drm/amdgpu: Initialize data to NULL in imu_v12_0_program_rlc_ram()

After a recent change in clang to expose uninitialized warnings from
const variables and pointers [1], there is a warning in
imu_v12_0_program_rlc_ram() because data is passed uninitialized to
program_imu_rlc_ram():

  drivers/gpu/drm/amd/amdgpu/imu_v12_0.c:374:30: error: variable 'data' is uninitialized when used here [-Werror,-Wuninitialized]
    374 |                         program_imu_rlc_ram(adev, data, (const u32)size);
        |                                                   ^~~~

As this warning happens early in clang's frontend, it does not realize
that due to the assignment of r to -EINVAL, program_imu_rlc_ram() is
never actually called, and even if it were, data would not be
dereferenced because size is 0.

Just initialize data to NULL to silence the warning, as the commit that
added program_imu_rlc_ram() mentioned it would eventually be used over
the old method, at which point data can be properly initialized and
used.

Cc: stable@vger.kernel.org
Closes: https://github.com/ClangBuiltLinux/linux/issues/2107
Fixes: 56159fffaab5 ("drm/amdgpu: use new method to program rlc ram")
Link: https://github.com/llvm/llvm-project/commit/2464313eef01c5b1edf0eccf57a32cdee01472c7
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix divide by zero when calculating min ODM factor

[WHY&HOW]
If the debug option is set to disable_dsc the max slice width and/or
dispclk can be zero. This causes a divide by zero when calculating the
min ODM combine factor. Add a check to ensure they are valid first.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org

Merge tag 'nova-next-v6.17-2025-07-18' of https://gitlab.freedesktop.org/drm/nova into drm-next

Nova changes for v6.17

DMA:

  - Merge topic/dma-features-2025-06-23 from alloc tree.

    - Clarify wording and be consistent in 'coherent' nomenclature.

    - Convert the read!() / write!() macros to return a Result.

    - Add as_slice() / write() methods in CoherentAllocation.

    - Fix doc-comment of dma_handle().

    - Expose count() and size() in CoherentAllocation and add the
      corresponding type invariants.

    - Implement CoherentAllocation::dma_handle_with_offset().

nova-core:

  - Various register!() macro improvements.

  - Custom Sleep / Delay helpers (until the actual abstractions land).

  - Add DMA object abstraction.

  - VBIOS

    - Image parser / iterator.

    - PMU table look up in FWSEC.

    - FWSEC ucode extraction.

  - Register sysmem flush page.

  - Falcon

    - Generic falcon boot code and HAL (Ampere).

    - GSP / SEC2 specific code.

  - FWSEC-FRTS

    - Compute layout of FRTS region (FbLayout and HAL).

    - Load into GSP falcon and execute.

  - Add Documentation for VBIOS layout, Devinit process, Fwsec operation
    and layout, Falcon basics.

  - Update and annotate TODO list.

  - Add Alexandre Courbot as co-maintainer.

Rust:

  - Make ETIMEDOUT error available.

  - Add size constants up to SZ_2G.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: "Danilo Krummrich" <dakr@kernel.org>
Link: https://lore.kernel.org/r/DBFKLDMUGZD9.Z93GN2N5B0FI@kernel.org

Merge tag 'amd-drm-next-6.17-2025-07-17' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.17-2025-07-17:

amdgpu:
- Partition fixes
- Reset fixes
- RAS fixes
- i2c fix
- MPC updates
- DSC cleanup
- EDID fixes
- Display idle D3 update
- IPS updates
- DMUB updates
- Retimer fix
- Replay fixes
- Fix DC memory leak
- Initial support for smartmux
- DCN 4.0.1 degamma LUT fix
- Per queue reset cleanups
- Track ring state associated with a fence
- SR-IOV fixes
- SMU fixes
- Per queue reset improvements for GC 9+ compute
- Per queue reset improvements for GC 10+ gfx
- Per queue reset improvements for SDMA 5+
- Per queue reset improvements for JPEG 2+
- Per queue reset improvements for VCN 2+
- GC 8 fix
- ISP updates

amdkfd:
- Enable KFD on LoongArch

radeon:
- Drop console lock during suspend/resume

UAPI:
- Add userq slot info to INFO IOCTL
  Used for IGT userq validation tests (https://lists.freedesktop.org/archives/igt-dev/2025-July/093228.html)

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250717213827.2061581-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>

Merge tag 'drm-misc-next-2025-07-17' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

drm-misc-next for 6.17:

UAPI Changes:

Cross-subsystem Changes:

Core Changes:

- mode_config: Change fb_create prototype to pass the drm_format_info
  and avoid redundant lookups in drivers
- sched: kunit improvements, memory leak fixes, reset handling
  improvements
- tests: kunit EDID update

Driver Changes:

- amdgpu: Hibernation fixes, structure lifetime fixes
- nouveau: sched improvements
- sitronix: Add Sitronix ST7567 Support

- bridge:
  - Make connector available to bridge detect hook

- panel:
  - More refcounting changes
  - New panels: BOE NE14QDM

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maxime Ripard <mripard@redhat.com>
Link: https://lore.kernel.org/r/20250717-efficient-kudu-of-fantasy-ff95e0@houat

Merge tag 'drm-xe-next-2025-07-15' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-next

Driver Changes:
 - Create and use XE_DEVICE_WA infrastructure (Atwood)
 - SRIOV: Mark BMG as SR-IOV capable (Michal)
 - Dont skip TLB invalidations on VF (Tejas)
 - Fix migration copy direction in access_memory (Auld)
 - General code clean-up (Lucas, Brost, Dr. David, Xin)
 - More missing XeLP workarounds (Tvrtko)
 - SRIOV: Relax VF/PF version negotiation (Michal)
 - SRIOV: LMTT invalidation (Michal)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/aHacDvF9IaVHI61C@intel.com

drm/amdgpu: move reset support type checks into the caller

Rather than checking in the callbacks, check if the reset
type is supported in the caller.

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sdma7: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sdma6: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sdma5.2: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/sdma5: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx12: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.
Drop the soft_recovery callbacks as the queue reset replaces
it.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx11: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.
Drop the soft_recovery callbacks as the queue reset replaces
it.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx10: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.
Drop the soft_recovery callbacks as the queue reset replaces
it.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx9.4.3: re-emit unprocessed state on kcq reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx9: re-emit unprocessed state on kcq reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add WARN_ON to the resource clear function

Set the dirty bit when the memory resource is not cleared
during BO release.

v2(Christian):
  - Drop the cleared flag set to false.
  - Improve the amdgpu_vram_mgr_set_clear_state() function.

v3:
  - Add back the resource clear flag set function call after
    being cleared during eviction (Christian).
  - Modified the patch subject name.

Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Use cached metrics data on SMUv13.0.6

Cached metrics data validity is 1ms on SMUv13.0.6 SOCs. It's not
reasonable for any client to query gpu_metrics at a faster rate and
constantly interrupt PMFW.

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>

drm/amd/pm: Use cached data for min/max clocks

If dpm tables are already populated on SMU v13.0.6 SOCs, use the cached
data. Otherwise, fetch values from firmware.

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>

gpu: nova-core: fix bounds check in PmuLookupTableEntry::new

data is sliced from 2..6, but the bounds check data.len() < 5
does not satisfy those bounds.

Fixes: 47c4846e4319 ("gpu: nova-core: vbios: Add support for FWSEC ucode extraction")
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Joel Fernandes <joelagnelf@nvidia.com>
Signed-off-by: Rhys Lloyd <krakow20@gmail.com>
Link: https://lore.kernel.org/r/20250713025108.9364-2-krakow20@gmail.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>

drm/amdgpu: Replace HQD terminology with slots naming

The term "HQD" is CP-specific and doesn't
accurately describe the queue resources for other IP blocks like SDMA,
VCN, or VPE. This change:

1. Renames `num_hqds` to `num_slots` in amdgpu_kms.c to better reflect
   the generic nature of the resource counting
2. Updates the UAPI struct member from `userq_num_hqds` to `userq_num_slots`
3. Maintains the same functionality while using more appropriate terminology

Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add user queue instance count in HW IP info

This change exposes the number of available user queue instances
for each hardware IP type (GFX, COMPUTE, SDMA) through the
drm_amdgpu_info_hw_ip interface.

Key changes:
1. Added userq_num_instance field to drm_amdgpu_info_hw_ip structure
2. Implemented counting of available HQD slots using:
   - mes.gfx_hqd_mask for GFX queues
   - mes.compute_hqd_mask for COMPUTE queues
   - mes.sdma_hqd_mask for SDMA queues
3. Only counts available instances when user queues are enabled
   (!disable_uq)

v2: using the adev->mes.gfx_hqd_mask[]/compute_hqd_mask[]/sdma_hqd_mask[] masks
  to determine the number of queue slots available for each engine type (Alex)
v3: rename userq_num_instance to userq_num_hqds (Alex)

Suggested-by: Alex Deucher <alexander.deucher@amd.com>
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>

drm/amd/amdgpu: Add helper functions for isp buffers

Accessing amdgpu internal data structures "struct amdgpu_device"
and "struct amdgpu_bo" in ISP V4L2 driver to alloc/free GART
buffers is not recommended.

Add new amdgpu_isp helper functions that takes opaque params
from ISP V4L2 driver and calls the amdgpu internal functions
amdgpu_bo_create_isp_user() and amdgpu_bo_create_kernel() to
alloc/free GART buffers.

Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/amdgpu: Initialize swnode for ISP MFD device

Create amd_isp_capture MFD device with swnode initialized to
isp specific software_node part of fwnode graph in amd_isp4
x86/platform driver. The isp driver use this swnode handle
to retrieve the critical properties (data-lanes, mipi phyid,
link-frequencies etc.) required for camera to work on AMD ISP4
based targets.

Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/gfx8: reset compute ring wptr on the GPU on resume

Commit 42cdf6f687da ("drm/amdgpu/gfx8: always restore kcq MQDs") made the
ring pointer always to be reset on resume from suspend. This caused compute
rings to fail since the reset was done without also resetting it for the
firmware. Reset wptr on the GPU to avoid a disconnect between the driver
and firmware wptr.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3911
Fixes: 42cdf6f687da ("drm/amdgpu/gfx8: always restore kcq MQDs")
Signed-off-by: Eeli Haapalainen <eeli.haapalainen@protonmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: fix null pointer access

Writing a string without delimiters (' ', '\n', '\0') to the under
gpu_od/fan_ctrl sysfs or pp_power_profile_mode for the CUSTOM profile
will result in a null pointer dereference.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4401
Signed-off-by: Umio Yasuno <coelacanth_dream@protonmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg: clean up reset type handling

Make the handling consistent with other IPs and across
JPEG versions.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: rework gmc_v9_0_get_coherence_flags v2

Avoid using the mapping here.

v2: use amdgpu_xgmi_same_hive() as suggested by Felix

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn3: implement ring reset

Use the new helpers to handle engine resets for VCN.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn2.5: implement ring reset

Use the new helpers to handle engine resets for VCN.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn2: implement ring reset

Use the new helpers to handle engine resets for VCN.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn: add a helper framework for engine resets

With engine resets we reset all queues on the engine rather
than just a single queue.  Add a framework to handle this
similar to SDMA.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn5: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn4.0.5: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn4.0.3: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn4: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg5.0.1: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg5: add queue reset

Add queue reset support for jpeg 5.0.0.
Use the new helpers to re-emit the unprocessed state
after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg4.0.5: add queue reset

Add queue reset support for jpeg 4.0.5.
Use the new helpers to re-emit the unprocessed state
after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg4.0.3: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg4: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg3: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg2.5: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg2: re-emit unprocessed state on ring reset

Re-emit the unprocessed state after resetting the queue.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Tested-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Remove unnecessary variable

Remove unnecessary variable ret from smu_v13_0_12_get_smu_metrics_data

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202507150618.WOfvWsQF-lkp@intel.com
Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Increase reset counter only on success

Increment the reset counter only if soft recovery succeeded. This is
consistent with a ring hard reset behaviour where counter gets
incremented only if hard reset succeeded.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Get max/min frequency on aldebaran VF

PMFW interface to get max/min frequencies is not available on aldebaran
VFs. Use data, if available, in DPM tables to get the max/min
frequencies.

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>

drm/amdgpu: make compute timeouts consistent

For kernel compute queues, align the timeout with
other kernel queues (10 sec).  This had previously
been set higher for OpenCL when it used kernel
queues, but now OpenCL uses KFD user queues which
don't have a timeout limitation. This also aligns
with SR-IOV which already used a shorter timeout.

Additionally the longer timeout negatively impacts
the user experience with kernel queues for interactive
applications.

Reviewed-by: Kent Russell <kent.russell@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Check SQ_CONFIG register support on SRIOV

On SRIOV environments, check if RLCG supports
SQ_CONFIG register programming.

Signed-off-by: Tony Yi <Tony.Yi@amd.com>
Reviewed-by: Zhigang Luo <zhigang.luo@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: track ring state associated with a fence

We need to know the wptr and sequence number associated
with a fence so that we can re-emit the unprocessed state
after a ring reset.  Pre-allocate storage space for
the ring buffer contents and add helpers to save off
and re-emit the unprocessed state so that it can be
re-emitted after the queue is reset.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: clean up GC reset functions

Make them consistent and use the reset flags.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: clean up jpeg reset functions

Make them consistent and use the reset flags.

Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn: don't enable per queue resets on SR-IOV

Power control is only available in bare metal.  SR-IOV
will need a different method.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/jpeg4: add additional ring reset error checking

Start and stop can fail, so add checks.

Fixes: 74894ffc7d0c ("drm/amdgpu: Add ring reset callback for JPEG4_0_0")
Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Sathishkumar S <sathishkumar.sundararaju@amd.com>

drm/amdgpu/jpeg3: add additional ring reset error checking

Start and stop can fail, so add checks.

Fixes: 03399d0bff25 ("drm/amdgpu: Add ring reset callback for JPEG3_0_0")
Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Sathishkumar S <sathishkumar.sundararaju@amd.com>

drm/amdgpu/jpeg2: add additional ring reset error checking

Start and stop can fail, so add checks.

Fixes: 500c04d2a708 ("drm/amdgpu: Add ring reset callback for JPEG2_0_0")
Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: Sathishkumar S <sathishkumar.sundararaju@amd.com>

drm/amdgpu: clean up sdma reset functions

Make them consistent and drop unneeded extra variables.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix missing unlocking in an error path in amdgpu_userq_create()

If kasprintf() fails, some mutex still need to be released to avoid locking
issue, as already done in all other error handling path.

Fixes: c03ea34cbf88 ("drm/amdgpu: add support of debugfs for mqd information")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Link: https://lore.kernel.org/all/366557fa7ca8173fd78c58336986ca56953369b9.1752087753.git.christophe.jaillet@wanadoo.fr/
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm: Make passing of format info to drm_helper_mode_fill_fb_struct() mandatory

Now that everyone passes along the format info to
drm_helper_mode_fill_fb_struct() we can make this behaviour
mandatory and drop the extra lookup.

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-20-ville.syrjala@linux.intel.com

drm/vmwgfx: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

Cc: Zack Rusin <zack.rusin@broadcom.com>
Cc: Broadcom internal kernel review list <bcm-kernel-feedback-list@broadcom.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-19-ville.syrjala@linux.intel.com

drm/virtio: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

Cc: David Airlie <airlied@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Cc: Gurchetan Singh <gurchetansingh@chromium.org>
Cc: Chia-I Wu <olvaffe@gmail.com>
Cc: virtualization@lists.linux.dev
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-18-ville.syrjala@linux.intel.com

drm/tegra: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

For the fbdev case a manual drm_get_format_info() lookup
is needed.

Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Mikko Perttunen <mperttunen@nvidia.com>
Cc: linux-tegra@vger.kernel.org
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-17-ville.syrjala@linux.intel.com

drm/msm: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookups.

For the fbdev case a manual drm_get_format_info() lookup
is needed.

Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
Cc: Sean Paul <sean@poorly.run>
Cc: Marijn Suijten <marijn.suijten@somainline.org>
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-16-ville.syrjala@linux.intel.com

drm/komeda: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

Cc: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-15-ville.syrjala@linux.intel.com

drm/i915: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

For the fbdev case a manual drm_get_format_info() lookup
is needed.

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-14-ville.syrjala@linux.intel.com

drm/gma500: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

For the fbdev case a manual drm_get_format_info() lookup
is needed.

Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-13-ville.syrjala@linux.intel.com

drm/exynos: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

For the fbdev case a manual drm_get_format_info() lookup
is needed.

Cc: Inki Dae <inki.dae@samsung.com>
Cc: Seung-Woo Kim <sw0312.kim@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-12-ville.syrjala@linux.intel.com

drm/armada: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

For the fbdev case a manual drm_get_format_info() lookup
is needed.

Cc: Russell King <linux@armlinux.org.uk>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-11-ville.syrjala@linux.intel.com

drm/amdgpu: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Plumb the format info from .fb_create() all the way to
drm_helper_mode_fill_fb_struct() to avoid the redundant
lookup.

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: amd-gfx@lists.freedesktop.org
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-10-ville.syrjala@linux.intel.com

drm/gem/afbc: Eliminate redundant drm_get_format_info()

Pass along the format info from .fb_create() to aliminate the
redundant drm_get_format_info() calls from the afbc code.

Cc: Sandy Huang <hjc@rock-chips.com>
Cc: "Heiko Stübner" <heiko@sntech.de>
Cc: Andy Yan <andy.yan@rock-chips.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-9-ville.syrjala@linux.intel.com

drm/gem: Pass along the format info from .fb_create() to drm_helper_mode_fill_fb_struct()

Pass along the format info from .fb_create() to eliminate the
redundant drm_get_format_info() calls from the gem fb code.

v2: Fix kernel docs (Laurent)

Cc: Dave Airlie <airlied@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Sandy Huang <hjc@rock-chips.com>
Cc: "Heiko Stübner" <heiko@sntech.de>
Cc: Andy Yan <andy.yan@rock-chips.com>
Cc: Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
Cc: virtualization@lists.linux.dev
Cc: spice-devel@lists.freedesktop.org
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-8-ville.syrjala@linux.intel.com

drm/malidp: Pass along the format info from .fb_create() malidp_verify_afbc_framebuffer_size()

Plumb the format info from .fb_create() all the way to
malidp_verify_afbc_framebuffer_size() to avoid the
redundant lookup.

Cc: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-7-ville.syrjala@linux.intel.com

drm: Allow the caller to pass in the format info to drm_helper_mode_fill_fb_struct()

Soon all drivers should have the format info already available in the
places where they call drm_helper_mode_fill_fb_struct(). Allow it to
be passed along into drm_helper_mode_fill_fb_struct() instead of doing
yet another redundant lookup.

Start by always passing in NULL and still doing the extra lookup.
The actual changes to avoid the lookup will follow.

Done with cocci (with some manual fixups):
@@
identifier dev, fb, mode_cmd;
expression get_format_info;
@@
void drm_helper_mode_fill_fb_struct(struct drm_device *dev,
                                    struct drm_framebuffer *fb,
+                                    const struct drm_format_info *info,
                                    const struct drm_mode_fb_cmd2 *mode_cmd)
{
...
- fb->format = get_format_info;
+ fb->format = info ?: get_format_info;
...
}

@@
identifier dev, fb, mode_cmd;
@@
void drm_helper_mode_fill_fb_struct(struct drm_device *dev,
                                    struct drm_framebuffer *fb,
+                                    const struct drm_format_info *info,
                                    const struct drm_mode_fb_cmd2 *mode_cmd);

@@
expression dev, fb, mode_cmd;
@@
drm_helper_mode_fill_fb_struct(dev, fb
+        ,NULL
       ,mode_cmd);

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Liviu Dudau <liviu.dudau@arm.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Seung-Woo Kim <sw0312.kim@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
Cc: Sean Paul <sean@poorly.run>
Cc: Marijn Suijten <marijn.suijten@somainline.org>
Cc: Lyude Paul <lyude@redhat.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Mikko Perttunen <mperttunen@nvidia.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Cc: Gurchetan Singh <gurchetansingh@chromium.org>
Cc: Chia-I Wu <olvaffe@gmail.com>
Cc: Zack Rusin <zack.rusin@broadcom.com>
Cc: Broadcom internal kernel review list <bcm-kernel-feedback-list@broadcom.com>
Cc: amd-gfx@lists.freedesktop.org
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: linux-tegra@vger.kernel.org
Cc: virtualization@lists.linux.dev
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-6-ville.syrjala@linux.intel.com

drm: Pass the format info to .fb_create()

Pass along the format information from the top to .fb_create()
so that we can avoid redundant (and somewhat expensive) lookups
in the drivers.

Done with cocci (with some manual fixups):
@@
identifier func =~ ".*create.*";
identifier dev, file, mode_cmd;
@@
struct drm_framebuffer *func(
       struct drm_device *dev,
       struct drm_file *file,
+      const struct drm_format_info *info,
       const struct drm_mode_fb_cmd2 *mode_cmd)
{
...
(
- const struct drm_format_info *info = drm_get_format_info(...);
|
- const struct drm_format_info *info;
...
- info = drm_get_format_info(...);
)
<...
- if (!info)
-    return ...;
...>
}

@@
identifier func =~ ".*create.*";
identifier dev, file, mode_cmd;
@@
struct drm_framebuffer *func(
       struct drm_device *dev,
       struct drm_file *file,
+      const struct drm_format_info *info,
       const struct drm_mode_fb_cmd2 *mode_cmd)
{
...
}

@find@
identifier fb_create_func =~ ".*create.*";
identifier dev, file, mode_cmd;
@@
struct drm_framebuffer *fb_create_func(
       struct drm_device *dev,
       struct drm_file *file,
+      const struct drm_format_info *info,
       const struct drm_mode_fb_cmd2 *mode_cmd);

@@
identifier find.fb_create_func;
expression dev, file, mode_cmd;
@@
fb_create_func(dev, file
+        ,info
       ,mode_cmd)

@@
expression dev, file, mode_cmd;
@@
drm_gem_fb_create(dev, file
+        ,info
       ,mode_cmd)

@@
expression dev, file, mode_cmd;
@@
drm_gem_fb_create_with_dirty(dev, file
+        ,info
       ,mode_cmd)

@@
expression dev, file_priv, mode_cmd;
identifier info, fb;
@@
info = drm_get_format_info(...);
...
fb = dev->mode_config.funcs->fb_create(dev, file_priv
+                                      ,info
                                       ,mode_cmd);

@@
identifier dev, file_priv, mode_cmd;
@@
struct drm_mode_config_funcs {
...
struct drm_framebuffer *(*fb_create)(struct drm_device *dev,
                                     struct drm_file *file_priv,
+                                     const struct drm_format_info *info,
                                     const struct drm_mode_fb_cmd2 *mode_cmd);
...
};

v2: Fix kernel docs (Laurent)
    Fix commit msg (Geert)

Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Liviu Dudau <liviu.dudau@arm.com>
Cc: Maxime Ripard <mripard@kernel.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Seung-Woo Kim <sw0312.kim@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Cc: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
Cc: Sean Paul <sean@poorly.run>
Cc: Marijn Suijten <marijn.suijten@somainline.org>
Cc: Marek Vasut <marex@denx.de>
Cc: Stefan Agner <stefan@agner.ch>
Cc: Lyude Paul <lyude@redhat.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Kieran Bingham <kieran.bingham+renesas@ideasonboard.com>
Cc: Biju Das <biju.das.jz@bp.renesas.com>
Cc: Sandy Huang <hjc@rock-chips.com>
Cc: "Heiko Stübner" <heiko@sntech.de>
Cc: Andy Yan <andy.yan@rock-chips.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Mikko Perttunen <mperttunen@nvidia.com>
Cc: Dave Stevenson <dave.stevenson@raspberrypi.com>
Cc: "Maíra Canal" <mcanal@igalia.com>
Cc: Raspberry Pi Kernel Maintenance <kernel-list@raspberrypi.com>
Cc: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Cc: Gurchetan Singh <gurchetansingh@chromium.org>
Cc: Chia-I Wu <olvaffe@gmail.com>
Cc: Zack Rusin <zack.rusin@broadcom.com>
Cc: Broadcom internal kernel review list <bcm-kernel-feedback-list@broadcom.com>
Cc: Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
Cc: amd-gfx@lists.freedesktop.org
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: virtualization@lists.linux.dev
Cc: spice-devel@lists.freedesktop.org
Cc: linux-renesas-soc@vger.kernel.org
Cc: linux-tegra@vger.kernel.org
Cc: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-5-ville.syrjala@linux.intel.com

drm: Look up the format info earlier

Look up the format info already in drm_internal_framebuffer_create()
so that we can later pass it along to .fb_create(). Currently various
drivers are doing additional lookups in their .fb_create()
implementations, and these lookups are rather expensive now (given
how many different pixel formats we have).

v2: Fix commit msg (Thomas)

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-4-ville.syrjala@linux.intel.com

drm: Pass pixel_format+modifier directly to drm_get_format_info()

Decouple drm_get_format_info() from struct drm_mode_fb_cmd2 and just
pass the pixel format+modifier combo in by hand.

We may want to use drm_get_format_info() outside of the normal
addfb paths where we won't have a struct drm_mode_fb_cmd2, and
creating a temporary one just for this seems silly.

Done with cocci:
@@
identifier dev, mode_cmd;
@@
struct drm_format_info *
drm_get_format_info(struct drm_device *dev,
-     const struct drm_mode_fb_cmd2 *mode_cmd
+     u32 pixel_format, u64 modifier
          )
{
<...
(
- mode_cmd->pixel_format
+ pixel_format
|
- mode_cmd->modifier[0]
+ modifier
)
...>
}

@@
identifier dev, mode_cmd;
@@
struct drm_format_info *
drm_get_format_info(struct drm_device *dev,
-     const struct drm_mode_fb_cmd2 *mode_cmd
+     u32 pixel_format, u64 modifier
          );

@@
expression dev, mode_cmd;
@@
- drm_get_format_info(dev, mode_cmd)
+ drm_get_format_info(dev, mode_cmd->pixel_format, mode_cmd->modifier[0])

v2: Fix kernel docs (Laurent)
    Drop drm_mode_fb_cmd2 forward declaration (Thomas)

Cc: Liviu Dudau <liviu.dudau@arm.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Inki Dae <inki.dae@samsung.com>
Cc: Seung-Woo Kim <sw0312.kim@samsung.com>
Cc: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Cc: Chun-Kuang Hu <chunkuang.hu@kernel.org>
Cc: Philipp Zabel <p.zabel@pengutronix.de>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
Cc: Sean Paul <sean@poorly.run>
Cc: Marijn Suijten <marijn.suijten@somainline.org>
Cc: Marek Vasut <marex@denx.de>
Cc: Stefan Agner <stefan@agner.ch>
Cc: Lyude Paul <lyude@redhat.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Sandy Huang <hjc@rock-chips.com>
Cc: "Heiko Stübner" <heiko@sntech.de>
Cc: Andy Yan <andy.yan@rock-chips.com>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Mikko Perttunen <mperttunen@nvidia.com>
Cc: linux-arm-msm@vger.kernel.org
Cc: freedreno@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Cc: linux-tegra@vger.kernel.org
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-3-ville.syrjala@linux.intel.com

drm: Pass pixel_format+modifier to .get_format_info()

Decouple .get_format_info() from struct drm_mode_fb_cmd2 and just
pass the pixel format+modifier combo in by hand.

We may want to use .get_format_info() outside of the normal
addfb paths where we won't have a struct drm_mode_fb_cmd2, and
creating a temporary one just for this seems silly.

v2: Fix intel_fb_get_format_info() docs (Laurent)

Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Rodrigo Siqueira <siqueira@igalia.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: amd-gfx@lists.freedesktop.org
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250701090722.13645-2-ville.syrjala@linux.intel.com

drm/tests: edid: Add edid-decode --check output

Some of our EDIDs are (rightfully) invalid, but most of them should be
valid.

Let's add the edid-decode --check of these EDIDs when they were
generated, so we know what to expect going forward, and a comment to
explicitly mention when we expect them to be broken.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250625-drm-update-edid-v1-3-2d963743ab9e@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/tests: edid: Update CTA-861 HDMI Vendor Specific Data Block

For some reason, the HDMI VSDBs in our kunit EDIDs had a length longer
than expected.

While this was harmless, we should get rid of it to make it somewhat
predictable.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250625-drm-update-edid-v1-2-2d963743ab9e@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/tests: edid: Fix monitor range limits

For some reason, some EDIDs used by kunit had Monitor Range Limits
making no sense, and not matching the edid-decode output in the comment.

While they were in the comments as:
    Display Range Limits:
      Monitor ranges (GTF): 50-70 Hz V, 30-70 kHz H, max dotclock 150 MHz

They were actually:
    Display Range Limits:
      Monitor ranges (GTF): 50-70 Hz V, 0-0 kHz H, max dotclock 1960 MHz

Fix that section of the EDIDs to match the expected edid-decode output.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250625-drm-update-edid-v1-1-2d963743ab9e@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/boe-himax8279d: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-14-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/boe-tv101wum-nl6: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-13-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/himax-hx83102: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-12-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/ilitek-ili9882t: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-11-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/lpm102a188a: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-10-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/jdi-lt070me05000: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-9-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/khadas-ts050: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-8-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/kd097d04: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-7-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/lg-sw43408: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-6-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/novatek-nt36672a: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-5-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/osd101t2587-53ts: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-4-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/vvx10f034n00: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-3-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/raspberrypi: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-2-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/panel/lq101r1sx01: Use refcounted allocation in place of devm_kzalloc()

Move to using the new API devm_drm_panel_alloc() to allocate the
panel. In the call to the new API, avoid using explicit type and use
__typeof() for more type safety.

Signed-off-by: Anusha Srivatsa <asrivats@redhat.com>
Link: https://lore.kernel.org/r/20250710-b4-driver-convert-last-part-july-v1-1-de73ba81b2f5@redhat.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>

drm/sitronix/st7571-i2c: Add support for the ST7567 Controller

The Sitronix ST7567 is a monochrome Dot Matrix LCD Controller that has SPI,
I2C and parallel interfaces. The st7571-i2c driver only has support for I2C
so displays using other transport interfaces are currently not supported.

The DRM_FORMAT_R1 pixel format and data commands are the same than what
is used by the ST7571 controller, so only is needed a different callback
that implements the expected initialization sequence for the ST7567 chip.

Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250715110411.448343-6-javierm@redhat.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>

dt-bindings: display: Add Sitronix ST7567 LCD Controller

Sitronix ST7567 is a monochrome Dot Matrix LCD Controller.

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250715110411.448343-5-javierm@redhat.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>

drm/sitronix/st7571-i2c: Add an indirection level to parse DT

Other Sitronix display controllers might need a different parsing DT
logic, so lets add a .parse_dt callback to struct st7571_panel_data.

Suggested-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250715110411.448343-4-javierm@redhat.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>

drm/sitronix/st7571-i2c: Log probe deferral cause for GPIO get failure

The driver already uses the dev_err_probe() helper (that only prints error
messages for the -EPROBE_DEFER case) when fails to get any other resource.

Also do the same when it fails to obtain the reset GPIO.

Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250715110411.448343-3-javierm@redhat.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>

drm/sitronix/st7571-i2c: Fix encoder callbacks function names

It seems the driver took some inspiration from ssd130x and some of the
functions (encoder callbacks) were not renamed to use a st7571_ prefix.

Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250715110411.448343-2-javierm@redhat.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>

drm/ttm: remove ttm_bo_validate_swapout test

The test is quite fragile since it tries to allocate halve available system
memory + 1 page.

If the system has either not enough memory to make the allocation work
with other things running in parallel or to much memory so the allocation
fails as to large/invalid the test will fail.

Completely remove the test. We already validate swapout on the device
level and that test seems to be stable.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250710144129.1803-2-christian.koenig@amd.com