summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu
AgeCommit message (Collapse)Author
2026-01-29drm/amdgpu/gfx11: adjust KGQ reset sequenceAlex Deucher
Kernel gfx queues do not need to be reinitialized or remapped after a reset. This fixes queue reset failures on APUs. v2: preserve init and remap for MMIO case. Fixes: b3e9bfd86658 ("drm/amdgpu/gfx11: add ring reset callbacks") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4789 Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-29drm/amdgpu/gfx12: fix wptr reset in KGQ initAlex Deucher
wptr is a 64 bit value and we need to update the full value, not just 32 bits. Align with what we already do for KCQs. Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Jesse Zhang <jesse.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-29drm/amdgpu/gfx11: fix wptr reset in KGQ initAlex Deucher
wptr is a 64 bit value and we need to update the full value, not just 32 bits. Align with what we already do for KCQs. Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Jesse Zhang <jesse.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-29drm/amdgpu/gfx10: fix wptr reset in KGQ initAlex Deucher
wptr is a 64 bit value and we need to update the full value, not just 32 bits. Align with what we already do for KCQs. Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Jesse Zhang <jesse.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-29drm/amdgpu: Use AMDGPU_MQD_SIZE_ALIGN in KGDLang Yu
Use AMDGPU_MQD_SIZE_ALIGN for both kernel and user queue. Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: David Belanger <david.belanger@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-29drm/amdgpu: Add a helper macro to align mqd sizeLang Yu
MES FW uses address(mqd_addr + sizeof(struct mqd) + 3*sizeof(uint32_t)) as fence address and writes a 32 bit fence value to this address. Driver needs to allocate some extra memory(at least 4 DWs) in addition to sizeof(struct mqd) as mqd memory(limited to gfx/compute/sdma queue). For gfx11/12, sizeof(struct mqd) < PAGE_SIZE, KGD allocates mqd memory with PAGE_SIZE aligned works. For gfx12.1, sizeof(struct mqd) == PAGE_SIZE, it doesn't work. KFD mqd manager hardcodes mqd size to PAGE_SIZE/MQD_SIZE across different IP versions to solve this issue. To avoid hardcoding in differnet places and across different IP versions. Let's use AMDGPU_MQD_SIZE_ALIGN instead. It is used in two places. 1. mqd memory alloction 2. mqd stride handling for multi xcc config v2: Use AMDGPU_GPU_PAGE_ALIGN. (Mukul) Signed-off-by: Lang Yu <lang.yu@amd.com> Reviewed-by: David Belanger <david.belanger@amd.com> (v1) Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Mukul Joshi <mukul.joshi@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-29drm/amdgpu: validate user queue size constraintsJesse.Zhang
Add validation to ensure user queue sizes meet hardware requirements: - Size must be a power of two for efficient ring buffer wrapping - Size must be at least AMDGPU_GPU_PAGE_SIZE to prevent undersized allocations This prevents invalid configurations that could lead to GPU faults or unexpected behavior. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-28drm/amdgpu: Fix cond_exec handling in amdgpu_ib_schedule()Alex Deucher
The EXEC_COUNT field must be > 0. In the gfx shadow handling we always emit a cond_exec packet after the gfx_shadow packet, but the EXEC_COUNT never gets patched. This leads to a hang when we try and reset queues on gfx11 APUs. Fixes: c68cbbfd54c6 ("drm/amdgpu: cleanup conditional execution") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4789 Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit ba205ac3d6e83f56c4f824f23f1b4522cb844ff3) Cc: stable@vger.kernel.org
2026-01-28drm/amdgpu/soc21: fix xclk for APUsAlex Deucher
The reference clock is supposed to be 100Mhz, but it appears to actually be slightly lower (99.81Mhz). Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14451 Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 637fee3954d4bd509ea9d95ad1780fc174489860) Cc: stable@vger.kernel.org
2026-01-28drm/amdgpu: Fix cond_exec handling in amdgpu_ib_schedule()Alex Deucher
The EXEC_COUNT field must be > 0. In the gfx shadow handling we always emit a cond_exec packet after the gfx_shadow packet, but the EXEC_COUNT never gets patched. This leads to a hang when we try and reset queues on gfx11 APUs. Fixes: c68cbbfd54c6 ("drm/amdgpu: cleanup conditional execution") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4789 Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-28drm/amdgpu/soc21: fix xclk for APUsAlex Deucher
The reference clock is supposed to be 100Mhz, but it appears to actually be slightly lower (99.81Mhz). Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/14451 Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-28BackMerge tag 'v6.19-rc7' into drm-nextDave Airlie
Linux 6.19-rc7 This is needed for msm and rust trees. Signed-off-by: Dave Airlie <airlied@redhat.com>
2026-01-27drm/amdgpu: fix NULL pointer dereference in amdgpu_gmc_filter_faults_removeJon Doron
On APUs such as Raven and Renoir (GC 9.1.0, 9.2.2, 9.3.0), the ih1 and ih2 interrupt ring buffers are not initialized. This is by design, as these secondary IH rings are only available on discrete GPUs. See vega10_ih_sw_init() which explicitly skips ih1/ih2 initialization when AMD_IS_APU is set. However, amdgpu_gmc_filter_faults_remove() unconditionally uses ih1 to get the timestamp of the last interrupt entry. When retry faults are enabled on APUs (noretry=0), this function is called from the SVM page fault recovery path, resulting in a NULL pointer dereference when amdgpu_ih_decode_iv_ts_helper() attempts to access ih->ring[]. The crash manifests as: BUG: kernel NULL pointer dereference, address: 0000000000000004 RIP: 0010:amdgpu_ih_decode_iv_ts_helper+0x22/0x40 [amdgpu] Call Trace: amdgpu_gmc_filter_faults_remove+0x60/0x130 [amdgpu] svm_range_restore_pages+0xae5/0x11c0 [amdgpu] amdgpu_vm_handle_fault+0xc8/0x340 [amdgpu] gmc_v9_0_process_interrupt+0x191/0x220 [amdgpu] amdgpu_irq_dispatch+0xed/0x2c0 [amdgpu] amdgpu_ih_process+0x84/0x100 [amdgpu] This issue was exposed by commit 1446226d32a4 ("drm/amdgpu: Remove GC HW IP 9.3.0 from noretry=1") which changed the default for Renoir APU from noretry=1 to noretry=0, enabling retry fault handling and thus exercising the buggy code path. Fix this by adding a check for ih1.ring_size before attempting to use it. Also restore the soft_ih support from commit dd299441654f ("drm/amdgpu: Rework retry fault removal"). This is needed if the hardware doesn't support secondary HW IH rings. v2: additional updates (Alex) Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3814 Fixes: dd299441654f ("drm/amdgpu: Rework retry fault removal") Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Jon Doron <jond@wiz.io> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 6ce8d536c80aa1f059e82184f0d1994436b1d526) Cc: stable@vger.kernel.org
2026-01-27drm/amdgpu: Send RMA CPER at bad page loadingKent Russell
Some older builds weren't sending RMA CPERs when the bad page threshold was exceeded. Newer builds have resolved this, but there could be systems out there with bad page numbers higher than the threshold, that haven't sent out an RMA CPER. To be thorough and safe, send an RMA CPER when we load the table, if the threshold is met or exceeded, instead of waiting for the next UE to trigger the CPER. Signed-off-by: Kent Russell <kent.russell@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-27drm/amdgpu: Fix jpeg ring test order in vcn_v4_0_3Jesse.Zhang
Fix the vcn reset sequence in vcn_v4_0_3_ring_reset() to restore JPEG power state and unlock the JPEG powergating mutex before running the JPEG ring post-reset helper. Fixes: d25c67fd9d6f ("drm/amdgpu/vcn4.0.3: rework reset handling") Reviewed-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>
2026-01-27drm/amdgpu: fix NULL pointer dereference in amdgpu_gmc_filter_faults_removeJon Doron
On APUs such as Raven and Renoir (GC 9.1.0, 9.2.2, 9.3.0), the ih1 and ih2 interrupt ring buffers are not initialized. This is by design, as these secondary IH rings are only available on discrete GPUs. See vega10_ih_sw_init() which explicitly skips ih1/ih2 initialization when AMD_IS_APU is set. However, amdgpu_gmc_filter_faults_remove() unconditionally uses ih1 to get the timestamp of the last interrupt entry. When retry faults are enabled on APUs (noretry=0), this function is called from the SVM page fault recovery path, resulting in a NULL pointer dereference when amdgpu_ih_decode_iv_ts_helper() attempts to access ih->ring[]. The crash manifests as: BUG: kernel NULL pointer dereference, address: 0000000000000004 RIP: 0010:amdgpu_ih_decode_iv_ts_helper+0x22/0x40 [amdgpu] Call Trace: amdgpu_gmc_filter_faults_remove+0x60/0x130 [amdgpu] svm_range_restore_pages+0xae5/0x11c0 [amdgpu] amdgpu_vm_handle_fault+0xc8/0x340 [amdgpu] gmc_v9_0_process_interrupt+0x191/0x220 [amdgpu] amdgpu_irq_dispatch+0xed/0x2c0 [amdgpu] amdgpu_ih_process+0x84/0x100 [amdgpu] This issue was exposed by commit 1446226d32a4 ("drm/amdgpu: Remove GC HW IP 9.3.0 from noretry=1") which changed the default for Renoir APU from noretry=1 to noretry=0, enabling retry fault handling and thus exercising the buggy code path. Fix this by adding a check for ih1.ring_size before attempting to use it. Also restore the soft_ih support from commit dd299441654f ("drm/amdgpu: Rework retry fault removal"). This is needed if the hardware doesn't support secondary HW IH rings. v2: additional updates (Alex) Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3814 Fixes: dd299441654f ("drm/amdgpu: Rework retry fault removal") Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Jon Doron <jond@wiz.io> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-27drm/amdgpu: Simplify sorting of the bo listTvrtko Ursulin
Sort function only cares about the sign so we can replace the conditionals with a single subtraction. Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-27drm/amdgpu/mes: Remove idr leftovers v2Tvrtko Ursulin
Commit cb17fff3a254 ("drm/amdgpu/mes: remove unused functions") removed most of the code using these IDRs but forgot to remove the struct members and init/destroy paths. There is also interrupt handling code in SDMA 5.0 and 5.2 which appears to be using it, but is is unreachable since nothing ever allocates the relevant IDR. We replace those with one time warnings just to avoid any functional difference, but it is also possible they should be removed. v2: also fix up gfx_v12_1.c and sdma_v7_1.c Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com> References: cb17fff3a254 ("drm/amdgpu/mes: remove unused functions") Cc: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu: fix type for wptr in ring backupAlex Deucher
Needs to be a u64. Fixes: 77cc0da39c7c ("drm/amdgpu: track ring state associated with a fence") Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 56fff1941abd3ca3b6f394979614ca7972552f7f)
2026-01-21drm/amdgpu: Fix validating flush_gpu_tlb_pasid()Timur Kristóf
When a function holds a lock and we return without unlocking it, it deadlocks the kernel. We should always unlock before returning. This commit fixes suspend/resume on SI. Tested on two Tahiti GPUs: FirePro W9000 and R9 280X. Fixes: f4db9913e4d3 ("drm/amdgpu: validate the flush_gpu_tlb_pasid()") Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Closes: https://lore.kernel.org/r/202601190121.z9C0uml5-lkp@intel.com/ Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit e3a6eff92bbd960b471966d9afccb4d584546d17)
2026-01-21drm/amdgpu: rename amdgpu_fence_driver_guilty_force_completion()Alex Deucher
The function no longer signals the fence so rename it to better match what it does. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu: fix type for wptr in ring backupAlex Deucher
Needs to be a u64. Fixes: 77cc0da39c7c ("drm/amdgpu: track ring state associated with a fence") Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu: mark invalid records with U64_MAXGangliang Xie
set retired_page of invalid ras records to U64_MAX, and skip them when reading ras records Signed-off-by: Gangliang Xie <ganglxie@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu: Avoid excessive dmesg logLijo Lazar
KIQ access is not guaranteed to work reliably under all reset situations. Avoid flooding dmesg with HDP flush failure messages. 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>
2026-01-21drm/amdgpu: Fix validating flush_gpu_tlb_pasid()Timur Kristóf
When a function holds a lock and we return without unlocking it, it deadlocks the kernel. We should always unlock before returning. This commit fixes suspend/resume on SI. Tested on two Tahiti GPUs: FirePro W9000 and R9 280X. Fixes: f4db9913e4d3 ("drm/amdgpu: validate the flush_gpu_tlb_pasid()") Reported-by: kernel test robot <lkp@intel.com> Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Closes: https://lore.kernel.org/r/202601190121.z9C0uml5-lkp@intel.com/ Signed-off-by: Timur Kristóf <timur.kristof@gmail.com> Signed-off-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Prike Liang <Prike.Liang@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu/vcn5.0.1: rework reset handlingJesse.Zhang
Resetting VCN resets the entire tile, including jpeg. When resetting the VCN, we need to ensure that JPEG data blocks are accessible and we also need to handle the JPEG queue. Add a helper function to restore the JPEG queue during the VCN reset. v2: split the jpeg helper in two, in the top helper we can stop the sched workqueues and attempt to wait for any outstanding fences. Then in the bottom helper, we can force completion, re-init the rings, and restart the sched workqueues (Alex) v3: merge patches 4 and 5 into one patch (Alex) Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu/vcn4.0.3: rework reset handlingJesse.Zhang
Resetting VCN resets the entire tile, including jpeg. When resetting the VCN, we need to ensure that JPEG data blocks are accessible and we also need to handle the JPEG queue. Add a helper function to restore the JPEG queue during the VCN reset. v2: split the jpeg helper in two, in the top helper we can stop the sched workqueues and attempt to wait for any outstanding fences. Then in the bottom helper, we can force completion, re-init the rings, and restart the sched workqueues (Alex) v3: merge patches 1 and 2 into one patch (Alex) Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-21drm/amdgpu/vcn4.0.3: implement DPG pause mode handling for VCN 4.0.3Jesse.Zhang
For MI projects, when Dynamic Power Gating (DPG) is enabled, VCN reset operations should be performed with DPG in pause mode. Otherwise, the hardware may perform undesirable reset operations Reviewed-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>
2026-01-20drm/amdgpu: fix error handling in ib_schedule()Alex Deucher
If fence emit fails, free the fence if necessary. Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 5eb680a06007f2f6ea333d11a4e29039da90614b)
2026-01-20drm/amdgpu: free hw_vm_fence when fail in amdgpu_job_allocJiqian Chen
If drm_sched_job_init fails, hw_vm_fence is not freed currently, then cause memory leak. Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Link: https://lore.kernel.org/amd-gfx/a5a828cb-0e4a-41f0-94c3-df31e5ddad52@amd.com/T/#t Signed-off-by: Jiqian Chen <Jiqian.Chen@amd.com> Reviewed-by: Amos Kong <kongjianjun@gmail.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 5d42ee457ccd1fb5da4c7f817825b2806ec36956)
2026-01-20drm/amdgpu: remove frame cntl for gfx v12Likun Gao
Remove emit_frame_cntl function for gfx v12, which is not support. Signed-off-by: Likun Gao <Likun.Gao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 5aaa5058dec5bfdcb24c42fe17ad91565a3037ca) Cc: stable@vger.kernel.org
2026-01-20drm/amdgpu: add new job idsAlex Deucher
Use this for gfx, sdma, vpe IB tests and kernel shaders. The end goal it to get rid of the direct IB submit without a job structure. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amdgpu: fix error handling in ib_schedule()Alex Deucher
If fence emit fails, free the fence if necessary. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amdgpu/jpeg4.0.3: remove redundant sr-iov checkAlex Deucher
The per queue reset flag is only set when sr-iov is disabled so this check is not necessary as the function will never be called on sr-iov. Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amdgpu: Improve IP discovery checksum failure loggingPerry Yuan
Enhance the error logging in amdgpu_discovery_verify_checksum() to print the calculated checksum, the expected checksum, the data size. This extra context helps quickly identify if the issue is a data corruption, a partially read binary, or an invalid table header without requiring additional instrumentation. Signed-off-by: Perry Yuan <perry.yuan@amd.com> Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amdgpu: free hw_vm_fence when fail in amdgpu_job_allocJiqian Chen
If drm_sched_job_init fails, hw_vm_fence is not freed currently, then cause memory leak. Fixes: db36632ea51e ("drm/amdgpu: clean up and unify hw fence handling") Link: https://lore.kernel.org/amd-gfx/a5a828cb-0e4a-41f0-94c3-df31e5ddad52@amd.com/T/#t Signed-off-by: Jiqian Chen <Jiqian.Chen@amd.com> Reviewed-by: Amos Kong <kongjianjun@gmail.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amd/amdgpu: Add independent hang detect work for user queue fenceJesse.Zhang
In error scenarios (e.g., malformed commands), user queue fences may never be signaled, causing processes to wait indefinitely. To address this while preserving the requirement of infinite fence waits, implement an independent timeout detection mechanism: 1. Initialize a hang detect work when creating a user queue (one-time setup) 2. Start the work with queue-type-specific timeout (gfx/compute/sdma) when the last fence is created via amdgpu_userq_signal_ioctl (per-fence timing) 3. Trigger queue reset logic if the timer expires before the fence is signaled v2: make timeout per queue type (adev->gfx_timeout vs adev->compute_timeout vs adev->sdma_timeout) to be consistent with kernel queues. (Alex) v3: The timeout detection must be independent from the fence, e.g. you don't wait for a timeout on the fence but rather have the timeout start as soon as the fence is initialized. (Christian) v4: replace the timer with the `hang_detect_work` delayed work. Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amdgpu: remove frame cntl for gfx v12Likun Gao
Remove emit_frame_cntl function for gfx v12, which is not support. Signed-off-by: Likun Gao <Likun.Gao@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-20drm/amdkfd: Move gfx9.4.3 and gfx 9.5 MQD to HBMPhilip Yang
To reduce queue switch latency further, move MQD to VRAM domain, CP access MQD and control stack via FB aperture, this requires contiguous pages. After MQD is initialized, updated or restored, flush HDP to guarantee the data is written to HBM and GPU cache is invalidated, then CP will read the new MQD. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2026-01-19Merge tag 'amd-drm-next-6.20-2026-01-16' of ↵Dave Airlie
https://gitlab.freedesktop.org/agd5f/linux into drm-next amd-drm-next-6.20-2026-01-16: amdgpu: - SR-IOV fixes - Rework SMU mailbox handling - Drop MMIO_REMAP domain - UserQ fixes - MES cleanups - Panel Replay updates - HDMI fixes - Backlight fixes - SMU 14.x fixes - SMU 15 updates amdkfd: - Fix a memory leak - Fixes for systems with non-4K pages - LDS/Scratch cleanup - MES process eviction fix Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexander.deucher@amd.com> Link: https://patch.msgid.link/20260116202609.23107-1-alexander.deucher@amd.com
2026-01-15Merge tag 'amd-drm-next-6.20-2026-01-09' of ↵Dave Airlie
https://gitlab.freedesktop.org/agd5f/linux into drm-next amd-drm-next-6.20-2026-01-09: amdgpu: - GPUVM updates - Initial support for larger GPU address spaces - Initial SMUIO 15.x support - Documentation updates - Initial PSP 15.x support - Initial IH 7.1 support - Initial IH 6.1.1 support - SMU 13.0.12 updates - RAS updates - Initial MMHUB 3.4 support - Initial MMHUB 4.2 support - Initial GC 12.1 support - Initial GC 11.5.4 support - HDMI fixes - Panel replay improvements - DML updates - DC FP fixes - Initial SDMA 6.1.4 support - Initial SDMA 7.1 support - Userq updates - DC HPD refactor - SwSMU cleanups and refactoring - TTM memory ops parallelization - DCN 3.5 fixes - DP audio fixes - Clang fixes - Misc spelling fixes and cleanups - Initial SDMA 7.11.4 support - Convert legacy DRM logging helpers to new drm logging helpers - Initial JPEG 5.3 support - Add support for changing UMA size via the driver - DC analog fixes - GC 9 gfx queue reset support - Initial SMU 15.x support amdkfd: - Reserved SDMA rework - Refactor SPM - Initial GC 12.1 support - Initial GC 11.5.4 support - Initial SDMA 7.1 support - Initial SDMA 6.1.4 support - Increase the kfd process hash table - Per context support - Topology fixes radeon: - Convert legacy DRM logging helpers to new drm logging helpers - Use devm for i2c adapters - Variable sized array fix - Misc cleanups UAPI: - KFD context support. Proposed userspace: https://github.com/ROCm/rocm-systems/pull/1705 https://github.com/ROCm/rocm-systems/pull/1701 - Add userq metadata queries for more queue types. Proposed userspace: https://gitlab.freedesktop.org/yogeshmohan/mesa/-/commits/userq_query From: Alex Deucher <alexander.deucher@amd.com> Link: https://patch.msgid.link/20260109154713.3242957-1-alexander.deucher@amd.com Signed-off-by: Dave Airlie <airlied@redhat.com>
2026-01-14drm/amd/display: Add an hdmi_hpd_debounce_delay_ms moduleIvan Lipski
[Why&How] Right now, the HDMI HPD filter is enabled by default at 1500ms. We want to disable it by default, as most modern displays with HDMI do not require it for DPMS mode. The HPD can instead be enabled as a driver parameter with a custom delay value in ms (up to 5000ms). Fixes: c918e75e1ed9 ("drm/amd/display: Add an HPD filter for HDMI") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4859 Signed-off-by: Ivan Lipski <ivan.lipski@amd.com> Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 6a681cd9034587fe3550868bacfbd639d1c6891f)
2026-01-14drm/amdgpu/userq: Fix fence reference leak on queue teardown v2Srinivasan Shanmugam
The user mode queue keeps a pointer to the most recent fence in userq->last_fence. This pointer holds an extra dma_fence reference. When the queue is destroyed, we free the fence driver and its xarray, but we forgot to drop the last_fence reference. Because of the missing dma_fence_put(), the last fence object can stay alive when the driver unloads. This leaves an allocated object in the amdgpu_userq_fence slab cache and triggers This is visible during driver unload as: BUG amdgpu_userq_fence: Objects remaining on __kmem_cache_shutdown() kmem_cache_destroy amdgpu_userq_fence: Slab cache still has objects Call Trace: kmem_cache_destroy amdgpu_userq_fence_slab_fini amdgpu_exit __do_sys_delete_module Fix this by putting userq->last_fence and clearing the pointer during amdgpu_userq_fence_driver_free(). This makes sure the fence reference is released and the slab cache is empty when the module exits. v2: Update to only release userq->last_fence with dma_fence_put() (Christian) Fixes: edc762a51c71 ("drm/amdgpu/userq: move some code around") Cc: Alex Deucher <alexander.deucher@amd.com> Cc: 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> (cherry picked from commit 8e051e38a8d45caf6a866d4ff842105b577953bb)
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> (cherry picked from commit 9163fe4d790fb4e16d6b0e23f55b43cddd3d4a65)
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> (cherry picked from commit f4db9913e4d3dabe9ff3ea6178f2c1bc286012b8)
2026-01-14drm/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> (cherry picked from commit d967509651601cddce7ff2a9f09479f3636f684d) Cc: stable@vger.kernel.org
2026-01-14drm/amdgpu: Use correct address to setup gart page table for vram accessXiaogang Chen
Use dst input parameter to setup gart page table entries instead of using fixed location. Fixes: 237d623ae659 ("drm/amdgpu/gart: Add helper to bind VRAM pages (v2)") Signed-off-by: Xiaogang Chen <xiaogang.chen@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit ca5d4db8db843be7ed35fc9334737490c2b58d32)
2026-01-14Revert duplicate "drm/amdgpu: disable peer-to-peer access for DCC-enabled ↵Peter Colberg
GC12 VRAM surfaces" This reverts commit 22a36e660d014925114feb09a2680bb3c2d1e279 once, which was merged twice due to an incorrect backmerge resolution. Fixes: ce0478b02ed2 ("Merge tag 'v6.18-rc6' into drm-next") Signed-off-by: Peter Colberg <pcolberg@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 38a0f4cf8c6147fd10baa206ab349f8ff724e391)
2026-01-14drm/amd: Clean up kfd node on surprise disconnectMario Limonciello (AMD)
When an eGPU is unplugged the KFD topology should also be destroyed for that GPU. This never happens because the fini_sw callbacks never get to run. Run them manually before calling amdgpu_device_ip_fini_early() when a device has already been disconnected. This location is intentionally chosen to make sure that the kfd locking refcount doesn't get incremented unintentionally. Cc: kent.russell@amd.com Closes: https://community.frame.work/t/amd-egpu-on-linux/8691/33 Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org> Reviewed-by: Kent Russell <kent.russell@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 6a23e7b4332c10f8b56c33a9c5431b52ecff9aab) Cc: stable@vger.kernel.org
2026-01-14drm/amdgpu: fix drm panic null pointer when driver not support atomicLu Yao
When driver not support atomic, fb using plane->fb rather than plane->state->fb. Fixes: fe151ed7af54 ("drm/amdgpu: add generic display panic helper code") Signed-off-by: Lu Yao <yaolu@kylinos.cn> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit 2f2a72de673513247cd6fae14e53f6c40c5841ef)
generated by cgit v1.2.3 (git 2.0.4) at 2026-04-14 11:37:21 +0000