6 +772,8 @@ static void i915_hpd_poll_init_work(struct work_struct *work) } drm_connector_list_iter_end(&conn_iter); + spin_unlock_irq(&dev_priv->irq_lock); + if (enabled) drm_kms_helper_poll_reschedule(&dev_priv->drm); @@ -774,8 +822,10 @@ void intel_hpd_poll_enable(struct drm_i915_private *dev_priv) * As well, there's no issue if we race here since we always reschedule * this worker anyway */ - queue_work(dev_priv->unordered_wq, - &dev_priv->display.hotplug.poll_init_work); + spin_lock_irq(&dev_priv->irq_lock); + queue_detection_work(dev_priv, + &dev_priv->display.hotplug.poll_init_work); + spin_unlock_irq(&dev_priv->irq_lock); } /** @@ -803,8 +853,11 @@ void intel_hpd_poll_disable(struct drm_i915_private *dev_priv) return; WRITE_ONCE(dev_priv->display.hotplug.poll_enabled, false); - queue_work(dev_priv->unordered_wq, - &dev_priv->display.hotplug.poll_init_work); + + spin_lock_irq(&dev_priv->irq_lock); + queue_detection_work(dev_priv, + &dev_priv->display.hotplug.poll_init_work); + spin_unlock_irq(&dev_priv->irq_lock); } void intel_hpd_init_early(struct drm_i915_private *i915) @@ -826,6 +879,20 @@ void intel_hpd_init_early(struct drm_i915_private *i915) i915->display.hotplug.hpd_short_storm_enabled = !HAS_DP_MST(i915); } +static bool cancel_all_detection_work(struct drm_i915_private *i915) +{ + bool was_pending = false; + + if (cancel_delayed_work_sync(&i915->display.hotplug.hotplug_work)) + was_pending = true; + if (cancel_work_sync(&i915->display.hotplug.poll_init_work)) + was_pending = true; + if (cancel_delayed_work_sync(&i915->display.hotplug.reenable_work)) + was_pending = true; + + return was_pending; +} + void intel_hpd_cancel_work(struct drm_i915_private *dev_priv) { if (!HAS_DISPLAY(dev_priv)) @@ -841,9 +908,13 @@ void intel_hpd_cancel_work(struct drm_i915_private *dev_priv) spin_unlock_irq(&dev_priv->irq_lock); cancel_work_sync(&dev_priv->display.hotplug.dig_port_work); - cancel_delayed_work_sync(&dev_priv->display.hotplug.hotplug_work); - cancel_work_sync(&dev_priv->display.hotplug.poll_init_work); - cancel_delayed_work_sync(&dev_priv->display.hotplug.reenable_work); + + /* + * All other work triggered by hotplug events should be canceled by + * now. + */ + if (cancel_all_detection_work(dev_priv)) + drm_dbg_kms(&dev_priv->drm, "Hotplug detection work still active\n"); } bool intel_hpd_disable(struct drm_i915_private *dev_priv, enum hpd_pin pin) @@ -873,6 +944,62 @@ void intel_hpd_enable(struct drm_i915_private *dev_priv, enum hpd_pin pin) spin_unlock_irq(&dev_priv->irq_lock); } +static void queue_work_for_missed_irqs(struct drm_i915_private *i915) +{ + bool queue_work = false; + enum hpd_pin pin; + + lockdep_assert_held(&i915->irq_lock); + + if (i915->display.hotplug.event_bits || + i915->display.hotplug.retry_bits) + queue_work = true; + + for_each_hpd_pin(pin) { + switch (i915->display.hotplug.stats[pin].state) { + case HPD_MARK_DISABLED: + queue_work = true; + break; + case HPD_ENABLED: + break; + default: + MISSING_CASE(i915->display.hotplug.stats[pin].state); + } + } + + if (queue_work) + queue_delayed_detection_work(i915, &i915->display.hotplug.hotplug_work, 0); +} + +void intel_hpd_enable_detection_work(struct drm_i915_private *i915) +{ + spin_lock_irq(&i915->irq_lock); + i915->display.hotplug.detection_work_enabled = true; + queue_work_for_missed_irqs(i915); + spin_unlock_irq(&i915->irq_lock); +} + +void intel_hpd_disable_detection_work(struct drm_i915_private *i915) +{ + spin_lock_irq(&i915->irq_lock); + i915->display.hotplug.detection_work_enabled = false; + spin_unlock_irq(&i915->irq_lock); + + cancel_all_detection_work(i915); +} + +bool intel_hpd_schedule_detection(struct drm_i915_private *i915) +{ + unsigned long flags; + bool ret; + + spin_lock_irqsave(&i915->irq_lock, flags); + ret = queue_delayed_detection_work(i915, &i915->display.hotplug.hotplug_work, 0); + spin_unlock_irqrestore(&i915->irq_lock, flags); + + return ret; +} + static int i915_hpd_storm_ctl_show(struct seq_file *m, void *data) { struct drm_i915_private *dev_priv = m->private; diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.h b/drivers/gpu/drm/i915/display/intel_hotplug.h index 424ae5dbf5a0..a17253ddec83 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug.h +++ b/drivers/gpu/drm/i915/display/intel_hotplug.h @@ -30,4 +30,8 @@ bool intel_hpd_disable(struct drm_i915_private *dev_priv, enum hpd_pin pin); void intel_hpd_enable(struct drm_i915_private *dev_priv, enum hpd_pin pin); void intel_hpd_debugfs_register(struct drm_i915_private *i915); +void intel_hpd_enable_detection_work(struct drm_i915_private *i915); +void intel_hpd_disable_detection_work(struct drm_i915_private *i915); +bool intel_hpd_schedule_detection(struct drm_i915_private *i915); + #endif /* __INTEL_HOTPLUG_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c index 04f62f27ad74..76076509f771 100644 --- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c +++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c @@ -163,12 +163,10 @@ static void intel_hpd_init_pins(struct drm_i915_private *dev_priv) (!HAS_PCH_SPLIT(dev_priv) || HAS_PCH_NOP(dev_priv))) return; - if (INTEL_PCH_TYPE(dev_priv) >= PCH_LNL) + if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTL) hpd->pch_hpd = hpd_mtp; else if (INTEL_PCH_TYPE(dev_priv) >= PCH_DG1) hpd->pch_hpd = hpd_sde_dg1; - else if (INTEL_PCH_TYPE(dev_priv) >= PCH_MTP) - hpd->pch_hpd = hpd_mtp; else if (INTEL_PCH_TYPE(dev_priv) >= PCH_ICP) hpd->pch_hpd = hpd_icp; else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_SPT(dev_priv)) @@ -1139,7 +1137,7 @@ static void xelpdp_hpd_irq_setup(struct drm_i915_private *i915) if (INTEL_PCH_TYPE(i915) >= PCH_LNL) xe2lpd_sde_hpd_irq_setup(i915); - else if (INTEL_PCH_TYPE(i915) >= PCH_MTP) + else if (INTEL_PCH_TYPE(i915) >= PCH_MTL) mtp_hpd_irq_setup(i915); } diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c index 1ce785db6a5e..f242bb320610 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.c +++ b/drivers/gpu/drm/i915/display/intel_opregion.c @@ -250,11 +250,36 @@ struct opregion_asle_ext { #define MAX_DSLP 1500 +#define OPREGION_SIZE (8 * 1024) + +struct intel_opregion { + struct drm_i915_private *i915; + + struct opregion_header *header; + struct opregion_acpi *acpi; + struct opregion_swsci *swsci; + u32 swsci_gbda_sub_functions; + u32 swsci_sbcb_sub_functions; + struct opregion_asle *asle; + struct opregion_asle_ext *asle_ext; + void *rvda; + void *vbt_firmware; + const void *vbt; + u32 vbt_size; + struct work_struct asle_work; + struct notifier_block acpi_notifier; +}; + static int check_swsci_function(struct drm_i915_private *i915, u32 function) { - struct opregion_swsci *swsci = i915->display.opregion.swsci; + struct intel_opregion *opregion = i915->display.opregion; + struct opregion_swsci *swsci; u32 main_function, sub_function; + if (!opregion) + return -ENODEV; + + swsci = opregion->swsci; if (!swsci) return -ENODEV; @@ -265,11 +290,11 @@ static int check_swsci_function(struct drm_i915_private *i915, u32 function) /* Check if we can call the function. See swsci_setup for details. */ if (main_function == SWSCI_SBCB) { - if ((i915->display.opregion.swsci_sbcb_sub_functions & + if ((opregion->swsci_sbcb_sub_functions & (1 << sub_function)) == 0) return -EINVAL; } else if (main_function == SWSCI_GBDA) { - if ((i915->display.opregion.swsci_gbda_sub_functions & + if ((opregion->swsci_gbda_sub_functions & (1 << sub_function)) == 0) return -EINVAL; } @@ -280,7 +305,7 @@ static int check_swsci_function(struct drm_i915_private *i915, u32 function) static int swsci(struct drm_i915_private *dev_priv, u32 function, u32 parm, u32 *parm_out) { - struct opregion_swsci *swsci = dev_priv->display.opregion.swsci; + struct opregion_swsci *swsci; struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); u32 scic, dslp; u16 swsci_val; @@ -290,6 +315,8 @@ static int swsci(struct drm_i915_private *dev_priv, if (ret) return ret; + swsci = dev_priv->display.opregion->swsci; + /* Driver sleep timeout in ms. */ dslp = swsci->dslp; if (!dslp) { @@ -462,7 +489,7 @@ static u32 asle_set_backlight(struct drm_i915_private *dev_priv, u32 bclp) { struct intel_connector *connector; struct drm_connector_list_iter conn_iter; - struct opregion_asle *asle = dev_priv->display.opregion.asle; + struct opregion_asle *asle = dev_priv->display.opregion->asle; drm_dbg(&dev_priv->drm, "bclp = 0x%08x\n", bclp); @@ -584,9 +611,8 @@ static void asle_work(struct work_struct *work) { struct intel_opregion *opregion = container_of(work, struct intel_opregion, asle_work); - struct drm_i915_private *dev_priv = - container_of(opregion, struct drm_i915_private, display.opregion); - struct opregion_asle *asle = dev_priv->display.opregion.asle; + struct drm_i915_private *dev_priv = opregion->i915; + struct opregion_asle *asle = opregion->asle; u32 aslc_stat = 0; u32 aslc_req; @@ -632,11 +658,17 @@ static void asle_work(struct work_struct *work) asle->aslc = aslc_stat; } -void intel_opregion_asle_intr(struct drm_i915_private *dev_priv) +bool intel_opregion_asle_present(struct drm_i915_private *i915) +{ + return i915->display.opregion && i915->display.opregion->asle; +} + +void intel_opregion_asle_intr(struct drm_i915_private *i915) { - if (dev_priv->display.opregion.asle) - queue_work(dev_priv->unordered_wq, - &dev_priv->display.opregion.asle_work); + struct intel_opregion *opregion = i915->display.opregion; + + if (opregion && opregion->asle) + queue_work(i915->unordered_wq, &opregion->asle_work); } #define ACPI_EV_DISPLAY_SWITCH (1<<0) @@ -692,7 +724,7 @@ static void set_did(struct intel_opregion *opregion, int i, u32 val) static void intel_didl_outputs(struct drm_i915_private *dev_priv) { - struct intel_opregion *opregion = &dev_priv->display.opregion; + struct intel_opregion *opregion = dev_priv->display.opregion; struct intel_connector *connector; struct drm_connector_list_iter conn_iter; int i = 0, max_outputs; @@ -731,7 +763,7 @@ static void intel_didl_outputs(struct drm_i915_private *dev_priv) static void intel_setup_cadls(struct drm_i915_private *dev_priv) { - struct intel_opregion *opregion = &dev_priv->display.opregion; + struct intel_opregion *opregion = dev_priv->display.opregion; struct intel_connector *connector; struct drm_connector_list_iter conn_iter; int i = 0; @@ -761,7 +793,7 @@ static void intel_setup_cadls(struct drm_i915_private *dev_priv) static void swsci_setup(struct drm_i915_private *dev_priv) { - struct intel_opregion *opregion = &dev_priv->display.opregion; + struct intel_opregion *opregion = dev_priv->display.opregion; bool requested_callbacks = false; u32 tmp; @@ -839,7 +871,7 @@ static const struct dmi_system_id intel_no_opregion_vbt[] = { static int intel_load_vbt_firmware(struct drm_i915_private *dev_priv) { - struct intel_opregion *opregion = &dev_priv->display.opregion; + struct intel_opregion *opregion = dev_priv->display.opregion; const struct firmware *fw = NULL; const char *name = dev_priv->display.params.vbt_firmware; int ret; @@ -879,7 +911,7 @@ static int intel_load_vbt_firmware(struct drm_i915_private *dev_priv) int intel_opregion_setup(struct drm_i915_private *dev_priv) { - struct intel_opregion *opregion = &dev_priv->display.opregion; + struct intel_opregion *opregion; struct pci_dev *pdev = to_pci_dev(dev_priv->drm.dev); u32 asls, mboxes; char buf[sizeof(OPREGION_SIGNATURE)]; @@ -902,11 +934,20 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv) return -ENOTSUPP; } + opregion = kzalloc(sizeof(*opregion), GFP_KERNEL); + if (!opregion) + return -ENOMEM; + + opregion->i915 = dev_priv; + dev_priv->display.opregion = opregion; + INIT_WORK(&opregion->asle_work, asle_work); base = memremap(asls, OPREGION_SIZE, MEMREMAP_WB); - if (!base) - return -ENOMEM; + if (!base) { + err = -ENOMEM; + goto err_memremap; + } memcpy(buf, base, sizeof(buf)); @@ -916,7 +957,6 @@ int intel_opregion_setup(struct drm_i915_private *dev_priv) goto err_out; } opregion->header = base; - opregion->lid_state = base + ACPI_CLID; drm_dbg(&dev_priv->drm, "ACPI OpRegion version %u.%u.%u\n", opregion->header->over.major, @@ -1034,6 +1074,10 @@ out: err_out: memunmap(base); +err_memremap: + kfree(opregion); + dev_priv->display.opregion = NULL; + return err; } @@ -1106,12 +1150,12 @@ const struct drm_edid *intel_opregion_get_edid(struct intel_connector *intel_con { struct drm_connector *connector = &intel_connector->base; struct drm_i915_private *i915 = to_i915(connector->dev); - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; const struct drm_edid *drm_edid; const void *edid; int len; - if (!opregion->asle_ext) + if (!opregion || !opregion->asle_ext) return NULL; edid = opregion->asle_ext->bddc; @@ -1132,10 +1176,28 @@ const struct drm_edid *intel_opregion_get_edid(struct intel_connector *intel_con return drm_edid; } +const void *intel_opregion_get_vbt(struct drm_i915_private *i915, size_t *size) +{ + struct intel_opregion *opregion = i915->display.opregion; + + if (!opregion || !opregion->vbt) + return NULL; + + if (size) + *size = opregion->vbt_size; + + return opregion->vbt; +} + bool intel_opregion_headless_sku(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; - struct opregion_header *header = opregion->header; + struct intel_opregion *opregion = i915->display.opregion; + struct opregion_header *header; + + if (!opregion) + return false; + + header = opregion->header; if (!header || header->over.major < 2 || (header->over.major == 2 && header->over.minor < 3)) @@ -1146,9 +1208,9 @@ bool intel_opregion_headless_sku(struct drm_i915_private *i915) void intel_opregion_register(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; - if (!opregion->header) + if (!opregion) return; if (opregion->acpi) { @@ -1162,7 +1224,7 @@ void intel_opregion_register(struct drm_i915_private *i915) static void intel_opregion_resume_display(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; if (opregion->acpi) { intel_didl_outputs(i915); @@ -1188,9 +1250,9 @@ static void intel_opregion_resume_display(struct drm_i915_private *i915) void intel_opregion_resume(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; - if (!opregion->header) + if (!opregion) return; if (HAS_DISPLAY(i915)) @@ -1201,12 +1263,12 @@ void intel_opregion_resume(struct drm_i915_private *i915) static void intel_opregion_suspend_display(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; if (opregion->asle) opregion->asle->ardy = ASLE_ARDY_NOT_READY; - cancel_work_sync(&i915->display.opregion.asle_work); + cancel_work_sync(&opregion->asle_work); if (opregion->acpi) opregion->acpi->drdy = 0; @@ -1214,9 +1276,9 @@ static void intel_opregion_suspend_display(struct drm_i915_private *i915) void intel_opregion_suspend(struct drm_i915_private *i915, pci_power_t state) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; - if (!opregion->header) + if (!opregion) return; intel_opregion_notify_adapter(i915, state); @@ -1227,11 +1289,11 @@ void intel_opregion_suspend(struct drm_i915_private *i915, pci_power_t state) void intel_opregion_unregister(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; intel_opregion_suspend(i915, PCI_D1); - if (!opregion->header) + if (!opregion) return; if (opregion->acpi_notifier.notifier_call) { @@ -1242,26 +1304,36 @@ void intel_opregion_unregister(struct drm_i915_private *i915) void intel_opregion_cleanup(struct drm_i915_private *i915) { - struct intel_opregion *opregion = &i915->display.opregion; + struct intel_opregion *opregion = i915->display.opregion; - if (!opregion->header) + if (!opregion) return; - /* just clear all opregion memory pointers now */ memunmap(opregion->header); - if (opregion->rvda) { + if (opregion->rvda) memunmap(opregion->rvda); - opregion->rvda = NULL; - } - if (opregion->vbt_firmware) { - kfree(opregion->vbt_firmware); - opregion->vbt_firmware = NULL; - } - opregion->header = NULL; - opregion->acpi = NULL; - opregion->swsci = NULL; - opregion->asle = NULL; - opregion->asle_ext = NULL; - opregion->vbt = NULL; - opregion->lid_state = NULL; + kfree(opregion->vbt_firmware); + kfree(opregion); + i915->display.opregion = NULL; +} + +static int intel_opregion_show(struct seq_file *m, void *unused) +{ + struct drm_i915_private *i915 = m->private; + struct intel_opregion *opregion = i915->display.opregion; + + if (opregion) + seq_write(m, opregion->header, OPREGION_SIZE); + + return 0; +} + +DEFINE_SHOW_ATTRIBUTE(intel_opregion); + +void intel_opregion_debugfs_register(struct drm_i915_private *i915) +{ + struct drm_minor *minor = i915->drm.primary; + + debugfs_create_file("i915_opregion", 0444, minor->debugfs_root, + i915, &intel_opregion_fops); } diff --git a/drivers/gpu/drm/i915/display/intel_opregion.h b/drivers/gpu/drm/i915/display/intel_opregion.h index fd2ea8ef0fa2..0bec224f711f 100644 --- a/drivers/gpu/drm/i915/display/intel_opregion.h +++ b/drivers/gpu/drm/i915/display/intel_opregion.h @@ -25,38 +25,13 @@ #ifndef _INTEL_OPREGION_H_ #define _INTEL_OPREGION_H_ -#include <linux/workqueue.h> #include <linux/pci.h> +#include <linux/types.h> struct drm_i915_private; struct intel_connector; struct intel_encoder; -struct opregion_header; -struct opregion_acpi; -struct opregion_swsci; -struct opregion_asle; -struct opregion_asle_ext; - -struct intel_opregion { - struct opregion_header *header; - struct opregion_acpi *acpi; - struct opregion_swsci *swsci; - u32 swsci_gbda_sub_functions; - u32 swsci_sbcb_sub_functions; - struct opregion_asle *asle; - struct opregion_asle_ext *asle_ext; - void *rvda; - void *vbt_firmware; - const void *vbt; - u32 vbt_size; - u32 *lid_state; - struct work_struct asle_work; - struct notifier_block acpi_notifier; -}; - -#define OPREGION_SIZE (8 * 1024) - #ifdef CONFIG_ACPI int intel_opregion_setup(struct drm_i915_private *dev_priv); @@ -69,6 +44,7 @@ void intel_opregion_resume(struct drm_i915_private *dev_priv); void intel_opregion_suspend(struct drm_i915_private *dev_priv, pci_power_t state); +bool intel_opregion_asle_present(struct drm_i915_private *i915); void intel_opregion_asle_intr(struct drm_i915_private *dev_priv); int intel_opregion_notify_encoder(struct intel_encoder *intel_encoder, bool enable); @@ -77,8 +53,12 @@ int intel_opregion_notify_adapter(struct drm_i915_private *dev_priv, int intel_opregion_get_panel_type(struct drm_i915_private *dev_priv); const struct drm_edid *intel_opregion_get_edid(struct intel_connector *connector); +const void *intel_opregion_get_vbt(struct drm_i915_private *i915, size_t *size); + bool intel_opregion_headless_sku(struct drm_i915_private *i915); +void intel_opregion_debugfs_register(struct drm_i915_private *i915); + #else /* CONFIG_ACPI*/ static inline int intel_opregion_setup(struct drm_i915_private *dev_priv) @@ -107,6 +87,11 @@ static inline void intel_opregion_suspend(struct drm_i915_private *dev_priv, { } +static inline bool intel_opregion_asle_present(struct drm_i915_private *i915) +{ + return false; +} + static inline void intel_opregion_asle_intr(struct drm_i915_private *dev_priv) { } @@ -134,11 +119,21 @@ intel_opregion_get_edid(struct intel_connector *connector) return NULL; } +static inline const void * +intel_opregion_get_vbt(struct drm_i915_private *i915, size_t *size) +{ + return NULL; +} + static inline bool intel_opregion_headless_sku(struct drm_i915_private *i915) { return false; } +static inline void intel_opregion_debugfs_register(struct drm_i915_private *i915) +{ +} + #endif /* CONFIG_ACPI */ #endif diff --git a/drivers/gpu/drm/i915/display/intel_panel.c b/drivers/gpu/drm/i915/display/intel_panel.c index 0d8e5320a4f8..073ea3166c36 100644 --- a/drivers/gpu/drm/i915/display/intel_panel.c +++ b/drivers/gpu/drm/i915/display/intel_panel.c @@ -37,6 +37,7 @@ #include "intel_backlight.h" #include "intel_connector.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_drrs.h" #include "intel_lvds_regs.h" @@ -683,6 +684,9 @@ intel_panel_detect(struct drm_connector *connector, bool force) if (!intel_display_device_enabled(i915)) return connector_status_disconnected; + if (!intel_display_driver_check_access(i915)) + return connector->status; + return connector_status_connected; } diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c index a55c09cbd0e4..ada1792df5b3 100644 --- a/drivers/gpu/drm/i915/display/intel_plane_initial.c +++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c @@ -3,9 +3,11 @@ * Copyright © 2021 Intel Corporation */ +#include "gem/i915_gem_lmem.h" #include "gem/i915_gem_region.h" #include "i915_drv.h" #include "intel_atomic_plane.h" +#include "intel_crtc.h" #include "intel_display.h" #include "intel_display_types.h" #include "intel_fb.h" @@ -13,20 +15,21 @@ #include "intel_plane_initial.h" static bool -intel_reuse_initial_plane_obj(struct drm_i915_private *i915, - const struct intel_initial_plane_config *plane_config, +intel_reuse_initial_plane_obj(struct intel_crtc *this, + const struct intel_initial_plane_config plane_configs[], struct drm_framebuffer **fb, struct i915_vma **vma) { + struct drm_i915_private *i915 = to_i915(this->base.dev); struct intel_crtc *crtc; for_each_intel_crtc(&i915->drm, crtc) { - struct intel_crtc_state *crtc_state = - to_intel_crtc_state(crtc->base.state); struct intel_plane *plane = to_intel_plane(crtc->base.primary); - struct intel_plane_state *plane_state = + const struct intel_plane_state *plane_state = to_intel_plane_state(plane->base.state); + const struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); if (!crtc_state->uapi.active) continue; @@ -34,7 +37,7 @@ intel_reuse_initial_plane_obj(struct drm_i915_private *i915, if (!plane_state->ggtt_vma) continue; - if (intel_plane_ggtt_offset(plane_state) == plane_config->base) { + if (plane_configs[this->pipe].base == plane_configs[crtc->pipe].base) { *fb = plane_state->hw.fb; *vma = plane_state->ggtt_vma; return true; @@ -44,12 +47,100 @@ intel_reuse_initial_plane_obj(struct drm_i915_private *i915, return false; } +static bool +initial_plane_phys_lmem(struct drm_i915_private *i915, + struct intel_initial_plane_config *plane_config) +{ + gen8_pte_t __iomem *gte = to_gt(i915)->ggtt->gsm; + struct intel_memory_region *mem; + dma_addr_t dma_addr; + gen8_pte_t pte; + u32 base; + + base = round_down(plane_config->base, I915_GTT_MIN_ALIGNMENT); + + gte += base / I915_GTT_PAGE_SIZE; + + pte = ioread64(gte); + if (!(pte & GEN12_GGTT_PTE_LM)) { + drm_err(&i915->drm, + "Initial plane programming missing PTE_LM bit\n"); + return false; + } + + dma_addr = pte & GEN12_GGTT_PTE_ADDR_MASK; + + if (IS_DGFX(i915)) + mem = i915->mm.regions[INTEL_REGION_LMEM_0]; + else + mem = i915->mm.stolen_region; + if (!mem) { + drm_dbg_kms(&i915->drm, + "Initial plane memory region not initialized\n"); + return false; + } + + /* + * On lmem we don't currently expect this to + * ever be placed in the stolen portion. + */ + if (dma_addr < mem->region.start || dma_addr > mem->region.end) { + drm_err(&i915->drm, + "Initial plane programming using invalid range, dma_addr=%pa (%s [%pa-%pa])\n", + &dma_addr, mem->region.name, &mem->region.start, &mem->region.end); + return false; + } + + drm_dbg(&i915->drm, + "Using dma_addr=%pa, based on initial plane programming\n", + &dma_addr); + + plane_config->phys_base = dma_addr - mem->region.start; + plane_config->mem = mem; + + return true; +} + +static bool +initial_plane_phys_smem(struct drm_i915_private *i915, + struct intel_initial_plane_config *plane_config) +{ + struct intel_memory_region *mem; + u32 base; + + base = round_down(plane_config->base, I915_GTT_MIN_ALIGNMENT); + + mem = i915->mm.stolen_region; + if (!mem) { + drm_dbg_kms(&i915->drm, + "Initial plane memory region not initialized\n"); + return false; + } + + /* FIXME get and validate the dma_addr from the PTE */ + plane_config->phys_base = base; + plane_config->mem = mem; + + return true; +} + +static bool +initial_plane_phys(struct drm_i915_private *i915, + struct intel_initial_plane_config *plane_config) +{ + if (IS_DGFX(i915) || HAS_LMEMBAR_SMEM_STOLEN(i915)) + return initial_plane_phys_lmem(i915, plane_config); + else + return initial_plane_phys_smem(i915, plane_config); +} + static struct i915_vma * initial_plane_vma(struct drm_i915_private *i915, struct intel_initial_plane_config *plane_config) { struct intel_memory_region *mem; struct drm_i915_gem_object *obj; + struct drm_mm_node orig_mm = {}; struct i915_vma *vma; resource_size_t phys_base; u32 base, size; @@ -58,45 +149,13 @@ initial_plane_vma(struct drm_i915_private *i915, if (plane_config->size == 0) return NULL; - base = round_down(plane_config->base, I915_GTT_MIN_ALIGNMENT); - if (IS_DGFX(i915)) { - gen8_pte_t __iomem *gte = to_gt(i915)->ggtt->gsm; - gen8_pte_t pte; - - gte += base / I915_GTT_PAGE_SIZE; - - pte = ioread64(gte); - if (!(pte & GEN12_GGTT_PTE_LM)) { - drm_err(&i915->drm, - "Initial plane programming missing PTE_LM bit\n"); - return NULL; - } - - phys_base = pte & I915_GTT_PAGE_MASK; - mem = i915->mm.regions[INTEL_REGION_LMEM_0]; - - /* - * We don't currently expect this to ever be placed in the - * stolen portion. - */ - if (phys_base >= resource_size(&mem->region)) { - drm_err(&i915->drm, - "Initial plane programming using invalid range, phys_base=%pa\n", - &phys_base); - return NULL; - } - - drm_dbg(&i915->drm, - "Using phys_base=%pa, based on initial plane programming\n", - &phys_base); - } else { - phys_base = base; - mem = i915->mm.stolen_region; - } - - if (!mem) + if (!initial_plane_phys(i915, plane_config)) return NULL; + phys_base = plane_config->phys_base; + mem = plane_config->mem; + + base = round_down(plane_config->base, I915_GTT_MIN_ALIGNMENT); size = round_up(plane_config->base + plane_config->size, mem->min_page_size); size -= base; @@ -108,14 +167,19 @@ initial_plane_vma(struct drm_i915_private *i915, */ if (IS_ENABLED(CONFIG_FRAMEBUFFER_CONSOLE) && mem == i915->mm.stolen_region && - size * 2 > i915->dsm.usable_size) + size * 2 > i915->dsm.usable_size) { + drm_dbg_kms(&i915->drm, "Initial FB size exceeds half of stolen, discarding\n"); return NULL; + } obj = i915_gem_object_create_region_at(mem, phys_base, size, I915_BO_ALLOC_USER | I915_BO_PREALLOC); - if (IS_ERR(obj)) + if (IS_ERR(obj)) { + drm_dbg_kms(&i915->drm, "Failed to preallocate initial FB in %s\n", + mem->region.name); return NULL; + } /* * Mark it WT ahead of time to avoid changing the @@ -139,23 +203,66 @@ initial_plane_vma(struct drm_i915_private *i915, goto err_obj; } + /* + * MTL GOP likes to place the framebuffer high up in ggtt, + * which can cause problems for ggtt_reserve_guc_top(). + * Try to pin it to a low ggtt address instead to avoid that. + */ + base = 0; + + if (base != plane_config->base) { + struct i915_ggtt *ggtt = to_gt(i915)->ggtt; + int ret; + + /* + * Make sure the original and new locations + * can't overlap. That would corrupt the original + * PTEs which are still being used for scanout. + */ + ret = i915_gem_gtt_reserve(&ggtt->vm, NULL, &orig_mm, + size, plane_config->base, + I915_COLOR_UNEVICTABLE, PIN_NOEVICT); + if (ret) + goto err_obj; + } + vma = i915_vma_instance(obj, &to_gt(i915)->ggtt->vm, NULL); if (IS_ERR(vma)) goto err_obj; +retry: pinctl = PIN_GLOBAL | PIN_OFFSET_FIXED | base; - if (HAS_GMCH(i915)) + if (!i915_gem_object_is_lmem(obj)) pinctl |= PIN_MAPPABLE; - if (i915_vma_pin(vma, 0, 0, pinctl)) + if (i915_vma_pin(vma, 0, 0, pinctl)) { + if (drm_mm_node_allocated(&orig_mm)) { + drm_mm_remove_node(&orig_mm); + /* + * Try again, but this time pin + * it to its original location. + */ + base = plane_config->base; + goto retry; + } goto err_obj; + } if (i915_gem_object_is_tiled(obj) && !i915_vma_is_map_and_fenceable(vma)) goto err_obj; + if (drm_mm_node_allocated(&orig_mm)) + drm_mm_remove_node(&orig_mm); + + drm_dbg_kms(&i915->drm, + "Initial plane fb bound to 0x%x in the ggtt (original 0x%x)\n", + i915_ggtt_offset(vma), plane_config->base); + return vma; err_obj: + if (drm_mm_node_allocated(&orig_mm)) + drm_mm_remove_node(&orig_mm); i915_gem_object_put(obj); return NULL; } @@ -210,10 +317,11 @@ err_vma: static void intel_find_initial_plane_obj(struct intel_crtc *crtc, - struct intel_initial_plane_config *plane_config) + struct intel_initial_plane_config plane_configs[]) { - struct drm_device *dev = crtc->base.dev; - struct drm_i915_private *dev_priv = to_i915(dev); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + struct intel_initial_plane_config *plane_config = + &plane_configs[crtc->pipe]; struct intel_plane *plane = to_intel_plane(crtc->base.primary); struct intel_plane_state *plane_state = @@ -239,7 +347,7 @@ intel_find_initial_plane_obj(struct intel_crtc *crtc, * Failed to alloc the obj, check to see if we should share * an fb with another CRTC instead */ - if (intel_reuse_initial_plane_obj(dev_priv, plane_config, &fb, &vma)) + if (intel_reuse_initial_plane_obj(crtc, plane_configs, &fb, &vma)) goto valid_fb; /* @@ -302,25 +410,36 @@ static void plane_config_fini(struct intel_initial_plane_config *plane_config) i915_vma_put(plane_config->vma); } -void intel_crtc_initial_plane_config(struct intel_crtc *crtc) +void intel_initial_plane_config(struct drm_i915_private *i915) { - struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); - struct intel_initial_plane_config plane_config = {}; + struct intel_initial_plane_config plane_configs[I915_MAX_PIPES] = {}; + struct intel_crtc *crtc; - /* - * Note that reserving the BIOS fb up front prevents us - * from stuffing other stolen allocations like the ring - * on top. This prevents some ugliness at boot time, and - * can even allow for smooth boot transitions if the BIOS - * fb is large enough for the active pipe configuration. - */ - dev_priv->display.funcs.display->get_initial_plane_config(crtc, &plane_config); + for_each_intel_crtc(&i915->drm, crtc) { + struct intel_initial_plane_config *plane_config = + &plane_configs[crtc->pipe]; - /* - * If the fb is shared between multiple heads, we'll - * just get the first one. - */ - intel_find_initial_plane_obj(crtc, &plane_config); + if (!to_intel_crtc_state(crtc->base.state)->uapi.active) + continue; + + /* + * Note that reserving the BIOS fb up front prevents us + * from stuffing other stolen allocations like the ring + * on top. This prevents some ugliness at boot time, and + * can even allow for smooth boot transitions if the BIOS + * fb is large enough for the active pipe configuration. + */ + i915->display.funcs.display->get_initial_plane_config(crtc, plane_config); - plane_config_fini(&plane_config); + /* + * If the fb is shared between multiple heads, we'll + * just get the first one. + */ + intel_find_initial_plane_obj(crtc, plane_configs); + + if (i915->display.funcs.display->fixup_initial_plane_config(crtc, plane_config)) + intel_crtc_wait_for_next_vblank(crtc); + + plane_config_fini(plane_config); + } } diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.h b/drivers/gpu/drm/i915/display/intel_plane_initial.h index c7e35ab3182b..64ab95239cd4 100644 --- a/drivers/gpu/drm/i915/display/intel_plane_initial.h +++ b/drivers/gpu/drm/i915/display/intel_plane_initial.h @@ -6,8 +6,8 @@ #ifndef __INTEL_PLANE_INITIAL_H__ #define __INTEL_PLANE_INITIAL_H__ -struct intel_crtc; +struct drm_i915_private; -void intel_crtc_initial_plane_config(struct intel_crtc *crtc); +void intel_initial_plane_config(struct drm_i915_private *i915); #endif diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c index a8fa3a20990e..2d65a538f83e 100644 --- a/drivers/gpu/drm/i915/display/intel_pps.c +++ b/drivers/gpu/drm/i915/display/intel_pps.c @@ -366,7 +366,7 @@ static bool intel_pps_is_valid(struct intel_dp *intel_dp) if (intel_dp->pps.pps_idx == 1 && INTEL_PCH_TYPE(i915) >= PCH_ICP && - INTEL_PCH_TYPE(i915) < PCH_MTP) + INTEL_PCH_TYPE(i915) <= PCH_ADP) return intel_de_read(i915, SOUTH_CHICKEN1) & ICP_SECOND_PPS_IO_SELECT; return true; diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 57bbf3e3af92..72cadad09db5 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -173,6 +173,12 @@ * irrelevant for normal operation. */ +#define CAN_PSR(intel_dp) ((intel_dp)->psr.sink_support && \ + (intel_dp)->psr.source_support) + +#define CAN_PANEL_REPLAY(intel_dp) ((intel_dp)->psr.sink_panel_replay_support && \ + (intel_dp)->psr.source_panel_replay_support) + bool intel_encoder_can_psr(struct intel_encoder *encoder) { if (intel_encoder_is_dp(encoder) || encoder->type == INTEL_OUTPUT_DP_MST) @@ -528,7 +534,7 @@ static void _psr_init_dpcd(struct intel_dp *intel_dp) intel_dp_get_sink_sync_latency(intel_dp); if (DISPLAY_VER(i915) >= 9 && - intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_IS_SUPPORTED) { + intel_dp->psr_dpcd[0] >= DP_PSR2_WITH_Y_COORD_IS_SUPPORTED) { bool y_req = intel_dp->psr_dpcd[1] & DP_PSR2_SU_Y_COORDINATE_REQUIRED; bool alpm = intel_dp_get_alpm_status(intel_dp); @@ -560,11 +566,8 @@ void intel_psr_init_dpcd(struct intel_dp *intel_dp) if (intel_dp->psr_dpcd[0]) _psr_init_dpcd(intel_dp); - if (intel_dp->psr.sink_psr2_support) { - intel_dp->psr.colorimetry_support = - intel_dp_get_colorimetry_status(intel_dp); + if (intel_dp->psr.sink_psr2_support) intel_dp_get_su_granularity(intel_dp); - } } static void hsw_psr_setup_aux(struct intel_dp *intel_dp) @@ -604,6 +607,18 @@ static void hsw_psr_setup_aux(struct intel_dp *intel_dp) aux_ctl); } +static bool psr2_su_region_et_valid(struct intel_dp *intel_dp) +{ + struct drm_i915_private *i915 = dp_to_i915(intel_dp); + + if (DISPLAY_VER(i915) >= 20 && + intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_ET_SUPPORTED && + !(intel_dp->psr.debug & I915_PSR_DEBUG_SU_REGION_ET_DISABLE)) + return true; + + return false; +} + static void intel_psr_enable_sink(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); @@ -619,6 +634,8 @@ static void intel_psr_enable_sink(struct intel_dp *intel_dp) DP_ALPM_LOCK_ERROR_IRQ_HPD_ENABLE); dpcd_val |= DP_PSR_ENABLE_PSR2 | DP_PSR_IRQ_HPD_WITH_CRC_ERRORS; + if (psr2_su_region_et_valid(intel_dp)) + dpcd_val |= DP_PSR_ENABLE_SU_REGION_ET; } else { if (intel_dp->psr.link_standby) dpcd_val |= DP_PSR_MAIN_LINK_ACTIVE; @@ -762,8 +779,8 @@ static u32 intel_psr2_get_tp_time(struct intel_dp *intel_dp) static int psr2_block_count_lines(struct intel_dp *intel_dp) { - return intel_dp->psr.io_wake_lines < 9 && - intel_dp->psr.fast_wake_lines < 9 ? 8 : 12; + return intel_dp->psr.alpm_parameters.io_wake_lines < 9 && + intel_dp->psr.alpm_parameters.fast_wake_lines < 9 ? 8 : 12; } static int psr2_block_count(struct intel_dp *intel_dp) @@ -800,6 +817,7 @@ static void dg2_activate_panel_replay(struct intel_dp *intel_dp) static void hsw_activate_psr2(struct intel_dp *intel_dp) { struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + struct intel_psr *psr = &intel_dp->psr; enum transcoder cpu_transcoder = intel_dp->psr.transcoder; u32 val = EDP_PSR2_ENABLE; u32 psr_val = 0; @@ -841,17 +859,18 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) */ int tmp; - tmp = map[intel_dp->psr.io_wake_lines - TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES]; + tmp = map[psr->alpm_parameters.io_wake_lines - + TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES]; val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(tmp + TGL_EDP_PSR2_IO_BUFFER_WAKE_MIN_LINES); - tmp = map[intel_dp->psr.fast_wake_lines - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES]; + tmp = map[psr->alpm_parameters.fast_wake_lines - TGL_EDP_PSR2_FAST_WAKE_MIN_LINES]; val |= TGL_EDP_PSR2_FAST_WAKE(tmp + TGL_EDP_PSR2_FAST_WAKE_MIN_LINES); } else if (DISPLAY_VER(dev_priv) >= 12) { - val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(intel_dp->psr.io_wake_lines); - val |= TGL_EDP_PSR2_FAST_WAKE(intel_dp->psr.fast_wake_lines); + val |= TGL_EDP_PSR2_IO_BUFFER_WAKE(psr->alpm_parameters.io_wake_lines); + val |= TGL_EDP_PSR2_FAST_WAKE(psr->alpm_parameters.fast_wake_lines); } else if (DISPLAY_VER(dev_priv) >= 9) { - val |= EDP_PSR2_IO_BUFFER_WAKE(intel_dp->psr.io_wake_lines); - val |= EDP_PSR2_FAST_WAKE(intel_dp->psr.fast_wake_lines); + val |= EDP_PSR2_IO_BUFFER_WAKE(psr->alpm_parameters.io_wake_lines); + val |= EDP_PSR2_FAST_WAKE(psr->alpm_parameters.fast_wake_lines); } if (intel_dp->psr.req_psr2_sdp_prior_scanline) @@ -869,6 +888,9 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp) intel_de_write(dev_priv, PSR2_MAN_TRK_CTL(cpu_transcoder), 0); } + if (psr2_su_region_et_valid(intel_dp)) + val |= LNL_EDP_PSR2_SU_REGION_ET_ENABLE; + /* * PSR2 HW is incorrectly using EDP_PSR_TP1_TP3_SEL and BSpec is * recommending keep this bit unset while PSR2 is enabled. @@ -1031,6 +1053,9 @@ static bool intel_psr2_sel_fetch_config_valid(struct intel_dp *intel_dp, return false; } + if (psr2_su_region_et_valid(intel_dp)) + crtc_state->enable_psr2_su_region_et = true; + return crtc_state->enable_psr2_sel_fetch = true; } @@ -1101,10 +1126,34 @@ static bool _compute_psr2_sdp_prior_scanline_indication(struct intel_dp *intel_d return true; } -static bool _compute_psr2_wake_times(struct intel_dp *intel_dp, +static bool _lnl_compute_alpm_params(struct intel_dp *intel_dp, struct intel_crtc_state *crtc_state) { struct drm_i915_private *i915 = dp_to_i915(intel_dp); + int check_entry_lines; + + if (DISPLAY_VER(i915) < 20) + return true; + + /* ALPM Entry Check = 2 + CEILING( 5us /tline ) */ + check_entry_lines = 2 + + intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode, 5); + + if (check_entry_lines > 15) + return false; + + if (i915->display.params.psr_safest_params) + check_entry_lines = 15; + + intel_dp->psr.alpm_parameters.check_entry_lines = check_entry_lines; + + return true; +} + +static bool _compute_alpm_params(struct intel_dp *intel_dp, + struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *i915 = dp_to_i915(intel_dp); int io_wake_lines, io_wake_time, fast_wake_lines, fast_wake_time; u8 max_wake_lines; @@ -1115,6 +1164,8 @@ static bool _compute_psr2_wake_times(struct intel_dp *intel_dp, * it is not enough -> use 45 us. */ fast_wake_time = 45; + + /* TODO: Check how we can use ALPM_CTL fast wake extended field */ max_wake_lines = 12; } else { io_wake_time = 50; @@ -1131,12 +1182,15 @@ static bool _compute_psr2_wake_times(struct intel_dp *intel_dp, fast_wake_lines > max_wake_lines) return false; + if (!_lnl_compute_alpm_params(intel_dp, crtc_state)) + return false; + if (i915->display.params.psr_safest_params) io_wake_lines = fast_wake_lines = max_wake_lines; /* According to Bspec lower limit should be set as 7 lines. */ - intel_dp->psr.io_wake_lines = max(io_wake_lines, 7); - intel_dp->psr.fast_wake_lines = max(fast_wake_lines, 7); + intel_dp->psr.alpm_parameters.io_wake_lines = max(io_wake_lines, 7); + intel_dp->psr.alpm_parameters.fast_wake_lines = max(fast_wake_lines, 7); return true; } @@ -1268,7 +1322,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp, return false; } - if (!_compute_psr2_wake_times(intel_dp, crtc_state)) { + if (!_compute_alpm_params(intel_dp, crtc_state)) { drm_dbg_kms(&dev_priv->drm, "PSR2 not enabled, Unable to use long enough wake times\n"); return false; @@ -1377,10 +1431,6 @@ void intel_psr_compute_config(struct intel_dp *intel_dp, return; crtc_state->has_psr2 = intel_psr2_config_valid(intel_dp, crtc_state); - - crtc_state->infoframes.enable |= intel_hdmi_infoframe_enable(DP_SDP_VSC); - intel_dp_compute_psr_vsc_sdp(intel_dp, crtc_state, conn_state, - &crtc_state->psr_vsc); } void intel_psr_get_config(struct intel_encoder *encoder, @@ -1504,6 +1554,21 @@ static void wm_optimization_wa(struct intel_dp *intel_dp, wa_16013835468_bit_get(intel_dp), 0); } +static void lnl_alpm_configure(struct intel_dp *intel_dp) +{ + struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + enum transcoder cpu_transcoder = intel_dp->psr.transcoder; + struct intel_psr *psr = &intel_dp->psr; + + if (DISPLAY_VER(dev_priv) < 20) + return; + + intel_de_write(dev_priv, ALPM_CTL(cpu_transcoder), + ALPM_CTL_EXTENDED_FAST_WAKE_ENABLE | + ALPM_CTL_ALPM_ENTRY_CHECK(psr->alpm_parameters.check_entry_lines) | + ALPM_CTL_EXTENDED_FAST_WAKE_TIME(psr->alpm_parameters.fast_wake_lines)); +} + static void intel_psr_enable_source(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state) { @@ -1569,6 +1634,8 @@ static void intel_psr_enable_source(struct intel_dp *intel_dp, intel_dp->psr.psr2_sel_fetch_enabled ? IGNORE_PSR2_HW_TRACKING : 0); + lnl_alpm_configure(intel_dp); + /* * Wa_16013835468 * Wa_14015648006 @@ -1634,7 +1701,6 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp, struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); enum phy phy = intel_port_to_phy(dev_priv, dig_port->base.port); - struct intel_encoder *encoder = &dig_port->base; u32 val; drm_WARN_ON(&dev_priv->drm, intel_dp->psr.enabled); @@ -1662,7 +1728,6 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp, drm_dbg_kms(&dev_priv->drm, "Enabling PSR%s\n", intel_dp->psr.psr2_enabled ? "2" : "1"); - intel_write_dp_vsc_sdp(encoder, crtc_state, &crtc_state->psr_vsc); intel_snps_phy_update_psr_power_state(dev_priv, phy, true); intel_psr_enable_sink(intel_dp); intel_psr_enable_source(intel_dp, crtc_state); @@ -1951,7 +2016,7 @@ void intel_psr2_program_trans_man_trk_ctl(const struct intel_crtc_state *crtc_st } static void psr2_man_trk_ctl_calc(struct intel_crtc_state *crtc_state, - struct drm_rect *clip, bool full_update) + bool full_update) { struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc); struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); @@ -1966,17 +2031,21 @@ static void psr2_man_trk_ctl_calc(struct intel_crtc_state *crtc_state, goto exit; } - if (clip->y1 == -1) + if (crtc_state->psr2_su_area.y1 == -1) goto exit; if (IS_ALDERLAKE_P(dev_priv) || DISPLAY_VER(dev_priv) >= 14) { - val |= ADLP_PSR2_MAN_TRK_CTL_SU_REGION_START_ADDR(clip->y1); - val |= ADLP_PSR2_MAN_TRK_CTL_SU_REGION_END_ADDR(clip->y2 - 1); + val |= ADLP_PSR2_MAN_TRK_CTL_SU_REGION_START_ADDR(crtc_state->psr2_su_area.y1); + val |= ADLP_PSR2_MAN_TRK_CTL_SU_REGION_END_ADDR(crtc_state->psr2_su_area.y2 - 1); } else { - drm_WARN_ON(crtc_state->uapi.crtc->dev, clip->y1 % 4 || clip->y2 % 4); + drm_WARN_ON(crtc_state->uapi.crtc->dev, + crtc_state->psr2_su_area.y1 % 4 || + crtc_state->psr2_su_area.y2 % 4); - val |= PSR2_MAN_TRK_CTL_SU_REGION_START_ADDR(clip->y1 / 4 + 1); - val |= PSR2_MAN_TRK_CTL_SU_REGION_END_ADDR(clip->y2 / 4 + 1); + val |= PSR2_MAN_TRK_CTL_SU_REGION_START_ADDR( + crtc_state->psr2_su_area.y1 / 4 + 1); + val |= PSR2_MAN_TRK_CTL_SU_REGION_END_ADDR( + crtc_state->psr2_su_area.y2 / 4 + 1); } exit: crtc_state->psr2_man_track_ctl = val; @@ -2002,8 +2071,7 @@ static void clip_area_update(struct drm_rect *overlap_damage_area, overlap_damage_area->y2 = damage_area->y2; } -static void intel_psr2_sel_fetch_pipe_alignment(const struct intel_crtc_state *crtc_state, - struct drm_rect *pipe_clip) +static void intel_psr2_sel_fetch_pipe_alignment(struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev); const struct drm_dsc_config *vdsc_cfg = &crtc_state->dsc.config; @@ -2016,9 +2084,32 @@ static void intel_psr2_sel_fetch_pipe_alignment(const struct intel_crtc_state *c else y_alignment = crtc_state->su_y_granularity; - pipe_clip->y1 -= pipe_clip->y1 % y_alignment; - if (pipe_clip->y2 % y_alignment) - pipe_clip->y2 = ((pipe_clip->y2 / y_alignment) + 1) * y_alignment; + crtc_state->psr2_su_area.y1 -= crtc_state->psr2_su_area.y1 % y_alignment; + if (crtc_state->psr2_su_area.y2 % y_alignment) + crtc_state->psr2_su_area.y2 = ((crtc_state->psr2_su_area.y2 / + y_alignment) + 1) * y_alignment; +} + +/* + * When early transport is in use we need to extend SU area to cover + * cursor fully when cursor is in SU area. + */ +static void +intel_psr2_sel_fetch_et_alignment(struct intel_crtc_state *crtc_state, + struct intel_plane_state *cursor_state) +{ + struct drm_rect inter; + + if (!crtc_state->enable_psr2_su_region_et || + !cursor_state->uapi.visible) + return; + + inter = crtc_state->psr2_su_area; + if (!drm_rect_intersect(&inter, &cursor_state->uapi.dst)) + return; + + clip_area_update(&crtc_state->psr2_su_area, &cursor_state->uapi.dst, + &crtc_state->pipe_src); } /* @@ -2061,8 +2152,8 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, { struct drm_i915_private *dev_priv = to_i915(state->base.dev); struct intel_crtc_state *crtc_state = intel_atomic_get_new_crtc_state(state, crtc); - struct drm_rect pipe_clip = { .x1 = 0, .y1 = -1, .x2 = INT_MAX, .y2 = -1 }; - struct intel_plane_state *new_plane_state, *old_plane_state; + struct intel_plane_state *new_plane_state, *old_plane_state, + *cursor_plane_state = NULL; struct intel_plane *plane; bool full_update = false; int i, ret; @@ -2075,6 +2166,11 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, goto skip_sel_fetch_set_loop; } + crtc_state->psr2_su_area.x1 = 0; + crtc_state->psr2_su_area.y1 = -1; + crtc_state->psr2_su_area.x2 = INT_MAX; + crtc_state->psr2_su_area.y2 = -1; + /* * Calculate minimal selective fetch area of each plane and calculate * the pipe damaged area. @@ -2109,14 +2205,14 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, if (old_plane_state->uapi.visible) { damaged_area.y1 = old_plane_state->uapi.dst.y1; damaged_area.y2 = old_plane_state->uapi.dst.y2; - clip_area_update(&pipe_clip, &damaged_area, + clip_area_update(&crtc_state->psr2_su_area, &damaged_area, &crtc_state->pipe_src); } if (new_plane_state->uapi.visible) { damaged_area.y1 = new_plane_state->uapi.dst.y1; damaged_area.y2 = new_plane_state->uapi.dst.y2; - clip_area_update(&pipe_clip, &damaged_area, + clip_area_update(&crtc_state->psr2_su_area, &damaged_area, &crtc_state->pipe_src); } continue; @@ -2124,7 +2220,7 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, /* If alpha changed mark the whole plane area as damaged */ damaged_area.y1 = new_plane_state->uapi.dst.y1; damaged_area.y2 = new_plane_state->uapi.dst.y2; - clip_area_update(&pipe_clip, &damaged_area, + clip_area_update(&crtc_state->psr2_su_area, &damaged_area, &crtc_state->pipe_src); continue; } @@ -2141,7 +2237,14 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, damaged_area.x1 += new_plane_state->uapi.dst.x1 - src.x1; damaged_area.x2 += new_plane_state->uapi.dst.x1 - src.x1; - clip_area_update(&pipe_clip, &damaged_area, &crtc_state->pipe_src); + clip_area_update(&crtc_state->psr2_su_area, &damaged_area, &crtc_state->pipe_src); + + /* + * Cursor plane new state is stored to adjust su area to cover + * cursor are fully. + */ + if (plane->id == PLANE_CURSOR) + cursor_plane_state = new_plane_state; } /* @@ -2150,7 +2253,7 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, * should identify cases where this happens and fix the area * calculation for those. */ - if (pipe_clip.y1 == -1) { + if (crtc_state->psr2_su_area.y1 == -1) { drm_info_once(&dev_priv->drm, "Selective fetch area calculation failed in pipe %c\n", pipe_name(crtc->pipe)); @@ -2164,13 +2267,17 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, if ((IS_DISPLAY_IP_STEP(dev_priv, IP_VER(14, 0), STEP_A0, STEP_B0) || IS_ALDERLAKE_P(dev_priv) || IS_TIGERLAKE(dev_priv)) && crtc_state->splitter.enable) - pipe_clip.y1 = 0; + crtc_state->psr2_su_area.y1 = 0; ret = drm_atomic_add_affected_planes(&state->base, &crtc->base); if (ret) return ret; - intel_psr2_sel_fetch_pipe_alignment(crtc_state, &pipe_clip); + /* Adjust su area to cover cursor fully as necessary */ + if (cursor_plane_state) + intel_psr2_sel_fetch_et_alignment(crtc_state, cursor_plane_state); + + intel_psr2_sel_fetch_pipe_alignment(crtc_state); /* * Now that we have the pipe damaged area check if it intersect with @@ -2185,7 +2292,7 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, !new_plane_state->uapi.visible) continue; - inter = pipe_clip; + inter = crtc_state->psr2_su_area; sel_fetch_area = &new_plane_state->psr2_sel_fetch_area; if (!drm_rect_intersect(&inter, &new_plane_state->uapi.dst)) { sel_fetch_area->y1 = -1; @@ -2230,7 +2337,7 @@ int intel_psr2_sel_fetch_update(struct intel_atomic_state *state, } skip_sel_fetch_set_loop: - psr2_man_trk_ctl_calc(crtc_state, &pipe_clip, full_update); + psr2_man_trk_ctl_calc(crtc_state, full_update); return 0; } @@ -2799,6 +2906,9 @@ void intel_psr_init(struct intel_dp *intel_dp) else intel_dp->psr.source_support = true; + /* Disable early transport for now */ + intel_dp->psr.debug |= I915_PSR_DEBUG_SU_REGION_ET_DISABLE; + /* Set link_standby x link_off defaults */ if (DISPLAY_VER(dev_priv) < 12) /* For new platforms up to TGL let's respect VBT back again */ diff --git a/drivers/gpu/drm/i915/display/intel_psr.h b/drivers/gpu/drm/i915/display/intel_psr.h index 143e0595c097..cde781df84d5 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.h +++ b/drivers/gpu/drm/i915/display/intel_psr.h @@ -21,12 +21,6 @@ struct intel_encoder; struct intel_plane; struct intel_plane_state; -#define CAN_PSR(intel_dp) ((intel_dp)->psr.sink_support && \ - (intel_dp)->psr.source_support) - -#define CAN_PANEL_REPLAY(intel_dp) ((intel_dp)->psr.sink_panel_replay_support && \ - (intel_dp)->psr.source_panel_replay_support) - bool intel_encoder_can_psr(struct intel_encoder *encoder); void intel_psr_init_dpcd(struct intel_dp *intel_dp); void intel_psr_pre_plane_update(struct intel_atomic_state *state, diff --git a/drivers/gpu/drm/i915/display/intel_psr_regs.h b/drivers/gpu/drm/i915/display/intel_psr_regs.h index efe4306b37e0..8427a736f639 100644 --- a/drivers/gpu/drm/i915/display/intel_psr_regs.h +++ b/drivers/gpu/drm/i915/display/intel_psr_regs.h @@ -159,6 +159,7 @@ #define TGL_EDP_PSR2_BLOCK_COUNT_MASK REG_BIT(28) #define TGL_EDP_PSR2_BLOCK_COUNT_NUM_2 REG_FIELD_PREP(TGL_EDP_PSR2_BLOCK_COUNT_MASK, 0) #define TGL_EDP_PSR2_BLOCK_COUNT_NUM_3 REG_FIELD_PREP(TGL_EDP_PSR2_BLOCK_COUNT_MASK, 1) +#define LNL_EDP_PSR2_SU_REGION_ET_ENABLE REG_BIT(27) #define EDP_Y_COORDINATE_ENABLE REG_BIT(25) /* display 10, 11 and 12 */ #define EDP_PSR2_SU_SDP_SCANLINE REG_BIT(25) /* display 13+ */ #define EDP_MAX_SU_DISABLE_TIME_MASK REG_GENMASK(24, 20) @@ -245,6 +246,11 @@ #define ADLP_PSR2_MAN_TRK_CTL_SF_SINGLE_FULL_FRAME REG_BIT(14) #define ADLP_PSR2_MAN_TRK_CTL_SF_CONTINUOS_FULL_FRAME REG_BIT(13) +/* PSR2 Early transport */ +#define _PIPE_SRCSZ_ERLY_TPT_A 0x70074 + +#define PIPE_SRCSZ_ERLY_TPT(trans) _MMIO_TRANS2(trans, _PIPE_SRCSZ_ERLY_TPT_A) + #define _SEL_FETCH_PLANE_BASE_1_A 0x70890 #define _SEL_FETCH_PLANE_BASE_2_A 0x708B0 #define _SEL_FETCH_PLANE_BASE_3_A 0x708D0 @@ -290,4 +296,61 @@ _SEL_FETCH_PLANE_OFFSET_1_A - \ _SEL_FETCH_PLANE_BASE_1_A) +#define _ALPM_CTL_A 0x60950 +#define ALPM_CTL(tran) _MMIO_TRANS2(tran, _ALPM_CTL_A) +#define ALPM_CTL_ALPM_ENABLE REG_BIT(31) +#define ALPM_CTL_ALPM_AUX_LESS_ENABLE REG_BIT(30) +#define ALPM_CTL_LOBF_ENABLE REG_BIT(29) +#define ALPM_CTL_EXTENDED_FAST_WAKE_ENABLE REG_BIT(28) +#define ALPM_CTL_KEEP_FEC_ENABLE_FOR_AUX_WAKE_SLEEP REG_BIT(27) +#define ALPM_CTL_RESTORE_OCCURED REG_BIT(26) +#define ALPM_CTL_RESTORE_TO_SLEEP REG_BIT(25) +#define ALPM_CTL_RESTORE_TO_DEEP_SLEEP REG_BIT(24) +#define ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_MASK REG_GENMASK(23, 21) +#define ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_50_SYMBOLS REG_FIELD_PREP(ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_MASK, 0) +#define ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_128_SYMBOLS REG_FIELD_PREP(ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_MASK, 1) +#define ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_256_SYMBOLS REG_FIELD_PREP(ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_MASK, 2) +#define ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_512_SYMBOLS REG_FIELD_PREP(ALPM_CTL_AUX_LESS_SLEEP_HOLD_TIME_MASK, 3) +#define ALPM_CTL_AUX_WAKE_SLEEP_HOLD_ENABLE REG_BIT(20) +#define ALPM_CTL_ALPM_ENTRY_CHECK_MASK REG_GENMASK(19, 16) +#define ALPM_CTL_ALPM_ENTRY_CHECK(val) REG_FIELD_PREP(ALPM_CTL_ALPM_ENTRY_CHECK_MASK, val) +#define ALPM_CTL_EXTENDED_FAST_WAKE_TIME_MASK REG_GENMASK(13, 8) +#define ALPM_CTL_EXTENDED_FAST_WAKE_MIN_LINES 5 +#define ALPM_CTL_EXTENDED_FAST_WAKE_TIME(lines) REG_FIELD_PREP(ALPM_CTL_EXTENDED_FAST_WAKE_TIME_MASK, (lines) - ALPM_CTL_EXTENDED_FAST_WAKE_MIN_LINES) +#define ALPM_CTL_AUX_LESS_WAKE_TIME_MASK REG_GENMASK(5, 0) +#define ALPM_CTL_AUX_LESS_WAKE_TIME(val) REG_FIELD_PREP(ALPM_CTL_AUX_LESS_WAKE_TIME_MASK, val) + +#define _ALPM_CTL2_A 0x60954 +#define ALPM_CTL2(tran) _MMIO_TRANS2(tran, _ALPM_CTL2_A) +#define ALPM_CTL2_SWITCH_TO_ACTIVE_LATENCY_MASK REG_GENMASK(28, 24) +#define ALPM_CTL2_SWITCH_TO_ACTIVE_LATENCY(val) REG_FIELD_PREP(ALPM_CTL2_SWITCH_TO_ACTIVE_LATENCY_MASK, val) +#define ALPM_CTL2_AUX_LESS_WAKE_TIME_EXTENSION_MASK REG_GENMASK(19, 16) +#define ALPM_CTL2_AUX_LESS_WAKE_TIME_EXTENSION(val) REG_FIELD_PREP(ALPM_CTL2_AUX_LESS_WAKE_TIME_EXTENSION_MASK, val) +#define ALPM_CTL2_NUMBER_OF_LTTPR_MASK REG_GENMASK(15, 12) +#define ALPM_CTL2_NUMBER_OF_LTTPR(val) REG_FIELD_PREP(ALPM_CTL2_NUMBER_OF_LTTPR_MASK, val) +#define ALPM_CTL2_LTTPR_AUX_LESS_SLEEP_HOLD_TIME_MASK REG_GENMASK(10, 8) +#define ALPM_CTL2_LTTPR_AUX_LESS_SLEEP_HOLD_TIME(val) REG_FIELD_PREP(ALPM_CTL2_LTTPR_AUX_LESS_SLEEP_HOLD_TIME_MASK, val) +#define ALPM_CTL2_FEC_DECODE_EN_POSITION_AFTER_WAKE_SR REG_BIT(4) +#define ALPM_CTL2_NUMBER_AUX_LESS_ML_PHY_SLEEP_SEQUENCES_MASK REG_GENMASK(2, 0) +#define ALPM_CTL2_NUMBER_AUX_LESS_ML_PHY_SLEEP_SEQUENCES(val) REG_FIELD_PREP(ALPM_CTL2_NUMBER_AUX_LESS_ML_PHY_SLEEP_SEQUENCES_MASK, val) + +#define _PORT_ALPM_CTL_A 0x16fa2c +#define PORT_ALPM_CTL(tran) _MMIO_TRANS2(tran, _PORT_ALPM_CTL_A) +#define PORT_ALPM_CTL_ALPM_AUX_LESS_ENABLE REG_BIT(31) +#define PORT_ALPM_CTL_MAX_PHY_SWING_SETUP_MASK REG_GENMASK(23, 20) +#define PORT_ALPM_CTL_MAX_PHY_SWING_SETUP(val) REG_FIELD_PREP(PORT_ALPM_CTL_MAX_PHY_SWING_SETUP_MASK, val) +#define PORT_ALPM_CTL_MAX_PHY_SWING_HOLD_MASK REG_GENMASK(19, 16) +#define PORT_ALPM_CTL_MAX_PHY_SWING_HOLD(val) REG_FIELD_PREP(PORT_ALPM_CTL_MAX_PHY_SWING_HOLD_MASK, val) +#define PORT_ALPM_CTL_SILENCE_PERIOD_MASK REG_GENMASK(7, 0) +#define PORT_ALPM_CTL_SILENCE_PERIOD(val) REG_FIELD_PREP(PORT_ALPM_CTL_SILENCE_PERIOD_MASK, val) + +#define _PORT_ALPM_LFPS_CTL_A 0x16fa30 +#define PORT_ALPM_LFPS_CTL(tran) _MMIO_TRANS2(tran, _PORT_ALPM_LFPS_CTL_A) +#define PORT_ALPM_LFPS_CTL_LFPS_START_POLARITY REG_BIT(31) +#define PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT_MASK REG_GENMASK(27, 24) +#define ALPM_CTL_EXTENDED_FAST_WAKE_MIN_LINES 5 +#define ALPM_CTL_EXTENDED_FAST_WAKE_TIME(lines) REG_FIELD_PREP(ALPM_CTL_EXTENDED_FAST_WAKE_TIME_MASK, (lines) - ALPM_CTL_EXTENDED_FAST_WAKE_MIN_LINES) +#define ALPM_CTL_AUX_LESS_WAKE_TIME_MASK REG_GENMASK(5, 0) +#define ALPM_CTL_AUX_LESS_WAKE_TIME(val) REG_FIELD_PREP(ALPM_CTL_AUX_LESS_WAKE_TIME_MASK, val) + #endif /* __INTEL_PSR_REGS_H__ */ diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c index acc6b6804105..2571ef5a1b21 100644 --- a/drivers/gpu/drm/i915/display/intel_sdvo.c +++ b/drivers/gpu/drm/i915/display/intel_sdvo.c @@ -44,6 +44,7 @@ #include "intel_connector.h" #include "intel_crtc.h" #include "intel_de.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_fdi.h" #include "intel_fifo_underrun.h" @@ -2140,6 +2141,9 @@ intel_sdvo_detect(struct drm_connector *connector, bool force) if (!intel_display_device_enabled(i915)) return connector_status_disconnected; + if (!intel_display_driver_check_access(i915)) + return connector->status; + if (!intel_sdvo_set_target_output(intel_sdvo, intel_sdvo_connector->output_flag)) return connector_status_unknown; @@ -2805,6 +2809,7 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, u16 type) } else { intel_connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; } + intel_connector->base.polled = intel_connector->polled; encoder->encoder_type = DRM_MODE_ENCODER_TMDS; connector->connector_type = DRM_MODE_CONNECTOR_DVID; @@ -2880,6 +2885,7 @@ intel_sdvo_analog_init(struct intel_sdvo *intel_sdvo, u16 type) intel_connector = &intel_sdvo_connector->base; connector = &intel_connector->base; intel_connector->polled = DRM_CONNECTOR_POLL_CONNECT; + intel_connector->base.polled = intel_connector->polled; encoder->encoder_type = DRM_MODE_ENCODER_DAC; connector->connector_type = DRM_MODE_CONNECTOR_VGA; diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c index dcf05e00e505..6b374d481cd9 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.c +++ b/drivers/gpu/drm/i915/display/intel_tc.c @@ -122,6 +122,15 @@ bool intel_tc_port_in_legacy_mode(struct intel_digital_port *dig_port) return intel_tc_port_in_mode(dig_port, TC_PORT_LEGACY); } +bool intel_tc_port_handles_hpd_glitches(struct intel_digital_port *dig_port) +{ + struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); + enum phy phy = intel_port_to_phy(i915, dig_port->base.port); + struct intel_tc_port *tc = to_tc_port(dig_port); + + return intel_phy_is_tc(i915, phy) && !tc->legacy_port; +} + /* * The display power domains used for TC ports depending on the * platform and TC mode (legacy, DP-alt, TBT): @@ -986,10 +995,11 @@ xelpdp_tc_phy_tcss_power_is_enabled(struct intel_tc_port *tc) { struct drm_i915_private *i915 = tc_to_i915(tc); enum port port = tc->dig_port->base.port; + i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port); assert_tc_cold_blocked(tc); - return intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_TCSS_POWER_STATE; + return intel_de_read(i915, reg) & XELPDP_TCSS_POWER_STATE; } static bool @@ -1012,16 +1022,17 @@ static void __xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool ena { struct drm_i915_private *i915 = tc_to_i915(tc); enum port port = tc->dig_port->base.port; + i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port); u32 val; assert_tc_cold_blocked(tc); - val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)); + val = intel_de_read(i915, reg); if (enable) val |= XELPDP_TCSS_POWER_REQUEST; else val &= ~XELPDP_TCSS_POWER_REQUEST; - intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val); + intel_de_write(i915, reg, val); } static bool xelpdp_tc_phy_enable_tcss_power(struct intel_tc_port *tc, bool enable) @@ -1055,26 +1066,28 @@ static void xelpdp_tc_phy_take_ownership(struct intel_tc_port *tc, bool take) { struct drm_i915_private *i915 = tc_to_i915(tc); enum port port = tc->dig_port->base.port; + i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port); u32 val; assert_tc_cold_blocked(tc); - val = intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)); + val = intel_de_read(i915, reg); if (take) val |= XELPDP_TC_PHY_OWNERSHIP; else val &= ~XELPDP_TC_PHY_OWNERSHIP; - intel_de_write(i915, XELPDP_PORT_BUF_CTL1(port), val); + intel_de_write(i915, reg, val); } static bool xelpdp_tc_phy_is_owned(struct intel_tc_port *tc) { struct drm_i915_private *i915 = tc_to_i915(tc); enum port port = tc->dig_port->base.port; + i915_reg_t reg = XELPDP_PORT_BUF_CTL1(i915, port); assert_tc_cold_blocked(tc); - return intel_de_read(i915, XELPDP_PORT_BUF_CTL1(port)) & XELPDP_TC_PHY_OWNERSHIP; + return intel_de_read(i915, reg) & XELPDP_TC_PHY_OWNERSHIP; } static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc) @@ -1590,7 +1603,7 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port, * connected ports are usable, and avoids exposing to the users objects they * can't really use. */ -bool intel_tc_port_connected_locked(struct intel_encoder *encoder) +bool intel_tc_port_connected(struct intel_encoder *encoder) { struct intel_digital_port *dig_port = enc_to_dig_port(encoder); struct drm_i915_private *i915 = to_i915(dig_port->base.base.dev); @@ -1605,19 +1618,6 @@ bool intel_tc_port_connected_locked(struct intel_encoder *encoder) return tc_phy_hpd_live_status(tc) & mask; } -bool intel_tc_port_connected(struct intel_encoder *encoder) -{ - struct intel_digital_port *dig_port = enc_to_dig_port(encoder); - struct intel_tc_port *tc = to_tc_port(dig_port); - bool is_connected; - - mutex_lock(&tc->lock); - is_connected = intel_tc_port_connected_locked(encoder); - mutex_unlock(&tc->lock); - - return is_connected; -} - static bool __intel_tc_port_link_needs_reset(struct intel_tc_port *tc) { bool ret; diff --git a/drivers/gpu/drm/i915/display/intel_tc.h b/drivers/gpu/drm/i915/display/intel_tc.h index 80a61e52850e..26c4265368c1 100644 --- a/drivers/gpu/drm/i915/display/intel_tc.h +++ b/drivers/gpu/drm/i915/display/intel_tc.h @@ -15,9 +15,9 @@ struct intel_encoder; bool intel_tc_port_in_tbt_alt_mode(struct intel_digital_port *dig_port); bool intel_tc_port_in_dp_alt_mode(struct intel_digital_port *dig_port); bool intel_tc_port_in_legacy_mode(struct intel_digital_port *dig_port); +bool intel_tc_port_handles_hpd_glitches(struct intel_digital_port *dig_port); bool intel_tc_port_connected(struct intel_encoder *encoder); -bool intel_tc_port_connected_locked(struct intel_encoder *encoder); u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port); int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port); diff --git a/drivers/gpu/drm/i915/display/intel_tv.c b/drivers/gpu/drm/i915/display/intel_tv.c index d4386cb3569e..a96bcfcf90a3 100644 --- a/drivers/gpu/drm/i915/display/intel_tv.c +++ b/drivers/gpu/drm/i915/display/intel_tv.c @@ -40,6 +40,7 @@ #include "intel_crtc.h" #include "intel_de.h" #include "intel_display_irq.h" +#include "intel_display_driver.h" #include "intel_display_types.h" #include "intel_dpll.h" #include "intel_hotplug.h" @@ -1327,7 +1328,7 @@ intel_tv_compute_config(struct intel_encoder *encoder, * the active portion. Hence following this formula seems * more trouble that it's worth. * - * if (GRAPHICS_VER(dev_priv) == 4) { + * if (DISPLAY_VER(dev_priv) == 4) { * num = cdclk * (tv_mode->oversample >> !tv_mode->progressive); * den = tv_mode->clock; * } else { @@ -1723,6 +1724,9 @@ intel_tv_detect(struct drm_connector *connector, if (!intel_display_device_enabled(i915)) return connector_status_disconnected; + if (!intel_display_driver_check_access(i915)) + return connector->status; + if (force) { struct drm_atomic_state *state; @@ -1990,6 +1994,7 @@ intel_tv_init(struct drm_i915_private *dev_priv) * More recent chipsets favour HDMI rather than integrated S-Video. */ intel_connector->polled = DRM_CONNECTOR_POLL_CONNECT; + intel_connector->base.polled = intel_connector->polled; drm_connector_init(&dev_priv->drm, connector, &intel_tv_connector_funcs, DRM_MODE_CONNECTOR_SVIDEO); diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c index fe256bf7b485..baf7354cb6e2 100644 --- a/drivers/gpu/drm/i915/display/intel_vblank.c +++ b/drivers/gpu/drm/i915/display/intel_vblank.c @@ -5,6 +5,7 @@ #include "i915_drv.h" #include "i915_reg.h" +#include "intel_crtc.h" #include "intel_de.h" #include "intel_display_types.h" #include "intel_vblank.h" @@ -581,3 +582,132 @@ void intel_crtc_update_active_timings(const struct intel_crtc_state *crtc_state, intel_vblank_section_exit(i915); spin_unlock_irqrestore(&i915->drm.vblank_time_lock, irqflags); } + +static int intel_mode_vblank_start(const struct drm_display_mode *mode) +{ + int vblank_start = mode->crtc_vblank_start; + + if (mode->flags & DRM_MODE_FLAG_INTERLACE) + vblank_start = DIV_ROUND_UP(vblank_start, 2); + + return vblank_start; +} + +void intel_vblank_evade_init(const struct intel_crtc_state *old_crtc_state, + const struct intel_crtc_state *new_crtc_state, + struct intel_vblank_evade_ctx *evade) +{ + struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->uapi.crtc); + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + const struct intel_crtc_state *crtc_state; + const struct drm_display_mode *adjusted_mode; + + evade->crtc = crtc; + + evade->need_vlv_dsi_wa = (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) && + intel_crtc_has_type(new_crtc_state, INTEL_OUTPUT_DSI); + + /* + * During fastsets/etc. the transcoder is still + * running with the old timings at this point. + * + * TODO: maybe just use the active timings here? + */ + if (intel_crtc_needs_modeset(new_crtc_state)) + crtc_state = new_crtc_state; + else + crtc_state = old_crtc_state; + + adjusted_mode = &crtc_state->hw.adjusted_mode; + + if (crtc->mode_flags & I915_MODE_FLAG_VRR) { + /* timing changes should happen with VRR disabled */ + drm_WARN_ON(crtc->base.dev, intel_crtc_needs_modeset(new_crtc_state) || + new_crtc_state->update_m_n || new_crtc_state->update_lrr); + + if (intel_vrr_is_push_sent(crtc_state)) + evade->vblank_start = intel_vrr_vmin_vblank_start(crtc_state); + else + evade->vblank_start = intel_vrr_vmax_vblank_start(crtc_state); + } else { + evade->vblank_start = intel_mode_vblank_start(adjusted_mode); + } + + /* FIXME needs to be calibrated sensibly */ + evade->min = evade->vblank_start - intel_usecs_to_scanlines(adjusted_mode, + VBLANK_EVASION_TIME_US); + evade->max = evade->vblank_start - 1; + + /* + * M/N and TRANS_VTOTAL are double buffered on the transcoder's + * undelayed vblank, so with seamless M/N and LRR we must evade + * both vblanks. + * + * DSB execution waits for the transcoder's undelayed vblank, + * hence we must kick off the commit before that. + */ + if (new_crtc_state->dsb || new_crtc_state->update_m_n || new_crtc_state->update_lrr) + evade->min -= adjusted_mode->crtc_vblank_start - adjusted_mode->crtc_vdisplay; +} + +/* must be called with vblank interrupt already enabled! */ +int intel_vblank_evade(struct intel_vblank_evade_ctx *evade) +{ + struct intel_crtc *crtc = evade->crtc; + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + long timeout = msecs_to_jiffies_timeout(1); + wait_queue_head_t *wq = drm_crtc_vblank_waitqueue(&crtc->base); + DEFINE_WAIT(wait); + int scanline; + + if (evade->min <= 0 || evade->max <= 0) + return 0; + + for (;;) { + /* + * prepare_to_wait() has a memory barrier, which guarantees + * other CPUs can see the task state update by the time we + * read the scanline. + */ + prepare_to_wait(wq, &wait, TASK_UNINTERRUPTIBLE); + + scanline = intel_get_crtc_scanline(crtc); + if (scanline < evade->min || scanline > evade->max) + break; + + if (!timeout) { + drm_err(&i915->drm, + "Potential atomic update failure on pipe %c\n", + pipe_name(crtc->pipe)); + break; + } + + local_irq_enable(); + + timeout = schedule_timeout(timeout); + + local_irq_disable(); + } + + finish_wait(wq, &wait); + + /* + * On VLV/CHV DSI the scanline counter would appear to + * increment approx. 1/3 of a scanline before start of vblank. + * The registers still get latched at start of vblank however. + * This means we must not write any registers on the first + * line of vblank (since not the whole line is actually in + * vblank). And unfortunately we can't use the interrupt to + * wait here since it will fire too soon. We could use the + * frame start interrupt instead since it will fire after the + * critical scanline, but that would require more changes + * in the interrupt code. So for now we'll just do the nasty + * thing and poll for the bad scanline to pass us by. + * + * FIXME figure out if BXT+ DSI suffers from this as well + */ + while (evade->need_vlv_dsi_wa && scanline == evade->vblank_start) + scanline = intel_get_crtc_scanline(crtc); + + return scanline; +} diff --git a/drivers/gpu/drm/i915/display/intel_vblank.h b/drivers/gpu/drm/i915/display/intel_vblank.h index 17636f140c71..ec6c3da3eeac 100644 --- a/drivers/gpu/drm/i915/display/intel_vblank.h +++ b/drivers/gpu/drm/i915/display/intel_vblank.h @@ -13,6 +13,18 @@ struct drm_crtc; struct intel_crtc; struct intel_crtc_state; +struct intel_vblank_evade_ctx { + struct intel_crtc *crtc; + int min, max, vblank_start; + bool need_vlv_dsi_wa; +}; + +void intel_vblank_evade_init(const struct intel_crtc_state *old_crtc_state, + const struct intel_crtc_state *new_crtc_state, + struct intel_vblank_evade_ctx *evade); +/* must be called with vblank interrupt already enabled! */ +int intel_vblank_evade(struct intel_vblank_evade_ctx *evade); + u32 i915_get_vblank_counter(struct drm_crtc *crtc); u32 g4x_get_vblank_counter(struct drm_crtc *crtc); bool intel_crtc_get_vblank_timestamp(struct drm_crtc *crtc, int *max_error, diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index 511dc1544854..392d93e97bf8 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -2624,3 +2624,31 @@ skl_get_initial_plane_config(struct intel_crtc *crtc, error: kfree(intel_fb); } + +bool skl_fixup_initial_plane_config(struct intel_crtc *crtc, + const struct intel_initial_plane_config *plane_config) +{ + struct drm_i915_private *i915 = to_i915(crtc->base.dev); + struct intel_plane *plane = to_intel_plane(crtc->base.primary); + const struct intel_plane_state *plane_state = + to_intel_plane_state(plane->base.state); + enum plane_id plane_id = plane->id; + enum pipe pipe = crtc->pipe; + u32 base; + + if (!plane_state->uapi.visible) + return false; + + base = intel_plane_ggtt_offset(plane_state); + + /* + * We may have moved the surface to a different + * part of ggtt, make the plane aware of that. + */ + if (plane_config->base == base) + return false; + + intel_de_write(i915, PLANE_SURF(pipe, plane_id), base); + + return true; +} diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.h b/drivers/gpu/drm/i915/display/skl_universal_plane.h index be64c201f9b3..e92e00c01b29 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.h +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.h @@ -22,6 +22,8 @@ skl_universal_plane_create(struct drm_i915_private *dev_priv, void skl_get_initial_plane_config(struct intel_crtc *crtc, struct intel_initial_plane_config *plane_config); +bool skl_fixup_initial_plane_config(struct intel_crtc *crtc, + const struct intel_initial_plane_config *plane_config); int skl_format_to_fourcc(int format, bool rgb_order, bool alpha); diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c index 56588d6e24ae..614f319d754e 100644 --- a/drivers/gpu/drm/i915/display/skl_watermark.c +++ b/drivers/gpu/drm/i915/display/skl_watermark.c @@ -443,12 +443,35 @@ static int intel_compute_sagv_mask(struct intel_atomic_state *state) for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) { + struct skl_pipe_wm *pipe_wm = &new_crtc_state->wm.skl.optimal; + new_bw_state = intel_atomic_get_bw_state(state); if (IS_ERR(new_bw_state)) return PTR_ERR(new_bw_state); old_bw_state = intel_atomic_get_old_bw_state(state); + /* + * We store use_sagv_wm in the crtc state rather than relying on + * that bw state since we have no convenient way to get at the + * latter from the plane commit hooks (especially in the legacy + * cursor case). + * + * drm_atomic_check_only() gets upset if we pull more crtcs + * into the state, so we have to calculate this based on the + * individual intel_crtc_can_enable_sagv() rather than + * the overall intel_can_enable_sagv(). Otherwise the + * crtcs not included in the commit would not switch to the + * SAGV watermarks when we are about to enable SAGV, and that + * would lead to underruns. This does mean extra power draw + * when only a subset of the crtcs are blocking SAGV as the + * other crtcs can't be allowed to use the more optimal + * normal (ie. non-SAGV) watermarks. + */ + pipe_wm->use_sagv_wm = !HAS_HW_SAGV_WM(i915) && + DISPLAY_VER(i915) >= 12 && + intel_crtc_can_enable_sagv(new_crtc_state); + if (intel_crtc_can_enable_sagv(new_crtc_state)) new_bw_state->pipe_sagv_reject &= ~BIT(crtc->pipe); else @@ -478,21 +501,6 @@ static int intel_compute_sagv_mask(struct intel_atomic_state *state) return ret; } - for_each_new_intel_crtc_in_state(state, crtc, - new_crtc_state, i) { - struct skl_pipe_wm *pipe_wm = &new_crtc_state->wm.skl.optimal; - - /* - * We store use_sagv_wm in the crtc state rather than relying on - * that bw state since we have no convenient way to get at the - * latter from the plane commit hooks (especially in the legacy - * cursor case) - */ - pipe_wm->use_sagv_wm = !HAS_HW_SAGV_WM(i915) && - DISPLAY_VER(i915) >= 12 && - intel_can_enable_sagv(i915, new_bw_state); - } - return 0; } @@ -1367,7 +1375,7 @@ skl_total_relative_data_rate(const struct intel_crtc_state *crtc_state) u64 data_rate = 0; for_each_plane_id_on_crtc(crtc, plane_id) { - if (plane_id == PLANE_CURSOR && DISPLAY_VER(i915) < 20) + if (plane_id == PLANE_CURSOR) continue; data_rate += crtc_state->rel_data_rate[plane_id]; @@ -1514,12 +1522,10 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state, return 0; /* Allocate fixed number of blocks for cursor. */ - if (DISPLAY_VER(i915) < 20) { - cursor_size = skl_cursor_allocation(crtc_state, num_active); - iter.size -= cursor_size; - skl_ddb_entry_init(&crtc_state->wm.skl.plane_ddb[PLANE_CURSOR], - alloc->end - cursor_size, alloc->end); - } + cursor_size = skl_cursor_allocation(crtc_state, num_active); + iter.size -= cursor_size; + skl_ddb_entry_init(&crtc_state->wm.skl.plane_ddb[PLANE_CURSOR], + alloc->end - cursor_size, alloc->end); iter.data_rate = skl_total_relative_data_rate(crtc_state); @@ -1533,7 +1539,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state, const struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; - if (plane_id == PLANE_CURSOR && DISPLAY_VER(i915) < 20) { + if (plane_id == PLANE_CURSOR) { const struct skl_ddb_entry *ddb = &crtc_state->wm.skl.plane_ddb[plane_id]; @@ -1581,7 +1587,7 @@ skl_crtc_allocate_plane_ddb(struct intel_atomic_state *state, const struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; - if (plane_id == PLANE_CURSOR && DISPLAY_VER(i915) < 20) + if (plane_id == PLANE_CURSOR) continue; if (DISPLAY_VER(i915) < 11 && diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 555022c0652c..d3a771afb083 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -2160,12 +2160,6 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb) #ifdef CONFIG_MMU_NOTIFIER if (!err && (eb->args->flags & __EXEC_USERPTR_USED)) { - read_lock(&eb->i915->mm.notifier_lock); - - /* - * count is always at least 1, otherwise __EXEC_USERPTR_USED - * could not have been set - */ for (i = 0; i < count; i++) { struct eb_vma *ev = &eb->vma[i]; struct drm_i915_gem_object *obj = ev->vma->obj; @@ -2177,8 +2171,6 @@ static int eb_move_to_gpu(struct i915_execbuffer *eb) if (err) break; } - - read_unlock(&eb->i915->mm.notifier_lock); } #endif diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c index 0d812f4d787d..3b27218aabe2 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c @@ -28,6 +28,13 @@ void i915_gem_suspend(struct drm_i915_private *i915) GEM_TRACE("%s\n", dev_name(i915->drm.dev)); intel_wakeref_auto(&i915->runtime_pm.userfault_wakeref, 0); + /* + * On rare occasions, we've observed the fence completion triggers + * free_engines asynchronously via rcu_call. Ensure those are done. + * This path is only called on suspend, so it's an acceptable cost. + */ + rcu_barrier(); + flush_workqueue(i915->wq); /* @@ -160,6 +167,9 @@ void i915_gem_suspend_late(struct drm_i915_private *i915) * machine in an unusable condition. */ + /* Like i915_gem_suspend, flush tasks staged from fence triggers */ + rcu_barrier(); + for_each_gt(gt, i915, i) intel_gt_suspend_late(gt); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_region.c b/drivers/gpu/drm/i915/gem/i915_gem_region.c index a4fb577eceb4..b09b74a2448b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_region.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_region.c @@ -129,7 +129,7 @@ i915_gem_object_create_region_at(struct intel_memory_region *mem, return ERR_PTR(-EINVAL); if (!(flags & I915_BO_ALLOC_GPU_ONLY) && - offset + size > mem->io_size && + offset + size > resource_size(&mem->io) && !i915_ggtt_has_aperture(to_gt(mem->i915)->ggtt)) return ERR_PTR(-ENOSPC); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c index 8c88075eeab2..ad6dd7f3259b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c @@ -541,7 +541,9 @@ static int i915_gem_init_stolen(struct intel_memory_region *mem) /* Exclude the reserved region from driver use */ mem->region.end = i915->dsm.reserved.start - 1; - mem->io_size = min(mem->io_size, resource_size(&mem->region)); + mem->io = DEFINE_RES_MEM(mem->io.start, + min(resource_size(&mem->io), + resource_size(&mem->region))); i915->dsm.usable_size = resource_size(&mem->region); @@ -752,7 +754,7 @@ static int _i915_gem_object_stolen_init(struct intel_memory_region *mem, * With discrete devices, where we lack a mappable aperture there is no * possible way to ever access this memory on the CPU side. */ - if (mem->type == INTEL_MEMORY_STOLEN_LOCAL && !mem->io_size && + if (mem->type == INTEL_MEMORY_STOLEN_LOCAL && !resource_size(&mem->io) && !(flags & I915_BO_ALLOC_GPU_ONLY)) return -ENOSPC; @@ -826,7 +828,6 @@ static const struct intel_memory_region_ops i915_region_stolen_smem_ops = { static int init_stolen_lmem(struct intel_memory_region *mem) { - struct drm_i915_private *i915 = mem->i915; int err; if (GEM_WARN_ON(resource_size(&mem->region) == 0)) @@ -838,14 +839,10 @@ static int init_stolen_lmem(struct intel_memory_region *mem) return 0; } - if (mem->io_size && - !io_mapping_init_wc(&mem->iomap, mem->io_start, mem->io_size)) + if (resource_size(&mem->io) && + !io_mapping_init_wc(&mem->iomap, mem->io.start, resource_size(&mem->io))) goto err_cleanup; - drm_dbg(&i915->drm, "Stolen Local memory IO start: %pa\n", - &mem->io_start); - drm_dbg(&i915->drm, "Stolen Local DSM base: %pa\n", &mem->region.start); - return 0; err_cleanup: @@ -855,7 +852,7 @@ err_cleanup: static int release_stolen_lmem(struct intel_memory_region *mem) { - if (mem->io_size) + if (resource_size(&mem->io)) io_mapping_fini(&mem->iomap); i915_gem_cleanup_stolen(mem->i915); return 0; @@ -938,13 +935,17 @@ i915_gem_stolen_lmem_setup(struct drm_i915_private *i915, u16 type, GEM_BUG_ON((dsm_base + dsm_size) > lmem_size); } else { /* Use DSM base address instead for stolen memory */ - dsm_base = intel_uncore_read64(uncore, GEN12_DSMBASE) & GEN12_BDSM_MASK; + dsm_base = intel_uncore_read64(uncore, GEN6_DSMBASE) & GEN11_BDSM_MASK; if (WARN_ON(lmem_size < dsm_base)) return ERR_PTR(-ENODEV); dsm_size = ALIGN_DOWN(lmem_size - dsm_base, SZ_1M); } - if (pci_resource_len(pdev, GEN12_LMEM_BAR) < lmem_size) { + if (i915_direct_stolen_access(i915)) { + drm_dbg(&i915->drm, "Using direct DSM access\n"); + io_start = intel_uncore_read64(uncore, GEN6_DSMBASE) & GEN11_BDSM_MASK; + io_size = dsm_size; + } else if (pci_resource_len(pdev, GEN12_LMEM_BAR) < lmem_size) { io_start = 0; io_size = 0; } else { diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index a6b0aaf30cbe..27dcfd8a34bb 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -142,13 +142,13 @@ i915_ttm_place_from_region(const struct intel_memory_region *mr, place->fpfn = offset >> PAGE_SHIFT; WARN_ON(overflows_type(place->fpfn + (size >> PAGE_SHIFT), place->lpfn)); place->lpfn = place->fpfn + (size >> PAGE_SHIFT); - } else if (mr->io_size && mr->io_size < mr->total) { + } else if (resource_size(&mr->io) && resource_size(&mr->io) < mr->total) { if (flags & I915_BO_ALLOC_GPU_ONLY) { place->flags |= TTM_PL_FLAG_TOPDOWN; } else { place->fpfn = 0; - WARN_ON(overflows_type(mr->io_size >> PAGE_SHIFT, place->lpfn)); - place->lpfn = mr->io_size >> PAGE_SHIFT; + WARN_ON(overflows_type(resource_size(&mr->io) >> PAGE_SHIFT, place->lpfn)); + place->lpfn = resource_size(&mr->io) >> PAGE_SHIFT; } } } @@ -1083,7 +1083,7 @@ static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) struct intel_memory_region *mr = obj->mm.placements[i]; unsigned int flags; - if (!mr->io_size && mr->type != INTEL_MEMORY_SYSTEM) + if (!resource_size(&mr->io) && mr->type != INTEL_MEMORY_SYSTEM) continue; flags = obj->flags; @@ -1094,8 +1094,9 @@ static vm_fault_t vm_fault_ttm(struct vm_fault *vmf) } if (err) { - drm_dbg(dev, "Unable to make resource CPU accessible(err = %pe)\n", - ERR_PTR(err)); + drm_dbg_ratelimited(dev, + "Unable to make resource CPU accessible(err = %pe)\n", + ERR_PTR(err)); dma_resv_unlock(bo->base.resv); ret = VM_FAULT_SIGBUS; goto out_rpm; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 1d3ebdf4069b..0e21ce9d3e5a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -42,7 +42,6 @@ #include "i915_drv.h" #include "i915_gem_ioctls.h" #include "i915_gem_object.h" -#include "i915_gem_userptr.h" #include "i915_scatterlist.h" #ifdef CONFIG_MMU_NOTIFIER @@ -61,36 +60,7 @@ static bool i915_gem_userptr_invalidate(struct mmu_interval_notifier *mni, const struct mmu_notifier_range *range, unsigned long cur_seq) { - struct drm_i915_gem_object *obj = container_of(mni, struct drm_i915_gem_object, userptr.notifier); - struct drm_i915_private *i915 = to_i915(obj->base.dev); - long r; - - if (!mmu_notifier_range_blockable(range)) - return false; - - write_lock(&i915->mm.notifier_lock); - mmu_interval_set_seq(mni, cur_seq); - - write_unlock(&i915->mm.notifier_lock); - - /* - * We don't wait when the process is exiting. This is valid - * because the object will be cleaned up anyway. - * - * This is also temporarily required as a hack, because we - * cannot currently force non-consistent batch buffers to preempt - * and reschedule by waiting on it, hanging processes on exit. - */ - if (current->flags & PF_EXITING) - return true; - - /* we will unbind on next submission, still have userptr pins */ - r = dma_resv_wait_timeout(obj->base.resv, DMA_RESV_USAGE_BOOKKEEP, false, - MAX_SCHEDULE_TIMEOUT); - if (r <= 0) - drm_err(&i915->drm, "(%ld) failed to wait for idle\n", r); - return true; } @@ -580,15 +550,3 @@ i915_gem_userptr_ioctl(struct drm_device *dev, #endif } -int i915_gem_init_userptr(struct drm_i915_private *dev_priv) -{ -#ifdef CONFIG_MMU_NOTIFIER - rwlock_init(&dev_priv->mm.notifier_lock); -#endif - - return 0; -} - -void i915_gem_cleanup_userptr(struct drm_i915_private *dev_priv) -{ -} diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.h b/drivers/gpu/drm/i915/gem/i915_gem_userptr.h deleted file mode 100644 index 8dadb2f8436d..000000000000 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.h +++ /dev/null @@ -1,14 +0,0 @@ -/* SPDX-License-Identifier: MIT */ -/* - * Copyright © 2021 Intel Corporation - */ - -#ifndef __I915_GEM_USERPTR_H__ -#define __I915_GEM_USERPTR_H__ - -struct drm_i915_private; - -int i915_gem_init_userptr(struct drm_i915_private *dev_priv); -void i915_gem_cleanup_userptr(struct drm_i915_private *dev_priv); - -#endif /* __I915_GEM_USERPTR_H__ */ diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index 2c51a2c452fc..99a9ade73956 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -1054,7 +1054,7 @@ static int igt_fill_mappable(struct intel_memory_region *mr, int err; total = 0; - size = mr->io_size; + size = resource_size(&mr->io); do { struct drm_i915_gem_object *obj; @@ -1315,28 +1315,28 @@ static int igt_mmap_migrate(void *arg) struct intel_memory_region *mixed[] = { mr, system }; struct intel_memory_region *single[] = { mr }; struct ttm_resource_manager *man = mr->region_private; - resource_size_t saved_io_size; + struct resource saved_io; int err; if (mr->private) continue; - if (!mr->io_size) + if (!resource_size(&mr->io)) continue; /* * For testing purposes let's force small BAR, if not already * present. */ - saved_io_size = mr->io_size; - if (mr->io_size == mr->total) { - resource_size_t io_size = mr->io_size; + saved_io = mr->io; + if (resource_size(&mr->io) == mr->total) { + resource_size_t io_size = resource_size(&mr->io); io_size = rounddown_pow_of_two(io_size >> 1); if (io_size < PAGE_SIZE) continue; - mr->io_size = io_size; + mr->io = DEFINE_RES_MEM(mr->io.start, io_size); i915_ttm_buddy_man_force_visible_size(man, io_size >> PAGE_SHIFT); } @@ -1396,9 +1396,9 @@ static int igt_mmap_migrate(void *arg) IGT_MMAP_MIGRATE_FAIL_GPU | IGT_MMAP_MIGRATE_UNFAULTABLE); out_io_size: - mr->io_size = saved_io_size; + mr->io = saved_io; i915_ttm_buddy_man_force_visible_size(man, - mr->io_size >> PAGE_SHIFT); + resource_size(&mr->io) >> PAGE_SHIFT); if (err) return err; } diff --git a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c index 86a04afff64b..e1bf13e3d307 100644 --- a/drivers/gpu/drm/i915/gt/gen8_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/gen8_engine_cs.c @@ -226,7 +226,7 @@ u32 *gen12_emit_aux_table_inv(struct intel_engine_cs *engine, u32 *cs) static int mtl_dummy_pipe_control(struct i915_request *rq) { /* Wa_14016712196 */ - if (IS_GFX_GT_IP_RANGE(rq->engine->gt, IP_VER(12, 70), IP_VER(12, 71)) || + if (IS_GFX_GT_IP_RANGE(rq->engine->gt, IP_VER(12, 70), IP_VER(12, 74)) || IS_DG2(rq->i915)) { u32 *cs; @@ -822,7 +822,7 @@ u32 *gen12_emit_fini_breadcrumb_rcs(struct i915_request *rq, u32 *cs) flags |= PIPE_CONTROL_FLUSH_L3; /* Wa_14016712196 */ - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71)) || IS_DG2(i915)) + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74)) || IS_DG2(i915)) /* dummy PIPE_CONTROL + depth flush */ cs = gen12_emit_pipe_control(cs, 0, PIPE_CONTROL_DEPTH_CACHE_FLUSH, 0); diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index 40687806d22a..1ade568ffbfa 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -1190,7 +1190,8 @@ static int intel_engine_init_tlb_invalidation(struct intel_engine_cs *engine) num = ARRAY_SIZE(xelpmp_regs); } } else { - if (GRAPHICS_VER_FULL(i915) == IP_VER(12, 71) || + if (GRAPHICS_VER_FULL(i915) == IP_VER(12, 74) || + GRAPHICS_VER_FULL(i915) == IP_VER(12, 71) || GRAPHICS_VER_FULL(i915) == IP_VER(12, 70) || GRAPHICS_VER_FULL(i915) == IP_VER(12, 50) || GRAPHICS_VER_FULL(i915) == IP_VER(12, 55)) { diff --git a/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c b/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c index 0f6406f0cca0..5f8d86e25993 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_heartbeat.c @@ -291,6 +291,9 @@ static int __intel_engine_pulse(struct intel_engine_cs *engine) heartbeat_commit(rq, &attr); GEM_BUG_ON(rq->sched.attr.priority < I915_PRIORITY_BARRIER); + /* Ensure the forced pulse gets a full period to execute */ + next_heartbeat(engine); + return 0; } diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c index 21a7e3191c18..ec1cbe229f0e 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c @@ -24,6 +24,7 @@ #include "intel_ring.h" #include "i915_drv.h" #include "i915_pci.h" +#include "i915_reg.h" #include "i915_request.h" #include "i915_scatterlist.h" #include "i915_utils.h" @@ -1152,13 +1153,20 @@ static unsigned int gen6_gttadr_offset(struct drm_i915_private *i915) static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size) { struct drm_i915_private *i915 = ggtt->vm.i915; + struct intel_uncore *uncore = ggtt->vm.gt->uncore; struct pci_dev *pdev = to_pci_dev(i915->drm.dev); phys_addr_t phys_addr; u32 pte_flags; int ret; GEM_WARN_ON(pci_resource_len(pdev, GEN4_GTTMMADR_BAR) != gen6_gttmmadr_size(i915)); - phys_addr = pci_resource_start(pdev, GEN4_GTTMMADR_BAR) + gen6_gttadr_offset(i915); + + if (i915_direct_stolen_access(i915)) { + drm_dbg(&i915->drm, "Using direct GSM access\n"); + phys_addr = intel_uncore_read64(uncore, GEN6_GSMBASE) & GEN11_BDSM_MASK; + } else { + phys_addr = pci_resource_start(pdev, GEN4_GTTMMADR_BAR) + gen6_gttadr_offset(i915); + } if (needs_wc_ggtt_mapping(i915)) ggtt->gsm = ioremap_wc(phys_addr, size); 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 f0dea54880af..c0b202223940 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_sysfs_pm.c @@ -176,27 +176,13 @@ static u32 get_residency(struct intel_gt *gt, enum intel_rc6_res_type id) return DIV_ROUND_CLOSEST_ULL(res, 1000); } -static u8 get_rc6_mask(struct intel_gt *gt) -{ - u8 mask = 0; - - if (HAS_RC6(gt->i915)) - mask |= BIT(0); - if (HAS_RC6p(gt->i915)) - mask |= BIT(1); - if (HAS_RC6pp(gt->i915)) - mask |= BIT(2); - - return mask; -} - static ssize_t rc6_enable_show(struct kobject *kobj, struct kobj_attribute *attr, char *buff) { struct intel_gt *gt = intel_gt_sysfs_get_drvdata(kobj, attr->attr.name); - return sysfs_emit(buff, "%x\n", get_rc6_mask(gt)); + return sysfs_emit(buff, "%x\n", gt->rc6.enabled); } static ssize_t rc6_enable_dev_show(struct device *dev, @@ -205,7 +191,7 @@ static ssize_t rc6_enable_dev_show(struct device *dev, { struct intel_gt *gt = intel_gt_sysfs_get_drvdata(&dev->kobj, attr->attr.name); - return sysfs_emit(buff, "%x\n", get_rc6_mask(gt)); + return sysfs_emit(buff, "%x\n", gt->rc6.enabled); } static u32 __rc6_residency_ms_show(struct intel_gt *gt) diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c index 86f73fe558ca..7811a8c9da06 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.c +++ b/drivers/gpu/drm/i915/gt/intel_gtt.c @@ -24,7 +24,8 @@ bool i915_ggtt_require_binder(struct drm_i915_private *i915) { /* Wa_13010847436 & Wa_14019519902 */ - return MEDIA_VER_FULL(i915) == IP_VER(13, 0); + return !i915_direct_stolen_access(i915) && + MEDIA_VER_FULL(i915) == IP_VER(13, 0); } static bool intel_ggtt_update_needs_vtd_wa(struct drm_i915_private *i915) diff --git a/drivers/gpu/drm/i915/gt/intel_mocs.c b/drivers/gpu/drm/i915/gt/intel_mocs.c index 353f93baaca0..25c1023eb5f9 100644 --- a/drivers/gpu/drm/i915/gt/intel_mocs.c +++ b/drivers/gpu/drm/i915/gt/intel_mocs.c @@ -495,7 +495,7 @@ static unsigned int get_mocs_settings(struct drm_i915_private *i915, memset(table, 0, sizeof(struct drm_i915_mocs_table)); table->unused_entries_index = I915_MOCS_PTE; - if (IS_GFX_GT_IP_RANGE(to_gt(i915), IP_VER(12, 70), IP_VER(12, 71))) { + if (IS_GFX_GT_IP_RANGE(to_gt(i915), IP_VER(12, 70), IP_VER(12, 74))) { table->size = ARRAY_SIZE(mtl_mocs_table); table->table = mtl_mocs_table; table->n_entries = MTL_NUM_MOCS_ENTRIES; diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c index 7090e4be29cb..8f4b3c8af09c 100644 --- a/drivers/gpu/drm/i915/gt/intel_rc6.c +++ b/drivers/gpu/drm/i915/gt/intel_rc6.c @@ -123,7 +123,7 @@ static void gen11_rc6_enable(struct intel_rc6 *rc6) * temporary wa and should be removed after fixing real cause * of forcewake timeouts. */ - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71))) + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) pg_enable = GEN9_MEDIA_PG_ENABLE | GEN11_MEDIA_SAMPLER_PG_ENABLE; diff --git a/drivers/gpu/drm/i915/gt/intel_region_lmem.c b/drivers/gpu/drm/i915/gt/intel_region_lmem.c index f8512aee58a8..51bb27e10a4f 100644 --- a/drivers/gpu/drm/i915/gt/intel_region_lmem.c +++ b/drivers/gpu/drm/i915/gt/intel_region_lmem.c @@ -144,8 +144,8 @@ region_lmem_init(struct intel_memory_region *mem) int ret; if (!io_mapping_init_wc(&mem->iomap, - mem->io_start, - mem->io_size)) + mem->io.start, + resource_size(&mem->io))) return -EIO; ret = intel_region_ttm_init(mem); @@ -240,7 +240,7 @@ static struct intel_memory_region *setup_lmem(struct intel_gt *gt) lmem_size -= tile_stolen; } else { /* Stolen starts from GSMBASE without CCS */ - lmem_size = intel_uncore_read64(&i915->uncore, GEN12_GSMBASE); + lmem_size = intel_uncore_read64(&i915->uncore, GEN6_GSMBASE); } i915_resize_lmem_bar(i915, lmem_size); @@ -273,14 +273,6 @@ static struct intel_memory_region *setup_lmem(struct intel_gt *gt) if (err) goto err_region_put; - drm_dbg(&i915->drm, "Local memory: %pR\n", &mem->region); - drm_dbg(&i915->drm, "Local memory IO start: %pa\n", - &mem->io_start); - drm_info(&i915->drm, "Local memory IO size: %pa\n", - &mem->io_size); - drm_info(&i915->drm, "Local memory available: %pa\n", - &lmem_size); - if (io_size < lmem_size) drm_info(&i915->drm, "Using a reduced BAR size of %lluMiB. Consider enabling 'Resizable BAR' or similar, if available in the BIOS.\n", (u64)io_size >> 20); diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 86756a252e20..d67d44611c28 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -789,8 +789,13 @@ static void xelpg_ctx_gt_tuning_init(struct intel_engine_cs *engine, dg2_ctx_gt_tuning_init(engine, wal); - if (IS_GFX_GT_IP_STEP(gt, IP_VER(12, 70), STEP_B0, STEP_FOREVER) || - IS_GFX_GT_IP_STEP(gt, IP_VER(12, 71), STEP_B0, STEP_FOREVER)) + /* + * Due to Wa_16014892111, the DRAW_WATERMARK tuning must be done in + * gen12_emit_indirect_ctx_rcs() rather than here on some early + * steppings. + */ + if (!(IS_GFX_GT_IP_STEP(gt, IP_VER(12, 70), STEP_A0, STEP_B0) || + IS_GFX_GT_IP_STEP(gt, IP_VER(12, 71), STEP_A0, STEP_B0))) wa_add(wal, DRAW_WATERMARK, VERT_WM_VAL, 0x3FF, 0, false); } @@ -820,6 +825,9 @@ static void xelpg_ctx_workarounds_init(struct intel_engine_cs *engine, /* Wa_18019271663 */ wa_masked_en(wal, CACHE_MODE_1, MSAA_OPTIMIZATION_REDUC_DISABLE); + + /* Wa_14019877138 */ + wa_mcr_masked_en(wal, XEHP_PSS_CHICKEN, FD_END_COLLECT); } static void fakewa_disable_nestedbb_mode(struct intel_engine_cs *engine, @@ -908,7 +916,7 @@ __intel_engine_init_ctx_wa(struct intel_engine_cs *engine, if (engine->class != RENDER_CLASS) goto done; - if (IS_GFX_GT_IP_RANGE(engine->gt, IP_VER(12, 70), IP_VER(12, 71))) + if (IS_GFX_GT_IP_RANGE(engine->gt, IP_VER(12, 70), IP_VER(12, 74))) xelpg_ctx_workarounds_init(engine, wal); else if (IS_PONTEVECCHIO(i915)) ; /* noop; none at this time */ @@ -1644,7 +1652,7 @@ pvc_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) static void xelpg_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) { - /* Wa_14018778641 / Wa_18018781329 */ + /* Wa_14018575942 / Wa_18018781329 */ wa_mcr_write_or(wal, COMP_MOD_CTRL, FORCE_MISS_FTLB); /* Wa_22016670082 */ @@ -1711,7 +1719,7 @@ xelpmp_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) */ static void gt_tuning_settings(struct intel_gt *gt, struct i915_wa_list *wal) { - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71))) { + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) { wa_mcr_write_or(wal, XEHP_L3SCQREG7, BLEND_FILL_CACHING_OPT_DIS); wa_mcr_write_or(wal, XEHP_SQCM, EN_32B_ACCESS); } @@ -1744,7 +1752,7 @@ gt_init_workarounds(struct intel_gt *gt, struct i915_wa_list *wal) return; } - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71))) + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) xelpg_gt_workarounds_init(gt, wal); else if (IS_PONTEVECCHIO(i915)) pvc_gt_workarounds_init(gt, wal); @@ -2217,7 +2225,7 @@ void intel_engine_init_whitelist(struct intel_engine_cs *engine) if (engine->gt->type == GT_MEDIA) ; /* none yet */ - else if (IS_GFX_GT_IP_RANGE(engine->gt, IP_VER(12, 70), IP_VER(12, 71))) + else if (IS_GFX_GT_IP_RANGE(engine->gt, IP_VER(12, 70), IP_VER(12, 74))) xelpg_whitelist_build(engine); else if (IS_PONTEVECCHIO(i915)) pvc_whitelist_build(engine); @@ -2829,7 +2837,7 @@ add_render_compute_tuning_settings(struct intel_gt *gt, { struct drm_i915_private *i915 = gt->i915; - if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71)) || IS_DG2(i915)) + if (IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74)) || IS_DG2(i915)) wa_mcr_write_clr_set(wal, RT_CTRL, STACKID_CTRL, STACKID_CTRL_512); /* @@ -2882,7 +2890,8 @@ general_render_compute_wa_init(struct intel_engine_cs *engine, struct i915_wa_li } if (IS_GFX_GT_IP_STEP(gt, IP_VER(12, 70), STEP_B0, STEP_FOREVER) || - IS_GFX_GT_IP_STEP(gt, IP_VER(12, 71), STEP_B0, STEP_FOREVER)) + IS_GFX_GT_IP_STEP(gt, IP_VER(12, 71), STEP_B0, STEP_FOREVER) || + IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 74), IP_VER(12, 74))) /* Wa_14017856879 */ wa_mcr_masked_en(wal, GEN9_ROW_CHICKEN3, MTL_DISABLE_FIX_FOR_EOT_FLUSH); diff --git a/drivers/gpu/drm/i915/gt/selftest_rc6.c b/drivers/gpu/drm/i915/gt/selftest_rc6.c index a7189c2d660c..1aa1446c8fb0 100644 --- a/drivers/gpu/drm/i915/gt/selftest_rc6.c +++ b/drivers/gpu/drm/i915/gt/selftest_rc6.c @@ -62,12 +62,12 @@ int live_rc6_manual(void *arg) dt = ktime_get(); rc0_power = librapl_energy_uJ(); - msleep(250); + msleep(1000); rc0_power = librapl_energy_uJ() - rc0_power; dt = ktime_sub(ktime_get(), dt); res[1] = rc6_residency(rc6); if ((res[1] - res[0]) >> 10) { - pr_err("RC6 residency increased by %lldus while disabled for 250ms!\n", + pr_err("RC6 residency increased by %lldus while disabled for 1000ms!\n", (res[1] - res[0]) >> 10); err = -EINVAL; goto out_unlock; diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c index 00b872b6380b..3941f2d6fa47 100644 --- a/drivers/gpu/drm/i915/gt/selftest_tlb.c +++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c @@ -206,8 +206,8 @@ static struct drm_i915_gem_object *create_lmem(struct intel_gt *gt) * of pages. To succeed with both allocations, especially in case of Small * BAR, try to allocate no more than quarter of mappable memory. */ - if (mr && size > mr->io_size / 4) - size = mr->io_size / 4; + if (mr && size > resource_size(&mr->io) / 4) + size = resource_size(&mr->io) / 4; return i915_gem_object_create_lmem(gt->i915, size, I915_BO_ALLOC_CONTIGUOUS); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c index 63724e17829a..f7372f736a77 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ads.c @@ -377,8 +377,13 @@ static int guc_mmio_regset_init(struct temp_regset *regset, CCS_MASK(engine->gt)) ret |= GUC_MMIO_REG_ADD(gt, regset, GEN12_RCU_MODE, true); + /* + * some of the WA registers are MCR registers. As it is safe to + * use MCR form for non-MCR registers, for code simplicity, all + * WA registers are added with MCR form. + */ for (i = 0, wa = wal->list; i < wal->count; i++, wa++) - ret |= GUC_MMIO_REG_ADD(gt, regset, wa->reg, wa->masked_reg); + ret |= GUC_MCR_REG_ADD(gt, regset, wa->mcr_reg, wa->masked_reg); /* Be extra paranoid and include all whitelist registers. */ for (i = 0; i < RING_MAX_NONPRIV_SLOTS; i++) @@ -394,13 +399,13 @@ static int guc_mmio_regset_init(struct temp_regset *regset, ret |= GUC_MMIO_REG_ADD(gt, regset, GEN9_LNCFCMOCS(i), false); if (GRAPHICS_VER(engine->i915) >= 12) { - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL0, false); - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL1, false); - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL2, false); - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL3, false); - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL4, false); - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL5, false); - ret |= GUC_MMIO_REG_ADD(gt, regset, EU_PERF_CNTL6, false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL0)), false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL1)), false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL2)), false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL3)), false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL4)), false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL5)), false); + ret |= GUC_MCR_REG_ADD(gt, regset, MCR_REG(i915_mmio_reg_offset(EU_PERF_CNTL6)), false); } return ret ? -1 : 0; 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 0f79cb658518..52332bb14339 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c @@ -184,7 +184,7 @@ static int guc_wait_ucode(struct intel_guc *guc) * in the seconds range. However, there is a limit on how long an * individual wait_for() can wait. So wrap it in a loop. */ - before_freq = intel_rps_read_actual_frequency(&uncore->gt->rps); + before_freq = intel_rps_read_actual_frequency(>->rps); before = ktime_get(); for (count = 0; count < GUC_LOAD_RETRY_LIMIT; count++) { ret = wait_for(guc_load_done(uncore, &status, &success), 1000); @@ -192,7 +192,7 @@ static int guc_wait_ucode(struct intel_guc *guc) break; guc_dbg(guc, "load still in progress, count = %d, freq = %dMHz, status = 0x%08X [0x%02X/%02X]\n", - count, intel_rps_read_actual_frequency(&uncore->gt->rps), status, + count, intel_rps_read_actual_frequency(>->rps), status, REG_FIELD_GET(GS_BOOTROM_MASK, status), REG_FIELD_GET(GS_UKERNEL_MASK, status)); } @@ -204,7 +204,7 @@ static int guc_wait_ucode(struct intel_guc *guc) u32 bootrom = REG_FIELD_GET(GS_BOOTROM_MASK, status); guc_info(guc, "load failed: status = 0x%08X, time = %lldms, freq = %dMHz, ret = %d\n", - status, delta_ms, intel_rps_read_actual_frequency(&uncore->gt->rps), ret); + status, delta_ms, intel_rps_read_actual_frequency(>->rps), ret); guc_info(guc, "load failed: status: Reset = %d, BootROM = 0x%02X, UKernel = 0x%02X, MIA = 0x%02X, Auth = 0x%02X\n", REG_FIELD_GET(GS_MIA_IN_RESET, status), bootrom, ukernel, @@ -254,11 +254,11 @@ static int guc_wait_ucode(struct intel_guc *guc) 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_rps_read_actual_frequency(>->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), + delta_ms, intel_rps_read_actual_frequency(>->rps), before_freq, status, count, ret); } 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 a259f1118c5a..f3dcae4b9d45 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -236,6 +236,13 @@ set_context_destroyed(struct intel_context *ce) ce->guc_state.sched_state |= SCHED_STATE_DESTROYED; } +static inline void +clr_context_destroyed(struct intel_context *ce) +{ + lockdep_assert_held(&ce->guc_state.lock); + ce->guc_state.sched_state &= ~SCHED_STATE_DESTROYED; +} + static inline bool context_pending_disable(struct intel_context *ce) { return ce->guc_state.sched_state & SCHED_STATE_PENDING_DISABLE; @@ -613,6 +620,8 @@ static int guc_submission_send_busy_loop(struct intel_guc *guc, u32 g2h_len_dw, bool loop) { + int ret; + /* * We always loop when a send requires a reply (i.e. g2h_len_dw > 0), * so we don't handle the case where we don't get a reply because we @@ -623,7 +632,11 @@ static int guc_submission_send_busy_loop(struct intel_guc *guc, if (g2h_len_dw) atomic_inc(&guc->outstanding_submission_g2h); - return intel_guc_send_busy_loop(guc, action, len, g2h_len_dw, loop); + ret = intel_guc_send_busy_loop(guc, action, len, g2h_len_dw, loop); + if (ret) + atomic_dec(&guc->outstanding_submission_g2h); + + return ret; } int intel_guc_wait_for_pending_msg(struct intel_guc *guc, @@ -1362,7 +1375,45 @@ static void guc_enable_busyness_worker(struct intel_guc *guc) static void guc_cancel_busyness_worker(struct intel_guc *guc) { - cancel_delayed_work_sync(&guc->timestamp.work); + /* + * There are many different call stacks that can get here. Some of them + * hold the reset mutex. The busyness worker also attempts to acquire the + * reset mutex. Synchronously flushing a worker thread requires acquiring + * the worker mutex. Lockdep sees this as a conflict. It thinks that the + * flush can deadlock because it holds the worker mutex while waiting for + * the reset mutex, but another thread is holding the reset mutex and might + * attempt to use other worker functions. + * + * In practice, this scenario does not exist because the busyness worker + * does not block waiting for the reset mutex. It does a try-lock on it and + * immediately exits if the lock is already held. Unfortunately, the mutex + * in question (I915_RESET_BACKOFF) is an i915 implementation which has lockdep + * annotation but not to the extent of explaining the 'might lock' is also a + * 'does not need to lock'. So one option would be to add more complex lockdep + * annotations to ignore the issue (if at all possible). A simpler option is to + * just not flush synchronously when a rest in progress. Given that the worker + * will just early exit and re-schedule itself anyway, there is no advantage + * to running it immediately. + * + * If a reset is not in progress, then the synchronous flush may be required. + * As noted many call stacks lead here, some during suspend and driver unload + * which do require a synchronous flush to make sure the worker is stopped + * before memory is freed. + * + * Trying to pass a 'need_sync' or 'in_reset' flag all the way down through + * every possible call stack is unfeasible. It would be too intrusive to many + * areas that really don't care about the GuC backend. However, there is the + * 'reset_in_progress' flag available, so just use that. + * + * And note that in the case of a reset occurring during driver unload + * (wedge_on_fini), skipping the cancel in _prepare (when the reset flag is set + * is fine because there is another cancel in _finish (when the reset flag is + * not). + */ + if (guc_to_gt(guc)->uc.reset_in_progress) + cancel_delayed_work(&guc->timestamp.work); + else + cancel_delayed_work_sync(&guc->timestamp.work); } static void __reset_guc_busyness_stats(struct intel_guc *guc) @@ -1613,6 +1664,11 @@ static void guc_flush_submissions(struct intel_guc *guc) spin_unlock_irqrestore(&sched_engine->lock, flags); } +void intel_guc_submission_flush_work(struct intel_guc *guc) +{ + flush_work(&guc->submission_state.destroyed_worker); +} + static void guc_flush_destroyed_contexts(struct intel_guc *guc); void intel_guc_submission_reset_prepare(struct intel_guc *guc) @@ -1948,8 +2004,16 @@ void intel_guc_submission_cancel_requests(struct intel_guc *guc) void intel_guc_submission_reset_finish(struct intel_guc *guc) { + /* + * Ensure the busyness worker gets cancelled even on a fatal wedge. + * Note that reset_prepare is not allowed to because it confuses lockdep. + */ + if (guc_submission_initialized(guc)) + guc_cancel_busyness_worker(guc); + /* Reset called during driver load or during wedge? */ if (unlikely(!guc_submission_initialized(guc) || + !intel_guc_is_fw_running(guc) || intel_gt_is_wedged(guc_to_gt(guc)))) { return; } @@ -3283,12 +3347,13 @@ static void guc_context_close(struct intel_context *ce) spin_unlock_irqrestore(&ce->guc_state.lock, flags); } -static inline void guc_lrc_desc_unpin(struct intel_context *ce) +static inline int guc_lrc_desc_unpin(struct intel_context *ce) { struct intel_guc *guc = ce_to_guc(ce); struct intel_gt *gt = guc_to_gt(guc); unsigned long flags; bool disabled; + int ret; GEM_BUG_ON(!intel_gt_pm_is_awake(gt)); GEM_BUG_ON(!ctx_id_mapped(guc, ce->guc_id.id)); @@ -3299,18 +3364,41 @@ static inline void guc_lrc_desc_unpin(struct intel_context *ce) spin_lock_irqsave(&ce->guc_state.lock, flags); disabled = submission_disabled(guc); if (likely(!disabled)) { + /* + * Take a gt-pm ref and change context state to be destroyed. + * NOTE: a G2H IRQ that comes after will put this gt-pm ref back + */ __intel_gt_pm_get(gt); set_context_destroyed(ce); clr_context_registered(ce); } spin_unlock_irqrestore(&ce->guc_state.lock, flags); + if (unlikely(disabled)) { release_guc_id(guc, ce); __guc_context_destroy(ce); - return; + return 0; } - deregister_context(ce, ce->guc_id.id); + /* + * GuC is active, lets destroy this context, but at this point we can still be racing + * with suspend, so we undo everything if the H2G fails in deregister_context so + * that GuC reset will find this context during clean up. + */ + ret = deregister_context(ce, ce->guc_id.id); + if (ret) { + spin_lock(&ce->guc_state.lock); + set_context_registered(ce); + clr_context_destroyed(ce); + spin_unlock(&ce->guc_state.lock); + /* + * As gt-pm is awake at function entry, intel_wakeref_put_async merely decrements + * the wakeref immediately but per function spec usage call this after unlock. + */ + intel_wakeref_put_async(>->wakeref); + } + + return ret; } static void __guc_context_destroy(struct intel_context *ce) @@ -3378,7 +3466,22 @@ static void deregister_destroyed_contexts(struct intel_guc *guc) if (!ce) break; - guc_lrc_desc_unpin(ce); + if (guc_lrc_desc_unpin(ce)) { + /* + * This means GuC's CT link severed mid-way which could happen + * in suspend-resume corner cases. In this case, put the + * context back into the destroyed_contexts list which will + * get picked up on the next context deregistration event or + * purged in a GuC sanitization event (reset/unload/wedged/...). + */ + spin_lock_irqsave(&guc->submission_state.lock, flags); + list_add_tail(&ce->destroyed_link, + &guc->submission_state.destroyed_contexts); + spin_unlock_irqrestore(&guc->submission_state.lock, flags); + /* Bail now since the list might never be emptied if h2gs fail */ + break; + } + } } @@ -3389,6 +3492,17 @@ static void destroyed_worker_func(struct work_struct *w) struct intel_gt *gt = guc_to_gt(guc); intel_wakeref_t wakeref; + /* + * In rare cases we can get here via async context-free fence-signals that + * come very late in suspend flow or very early in resume flows. In these + * cases, GuC won't be ready but just skipping it here is fine as these + * pending-destroy-contexts get destroyed totally at GuC reset time at the + * end of suspend.. OR.. this worker can be picked up later on the next + * context destruction trigger after resume-completes + */ + if (!intel_guc_is_ready(guc)) + return; + with_intel_gt_pm(gt, wakeref) deregister_destroyed_contexts(guc); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h index c57b29cdb1a6..b6df75622d3b 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.h @@ -38,6 +38,8 @@ int intel_guc_wait_for_pending_msg(struct intel_guc *guc, bool interruptible, long timeout); +void intel_guc_submission_flush_work(struct intel_guc *guc); + static inline bool intel_guc_submission_is_supported(struct intel_guc *guc) { return guc->submission_supported; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c index ba9e07fc2b57..0945b177d5f9 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c @@ -6,6 +6,7 @@ #include <linux/types.h> #include "gt/intel_gt.h" +#include "gt/intel_rps.h" #include "intel_guc_reg.h" #include "intel_huc.h" #include "intel_huc_print.h" @@ -447,17 +448,68 @@ static const char *auth_mode_string(struct intel_huc *huc, return partial ? "clear media" : "all workloads"; } +/* + * Use a longer timeout for debug builds so that problems can be detected + * and analysed. But a shorter timeout for releases so that user's don't + * wait forever to find out there is a problem. Note that the only reason + * an end user should hit the timeout is in case of extreme thermal throttling. + * And a system that is that hot during boot is probably dead anyway! + */ +#if defined(CONFIG_DRM_I915_DEBUG_GEM) +#define HUC_LOAD_RETRY_LIMIT 20 +#else +#define HUC_LOAD_RETRY_LIMIT 3 +#endif + int intel_huc_wait_for_auth_complete(struct intel_huc *huc, enum intel_huc_authentication_type type) { struct intel_gt *gt = huc_to_gt(huc); - int ret; + struct intel_uncore *uncore = gt->uncore; + ktime_t before, after, delta; + int ret, count; + u64 delta_ms; + u32 before_freq; - ret = __intel_wait_for_register(gt->uncore, - huc->status[type].reg, - huc->status[type].mask, - huc->status[type].value, - 2, 50, NULL); + /* + * The KMD requests maximum frequency during driver load, however thermal + * throttling can force the frequency down to minimum (although the board + * really should never get that hot in real life!). IFWI issues have been + * seen to cause sporadic failures to grant the higher frequency. And at + * minimum frequency, the authentication time can be in the seconds range. + * Note that there is a limit on how long an individual wait_for() can wait. + * So wrap it in a loop. + */ + before_freq = intel_rps_read_actual_frequency(>->rps); + before = ktime_get(); + for (count = 0; count < HUC_LOAD_RETRY_LIMIT; count++) { + ret = __intel_wait_for_register(gt->uncore, + huc->status[type].reg, + huc->status[type].mask, + huc->status[type].value, + 2, 1000, NULL); + if (!ret) + break; + + huc_dbg(huc, "auth still in progress, count = %d, freq = %dMHz, status = 0x%08X\n", + count, intel_rps_read_actual_frequency(>->rps), + huc->status[type].reg.reg); + } + after = ktime_get(); + delta = ktime_sub(after, before); + delta_ms = ktime_to_ms(delta); + + if (delta_ms > 50) { + huc_warn(huc, "excessive auth time: %lldms! [status = 0x%08X, count = %d, ret = %d]\n", + delta_ms, huc->status[type].reg.reg, count, ret); + huc_warn(huc, "excessive auth time: [freq = %dMHz, before = %dMHz, perf_limit_reasons = 0x%08X]\n", + intel_rps_read_actual_frequency(>->rps), before_freq, + intel_uncore_read(uncore, intel_gt_perf_limit_reasons_reg(gt))); + } else { + huc_dbg(huc, "auth took %lldms, freq = %dMHz, before = %dMHz, status = 0x%08X, count = %d, ret = %d\n", + delta_ms, intel_rps_read_actual_frequency(>->rps), + before_freq, huc->status[type].reg.reg, count, ret); + } /* mark the load process as complete even if the wait failed */ delayed_huc_load_complete(huc); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index 3872d309ed31..6dfe5d9456c6 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -640,7 +640,7 @@ void intel_uc_reset_finish(struct intel_uc *uc) uc->reset_in_progress = false; /* Firmware expected to be running when this function is called */ - if (intel_guc_is_fw_running(guc) && intel_uc_uses_guc_submission(uc)) + if (intel_uc_uses_guc_submission(uc)) intel_guc_submission_reset_finish(guc); } @@ -690,6 +690,8 @@ void intel_uc_suspend(struct intel_uc *uc) return; } + intel_guc_submission_flush_work(guc); + with_intel_runtime_pm(&uc_to_gt(uc)->i915->runtime_pm, wakeref) { err = intel_guc_suspend(guc); if (err) diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c index faf21be724c3..4f74d867fe1a 100644 --- a/drivers/gpu/drm/i915/gvt/kvmgt.c +++ b/drivers/gpu/drm/i915/gvt/kvmgt.c @@ -574,7 +574,7 @@ int intel_gvt_set_opregion(struct intel_vgpu *vgpu) ret = intel_vgpu_register_reg(vgpu, PCI_VENDOR_ID_INTEL | VFIO_REGION_TYPE_PCI_VENDOR_TYPE, VFIO_REGION_SUBTYPE_INTEL_IGD_OPREGION, - &intel_vgpu_regops_opregion, OPREGION_SIZE, + &intel_vgpu_regops_opregion, INTEL_GVT_OPREGION_SIZE, VFIO_REGION_INFO_FLAG_READ, base); return ret; diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index db99c2ef66db..990eaa029d9c 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -147,7 +147,7 @@ static const char *i915_cache_level_str(struct drm_i915_gem_object *obj) { struct drm_i915_private *i915 = obj_to_i915(obj); - if (IS_GFX_GT_IP_RANGE(to_gt(i915), IP_VER(12, 70), IP_VER(12, 71))) { + if (IS_GFX_GT_IP_RANGE(to_gt(i915), IP_VER(12, 70), IP_VER(12, 74))) { switch (obj->pat_index) { case 0: return " WB"; case 1: return " WT"; diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index c1cf5ba7ba14..9ee902d5b72c 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -1004,8 +1004,10 @@ void i915_driver_shutdown(struct drm_i915_private *i915) intel_runtime_pm_disable(&i915->runtime_pm); intel_power_domains_disable(i915); + intel_fbdev_set_suspend(&i915->drm, FBINFO_STATE_SUSPENDED, true); if (HAS_DISPLAY(i915)) { drm_kms_helper_poll_disable(&i915->drm); + intel_display_driver_disable_user_access(i915); drm_atomic_helper_shutdown(&i915->drm); } @@ -1015,6 +1017,9 @@ void i915_driver_shutdown(struct drm_i915_private *i915) intel_runtime_pm_disable_interrupts(i915); intel_hpd_cancel_work(i915); + if (HAS_DISPLAY(i915)) + intel_display_driver_suspend_access(i915); + intel_suspend_encoders(i915); intel_shutdown_encoders(i915); @@ -1081,8 +1086,11 @@ static int i915_drm_suspend(struct drm_device *dev) /* We do a lot of poking in a lot of registers, make sure they work * properly. */ intel_power_domains_disable(dev_priv); - if (HAS_DISPLAY(dev_priv)) + intel_fbdev_set_suspend(dev, FBINFO_STATE_SUSPENDED, true); + if (HAS_DISPLAY(dev_priv)) { drm_kms_helper_poll_disable(dev); + intel_display_driver_disable_user_access(dev_priv); + } pci_save_state(pdev); @@ -1093,6 +1101,9 @@ static int i915_drm_suspend(struct drm_device *dev) intel_runtime_pm_disable_interrupts(dev_priv); intel_hpd_cancel_work(dev_priv); + if (HAS_DISPLAY(dev_priv)) + intel_display_driver_suspend_access(dev_priv); + intel_suspend_encoders(dev_priv); /* Must be called before GGTT is suspended. */ @@ -1104,8 +1115,6 @@ static int i915_drm_suspend(struct drm_device *dev) opregion_target_state = suspend_to_idle(dev_priv) ? PCI_D1 : PCI_D3cold; intel_opregion_suspend(dev_priv, opregion_target_state); - intel_fbdev_set_suspend(dev, FBINFO_STATE_SUSPENDED, true); - dev_priv->suspend_count++; intel_dmc_suspend(dev_priv); @@ -1244,15 +1253,21 @@ static int i915_drm_resume(struct drm_device *dev) intel_display_driver_init_hw(dev_priv); intel_clock_gating_init(dev_priv); + + if (HAS_DISPLAY(dev_priv)) + intel_display_driver_resume_access(dev_priv); + intel_hpd_init(dev_priv); /* MST sideband requires HPD interrupts enabled */ intel_dp_mst_resume(dev_priv); intel_display_driver_resume(dev_priv); - intel_hpd_poll_disable(dev_priv); - if (HAS_DISPLAY(dev_priv)) + if (HAS_DISPLAY(dev_priv)) { + intel_display_driver_enable_user_access(dev_priv); drm_kms_helper_poll_enable(dev); + } + intel_hpd_poll_disable(dev_priv); intel_opregion_resume(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 861567362abd..e81b3b2858ac 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -165,14 +165,6 @@ struct i915_gem_mm { struct notifier_block vmap_notifier; struct shrinker *shrinker; -#ifdef CONFIG_MMU_NOTIFIER - /** - * notifier_lock for mmu notifiers, memory may not be allocated - * while holding this lock. - */ - rwlock_t notifier_lock; -#endif - /* shrinker accounting, also useful for userland debugging */ u64 shrink_memory; u32 shrink_count; diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 92758b6b41f0..1391c01d7663 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -48,7 +48,6 @@ #include "gem/i915_gem_object_frontbuffer.h" #include "gem/i915_gem_pm.h" #include "gem/i915_gem_region.h" -#include "gem/i915_gem_userptr.h" #include "gt/intel_engine_user.h" #include "gt/intel_gt.h" #include "gt/intel_gt_pm.h" @@ -1165,10 +1164,6 @@ int i915_gem_init(struct drm_i915_private *dev_priv) if (intel_vgpu_active(dev_priv) && !intel_vgpu_has_huge_gtt(dev_priv)) RUNTIME_INFO(dev_priv)->page_sizes = I915_GTT_PAGE_SIZE_4K; - ret = i915_gem_init_userptr(dev_priv); - if (ret) - return ret; - for_each_gt(gt, dev_priv, i) { intel_uc_fetch_firmwares(>->uc); intel_wopcm_init(>->wopcm); diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index d04660b60046..a0b784ebaddd 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -1157,7 +1157,7 @@ i915_vma_coredump_create(const struct intel_gt *gt, dma_addr_t offset = dma - mem->region.start; void __iomem *s; - if (offset + PAGE_SIZE > mem->io_size) { + if (offset + PAGE_SIZE > resource_size(&mem->io)) { ret = -EINVAL; break; } diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 2d695818f006..bd9d812b1afa 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -3225,7 +3225,7 @@ u32 i915_perf_oa_timestamp_frequency(struct drm_i915_private *i915) struct intel_gt *gt = to_gt(i915); /* Wa_18013179988 */ - if (IS_DG2(i915) || IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 71))) { + if (IS_DG2(i915) || IS_GFX_GT_IP_RANGE(gt, IP_VER(12, 70), IP_VER(12, 74))) { intel_wakeref_t wakeref; u32 reg, shift; diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c index 00871ef99792..3baa2f54a86e 100644 --- a/drivers/gpu/drm/i915/i915_query.c +++ b/drivers/gpu/drm/i915/i915_query.c @@ -502,7 +502,7 @@ static int query_memregion_info(struct drm_i915_private *i915, info.probed_size = mr->total; if (mr->type == INTEL_MEMORY_LOCAL) - info.probed_cpu_visible_size = mr->io_size; + info.probed_cpu_visible_size = resource_size(&mr->io); else info.probed_cpu_visible_size = mr->total; @@ -551,6 +551,38 @@ static int query_hwconfig_blob(struct drm_i915_private *i915, return hwconfig->size; } +static int +query_guc_submission_version(struct drm_i915_private *i915, + struct drm_i915_query_item *query) +{ + struct drm_i915_query_guc_submission_version __user *query_ptr = + u64_to_user_ptr(query->data_ptr); + struct drm_i915_query_guc_submission_version ver; + struct intel_guc *guc = &to_gt(i915)->uc.guc; + const size_t size = sizeof(ver); + int ret; + + if (!intel_uc_uses_guc_submission(&to_gt(i915)->uc)) + return -ENODEV; + + ret = copy_query_item(&ver, size, size, query); + if (ret != 0) + return ret; + + if (ver.branch || ver.major || ver.minor || ver.patch) + return -EINVAL; + + ver.branch = 0; + ver.major = guc->submission_version.major; + ver.minor = guc->submission_version.minor; + ver.patch = guc->submission_version.patch; + + if (copy_to_user(query_ptr, &ver, size)) + return -EFAULT; + + return 0; +} + static int (* const i915_query_funcs[])(struct drm_i915_private *dev_priv, struct drm_i915_query_item *query_item) = { query_topology_info, @@ -559,6 +591,7 @@ static int (* const i915_query_funcs[])(struct drm_i915_private *dev_priv, query_memregion_info, query_hwconfig_blob, query_geometry_subslices, + query_guc_submission_version, }; int i915_query_ioctl(struct drm_device *dev, void *data, struct drm_file *file) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 27dc903f0553..e00557e1a57f 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3059,6 +3059,7 @@ #define MCURSOR_MODE_64_ARGB_AX (0x20 | MCURSOR_MODE_64_32B_AX) #define _CURABASE 0x70084 #define _CURAPOS 0x70088 +#define _CURAPOS_ERLY_TPT 0x7008c #define CURSOR_POS_Y_SIGN REG_BIT(31) #define CURSOR_POS_Y_MASK REG_GENMASK(30, 16) #define CURSOR_POS_Y(y) REG_FIELD_PREP(CURSOR_POS_Y_MASK, (y)) @@ -3087,6 +3088,7 @@ #define CURCNTR(pipe) _MMIO_CURSOR2(pipe, _CURACNTR) #define CURBASE(pipe) _MMIO_CURSOR2(pipe, _CURABASE) #define CURPOS(pipe) _MMIO_CURSOR2(pipe, _CURAPOS) +#define CURPOS_ERLY_TPT(pipe) _MMIO_CURSOR2(pipe, _CURAPOS_ERLY_TPT) #define CURSIZE(pipe) _MMIO_CURSOR2(pipe, _CURASIZE) #define CUR_FBC_CTL(pipe) _MMIO_CURSOR2(pipe, _CUR_FBC_CTL_A) #define CUR_CHICKEN(pipe) _MMIO_CURSOR2(pipe, _CUR_CHICKEN_A) @@ -5412,6 +5414,9 @@ #define GEN6_PCODE_FREQ_RING_RATIO_SHIFT 16 #define GEN6_PCODE_DATA1 _MMIO(0x13812C) +#define MTL_PCODE_STOLEN_ACCESS _MMIO(0x138914) +#define STOLEN_ACCESS_ALLOWED 0x1 + /* IVYBRIDGE DPF */ #define GEN7_L3CDERRST1(slice) _MMIO(0xB008 + (slice) * 0x200) /* L3CD Error Status 1 */ #define GEN7_L3CDERRST1_ROW_MASK (0x7ff << 14) @@ -5652,6 +5657,10 @@ enum skl_power_gate { #define DP_TP_CTL_MODE_SST (0 << 27) #define DP_TP_CTL_MODE_MST (1 << 27) #define DP_TP_CTL_FORCE_ACT (1 << 25) +#define DP_TP_CTL_TRAIN_PAT4_SEL_MASK (3 << 19) +#define DP_TP_CTL_TRAIN_PAT4_SEL_TP4A (0 << 19) +#define DP_TP_CTL_TRAIN_PAT4_SEL_TP4B (1 << 19) +#define DP_TP_CTL_TRAIN_PAT4_SEL_TP4C (2 << 19) #define DP_TP_CTL_ENHANCED_FRAME_ENABLE (1 << 18) #define DP_TP_CTL_FDI_AUTOTRAIN (1 << 15) #define DP_TP_CTL_LINK_TRAIN_MASK (7 << 8) @@ -5684,6 +5693,8 @@ enum skl_power_gate { /* Known as DDI_CTL_DE in MTL+ */ #define DDI_BUF_CTL(port) _MMIO_PORT(port, _DDI_BUF_CTL_A, _DDI_BUF_CTL_B) #define DDI_BUF_CTL_ENABLE (1 << 31) +#define XE2LPD_DDI_BUF_D2D_LINK_ENABLE REG_BIT(29) +#define XE2LPD_DDI_BUF_D2D_LINK_STATE REG_BIT(28) #define DDI_BUF_TRANS_SELECT(n) ((n) << 24) #define DDI_BUF_EMP_MASK (0xf << 24) #define DDI_BUF_PHY_LINK_RATE(r) ((r) << 20) @@ -6314,9 +6325,10 @@ enum skl_power_gate { #define GMS_MASK REG_GENMASK(15, 8) #define GGMS_MASK REG_GENMASK(7, 6) -#define GEN12_GSMBASE _MMIO(0x108100) -#define GEN12_DSMBASE _MMIO(0x1080C0) -#define GEN12_BDSM_MASK REG_GENMASK64(63, 20) +#define GEN6_GSMBASE _MMIO(0x108100) +#define GEN6_DSMBASE _MMIO(0x1080C0) +#define GEN6_BDSM_MASK REG_GENMASK64(31, 20) +#define GEN11_BDSM_MASK REG_GENMASK64(63, 20) #define XEHP_CLOCK_GATE_DIS _MMIO(0x101014) #define SGSI_SIDECLK_DIS REG_BIT(17) diff --git a/drivers/gpu/drm/i915/i915_syncmap.c b/drivers/gpu/drm/i915/i915_syncmap.c index 60404dbb2e9f..df6437c37373 100644 --- a/drivers/gpu/drm/i915/i915_syncmap.c +++ b/drivers/gpu/drm/i915/i915_syncmap.c @@ -75,13 +75,10 @@ struct i915_syncmap { unsigned int height; unsigned int bitmap; struct i915_syncmap *parent; - /* - * Following this header is an array of either seqno or child pointers: - * union { - * u32 seqno[KSYNCMAP]; - * struct i915_syncmap *child[KSYNCMAP]; - * }; - */ + union { + DECLARE_FLEX_ARRAY(u32, seqno); + DECLARE_FLEX_ARRAY(struct i915_syncmap *, child); + }; }; /** @@ -99,13 +96,13 @@ void i915_syncmap_init(struct i915_syncmap **root) static inline u32 *__sync_seqno(struct i915_syncmap *p) { GEM_BUG_ON(p->height); - return (u32 *)(p + 1); + return p->seqno; } static inline struct i915_syncmap **__sync_child(struct i915_syncmap *p) { GEM_BUG_ON(!p->height); - return (struct i915_syncmap **)(p + 1); + return p->child; } static inline unsigned int @@ -200,7 +197,7 @@ __sync_alloc_leaf(struct i915_syncmap *parent, u64 id) { struct i915_syncmap *p; - p = kmalloc(sizeof(*p) + KSYNCMAP * sizeof(u32), GFP_KERNEL); + p = kmalloc(struct_size(p, seqno, KSYNCMAP), GFP_KERNEL); if (unlikely(!p)) return NULL; @@ -282,7 +279,7 @@ static noinline int __sync_set(struct i915_syncmap **root, u64 id, u32 seqno) unsigned int above; /* Insert a join above the current layer */ - next = kzalloc(sizeof(*next) + KSYNCMAP * sizeof(next), + next = kzalloc(struct_size(next, child, KSYNCMAP), GFP_KERNEL); if (unlikely(!next)) return -ENOMEM; diff --git a/drivers/gpu/drm/i915/i915_utils.c b/drivers/gpu/drm/i915/i915_utils.c index 29fd02bf5ea8..6f9e7b354b54 100644 --- a/drivers/gpu/drm/i915/i915_utils.c +++ b/drivers/gpu/drm/i915/i915_utils.c @@ -8,6 +8,7 @@ #include <drm/drm_drv.h> #include "i915_drv.h" +#include "i915_reg.h" #include "i915_utils.h" #define FDO_BUG_MSG "Please file a bug on drm/i915; see " FDO_BUG_URL " for details." @@ -125,3 +126,19 @@ bool i915_vtd_active(struct drm_i915_private *i915) /* Running as a guest, we assume the host is enforcing VT'd */ return i915_run_as_guest(); } + +bool i915_direct_stolen_access(struct drm_i915_private *i915) +{ + /* + * Wa_22018444074 + * + * Access via BAR can hang MTL, go directly to GSM/DSM, + * except for VM guests which won't have access to it. + * + * Normally this would not work but on MTL the system firmware + * should have relaxed the access permissions sufficiently. + * 0x138914==0x1 indicates that the firmware has done its job. + */ + return IS_METEORLAKE(i915) && !i915_run_as_guest() && + intel_uncore_read(&i915->uncore, MTL_PCODE_STOLEN_ACCESS) == STOLEN_ACCESS_ALLOWED; +} diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h index f98577967b7f..b45ef0560611 100644 --- a/drivers/gpu/drm/i915/i915_utils.h +++ b/drivers/gpu/drm/i915/i915_utils.h @@ -391,4 +391,6 @@ static inline bool i915_run_as_guest(void) bool i915_vtd_active(struct drm_i915_private *i915); +bool i915_direct_stolen_access(struct drm_i915_private *i915); + #endif /* !__I915_UTILS_H */ diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c index 60a03340bbd4..52d998e5c21a 100644 --- a/drivers/gpu/drm/i915/intel_memory_region.c +++ b/drivers/gpu/drm/i915/intel_memory_region.c @@ -50,7 +50,7 @@ static int __iopagetest(struct intel_memory_region *mem, if (memchr_inv(result, value, sizeof(result))) { dev_err(mem->i915->drm.dev, "Failed to read back from memory region:%pR at [%pa + %pa] for %ps; wrote %x, read (%x, %x, %x)\n", - &mem->region, &mem->io_start, &offset, caller, + &mem->region, &mem->io.start, &offset, caller, value, result[0], result[1], result[2]); return -EINVAL; } @@ -67,11 +67,11 @@ static int iopagetest(struct intel_memory_region *mem, int err; int i; - va = ioremap_wc(mem->io_start + offset, PAGE_SIZE); + va = ioremap_wc(mem->io.start + offset, PAGE_SIZE); if (!va) { dev_err(mem->i915->drm.dev, "Failed to ioremap memory region [%pa + %pa] for %ps\n", - &mem->io_start, &offset, caller); + &mem->io.start, &offset, caller); return -EFAULT; } @@ -102,10 +102,10 @@ static int iomemtest(struct intel_memory_region *mem, resource_size_t last, page; int err; - if (mem->io_size < PAGE_SIZE) + if (resource_size(&mem->io) < PAGE_SIZE) return 0; - last = mem->io_size - PAGE_SIZE; + last = resource_size(&mem->io) - PAGE_SIZE; /* * Quick test to check read/write access to the iomap (backing store). @@ -207,7 +207,7 @@ static int intel_memory_region_memtest(struct intel_memory_region *mem, struct drm_i915_private *i915 = mem->i915; int err = 0; - if (!mem->io_start) + if (!mem->io.start) return 0; if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) || i915->params.memtest) @@ -252,8 +252,7 @@ intel_memory_region_create(struct drm_i915_private *i915, mem->i915 = i915; mem->region = DEFINE_RES_MEM(start, size); - mem->io_start = io_start; - mem->io_size = io_size; + mem->io = DEFINE_RES_MEM(io_start, io_size); mem->min_page_size = min_page_size; mem->ops = ops; mem->total = size; @@ -373,6 +372,24 @@ int intel_memory_regions_hw_probe(struct drm_i915_private *i915) i915->mm.regions[i] = mem; } + for (i = 0; i < ARRAY_SIZE(i915->mm.regions); i++) { + struct intel_memory_region *mem = i915->mm.regions[i]; + u64 region_size, io_size; + + if (!mem) + continue; + + region_size = resource_size(&mem->region) >> 20; + io_size = resource_size(&mem->io) >> 20; + + if (resource_size(&mem->io)) + drm_dbg(&i915->drm, "Memory region(%d): %s: %llu MiB %pR, io: %llu MiB %pR\n", + mem->id, mem->name, region_size, &mem->region, io_size, &mem->io); + else + drm_dbg(&i915->drm, "Memory region(%d): %s: %llu MiB %pR, io: n/a\n", + mem->id, mem->name, region_size, &mem->region); + } + return 0; out_cleanup: diff --git a/drivers/gpu/drm/i915/intel_memory_region.h b/drivers/gpu/drm/i915/intel_memory_region.h index 9ba36454e51b..40810cfb3fd9 100644 --- a/drivers/gpu/drm/i915/intel_memory_region.h +++ b/drivers/gpu/drm/i915/intel_memory_region.h @@ -71,8 +71,7 @@ struct intel_memory_region { struct io_mapping iomap; struct resource region; - resource_size_t io_start; - resource_size_t io_size; + struct resource io; resource_size_t min_page_size; resource_size_t total; diff --git a/drivers/gpu/drm/i915/intel_region_ttm.c b/drivers/gpu/drm/i915/intel_region_ttm.c index bf6097e7433d..04525d92bec5 100644 --- a/drivers/gpu/drm/i915/intel_region_ttm.c +++ b/drivers/gpu/drm/i915/intel_region_ttm.c @@ -87,7 +87,7 @@ int intel_region_ttm_init(struct intel_memory_region *mem) ret = i915_ttm_buddy_man_init(bdev, mem_type, false, resource_size(&mem->region), - mem->io_size, + resource_size(&mem->io), mem->min_page_size, PAGE_SIZE); if (ret) return ret; @@ -219,16 +219,16 @@ intel_region_ttm_resource_alloc(struct intel_memory_region *mem, goto out; } place.lpfn = place.fpfn + (size >> PAGE_SHIFT); - } else if (mem->io_size && mem->io_size < mem->total) { + } else if (resource_size(&mem->io) && resource_size(&mem->io) < mem->total) { if (flags & I915_BO_ALLOC_GPU_ONLY) { place.flags |= TTM_PL_FLAG_TOPDOWN; } else { place.fpfn = 0; - if (WARN_ON(overflows_type(mem->io_size >> PAGE_SHIFT, place.lpfn))) { + if (WARN_ON(overflows_type(resource_size(&mem->io) >> PAGE_SHIFT, place.lpfn))) { ret = -E2BIG; goto out; } - place.lpfn = mem->io_size >> PAGE_SHIFT; + place.lpfn = resource_size(&mem->io) >> PAGE_SHIFT; } } diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index dfefad5a5fec..76400e9c40f0 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1800,7 +1800,10 @@ static const struct intel_forcewake_range __mtl_fw_ranges[] = { GEN_FW_RANGE(0x24000, 0x2ffff, 0), /* 0x24000 - 0x2407f: always on 0x24080 - 0x2ffff: reserved */ - GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_GT) + GEN_FW_RANGE(0x30000, 0x3ffff, FORCEWAKE_GT), + GEN_FW_RANGE(0x40000, 0x1901ef, 0), + GEN_FW_RANGE(0x1901f0, 0x1901f3, FORCEWAKE_GT) + /* FIXME: WA to wake GT while triggering H2G */ }; /* diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c index d985d9bae2e8..ae6070b5bf07 100644 --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c @@ -544,8 +544,8 @@ static u64 igt_object_mappable_total(struct drm_i915_gem_object *obj) u64 start = drm_buddy_block_offset(block); u64 end = start + drm_buddy_block_size(mm, block); - if (start < mr->io_size) - total += min_t(u64, end, mr->io_size) - start; + if (start < resource_size(&mr->io)) + total += min_t(u64, end, resource_size(&mr->io)) - start; } return total; diff --git a/drivers/gpu/drm/i915/soc/intel_pch.c b/drivers/gpu/drm/i915/soc/intel_pch.c index 240beafb38ed..3cad6dac06b0 100644 --- a/drivers/gpu/drm/i915/soc/intel_pch.c +++ b/drivers/gpu/drm/i915/soc/intel_pch.c @@ -140,11 +140,6 @@ intel_pch_type(const struct drm_i915_private *dev_priv, unsigned short id) drm_WARN_ON(&dev_priv->drm, !IS_ALDERLAKE_S(dev_priv) && !IS_ALDERLAKE_P(dev_priv)); return PCH_ADP; - case INTEL_PCH_MTP_DEVICE_ID_TYPE: - case INTEL_PCH_MTP2_DEVICE_ID_TYPE: - drm_dbg_kms(&dev_priv->drm, "Found Meteor Lake PCH\n"); - drm_WARN_ON(&dev_priv->drm, !IS_METEORLAKE(dev_priv)); - return PCH_MTP; default: return PCH_NONE; } @@ -173,9 +168,7 @@ intel_virt_detect_pch(const struct drm_i915_private *dev_priv, * make an educated guess as to which PCH is really there. */ - if (IS_METEORLAKE(dev_priv)) - id = INTEL_PCH_MTP_DEVICE_ID_TYPE; - else if (IS_ALDERLAKE_S(dev_priv) || IS_ALDERLAKE_P(dev_priv)) + if (IS_ALDERLAKE_S(dev_priv) || IS_ALDERLAKE_P(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; @@ -225,6 +218,13 @@ void intel_detect_pch(struct drm_i915_private *dev_priv) if (DISPLAY_VER(dev_priv) >= 20) { dev_priv->pch_type = PCH_LNL; return; + } else if (IS_METEORLAKE(dev_priv)) { + /* + * Both north display and south display are on the SoC die. + * The real PCH is uninvolved in display. + */ + dev_priv->pch_type = PCH_MTL; + return; } else if (IS_DG2(dev_priv)) { dev_priv->pch_type = PCH_DG2; return; diff --git a/drivers/gpu/drm/i915/soc/intel_pch.h b/drivers/gpu/drm/i915/soc/intel_pch.h index 1b03ea60a7a8..89e89ede265d 100644 --- a/drivers/gpu/drm/i915/soc/intel_pch.h +++ b/drivers/gpu/drm/i915/soc/intel_pch.h @@ -25,11 +25,11 @@ enum intel_pch { PCH_ICP, /* Ice Lake/Jasper Lake PCH */ PCH_TGP, /* Tiger Lake/Mule Creek Canyon PCH */ PCH_ADP, /* Alder Lake PCH */ - PCH_MTP, /* Meteor Lake PCH */ /* Fake PCHs, functionality handled on the same PCI dev */ PCH_DG1 = 1024, PCH_DG2, + PCH_MTL, PCH_LNL, }; @@ -59,16 +59,12 @@ enum intel_pch { #define INTEL_PCH_ADP2_DEVICE_ID_TYPE 0x5180 #define INTEL_PCH_ADP3_DEVICE_ID_TYPE 0x7A00 #define INTEL_PCH_ADP4_DEVICE_ID_TYPE 0x5480 -#define INTEL_PCH_MTP_DEVICE_ID_TYPE 0x7E00 -#define INTEL_PCH_MTP2_DEVICE_ID_TYPE 0xAE00 #define INTEL_PCH_P2X_DEVICE_ID_TYPE 0x7100 #define INTEL_PCH_P3X_DEVICE_ID_TYPE 0x7000 #define INTEL_PCH_QEMU_DEVICE_ID_TYPE 0x2900 /* qemu q35 has 2918 */ #define INTEL_PCH_TYPE(dev_priv) ((dev_priv)->pch_type) #define INTEL_PCH_ID(dev_priv) ((dev_priv)->pch_id) -#define HAS_PCH_LNL(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_LNL) -#define HAS_PCH_MTP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_MTP) #define HAS_PCH_DG2(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_DG2) #define HAS_PCH_ADP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_ADP) #define HAS_PCH_DG1(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_DG1) diff --git a/drivers/gpu/drm/radeon/atom-bits.h b/drivers/gpu/drm/radeon/atom-bits.h index e8fae5c77514..2bfd6d0ff050 100644 --- a/drivers/gpu/drm/radeon/atom-bits.h +++ b/drivers/gpu/drm/radeon/atom-bits.h @@ -33,7 +33,7 @@ static inline uint8_t get_u8(void *bios, int ptr) #define CU8(ptr) get_u8(ctx->bios, (ptr)) static inline uint16_t get_u16(void *bios, int ptr) { - return get_u8(bios ,ptr)|(((uint16_t)get_u8(bios, ptr+1))<<8); + return get_u8(bios, ptr)|(((uint16_t)get_u8(bios, ptr+1))<<8); } #define U16(ptr) get_u16(ctx->ctx->bios, (ptr)) #define CU16(ptr) get_u16(ctx->bios, (ptr)) diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c index ceb6d772ef94..5bc3e6b41c34 100644 --- a/drivers/gpu/drm/radeon/atom.c +++ b/drivers/gpu/drm/radeon/atom.c @@ -60,6 +60,7 @@ typedef struct { struct atom_context *ctx; uint32_t *ps, *ws; + int ps_size, ws_size; int ps_shift; uint16_t start; unsigned last_jump; @@ -68,8 +69,8 @@ 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 params_size); +int atom_execute_table(struct atom_context *ctx, int index, uint32_t *params, int params_size); static uint32_t atom_arg_mask[8] = { 0xFFFFFFFF, 0x0000FFFF, 0x00FFFF00, 0xFFFF0000, @@ -221,7 +222,10 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, (*ptr)++; /* get_unaligned_le32 avoids unaligned accesses from atombios * tables, noticed on a DEC Alpha. */ - val = get_unaligned_le32((u32 *)&ctx->ps[idx]); + if (idx < ctx->ps_size) + val = get_unaligned_le32((u32 *)&ctx->ps[idx]); + else + pr_info("PS index out of range: %i > %i\n", idx, ctx->ps_size); if (print) DEBUG("PS[0x%02X,0x%04X]", idx, val); break; @@ -259,7 +263,10 @@ static uint32_t atom_get_src_int(atom_exec_context *ctx, uint8_t attr, val = gctx->reg_block; break; default: - val = ctx->ws[idx]; + if (idx < ctx->ws_size) + val = ctx->ws[idx]; + else + pr_info("WS index out of range: %i > %i\n", idx, ctx->ws_size); } break; case ATOM_ARG_ID: @@ -494,6 +501,10 @@ static void atom_put_dst(atom_exec_context *ctx, int arg, uint8_t attr, idx = U8(*ptr); (*ptr)++; DEBUG("PS[0x%02X]", idx); + if (idx >= ctx->ps_size) { + pr_info("PS index out of range: %i > %i\n", idx, ctx->ps_size); + return; + } ctx->ps[idx] = cpu_to_le32(val); break; case ATOM_ARG_WS: @@ -526,6 +537,10 @@ static void atom_put_dst(atom_exec_context *ctx, int arg, uint8_t attr, gctx->reg_block = val; break; default: + if (idx >= ctx->ws_size) { + pr_info("WS index out of range: %i > %i\n", idx, ctx->ws_size); + return; + } ctx->ws[idx] = val; } break; @@ -623,7 +638,7 @@ static void atom_op_calltable(atom_exec_context *ctx, int *ptr, int arg) else SDEBUG(" table: %d\n", idx); if (U16(ctx->ctx->cmd_table + 4 + 2 * idx)) - r = atom_execute_table_locked(ctx->ctx, idx, ctx->ps + ctx->ps_shift); + r = atom_execute_table_locked(ctx->ctx, idx, ctx->ps + ctx->ps_shift, ctx->ps_size - ctx->ps_shift); if (r) { ctx->abort = true; } @@ -1152,7 +1167,7 @@ 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 params_size) { int base = CU16(ctx->cmd_table + 4 + 2 * index); int len, ws, ps, ptr; @@ -1174,12 +1189,16 @@ static int atom_execute_table_locked(struct atom_context *ctx, int index, uint32 ectx.ps_shift = ps / 4; ectx.start = base; ectx.ps = params; + ectx.ps_size = params_size; ectx.abort = false; ectx.last_jump = 0; - if (ws) + if (ws) { ectx.ws = kcalloc(4, ws, GFP_KERNEL); - else + ectx.ws_size = ws; + } else { ectx.ws = NULL; + ectx.ws_size = 0; + } debug_depth++; while (1) { @@ -1212,7 +1231,7 @@ 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 params_size) { int r; @@ -1228,16 +1247,16 @@ int atom_execute_table_scratch_unlocked(struct atom_context *ctx, int index, uin /* reset divmul */ ctx->divmul[0] = 0; ctx->divmul[1] = 0; - r = atom_execute_table_locked(ctx, index, params); + r = atom_execute_table_locked(ctx, index, params, params_size); mutex_unlock(&ctx->mutex); 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 params_size) { int r; mutex_lock(&ctx->scratch_mutex); - r = atom_execute_table_scratch_unlocked(ctx, index, params); + r = atom_execute_table_scratch_unlocked(ctx, index, params, params_size); mutex_unlock(&ctx->scratch_mutex); return r; } @@ -1335,7 +1354,7 @@ int atom_asic_init(struct atom_context *ctx) if (!CU16(ctx->cmd_table + 4 + 2 * ATOM_CMD_INIT)) return 1; - ret = atom_execute_table(ctx, ATOM_CMD_INIT, ps); + ret = atom_execute_table(ctx, ATOM_CMD_INIT, ps, 16); if (ret) return ret; @@ -1343,7 +1362,7 @@ int atom_asic_init(struct atom_context *ctx) if (rdev->family < CHIP_R600) { if (CU16(ctx->cmd_table + 4 + 2 * ATOM_CMD_SPDFANCNTL)) - atom_execute_table(ctx, ATOM_CMD_SPDFANCNTL, ps); + atom_execute_table(ctx, ATOM_CMD_SPDFANCNTL, ps, 16); } return ret; } diff --git a/drivers/gpu/drm/radeon/atom.h b/drivers/gpu/drm/radeon/atom.h index 5de0563b63d2..5bf06c0bd6ff 100644 --- a/drivers/gpu/drm/radeon/atom.h +++ b/drivers/gpu/drm/radeon/atom.h @@ -145,8 +145,8 @@ struct atom_context { extern int atom_debug; struct atom_context *atom_parse(struct card_info *, void *); -int atom_execute_table(struct atom_context *, int, uint32_t *); -int atom_execute_table_scratch_unlocked(struct atom_context *, int, uint32_t *); +int atom_execute_table(struct atom_context *, int, uint32_t *, int); +int atom_execute_table_scratch_unlocked(struct atom_context *, int, uint32_t *, int); int atom_asic_init(struct atom_context *); void atom_destroy(struct atom_context *); bool atom_parse_data_header(struct atom_context *ctx, int index, uint16_t *size, diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index ade13173921b..9b3a3a9d60e2 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -77,7 +77,7 @@ static void atombios_overscan_setup(struct drm_crtc *crtc, args.usOverscanTop = cpu_to_le16(radeon_crtc->v_border); break; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void atombios_scaler_setup(struct drm_crtc *crtc) @@ -157,7 +157,7 @@ static void atombios_scaler_setup(struct drm_crtc *crtc) break; } } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); if ((is_tv || is_cv) && rdev->family >= CHIP_RV515 && rdev->family <= CHIP_R580) { atom_rv515_force_tv_scaler(rdev, radeon_crtc); @@ -178,7 +178,7 @@ static void atombios_lock_crtc(struct drm_crtc *crtc, int lock) args.ucCRTC = radeon_crtc->crtc_id; args.ucEnable = lock; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void atombios_enable_crtc(struct drm_crtc *crtc, int state) @@ -194,7 +194,7 @@ static void atombios_enable_crtc(struct drm_crtc *crtc, int state) args.ucCRTC = radeon_crtc->crtc_id; args.ucEnable = state; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void atombios_enable_crtc_memreq(struct drm_crtc *crtc, int state) @@ -210,7 +210,7 @@ static void atombios_enable_crtc_memreq(struct drm_crtc *crtc, int state) args.ucCRTC = radeon_crtc->crtc_id; args.ucEnable = state; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static const u32 vga_control_regs[6] = @@ -242,7 +242,7 @@ static void atombios_blank_crtc(struct drm_crtc *crtc, int state) args.ucCRTC = radeon_crtc->crtc_id; args.ucBlanking = state; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); if (ASIC_IS_DCE8(rdev)) WREG32(vga_control_regs[radeon_crtc->crtc_id], vga_control); @@ -261,7 +261,7 @@ static void atombios_powergate_crtc(struct drm_crtc *crtc, int state) args.ucDispPipeId = radeon_crtc->crtc_id; args.ucEnable = state; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void atombios_crtc_dpms(struct drm_crtc *crtc, int mode) @@ -343,7 +343,7 @@ atombios_set_crtc_dtd_timing(struct drm_crtc *crtc, args.susModeMiscInfo.usAccess = cpu_to_le16(misc); args.ucCRTC = radeon_crtc->crtc_id; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void atombios_crtc_set_timing(struct drm_crtc *crtc, @@ -389,7 +389,7 @@ static void atombios_crtc_set_timing(struct drm_crtc *crtc, args.susModeMiscInfo.usAccess = cpu_to_le16(misc); args.ucCRTC = radeon_crtc->crtc_id; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void atombios_disable_ss(struct radeon_device *rdev, int pll_id) @@ -546,7 +546,7 @@ static void atombios_crtc_program_ss(struct radeon_device *rdev, args.lvds_ss.ucSpreadSpectrumStepSize_Delay |= (ss->delay & 7) << 4; args.lvds_ss.ucEnable = enable; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } union adjust_pixel_clock { @@ -692,7 +692,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, ADJUST_DISPLAY_CONFIG_SS_ENABLE; atom_execute_table(rdev->mode_info.atom_context, - index, (uint32_t *)&args); + index, (uint32_t *)&args, sizeof(args)); adjusted_clock = le16_to_cpu(args.v1.usPixelClock) * 10; break; case 3: @@ -725,7 +725,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, args.v3.sInput.ucExtTransmitterID = 0; atom_execute_table(rdev->mode_info.atom_context, - index, (uint32_t *)&args); + index, (uint32_t *)&args, sizeof(args)); adjusted_clock = le32_to_cpu(args.v3.sOutput.ulDispPllFreq) * 10; if (args.v3.sOutput.ucRefDiv) { radeon_crtc->pll_flags |= RADEON_PLL_USE_FRAC_FB_DIV; @@ -809,7 +809,7 @@ static void atombios_crtc_set_disp_eng_pll(struct radeon_device *rdev, DRM_ERROR("Unknown table version %d %d\n", frev, crev); return; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void atombios_crtc_program_pll(struct drm_crtc *crtc, @@ -949,7 +949,7 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc, return; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static bool atombios_crtc_prepare_pll(struct drm_crtc *crtc, struct drm_display_mode *mode) diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 009333645438..fca8b08535a5 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -112,7 +112,7 @@ static int radeon_process_aux_ch(struct radeon_i2c_chan *chan, if (ASIC_IS_DCE4(rdev)) args.v2.ucHPD_ID = chan->rec.hpd; - atom_execute_table_scratch_unlocked(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table_scratch_unlocked(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *ack = args.v1.ucReplyStatus; @@ -354,7 +354,7 @@ static u8 radeon_dp_encoder_service(struct radeon_device *rdev, args.ucLaneNum = lane_num; args.ucStatus = 0; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); return args.ucStatus; } diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index 6e537c5bd295..2bff0d9e20f5 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -119,12 +119,12 @@ atombios_set_backlight_level(struct radeon_encoder *radeon_encoder, u8 level) index = GetIndexIntoMasterTable(COMMAND, LCD1OutputControl); if (dig->backlight_level == 0) { args.ucAction = ATOM_LCD_BLOFF; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } else { args.ucAction = ATOM_LCD_BL_BRIGHTNESS_CONTROL; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); args.ucAction = ATOM_LCD_BLON; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } break; case ENCODER_OBJECT_ID_INTERNAL_UNIPHY: @@ -389,7 +389,7 @@ atombios_dac_setup(struct drm_encoder *encoder, int action) } args.usPixelClock = cpu_to_le16(radeon_encoder->pixel_clock / 10); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } @@ -445,7 +445,7 @@ atombios_tv_setup(struct drm_encoder *encoder, int action) args.sTVEncoder.usPixelClock = cpu_to_le16(radeon_encoder->pixel_clock / 10); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } @@ -546,7 +546,7 @@ atombios_dvo_setup(struct drm_encoder *encoder, int action) break; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } union lvds_encoder_control { @@ -664,7 +664,7 @@ atombios_digital_setup(struct drm_encoder *encoder, int action) break; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } int @@ -979,7 +979,7 @@ atombios_dig_encoder_setup2(struct drm_encoder *encoder, int action, int panel_m break; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } @@ -1361,7 +1361,7 @@ atombios_dig_transmitter_setup2(struct drm_encoder *encoder, int action, uint8_t break; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void @@ -1397,7 +1397,7 @@ atombios_set_edp_panel_power(struct drm_connector *connector, int action) args.v1.ucAction = action; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); /* wait for the panel to power up */ if (action == ATOM_TRANSMITTER_ACTION_POWER_ON) { @@ -1519,7 +1519,7 @@ atombios_external_encoder_setup(struct drm_encoder *encoder, DRM_ERROR("Unknown table version: %d, %d\n", frev, crev); return; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } static void @@ -1554,7 +1554,7 @@ atombios_yuv_setup(struct drm_encoder *encoder, bool enable) args.ucEnable = ATOM_ENABLE; args.ucCRTC = radeon_crtc->crtc_id; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); WREG32(reg, temp); } @@ -1618,10 +1618,10 @@ radeon_atom_encoder_dpms_avivo(struct drm_encoder *encoder, int mode) if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_DDI) { u32 reg = RREG32(RADEON_BIOS_3_SCRATCH); WREG32(RADEON_BIOS_3_SCRATCH, reg & ~ATOM_S3_DFP2I_ACTIVE); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); WREG32(RADEON_BIOS_3_SCRATCH, reg); } else - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { if (rdev->mode_info.bl_encoder) { struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; @@ -1629,7 +1629,7 @@ radeon_atom_encoder_dpms_avivo(struct drm_encoder *encoder, int mode) atombios_set_backlight_level(radeon_encoder, dig->backlight_level); } else { args.ucAction = ATOM_LCD_BLON; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } } break; @@ -1637,10 +1637,10 @@ radeon_atom_encoder_dpms_avivo(struct drm_encoder *encoder, int mode) case DRM_MODE_DPMS_SUSPEND: case DRM_MODE_DPMS_OFF: args.ucAction = ATOM_DISABLE; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { args.ucAction = ATOM_LCD_BLOFF; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } break; } @@ -1983,7 +1983,7 @@ atombios_set_encoder_crtc_source(struct drm_encoder *encoder) return; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); /* update scratch regs with new routing */ radeon_atombios_encoder_crtc_scratch_regs(encoder, radeon_crtc->crtc_id); @@ -2311,7 +2311,7 @@ atombios_dac_load_detect(struct drm_encoder *encoder, struct drm_connector *conn args.sDacload.ucMisc = DAC_LOAD_MISC_YPrPb; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); return true; } else diff --git a/drivers/gpu/drm/radeon/atombios_i2c.c b/drivers/gpu/drm/radeon/atombios_i2c.c index ab4d21072191..730f0b25312b 100644 --- a/drivers/gpu/drm/radeon/atombios_i2c.c +++ b/drivers/gpu/drm/radeon/atombios_i2c.c @@ -78,7 +78,7 @@ static int radeon_process_i2c_ch(struct radeon_i2c_chan *chan, args.ucSlaveAddr = slave_addr << 1; args.ucLineNumber = chan->rec.i2c_id; - atom_execute_table_scratch_unlocked(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table_scratch_unlocked(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); /* error */ if (args.ucStatus != HW_ASSISTED_I2C_STATUS_SUCCESS) { diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c index 4e64ed38c439..70931b04bbac 100644 --- a/drivers/gpu/drm/radeon/btc_dpm.c +++ b/drivers/gpu/drm/radeon/btc_dpm.c @@ -53,8 +53,7 @@ extern int ni_mc_load_microcode(struct radeon_device *rdev); //********* BARTS **************// -static const u32 barts_cgcg_cgls_default[] = -{ +static const u32 barts_cgcg_cgls_default[] = { /* Register, Value, Mask bits */ 0x000008f8, 0x00000010, 0xffffffff, 0x000008fc, 0x00000000, 0xffffffff, @@ -107,8 +106,7 @@ static const u32 barts_cgcg_cgls_default[] = }; #define BARTS_CGCG_CGLS_DEFAULT_LENGTH sizeof(barts_cgcg_cgls_default) / (3 * sizeof(u32)) -static const u32 barts_cgcg_cgls_disable[] = -{ +static const u32 barts_cgcg_cgls_disable[] = { 0x000008f8, 0x00000010, 0xffffffff, 0x000008fc, 0xffffffff, 0xffffffff, 0x000008f8, 0x00000011, 0xffffffff, @@ -162,8 +160,7 @@ static const u32 barts_cgcg_cgls_disable[] = }; #define BARTS_CGCG_CGLS_DISABLE_LENGTH sizeof(barts_cgcg_cgls_disable) / (3 * sizeof(u32)) -static const u32 barts_cgcg_cgls_enable[] = -{ +static const u32 barts_cgcg_cgls_enable[] = { /* 0x0000c124, 0x84180000, 0x00180000, */ 0x00000644, 0x000f7892, 0x001f4080, 0x000008f8, 0x00000010, 0xffffffff, @@ -217,8 +214,7 @@ static const u32 barts_cgcg_cgls_enable[] = }; #define BARTS_CGCG_CGLS_ENABLE_LENGTH sizeof(barts_cgcg_cgls_enable) / (3 * sizeof(u32)) -static const u32 barts_mgcg_default[] = -{ +static const u32 barts_mgcg_default[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x00005448, 0x00000100, 0xffffffff, 0x000055e4, 0x00600100, 0xffffffff, @@ -366,8 +362,7 @@ static const u32 barts_mgcg_default[] = }; #define BARTS_MGCG_DEFAULT_LENGTH sizeof(barts_mgcg_default) / (3 * sizeof(u32)) -static const u32 barts_mgcg_disable[] = -{ +static const u32 barts_mgcg_disable[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x000008f8, 0x00000000, 0xffffffff, 0x000008fc, 0xffffffff, 0xffffffff, @@ -381,8 +376,7 @@ static const u32 barts_mgcg_disable[] = }; #define BARTS_MGCG_DISABLE_LENGTH sizeof(barts_mgcg_disable) / (3 * sizeof(u32)) -static const u32 barts_mgcg_enable[] = -{ +static const u32 barts_mgcg_enable[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x000008f8, 0x00000000, 0xffffffff, 0x000008fc, 0x00000000, 0xffffffff, @@ -397,8 +391,7 @@ static const u32 barts_mgcg_enable[] = #define BARTS_MGCG_ENABLE_LENGTH sizeof(barts_mgcg_enable) / (3 * sizeof(u32)) //********* CAICOS **************// -static const u32 caicos_cgcg_cgls_default[] = -{ +static const u32 caicos_cgcg_cgls_default[] = { 0x000008f8, 0x00000010, 0xffffffff, 0x000008fc, 0x00000000, 0xffffffff, 0x000008f8, 0x00000011, 0xffffffff, @@ -450,8 +443,7 @@ static const u32 caicos_cgcg_cgls_default[] = }; #define CAICOS_CGCG_CGLS_DEFAULT_LENGTH sizeof(caicos_cgcg_cgls_default) / (3 * sizeof(u32)) -static const u32 caicos_cgcg_cgls_disable[] = -{ +static const u32 caicos_cgcg_cgls_disable[] = { 0x000008f8, 0x00000010, 0xffffffff, 0x000008fc, 0xffffffff, 0xffffffff, 0x000008f8, 0x00000011, 0xffffffff, @@ -505,8 +497,7 @@ static const u32 caicos_cgcg_cgls_disable[] = }; #define CAICOS_CGCG_CGLS_DISABLE_LENGTH sizeof(caicos_cgcg_cgls_disable) / (3 * sizeof(u32)) -static const u32 caicos_cgcg_cgls_enable[] = -{ +static const u32 caicos_cgcg_cgls_enable[] = { /* 0x0000c124, 0x84180000, 0x00180000, */ 0x00000644, 0x000f7892, 0x001f4080, 0x000008f8, 0x00000010, 0xffffffff, @@ -560,8 +551,7 @@ static const u32 caicos_cgcg_cgls_enable[] = }; #define CAICOS_CGCG_CGLS_ENABLE_LENGTH sizeof(caicos_cgcg_cgls_enable) / (3 * sizeof(u32)) -static const u32 caicos_mgcg_default[] = -{ +static const u32 caicos_mgcg_default[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x00005448, 0x00000100, 0xffffffff, 0x000055e4, 0x00600100, 0xffffffff, @@ -640,8 +630,7 @@ static const u32 caicos_mgcg_default[] = }; #define CAICOS_MGCG_DEFAULT_LENGTH sizeof(caicos_mgcg_default) / (3 * sizeof(u32)) -static const u32 caicos_mgcg_disable[] = -{ +static const u32 caicos_mgcg_disable[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x000008f8, 0x00000000, 0xffffffff, 0x000008fc, 0xffffffff, 0xffffffff, @@ -655,8 +644,7 @@ static const u32 caicos_mgcg_disable[] = }; #define CAICOS_MGCG_DISABLE_LENGTH sizeof(caicos_mgcg_disable) / (3 * sizeof(u32)) -static const u32 caicos_mgcg_enable[] = -{ +static const u32 caicos_mgcg_enable[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x000008f8, 0x00000000, 0xffffffff, 0x000008fc, 0x00000000, 0xffffffff, @@ -671,8 +659,7 @@ static const u32 caicos_mgcg_enable[] = #define CAICOS_MGCG_ENABLE_LENGTH sizeof(caicos_mgcg_enable) / (3 * sizeof(u32)) //********* TURKS **************// -static const u32 turks_cgcg_cgls_default[] = -{ +static const u32 turks_cgcg_cgls_default[] = { 0x000008f8, 0x00000010, 0xffffffff, 0x000008fc, 0x00000000, 0xffffffff, 0x000008f8, 0x00000011, 0xffffffff, @@ -724,8 +711,7 @@ static const u32 turks_cgcg_cgls_default[] = }; #define TURKS_CGCG_CGLS_DEFAULT_LENGTH sizeof(turks_cgcg_cgls_default) / (3 * sizeof(u32)) -static const u32 turks_cgcg_cgls_disable[] = -{ +static const u32 turks_cgcg_cgls_disable[] = { 0x000008f8, 0x00000010, 0xffffffff, 0x000008fc, 0xffffffff, 0xffffffff, 0x000008f8, 0x00000011, 0xffffffff, @@ -779,8 +765,7 @@ static const u32 turks_cgcg_cgls_disable[] = }; #define TURKS_CGCG_CGLS_DISABLE_LENGTH sizeof(turks_cgcg_cgls_disable) / (3 * sizeof(u32)) -static const u32 turks_cgcg_cgls_enable[] = -{ +static const u32 turks_cgcg_cgls_enable[] = { /* 0x0000c124, 0x84180000, 0x00180000, */ 0x00000644, 0x000f7892, 0x001f4080, 0x000008f8, 0x00000010, 0xffffffff, @@ -835,8 +820,7 @@ static const u32 turks_cgcg_cgls_enable[] = #define TURKS_CGCG_CGLS_ENABLE_LENGTH sizeof(turks_cgcg_cgls_enable) / (3 * sizeof(u32)) // These are the sequences for turks_mgcg_shls -static const u32 turks_mgcg_default[] = -{ +static const u32 turks_mgcg_default[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x00005448, 0x00000100, 0xffffffff, 0x000055e4, 0x00600100, 0xffffffff, @@ -935,8 +919,7 @@ static const u32 turks_mgcg_default[] = }; #define TURKS_MGCG_DEFAULT_LENGTH sizeof(turks_mgcg_default) / (3 * sizeof(u32)) -static const u32 turks_mgcg_disable[] = -{ +static const u32 turks_mgcg_disable[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x000008f8, 0x00000000, 0xffffffff, 0x000008fc, 0xffffffff, 0xffffffff, @@ -950,8 +933,7 @@ static const u32 turks_mgcg_disable[] = }; #define TURKS_MGCG_DISABLE_LENGTH sizeof(turks_mgcg_disable) / (3 * sizeof(u32)) -static const u32 turks_mgcg_enable[] = -{ +static const u32 turks_mgcg_enable[] = { 0x0000802c, 0xc0000000, 0xffffffff, 0x000008f8, 0x00000000, 0xffffffff, 0x000008fc, 0x00000000, 0xffffffff, @@ -972,8 +954,7 @@ static const u32 turks_mgcg_enable[] = //********* BARTS **************// -static const u32 barts_sysls_default[] = -{ +static const u32 barts_sysls_default[] = { /* Register, Value, Mask bits */ 0x000055e8, 0x00000000, 0xffffffff, 0x0000d0bc, 0x00000000, 0xffffffff, @@ -993,8 +974,7 @@ static const u32 barts_sysls_default[] = }; #define BARTS_SYSLS_DEFAULT_LENGTH sizeof(barts_sysls_default) / (3 * sizeof(u32)) -static const u32 barts_sysls_disable[] = -{ +static const u32 barts_sysls_disable[] = { 0x000055e8, 0x00000000, 0xffffffff, 0x0000d0bc, 0x00000000, 0xffffffff, 0x000015c0, 0x00041401, 0xffffffff, @@ -1013,8 +993,7 @@ static const u32 barts_sysls_disable[] = }; #define BARTS_SYSLS_DISABLE_LENGTH sizeof(barts_sysls_disable) / (3 * sizeof(u32)) -static const u32 barts_sysls_enable[] = -{ +static const u32 barts_sysls_enable[] = { 0x000055e8, 0x00000001, 0xffffffff, 0x0000d0bc, 0x00000100, 0xffffffff, 0x000015c0, 0x000c1401, 0xffffffff, @@ -1034,8 +1013,7 @@ static const u32 barts_sysls_enable[] = #define BARTS_SYSLS_ENABLE_LENGTH sizeof(barts_sysls_enable) / (3 * sizeof(u32)) //********* CAICOS **************// -static const u32 caicos_sysls_default[] = -{ +static const u32 caicos_sysls_default[] = { 0x000055e8, 0x00000000, 0xffffffff, 0x0000d0bc, 0x00000000, 0xffffffff, 0x000015c0, 0x000c1401, 0xffffffff, @@ -1053,8 +1031,7 @@ static const u32 caicos_sysls_default[] = }; #define CAICOS_SYSLS_DEFAULT_LENGTH sizeof(caicos_sysls_default) / (3 * sizeof(u32)) -static const u32 caicos_sysls_disable[] = -{ +static const u32 caicos_sysls_disable[] = { 0x000055e8, 0x00000000, 0xffffffff, 0x0000d0bc, 0x00000000, 0xffffffff, 0x000015c0, 0x00041401, 0xffffffff, @@ -1072,8 +1049,7 @@ static const u32 caicos_sysls_disable[] = }; #define CAICOS_SYSLS_DISABLE_LENGTH sizeof(caicos_sysls_disable) / (3 * sizeof(u32)) -static const u32 caicos_sysls_enable[] = -{ +static const u32 caicos_sysls_enable[] = { 0x000055e8, 0x00000001, 0xffffffff, 0x0000d0bc, 0x00000100, 0xffffffff, 0x000015c0, 0x000c1401, 0xffffffff, @@ -1092,8 +1068,7 @@ static const u32 caicos_sysls_enable[] = #define CAICOS_SYSLS_ENABLE_LENGTH sizeof(caicos_sysls_enable) / (3 * sizeof(u32)) //********* TURKS **************// -static const u32 turks_sysls_default[] = -{ +static const u32 turks_sysls_default[] = { 0x000055e8, 0x00000000, 0xffffffff, 0x0000d0bc, 0x00000000, 0xffffffff, 0x000015c0, 0x000c1401, 0xffffffff, @@ -1112,8 +1087,7 @@ static const u32 turks_sysls_default[] = }; #define TURKS_SYSLS_DEFAULT_LENGTH sizeof(turks_sysls_default) / (3 * sizeof(u32)) -static const u32 turks_sysls_disable[] = -{ +static const u32 turks_sysls_disable[] = { 0x000055e8, 0x00000000, 0xffffffff, 0x0000d0bc, 0x00000000, 0xffffffff, 0x000015c0, 0x00041401, 0xffffffff, @@ -1132,8 +1106,7 @@ static const u32 turks_sysls_disable[] = }; #define TURKS_SYSLS_DISABLE_LENGTH sizeof(turks_sysls_disable) / (3 * sizeof(u32)) -static const u32 turks_sysls_enable[] = -{ +static const u32 turks_sysls_enable[] = { 0x000055e8, 0x00000001, 0xffffffff, 0x0000d0bc, 0x00000100, 0xffffffff, 0x000015c0, 0x000c1401, 0xffffffff, @@ -1154,8 +1127,7 @@ static const u32 turks_sysls_enable[] = #endif -u32 btc_valid_sclk[40] = -{ +u32 btc_valid_sclk[40] = { 5000, 10000, 15000, 20000, 25000, 30000, 35000, 40000, 45000, 50000, 55000, 60000, 65000, 70000, 75000, 80000, 85000, 90000, 95000, 100000, 105000, 110000, 11500, 120000, 125000, 130000, 135000, 140000, 145000, 150000, @@ -1194,7 +1166,7 @@ void btc_apply_voltage_dependency_rules(struct radeon_clock_voltage_dependency_t if ((table == NULL) || (table->count == 0)) return; - for (i= 0; i < table->count; i++) { + for (i = 0; i < table->count; i++) { if (clock <= table->entries[i].clk) { if (*voltage < table->entries[i].v) *voltage = (u16)((table->entries[i].v < max_voltage) ? @@ -1441,7 +1413,7 @@ void btc_program_mgcg_hw_sequence(struct radeon_device *rdev, u32 i, length = count * 3; u32 tmp; - for (i = 0; i < length; i+=3) { + for (i = 0; i < length; i += 3) { tmp = RREG32(sequence[i]); tmp &= ~sequence[i+2]; tmp |= sequence[i+1] & sequence[i+2]; @@ -2003,7 +1975,7 @@ static int btc_copy_vbios_mc_reg_table(struct atom_mc_reg_table *table, for (i = 0; i < table->num_entries; i++) { eg_table->mc_reg_table_entry[i].mclk_max = table->mc_reg_table_entry[i].mclk_max; - for(j = 0; j < table->last; j++) + for (j = 0; j < table->last; j++) eg_table->mc_reg_table_entry[i].mc_data[j] = table->mc_reg_table_entry[i].mc_data[j]; } diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c index b8f4dac68d85..abe9d65cc460 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.c +++ b/drivers/gpu/drm/radeon/ci_dpm.c @@ -46,36 +46,31 @@ #define VOLTAGE_VID_OFFSET_SCALE1 625 #define VOLTAGE_VID_OFFSET_SCALE2 100 -static const struct ci_pt_defaults defaults_hawaii_xt = -{ +static const struct ci_pt_defaults defaults_hawaii_xt = { 1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0xB0000, { 0x2E, 0x00, 0x00, 0x88, 0x00, 0x00, 0x72, 0x60, 0x51, 0xA7, 0x79, 0x6B, 0x90, 0xBD, 0x79 }, { 0x217, 0x217, 0x217, 0x242, 0x242, 0x242, 0x269, 0x269, 0x269, 0x2A1, 0x2A1, 0x2A1, 0x2C9, 0x2C9, 0x2C9 } }; -static const struct ci_pt_defaults defaults_hawaii_pro = -{ +static const struct ci_pt_defaults defaults_hawaii_pro = { 1, 0xF, 0xFD, 0x19, 5, 0x14, 0, 0x65062, { 0x2E, 0x00, 0x00, 0x88, 0x00, 0x00, 0x72, 0x60, 0x51, 0xA7, 0x79, 0x6B, 0x90, 0xBD, 0x79 }, { 0x217, 0x217, 0x217, 0x242, 0x242, 0x242, 0x269, 0x269, 0x269, 0x2A1, 0x2A1, 0x2A1, 0x2C9, 0x2C9, 0x2C9 } }; -static const struct ci_pt_defaults defaults_bonaire_xt = -{ +static const struct ci_pt_defaults defaults_bonaire_xt = { 1, 0xF, 0xFD, 0x19, 5, 45, 0, 0xB0000, { 0x79, 0x253, 0x25D, 0xAE, 0x72, 0x80, 0x83, 0x86, 0x6F, 0xC8, 0xC9, 0xC9, 0x2F, 0x4D, 0x61 }, { 0x17C, 0x172, 0x180, 0x1BC, 0x1B3, 0x1BD, 0x206, 0x200, 0x203, 0x25D, 0x25A, 0x255, 0x2C3, 0x2C5, 0x2B4 } }; -static const struct ci_pt_defaults defaults_saturn_xt = -{ +static const struct ci_pt_defaults defaults_saturn_xt = { 1, 0xF, 0xFD, 0x19, 5, 55, 0, 0x70000, { 0x8C, 0x247, 0x249, 0xA6, 0x80, 0x81, 0x8B, 0x89, 0x86, 0xC9, 0xCA, 0xC9, 0x4D, 0x4D, 0x4D }, { 0x187, 0x187, 0x187, 0x1C7, 0x1C7, 0x1C7, 0x210, 0x210, 0x210, 0x266, 0x266, 0x266, 0x2C9, 0x2C9, 0x2C9 } }; -static const struct ci_pt_config_reg didt_config_ci[] = -{ +static const struct ci_pt_config_reg didt_config_ci[] = { { 0x10, 0x000000ff, 0, 0x0, CISLANDS_CONFIGREG_DIDT_IND }, { 0x10, 0x0000ff00, 8, 0x0, CISLANDS_CONFIGREG_DIDT_IND }, { 0x10, 0x00ff0000, 16, 0x0, CISLANDS_CONFIGREG_DIDT_IND }, @@ -1216,7 +1211,7 @@ static void ci_thermal_initialize(struct radeon_device *rdev) if (rdev->pm.fan_pulses_per_revolution) { tmp = RREG32_SMC(CG_TACH_CTRL) & ~EDGE_PER_REV_MASK; - tmp |= EDGE_PER_REV(rdev->pm.fan_pulses_per_revolution -1); + tmp |= EDGE_PER_REV(rdev->pm.fan_pulses_per_revolution - 1); WREG32_SMC(CG_TACH_CTRL, tmp); } @@ -3333,7 +3328,7 @@ static int ci_populate_all_memory_levels(struct radeon_device *rdev) } static void ci_reset_single_dpm_table(struct radeon_device *rdev, - struct ci_single_dpm_table* dpm_table, + struct ci_single_dpm_table *dpm_table, u32 count) { u32 i; @@ -3343,7 +3338,7 @@ static void ci_reset_single_dpm_table(struct radeon_device *rdev, dpm_table->dpm_levels[i].enabled = false; } -static void ci_setup_pcie_table_entry(struct ci_single_dpm_table* dpm_table, +static void ci_setup_pcie_table_entry(struct ci_single_dpm_table *dpm_table, u32 index, u32 pcie_gen, u32 pcie_lanes) { dpm_table->dpm_levels[index].value = pcie_gen; @@ -3503,7 +3498,7 @@ static int ci_find_boot_level(struct ci_single_dpm_table *table, u32 i; int ret = -EINVAL; - for(i = 0; i < table->count; i++) { + for (i = 0; i < table->count; i++) { if (value == table->dpm_levels[i].value) { *boot_level = i; ret = 0; @@ -4304,7 +4299,7 @@ static int ci_set_mc_special_registers(struct radeon_device *rdev, for (i = 0, j = table->last; i < table->last; i++) { if (j >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE) return -EINVAL; - switch(table->mc_reg_address[i].s1 << 2) { + switch (table->mc_reg_address[i].s1 << 2) { case MC_SEQ_MISC1: temp_reg = RREG32(MC_PMG_CMD_EMRS); table->mc_reg_address[j].s1 = MC_PMG_CMD_EMRS >> 2; @@ -4369,7 +4364,7 @@ static bool ci_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg) { bool result = true; - switch(in_reg) { + switch (in_reg) { case MC_SEQ_RAS_TIMING >> 2: *out_reg = MC_SEQ_RAS_TIMING_LP >> 2; break; @@ -4508,7 +4503,7 @@ static int ci_register_patching_mc_seq(struct radeon_device *rdev, for (i = 0; i < table->last; i++) { if (table->last >= SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE) return -EINVAL; - switch(table->mc_reg_address[i].s1 >> 2) { + switch (table->mc_reg_address[i].s1 >> 2) { case MC_SEQ_MISC1: for (k = 0; k < table->num_entries; k++) { if ((table->mc_reg_table_entry[k].mclk_max == 125000) || @@ -4683,7 +4678,7 @@ static void ci_convert_mc_reg_table_entry_to_smc(struct radeon_device *rdev, struct ci_power_info *pi = ci_get_pi(rdev); u32 i = 0; - for(i = 0; i < pi->mc_reg_table.num_entries; i++) { + for (i = 0; i < pi->mc_reg_table.num_entries; i++) { if (memory_clock <= pi->mc_reg_table.mc_reg_table_entry[i].mclk_max) break; } diff --git a/drivers/gpu/drm/radeon/ci_dpm.h b/drivers/gpu/drm/radeon/ci_dpm.h index ac12db5f2cf7..74b95c200222 100644 --- a/drivers/gpu/drm/radeon/ci_dpm.h +++ b/drivers/gpu/drm/radeon/ci_dpm.h @@ -87,8 +87,7 @@ struct ci_mc_reg_table { SMU7_Discrete_MCRegisterAddress mc_reg_address[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE]; }; -struct ci_ulv_parm -{ +struct ci_ulv_parm { bool supported; u32 cg_ulv_parameter; u32 volt_change_delay; @@ -113,8 +112,7 @@ struct ci_dpm_level_enable_mask { u32 pcie_dpm_enable_mask; }; -struct ci_vbios_boot_state -{ +struct ci_vbios_boot_state { u16 mvdd_bootup_value; u16 vddc_bootup_value; u16 vddci_bootup_value; diff --git a/drivers/gpu/drm/radeon/clearstate_cayman.h b/drivers/gpu/drm/radeon/clearstate_cayman.h index 4774e04c4da6..7693fb6624a3 100644 --- a/drivers/gpu/drm/radeon/clearstate_cayman.h +++ b/drivers/gpu/drm/radeon/clearstate_cayman.h @@ -23,8 +23,7 @@ #include "clearstate_defs.h" -static const u32 SECT_CONTEXT_def_1[] = -{ +static const u32 SECT_CONTEXT_def_1[] = { 0x00000000, // DB_RENDER_CONTROL 0x00000000, // DB_COUNT_CONTROL 0x00000000, // DB_DEPTH_VIEW @@ -514,8 +513,7 @@ static const u32 SECT_CONTEXT_def_1[] = 0x00000000, // CB_BLEND6_CONTROL 0x00000000, // CB_BLEND7_CONTROL }; -static const u32 SECT_CONTEXT_def_2[] = -{ +static const u32 SECT_CONTEXT_def_2[] = { 0x00000000, // PA_CL_POINT_X_RAD 0x00000000, // PA_CL_POINT_Y_RAD 0x00000000, // PA_CL_POINT_SIZE @@ -523,8 +521,7 @@ static const u32 SECT_CONTEXT_def_2[] = 0x00000000, // VGT_DMA_BASE_HI 0x00000000, // VGT_DMA_BASE }; -static const u32 SECT_CONTEXT_def_3[] = -{ +static const u32 SECT_CONTEXT_def_3[] = { 0x00000000, // DB_DEPTH_CONTROL 0x00000000, // DB_EQAA 0x00000000, // CB_COLOR_CONTROL diff --git a/drivers/gpu/drm/radeon/clearstate_ci.h b/drivers/gpu/drm/radeon/clearstate_ci.h index c1b6c22dbed7..0045d42aa27c 100644 --- a/drivers/gpu/drm/radeon/clearstate_ci.h +++ b/drivers/gpu/drm/radeon/clearstate_ci.h @@ -23,8 +23,7 @@ #include "clearstate_defs.h" -static const unsigned int ci_SECT_CONTEXT_def_1[] = -{ +static const unsigned int ci_SECT_CONTEXT_def_1[] = { 0x00000000, // DB_RENDER_CONTROL 0x00000000, // DB_COUNT_CONTROL 0x00000000, // DB_DEPTH_VIEW diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index a424b86008b8..c634dc28e6c3 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2514,8 +2514,7 @@ static void evergreen_agp_enable(struct radeon_device *rdev) WREG32(VM_CONTEXT1_CNTL, 0); } -static const unsigned ni_dig_offsets[] = -{ +static const unsigned ni_dig_offsets[] = { NI_DIG0_REGISTER_OFFSET, NI_DIG1_REGISTER_OFFSET, NI_DIG2_REGISTER_OFFSET, @@ -2524,8 +2523,7 @@ static const unsigned ni_dig_offsets[] = NI_DIG5_REGISTER_OFFSET }; -static const unsigned ni_tx_offsets[] = -{ +static const unsigned ni_tx_offsets[] = { NI_DCIO_UNIPHY0_UNIPHY_TX_CONTROL1, NI_DCIO_UNIPHY1_UNIPHY_TX_CONTROL1, NI_DCIO_UNIPHY2_UNIPHY_TX_CONTROL1, @@ -2534,8 +2532,7 @@ static const unsigned ni_tx_offsets[] = NI_DCIO_UNIPHY5_UNIPHY_TX_CONTROL1 }; -static const unsigned evergreen_dp_offsets[] = -{ +static const unsigned evergreen_dp_offsets[] = { EVERGREEN_DP0_REGISTER_OFFSET, EVERGREEN_DP1_REGISTER_OFFSET, EVERGREEN_DP2_REGISTER_OFFSET, @@ -2544,8 +2541,7 @@ static const unsigned evergreen_dp_offsets[] = EVERGREEN_DP5_REGISTER_OFFSET }; -static const unsigned evergreen_disp_int_status[] = -{ +static const unsigned evergreen_disp_int_status[] = { DISP_INTERRUPT_STATUS, DISP_INTERRUPT_STATUS_CONTINUE, DISP_INTERRUPT_STATUS_CONTINUE2, @@ -2643,7 +2639,7 @@ static void evergreen_blank_dp_output(struct radeon_device *rdev, return; } - stream_ctrl &=~EVERGREEN_DP_VID_STREAM_CNTL_ENABLE; + stream_ctrl &= ~EVERGREEN_DP_VID_STREAM_CNTL_ENABLE; WREG32(EVERGREEN_DP_VID_STREAM_CNTL + evergreen_dp_offsets[dig_fe], stream_ctrl); @@ -2655,7 +2651,7 @@ static void evergreen_blank_dp_output(struct radeon_device *rdev, stream_ctrl = RREG32(EVERGREEN_DP_VID_STREAM_CNTL + evergreen_dp_offsets[dig_fe]); } - if (counter >= 32 ) + if (counter >= 32) DRM_ERROR("counter exceeds %d\n", counter); fifo_ctrl = RREG32(EVERGREEN_DP_STEER_FIFO + evergreen_dp_offsets[dig_fe]); @@ -2716,7 +2712,7 @@ void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *sav /*for now we do it this manually*/ /**/ if (ASIC_IS_DCE5(rdev) && - evergreen_is_dp_sst_stream_enabled(rdev, i ,&dig_fe)) + evergreen_is_dp_sst_stream_enabled(rdev, i, &dig_fe)) evergreen_blank_dp_output(rdev, dig_fe); /*we could remove 6 lines below*/ /* XXX this is a hack to avoid strange behavior with EFI on certain systems */ @@ -3597,7 +3593,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev) sq_lds_resource_mgmt = RREG32(SQ_LDS_RESOURCE_MGMT); - sq_gpr_resource_mgmt_1 = NUM_PS_GPRS((rdev->config.evergreen.max_gprs - (4 * 2))* 12 / 32); + sq_gpr_resource_mgmt_1 = NUM_PS_GPRS((rdev->config.evergreen.max_gprs - (4 * 2)) * 12 / 32); sq_gpr_resource_mgmt_1 |= NUM_VS_GPRS((rdev->config.evergreen.max_gprs - (4 * 2)) * 6 / 32); sq_gpr_resource_mgmt_1 |= NUM_CLAUSE_TEMP_GPRS(4); sq_gpr_resource_mgmt_2 = NUM_GS_GPRS((rdev->config.evergreen.max_gprs - (4 * 2)) * 4 / 32); diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index 0de79f3a7e3f..1fe6e0d883c7 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -33,8 +33,8 @@ #include "evergreen_reg_safe.h" #include "cayman_reg_safe.h" -#define MAX(a,b) (((a)>(b))?(a):(b)) -#define MIN(a,b) (((a)<(b))?(a):(b)) +#define MAX(a, b) (((a) > (b)) ? (a) : (b)) +#define MIN(a, b) (((a) < (b)) ? (a) : (b)) #define REG_SAFE_BM_SIZE ARRAY_SIZE(evergreen_reg_safe_bm) diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h index b436badf9efa..3ff9fda54aa3 100644 --- a/drivers/gpu/drm/radeon/evergreen_reg.h +++ b/drivers/gpu/drm/radeon/evergreen_reg.h @@ -265,8 +265,8 @@ #define NI_DIG_BE_CNTL 0x7140 -# define NI_DIG_BE_CNTL_FE_SOURCE_SELECT(x) (((x) >> 8 ) & 0x3F) -# define NI_DIG_FE_CNTL_MODE(x) (((x) >> 16) & 0x7 ) +# define NI_DIG_BE_CNTL_FE_SOURCE_SELECT(x) (((x) >> 8) & 0x3F) +# define NI_DIG_FE_CNTL_MODE(x) (((x) >> 16) & 0x7) #define NI_DIG_BE_EN_CNTL 0x7144 # define NI_DIG_BE_EN_CNTL_ENABLE (1 << 0) @@ -284,7 +284,7 @@ #define EVERGREEN_DP_VID_STREAM_CNTL 0x730C # define EVERGREEN_DP_VID_STREAM_CNTL_ENABLE (1 << 0) -# define EVERGREEN_DP_VID_STREAM_STATUS (1 <<16) +# define EVERGREEN_DP_VID_STREAM_STATUS (1 << 16) #define EVERGREEN_DP_STEER_FIFO 0x7310 # define EVERGREEN_DP_STEER_FIFO_RESET (1 << 0) #define EVERGREEN_DP_SEC_CNTL 0x7280 @@ -302,8 +302,8 @@ # define EVERGREEN_DP_SEC_SS_EN (1 << 28) /*DCIO_UNIPHY block*/ -#define NI_DCIO_UNIPHY0_UNIPHY_TX_CONTROL1 (0x6600 -0x6600) -#define NI_DCIO_UNIPHY1_UNIPHY_TX_CONTROL1 (0x6640 -0x6600) +#define NI_DCIO_UNIPHY0_UNIPHY_TX_CONTROL1 (0x6600 - 0x6600) +#define NI_DCIO_UNIPHY1_UNIPHY_TX_CONTROL1 (0x6640 - 0x6600) #define NI_DCIO_UNIPHY2_UNIPHY_TX_CONTROL1 (0x6680 - 0x6600) #define NI_DCIO_UNIPHY3_UNIPHY_TX_CONTROL1 (0x66C0 - 0x6600) #define NI_DCIO_UNIPHY4_UNIPHY_TX_CONTROL1 (0x6700 - 0x6600) diff --git a/drivers/gpu/drm/radeon/evergreen_smc.h b/drivers/gpu/drm/radeon/evergreen_smc.h index 3a03ba37d043..b34d54b567b7 100644 --- a/drivers/gpu/drm/radeon/evergreen_smc.h +++ b/drivers/gpu/drm/radeon/evergreen_smc.h @@ -29,8 +29,7 @@ #define SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE 16 -struct SMC_Evergreen_MCRegisterAddress -{ +struct SMC_Evergreen_MCRegisterAddress { uint16_t s0; uint16_t s1; }; @@ -38,15 +37,13 @@ struct SMC_Evergreen_MCRegisterAddress typedef struct SMC_Evergreen_MCRegisterAddress SMC_Evergreen_MCRegisterAddress; -struct SMC_Evergreen_MCRegisterSet -{ +struct SMC_Evergreen_MCRegisterSet { uint32_t value[SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE]; }; typedef struct SMC_Evergreen_MCRegisterSet SMC_Evergreen_MCRegisterSet; -struct SMC_Evergreen_MCRegisters -{ +struct SMC_Evergreen_MCRegisters { uint8_t last; uint8_t reserved[3]; SMC_Evergreen_MCRegisterAddress address[SMC_EVERGREEN_MC_REGISTER_ARRAY_SIZE]; diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c index f7735da07feb..55dbf450bd9c 100644 --- a/drivers/gpu/drm/radeon/kv_dpm.c +++ b/drivers/gpu/drm/radeon/kv_dpm.c @@ -64,8 +64,7 @@ extern void cik_exit_rlc_safe_mode(struct radeon_device *rdev); extern void cik_update_cg(struct radeon_device *rdev, u32 block, bool enable); -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 }, @@ -931,9 +930,9 @@ static void kv_calculate_dfs_bypass_settings(struct radeon_device *rdev) 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; @@ -1577,7 +1576,7 @@ static void kv_set_valid_clock_range(struct radeon_device *rdev, 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; diff --git a/drivers/gpu/drm/radeon/kv_smc.c b/drivers/gpu/drm/radeon/kv_smc.c index c0a59527e7b8..65831cca6730 100644 --- a/drivers/gpu/drm/radeon/kv_smc.c +++ b/drivers/gpu/drm/radeon/kv_smc.c @@ -189,7 +189,7 @@ int kv_copy_bytes_to_smc(struct radeon_device *rdev, if (ret) return ret; - original_data= RREG32(SMC_IND_DATA_0); + original_data = RREG32(SMC_IND_DATA_0); extra_shift = 8 * (4 - byte_count); diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index 927e5f42e97d..9f0881ab3105 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -66,8 +66,7 @@ void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v) spin_unlock_irqrestore(&rdev->smc_idx_lock, flags); } -static const u32 tn_rlc_save_restore_register_list[] = -{ +static const u32 tn_rlc_save_restore_register_list[] = { 0x98fc, 0x98f0, 0x9834, @@ -216,8 +215,7 @@ MODULE_FIRMWARE("radeon/ARUBA_me.bin"); MODULE_FIRMWARE("radeon/ARUBA_rlc.bin"); -static const u32 cayman_golden_registers2[] = -{ +static const u32 cayman_golden_registers2[] = { 0x3e5c, 0xffffffff, 0x00000000, 0x3e48, 0xffffffff, 0x00000000, 0x3e4c, 0xffffffff, 0x00000000, @@ -226,8 +224,7 @@ static const u32 cayman_golden_registers2[] = 0x3e60, 0xffffffff, 0x00000000 }; -static const u32 cayman_golden_registers[] = -{ +static const u32 cayman_golden_registers[] = { 0x5eb4, 0xffffffff, 0x00000002, 0x5e78, 0x8f311ff1, 0x001000f0, 0x3f90, 0xffff0000, 0xff000000, @@ -267,16 +264,14 @@ static const u32 cayman_golden_registers[] = 0x8974, 0xffffffff, 0x00000000 }; -static const u32 dvst_golden_registers2[] = -{ +static const u32 dvst_golden_registers2[] = { 0x8f8, 0xffffffff, 0, 0x8fc, 0x00380000, 0, 0x8f8, 0xffffffff, 1, 0x8fc, 0x0e000000, 0 }; -static const u32 dvst_golden_registers[] = -{ +static const u32 dvst_golden_registers[] = { 0x690, 0x3fff3fff, 0x20c00033, 0x918c, 0x0fff0fff, 0x00010006, 0x91a8, 0x0fff0fff, 0x00010006, @@ -333,8 +328,7 @@ static const u32 dvst_golden_registers[] = 0x8974, 0xffffffff, 0x00000000 }; -static const u32 scrapper_golden_registers[] = -{ +static const u32 scrapper_golden_registers[] = { 0x690, 0x3fff3fff, 0x20c00033, 0x918c, 0x0fff0fff, 0x00010006, 0x918c, 0x0fff0fff, 0x00010006, @@ -624,7 +618,7 @@ static const u32 cayman_io_mc_regs[BTC_IO_MC_REGS_SIZE][2] = { int ni_mc_load_microcode(struct radeon_device *rdev) { const __be32 *fw_data; - u32 mem_type, running, blackout = 0; + u32 mem_type, running; u32 *io_mc_regs; int i, ucode_size, regs_size; @@ -659,11 +653,6 @@ int ni_mc_load_microcode(struct radeon_device *rdev) running = RREG32(MC_SEQ_SUP_CNTL) & RUN_MASK; if ((mem_type == MC_SEQ_MISC0_GDDR5_VALUE) && (running == 0)) { - if (running) { - blackout = RREG32(MC_SHARED_BLACKOUT_CNTL); - WREG32(MC_SHARED_BLACKOUT_CNTL, 1); - } - /* reset the engine and set to writable */ WREG32(MC_SEQ_SUP_CNTL, 0x00000008); WREG32(MC_SEQ_SUP_CNTL, 0x00000010); @@ -689,9 +678,6 @@ int ni_mc_load_microcode(struct radeon_device *rdev) break; udelay(1); } - - if (running) - WREG32(MC_SHARED_BLACKOUT_CNTL, blackout); } return 0; @@ -754,7 +740,8 @@ int ni_init_microcode(struct radeon_device *rdev) rlc_req_size = ARUBA_RLC_UCODE_SIZE * 4; mc_req_size = 0; break; - default: BUG(); + default: + BUG(); } DRM_INFO("Loading %s Microcode\n", chip_name); diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c index 3e1c1a392fb7..e08559c44a5c 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.c +++ b/drivers/gpu/drm/radeon/ni_dpm.c @@ -3103,9 +3103,6 @@ static int ni_init_simplified_leakage_table(struct radeon_device *rdev, u32 smc_leakage, max_leakage = 0; u32 scaling_factor; - if (!leakage_table) - return -EINVAL; - table_size = leakage_table->count; if (eg_pi->vddc_voltage_table.count != table_size) diff --git a/drivers/gpu/drm/radeon/ni_dpm.h b/drivers/gpu/drm/radeon/ni_dpm.h index 74e301936906..4e3e7303e035 100644 --- a/drivers/gpu/drm/radeon/ni_dpm.h +++ b/drivers/gpu/drm/radeon/ni_dpm.h @@ -59,8 +59,7 @@ struct ni_mc_reg_table { #define NISLANDS_MCREGISTERTABLE_FIRST_DRIVERSTATE_SLOT 2 -enum ni_dc_cac_level -{ +enum ni_dc_cac_level { NISLANDS_DCCAC_LEVEL_0 = 0, NISLANDS_DCCAC_LEVEL_1, NISLANDS_DCCAC_LEVEL_2, @@ -72,8 +71,7 @@ enum ni_dc_cac_level NISLANDS_DCCAC_MAX_LEVELS }; -struct ni_leakage_coeffients -{ +struct ni_leakage_coeffients { u32 at; u32 bt; u32 av; @@ -83,8 +81,7 @@ struct ni_leakage_coeffients u32 t_ref; }; -struct ni_cac_data -{ +struct ni_cac_data { struct ni_leakage_coeffients leakage_coefficients; u32 i_leakage; s32 leakage_minimum_temperature; @@ -100,8 +97,7 @@ struct ni_cac_data u8 lts_truncate_n; }; -struct ni_cac_weights -{ +struct ni_cac_weights { u32 weight_tcp_sig0; u32 weight_tcp_sig1; u32 weight_ta_sig; diff --git a/drivers/gpu/drm/radeon/nislands_smc.h b/drivers/gpu/drm/radeon/nislands_smc.h index 42f3bab0f9ee..097893c38915 100644 --- a/drivers/gpu/drm/radeon/nislands_smc.h +++ b/drivers/gpu/drm/radeon/nislands_smc.h @@ -27,8 +27,7 @@ #define NISLANDS_MAX_SMC_PERFORMANCE_LEVELS_PER_SWSTATE 16 -struct PP_NIslands_Dpm2PerfLevel -{ +struct PP_NIslands_Dpm2PerfLevel { uint8_t MaxPS; uint8_t TgtAct; uint8_t MaxPS_StepInc; @@ -44,8 +43,7 @@ struct PP_NIslands_Dpm2PerfLevel typedef struct PP_NIslands_Dpm2PerfLevel PP_NIslands_Dpm2PerfLevel; -struct PP_NIslands_DPM2Parameters -{ +struct PP_NIslands_DPM2Parameters { uint32_t TDPLimit; uint32_t NearTDPLimit; uint32_t SafePowerLimit; @@ -53,8 +51,7 @@ struct PP_NIslands_DPM2Parameters }; typedef struct PP_NIslands_DPM2Parameters PP_NIslands_DPM2Parameters; -struct NISLANDS_SMC_SCLK_VALUE -{ +struct NISLANDS_SMC_SCLK_VALUE { uint32_t vCG_SPLL_FUNC_CNTL; uint32_t vCG_SPLL_FUNC_CNTL_2; uint32_t vCG_SPLL_FUNC_CNTL_3; @@ -66,8 +63,7 @@ struct NISLANDS_SMC_SCLK_VALUE typedef struct NISLANDS_SMC_SCLK_VALUE NISLANDS_SMC_SCLK_VALUE; -struct NISLANDS_SMC_MCLK_VALUE -{ +struct NISLANDS_SMC_MCLK_VALUE { uint32_t vMPLL_FUNC_CNTL; uint32_t vMPLL_FUNC_CNTL_1; uint32_t vMPLL_FUNC_CNTL_2; @@ -84,8 +80,7 @@ struct NISLANDS_SMC_MCLK_VALUE typedef struct NISLANDS_SMC_MCLK_VALUE NISLANDS_SMC_MCLK_VALUE; -struct NISLANDS_SMC_VOLTAGE_VALUE -{ +struct NISLANDS_SMC_VOLTAGE_VALUE { uint16_t value; uint8_t index; uint8_t padding; @@ -93,8 +88,7 @@ struct NISLANDS_SMC_VOLTAGE_VALUE typedef struct NISLANDS_SMC_VOLTAGE_VALUE NISLANDS_SMC_VOLTAGE_VALUE; -struct NISLANDS_SMC_HW_PERFORMANCE_LEVEL -{ +struct NISLANDS_SMC_HW_PERFORMANCE_LEVEL { uint8_t arbValue; uint8_t ACIndex; uint8_t displayWatermark; @@ -132,8 +126,7 @@ struct NISLANDS_SMC_HW_PERFORMANCE_LEVEL typedef struct NISLANDS_SMC_HW_PERFORMANCE_LEVEL NISLANDS_SMC_HW_PERFORMANCE_LEVEL; -struct NISLANDS_SMC_SWSTATE -{ +struct NISLANDS_SMC_SWSTATE { uint8_t flags; uint8_t levelCount; uint8_t padding2; @@ -156,8 +149,7 @@ struct NISLANDS_SMC_SWSTATE_SINGLE { #define NISLANDS_SMC_VOLTAGEMASK_VDDCI 2 #define NISLANDS_SMC_VOLTAGEMASK_MAX 4 -struct NISLANDS_SMC_VOLTAGEMASKTABLE -{ +struct NISLANDS_SMC_VOLTAGEMASKTABLE { uint8_t highMask[NISLANDS_SMC_VOLTAGEMASK_MAX]; uint32_t lowMask[NISLANDS_SMC_VOLTAGEMASK_MAX]; }; @@ -166,8 +158,7 @@ typedef struct NISLANDS_SMC_VOLTAGEMASKTABLE NISLANDS_SMC_VOLTAGEMASKTABLE; #define NISLANDS_MAX_NO_VREG_STEPS 32 -struct NISLANDS_SMC_STATETABLE -{ +struct NISLANDS_SMC_STATETABLE { uint8_t thermalProtectType; uint8_t systemFlags; uint8_t maxVDDCIndexInPPTable; @@ -203,8 +194,7 @@ typedef struct NISLANDS_SMC_STATETABLE NISLANDS_SMC_STATETABLE; #define SMC_NISLANDS_LKGE_LUT_NUM_OF_VOLT_ENTRIES 16 #define SMC_NISLANDS_BIF_LUT_NUM_OF_ENTRIES 4 -struct SMC_NISLANDS_MC_TPP_CAC_TABLE -{ +struct SMC_NISLANDS_MC_TPP_CAC_TABLE { uint32_t tpp[SMC_NISLANDS_MC_TPP_CAC_NUM_OF_ENTRIES]; uint32_t cacValue[SMC_NISLANDS_MC_TPP_CAC_NUM_OF_ENTRIES]; }; @@ -212,8 +202,7 @@ struct SMC_NISLANDS_MC_TPP_CAC_TABLE typedef struct SMC_NISLANDS_MC_TPP_CAC_TABLE SMC_NISLANDS_MC_TPP_CAC_TABLE; -struct PP_NIslands_CACTABLES -{ +struct PP_NIslands_CACTABLES { uint32_t cac_bif_lut[SMC_NISLANDS_BIF_LUT_NUM_OF_ENTRIES]; uint32_t cac_lkge_lut[SMC_NISLANDS_LKGE_LUT_NUM_OF_TEMP_ENTRIES][SMC_NISLANDS_LKGE_LUT_NUM_OF_VOLT_ENTRIES]; @@ -257,8 +246,7 @@ typedef struct PP_NIslands_CACTABLES PP_NIslands_CACTABLES; #define SMC_NISLANDS_MC_REGISTER_ARRAY_SIZE 32 #define SMC_NISLANDS_MC_REGISTER_ARRAY_SET_COUNT 20 -struct SMC_NIslands_MCRegisterAddress -{ +struct SMC_NIslands_MCRegisterAddress { uint16_t s0; uint16_t s1; }; @@ -266,15 +254,13 @@ struct SMC_NIslands_MCRegisterAddress typedef struct SMC_NIslands_MCRegisterAddress SMC_NIslands_MCRegisterAddress; -struct SMC_NIslands_MCRegisterSet -{ +struct SMC_NIslands_MCRegisterSet { uint32_t value[SMC_NISLANDS_MC_REGISTER_ARRAY_SIZE]; }; typedef struct SMC_NIslands_MCRegisterSet SMC_NIslands_MCRegisterSet; -struct SMC_NIslands_MCRegisters -{ +struct SMC_NIslands_MCRegisters { uint8_t last; uint8_t reserved[3]; SMC_NIslands_MCRegisterAddress address[SMC_NISLANDS_MC_REGISTER_ARRAY_SIZE]; @@ -283,8 +269,7 @@ struct SMC_NIslands_MCRegisters typedef struct SMC_NIslands_MCRegisters SMC_NIslands_MCRegisters; -struct SMC_NIslands_MCArbDramTimingRegisterSet -{ +struct SMC_NIslands_MCArbDramTimingRegisterSet { uint32_t mc_arb_dram_timing; uint32_t mc_arb_dram_timing2; uint8_t mc_arb_rfsh_rate; @@ -293,8 +278,7 @@ struct SMC_NIslands_MCArbDramTimingRegisterSet typedef struct SMC_NIslands_MCArbDramTimingRegisterSet SMC_NIslands_MCArbDramTimingRegisterSet; -struct SMC_NIslands_MCArbDramTimingRegisters -{ +struct SMC_NIslands_MCArbDramTimingRegisters { uint8_t arb_current; uint8_t reserved[3]; SMC_NIslands_MCArbDramTimingRegisterSet data[20]; @@ -302,8 +286,7 @@ struct SMC_NIslands_MCArbDramTimingRegisters typedef struct SMC_NIslands_MCArbDramTimingRegisters SMC_NIslands_MCArbDramTimingRegisters; -struct SMC_NISLANDS_SPLL_DIV_TABLE -{ +struct SMC_NISLANDS_SPLL_DIV_TABLE { uint32_t freq[256]; uint32_t ss[256]; }; diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index cfeca2694d5f..86b8b770af19 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -1327,7 +1327,7 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p, return -EINVAL; } track->num_arrays = c; - for (i = 0; i < (c - 1); i+=2, idx+=3) { + for (i = 0; i < (c - 1); i += 2, idx += 3) { r = radeon_cs_packet_next_reloc(p, &reloc, 0); if (r) { DRM_ERROR("No reloc for packet3 %d\n", diff --git a/drivers/gpu/drm/radeon/r300_reg.h b/drivers/gpu/drm/radeon/r300_reg.h index 9d341cff63ee..d776f929d5c3 100644 --- a/drivers/gpu/drm/radeon/r300_reg.h +++ b/drivers/gpu/drm/radeon/r300_reg.h @@ -825,7 +825,7 @@ # define R300_TX_MIN_FILTER_ANISO_LINEAR (0 << 13) # define R300_TX_MIN_FILTER_ANISO_NEAREST_MIP_NEAREST (1 << 13) # define R300_TX_MIN_FILTER_ANISO_NEAREST_MIP_LINEAR (2 << 13) -# define R300_TX_MIN_FILTER_MASK ( (15 << 11) | (3 << 13) ) +# define R300_TX_MIN_FILTER_MASK ((15 << 11) | (3 << 13)) # define R300_TX_MAX_ANISO_1_TO_1 (0 << 21) # define R300_TX_MAX_ANISO_2_TO_1 (2 << 21) # define R300_TX_MAX_ANISO_4_TO_1 (4 << 21) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index a17b95eec65f..b5e97d95a19f 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -99,8 +99,7 @@ MODULE_FIRMWARE("radeon/SUMO_me.bin"); MODULE_FIRMWARE("radeon/SUMO2_pfp.bin"); MODULE_FIRMWARE("radeon/SUMO2_me.bin"); -static const u32 crtc_offsets[2] = -{ +static const u32 crtc_offsets[2] = { 0, AVIVO_D2CRTC_H_TOTAL - AVIVO_D1CRTC_H_TOTAL }; diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c index 9d2bcb9551e6..64980a61d38a 100644 --- a/drivers/gpu/drm/radeon/r600_dpm.c +++ b/drivers/gpu/drm/radeon/r600_dpm.c @@ -28,8 +28,7 @@ #include "r600_dpm.h" #include "atom.h" -const u32 r600_utc[R600_PM_NUMBER_OF_TC] = -{ +const u32 r600_utc[R600_PM_NUMBER_OF_TC] = { R600_UTC_DFLT_00, R600_UTC_DFLT_01, R600_UTC_DFLT_02, @@ -47,8 +46,7 @@ const u32 r600_utc[R600_PM_NUMBER_OF_TC] = R600_UTC_DFLT_14, }; -const u32 r600_dtc[R600_PM_NUMBER_OF_TC] = -{ +const u32 r600_dtc[R600_PM_NUMBER_OF_TC] = { R600_DTC_DFLT_00, R600_DTC_DFLT_01, R600_DTC_DFLT_02, diff --git a/drivers/gpu/drm/radeon/r600_dpm.h b/drivers/gpu/drm/radeon/r600_dpm.h index 6e4d22ed2a00..5c2513c84c48 100644 --- a/drivers/gpu/drm/radeon/r600_dpm.h +++ b/drivers/gpu/drm/radeon/r600_dpm.h @@ -119,8 +119,7 @@ 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/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 3d3d2109dfeb..3e5ff17e3caf 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1355,14 +1355,12 @@ struct radeon_dpm_thermal { bool high_to_low; }; -enum radeon_clk_action -{ +enum radeon_clk_action { RADEON_SCLK_UP = 1, RADEON_SCLK_DOWN }; -struct radeon_blacklist_clocks -{ +struct radeon_blacklist_clocks { u32 sclk; u32 mclk; enum radeon_clk_action action; diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index 802b5af19261..b5a0109b2e2c 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -2400,10 +2400,10 @@ int radeon_asic_init(struct radeon_device *rdev) case CHIP_RS880: rdev->asic = &rs780_asic; /* 760G/780V/880V don't have UVD */ - if ((rdev->pdev->device == 0x9616)|| - (rdev->pdev->device == 0x9611)|| - (rdev->pdev->device == 0x9613)|| - (rdev->pdev->device == 0x9711)|| + if ((rdev->pdev->device == 0x9616) || + (rdev->pdev->device == 0x9611) || + (rdev->pdev->device == 0x9613) || + (rdev->pdev->device == 0x9711) || (rdev->pdev->device == 0x9713)) rdev->has_uvd = false; else diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index 3596ea4a8b60..bb1f0a3371ab 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -2852,7 +2852,7 @@ int radeon_atom_get_clock_dividers(struct radeon_device *rdev, args.v1.ucAction = clock_type; args.v1.ulClock = cpu_to_le32(clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); dividers->post_div = args.v1.ucPostDiv; dividers->fb_div = args.v1.ucFbDiv; @@ -2866,7 +2866,7 @@ int radeon_atom_get_clock_dividers(struct radeon_device *rdev, args.v2.ucAction = clock_type; args.v2.ulClock = cpu_to_le32(clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); dividers->post_div = args.v2.ucPostDiv; dividers->fb_div = le16_to_cpu(args.v2.usFbDiv); @@ -2881,7 +2881,7 @@ int radeon_atom_get_clock_dividers(struct radeon_device *rdev, if (clock_type == COMPUTE_ENGINE_PLL_PARAM) { args.v3.ulClockParams = cpu_to_le32((clock_type << 24) | clock); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); dividers->post_div = args.v3.ucPostDiv; dividers->enable_post_div = (args.v3.ucCntlFlag & @@ -2901,7 +2901,7 @@ int radeon_atom_get_clock_dividers(struct radeon_device *rdev, if (strobe_mode) args.v5.ucInputFlag = ATOM_PLL_INPUT_FLAG_PLL_STROBE_MODE_EN; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); dividers->post_div = args.v5.ucPostDiv; dividers->enable_post_div = (args.v5.ucCntlFlag & @@ -2920,7 +2920,7 @@ int radeon_atom_get_clock_dividers(struct radeon_device *rdev, /* fusion */ args.v4.ulClock = cpu_to_le32(clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); dividers->post_divider = dividers->post_div = args.v4.ucPostDiv; dividers->real_clock = le32_to_cpu(args.v4.ulClock); @@ -2931,7 +2931,7 @@ int radeon_atom_get_clock_dividers(struct radeon_device *rdev, args.v6_in.ulClock.ulComputeClockFlag = clock_type; args.v6_in.ulClock.ulClockFreq = cpu_to_le32(clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); dividers->whole_fb_div = le16_to_cpu(args.v6_out.ulFbDiv.usFbDiv); dividers->frac_fb_div = le16_to_cpu(args.v6_out.ulFbDiv.usFbDivFrac); @@ -2972,7 +2972,7 @@ int radeon_atom_get_memory_pll_dividers(struct radeon_device *rdev, if (strobe_mode) args.ucInputFlag |= MPLL_INPUT_FLAG_STROBE_MODE_EN; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); mpll_param->clkfrac = le16_to_cpu(args.ulFbDiv.usFbDivFrac); mpll_param->clkf = le16_to_cpu(args.ulFbDiv.usFbDiv); @@ -3005,7 +3005,7 @@ void radeon_atom_set_clock_gating(struct radeon_device *rdev, int enable) args.ucEnable = enable; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } uint32_t radeon_atom_get_engine_clock(struct radeon_device *rdev) @@ -3013,7 +3013,7 @@ uint32_t radeon_atom_get_engine_clock(struct radeon_device *rdev) GET_ENGINE_CLOCK_PS_ALLOCATION args; int index = GetIndexIntoMasterTable(COMMAND, GetEngineClock); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); return le32_to_cpu(args.ulReturnEngineClock); } @@ -3022,7 +3022,7 @@ uint32_t radeon_atom_get_memory_clock(struct radeon_device *rdev) GET_MEMORY_CLOCK_PS_ALLOCATION args; int index = GetIndexIntoMasterTable(COMMAND, GetMemoryClock); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); return le32_to_cpu(args.ulReturnMemoryClock); } @@ -3034,7 +3034,7 @@ void radeon_atom_set_engine_clock(struct radeon_device *rdev, args.ulTargetEngineClock = cpu_to_le32(eng_clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void radeon_atom_set_memory_clock(struct radeon_device *rdev, @@ -3048,7 +3048,7 @@ void radeon_atom_set_memory_clock(struct radeon_device *rdev, args.ulTargetMemoryClock = cpu_to_le32(mem_clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void radeon_atom_set_engine_dram_timings(struct radeon_device *rdev, @@ -3067,7 +3067,7 @@ void radeon_atom_set_engine_dram_timings(struct radeon_device *rdev, if (mem_clock) args.sReserved.ulClock = cpu_to_le32(mem_clock & SET_CLOCK_FREQ_MASK); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void radeon_atom_update_memory_dll(struct radeon_device *rdev, @@ -3078,7 +3078,7 @@ void radeon_atom_update_memory_dll(struct radeon_device *rdev, args = cpu_to_le32(mem_clock); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } void radeon_atom_set_ac_timing(struct radeon_device *rdev, @@ -3090,7 +3090,7 @@ void radeon_atom_set_ac_timing(struct radeon_device *rdev, args.ulTargetMemoryClock = cpu_to_le32(tmp); /* 10 khz */ - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } union set_voltage { @@ -3134,7 +3134,7 @@ void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 v return; } - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); } int radeon_atom_get_max_vddc(struct radeon_device *rdev, u8 voltage_type, @@ -3155,7 +3155,7 @@ int radeon_atom_get_max_vddc(struct radeon_device *rdev, u8 voltage_type, args.v2.ucVoltageMode = 0; args.v2.usVoltageLevel = 0; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *voltage = le16_to_cpu(args.v2.usVoltageLevel); break; @@ -3164,7 +3164,7 @@ int radeon_atom_get_max_vddc(struct radeon_device *rdev, u8 voltage_type, args.v3.ucVoltageMode = ATOM_GET_VOLTAGE_LEVEL; args.v3.usVoltageLevel = cpu_to_le16(voltage_id); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *voltage = le16_to_cpu(args.v3.usVoltageLevel); break; @@ -3200,7 +3200,7 @@ int radeon_atom_get_leakage_id_from_vbios(struct radeon_device *rdev, args.v3.ucVoltageMode = ATOM_GET_LEAKAGE_ID; args.v3.usVoltageLevel = 0; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *leakage_id = le16_to_cpu(args.v3.usVoltageLevel); break; @@ -3327,7 +3327,7 @@ int radeon_atom_get_voltage_evv(struct radeon_device *rdev, args.in.ulSCLKFreq = cpu_to_le32(rdev->pm.dpm.dyn_state.vddc_dependency_on_sclk.entries[entry_id].clk); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *voltage = le16_to_cpu(args.evv_out.usVoltageLevel); @@ -3353,7 +3353,7 @@ int radeon_atom_get_voltage_gpio_settings(struct radeon_device *rdev, args.v2.ucVoltageMode = SET_ASIC_VOLTAGE_MODE_GET_GPIOMASK; args.v2.usVoltageLevel = cpu_to_le16(voltage_level); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *gpio_mask = le32_to_cpu(*(u32 *)&args.v2); @@ -3361,7 +3361,7 @@ int radeon_atom_get_voltage_gpio_settings(struct radeon_device *rdev, args.v2.ucVoltageMode = SET_ASIC_VOLTAGE_MODE_GET_GPIOVAL; args.v2.usVoltageLevel = cpu_to_le16(voltage_level); - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args, sizeof(args)); *gpio_value = le32_to_cpu(*(u32 *)&args.v2); break; diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c index 595354e3ce0b..f557535c1d7b 100644 --- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c +++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c @@ -61,19 +61,23 @@ struct atpx_mux { u16 mux; } __packed; -bool radeon_has_atpx(void) { +bool radeon_has_atpx(void) +{ return radeon_atpx_priv.atpx_detected; } -bool radeon_has_atpx_dgpu_power_cntl(void) { +bool radeon_has_atpx_dgpu_power_cntl(void) +{ return radeon_atpx_priv.atpx.functions.power_cntl; } -bool radeon_is_atpx_hybrid(void) { +bool radeon_is_atpx_hybrid(void) +{ return radeon_atpx_priv.atpx.is_hybrid; } -bool radeon_atpx_dgpu_req_power_for_displays(void) { +bool radeon_atpx_dgpu_req_power_for_displays(void) +{ return radeon_atpx_priv.atpx.dgpu_req_power_for_displays; } diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c index 91b58fbc2be7..74753bb26d33 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.c +++ b/drivers/gpu/drm/radeon/radeon_audio.c @@ -37,15 +37,14 @@ void dce6_audio_enable(struct radeon_device *rdev, struct r600_audio_pin *pin, u8 enable_mask); -struct r600_audio_pin* r600_audio_get_pin(struct radeon_device *rdev); -struct r600_audio_pin* dce6_audio_get_pin(struct radeon_device *rdev); +struct r600_audio_pin *r600_audio_get_pin(struct radeon_device *rdev); +struct r600_audio_pin *dce6_audio_get_pin(struct radeon_device *rdev); static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode); static void radeon_audio_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode); -static const u32 pin_offsets[7] = -{ +static const u32 pin_offsets[7] = { (0x5e00 - 0x5e00), (0x5e18 - 0x5e00), (0x5e30 - 0x5e00), @@ -361,7 +360,7 @@ static void radeon_audio_write_latency_fields(struct drm_encoder *encoder, radeon_encoder->audio->write_latency_fields(encoder, connector, mode); } -struct r600_audio_pin* radeon_audio_get_pin(struct drm_encoder *encoder) +struct r600_audio_pin *radeon_audio_get_pin(struct drm_encoder *encoder) { struct radeon_device *rdev = encoder->dev->dev_private; struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); @@ -528,7 +527,7 @@ static void radeon_audio_calc_cts(unsigned int clock, int *CTS, int *N, int freq *N, *CTS, freq); } -static const struct radeon_hdmi_acr* radeon_audio_acr(unsigned int clock) +static const struct radeon_hdmi_acr *radeon_audio_acr(unsigned int clock) { static struct radeon_hdmi_acr res; u8 i; diff --git a/drivers/gpu/drm/radeon/radeon_audio.h b/drivers/gpu/drm/radeon/radeon_audio.h index dacaaa007051..a073dadd0638 100644 --- a/drivers/gpu/drm/radeon/radeon_audio.h +++ b/drivers/gpu/drm/radeon/radeon_audio.h @@ -34,8 +34,7 @@ struct cea_sad; #define WREG32_ENDPOINT(block, reg, v) \ radeon_audio_endpoint_wreg(rdev, (block), (reg), (v)) -struct radeon_audio_basic_funcs -{ +struct radeon_audio_basic_funcs { u32 (*endpoint_rreg)(struct radeon_device *rdev, u32 offset, u32 reg); void (*endpoint_wreg)(struct radeon_device *rdev, u32 offset, u32 reg, u32 v); @@ -43,8 +42,7 @@ struct radeon_audio_basic_funcs struct r600_audio_pin *pin, u8 enable_mask); }; -struct radeon_audio_funcs -{ +struct radeon_audio_funcs { void (*select_pin)(struct drm_encoder *encoder); struct r600_audio_pin* (*get_pin)(struct radeon_device *rdev); void (*write_latency_fields)(struct drm_encoder *encoder, diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 59c4db13d90a..546381a5c918 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -603,8 +603,7 @@ 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]; @@ -632,14 +631,12 @@ 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/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index b73fd9ab0252..4482c8c5f5ce 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -587,7 +587,7 @@ static ssize_t radeon_hwmon_set_pwm1_enable(struct device *dev, int err; int value; - if(!rdev->asic->dpm.fan_ctrl_set_mode) + if (!rdev->asic->dpm.fan_ctrl_set_mode) return -EINVAL; err = kstrtoint(buf, 10, &value); @@ -789,7 +789,7 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, return 0; /* Skip vddc attribute if get_current_vddc is not implemented */ - if(attr == &sensor_dev_attr_in0_input.dev_attr.attr && + if (attr == &sensor_dev_attr_in0_input.dev_attr.attr && !rdev->asic->dpm.get_current_vddc) return 0; diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c index 922a29e58880..d7f552d441ab 100644 --- a/drivers/gpu/drm/radeon/rs400.c +++ b/drivers/gpu/drm/radeon/rs400.c @@ -86,7 +86,7 @@ int rs400_gart_init(struct radeon_device *rdev) return 0; } /* Check gart size */ - switch(rdev->mc.gtt_size / (1024 * 1024)) { + switch (rdev->mc.gtt_size / (1024 * 1024)) { case 32: case 64: case 128: @@ -116,7 +116,7 @@ int rs400_gart_enable(struct radeon_device *rdev) tmp |= RS690_DIS_OUT_OF_PCI_GART_ACCESS; WREG32_MC(RS690_AIC_CTRL_SCRATCH, tmp); /* Check gart size */ - switch(rdev->mc.gtt_size / (1024 * 1024)) { + switch (rdev->mc.gtt_size / (1024 * 1024)) { case 32: size_reg = RS480_VA_SIZE_32MB; break; diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c index 8cf87a0a2b2a..5c162778899b 100644 --- a/drivers/gpu/drm/radeon/rs600.c +++ b/drivers/gpu/drm/radeon/rs600.c @@ -54,8 +54,7 @@ static void rs600_gpu_init(struct radeon_device *rdev); int rs600_mc_wait_for_idle(struct radeon_device *rdev); -static const u32 crtc_offsets[2] = -{ +static const u32 crtc_offsets[2] = { 0, AVIVO_D2CRTC_H_TOTAL - AVIVO_D1CRTC_H_TOTAL }; diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index 76260fdfbaa7..79709d26d983 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -42,8 +42,7 @@ static void rv515_gpu_init(struct radeon_device *rdev); int rv515_mc_wait_for_idle(struct radeon_device *rdev); -static const u32 crtc_offsets[2] = -{ +static const u32 crtc_offsets[2] = { 0, AVIVO_D2CRTC_H_TOTAL - AVIVO_D1CRTC_H_TOTAL }; diff --git a/drivers/gpu/drm/radeon/rv6xx_dpm.h b/drivers/gpu/drm/radeon/rv6xx_dpm.h index 8035d53ebea6..020c0dc8361d 100644 --- a/drivers/gpu/drm/radeon/rv6xx_dpm.h +++ b/drivers/gpu/drm/radeon/rv6xx_dpm.h @@ -28,8 +28,7 @@ #include "r600_dpm.h" /* Represents a single SCLK step. */ -struct rv6xx_sclk_stepping -{ +struct rv6xx_sclk_stepping { u32 vco_frequency; u32 post_divider; }; diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c index ef2f1a048cfe..e3e1f6833f12 100644 --- a/drivers/gpu/drm/radeon/rv770_dpm.c +++ b/drivers/gpu/drm/radeon/rv770_dpm.c @@ -1010,7 +1010,7 @@ int rv770_populate_initial_mvdd_value(struct radeon_device *rdev, struct rv7xx_power_info *pi = rv770_get_pi(rdev); if ((pi->s0_vid_lower_smio_cntl & pi->mvdd_mask_low) == - (pi->mvdd_low_smio[MVDD_LOW_INDEX] & pi->mvdd_mask_low) ) { + (pi->mvdd_low_smio[MVDD_LOW_INDEX] & pi->mvdd_mask_low)) { voltage->index = MVDD_LOW_INDEX; voltage->value = cpu_to_be16(MVDD_LOW_VALUE); } else { @@ -1260,7 +1260,7 @@ static int rv770_construct_vddc_table(struct radeon_device *rdev) pi->vddc_mask_low = gpio_mask; if (i > 0) { if ((pi->vddc_table[i].low_smio != - pi->vddc_table[i - 1].low_smio ) || + pi->vddc_table[i - 1].low_smio) || (pi->vddc_table[i].high_smio != pi->vddc_table[i - 1].high_smio)) vddc_index++; diff --git a/drivers/gpu/drm/radeon/rv770_smc.h b/drivers/gpu/drm/radeon/rv770_smc.h index 3b2c963c4880..d8e8f70135f2 100644 --- a/drivers/gpu/drm/radeon/rv770_smc.h +++ b/drivers/gpu/drm/radeon/rv770_smc.h @@ -31,8 +31,7 @@ #define RV770_SMC_PERFORMANCE_LEVELS_PER_SWSTATE 3 -struct RV770_SMC_SCLK_VALUE -{ +struct RV770_SMC_SCLK_VALUE { uint32_t vCG_SPLL_FUNC_CNTL; uint32_t vCG_SPLL_FUNC_CNTL_2; uint32_t vCG_SPLL_FUNC_CNTL_3; @@ -43,8 +42,7 @@ struct RV770_SMC_SCLK_VALUE typedef struct RV770_SMC_SCLK_VALUE RV770_SMC_SCLK_VALUE; -struct RV770_SMC_MCLK_VALUE -{ +struct RV770_SMC_MCLK_VALUE { uint32_t vMPLL_AD_FUNC_CNTL; uint32_t vMPLL_AD_FUNC_CNTL_2; uint32_t vMPLL_DQ_FUNC_CNTL; @@ -59,8 +57,7 @@ struct RV770_SMC_MCLK_VALUE typedef struct RV770_SMC_MCLK_VALUE RV770_SMC_MCLK_VALUE; -struct RV730_SMC_MCLK_VALUE -{ +struct RV730_SMC_MCLK_VALUE { uint32_t vMCLK_PWRMGT_CNTL; uint32_t vDLL_CNTL; uint32_t vMPLL_FUNC_CNTL; @@ -73,8 +70,7 @@ struct RV730_SMC_MCLK_VALUE typedef struct RV730_SMC_MCLK_VALUE RV730_SMC_MCLK_VALUE; -struct RV770_SMC_VOLTAGE_VALUE -{ +struct RV770_SMC_VOLTAGE_VALUE { uint16_t value; uint8_t index; uint8_t padding; @@ -82,16 +78,14 @@ struct RV770_SMC_VOLTAGE_VALUE typedef struct RV770_SMC_VOLTAGE_VALUE RV770_SMC_VOLTAGE_VALUE; -union RV7XX_SMC_MCLK_VALUE -{ +union RV7XX_SMC_MCLK_VALUE { RV770_SMC_MCLK_VALUE mclk770; RV730_SMC_MCLK_VALUE mclk730; }; typedef union RV7XX_SMC_MCLK_VALUE RV7XX_SMC_MCLK_VALUE, *LPRV7XX_SMC_MCLK_VALUE; -struct RV770_SMC_HW_PERFORMANCE_LEVEL -{ +struct RV770_SMC_HW_PERFORMANCE_LEVEL { uint8_t arbValue; union{ uint8_t seqValue; @@ -126,8 +120,7 @@ struct RV770_SMC_HW_PERFORMANCE_LEVEL typedef struct RV770_SMC_HW_PERFORMANCE_LEVEL RV770_SMC_HW_PERFORMANCE_LEVEL; -struct RV770_SMC_SWSTATE -{ +struct RV770_SMC_SWSTATE { uint8_t flags; uint8_t padding1; uint8_t padding2; @@ -142,8 +135,7 @@ typedef struct RV770_SMC_SWSTATE RV770_SMC_SWSTATE; #define RV770_SMC_VOLTAGEMASK_VDDCI 2 #define RV770_SMC_VOLTAGEMASK_MAX 4 -struct RV770_SMC_VOLTAGEMASKTABLE -{ +struct RV770_SMC_VOLTAGEMASKTABLE { uint8_t highMask[RV770_SMC_VOLTAGEMASK_MAX]; uint32_t lowMask[RV770_SMC_VOLTAGEMASK_MAX]; }; @@ -152,8 +144,7 @@ typedef struct RV770_SMC_VOLTAGEMASKTABLE RV770_SMC_VOLTAGEMASKTABLE; #define MAX_NO_VREG_STEPS 32 -struct RV770_SMC_STATETABLE -{ +struct RV770_SMC_STATETABLE { uint8_t thermalProtectType; uint8_t systemFlags; uint8_t maxVDDCIndexInPPTable; diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 85e9cba49cec..93f197d96d8f 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -138,8 +138,7 @@ static void si_fini_pg(struct radeon_device *rdev); static void si_fini_cg(struct radeon_device *rdev); static void si_rlc_stop(struct radeon_device *rdev); -static const u32 crtc_offsets[] = -{ +static const u32 crtc_offsets[] = { EVERGREEN_CRTC0_REGISTER_OFFSET, EVERGREEN_CRTC1_REGISTER_OFFSET, EVERGREEN_CRTC2_REGISTER_OFFSET, @@ -148,8 +147,7 @@ static const u32 crtc_offsets[] = EVERGREEN_CRTC5_REGISTER_OFFSET }; -static const u32 si_disp_int_status[] = -{ +static const u32 si_disp_int_status[] = { DISP_INTERRUPT_STATUS, DISP_INTERRUPT_STATUS_CONTINUE, DISP_INTERRUPT_STATUS_CONTINUE2, @@ -162,8 +160,7 @@ static const u32 si_disp_int_status[] = #define DC_HPDx_INT_CONTROL(x) (DC_HPD1_INT_CONTROL + (x * 0xc)) #define DC_HPDx_INT_STATUS_REG(x) (DC_HPD1_INT_STATUS + (x * 0xc)) -static const u32 verde_rlc_save_restore_register_list[] = -{ +static const u32 verde_rlc_save_restore_register_list[] = { (0x8000 << 16) | (0x98f4 >> 2), 0x00000000, (0x8040 << 16) | (0x98f4 >> 2), @@ -384,8 +381,7 @@ static const u32 verde_rlc_save_restore_register_list[] = 0x00000000 }; -static const u32 tahiti_golden_rlc_registers[] = -{ +static const u32 tahiti_golden_rlc_registers[] = { 0xc424, 0xffffffff, 0x00601005, 0xc47c, 0xffffffff, 0x10104040, 0xc488, 0xffffffff, 0x0100000a, @@ -394,8 +390,7 @@ static const u32 tahiti_golden_rlc_registers[] = 0xf4a8, 0xffffffff, 0x00000000 }; -static const u32 tahiti_golden_registers[] = -{ +static const u32 tahiti_golden_registers[] = { 0x9a10, 0x00010000, 0x00018208, 0x9830, 0xffffffff, 0x00000000, 0x9834, 0xf00fffff, 0x00000400, @@ -429,13 +424,11 @@ static const u32 tahiti_golden_registers[] = 0x15c0, 0x000c0fc0, 0x000c0400 }; -static const u32 tahiti_golden_registers2[] = -{ +static const u32 tahiti_golden_registers2[] = { 0xc64, 0x00000001, 0x00000001 }; -static const u32 pitcairn_golden_rlc_registers[] = -{ +static const u32 pitcairn_golden_rlc_registers[] = { 0xc424, 0xffffffff, 0x00601004, 0xc47c, 0xffffffff, 0x10102020, 0xc488, 0xffffffff, 0x01000020, @@ -443,8 +436,7 @@ static const u32 pitcairn_golden_rlc_registers[] = 0xc30c, 0xffffffff, 0x800000a4 }; -static const u32 pitcairn_golden_registers[] = -{ +static const u32 pitcairn_golden_registers[] = { 0x9a10, 0x00010000, 0x00018208, 0x9830, 0xffffffff, 0x00000000, 0x9834, 0xf00fffff, 0x00000400, @@ -474,8 +466,7 @@ static const u32 pitcairn_golden_registers[] = 0x15c0, 0x000c0fc0, 0x000c0400 }; -static const u32 verde_golden_rlc_registers[] = -{ +static const u32 verde_golden_rlc_registers[] = { 0xc424, 0xffffffff, 0x033f1005, 0xc47c, 0xffffffff, 0x10808020, 0xc488, 0xffffffff, 0x00800008, @@ -483,8 +474,7 @@ static const u32 verde_golden_rlc_registers[] = 0xc30c, 0xffffffff, 0x80010014 }; -static const u32 verde_golden_registers[] = -{ +static const u32 verde_golden_registers[] = { 0x9a10, 0x00010000, 0x00018208, 0x9830, 0xffffffff, 0x00000000, 0x9834, 0xf00fffff, 0x00000400, @@ -539,8 +529,7 @@ static const u32 verde_golden_registers[] = 0x15c0, 0x000c0fc0, 0x000c0400 }; -static const u32 oland_golden_rlc_registers[] = -{ +static const u32 oland_golden_rlc_registers[] = { 0xc424, 0xffffffff, 0x00601005, 0xc47c, 0xffffffff, 0x10104040, 0xc488, 0xffffffff, 0x0100000a, @@ -548,8 +537,7 @@ static const u32 oland_golden_rlc_registers[] = 0xc30c, 0xffffffff, 0x800000f4 }; -static const u32 oland_golden_registers[] = -{ +static const u32 oland_golden_registers[] = { 0x9a10, 0x00010000, 0x00018208, 0x9830, 0xffffffff, 0x00000000, 0x9834, 0xf00fffff, 0x00000400, @@ -579,8 +567,7 @@ static const u32 oland_golden_registers[] = 0x15c0, 0x000c0fc0, 0x000c0400 }; -static const u32 hainan_golden_registers[] = -{ +static const u32 hainan_golden_registers[] = { 0x9a10, 0x00010000, 0x00018208, 0x9830, 0xffffffff, 0x00000000, 0x9834, 0xf00fffff, 0x00000400, @@ -608,13 +595,11 @@ static const u32 hainan_golden_registers[] = 0x15c0, 0x000c0fc0, 0x000c0400 }; -static const u32 hainan_golden_registers2[] = -{ +static const u32 hainan_golden_registers2[] = { 0x98f8, 0xffffffff, 0x02010001 }; -static const u32 tahiti_mgcg_cgcg_init[] = -{ +static const u32 tahiti_mgcg_cgcg_init[] = { 0xc400, 0xffffffff, 0xfffffffc, 0x802c, 0xffffffff, 0xe0000000, 0x9a60, 0xffffffff, 0x00000100, @@ -743,8 +728,7 @@ static const u32 tahiti_mgcg_cgcg_init[] = 0xd8c0, 0xfffffff0, 0x00000100 }; -static const u32 pitcairn_mgcg_cgcg_init[] = -{ +static const u32 pitcairn_mgcg_cgcg_init[] = { 0xc400, 0xffffffff, 0xfffffffc, 0x802c, 0xffffffff, 0xe0000000, 0x9a60, 0xffffffff, 0x00000100, @@ -841,8 +825,7 @@ static const u32 pitcairn_mgcg_cgcg_init[] = 0xd8c0, 0xfffffff0, 0x00000100 }; -static const u32 verde_mgcg_cgcg_init[] = -{ +static const u32 verde_mgcg_cgcg_init[] = { 0xc400, 0xffffffff, 0xfffffffc, 0x802c, 0xffffffff, 0xe0000000, 0x9a60, 0xffffffff, 0x00000100, @@ -941,8 +924,7 @@ static const u32 verde_mgcg_cgcg_init[] = 0xd8c0, 0xfffffff0, 0x00000100 }; -static const u32 oland_mgcg_cgcg_init[] = -{ +static const u32 oland_mgcg_cgcg_init[] = { 0xc400, 0xffffffff, 0xfffffffc, 0x802c, 0xffffffff, 0xe0000000, 0x9a60, 0xffffffff, 0x00000100, @@ -1021,8 +1003,7 @@ static const u32 oland_mgcg_cgcg_init[] = 0xd8c0, 0xfffffff0, 0x00000100 }; -static const u32 hainan_mgcg_cgcg_init[] = -{ +static const u32 hainan_mgcg_cgcg_init[] = { 0xc400, 0xffffffff, 0xfffffffc, 0x802c, 0xffffffff, 0xe0000000, 0x9a60, 0xffffffff, 0x00000100, @@ -1098,8 +1079,7 @@ static const u32 hainan_mgcg_cgcg_init[] = 0xd8c0, 0xfffffff0, 0x00000100 }; -static u32 verde_pg_init[] = -{ +static u32 verde_pg_init[] = { 0x353c, 0xffffffff, 0x40000, 0x3538, 0xffffffff, 0x200010ff, 0x353c, 0xffffffff, 0x0, @@ -1768,7 +1748,8 @@ static int si_init_microcode(struct radeon_device *rdev) mc_req_size = mc2_req_size = OLAND_MC_UCODE_SIZE * 4; smc_req_size = ALIGN(HAINAN_SMC_UCODE_SIZE, 4); break; - default: BUG(); + default: + BUG(); } /* this memory configuration requires special firmware */ diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index fbf968e3f6d7..9deb91970d4d 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -46,8 +46,7 @@ #define SCLK_MIN_DEEPSLEEP_FREQ 1350 -static const struct si_cac_config_reg cac_weights_tahiti[] = -{ +static const struct si_cac_config_reg cac_weights_tahiti[] = { { 0x0, 0x0000ffff, 0, 0xc, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x0, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x101, SISLANDS_CACCONFIG_CGIND }, @@ -111,8 +110,7 @@ static const struct si_cac_config_reg cac_weights_tahiti[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg lcac_tahiti[] = -{ +static const struct si_cac_config_reg lcac_tahiti[] = { { 0x143, 0x0001fffe, 1, 0x3, SISLANDS_CACCONFIG_CGIND }, { 0x143, 0x00000001, 0, 0x1, SISLANDS_CACCONFIG_CGIND }, { 0x146, 0x0001fffe, 1, 0x3, SISLANDS_CACCONFIG_CGIND }, @@ -203,13 +201,11 @@ static const struct si_cac_config_reg lcac_tahiti[] = }; -static const struct si_cac_config_reg cac_override_tahiti[] = -{ +static const struct si_cac_config_reg cac_override_tahiti[] = { { 0xFFFFFFFF } }; -static const struct si_powertune_data powertune_data_tahiti = -{ +static const struct si_powertune_data powertune_data_tahiti = { ((1 << 16) | 27027), 6, 0, @@ -239,8 +235,7 @@ static const struct si_powertune_data powertune_data_tahiti = true }; -static const struct si_dte_data dte_data_tahiti = -{ +static const struct si_dte_data dte_data_tahiti = { { 1159409, 0, 0, 0, 0 }, { 777, 0, 0, 0, 0 }, 2, @@ -257,8 +252,7 @@ static const struct si_dte_data dte_data_tahiti = false }; -static const struct si_dte_data dte_data_tahiti_pro = -{ +static const struct si_dte_data dte_data_tahiti_pro = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -275,8 +269,7 @@ static const struct si_dte_data dte_data_tahiti_pro = true }; -static const struct si_dte_data dte_data_new_zealand = -{ +static const struct si_dte_data dte_data_new_zealand = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0 }, { 0x29B, 0x3E9, 0x537, 0x7D2, 0 }, 0x5, @@ -293,8 +286,7 @@ static const struct si_dte_data dte_data_new_zealand = true }; -static const struct si_dte_data dte_data_aruba_pro = -{ +static const struct si_dte_data dte_data_aruba_pro = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -311,8 +303,7 @@ static const struct si_dte_data dte_data_aruba_pro = true }; -static const struct si_dte_data dte_data_malta = -{ +static const struct si_dte_data dte_data_malta = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -329,8 +320,7 @@ static const struct si_dte_data dte_data_malta = true }; -static struct si_cac_config_reg cac_weights_pitcairn[] = -{ +static struct si_cac_config_reg cac_weights_pitcairn[] = { { 0x0, 0x0000ffff, 0, 0x8a, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x0, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x0, SISLANDS_CACCONFIG_CGIND }, @@ -394,8 +384,7 @@ static struct si_cac_config_reg cac_weights_pitcairn[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg lcac_pitcairn[] = -{ +static const struct si_cac_config_reg lcac_pitcairn[] = { { 0x98, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, { 0x98, 0x00000001, 0, 0x1, SISLANDS_CACCONFIG_CGIND }, { 0x104, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, @@ -485,13 +474,11 @@ static const struct si_cac_config_reg lcac_pitcairn[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_override_pitcairn[] = -{ +static const struct si_cac_config_reg cac_override_pitcairn[] = { { 0xFFFFFFFF } }; -static const struct si_powertune_data powertune_data_pitcairn = -{ +static const struct si_powertune_data powertune_data_pitcairn = { ((1 << 16) | 27027), 5, 0, @@ -521,8 +508,7 @@ static const struct si_powertune_data powertune_data_pitcairn = true }; -static const struct si_dte_data dte_data_pitcairn = -{ +static const struct si_dte_data dte_data_pitcairn = { { 0, 0, 0, 0, 0 }, { 0, 0, 0, 0, 0 }, 0, @@ -539,8 +525,7 @@ static const struct si_dte_data dte_data_pitcairn = false }; -static const struct si_dte_data dte_data_curacao_xt = -{ +static const struct si_dte_data dte_data_curacao_xt = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -557,8 +542,7 @@ static const struct si_dte_data dte_data_curacao_xt = true }; -static const struct si_dte_data dte_data_curacao_pro = -{ +static const struct si_dte_data dte_data_curacao_pro = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -575,8 +559,7 @@ static const struct si_dte_data dte_data_curacao_pro = true }; -static const struct si_dte_data dte_data_neptune_xt = -{ +static const struct si_dte_data dte_data_neptune_xt = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -593,8 +576,7 @@ static const struct si_dte_data dte_data_neptune_xt = true }; -static const struct si_cac_config_reg cac_weights_chelsea_pro[] = -{ +static const struct si_cac_config_reg cac_weights_chelsea_pro[] = { { 0x0, 0x0000ffff, 0, 0x82, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x4F, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x153, SISLANDS_CACCONFIG_CGIND }, @@ -658,8 +640,7 @@ static const struct si_cac_config_reg cac_weights_chelsea_pro[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_chelsea_xt[] = -{ +static const struct si_cac_config_reg cac_weights_chelsea_xt[] = { { 0x0, 0x0000ffff, 0, 0x82, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x4F, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x153, SISLANDS_CACCONFIG_CGIND }, @@ -723,8 +704,7 @@ static const struct si_cac_config_reg cac_weights_chelsea_xt[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_heathrow[] = -{ +static const struct si_cac_config_reg cac_weights_heathrow[] = { { 0x0, 0x0000ffff, 0, 0x82, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x4F, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x153, SISLANDS_CACCONFIG_CGIND }, @@ -788,8 +768,7 @@ static const struct si_cac_config_reg cac_weights_heathrow[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_cape_verde_pro[] = -{ +static const struct si_cac_config_reg cac_weights_cape_verde_pro[] = { { 0x0, 0x0000ffff, 0, 0x82, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x4F, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x153, SISLANDS_CACCONFIG_CGIND }, @@ -853,8 +832,7 @@ static const struct si_cac_config_reg cac_weights_cape_verde_pro[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_cape_verde[] = -{ +static const struct si_cac_config_reg cac_weights_cape_verde[] = { { 0x0, 0x0000ffff, 0, 0x82, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x4F, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x153, SISLANDS_CACCONFIG_CGIND }, @@ -918,8 +896,7 @@ static const struct si_cac_config_reg cac_weights_cape_verde[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg lcac_cape_verde[] = -{ +static const struct si_cac_config_reg lcac_cape_verde[] = { { 0x98, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, { 0x98, 0x00000001, 0, 0x1, SISLANDS_CACCONFIG_CGIND }, { 0x104, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, @@ -977,13 +954,11 @@ static const struct si_cac_config_reg lcac_cape_verde[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_override_cape_verde[] = -{ +static const struct si_cac_config_reg cac_override_cape_verde[] = { { 0xFFFFFFFF } }; -static const struct si_powertune_data powertune_data_cape_verde = -{ +static const struct si_powertune_data powertune_data_cape_verde = { ((1 << 16) | 0x6993), 5, 0, @@ -1013,8 +988,7 @@ static const struct si_powertune_data powertune_data_cape_verde = true }; -static const struct si_dte_data dte_data_cape_verde = -{ +static const struct si_dte_data dte_data_cape_verde = { { 0, 0, 0, 0, 0 }, { 0, 0, 0, 0, 0 }, 0, @@ -1031,8 +1005,7 @@ static const struct si_dte_data dte_data_cape_verde = false }; -static const struct si_dte_data dte_data_venus_xtx = -{ +static const struct si_dte_data dte_data_venus_xtx = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x71C, 0xAAB, 0xE39, 0x11C7, 0x0 }, 5, @@ -1049,8 +1022,7 @@ static const struct si_dte_data dte_data_venus_xtx = true }; -static const struct si_dte_data dte_data_venus_xt = -{ +static const struct si_dte_data dte_data_venus_xt = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0xBDA, 0x11C7, 0x17B4, 0x1DA1, 0x0 }, 5, @@ -1067,8 +1039,7 @@ static const struct si_dte_data dte_data_venus_xt = true }; -static const struct si_dte_data dte_data_venus_pro = -{ +static const struct si_dte_data dte_data_venus_pro = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x11C7, 0x1AAB, 0x238E, 0x2C72, 0x0 }, 5, @@ -1085,8 +1056,7 @@ static const struct si_dte_data dte_data_venus_pro = true }; -static struct si_cac_config_reg cac_weights_oland[] = -{ +static struct si_cac_config_reg cac_weights_oland[] = { { 0x0, 0x0000ffff, 0, 0x82, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x4F, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x153, SISLANDS_CACCONFIG_CGIND }, @@ -1150,8 +1120,7 @@ static struct si_cac_config_reg cac_weights_oland[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_mars_pro[] = -{ +static const struct si_cac_config_reg cac_weights_mars_pro[] = { { 0x0, 0x0000ffff, 0, 0x43, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x29, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0xAF, SISLANDS_CACCONFIG_CGIND }, @@ -1215,8 +1184,7 @@ static const struct si_cac_config_reg cac_weights_mars_pro[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_mars_xt[] = -{ +static const struct si_cac_config_reg cac_weights_mars_xt[] = { { 0x0, 0x0000ffff, 0, 0x43, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x29, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0xAF, SISLANDS_CACCONFIG_CGIND }, @@ -1280,8 +1248,7 @@ static const struct si_cac_config_reg cac_weights_mars_xt[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_oland_pro[] = -{ +static const struct si_cac_config_reg cac_weights_oland_pro[] = { { 0x0, 0x0000ffff, 0, 0x43, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x29, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0xAF, SISLANDS_CACCONFIG_CGIND }, @@ -1345,8 +1312,7 @@ static const struct si_cac_config_reg cac_weights_oland_pro[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_weights_oland_xt[] = -{ +static const struct si_cac_config_reg cac_weights_oland_xt[] = { { 0x0, 0x0000ffff, 0, 0x43, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x29, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0xAF, SISLANDS_CACCONFIG_CGIND }, @@ -1410,8 +1376,7 @@ static const struct si_cac_config_reg cac_weights_oland_xt[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg lcac_oland[] = -{ +static const struct si_cac_config_reg lcac_oland[] = { { 0x98, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, { 0x98, 0x00000001, 0, 0x1, SISLANDS_CACCONFIG_CGIND }, { 0x104, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, @@ -1457,8 +1422,7 @@ static const struct si_cac_config_reg lcac_oland[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg lcac_mars_pro[] = -{ +static const struct si_cac_config_reg lcac_mars_pro[] = { { 0x98, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, { 0x98, 0x00000001, 0, 0x1, SISLANDS_CACCONFIG_CGIND }, { 0x104, 0x0001fffe, 1, 0x2, SISLANDS_CACCONFIG_CGIND }, @@ -1504,13 +1468,11 @@ static const struct si_cac_config_reg lcac_mars_pro[] = { 0xFFFFFFFF } }; -static const struct si_cac_config_reg cac_override_oland[] = -{ +static const struct si_cac_config_reg cac_override_oland[] = { { 0xFFFFFFFF } }; -static const struct si_powertune_data powertune_data_oland = -{ +static const struct si_powertune_data powertune_data_oland = { ((1 << 16) | 0x6993), 5, 0, @@ -1540,8 +1502,7 @@ static const struct si_powertune_data powertune_data_oland = true }; -static const struct si_powertune_data powertune_data_mars_pro = -{ +static const struct si_powertune_data powertune_data_mars_pro = { ((1 << 16) | 0x6993), 5, 0, @@ -1571,8 +1532,7 @@ static const struct si_powertune_data powertune_data_mars_pro = true }; -static const struct si_dte_data dte_data_oland = -{ +static const struct si_dte_data dte_data_oland = { { 0, 0, 0, 0, 0 }, { 0, 0, 0, 0, 0 }, 0, @@ -1589,8 +1549,7 @@ static const struct si_dte_data dte_data_oland = false }; -static const struct si_dte_data dte_data_mars_pro = -{ +static const struct si_dte_data dte_data_mars_pro = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -1607,8 +1566,7 @@ static const struct si_dte_data dte_data_mars_pro = true }; -static const struct si_dte_data dte_data_sun_xt = -{ +static const struct si_dte_data dte_data_sun_xt = { { 0x1E8480, 0x3D0900, 0x989680, 0x2625A00, 0x0 }, { 0x0, 0x0, 0x0, 0x0, 0x0 }, 5, @@ -1626,8 +1584,7 @@ static const struct si_dte_data dte_data_sun_xt = }; -static const struct si_cac_config_reg cac_weights_hainan[] = -{ +static const struct si_cac_config_reg cac_weights_hainan[] = { { 0x0, 0x0000ffff, 0, 0x2d9, SISLANDS_CACCONFIG_CGIND }, { 0x0, 0xffff0000, 16, 0x22b, SISLANDS_CACCONFIG_CGIND }, { 0x1, 0x0000ffff, 0, 0x21c, SISLANDS_CACCONFIG_CGIND }, @@ -1691,8 +1648,7 @@ static const struct si_cac_config_reg cac_weights_hainan[] = { 0xFFFFFFFF } }; -static const struct si_powertune_data powertune_data_hainan = -{ +static const struct si_powertune_data powertune_data_hainan = { ((1 << 16) | 0x6993), 5, 0, diff --git a/drivers/gpu/drm/radeon/si_dpm.h b/drivers/gpu/drm/radeon/si_dpm.h index aa857906ef93..4887edebd348 100644 --- a/drivers/gpu/drm/radeon/si_dpm.h +++ b/drivers/gpu/drm/radeon/si_dpm.h @@ -26,15 +26,13 @@ #include "ni_dpm.h" #include "sislands_smc.h" -enum si_cac_config_reg_type -{ +enum si_cac_config_reg_type { SISLANDS_CACCONFIG_MMR = 0, SISLANDS_CACCONFIG_CGIND, SISLANDS_CACCONFIG_MAX }; -struct si_cac_config_reg -{ +struct si_cac_config_reg { u32 offset; u32 mask; u32 shift; @@ -42,8 +40,7 @@ struct si_cac_config_reg enum si_cac_config_reg_type type; }; -struct si_powertune_data -{ +struct si_powertune_data { u32 cac_window; u32 l2_lta_window_size_default; u8 lts_truncate_default; @@ -56,8 +53,7 @@ struct si_powertune_data bool enable_powertune_by_default; }; -struct si_dyn_powertune_data -{ +struct si_dyn_powertune_data { u32 cac_leakage; s32 leakage_minimum_temperature; u32 wintime; @@ -68,8 +64,7 @@ struct si_dyn_powertune_data bool disable_uvd_powertune; }; -struct si_dte_data -{ +struct si_dte_data { u32 tau[SMC_SISLANDS_DTE_MAX_FILTER_STAGES]; u32 r[SMC_SISLANDS_DTE_MAX_FILTER_STAGES]; u32 k; @@ -122,8 +117,7 @@ struct si_mc_reg_table { #define SISLANDS_MCREGISTERTABLE_ULV_SLOT 2 #define SISLANDS_MCREGISTERTABLE_FIRST_DRIVERSTATE_SLOT 3 -struct si_leakage_voltage_entry -{ +struct si_leakage_voltage_entry { u16 voltage; u16 leakage_index; }; @@ -131,8 +125,7 @@ struct si_leakage_voltage_entry #define SISLANDS_LEAKAGE_INDEX0 0xff01 #define SISLANDS_MAX_LEAKAGE_COUNT 4 -struct si_leakage_voltage -{ +struct si_leakage_voltage { u16 count; struct si_leakage_voltage_entry entries[SISLANDS_MAX_LEAKAGE_COUNT]; }; diff --git a/drivers/gpu/drm/radeon/smu7.h b/drivers/gpu/drm/radeon/smu7.h index 75a380a15292..985d720dbc0d 100644 --- a/drivers/gpu/drm/radeon/smu7.h +++ b/drivers/gpu/drm/radeon/smu7.h @@ -82,8 +82,7 @@ #define SCRATCH_B_CURR_SAMU_INDEX_MASK (0x7<<SCRATCH_B_CURR_SAMU_INDEX_SHIFT) -struct SMU7_PIDController -{ +struct SMU7_PIDController { uint32_t Ki; int32_t LFWindupUL; int32_t LFWindupLL; @@ -117,8 +116,7 @@ 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/radeon/smu7_discrete.h b/drivers/gpu/drm/radeon/smu7_discrete.h index 0b0b404ff091..1f63cbbd6515 100644 --- a/drivers/gpu/drm/radeon/smu7_discrete.h +++ b/drivers/gpu/drm/radeon/smu7_discrete.h @@ -35,8 +35,7 @@ #define SMU7_NUM_GPU_TES 1 #define SMU7_NUM_NON_TES 2 -struct SMU7_SoftRegisters -{ +struct SMU7_SoftRegisters { uint32_t RefClockFrequency; uint32_t PmTimerP; uint32_t FeatureEnables; @@ -89,8 +88,7 @@ struct SMU7_SoftRegisters typedef struct SMU7_SoftRegisters SMU7_SoftRegisters; -struct SMU7_Discrete_VoltageLevel -{ +struct SMU7_Discrete_VoltageLevel { uint16_t Voltage; uint16_t StdVoltageHiSidd; uint16_t StdVoltageLoSidd; @@ -100,8 +98,7 @@ struct SMU7_Discrete_VoltageLevel typedef struct SMU7_Discrete_VoltageLevel SMU7_Discrete_VoltageLevel; -struct SMU7_Discrete_GraphicsLevel -{ +struct SMU7_Discrete_GraphicsLevel { uint32_t Flags; uint32_t MinVddc; uint32_t MinVddcPhases; @@ -131,8 +128,7 @@ struct SMU7_Discrete_GraphicsLevel typedef struct SMU7_Discrete_GraphicsLevel SMU7_Discrete_GraphicsLevel; -struct SMU7_Discrete_ACPILevel -{ +struct SMU7_Discrete_ACPILevel { uint32_t Flags; uint32_t MinVddc; uint32_t MinVddcPhases; @@ -153,8 +149,7 @@ struct SMU7_Discrete_ACPILevel typedef struct SMU7_Discrete_ACPILevel SMU7_Discrete_ACPILevel; -struct SMU7_Discrete_Ulv -{ +struct SMU7_Discrete_Ulv { uint32_t CcPwrDynRm; uint32_t CcPwrDynRm1; uint16_t VddcOffset; @@ -165,8 +160,7 @@ struct SMU7_Discrete_Ulv typedef struct SMU7_Discrete_Ulv SMU7_Discrete_Ulv; -struct SMU7_Discrete_MemoryLevel -{ +struct SMU7_Discrete_MemoryLevel { uint32_t MinVddc; uint32_t MinVddcPhases; uint32_t MinVddci; @@ -206,8 +200,7 @@ struct SMU7_Discrete_MemoryLevel typedef struct SMU7_Discrete_MemoryLevel SMU7_Discrete_MemoryLevel; -struct SMU7_Discrete_LinkLevel -{ +struct SMU7_Discrete_LinkLevel { uint8_t PcieGenSpeed; uint8_t PcieLaneCount; uint8_t EnabledForActivity; @@ -220,8 +213,7 @@ struct SMU7_Discrete_LinkLevel typedef struct SMU7_Discrete_LinkLevel SMU7_Discrete_LinkLevel; -struct SMU7_Discrete_MCArbDramTimingTableEntry -{ +struct SMU7_Discrete_MCArbDramTimingTableEntry { uint32_t McArbDramTiming; uint32_t McArbDramTiming2; uint8_t McArbBurstTime; @@ -230,15 +222,13 @@ struct SMU7_Discrete_MCArbDramTimingTableEntry typedef struct SMU7_Discrete_MCArbDramTimingTableEntry SMU7_Discrete_MCArbDramTimingTableEntry; -struct SMU7_Discrete_MCArbDramTimingTable -{ +struct SMU7_Discrete_MCArbDramTimingTable { SMU7_Discrete_MCArbDramTimingTableEntry entries[SMU__NUM_SCLK_DPM_STATE][SMU__NUM_MCLK_DPM_LEVELS]; }; typedef struct SMU7_Discrete_MCArbDramTimingTable SMU7_Discrete_MCArbDramTimingTable; -struct SMU7_Discrete_UvdLevel -{ +struct SMU7_Discrete_UvdLevel { uint32_t VclkFrequency; uint32_t DclkFrequency; uint16_t MinVddc; @@ -250,8 +240,7 @@ struct SMU7_Discrete_UvdLevel typedef struct SMU7_Discrete_UvdLevel SMU7_Discrete_UvdLevel; -struct SMU7_Discrete_ExtClkLevel -{ +struct SMU7_Discrete_ExtClkLevel { uint32_t Frequency; uint16_t MinVoltage; uint8_t MinPhases; @@ -260,8 +249,7 @@ struct SMU7_Discrete_ExtClkLevel typedef struct SMU7_Discrete_ExtClkLevel SMU7_Discrete_ExtClkLevel; -struct SMU7_Discrete_StateInfo -{ +struct SMU7_Discrete_StateInfo { uint32_t SclkFrequency; uint32_t MclkFrequency; uint32_t VclkFrequency; @@ -285,8 +273,7 @@ struct SMU7_Discrete_StateInfo typedef struct SMU7_Discrete_StateInfo SMU7_Discrete_StateInfo; -struct SMU7_Discrete_DpmTable -{ +struct SMU7_Discrete_DpmTable { SMU7_PIDController GraphicsPIDController; SMU7_PIDController MemoryPIDController; SMU7_PIDController LinkPIDController; @@ -406,23 +393,20 @@ typedef struct SMU7_Discrete_DpmTable SMU7_Discrete_DpmTable; #define SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE 16 #define SMU7_DISCRETE_MC_REGISTER_ARRAY_SET_COUNT SMU7_MAX_LEVELS_MEMORY -struct SMU7_Discrete_MCRegisterAddress -{ +struct SMU7_Discrete_MCRegisterAddress { uint16_t s0; uint16_t s1; }; typedef struct SMU7_Discrete_MCRegisterAddress SMU7_Discrete_MCRegisterAddress; -struct SMU7_Discrete_MCRegisterSet -{ +struct SMU7_Discrete_MCRegisterSet { uint32_t value[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE]; }; typedef struct SMU7_Discrete_MCRegisterSet SMU7_Discrete_MCRegisterSet; -struct SMU7_Discrete_MCRegisters -{ +struct SMU7_Discrete_MCRegisters { uint8_t last; uint8_t reserved[3]; SMU7_Discrete_MCRegisterAddress address[SMU7_DISCRETE_MC_REGISTER_ARRAY_SIZE]; @@ -431,8 +415,7 @@ struct SMU7_Discrete_MCRegisters typedef struct SMU7_Discrete_MCRegisters SMU7_Discrete_MCRegisters; -struct SMU7_Discrete_FanTable -{ +struct SMU7_Discrete_FanTable { uint16_t FdoMode; int16_t TempMin; int16_t TempMed; diff --git a/drivers/gpu/drm/radeon/smu7_fusion.h b/drivers/gpu/drm/radeon/smu7_fusion.h index 78ada9ffd508..e130f52fe8d6 100644 --- a/drivers/gpu/drm/radeon/smu7_fusion.h +++ b/drivers/gpu/drm/radeon/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/radeon/sumo_dpm.c b/drivers/gpu/drm/radeon/sumo_dpm.c index d49c145db437..21d27e6235f3 100644 --- a/drivers/gpu/drm/radeon/sumo_dpm.c +++ b/drivers/gpu/drm/radeon/sumo_dpm.c @@ -33,8 +33,7 @@ #define SUMO_MINIMUM_ENGINE_CLOCK 800 #define BOOST_DPM_LEVEL 7 -static const u32 sumo_utc[SUMO_PM_NUMBER_OF_TC] = -{ +static const u32 sumo_utc[SUMO_PM_NUMBER_OF_TC] = { SUMO_UTC_DFLT_00, SUMO_UTC_DFLT_01, SUMO_UTC_DFLT_02, @@ -52,8 +51,7 @@ static const u32 sumo_utc[SUMO_PM_NUMBER_OF_TC] = SUMO_UTC_DFLT_14, }; -static const u32 sumo_dtc[SUMO_PM_NUMBER_OF_TC] = -{ +static const u32 sumo_dtc[SUMO_PM_NUMBER_OF_TC] = { SUMO_DTC_DFLT_00, SUMO_DTC_DFLT_01, SUMO_DTC_DFLT_02, @@ -109,11 +107,11 @@ static void sumo_mg_clockgating_enable(struct radeon_device *rdev, bool enable) local1 = RREG32(CG_CGTT_LOCAL_1); if (enable) { - WREG32(CG_CGTT_LOCAL_0, (0 & CGCG_CGTT_LOCAL0_MASK) | (local0 & ~CGCG_CGTT_LOCAL0_MASK) ); - WREG32(CG_CGTT_LOCAL_1, (0 & CGCG_CGTT_LOCAL1_MASK) | (local1 & ~CGCG_CGTT_LOCAL1_MASK) ); + WREG32(CG_CGTT_LOCAL_0, (0 & CGCG_CGTT_LOCAL0_MASK) | (local0 & ~CGCG_CGTT_LOCAL0_MASK)); + WREG32(CG_CGTT_LOCAL_1, (0 & CGCG_CGTT_LOCAL1_MASK) | (local1 & ~CGCG_CGTT_LOCAL1_MASK)); } else { - WREG32(CG_CGTT_LOCAL_0, (0xFFFFFFFF & CGCG_CGTT_LOCAL0_MASK) | (local0 & ~CGCG_CGTT_LOCAL0_MASK) ); - WREG32(CG_CGTT_LOCAL_1, (0xFFFFCFFF & CGCG_CGTT_LOCAL1_MASK) | (local1 & ~CGCG_CGTT_LOCAL1_MASK) ); + WREG32(CG_CGTT_LOCAL_0, (0xFFFFFFFF & CGCG_CGTT_LOCAL0_MASK) | (local0 & ~CGCG_CGTT_LOCAL0_MASK)); + WREG32(CG_CGTT_LOCAL_1, (0xFFFFCFFF & CGCG_CGTT_LOCAL1_MASK) | (local1 & ~CGCG_CGTT_LOCAL1_MASK)); } } @@ -702,9 +700,9 @@ static void sumo_post_notify_alt_vddnb_change(struct radeon_device *rdev, u32 nbps1_new = 0; if (old_ps != NULL) - nbps1_old = (old_ps->flags & SUMO_POWERSTATE_FLAGS_FORCE_NBPS1_STATE)? 1 : 0; + nbps1_old = (old_ps->flags & SUMO_POWERSTATE_FLAGS_FORCE_NBPS1_STATE) ? 1 : 0; - nbps1_new = (new_ps->flags & SUMO_POWERSTATE_FLAGS_FORCE_NBPS1_STATE)? 1 : 0; + nbps1_new = (new_ps->flags & SUMO_POWERSTATE_FLAGS_FORCE_NBPS1_STATE) ? 1 : 0; if (nbps1_old == 0 && nbps1_new == 1) sumo_smu_notify_alt_vddnb_change(rdev, 1, 1); diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c index ef1cc7bad20a..b9a2c7ccc881 100644 --- a/drivers/gpu/drm/radeon/trinity_dpm.c +++ b/drivers/gpu/drm/radeon/trinity_dpm.c @@ -39,8 +39,7 @@ #ifndef TRINITY_MGCG_SEQUENCE #define TRINITY_MGCG_SEQUENCE 100 -static const u32 trinity_mgcg_shls_default[] = -{ +static const u32 trinity_mgcg_shls_default[] = { /* Register, Value, Mask */ 0x0000802c, 0xc0000000, 0xffffffff, 0x00003fc4, 0xc0000000, 0xffffffff, @@ -122,8 +121,7 @@ static const u32 trinity_mgcg_shls_default[] = #ifndef TRINITY_SYSLS_SEQUENCE #define TRINITY_SYSLS_SEQUENCE 100 -static const u32 trinity_sysls_disable[] = -{ +static const u32 trinity_sysls_disable[] = { /* Register, Value, Mask */ 0x0000d0c0, 0x00000000, 0xffffffff, 0x0000d8c0, 0x00000000, 0xffffffff, @@ -146,8 +144,7 @@ static const u32 trinity_sysls_disable[] = 0x00006dfc, 0x0000007f, 0xffffffff }; -static const u32 trinity_sysls_enable[] = -{ +static const u32 trinity_sysls_enable[] = { /* Register, Value, Mask */ 0x000055e8, 0x00000001, 0xffffffff, 0x0000d0bc, 0x00000100, 0xffffffff, @@ -169,8 +166,7 @@ static const u32 trinity_sysls_enable[] = }; #endif -static const u32 trinity_override_mgpg_sequences[] = -{ +static const u32 trinity_override_mgpg_sequences[] = { /* Register, Value */ 0x00000200, 0xE030032C, 0x00000204, 0x00000FFF, @@ -366,9 +362,9 @@ static void trinity_mg_clockgating_enable(struct radeon_device *rdev, local1 = RREG32_CG(CG_CGTT_LOCAL_1); WREG32_CG(CG_CGTT_LOCAL_0, - (0x00380000 & CGCG_CGTT_LOCAL0_MASK) | (local0 & ~CGCG_CGTT_LOCAL0_MASK) ); + (0x00380000 & CGCG_CGTT_LOCAL0_MASK) | (local0 & ~CGCG_CGTT_LOCAL0_MASK)); WREG32_CG(CG_CGTT_LOCAL_1, - (0x0E000000 & CGCG_CGTT_LOCAL1_MASK) | (local1 & ~CGCG_CGTT_LOCAL1_MASK) ); + (0x0E000000 & CGCG_CGTT_LOCAL1_MASK) | (local1 & ~CGCG_CGTT_LOCAL1_MASK)); WREG32(CGTS_SM_CTRL_REG, CGTS_SM_CTRL_REG_ENABLE); } else { @@ -378,9 +374,9 @@ static void trinity_mg_clockgating_enable(struct radeon_device *rdev, local1 = RREG32_CG(CG_CGTT_LOCAL_1); WREG32_CG(CG_CGTT_LOCAL_0, - CGCG_CGTT_LOCAL0_MASK | (local0 & ~CGCG_CGTT_LOCAL0_MASK) ); + CGCG_CGTT_LOCAL0_MASK | (local0 & ~CGCG_CGTT_LOCAL0_MASK)); WREG32_CG(CG_CGTT_LOCAL_1, - CGCG_CGTT_LOCAL1_MASK | (local1 & ~CGCG_CGTT_LOCAL1_MASK) ); + CGCG_CGTT_LOCAL1_MASK | (local1 & ~CGCG_CGTT_LOCAL1_MASK)); } } @@ -1434,7 +1430,7 @@ static void trinity_adjust_uvd_state(struct radeon_device *rdev, if (pi->uvd_dpm && r600_is_uvd_state(rps->class, rps->class2)) { high_index = trinity_get_uvd_clock_index(rdev, rps); - switch(high_index) { + switch (high_index) { case 3: case 2: low_index = 1; diff --git a/drivers/gpu/drm/radeon/trinity_dpm.h b/drivers/gpu/drm/radeon/trinity_dpm.h index c261657750ca..431e2b68d21e 100644 --- a/drivers/gpu/drm/radeon/trinity_dpm.h +++ b/drivers/gpu/drm/radeon/trinity_dpm.h @@ -64,8 +64,7 @@ struct trinity_ps { #define TRINITY_NUM_NBPSTATES 4 -struct trinity_uvd_clock_table_entry -{ +struct trinity_uvd_clock_table_entry { u32 vclk; u32 dclk; u8 vclk_did; diff --git a/drivers/gpu/drm/radeon/uvd_v1_0.c b/drivers/gpu/drm/radeon/uvd_v1_0.c index 58557c2263a7..5684639d20a6 100644 --- a/drivers/gpu/drm/radeon/uvd_v1_0.c +++ b/drivers/gpu/drm/radeon/uvd_v1_0.c @@ -142,7 +142,7 @@ int uvd_v1_0_resume(struct radeon_device *rdev) addr = (rdev->uvd.gpu_addr >> 32) & 0xFF; WREG32(UVD_LMI_EXT40_ADDR, addr | (0x9 << 16) | (0x1 << 31)); - WREG32(UVD_FW_START, *((uint32_t*)rdev->uvd.cpu_addr)); + WREG32(UVD_FW_START, *((uint32_t *)rdev->uvd.cpu_addr)); return 0; } diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c index ccf83c12b545..866d1dd6eeb4 100644 --- a/drivers/gpu/drm/xe/display/xe_plane_initial.c +++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c @@ -10,6 +10,7 @@ #include "i915_drv.h" #include "intel_atomic_plane.h" +#include "intel_crtc.h" #include "intel_display.h" #include "intel_display_types.h" #include "intel_fb.h" @@ -18,19 +19,20 @@ #include "intel_plane_initial.h" static bool -intel_reuse_initial_plane_obj(struct drm_i915_private *i915, - const struct intel_initial_plane_config *plane_config, +intel_reuse_initial_plane_obj(struct intel_crtc *this, + const struct intel_initial_plane_config plane_configs[], struct drm_framebuffer **fb) { + struct drm_i915_private *i915 = to_i915(this->base.dev); struct intel_crtc *crtc; for_each_intel_crtc(&i915->drm, crtc) { - struct intel_crtc_state *crtc_state = - to_intel_crtc_state(crtc->base.state); struct intel_plane *plane = to_intel_plane(crtc->base.primary); - struct intel_plane_state *plane_state = + const struct intel_plane_state *plane_state = to_intel_plane_state(plane->base.state); + const struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); if (!crtc_state->uapi.active) continue; @@ -38,7 +40,7 @@ intel_reuse_initial_plane_obj(struct drm_i915_private *i915, if (!plane_state->ggtt_vma) continue; - if (intel_plane_ggtt_offset(plane_state) == plane_config->base) { + if (plane_configs[this->pipe].base == plane_configs[crtc->pipe].base) { *fb = plane_state->hw.fb; return true; } @@ -178,10 +180,10 @@ err_bo: static void intel_find_initial_plane_obj(struct intel_crtc *crtc, - struct intel_initial_plane_config *plane_config) + struct intel_initial_plane_config plane_configs[]) { - struct drm_device *dev = crtc->base.dev; - struct drm_i915_private *dev_priv = to_i915(dev); + struct intel_initial_plane_config *plane_config = + &plane_configs[crtc->pipe]; struct intel_plane *plane = to_intel_plane(crtc->base.primary); struct intel_plane_state *plane_state = @@ -201,7 +203,7 @@ intel_find_initial_plane_obj(struct intel_crtc *crtc, if (intel_alloc_initial_plane_obj(crtc, plane_config)) fb = &plane_config->fb->base; - else if (!intel_reuse_initial_plane_obj(dev_priv, plane_config, &fb)) + else if (!intel_reuse_initial_plane_obj(crtc, plane_configs, &fb)) goto nofb; plane_state->uapi.rotation = plane_config->rotation; @@ -267,25 +269,36 @@ static void plane_config_fini(struct intel_initial_plane_config *plane_config) } } -void intel_crtc_initial_plane_config(struct intel_crtc *crtc) +void intel_initial_plane_config(struct drm_i915_private *i915) { - struct xe_device *xe = to_xe_device(crtc->base.dev); - struct intel_initial_plane_config plane_config = {}; + struct intel_initial_plane_config plane_configs[I915_MAX_PIPES] = {}; + struct intel_crtc *crtc; - /* - * Note that reserving the BIOS fb up front prevents us - * from stuffing other stolen allocations like the ring - * on top. This prevents some ugliness at boot time, and - * can even allow for smooth boot transitions if the BIOS - * fb is large enough for the active pipe configuration. - */ - xe->display.funcs.display->get_initial_plane_config(crtc, &plane_config); + for_each_intel_crtc(&i915->drm, crtc) { + struct intel_initial_plane_config *plane_config = + &plane_configs[crtc->pipe]; - /* - * If the fb is shared between multiple heads, we'll - * just get the first one. - */ - intel_find_initial_plane_obj(crtc, &plane_config); + if (!to_intel_crtc_state(crtc->base.state)->uapi.active) + continue; - plane_config_fini(&plane_config); + /* + * Note that reserving the BIOS fb up front prevents us + * from stuffing other stolen allocations like the ring + * on top. This prevents some ugliness at boot time, and + * can even allow for smooth boot transitions if the BIOS + * fb is large enough for the active pipe configuration. + */ + i915->display.funcs.display->get_initial_plane_config(crtc, plane_config); + + /* + * If the fb is shared between multiple heads, we'll + * just get the first one. + */ + intel_find_initial_plane_obj(crtc, plane_configs); + + if (i915->display.funcs.display->fixup_initial_plane_config(crtc, plane_config)) + intel_crtc_wait_for_next_vblank(crtc); + + plane_config_fini(plane_config); + } } diff --git a/include/drm/display/drm_dp.h b/include/drm/display/drm_dp.h index 3731828825bd..281afff6ee4e 100644 --- a/include/drm/display/drm_dp.h +++ b/include/drm/display/drm_dp.h @@ -718,6 +718,7 @@ # define DP_PSR_SU_REGION_SCANLINE_CAPTURE BIT(4) /* eDP 1.4a */ # define DP_PSR_IRQ_HPD_WITH_CRC_ERRORS BIT(5) /* eDP 1.4a */ # define DP_PSR_ENABLE_PSR2 BIT(6) /* eDP 1.4a */ +# define DP_PSR_ENABLE_SU_REGION_ET BIT(7) /* eDP 1.5 */ #define DP_ADAPTER_CTRL 0x1a0 # define DP_ADAPTER_CTRL_FORCE_LOAD_SENSE (1 << 0) diff --git a/include/drm/drm_print.h b/include/drm/drm_print.h index 39ea79b5c0f0..9cc473e5d353 100644 --- a/include/drm/drm_print.h +++ b/include/drm/drm_print.h @@ -614,6 +614,9 @@ void __drm_err(const char *format, ...); drm_dev_printk(drm_ ? drm_->dev : NULL, KERN_DEBUG, fmt, ## __VA_ARGS__); \ }) +#define drm_dbg_ratelimited(drm, fmt, ...) \ + __DRM_DEFINE_DBG_RATELIMITED(DRIVER, drm, fmt, ## __VA_ARGS__) + #define drm_dbg_kms_ratelimited(drm, fmt, ...) \ __DRM_DEFINE_DBG_RATELIMITED(KMS, drm, fmt, ## __VA_ARGS__) diff --git a/include/drm/i915_pciids.h b/include/drm/i915_pciids.h index fcf1849aa47c..07779a11758e 100644 --- a/include/drm/i915_pciids.h +++ b/include/drm/i915_pciids.h @@ -751,10 +751,13 @@ /* MTL */ #define INTEL_MTL_IDS(info) \ INTEL_VGA_DEVICE(0x7D40, info), \ + INTEL_VGA_DEVICE(0x7D41, info), \ INTEL_VGA_DEVICE(0x7D45, info), \ + INTEL_VGA_DEVICE(0x7D51, info), \ INTEL_VGA_DEVICE(0x7D55, info), \ INTEL_VGA_DEVICE(0x7D60, info), \ INTEL_VGA_DEVICE(0x7D67, info), \ + INTEL_VGA_DEVICE(0x7DD1, info), \ INTEL_VGA_DEVICE(0x7DD5, info) #endif /* _I915_PCIIDS_H */ diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index ad21c613fec8..96e32dafd4f0 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -865,6 +865,8 @@ struct drm_amdgpu_cs_chunk_cp_gfx_shadow { #define AMDGPU_INFO_SENSOR_PEAK_PSTATE_GFX_SCLK 0xa /* Subquery id: Query GPU peak pstate memory clock */ #define AMDGPU_INFO_SENSOR_PEAK_PSTATE_GFX_MCLK 0xb + /* Subquery id: Query input GPU power */ + #define AMDGPU_INFO_SENSOR_GPU_INPUT_POWER 0xc /* Number of VRAM page faults on CPU access. */ #define AMDGPU_INFO_NUM_VRAM_CPU_PAGE_FAULTS 0x1E #define AMDGPU_INFO_VRAM_LOST_COUNTER 0x1F diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index fd4f9574d177..bd87386a8243 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -3013,6 +3013,7 @@ struct drm_i915_query_item { * - %DRM_I915_QUERY_MEMORY_REGIONS (see struct drm_i915_query_memory_regions) * - %DRM_I915_QUERY_HWCONFIG_BLOB (see `GuC HWCONFIG blob uAPI`) * - %DRM_I915_QUERY_GEOMETRY_SUBSLICES (see struct drm_i915_query_topology_info) + * - %DRM_I915_QUERY_GUC_SUBMISSION_VERSION (see struct drm_i915_query_guc_submission_version) */ __u64 query_id; #define DRM_I915_QUERY_TOPOLOGY_INFO 1 @@ -3021,6 +3022,7 @@ struct drm_i915_query_item { #define DRM_I915_QUERY_MEMORY_REGIONS 4 #define DRM_I915_QUERY_HWCONFIG_BLOB 5 #define DRM_I915_QUERY_GEOMETRY_SUBSLICES 6 +#define DRM_I915_QUERY_GUC_SUBMISSION_VERSION 7 /* Must be kept compact -- no holes and well documented */ /** @@ -3567,6 +3569,16 @@ struct drm_i915_query_memory_regions { }; /** + * struct drm_i915_query_guc_submission_version - query GuC submission interface version + */ +struct drm_i915_query_guc_submission_version { + __u32 branch; + __u32 major; + __u32 minor; + __u32 patch; +}; + +/** * DOC: GuC HWCONFIG blob uAPI * * The GuC produces a blob with information about the current device. diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h index f0ed68974c54..9ce46edc62a5 100644 --- a/include/uapi/linux/kfd_ioctl.h +++ b/include/uapi/linux/kfd_ioctl.h @@ -40,9 +40,10 @@ * - 1.12 - Add DMA buf export ioctl * - 1.13 - Add debugger API * - 1.14 - Update kfd_event_data + * - 1.15 - Enable managing mappings in compute VMs with GEM_VA ioctl */ #define KFD_IOCTL_MAJOR_VERSION 1 -#define KFD_IOCTL_MINOR_VERSION 14 +#define KFD_IOCTL_MINOR_VERSION 15 struct kfd_ioctl_get_version_args { __u32 major_version; /* from KFD */ |