summaryrefslogtreecommitdiffstats
path: root/drivers
AgeCommit message (Collapse)AuthorLines
2026-03-23drm/amd/pm: Skip redundant UCLK restore in smu_v13_0_6Asad Kamal-8/+11
Only reapply UCLK soft limits during PP_OD_RESTORE_DEFAULT when the current max differs from the DPM table max. This avoids redundant SMC updates and prevents -EINVAL on restore when no change is needed. Fixes: b7a900344546 ("drm/amd/pm: Allow setting max UCLK on SMU v13.0.6") Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 17f11bbbc76c8e83c8474ea708316b1e3631d927)
2026-03-23drm/amd/display: Fix drm_edid leak in amdgpu_dmAlex Hung-1/+2
[WHAT] When a sink is connected, aconnector->drm_edid was overwritten without freeing the previous allocation, causing a memory leak on resume. [HOW] Free the previous drm_edid before updating it. Reviewed-by: Roman Li <roman.li@amd.com> Signed-off-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 52024a94e7111366141cfc5d888b2ef011f879e5) Cc: stable@vger.kernel.org
2026-03-23drm/amdgpu: prevent immediate PASID reuse caseEric Huang-13/+34
PASID resue could cause interrupt issue when process immediately runs into hw state left by previous process exited with the same PASID, it's possible that page faults are still pending in the IH ring buffer when the process exits and frees up its PASID. To prevent the case, it uses idr cyclic allocator same as kernel pid's. Signed-off-by: Eric Huang <jinhuieric.huang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 8f1de51f49be692de137c8525106e0fce2d1912d) Cc: stable@vger.kernel.org
2026-03-23drm/amdgpu: fix strsep() corrupting lockup_timeout on multi-GPU (v3)Ruijing Dong-2/+11
amdgpu_device_get_job_timeout_settings() passes a pointer directly to the global amdgpu_lockup_timeout[] buffer into strsep(). strsep() destructively replaces delimiter characters with '\0' in-place. On multi-GPU systems, this function is called once per device. When a multi-value setting like "0,0,0,-1" is used, the first GPU's call transforms the global buffer into "0\00\00\0-1". The second GPU then sees only "0" (terminated at the first '\0'), parses a single value, hits the single-value fallthrough (index == 1), and applies timeout=0 to all rings — causing immediate false job timeouts. Fix this by copying into a stack-local array before calling strsep(), so the global module parameter buffer remains intact across calls. The buffer is AMDGPU_MAX_TIMEOUT_PARAM_LENGTH (256) bytes, which is safe for the stack. v2: wrap commit message to 72 columns, add Assisted-by tag. v3: use stack array with strscpy() instead of kstrdup()/kfree() to avoid unnecessary heap allocation (Christian). This patch was developed with assistance from Claude (claude-opus-4-6). Assisted-by: Claude:claude-opus-4-6 Reviewed-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Ruijing Dong <ruijing.dong@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 94d79f51efecb74be1d88dde66bdc8bfcca17935) Cc: stable@vger.kernel.org
2026-03-23drm/amd/display: Do not skip unrelated mode changes in DSC validationYussuf Khalil-1/+9
Starting with commit 17ce8a6907f7 ("drm/amd/display: Add dsc pre-validation in atomic check"), amdgpu resets the CRTC state mode_changed flag to false when recomputing the DSC configuration results in no timing change for a particular stream. However, this is incorrect in scenarios where a change in MST/DSC configuration happens in the same KMS commit as another (unrelated) mode change. For example, the integrated panel of a laptop may be configured differently (e.g., HDR enabled/disabled) depending on whether external screens are attached. In this case, plugging in external DP-MST screens may result in the mode_changed flag being dropped incorrectly for the integrated panel if its DSC configuration did not change during precomputation in pre_validate_dsc(). At this point, however, dm_update_crtc_state() has already created new streams for CRTCs with DSC-independent mode changes. In turn, amdgpu_dm_commit_streams() will never release the old stream, resulting in a memory leak. amdgpu_dm_atomic_commit_tail() will never acquire a reference to the new stream either, which manifests as a use-after-free when the stream gets disabled later on: BUG: KASAN: use-after-free in dc_stream_release+0x25/0x90 [amdgpu] Write of size 4 at addr ffff88813d836524 by task kworker/9:9/29977 Workqueue: events drm_mode_rmfb_work_fn Call Trace: <TASK> dump_stack_lvl+0x6e/0xa0 print_address_description.constprop.0+0x88/0x320 ? dc_stream_release+0x25/0x90 [amdgpu] print_report+0xfc/0x1ff ? srso_alias_return_thunk+0x5/0xfbef5 ? __virt_addr_valid+0x225/0x4e0 ? dc_stream_release+0x25/0x90 [amdgpu] kasan_report+0xe1/0x180 ? dc_stream_release+0x25/0x90 [amdgpu] kasan_check_range+0x125/0x200 dc_stream_release+0x25/0x90 [amdgpu] dc_state_destruct+0x14d/0x5c0 [amdgpu] dc_state_release.part.0+0x4e/0x130 [amdgpu] dm_atomic_destroy_state+0x3f/0x70 [amdgpu] drm_atomic_state_default_clear+0x8ee/0xf30 ? drm_mode_object_put.part.0+0xb1/0x130 __drm_atomic_state_free+0x15c/0x2d0 atomic_remove_fb+0x67e/0x980 Since there is no reliable way of figuring out whether a CRTC has unrelated mode changes pending at the time of DSC validation, remember the value of the mode_changed flag from before the point where a CRTC was marked as potentially affected by a change in DSC configuration. Reset the mode_changed flag to this earlier value instead in pre_validate_dsc(). Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/5004 Fixes: 17ce8a6907f7 ("drm/amd/display: Add dsc pre-validation in atomic check") Signed-off-by: Yussuf Khalil <dev@pp3345.net> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit cc7c7121ae082b7b82891baa7280f1ff2608f22b)
2026-03-23spi: meson-spicc: Fix double-put in remove pathFelix Gu-2/+0
meson_spicc_probe() registers the controller with devm_spi_register_controller(), so teardown already drops the controller reference via devm cleanup. Calling spi_controller_put() again in meson_spicc_remove() causes a double-put. Fixes: 8311ee2164c5 ("spi: meson-spicc: fix memory leak in meson_spicc_remove") Signed-off-by: Felix Gu <ustc.gu@gmail.com> Reviewed-by: Johan Hovold <johan@kernel.org> Link: https://patch.msgid.link/20260322-rockchip-v1-1-fac3f0c6dad8@gmail.com Signed-off-by: Mark Brown <broonie@kernel.org>
2026-03-23drm/amd/pm: disable OD_FAN_CURVE if temp or pwm range invalid for smu v13Yang Wang-2/+64
Forcibly disable the OD_FAN_CURVE feature when temperature or PWM range is invalid, otherwise PMFW will reject this configuration on smu v13.0.x example: $ sudo cat /sys/bus/pci/devices/<BDF>/gpu_od/fan_ctrl/fan_curve OD_FAN_CURVE: 0: 0C 0% 1: 0C 0% 2: 0C 0% 3: 0C 0% 4: 0C 0% OD_RANGE: FAN_CURVE(hotspot temp): 0C 0C FAN_CURVE(fan speed): 0% 0% $ echo "0 50 40" | sudo tee fan_curve kernel log: [ 756.442527] amdgpu 0000:03:00.0: amdgpu: Fan curve temp setting(50) must be within [0, 0]! [ 777.345800] amdgpu 0000:03:00.0: amdgpu: Fan curve temp setting(50) must be within [0, 0]! Closes: https://github.com/ROCm/amdgpu/issues/208 Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amdgpu: Use stack variable to fetch nps infoLijo Lazar-17/+16
Instead of a dynamic allocation, use stack variable and let the caller pass the maximum ranges that can be held in the buffer. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Rename enum 'pixel_format' to 'dc_pixel_format'Hou Wenlong-10/+11
Rename the enum 'pixel_format' to 'dc_pixel_format' to avoid potential name conflicts with the pixel_format struct defined in include/video/pixel_format.h. Signed-off-by: Hou Wenlong <houwenlong.hwl@antgroup.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Enable user specified gfx clock rangesAsad Kamal-0/+1
Enable user specified gfx clock ranges for smu_15_0_8 Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amdgpu: Add smu v15_0_8 ip blockHawking Zhang-0/+4
Add smu v15_0_8 ip block Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Likun Gao <Likun.Gao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add NPM support for smu_v15_0_8Asad Kamal-0/+49
Add node power management support for smu_v15_0_8 Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add baseboard temperature metrics supportAsad Kamal-13/+184
Add baseboard temperature metrics support via system metrics table for smu_v15_0_8 v4: Add separate function to fill baseboard temperature, use 16, remove casting v5: Optimize to use single switch case (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add gpuboard temperature metrics supportAsad Kamal-0/+278
Add gpuboard temperature metrics support via system metrics table for smu_v15_0_8 v3: Use per sensor attr id (Lijo) v4: Use s16 for temp, remove cast, use separate function to fill gpuboard temperature metrics data (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add read sensor supportAsad Kamal-0/+112
Add read sensor support for smu_v15_0_8 v2: Remove gfx voltage support (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add ppt1 supportAsad Kamal-1/+74
Add ppt1 support for smu_v15_0_8 Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add get_thermal_temperature_range supportAsad Kamal-0/+33
Add get_thermal_temperature_range support smu_v15_0_8 v2: Remove sriov check (Lijo) v3: Restrict to 1VF mode(Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add od_edit_dpm_table supportAsad Kamal-1/+306
Add od_edit_dpm_table support for smu_v15_0_8 v2: Skip Gl2clk/Fclk (Lijo) v3: sqaush in set_performance_support (Asad) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: add populate_umd_state_clk supportAsad Kamal-1/+17
add populate_umd_state_clk support for smu 15.0.8 v2: remove gl2clk/socclk/fclk, restrict to only current min/max (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Fix NULL pointer assumptions in dcn42_init_hw()Srinivasan Shanmugam-7/+17
dcn42_init_hw() calls update_bw_bounding_box() when FAMS2 is disabled or when the dchub reference clock changes. However the existing condition mixes the callback pointer check with only one side of the || expression: ((!fams2_enable && update_bw_bounding_box) || freq_changed) This allows the block to be entered through the freq_changed path even when update_bw_bounding_box() is NULL. The function is then called unconditionally inside the block, which can lead to a NULL pointer dereference. Additionally, the code dereferences dc->clk_mgr->bw_params without verifying that dc->clk_mgr and bw_params are valid. Restructure the condition so that the update trigger remains the same (FAMS2 disabled or dchub ref clock changed), but guard the call with explicit checks for: - update_bw_bounding_box callback - dc->clk_mgr - dc->clk_mgr->bw_params Also introduce a helper boolean (dchub_ref_freq_changed) to improve readability of the clock-change condition. This fixes Smatch warnings about inconsistent NULL assumptions in dcn42_init_hw(). drivers/gpu/drm/amd/amdgpu/../display/dc/hwss/dcn42/dcn42_hwseq.c:264 dcn42_init_hw() error: we previously assumed 'dc->clk_mgr' could be null (see line 253) drivers/gpu/drm/amd/amdgpu/../display/dc/hwss/dcn42/dcn42_hwseq.c:278 dcn42_init_hw() error: we previously assumed 'dc->res_pool->funcs->update_bw_bounding_box' could be null (see line 274) Cc: Roman Li <roman.li@amd.com> Cc: Alex Hung <alex.hung@amd.com> Cc: Jerry Zuo <jerry.zuo@amd.com> Cc: Sun peng Li <sunpeng.li@amd.com> Cc: Tom Chung <chiahsuan.chung@amd.com> Cc: Dan Carpenter <dan.carpenter@linaro.org> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Add clk_mgr NULL checks in dcn32_initialize_min_clocks()Srinivasan Shanmugam-6/+11
dcn32_init_hw() checks dc->clk_mgr before calling init_clocks(), so the clock manager is not treated as unconditionally present on this path. However, dcn32_initialize_min_clocks() later dereferences dc->clk_mgr, bw_params, and clk_mgr callbacks without validating them. Add the required guards in dcn32_initialize_min_clocks() before accessing clk_mgr-dependent state, and check callback presence before calling get_dispclk_from_dentist() and update_clocks(). Also guard the later update_bw_bounding_box() call in the FAMS2-disabled path since it also dereferences dc->clk_mgr->bw_params. This keeps clk_mgr handling consistent in the DCN32 HW init flow and avoids possible NULL pointer dereferences reported by Smatch. Fixes the below: drivers/gpu/drm/amd/amdgpu/../display/dc/hwss/dcn32/dcn32_hwseq.c:1012 dcn32_init_hw() error: we previously assumed 'dc->clk_mgr' could be null (see line 978) Cc: Roman Li <roman.li@amd.com> Cc: Alex Hung <alex.hung@amd.com> Cc: Jerry Zuo <jerry.zuo@amd.com> Cc: Sun peng Li <sunpeng.li@amd.com> Cc: Tom Chung <chiahsuan.chung@amd.com> Cc: Dan Carpenter <dan.carpenter@linaro.org> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add emit clock supportAsad Kamal-0/+198
Add emit clock support and fetching other metrics data like temperature, clock for smu_v15_0_8 v2: Use umc count for hbm stack temperature (Lijo) v3: Use correct logic for hbm stacks (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: add set{get}_power_limit support for smu 15.0.8Yang Wang-0/+35
export .set_power_limit & .get_power_limit interface for smu 15.0.8 Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: add get_unique_id support for smu 15.0.8Yang Wang-0/+10
export .get_unique_id interface for smu 15.0.8 Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: add get_gpu_metrics support for 15.0.8Yang Wang-4/+303
export .get_gpu_metrics interface for 15.0.8 v2: Remove members already exposed by other interfaces, use mask, logical conversion (Lijo) v3: Use correct logic for hbm stacks loop (Lijo) Remove buffer allocation v4: Make out of bound check outside loop (Lijo) v5: fix locking in error case (Alex) Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add get_pm_metrics support for smu 15.0.8Asad Kamal-0/+65
export .get_pm_metrics interface for smu 15.0.8. v2: Make tmo as unsigned (Lijo) Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add default dpm table support for smu 15.0.8Asad Kamal-0/+213
Add default dpm table support for smu 15.0.8 v2: Remove lclk, move pptable check up, add missing clk (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Update dpm table structs for smu_v15_0Asad Kamal-77/+49
Update dpm table structs to use common definitions for smu_15_0 Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Setup driver pptable for smu 15.0.8Yang Wang-5/+257
Setup driver pptable and initialize data from static metrics table for smu_v15_0_8 v2: Remove unrelated changes and update description (Lijo) v3: Use ARRAY_SIZE (Lijo) v4: Move structure to header file v5: squash in static metrics support (Asad) Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add mode2 support for smu_v15_0_8Asad Kamal-0/+44
Add initial mode2 support for smu_v15_0_8 v2: Move out non smu code, remove pci save/restore logic (Lijo) v3: squash in updated msg (Alex) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add initial support for smu v15_0_8Hawking Zhang-52/+564
smu v15_0_8 is the new generation of smu ip block v2: Squash in rebase changes (Alex) v3: Squash in fw version check changes (Alex) Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Likun Gao <Likun.Gao@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amdgpu/userq: cleanup amdgpu_userq_get/put where not neededSunil Khatri-18/+3
amdgpu_userq_put/get are not needed in case we already holding the userq_mutex and reference is valid already from queue create time or from signal ioctl. These additional get/put could be a potential reason for deadlock in case the ref count reaches zero and destroy is called which again try to take the userq_mutex. Due to the above change we avoid deadlock between suspend/restore calling destroy queues trying to take userq_mutex again. Cc: Prike Liang <Prike.Liang@amd.com> Signed-off-by: Sunil Khatri <sunil.khatri@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add custom fclk setting supportAsad Kamal-1/+83
Add custom fclk setting support for smu_v13_x_x v2: Move uclk fix to separate patch, return EOPNOTSUPP in case of dpm disabled (Lijo) v3: remove dpm check for filling fclk pstate table (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Add OD_FCLK interfaceAsad Kamal-7/+16
Add OD_FCLK interface to set customa fclk max v2: Merge patch1 & 3, check EOPNOTSUPP for all clks (Lijo) v3: Remove redundant check (Lijo) Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Return -EOPNOTSUPP for unsupported OD_MCLK on smu_v13_0_6Asad Kamal-1/+1
When SET_UCLK_MAX capability is absent, return -EOPNOTSUPP from smu_v13_0_6_emit_clk_levels() for OD_MCLK instead of 0. This makes unsupported OD_MCLK reporting consistent with other clock types and allows callers to skip the entry cleanly. Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/pm: Skip redundant UCLK restore in smu_v13_0_6Asad Kamal-8/+11
Only reapply UCLK soft limits during PP_OD_RESTORE_DEFAULT when the current max differs from the DPM table max. This avoids redundant SMC updates and prevents -EINVAL on restore when no change is needed. Fixes: b7a900344546 ("drm/amd/pm: Allow setting max UCLK on SMU v13.0.6") Signed-off-by: Asad Kamal <asad.kamal@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amdgpu: Add amdgpu_regs_pcie64 debugfs nodeStanley.Yang-0/+112
Add amdgpu_regs_pcie64 debugfs node to read/write 64bit PCIE registers. Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Promote DC to 3.2.375Taimur Hassan-1/+1
This version brings along following fixes: - Rework YCbCr422 DSC policy - Restore full update for tiling change to linear - add dccg FGCG mask init - Remove unnecessary completion flag for secure display - Agument live + capture with CVT case. - remove dc_clock_limit for apu - Fix Signed/Unsigned Int Usage Compiler Warning - Hardcode dtbclk value in bw_params - Revert inbox0 lock for cursor due to deadlock - Add 3DLUT DMA broadcast support - Fix Silence warnings - export get_power_profile interface for later use - pg cntl update based on previous asic. - remove disable_sutter touch pstate debug code - Refactor DC update checks - Fix drm_edid leak in amdgpu_dm - Add Extra SMU Log for dtbclk - Clamp min DS DCFCLK value to DCN limit - Update dpia supported configuration - Multiple DCN42 updates Acked-by: ChiaHsuan Chung <ChiaHsuan.Chung@amd.com> Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Rework YCbCr422 DSC policyRelja Vojvodic-20/+25
- Reworked YCbCr4:2:2 Native/Simple policy decision making with DSC enabled based on DSC caps and stream signal type Reviewed-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Relja Vojvodic <Relja.Vojvodic@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Restore full update for tiling change to linearJoshua Aberback-21/+9
[Why] There was previously a dc debug flag to indicate that tiling changes should only be a medium update instead of full. The function get_plane_info_type was refactored to not rely on dc state, but in the process the logic was unintentionally changed, which leads to screen corruption in some cases. [How] - add flag to tiling struct to avoid full update when necessary Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Joshua Aberback <joshua.aberback@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: add dccg FGCG mask initCharlene Liu-0/+1
[why] missing DCCG_GLOBAL_FGCG_REP_DIS mask macro init Reviewed-by: Swapnil Patel <swapnil.patel@amd.com> Signed-off-by: Charlene Liu <Charlene.Liu@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Remove unnecessary completion flag for secure displayWayne Lin-8/+0
The completion flag is not used in secure display today. Remove unnecessary code. Reviewed-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Wayne Lin <Wayne.Lin@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Agument live + capture with CVT case.ChunTao Tso-2/+7
1. Add LIVE_CAPTURE_WITH_CVT bit (bit[2]) in union replay_optimization to control this feature via DalRegKey_ReplayOptimization. 2. Check the bit in mod_power_set_live_capture_with_cvt_activate function before enabling live capture with CVT. 3. Use LIVE_CAPTURE_WITH_CVT to control if Replay want to send CVT in live + capture or not. Reviewed-by: Leon Huang <leon.huang1@amd.com> Signed-off-by: ChunTao Tso <chuntao.tso@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: remove dc_clock_limit for apuCharlene Liu-1/+1
[why] current apu pmfw does not support dc_clock_limit Reviewed-by: Roman Li <roman.li@amd.com> Signed-off-by: Charlene Liu <Charlene.Liu@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Fix Signed/Unsigned Int Usage Compiler WarningGaghik Khachatrian-1/+1
[Why] Compiler generates compiler warnings when signed enum constants or literal -1 are implicitly converted to unsigned integer types, cluttering build output and masking genuine issues. [How] Use UINT_MAX as the invalid sentinel for unsigned IDs and align loop/index types to unsigned where appropriate to remove implicit signed-to-unsigned conversions, with no functional behavior change. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Gaghik Khachatrian <gaghik.khachatrian@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Hardcode dtbclk value in bw_paramsMatthew Stewart-5/+2
[why&how] dtbclk should always be 600MHz. Previous logic was to get the real value from SMU, but this returns 0 when dtbclk is off. Not a problem during boot when pre-OS enables dtbclk, but PnP was broken due to this. Reviewed-by: Charlene Liu <charlene.liu@amd.com> Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Matthew Stewart <Matthew.Stewart2@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Move DPM clk read to clk_mgr_construct in DCN42Ivan Lipski-0/+5
[Why&How] The DPM clocks on DCN42 are currently read on every dm_resume, which can cause in gpu memory freeing while the device is still in suspend. Move the DPM clock read functionality to clk_mgr_construct() so it completes once on driver enablement. Reviewed-by: Charlene Liu <charlene.liu@amd.com> Signed-off-by: Ivan Lipski <ivan.lipski@amd.com> Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Revert inbox0 lock for cursor due to deadlockNicholas Kazlauskas-26/+4
[Why] A deadlock occurs when using inbox0 lock for cursor operations on PSR-SU and Replays that does not when using the inbox1 locking path. This is because of a priority inversion issue where inbox1 work cannot be serviced while holding the HW lock from driver and sending cursor notifications to DMUB. Typically the lower priority of inbox1 for the lock command would allow the PSR and Replay FSMs to complete their transition prior to giving driver the lock but this is no longer the case with inbox0 having the highest priority in servicing. [How] This will reintroduce any synchronization bugs that were there with Replay or PSR-SU touching the cursor at the same time as driver. Reviewed-by: Charlene Liu <charlene.liu@amd.com> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Add 3DLUT DMA broadcast supportDillon Varone-48/+67
[WHY&HOW] A single HUBP can be used to fetch 3DLUT and broadcast to a single HUBP. Add logic to select the top pipe for a given plane and use it's HUBP as the broadcast source for multiple MPC's. Reviewed-by: Ilya Bakoulin <ilya.bakoulin@amd.com> Signed-off-by: Dillon Varone <Dillon.Varone@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-03-23drm/amd/display: Fix Silence warningsGaghik Khachatrian-0/+9
Also affects: freesync, hdcp, info_packet, power [Why] Resolve compiler warnings by marking unused parameters explicitly. [How] In .c/.h keep parameter names in signatures and add a line with `(void)param;` inside the function body Preserved function signatures and avoids breaking code paths that may reference the parameter under conditional compilation. Reviewed-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Gaghik Khachatrian <gaghik.khachatrian@amd.com> Signed-off-by: Chuanyu Tseng <chuanyu.tseng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit v1.2.3 (git 2.46.0) at 2026-07-07 04:18:52 +0000