Pull drm next fixes from Dave Airlie:
"This is the first of two fixes for the merge PRs, the other is based
on 7.0 branch. This mostly AMD fixes, a couple of weeks of backlog
built up and this weeks. The main complaint I've seen is some boot
warnings around the FP code handling which this should fix. Otherwise
a single rcar-du and a single i915 fix.
amdgpu:
- SMU 14 fixes
- Partition fixes
- SMUIO 15.x fix
- SR-IOV fixes
- JPEG fix
- PSP 15.x fix
- NBIF fix
- Devcoredump fixes
- DPC fix
- RAS fixes
- Aldebaran smu fix
- IP discovery fix
- SDMA 7.1 fix
- Runtime pm fix
- MES 12.1 fix
- DML2 fixes
- DCN 4.2 fixes
- YCbCr fixes
- Freesync fixes
- ISM fixes
- Overlay cursor fix
- DC FP fixes
- UserQ locking fixes
- DC idle state manager fix
- ASPM fix
- GPUVM SVM fix
- DCE 6 fix
amdkfd:
- Fix memory clear handling
- num_of_nodes bounds check fix
i915:
- Fix uninitialized variable in the alignment loop [psr]
rcar-du:
- fix NULL-ptr crash"
* tag 'drm-next-2026-04-24' of https://gitlab.freedesktop.org/drm/kernel: (75 commits)
drm/amdkfd: Add upper bound check for num_of_nodes
drm: rcar-du: Fix crash when no CMM is available
drm/amd/display: Disable 10-bit truncation and dithering on DCE 6.x
drm/amdgpu: OR init_pte_flags into invalid leaf PTE updates
drm/amd: Adjust ASPM support quirk to cover more Intel hosts
drm/amd/display: Undo accidental fix revert in amdgpu_dm_ism.c
drm/i915/psr: Init variable to avoid early exit from et alignment loop
drm/amdgpu: drop userq fence driver refs out of fence process()
drm/amdgpu/userq: unpin and unref doorbell and wptr outside mutex
drm/amdgpu/userq: use pm_runtime_resume_and_get and fix err handling
drm/amdgpu/userq: unmap_helper dont return the queue state
drm/amdgpu/userq: unmap is to be called before freeing doorbell/wptr bo
drm/amdgpu/userq: hold root bo lock in caller of input_va_validate
drm/amdgpu/userq: caller to take reserv lock for vas_list_cleanup
drm/amdgpu/userq: create_mqd does not need userq_mutex
drm/amdgpu/userq: dont lock root bo with userq_mutex held
drm/amdgpu/userq: fix kerneldoc for amdgpu_userq_ensure_ev_fence
drm/amdgpu/userq: clean the VA mapping list for failed queue creation
drm/amdgpu/userq: avoid uneccessary locking in amdgpu_userq_create
drm/amd/display: Fix ISM teardown crash from NULL dc dereference
...
GFX V11 has GC block as default off IP.
Every time AMDGPU driver sends a request to PMFW
to unload MP1, PMFW will put GC in reset and
power down the voltage.Hence, skipping reset
for APUs with GFX V11 or later to avoid reset
related failures.
Fixes: 34355e6183 ("drm/amdgpu: Fix GFX hang on SteamDeck when amdgpu is reloaded")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Shubhankar Milind Sardeshpande <Shubhankar.MilindSardeshpande@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit d0a8cadffc)
Cc: stable@vger.kernel.org
According to our documentation, the RMA should only occur when the
threshold has been exceeded, not met.
Fixes: 5028a24aa8 ("drm/amdgpu: Send applicable RMA CPERs at end of RAS init")
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>
(cherry picked from commit 8bc09a7d0e)
svm_range_restore_pages might reserve the root bo so it must
be called after unreserving it.
Fixes: 1b135c6da0 ("drm/amdgpu: extract amdgpu_vm_lock_by_pasid from amdgpu_vm_handle_fault")
Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 5cdc219fe8)
This commit fixes amdgpu to work on the Radeon HD 7870 XT
which has never worked with the Linux open source drivers before.
Some boards have "harvested" chips, meaning that some parts of
the chip are disabled and fused, and it's sold for cheaper and
under a different marketing name.
On a harvested chip, any of the following can be disabled:
- CUs (Compute Units)
- RBs (Render Backend, aka. ROP)
- Memory channels (ie. the chip has a lower bandwidth)
- TCCs (ie. less L2 cache)
Handle chips with harvested TCCs by patching the registers
that configure how TCCs are mapped.
If some TCCs are disabled, we need to make sure that
the disabled TCCs are not used, and the remaining TCCs
are used optimally.
TCP_CHAN_STEER_LO/HI control which TCC is used by TCP channels.
TCP_ADDR_CONFIG.NUM_TCC_BANKS controls how many channels are used.
Note that the TCC configuration is highly relevant to performance.
Suboptimal configuration (eg. CHAN_STEER=0) can significantly
reduce gaming performance.
For optimal performance:
- Rely on the CHAN_STEER from the golden registers table,
only skip disabled TCCs but keep the mapping order.
- Limit NUM_TCC_BANKS to number of active TCCs to avoid thrashing,
which performs better than using the same TCC twice.
v2:
- Also consider CGTS_USER_TCC_DISABLE for disabled TCCs.
Link: https://bugs.freedesktop.org/show_bug.cgi?id=60879
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/2664
Fixes: 2cd46ad223 ("drm/amdgpu: add graphic pipeline implementation for si v8")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 00218d1552)
UVD 3.1 firmware validation seems to always fail after
attempting it when it had already been validated.
(This works similarly with the VCE 1.0 as well.)
Don't attempt repeating the validation when it's already done.
This caused issues in situations when the system isn't able
to suspend the GPU properly and so the GPU isn't actually
powered down. Then amdgpu would fail when calling the IP
block resume function.
Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/2887
Fixes: bb7978111d ("drm/amdgpu: fix SI UVD firmware validate resume fail")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 889a2cfd88)
There were multiple issues in that code.
First of all the order between the reset semaphore and the mm_lock was
wrong (e.g. copy_to_user) was called while holding the lock.
Then we allocated memory while holding the reset semaphore which is also
a pretty big bug and can deadlock.
Then we used down_read_trylock() instead of waiting for the reset to
finish.
Signed-off-by: Christian König <christian.koenig@amd.com>
Fixes: 9e823f3070 ("drm/amdgpu: Block MMR_READ IOCTL in reset")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 361b6e6b30)
When the GART placement is set to AMDGPU_GART_PLACEMENT_LOW:
Make sure that GART does not overlap with VRAM when
VRAM is configured to be in the low address space.
Solve this according to the following logic:
- When GART fits before VRAM, use zero address for GART
- Otherwise, put GART after the end of VRAM, aligned to 4 GiB
Previously, I had assumed this was not possible
so it was OK to not handle it, but now we got a report
from a user who has a board that is configured this way.
Fixes: 917f91d8d8 ("drm/amdgpu/gmc: add a way to force a particular placement for GART")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 3d9de5d86a)
When compiled without CONFIG_HSA_AMD KFD will return -ENOENT.
As other errors will cause KFD functionality issues this is the
only error code that should be ignored at init.
Reviewed-by: Kent Russell <kent.russell@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 4259a25341)
When new_addition is true, amdgpu_userq_vm_validate() calls
drm_exec_fini(&exec) before iterating over the collected HMM ranges and
calling amdgpu_ttm_tt_get_user_pages().
If amdgpu_ttm_tt_get_user_pages() fails in that path, the code jumps to
unlock_all and calls drm_exec_fini(&exec) a second time on the same
exec object. drm_exec_fini() is not idempotent: it frees exec->objects
and may also drop exec->contended and finalize the ww acquire context.
Route that error path directly to the range cleanup once exec has
already been finalized.
Fixes: 42f1487884 ("drm/amdgpu/userqueue: validate userptrs for userqueues")
Issue found using a prototype static analysis tool
and confirmed by code review.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Hongyan Xu <getshell@seu.edu.cn>
Signed-off-by: Slavin Liu <220245772@seu.edu.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 2802952e4a)
RDNA4 (GFX 12) hardware removes the GDS, GWS, and OA on-chip memory
resources. The gfx_v12_0 initialisation code correctly leaves
adev->gds.gds_size, adev->gds.gws_size, and adev->gds.oa_size at
zero to reflect this.
amdgpu_ttm_init() unconditionally calls amdgpu_ttm_init_on_chip() for
each of these resources regardless of size. When the size is zero,
amdgpu_ttm_init_on_chip() forwards the call to ttm_range_man_init(),
which calls drm_mm_init(mm, 0, 0). drm_mm_init() immediately fires
DRM_MM_BUG_ON(start + size <= start) -- trivially true when size is
zero -- crashing the kernel during modprobe of amdgpu on an RX 9070 XT.
Guard against this by returning 0 early from
amdgpu_ttm_init_on_chip() when size_in_page is zero. This skips TTM
resource manager registration for hardware resources that are absent,
without affecting any other GPU type.
DRM_MM_BUG_ON() only asserts if CONFIG_DRM_DEBUG_MM is enabled in
the kernel config. This is apparently rarely enabled as these chips
have been in the market for over a year and this issue was only reported
now.
Link: https://lore.kernel.org/all/bug-221376-2300@https.bugzilla.kernel.org%2F/
Link: https://bugzilla.kernel.org/show_bug.cgi?id=221376
Oops-Analysis: http://oops.fenrus.org/reports/bugzilla.korg/221376/report.html
Assisted-by: GitHub Copilot:Claude Sonnet 4.6 linux-kernel-oops-x86.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: amd-gfx@lists.freedesktop.org
Cc: dri-devel@lists.freedesktop.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 5719ce5865)
Cc: stable@vger.kernel.org
Pull more drm updates from Dave Airlie:
"This is a followup which is mostly next material with some fixes.
Alex pointed out I missed one of his AMD MRs from last week, so I
added that, then Jani sent the pipe reordering stuff, otherwise it's
just some minor i915 fixes and a dma-buf fix.
drm:
- Add support for AMD VSDB parsing to drm_edid
dma-buf:
- fix documentation formatting
i915:
- add support for reordered pipes to support joined pipes better
- Fix VESA backlight possible check condition
- Verify the correct plane DDB entry
amdgpu:
- Audio regression fix
- Use drm edid parser for AMD VSDB
- Misc cleanups
- VCE cs parse fixes
- VCN cs parse fixes
- RAS fixes
- Clean up and unify vram reservation handling
- GPU Partition updates
- system_wq cleanups
- Add CONFIG_GCOV_PROFILE_AMDGPU kconfig option
- SMU vram copy updates
- SMU 13/14/15 fixes
- UserQ fixes
- Replace pasid idr with an xarray
- Dither handling fix
- Enable amdgpu by default for CIK APUs
- Add IBs to devcoredump
amdkfd:
- system_wq cleanups
radeon:
- system_wq cleanups"
* tag 'drm-next-2026-04-22' of https://gitlab.freedesktop.org/drm/kernel: (62 commits)
drm/i915/display: change pipe allocation order for discrete platforms
drm/i915/wm: Verify the correct plane DDB entry
drm/i915/backlight: Fix VESA backlight possible check condition
drm/i915: Walk crtcs in pipe order
drm/i915/joiner: Make joiner "nomodeset" state copy independent of pipe order
dma-buf: fix htmldocs error for dma_buf_attach_revocable
drm/amdgpu: dump job ibs in the devcoredump
drm/amdgpu: store ib info for devcoredump
drm/amdgpu: extract amdgpu_vm_lock_by_pasid from amdgpu_vm_handle_fault
drm/amdgpu: Use amdgpu by default for CIK APUs too
drm/amd/display: Remove unused NUM_ELEMENTS macros
drm/amd/display: Replace inline NUM_ELEMENTS macro with ARRAY_SIZE
drm/amdgpu: save ring content before resetting the device
drm/amdgpu: make userq fence_drv drop explicit in queue destroy
drm/amdgpu: rework userq fence driver alloc/destroy
drm/amdgpu/userq: use dma_fence_wait_timeout without test for signalled
drm/amdgpu/userq: call dma_resv_wait_timeout without test for signalled
drm/amdgpu/userq: add the return code too in error condition
drm/amdgpu/userq: fence wait for max time in amdgpu_userq_wait_for_signal
drm/amd/display: Change dither policy for 10 bpc output back to dithering
...
Invalid leaf clears that only set AMDGPU_PTE_EXECUTABLE match the old
GMC9 fault-priority workaround but omit adev->gmc.init_pte_flags.
On GFX12 that includes AMDGPU_PTE_IS_PTE; without it, some cleared
PTEs can fault as no-retry and bypass the SVM/XNACK handler when a
VA is reused after a BO unmap.
Apply init_pte_flags in amdgpu_vm_pte_update_flags() alongside
EXECUTABLE so range-driven clears (e.g. amdgpu_vm_clear_freed) match
amdgpu_vm_pt_clear() for leaf templates.
Signed-off-by: Siwei He <siwei.he@amd.com>
Reviewed-by: Philip Yang <philip.yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 9d47b2c36b)
amdgpu_userq_wait_ioctl() takes extra references on waited-on fence
drivers and stores them in waitq->fence_drv_xa. When a new userq fence is
created, those references are transferred into userq_fence->fence_drv_array
so they can be released when the fence completes.
However, those inherited references are currently only dropped from
amdgpu_userq_fence_driver_process(). If a fence never reaches that path,
such as it is already signaled when created, so we need to explicitly release
those fences in that case.
v2: use a list(list_cut_before) for managing the signal userq driver fences.(Christian)
Link: https://patchwork.freedesktop.org/patch/718078/?series=164763&rev=2
v3: Doesn't cache the userq first unsignaled fence and use the cut before list
head directly.(Christian)
Cc: Alex Deucher <alexander.deucher@amd.com>
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>
In amdgpu_userq_destroy once unmap_helpder is called within mutex
there is no need to hold mutex.
This helps in avoiding a deadlock between doorbell and wptr ww mutex
and we could unpin and unref these bos outside mutex safely.
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>
Use pm_runtime_resume_and_get instead of pm_runtime_get_sync as it
return error but put the reference in the function itself.
In goto statements we need to drop the pm reference too.
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We check for return value of amdgpu_userq_unmap_helper and
compare it against the queue->state which is logically
wrong and we should just check for failure and do the needfull.
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>
Unmap the queue after freeing doorbell and wptr memory is completely
wrong. Any operation on the queue needs the doorbell and wptr to be
valid and hence fixing the ordering.
Also since we are using amdgpu_bo_reserve in non interruptrable mode
so there is no need to check for its return values.
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>
In function amdgpu_userq_buffer_vas_list_cleanup, remove the
reservation lock for vm and caller should make sure it's taken
before locking userq_mutex.
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>
Reshuffle the code to run create_mqd outside the mutex.
code here is mostly setting up software structure init
before actually registering the userqueue in the xa and
to the driver.
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>
Do not hold reservation lock for root bo if userq_mutex
is already held in the call flow this cause a lock
issue with ttm_bo_delayed_delete.
Its better to lock the vm->root.bo first and then go ahead
with userq_mutex so userq_mutex threads dont get stuck until
the reservation lock is held.
In this case it helps in the function amdgpu_userq_buffer_vas_mapped
for each queue during restore_all.
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>
If the queue creation failed during mapping of the important VA's
like queue_va, rptr_va and wptr_va. These needs to be cleaned
as queue destroy will not be called for such queues as user never
get call to creation failure.
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
