aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd (follow)
AgeCommit message (Collapse)AuthorFilesLines
2025-08-27drm/amdgpu/gfx12: set MQD as appriopriate for queue typesAlex Deucher1-2/+6
Set the MQD as appropriate for the kernel vs user queues. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 7b9110f2897957efd9715b52fc01986509729db3) Cc: stable@vger.kernel.org
2025-08-27drm/amdgpu/gfx11: set MQD as appriopriate for queue typesAlex Deucher1-2/+6
Set the MQD as appropriate for the kernel vs user queues. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 063d6683208722b1875f888a45084e3d112701ac) Cc: stable@vger.kernel.org
2025-08-27drm/amdgpu/userq: fix error handling of invalid doorbellAlex Deucher1-0/+1
If the doorbell is invalid, be sure to set the r to an error state so the function returns an error. Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: update firmware version checks for user queue supportJesse.Zhang1-3/+3
The minimum firmware versions required for user queue functionality have been increased to address an issue where the queue privilege state was lost during queue connect operations. The problem occurred because the privilege state was being restored to its initial value at the beginning of the function, overwriting the state that was properly set during the queue connect case. This commit updates the minimum version requirements: - ME firmware from 2390 to 2420 - PFP firmware from 2530 to 2580 - MEC firmware from 2600 to 2650 - MES firmware remains at 120 These updated firmware versions contain the necessary fixes to properly maintain queue privilege state throughout connect operations. Fixes: 61ca97e9590c ("drm/amdgpu: Add fw minimum version check for usermode queue") Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/amdgpu: disable hwmon power1_cap* for gfx 11.0.3 on vf modeYang Wang1-8/+10
the PPSMC_MSG_GetPptLimit msg is not valid for gfx 11.0.3 on vf mode, so skiped to create power1_cap* hwmon sysfs node. Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Asad Kamal <asad.kamal@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/vpe: cancel delayed work in hw_finiAlex Deucher1-4/+2
We need to cancel any outstanding work at both suspend and driver teardown. Move the cancel to hw_fini which gets called in both cases. Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/vcn: remove unused code in vcn_v1_0.cDavid (Ming Qiang) Wu1-2/+0
Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/amdgpu : Use the MES INV_TLBS API for tlb invalidation on gfx12Shaoyun Liu3-0/+68
From MES version 0x81, it provide the new API INV_TLBS that support invalidate tlbs with PASID. Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/include : Update MES v12 API header(INV_TLBS)Shaoyun Liu1-0/+30
The requirement from driver side is to have an API that can do the tlb invalidation on dedicate pasid since driver don't know the vmid and process mapping. Make the API generic to support different tlb invalidation related request. Driver can specify pasid, vmid, hub_id and vm address range need to be invalidated. With this API the old INV_GART in MISC Op can be deprecated. Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: fix shift-out-of-bounds in amdgpu_debugfs_jpeg_sched_mask_setJesse.Zhang1-1/+1
Fix a UBSAN shift-out-of-bounds warning in amdgpu_debugfs_jpeg_sched_mask_set when the shift exponent reaches or exceeds 32 bits. The issue occurred because a 32-bit integer '1' was being shifted by up to 32 bits, which is undefined behavior. Replace '1' with '1ULL' to ensure 64-bit arithmetic, matching the u64 type of 'val' and preventing the shift overflow. This is consistent with the existing mask calculation that already uses 1ULL. The error manifested as: UBSAN: shift-out-of-bounds in drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c:373:17 shift exponent 32 is too large for 32-bit type 'int' v2: remove debug log Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Suggested-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: DC v3.2.347Nicholas Carbones1-1/+1
DC Release v3.2.347 * Firmware releases for multiple asics * CodeQL fixes * Fix for double cursor with 180 degree rotation on large resolutions * Misc bug fixes for DSC, PSR/Replay, DPIA etc. Signed-off-by: Nicholas Carbones <ncarbone@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Reviewed-by: Leo Li <sunpeng.li@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: [FW Promotion] Release 0.1.24.0Taimur Hassan1-0/+4
Add two new IPS residency data modes. Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Reviewed-by: Leo Li <sunpeng.li@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27Reapply "drm/amdgpu: fix incorrect vm flags to map bo"Jack Xiao1-2/+2
It should use vm flags instead of pte flags to specify bo vm attributes. This reverts commit 1263ceea2a1327014d9de2858a122f3c27dfa4dd. Reapply this patch with the proper fixes tag. Fixes: 6716a823d18d ("drm/amdgpu: rework how PTE flags are generated v3") Signed-off-by: Jack Xiao <Jack.Xiao@amd.com> Reviewed-by: Likun Gao <Likun.Gao@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27Revert "drm/amdgpu: fix incorrect vm flags to map bo"Alex Deucher1-2/+2
This reverts commit b08425fa77ad2f305fe57a33dceb456be03b653f. Revert this to align with 6.17 because the fixes tag was wrong on this commit. Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Consider sink max slice width limitation for dscDillon Varone1-0/+5
[WHY&HOW] The sink max slice width limitation should be considered for DSC, but was removed in "refactor DSC cap calculations". This patch adds it back and takes the valid minimum between the sink and source. Signed-off-by: Dillon Varone <Dillon.Varone@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Array offset used before range checkClay King5-5/+5
Consolidating multiple CodeQL Fixes for alerts with rule id: cpp/offset-use-before-range-check Reviewed-by: Joshua Aberback <joshua.aberback@amd.com> Signed-off-by: Clay King <clayking@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: wait for otg update pending latch before clock optimizationYihan Zhu4-0/+22
[WHY & HOW] OTG pending update unlatched will cause system fail, wait OTG fully disabled to avoid this error. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Yihan Zhu <Yihan.Zhu@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Incorrect 'not' operator usageClay King8-8/+8
Consolidating multiple CodeQL Fixes for alerts with rule id: cpp/incorrect-not-operator-usage Reviewed-by: Joshua Aberback <joshua.aberback@amd.com> Signed-off-by: Clay King <clayking@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Multiplication result converted to larger typeClay King7-16/+16
Consolidating multiple CodeQL Fixes for alerts with rule id: cpp/integer-multiplication-cast-to-long Reviewed-by: Joshua Aberback <joshua.aberback@amd.com> Signed-off-by: Clay King <clayking@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Support HW cursor 180 rot for any number of pipe splitsIvan Lipski1-46/+27
[Why] For the HW cursor, its current position in the pipe_ctx->stream struct is not affected by the 180 rotation, i. e. the top left corner is still at 0,0. However, the DPP & HUBP set_cursor_position functions require rotated position. The current approach is hard-coded for ODM 2:1, thus it's failing for ODM 4:1, resulting in a double cursor. [How] Instead of calculating the new cursor position relatively to the viewports, we calculate it using a viewavable clip_rect of each plane. The clip_rects are first offset and scaled to the same space as the src_rect, i. e. Stream space -> Plane space. In case of a pipe split, which divides the plane into 2 or more viewports, the clip_rect is the union of all the viewports of the given plane. With the assumption that the viewports in HUBP's set_cursor_position are in the Plane space as well, it should produce a correct cursor position for any number of pipe splits. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Ivan Lipski <ivan.lipski@amd.com> Signed-off-by: Leo Li <sunpeng.li@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Decrease stack size in logging pathReza Amini1-2/+2
[why] Reducing stack size can avoid stack over flow [how] Make local variables const and static so they are not on the stack. Reviewed-by: Dillon Varone <dillon.varone@amd.com> Signed-off-by: Reza Amini <reza.amini@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: track dpia supportAusef Yousof2-3/+5
[why&how] initialize a flag to track if we previously supported dpia and write that to boot options Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com> Signed-off-by: Ausef Yousof <Ausef.Yousof@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Reserve instance index notified by DMUBCruise Hung3-5/+7
[Why] Reserve instance index notified by DMUB. [How] Add new variable for instance index. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Cruise Hung <Cruise.Hung@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Add control flags to force PSR / replayKarthi Kandasamy4-1/+10
To change PSR/Replay behavior based on OS preferences, add some config options. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Karthi Kandasamy <karthi.kandasamy@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/vpe: add ring reset supportAlex Deucher1-1/+24
Implement ring reset for VPE. Similar to VCN and JPEG, just powergate the the IP to reset it. v2: Properly set per queue reset flag Reviewed-by: Jesse Zhang <jesse.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/vcn: drop extra cancel_delayed_work_sync()Alex Deucher1-2/+0
We already call this in the hw_fini() methods for all VCN instances, so no need to call it again in amdgpu_vcn_suspend(). Tested-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Reviewed-by: Leo Liu <leo.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdkfd: Tie UNMAP_LATENCY to queue_preemptionAmber Lin1-1/+8
When KFD asks CP to preempt queues, other than preempt CP queues, CP also requests SDMA to preempt SDMA queues with UNMAP_LATENCY timeout. Currently queue_preemption_timeout_ms is 9000 ms by default but can be configured via module parameter. KFD_UNMAP_LATENCY_MS is hard coded as 4000 ms though. This patch ties KFD_UNMAP_LATENCY_MS to queue_preemption_timeout_ms so in a slow system such as emulator, both CP and SDMA slowness are taken into account. Signed-off-by: Amber Lin <Amber.Lin@amd.com> Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/pm: Update SMU v13.0.6 PPT caps initializationJesse.Zhang1-1/+4
Update the conditions for setting the SMU vcn reset caps in the SMU v13.0.6 PPT initialization function. Specifically: - Add support for VCN reset capability for firmware versions 0x00558200 and above when the program version is 0. - Add support for VCN reset capability for firmware versions 0x05551800 and above when the program version is 5. v2: correct the smu mp1 version for program 5 (Lijo) Suggested-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdkfd: fix vram allocation failure for a special caseEric Huang1-1/+6
When it only allocates vram without va, which is 0, and a SVM range allocated stays in this range, the vram allocation returns failure. It should be skipped for this case from SVM usage check. Signed-off-by: Eric Huang <jinhuieric.huang@amd.com> Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdkfd: Allow device error to be loggedSunday Clement1-5/+2
The addition of a WARN_ON() check in order to return early in the kq_initialize function retroactively causes the default case in the following switch statement to never be executed, preventing dev_err from logging device errors in the kernel. Both logs are now checked in the default case. Signed-off-by: Sunday Clement <Sunday.Clement@amd.com> Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: clean-up dead code in dml2_mall_phantomBrahmajit Das1-4/+2
pipe_idx in funtion dml2_svp_validate_static_schedulabilit, although set is never actually used. While building with GCC 16 this gives a warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dml2/dml2_mall_phantom.c: In function ‘set_phantom_stream_timing’: drivers/gpu/drm/amd/amdgpu/../display/dc/dml2/dml2_mall_phantom.c:657:25: warning: variable ‘pipe_idx’ set but not used [-Wunused-but-set-variable=] 657 | unsigned int i, pipe_idx; | ^~~~~~~~ Signed-off-by: Brahmajit Das <listout@listout.xyz> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: Add support for dpc to the productCe Sun1-1/+1
Add support for dpc to the product Signed-off-by: Ce Sun <cesun102@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: remove redundant AMDGPU_HAS_VRAMYifan Zhang2-9/+1
AMDGPU_HAS_VRAM is redundant with is_app_apu, as both refer to APUs with no carve-out. Since AMDGPU_HAS_VRAM only occurs once, remove AMDGPU_HAS_VRAM definition. The tmr allocation can be covered with AMDGPU_GEM_DOMAIN_GTT | AMDGPU_GEM_DOMAIN_VRAM in both vram and non vram ASICs. Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: Add support for dpc to a series of productsCe Sun1-1/+1
Add support for dpc to a series of products Signed-off-by: Ce Sun <cesun102@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: Correct the loss of aca bank reg infoCe Sun4-32/+29
By polling, poll ACA bank count to ensure that valid ACA bank reg info can be obtained v2: add corresponding delay before send msg to SMU to query mca bank info (Stanley) v3: the loop cannot exit. (Thomas) v4: remove amdgpu_aca_clear_bank_count. (Kevin) v5: continuously inject ce. If a creation interruption occurs at this time, bank reg info will be lost. (Thomas) v5: each cycle is delayed by 100ms. (Tao) Signed-off-by: Ce Sun <cesun102@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: Add a mutex lock to protect poison injectionCe Sun2-0/+5
When poison is triggered multiple times, competition will occur. Add a mutex lock to protect poison injection Signed-off-by: Ce Sun <cesun102@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: Correct the counts of nr_banks and nr_errorsCe Sun1-0/+2
Correct the counts of nr_banks and nr_errors Signed-off-by: Ce Sun <cesun102@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amd/display: Remove redundant header filesLiao Yuanhong1-2/+0
The header file "dc_stream.h" is already included on line 1507. Remove the redundant include. This is because the header file was initially included towards the latter part of the code. Subsequent commits had to include the header file again earlier in the code. In my opinion, this doesn't count as a fix; it just requires removing the redundant header inclusion. Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/fence: Remove redundant 0 value initializationLiao Yuanhong1-1/+0
The amdgpu_fence struct is already zeroed by kzalloc(). It's redundant to initialize am_fence->context to 0. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu: Allocate psp fw private buffer in vramHawking Zhang1-2/+1
It's not necessarily to allocate psp firmware private buffer in different memory domain in sriov and bare metal environment Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Le Ma <le.ma@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/gfx12: set MQD as appriopriate for queue typesAlex Deucher1-2/+6
Set the MQD as appropriate for the kernel vs user queues. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-27drm/amdgpu/gfx11: set MQD as appriopriate for queue typesAlex Deucher1-2/+6
Set the MQD as appropriate for the kernel vs user queues. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2025-08-23Merge tag 'drm-misc-fixes-2025-08-21' of ↵Dave Airlie3-3/+4
https://gitlab.freedesktop.org/drm/misc/kernel into drm-fixes A bunch of fixes for 6.17: - analogix_dp: devm_drm_bridge_alloc() error handling fix - gaudi: Memory deallocation fix - gpuvm: Documentation warning fix - hibmc: Various misc fixes - nouveau: Memory leak fixes, typos - panic: u64 division handling on 32 bits architecture fix - rockchip: Kconfig fix, register caching fix - rust: memory layout and safety fixes - tests: Endianness fixes Signed-off-by: Dave Airlie <airlied@redhat.com> From: Maxime Ripard <mripard@redhat.com> Link: https://lore.kernel.org/r/20250821-economic-dandelion-rooster-c57fa9@houat
2025-08-22drm/amdgpu: Pin buffers while vmap'ing exported dma-buf objectsThomas Zimmermann1-2/+32
Current dma-buf vmap semantics require that the mapped buffer remains in place until the corresponding vunmap has completed. For GEM-SHMEM, this used to be guaranteed by a pin operation while creating an S/G table in import. GEM-SHMEN can now import dma-buf objects without creating the S/G table, so the pin is missing. Leads to page-fault errors, such as the one shown below. [ 102.101726] BUG: unable to handle page fault for address: ffffc90127000000 [...] [ 102.157102] RIP: 0010:udl_compress_hline16+0x219/0x940 [udl] [...] [ 102.243250] Call Trace: [ 102.245695] <TASK> [ 102.2477V95] ? validate_chain+0x24e/0x5e0 [ 102.251805] ? __lock_acquire+0x568/0xae0 [ 102.255807] udl_render_hline+0x165/0x341 [udl] [ 102.260338] ? __pfx_udl_render_hline+0x10/0x10 [udl] [ 102.265379] ? local_clock_noinstr+0xb/0x100 [ 102.269642] ? __lock_release.isra.0+0x16c/0x2e0 [ 102.274246] ? mark_held_locks+0x40/0x70 [ 102.278177] udl_primary_plane_helper_atomic_update+0x43e/0x680 [udl] [ 102.284606] ? __pfx_udl_primary_plane_helper_atomic_update+0x10/0x10 [udl] [ 102.291551] ? lockdep_hardirqs_on_prepare.part.0+0x92/0x170 [ 102.297208] ? lockdep_hardirqs_on+0x88/0x130 [ 102.301554] ? _raw_spin_unlock_irq+0x24/0x50 [ 102.305901] ? wait_for_completion_timeout+0x2bb/0x3a0 [ 102.311028] ? drm_atomic_helper_calc_timestamping_constants+0x141/0x200 [ 102.317714] ? drm_atomic_helper_commit_planes+0x3b6/0x1030 [ 102.323279] drm_atomic_helper_commit_planes+0x3b6/0x1030 [ 102.328664] drm_atomic_helper_commit_tail+0x41/0xb0 [ 102.333622] commit_tail+0x204/0x330 [...] [ 102.529946] ---[ end trace 0000000000000000 ]--- [ 102.651980] RIP: 0010:udl_compress_hline16+0x219/0x940 [udl] In this stack strace, udl (based on GEM-SHMEM) imported and vmap'ed a dma-buf from amdgpu. Amdgpu relocated the buffer, thereby invalidating the mapping. Provide a custom dma-buf vmap method in amdgpu that pins the object before mapping it's buffer's pages into kernel address space. Do the opposite in vunmap. Note that dma-buf vmap differs from GEM vmap in how it handles relocation. While dma-buf vmap keeps the buffer in place, GEM vmap requires the caller to keep the buffer in place. Hence, this fix is in amdgpu's dma-buf code instead of its GEM code. A discussion of various approaches to solving the problem is available at [1]. v3: - try (GTT | VRAM); drop CPU domain (Christian) v2: - only use mapable domains (Christian) - try pinning to domains in preferred order Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Fixes: 660cd44659a0 ("drm/shmem-helper: Import dmabuf without mapping its sg_table") Reported-by: Thomas Zimmermann <tzimmermann@suse.de> Closes: https://lore.kernel.org/dri-devel/ba1bdfb8-dbf7-4372-bdcb-df7e0511c702@suse.de/ Cc: Shixiong Ou <oushixiong@kylinos.cn> Cc: Thomas Zimmermann <tzimmermann@suse.de> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Maxime Ripard <mripard@kernel.org> Cc: David Airlie <airlied@gmail.com> Cc: Simona Vetter <simona@ffwll.ch> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: "Christian König" <christian.koenig@amd.com> Cc: dri-devel@lists.freedesktop.org Cc: linux-media@vger.kernel.org Cc: linaro-mm-sig@lists.linaro.org Link: https://lore.kernel.org/dri-devel/9792c6c3-a2b8-4b2b-b5ba-fba19b153e21@suse.de/ # [1] Reviewed-by: Christian König <christian.koenig@amd.com> Link: https://lore.kernel.org/r/20250821064031.39090-1-tzimmermann@suse.de
2025-08-20Merge drm/drm-fixes into drm-misc-fixesMaxime Ripard5-11/+33
Update drm-misc-fixes to -rc2. Signed-off-by: Maxime Ripard <mripard@kernel.org>
2025-08-18drm/amd/display: Fix DP audio DTO1 clock source on DCE 6.Timur Kristóf1-15/+6
On DCE 6, DP audio was not working. However, it worked when an HDMI monitor was also plugged in. Looking at dce_aud_wall_dto_setup it seems that the main difference is that we use DTO1 when only DP is plugged in. When programming DTO1, it uses audio_dto_source_clock_in_khz which is set from get_dp_ref_freq_khz The dce60_get_dp_ref_freq_khz implementation looks incorrect, because DENTIST_DISPCLK_CNTL seems to be always zero on DCE 6, so it isn't usable. I compared dce60_get_dp_ref_freq_khz to the legacy display code, specifically dce_v6_0_audio_set_dto, and it turns out that in case of DCE 6, it needs to use the display clock. With that, DP audio started working on Pitcairn, Oland and Cape Verde. However, it still didn't work on Tahiti. Despite having the same DCE version, Tahiti seems to have a different audio device. After some trial and error I realized that it works with the default display clock as reported by the VBIOS, not the current display clock. The patch was tested on all four SI GPUs: * Pitcairn (DCE 6.0) * Oland (DCE 6.4) * Cape Verde (DCE 6.0) * Tahiti (DCE 6.0 but different) The testing was done on Samsung Odyssey G7 LS28BG700EPXEN on each of the above GPUs, at the following settings: * 4K 60 Hz * 1080p 60 Hz * 1080p 144 Hz Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 645cc7863da5de700547d236697dffd6760cf051) Cc: stable@vger.kernel.org
2025-08-18drm/amd/display: Fix fractional fb divider in set_pixel_clock_v3Timur Kristóf1-1/+1
For later VBIOS versions, the fractional feedback divider is calculated as the remainder of dividing the feedback divider by a factor, which is set to 1000000. For reference, see: - calculate_fb_and_fractional_fb_divider - calc_pll_max_vco_construct However, in case of old VBIOS versions that have set_pixel_clock_v3, they only have 1 byte available for the fractional feedback divider, and it's expected to be set to the remainder from dividing the feedback divider by 10. For reference see the legacy display code: - amdgpu_pll_compute - amdgpu_atombios_crtc_program_pll This commit fixes set_pixel_clock_v3 by dividing the fractional feedback divider passed to the function by 100000. Fixes: 4562236b3bc0 ("drm/amd/dc: Add dc display driver (v2)") Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 027e7acc7e17802ebf28e1edb88a404836ad50d6) Cc: stable@vger.kernel.org
2025-08-18drm/amd/display: Don't print errors for nonexistent connectorsTimur Kristóf2-5/+15
When getting the number of connectors, the VBIOS reports the number of valid indices, but it doesn't say which indices are valid, and not every valid index has an actual connector. If we don't find a connector on an index, that is not an error. Considering these are not actual errors, don't litter the logs. Fixes: 60df5628144b ("drm/amd/display: handle invalid connector indices") Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 249d4bc5f1935f04bb45b3b63c0f8922565124f7)
2025-08-18drm/amd/display: Don't warn when missing DCE encoder capsTimur Kristóf1-4/+4
On some GPUs the VBIOS just doesn't have encoder caps, or maybe not for every encoder. This isn't really a problem and it's handled well, so let's not litter the logs with it. Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 33e0227ee96e62d034781e91f215e32fd0b1d512)
2025-08-18drm/amd/display: Fill display clock and vblank time in ↵Timur Kristóf3-11/+3
dce110_fill_display_configs Also needed by DCE 6. This way the code that gathers this info can be shared between different DCE versions and doesn't have to be repeated. Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com> Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 8107432dff37db26fcb641b6cebeae8981cd73a0) Cc: stable@vger.kernel.org
generated by cgit v1.2.3 (git 2.39.1) at 2025-12-25 13:16:45 +0000