summaryrefslogtreecommitdiffstats
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorLines
2025-08-04drm/amd/pm: Use cached metrics data on aldebaranLijo Lazar-1/+1
Cached metrics data validity is 1ms on aldebaran. It's not reasonable for any client to query gpu_metrics at a faster rate and constantly interrupt PMFW. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Add NULL check for asic_funcsLijo Lazar-1/+2
If driver load fails too early, asic_funcs pointer remains unassigned. Add NULL check to sanitize unwind path. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Promote DC to 3.2.344Taimur Hassan-1/+1
Summary: * Add interface to log hw state when underflow happens * Fix hubp programming of 3dlut fast load * Avoid Read Remote DPCD Many Times * More liberal vmin/vmax update for freesync * Fix dmub access race condition Acked-by: Sun peng (Leo) Li <sunpeng.li@amd.com> Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Adding interface to log hw state when underflow happensMuhammad Ahmed-7/+236
[why] Will help us better debug underflow issues. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Muhammad Ahmed <Muhammad.Ahmed@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Toggle for Disable Force Pstate Allow on DisableRyan Seto-14/+24
[Why & How] In theory, driver should be able to support disabling force pstate allow after hardware release however this behavior is not tested yet. Introducing a new toggle to disable the force on the fly. Reviewed-by: Dillon Varone <dillon.varone@amd.com> Signed-off-by: Ryan Seto <ryanseto@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Fixing hubp programming of 3dlut fast loadReza Amini-15/+148
[why] HUBP needs to know the size of the lut's destination in MPC. This is currently defaulted to 17, and needs to be set for specific lut size. [how] Define and apply the missing hubp field. Taking this opportunity to consolidate the programming of 3dlut into a hubp and mpc function. Reviewed-by: Krunoslav Kovac <krunoslav.kovac@amd.com> Signed-off-by: Reza Amini <reza.amini@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: limited pll vco w/a v2Jingwen Zhu-1/+4
[Why/How] The w/a will cause reboot black screen issue. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Jingwen Zhu <Jingwen.Zhu@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Avoid Read Remote DPCD Many TimesFangzhi Zuo-3/+6
Reading remote dpcd is time consuming. Instead of reading each byte one by one, read 16 bytes together. Reviewed-by: ChiaHsuan (Tom) Chung <chiahsuan.chung@amd.com> Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com> Signed-off-by: Wayne Lin <Wayne.Lin@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Revert "drm/amd/display: Fix AMDGPU_MAX_BL_LEVEL value"Mario Limonciello-4/+4
This reverts commit 66abb996999de0d440a02583a6e70c2c24deab45. This broke custom brightness curves but it wasn't obvious because of other related changes. Custom brightness curves are always from a 0-255 input signal. The correct fix was to fix the default value which was done by [1]. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4412 Link: https://lore.kernel.org/amd-gfx/0f094c4b-d2a3-42cd-824c-dc2858a5618d@kernel.org/T/#m69f875a7e69aa22df3370b3e3a9e69f4a61fdaf2 Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: update dpp/disp clock from smu clock tablePaul Hsieh-3/+33
[Why] The reason some high-resolution monitors fail to display properly is that this platform does not support sufficiently high DPP and DISP clock frequencies [How] Update DISP and DPP clocks from the smu clock table then DML can filter these mode if not support. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Paul Hsieh <Paul.Hsieh@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: more liberal vmin/vmax update for freesyncAurabindo Pillai-8/+20
[Why] FAMS2 expects vmin/vmax to be updated in the case when freesync is off, but supported. But we only update it when freesync is enabled. [How] Change the vsync handler such that dc_stream_adjust_vmin_vmax() its called irrespective of whether freesync is enabled. If freesync is supported, then there is no harm in updating vmin/vmax registers. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3546 Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Ray Wu <ray.wu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Reviewed-by: ChiaHsuan Chung <chiahsuan.chung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: fix dmub access race conditionAurabindo Pillai-6/+63
Accessing DC from amdgpu_dm is usually preceded by acquisition of dc_lock mutex. Most of the DC API that DM calls are under a DC lock. However, there are a few that are not. Some DC API called from interrupt context end up sending DMUB commands via a DC API, while other threads were using DMUB. This was apparent from a race between calls for setting idle optimization enable/disable and the DC API to set vmin/vmax. Offload the call to dc_stream_adjust_vmin_vmax() to a thread instead of directly calling them from the interrupt handler such that it waits for dc_lock. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Adjust AUX-less ALPM settingDuncan Ma-1/+1
[Why & How] Change ACDS period to support LTTPR. Reviewed-by: Charlene Liu <charlene.liu@amd.com> Signed-off-by: Duncan Ma <Duncan.Ma@amd.com> Signed-off-by: Roman Li <roman.li@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: fix a Null pointer dereference vulnerabilitySiyang Liu-9/+10
[Why] A null pointer dereference vulnerability exists in the AMD display driver's (DC module) cleanup function dc_destruct(). When display control context (dc->ctx) construction fails (due to memory allocation failure), this pointer remains NULL. During subsequent error handling when dc_destruct() is called, there's no NULL check before dereferencing the perf_trace member (dc->ctx->perf_trace), causing a kernel null pointer dereference crash. [How] Check if dc->ctx is non-NULL before dereferencing. Link: https://lore.kernel.org/r/tencent_54FF4252EDFB6533090A491A25EEF3EDBF06@qq.com Co-developed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> (Updated commit text and removed unnecessary error message) Signed-off-by: Siyang Liu <Security@tencent.com> Signed-off-by: Roman Li <roman.li@amd.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Initialize vcn v5_0_1 ras functionMangesh Gadre-0/+8
Initialize vcn v5_0_1 ras function Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com> Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Add primary plane to commits for correct VRR handlingMichel Dänzer-0/+9
amdgpu_dm_commit_planes calls update_freesync_state_on_stream only for the primary plane. If a commit affects a CRTC but not its primary plane, it would previously not trigger a refresh cycle or affect LFC, violating current UAPI semantics. Fixes e.g. atomic commits affecting only the cursor plane being limited to the minimum refresh rate. Don't do this for the legacy cursor ioctls though, it would break the UAPI semantics for those. Suggested-by: Xaver Hugl <xaver.hugl@kde.org> Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3034 Signed-off-by: Michel Dänzer <mdaenzer@redhat.com> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: skip mgpu fan boost for multi-vfYunxiang Li-1/+1
On multi-vf setup if the VM have two vf assigned, perhaps from two different gpus, mgpu fan boost will fail. Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Initialize jpeg v5_0_1 ras functionMangesh Gadre-0/+8
Initialize jpeg v5_0_1 ras function Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com> Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Skip poison aca bank from UE channelXiang Liu-21/+30
Avoid GFX poison consumption errors logged when fatal error occurs. Signed-off-by: Xiang Liu <xiang.liu@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: fix link error for !PM_SLEEPArnd Bergmann-10/+10
When power management is not enabled in the kernel build, the newly added hibernation changes cause a link failure: arm-linux-gnueabi-ld: drivers/gpu/drm/amd/amdgpu/amdgpu_drv.o: in function `amdgpu_pmops_thaw': amdgpu_drv.c:(.text+0x1514): undefined reference to `pm_hibernate_is_recovering' Make the power management code in this driver conditional on CONFIG_PM and CONFIG_PM_SLEEP Fixes: 530694f54dd5 ("drm/amdgpu: do not resume device in thaw for normal hibernation") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20250714081635.4071570-1-arnd@kernel.org Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: add more cyan skillfish devicesAlex Deucher-1/+12
Add PCI IDs to support display probe for cyan skillfish family of SOCs. Acked-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: update mmhub 3.3 client id mappingsAlex Deucher-1/+104
Update the client id mapping so the correct clients get printed when there is a mmhub page fault. v2: fix typos spotted by David Wu. v3: fix additional typo spotted by David. Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: update mmhub 3.0.1 client id mappingsAlex Deucher-25/+32
Update the client id mapping so the correct clients get printed when there is a mmhub page fault. Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN2_5Sathishkumar S-76/+7
Use generic vcn devcoredump helper functions for VCN2_5 and VCN2_6 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN2_0_0Sathishkumar S-74/+5
Use generic vcn devcoredump helper functions for VCN2_0_0 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN3_0Sathishkumar S-73/+5
Use generic vcn devcoredump helper functions for VCN3_0 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN4_0_3Sathishkumar S-74/+5
Use generic vcn devcoredump helper functions for VCN4_0_3 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN4_0_5Sathishkumar S-75/+5
Use generic vcn devcoredump helper functions for VCN4_0_5 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN4_0_0Sathishkumar S-75/+5
Use generic vcn devcoredump helper functions for VCN4_0_0 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Register dump cleanup in VCN5Sathishkumar S-91/+44
Use generic vcn devcoredump helper functions for VCN5 Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Add new error code for VCN/JPEG new chainStanley.Yang-3/+4
Add VIDS and JPEG8/9 S|D chain error code for VCN/JPEG v5.0.1. Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Fix vcn v5.0.1 poison irq call traceStanley.Yang-5/+12
Why: [13014.890792] Call Trace: [13014.890793] <TASK> [13014.890795] ? show_trace_log_lvl+0x1d6/0x2ea [13014.890799] ? show_trace_log_lvl+0x1d6/0x2ea [13014.890800] ? vcn_v5_0_1_hw_fini+0xe9/0x110 [amdgpu] [13014.890872] ? show_regs.part.0+0x23/0x29 [13014.890873] ? show_regs.cold+0x8/0xd [13014.890874] ? amdgpu_irq_put+0xc6/0xe0 [amdgpu] [13014.890934] ? __warn+0x8c/0x100 [13014.890936] ? amdgpu_irq_put+0xc6/0xe0 [amdgpu] [13014.890995] ? report_bug+0xa4/0xd0 [13014.890999] ? handle_bug+0x39/0x90 [13014.891001] ? exc_invalid_op+0x19/0x70 [13014.891003] ? asm_exc_invalid_op+0x1b/0x20 [13014.891005] ? amdgpu_irq_put+0xc6/0xe0 [amdgpu] [13014.891065] ? amdgpu_irq_put+0x63/0xe0 [amdgpu] [13014.891124] vcn_v5_0_1_hw_fini+0xe9/0x110 [amdgpu] [13014.891189] amdgpu_ip_block_hw_fini+0x3b/0x78 [amdgpu] [13014.891309] amdgpu_device_fini_hw+0x3c1/0x479 [amdgpu] How: Add omitted vcn poison irq get call. Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu/vcn: Add regdump helper functionsSathishkumar S-0/+95
Add generic helper functions for vcn devcoredump support which can be re-used for all vcn versions. Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Acked-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/amdgpu: Release xcp drm memory after unplugMeng Li-9/+49
Add a new API amdgpu_xcp_drm_dev_free(). After unplug xcp device, need to release xcp drm memory etc. Co-developed-by: Jiang Liu <gerry@linux.alibaba.com> Signed-off-by: Jiang Liu <gerry@linux.alibaba.com> Signed-off-by: Meng Li <li.meng@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Retain job->vm in amdgpu_job_prepare_jobYuanShang-7/+0
The field job->vm is used in function amdgpu_job_run to get the page table re-generation counter and decide whether the job should be skipped. Specifically, function amdgpu_vm_generation checks if the VM is valid for this job to use. For instance, if a gfx job depends on a cancelled sdma job from entity vm->delayed, then the gfx job should be skipped. Fixes: 26c95e838e63 ("drm/amdgpu: set the VM pointer to NULL in amdgpu_job_prepare") Signed-off-by: YuanShang <YuanShang.Mao@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd: Use drm_*() macros instead of DRM_*() for amdgpu_csMario Limonciello-10/+10
Some of the IOCTL messages can be called for different GPUs and it might not be obvious which one called them from a problem. Using the drm_*() macros the correct device will be shown in the messages. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Link: https://lore.kernel.org/r/20250715212420.2254925-1-superm1@kernel.org Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: use kmalloc_array() instead of kmalloc()Yunshui Jiang-2/+2
Use kmalloc_array() instead of kmalloc() with multiplication. kmalloc_array() is a safer way because of its multiply overflow check. Signed-off-by: Yunshui Jiang <jiangyunshui@kylinos.cn> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Fix unintended error log in VCN5_0_0Sathishkumar S-2/+3
The error log is supposed to be gaurded under if failure condition. Fixes: faab5ea08367 ("drm/amdgpu: Check vcn sram load return value") Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Fix DCE 6.0 and 6.4 PLL programming.Timur Kristóf-14/+25
Apparently, both DCE 6.0 and 6.4 have 3 PLLs, but PLL0 can only be used for DP. Make sure to initialize the correct amount of PLLs in DC for these DCE versions and use PLL0 only for DP. Also, on DCE 6.0 and 6.4, the PLL0 needs to be powered on at initialization as opposed to DCE 6.1 and 7.x which use a different clock source for DFS. The following functions were used as reference from the old radeon driver implementation of DCE 6.x: - radeon_atom_pick_pll - atombios_crtc_set_disp_eng_pll Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd/display: Don't overwrite dce60_clk_mgrTimur Kristóf-1/+0
dc_clk_mgr_create accidentally overwrites the dce60_clk_mgr with the dce_clk_mgr, causing incorrect behaviour on DCE6. Fix it by removing the extra dce_clk_mgr_construct. Fixes: 62eab49faae7 ("drm/amd/display: hide VGH asic specific structs") Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Effective health check before resetCe Sun-18/+8
Move amdgpu_device_health_check into amdgpu_device_gpu_recover to ensure that if the device is present can be checked before reset The reason is: 1.During the dpc event, the device where the dpc event occurs is not present on the bus 2.When both dpc event and ATHUB event occur simultaneously,the dpc thread holds the reset domain lock when detecting error,and the gpu recover thread acquires the hive lock.The device is simultaneously in the states of amdgpu_ras_in_recovery and occurs_dpc,so gpu recover thread will not go to amdgpu_device_health_check.It waits for the reset domain lock held by the dpc thread, but dpc thread has not released the reset domain lock.In the dpc callback slot_reset,to obtain the hive lock, the hive lock is held by the gpu recover thread at this time.So a deadlock occurred Signed-off-by: Ce Sun <cesun102@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Avoid rma causes GPU duplicate resetCe Sun-7/+11
Try to ensure poison creation handle is completed in time to set device rma value. Signed-off-by: Ce Sun <cesun102@amd.com> Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Update IPID value for bad page threshold CPERXiang Liu-2/+6
Update the IPID register value for bad page threshold CPER according to the latest definition. Signed-off-by: Xiang Liu <xiang.liu@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Fix kdoc style in amdgpu_fence.cSrinivasan Shanmugam-1/+1
The initial comment block before amdgpu_fence_driver_guilty_force_completion() incorrectly used '/**' but is not a kernel-doc comment, causing build warnings. Fixes the below with gcc W=1: drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c:742: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst * Kernel queue reset handling Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdkfd: Fix checkpoint-restore on multi-xccDavid Yat Sin-16/+67
GPUs with multi-xcc have multiple MQDs per queue. This patch saves and restores all the MQDs within the partition. Signed-off-by: David Yat Sin <David.YatSin@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd: Restore cached manual clock settings during resumeMario Limonciello-0/+10
If the SCLK limits have been set before S3 they will not be restored. The limits are however cached in the driver and so they can be restored by running a commit sequence during resume. Acked-by: Alex Deucher <alexander.deucher@amd.com> Link: https://lore.kernel.org/r/20250725031222.3015095-3-superm1@kernel.org Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amd: Restore cached power limit during resumeMario Limonciello-0/+6
The power limit will be cached in smu->current_power_limit but if the ASIC goes into S3 this value won't be restored. Restore the value during SMU resume. Acked-by: Alex Deucher <alexander.deucher@amd.com> Link: https://lore.kernel.org/r/20250725031222.3015095-2-superm1@kernel.org Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: Fix build error when CONFIG_SUSPEND is disabledPerry Yuan-0/+4
The variable `pm_suspend_target_state` is conditionally defined only when `CONFIG_SUSPEND` is enabled (see `include/linux/suspend.h`). Directly referencing it without guarding by `#ifdef CONFIG_SUSPEND` causes build failures when suspend functionality is disabled (e.g., `CONFIG_SUSPEND=n`). Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Perry Yuan <perry.yuan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04drm/amdgpu: rework how PTE flags are generated v3Christian König-186/+158
Previously we tried to keep the HW specific PTE flags in each mapping, but for CRIU that isn't sufficient any more since the original value is needed for the checkpoint procedure. So rework the whole handling, nuke the early mapping function, keep the UAPI flags in each mapping instead of the HW flags and translate them to the HW flags while filling in the PTEs. Only tested on Navi 23 for now, so probably needs quite a bit of more work. v2: fix KFD and SVN handling v3: one more SVN fix pointed out by Felix v4: squash in gfx12 fix from David Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-04Mark xe driver as BROKEN if kernel page size is not 4kBSimon Richter-0/+1
This driver, for the time being, assumes that the kernel page size is 4kB, so it fails on loong64 and aarch64 with 16kB pages, and ppc64el with 64kB pages. Signed-off-by: Simon Richter <Simon.Richter@hogyros.de> Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Cc: stable@vger.kernel.org # v6.8+ Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> Link: https://lore.kernel.org/r/20250802024152.3021-1-Simon.Richter@hogyros.de (cherry picked from commit 0521a868222ffe636bf202b6e9d29292c1e19c62) Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
generated by cgit v1.2.3 (git 2.46.0) at 2026-05-03 11:59:08 +0000