summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd
AgeCommit message (Collapse)Author
2026-01-14drm/amdkfd: No need to suspend whole MES to evict processHarish Kasiviswanathan
Each queue of the process is individually removed and there is not need to suspend whole mes. Suspending mes stops kernel mode queues also causing unnecessary timeouts when running mixed work loads Fixes: 079ae5118e1f ("drm/amdkfd: fix suspend/resume all calls in mes based eviction path") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4765 Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amd/pm: Deprecate print_clk_levels callbackLijo Lazar
Use emit_clk_levels instead. Also, remove the unused helper function for getting sysfs buffer offset. 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>
2026-01-14drm/amd/pm: Use emit clock levels in SMU v15.0.0Lijo Lazar
print_clk_levels is no longer used, use emit_clk_levels instead. 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>
2026-01-14Revert "drm/amdgpu: don't attach the tlb fence for SI"Prike Liang
This reverts commit 820b3d376e8a102c6aeab737ec6edebbbb710e04. It’s better to validate VM TLB flushes in the flush‑TLB backend rather than in the generic VM layer. Reverting this patch depends on commit fa7c231fc2b0 ("drm/amdgpu: validate the flush_gpu_tlb_pasid()") being present in the tree. Signed-off-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amdgpu: validate the flush_gpu_tlb_pasid()Prike Liang
Validate flush_gpu_tlb_pasid() availability before flushing tlb. Signed-off-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amdkfd: Switch to using GC VERSION to decide LDS/Scratch baseLang Yu
Next generation GC IP with 4-level page table needs to use the same LDS/Scratch base with 5-level page table, use GC VERSION to decide is more appropriate. Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amd/pm: fix smu overdrive data type wrong issue on smu 14.0.2Yang Wang
resolving the issue of incorrect type definitions potentially causing calculation errors. Fixes: 54f7f3ca982a ("drm/amdgpu/swm14: Update power limit logic") Signed-off-by: Yang Wang <kevinyang.wang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amdkfd: kfd driver supports hot unplug/replug amdgpu devicesXiaogang Chen
This patch allows kfd driver function correctly when AMD gpu devices got unplug/replug at run time. When an AMD gpu device got unplug kfd driver gracefully terminates existing kfd processes after stops all queues by sending SIGBUS to user process. After that user space can still use remaining AMD gpu devices. When all AMD gpu devices at system got removed kfd driver will not response new requests. Unplugged AMD gpu devices can be re-plugged. kfd driver will use added devices to function as usual. The purpose of this patch is having kfd driver behavior as expected during and after AMD gpu devices unplug/replug at run time. Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amd/pm: fix issue of missing '*' on pp_dpm_xxx nodesYang Wang
refine the code to fix '*' missing on pp_dpm_xxx series node. e.g.: missing '*' on navi10 pp_dpm_sclk $ cat /sys/class/drm/card0/device/pp_dpm_sclk 0: 300Mhz 1: 1930Mhz (missing symbol '*') Fixes: a08ea4bc7711 ("drm/amd/pm: Add a helper to show dpm table") 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>
2026-01-14drm/amdkfd: Fix GART PTE for non-4K pagesize in svm_migrate_gart_map()Donet Tom
In svm_migrate_gart_map(), while migrating GART mapping, the number of bytes copied for the GART table only accounts for CPU pages. On non-4K systems, each CPU page can contain multiple GPU pages, and the GART requires one 8-byte PTE per GPU page. As a result, an incorrect size was passed to the DMA, causing only a partial update of the GART table. Fix this function to work correctly on non-4K page-size systems by accounting for the number of GPU pages per CPU page when calculating the number of bytes to be copied. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com> Signed-off-by: Donet Tom <donettom@linux.ibm.com> Signed-off-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amdkfd: Fix SVM map/unmap address conversion for non-4k page sizesDonet Tom
SVM range size is tracked using the system page size. The range start and end are aligned to system page-sized PFNs, so the total SVM range size equals the total number of pages in the SVM range multiplied by the system page size. The SVM range map/unmap functions pass these system page-sized PFN numbers to amdgpu_vm_update_range(), which expects PFNs based on the GPU page size (4K). On non-4K page systems, this mismatch causes only part of the SVM range to be mapped in the GPU page table, while the rest remains unmapped. If the GPU accesses an unmapped address within the same range, it results in a GPU page fault. To fix this, the required conversion has been added in both svm_range_map_to_gpu() and svm_range_unmap_from_gpu(), ensuring that all pages in the SVM range are correctly mapped on non-4K systems. Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com> Signed-off-by: Donet Tom <donettom@linux.ibm.com> Signed-off-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amdkfd: Relax size checking during queue buffer getDonet Tom
HW-supported EOP buffer sizes are 4K and 32K. On systems that do not use 4K pages, the minimum buffer object (BO) allocation size is PAGE_SIZE (for example, 64K). During queue buffer acquisition, the driver currently checks the allocated BO size against the supported EOP buffer size. Since the allocated BO is larger than the expected size, this check fails, preventing queue creation. Relax the strict size validation and allow PAGE_SIZE-sized BOs to be used. Only the required 4K region of the buffer will be used as the EOP buffer and avoids queue creation failures on non-4K page systems. Acked-by: Christian König <christian.koenig@amd.com> Suggested-by: Philip Yang <yangp@amd.com> Signed-off-by: Donet Tom <donettom@linux.ibm.com> Signed-off-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/amd/display: Initialise backlight level values from hwVivek Das Mohapatra
Internal backlight levels are initialised from ACPI but the values are sometimes out of sync with the levels in effect until there has been a read from hardware (eg triggered by reading from sysfs). This means that the first drm_commit can cause the levels to be set to a different value than the actual starting one, which results in a sudden change in brightness. This path shows the problem (when the values are out of sync): amdgpu_dm_atomic_commit_tail() -> amdgpu_dm_commit_streams() -> amdgpu_dm_backlight_set_level(..., dm->brightness[n]) This patch calls the backlight ops get_brightness explicitly at the end of backlight registration to make sure dm->brightness[n] is in sync with the actual hardware levels. Fixes: 2fe87f54abdc ("drm/amd/display: Set default brightness according to ACPI") Signed-off-by: Vivek Das Mohapatra <vivek@collabora.com> Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-14drm/atomic: add max_size check to drm_property_replace_blob_from_id()Dmitry Baryshkov
The function drm_property_replace_blob_from_id() allows checking whether the blob size is equal to a predefined value. In case of variable-size properties (like the gamma / degamma LUTs) we might want to check for the blob size against the maximum, allowing properties of the size lesser than the max supported by the hardware. Extend the function in order to support such checks. Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de> Link: https://patch.msgid.link/20260106-drm-fix-lut-checks-v3-2-f7f979eb73c8@oss.qualcomm.com Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
2026-01-10drm/amd/display: Promote DC to 3.2.365Taimur Hassan
This version brings along the following updates: - Cleanup, refactoring of panel replay code to prepare for non-eDP replay - Switch to drm_dbg_macros instead of DRM_DEBUG variants - Add pwait status to DMCUB debug logging - Adjust PHY FSM transition to TX_EN-to-PLL_ON for TMDS on DCN35 - Always update divider settings for DP tunnel - correct clip x assignment in cursor programming - Bump the HDMI clock to 340MHz Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Acked-by: Leo Li <sunpeng.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: [FW Promotion] Release 0.1.42.0Taimur Hassan
Summary for changes in firmware: * Fix 24/30FPS full screen video low residency * Fix 60Hz video playback freeze * Add Panel Replay command for VESA replay Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Acked-by: Leo Li <sunpeng.li@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: switch to drm_dbg_ macros instead of DRM_DEBUG_ variantsAurabindo Pillai
Device pointer sources used: - connector->dev - when a DRM connector was available - old_plane_crtc->dev / new_plane_crtc->dev - for plane state functions - pass in drm_device - for the stream scaling function - aconnector->base.dev - for the VSDB parsing function Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: switch to drm_dbg_kms() from DRM_DEBUG_KMSAurabindo Pillai
DRM_DEBUG_ class of macros are depricated. Recommended drm_dbg_kms() has the advantage of being able to distinguish the logs from devices in a multi-gpu environment. Where a pointer to struct amdgpu_device is available, use that to get the drm device. Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com> Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Add pwait status to DMCUB debug loggingNicholas Kazlauskas
[Why] To know if DMCUB is idle at the time of the debug data being collected. [How] Extend the logging to include the field. It's already captured as part of each ASIC's get_diagnostic_data. Reviewed-by: Dillon Varone <dillon.varone@amd.com> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Adjust PHY FSM transition to TX_EN-to-PLL_ON for TMDS on DCN35Nicholas Kazlauskas
[Why] A backport of the change made for DCN401 that addresses an issue where we turn off the PHY PLL when disabling TMDS output, which causes the OTG to remain stuck. The OTG being stuck can lead to a hang in the DCHVM's ability to ACK invalidations when it thinks the HUBP is still on but it's not receiving global sync. The transition to PLL_ON needs to be atomic as there's no guarantee that the thread isn't pre-empted or is able to complete before the IOMMU watchdog times out. [How] Backport the implementation from dcn401 back to dcn35. There's a functional difference in when the eDP output is disabled in dcn401 code so we don't want to utilize it directly. Reviewed-by: Yihan Zhu <yihan.zhu@amd.com> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Add replay_events in replay settingsPeichen Huang
Add a new field to replay_settings. Reviewed-by: Robin Chen <robin.chen@amd.com> Signed-off-by: Peichen Huang <PeiChen.Huang@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: init code for external panel replayPeichen Huang
Prepare for supporting panel replay on external monitors. Reviewed-by: Robin Chen <robin.chen@amd.com> Reviewed-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Peichen Huang <PeiChen.Huang@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: disable replay when crc source is enabledRay Wu
[Why] IGT CRC tests fail on replay panels due to invalid CRC values captured when replay is active. [How] - Disable replay when CRC source is enabled; set flag to prevent unexpected re-enable - Reset flag when CRC source is disabled to allow replay Reviewed-by: ChiaHsuan (Tom) Chung <chiahsuan.chung@amd.com> Signed-off-by: Ray Wu <ray.wu@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Always update divider settings for DP tunnelCruise Hung
[Why] When transitioning from 640x480 at RBRx1 to HBR3x1, both output pixel mode and pixel rate divider should update. The needs_divider_update flag was only for 8b10b and 128b132b transition. [How] For DP tunneling, always update divider settings. Reviewed-by: Jerry Zuo <jerry.zuo@amd.com> Signed-off-by: Cruise Hung <Cruise.Hung@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: revert "write default Vesa Aux backlight control in dmub"Derek Lai
This reverts commit 08f133e932cc. Commit causing backlight darker when the system wakes up from standby mode, and green screen w/ secondary screen only when the system wakes up from standby mode. Reviewed-by: Charlene Liu <charlene.liu@amd.com> Signed-off-by: Derek Lai <Derek.Lai@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: correct clip x assignment in cursor programmingAusef Yousof
[why&how] primary planes for external displays getting incorrect clip values, detect such a scenario and pass correct parameters Reviewed-by: Ovidiu (Ovi) Bunea <ovidiu.bunea@amd.com> Signed-off-by: Ausef Yousof <auyousof@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Add global fgcg function prototype to DCCGLeo Chen
[Why & How] Add the ability to enable/disable global fine grain clock gating within DCCG. Reviewed-by: Charlene Liu <charlene.liu@amd.com> Signed-off-by: Leo Chen <leo.chen@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: only power down dig on phy endpointsDmytro Laktyushkin
This avoids any issues with dpia endpoints Reviewed-by: Charlene Liu <charlene.liu@amd.com> Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: PR error HPD_IRQ handlingJack Chang
[Why & How] Add error handling for IRQ_HPD in Panel Replay Reviewed-by: Robin Chen <robin.chen@amd.com> Reviewed-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Jack Chang <jack.chang@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: DPCD for Selective UpdateJack Chang
[Why&How] Add flow to read selective update related info from DPCD, and pass the info to DMUB. Reviewed-by: Robin Chen <robin.chen@amd.com> Reviewed-by: Wenjing Liu <wenjing.liu@amd.com> Signed-off-by: Jack Chang <jack.chang@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Bump the HDMI clock to 340MHzMario Limonciello
[Why] DP-HDMI dongles can execeed bandwidth requirements on high resolution monitors. This can lead to pruning the high resolution modes. HDMI 1.3 bumped the clock to 340MHz, but display code never matched it. [How] Set default to (DVI) 165MHz. Once HDMI display is identified update to 340MHz. Reported-by: Dianne Skoll <dianne@skoll.ca> Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4780 Reviewed-by: Chris Park <chris.park@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Show link name in PSR status messageMario Limonciello (AMD)
[Why] The PSR message was moved in commit 4321742c394e ("drm/amd/display: Move PSR support message into amdgpu_dm"). This message however shows for every single link without showing which link is which. This can send a confusing message to the user. [How] Add link name into the message. Fixes: 4321742c394e ("drm/amd/display: Move PSR support message into amdgpu_dm") Reviewed-by: Alex Hung <alex.hung@amd.com> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Remove unused DMUB replay commandsRobin Chen
[WHY] Remove unused DMUB Replay set version command and related code. Reviewed-by: Jack Chang <jack.chang@amd.com> Signed-off-by: Robin Chen <robin.chen@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: Re-implement minimal transition deferralJoshua Aberback
[Why] The update v3 path got refactored into new functions, which happened just before the previous implementation was submitted, which resulted in the optimizations not executing. This commit re-implements the same logic in the new codepath. Reviewed-by: Aric Cyr <aric.cyr@amd.com> Signed-off-by: Joshua Aberback <joshua.aberback@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/display: move panel replay out from edpPeichen Huang
[WHY] Panel Replay is not an eDP-specific function. [HOW] Create new Panel Replay source files and move the Panel Replay functions from the eDP files to the new files. Additionally, create a new link_service construct function to assign the related function pointers. Reviewed-by: Robin Chen <robin.chen@amd.com> Signed-off-by: Peichen Huang <PeiChen.Huang@amd.com> Signed-off-by: Matthew Stewart <matthew.stewart2@amd.com> Tested-by: Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amdgpu/mes: Simplify hqd mask initializationLang Yu
"adev->mes.compute_hqd_mask[i] = adev->gfx.disable_kq ? 0xF" is actually incorrect for MEC with 8 queues per pipe. Let's get rid of version check and hardcode, calculate hqd mask with number of queues per pipe and number of gfx/compute queues kernel used. Currently, only MEC1 is used for both kernel/user compute queue. To enable other MEC, we need to redistribute queues per pipe and adjust queue resource shared with kfd that needs a separate patch. Just skip other MEC for now to avoid potential issues. v2: Force reserved queues to 0 if kernel queue is explicitly disabled. Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amdgpu: Refactor amdgpu_gem_va_ioctl for Handling Last Fence Update and ↵Srinivasan Shanmugam
Timeline Management v7 When GPU memory mappings are updated, the driver returns a fence so userspace knows when the update is finished. The previous refactor could pick the wrong fence or rely on checks that are not safe for GPU mappings that stay valid even when memory is missing. In some cases this could return an invalid fence or cause fence reference counting problems. Fix this by (v5,v6, per Christian): - Starting from the VM’s existing last update fence, so a valid and meaningful fence is always returned even when no new work is required. - Selecting the VM-level fence only for always-valid / PRT mappings using the required combined bo_va + bo guard. - Using the per-BO page table update fence for normal MAP and REPLACE operations. - For UNMAP and CLEAR, returning the fence provided by amdgpu_vm_clear_freed(), which may remain unchanged when nothing needs clearing. - Keeping fence reference counting balanced. v7: Drop the extra bo_va/bo NULL guard since amdgpu_vm_is_bo_always_valid() handles NULL BOs correctly (including PRT). (Christian) This makes VM timeline fences correct and prevents crashes caused by incorrect fence handling. Fixes: bd8150a1b337 ("drm/amdgpu: Refactor amdgpu_gem_va_ioctl for Handling Last Fence Update and Timeline Management v4") Suggested-by: Christian König <christian.koenig@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amdkfd: fix a memory leak in device_queue_manager_init()Haoxiang Li
If dqm->ops.initialize() fails, add deallocate_hiq_sdma_mqd() to release the memory allocated by allocate_hiq_sdma_mqd(). Move deallocate_hiq_sdma_mqd() up to ensure proper function visibility at the point of use. Fixes: 11614c36bc8f ("drm/amdkfd: Allocate MQD trunk for HIQ and SDMA") Signed-off-by: Haoxiang Li <lihaoxiang@isrc.iscas.ac.cn> Signed-off-by: Felix Kuehling <felix.kuehling@amd.com> Reviewed-by: Oak Zeng <Oak.Zeng@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amdgpu: make sure userqs are enabled in userq IOCTLsAlex Deucher
These IOCTLs shouldn't be called when userqs are not enabled. Make sure they are enabled before executing the IOCTLs. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amdgpu: Slightly simplify base_addr_show()Christophe JAILLET
sysfs_emit_at() never returns a negative error code. It returns 0 or the number of characters written in the buffer. Remove the useless tests. This simplifies the logic and saves a few lines of code. Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amdgpu: Drop MMIO_REMAP domain bit and keep it InternalChristian König
"AMDGPU_GEM_DOMAIN_MMIO_REMAP" - Never activated as UAPI and it turned out that this was to inflexible. Allocate the MMIO_REMAP buffer object as a regular GEM BO and explicitly move it into the fixed AMDGPU_PL_MMIO_REMAP placement at the TTM level. This avoids relying on GEM domain bits for MMIO_REMAP, keeps the placement purely internal, and makes the lifetime and pinning of the global MMIO_REMAP BO explicit. The BO is pinned in TTM so it cannot be migrated or evicted. The corresponding free path relies on normal DRM teardown ordering, where no further user ioctls can access the global BO once TTM teardown begins. v2 (Srini): - Updated patch title. - Drop use of AMDGPU_GEM_DOMAIN_MMIO_REMAP in amdgpu_ttm.c. The MMIO_REMAP domain bit is removed from UAPI, so keep the MMIO_REMAP BO allocation domain-less (bp.domain = 0) and rely on the TTM placement (AMDGPU_PL_MMIO_REMAP) for backing/pinning. - Keep fdinfo/mem-stats visibility for MMIO_REMAP by classifying BOs based on bo->tbo.resource->mem_type == AMDGPU_PL_MMIO_REMAP, since the domain bit is removed. v3: Squash patches #1 & #3 Fixes: 056132483724 ("drm/amdgpu/uapi: Introduce AMDGPU_GEM_DOMAIN_MMIO_REMAP") Fixes: 2a7a794eb82c ("drm/amdgpu/ttm: Allocate/Free 4K MMIO_REMAP Singleton") Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Leo Liu <leo.liu@amd.com> Cc: Ruijing Dong <ruijing.dong@amd.com> Cc: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-10drm/amd/pm: Use message control for debug mailboxLijo Lazar
Migrate existing debug message mechanism so that it uses debug message callbacks in message control block. 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>
2026-01-10drm/amd/pm: Add debug message callbackLijo Lazar
Add callback in message control to send message through debug mailbox. 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>
2026-01-10drm/amd/pm: Drop unused ppt callback definitionsLijo Lazar
SMU message related ppt callbacks are not used. Drop from ppt_funcs. 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>
2026-01-10drm/amd/pm: Drop unused ppt callback from SMUv15Lijo Lazar
SMU message related ppt callbacks are not used. Drop from SMUv15. 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>
2026-01-10drm/amd/pm: Drop unused ppt callback from SMUv14Lijo Lazar
SMU message related ppt callbacks are not used. Drop from SMUv14. 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>
2026-01-10drm/amd/pm: Drop unused ppt callback from SMUv13Lijo Lazar
SMU message related ppt callbacks are not used. Drop from SMUv13. 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>
2026-01-10drm/amd/pm: Drop unused ppt callback from SMUv12Lijo Lazar
SMU message related ppt callbacks are not used. Drop from SMUv12. 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>
2026-01-10drm/amd/pm: Drop unused ppt callback from SMUv11Lijo Lazar
SMU message related ppt callbacks are not used. Drop from SMUv11. 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>
2026-01-10drm/amd/pm: Drop legacy message related fieldsLijo Lazar
Remove legacy message related fields from smu context. 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>
generated by cgit v1.2.3 (git 2.25.1) at 2026-09-23 00:49:59 +0000