drm/imagination: take paired job reference

For paired jobs, have the fragment job take a reference on the
geometry job, so that the geometry job cannot be freed until
the fragment job has finished w

drm/imagination: take paired job reference

For paired jobs, have the fragment job take a reference on the
geometry job, so that the geometry job cannot be freed until
the fragment job has finished with it.

The geometry job structure is accessed when the fragment job is being
prepared by the GPU scheduler. Taking the reference prevents the
geometry job being freed until the fragment job no longer requires it.

Fixes a use after free bug detected by KASAN:

[ 124.256386] BUG: KASAN: slab-use-after-free in pvr_queue_prepare_job+0x108/0x868 [powervr]
[ 124.264893] Read of size 1 at addr ffff0000084cb960 by task kworker/u16:4/63

Cc: [email protected]
Fixes: eaf01ee5ba28 ("drm/imagination: Implement job submission and scheduling")
Signed-off-by: Brendan King <[email protected]>
Reviewed-by: Matt Coster <[email protected]>
Link: https://lore.kernel.org/r/20250318-ddkopsrc-1337-use-after-free-in-pvr_queue_prepare_job-v1-1-80fb30d044a6@imgtec.com
Signed-off-by: Matt Coster <[email protected]>

show more ...

drm/imagination: only init job done fences once

Ensure job done fences are only initialised once.

This fixes a memory manager not clean warning from drm_mm_takedown
on module unload.

Cc: stable@vg

drm/imagination: only init job done fences once

Ensure job done fences are only initialised once.

This fixes a memory manager not clean warning from drm_mm_takedown
on module unload.

Cc: [email protected]
Fixes: eaf01ee5ba28 ("drm/imagination: Implement job submission and scheduling")
Signed-off-by: Brendan King <[email protected]>
Reviewed-by: Matt Coster <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Signed-off-by: Matt Coster <[email protected]>

show more ...

drm/imagination: avoid deadlock on fence release

Do scheduler queue fence release processing on a workqueue, rather
than in the release function itself.

Fixes deadlock issues such as the following:

drm/imagination: avoid deadlock on fence release

Do scheduler queue fence release processing on a workqueue, rather
than in the release function itself.

Fixes deadlock issues such as the following:

[ 607.400437] ============================================
[ 607.405755] WARNING: possible recursive locking detected
[ 607.415500] --------------------------------------------
[ 607.420817] weston:zfq0/24149 is trying to acquire lock:
[ 607.426131] ffff000017d041a0 (reservation_ww_class_mutex){+.+.}-{3:3}, at: pvr_gem_object_vunmap+0x40/0xc0 [powervr]
[ 607.436728]
but task is already holding lock:
[ 607.442554] ffff000017d105a0 (reservation_ww_class_mutex){+.+.}-{3:3}, at: dma_buf_ioctl+0x250/0x554
[ 607.451727]
other info that might help us debug this:
[ 607.458245] Possible unsafe locking scenario:

[ 607.464155] CPU0
[ 607.466601] ----
[ 607.469044] lock(reservation_ww_class_mutex);
[ 607.473584] lock(reservation_ww_class_mutex);
[ 607.478114]
*** DEADLOCK ***

Cc: [email protected]
Fixes: eaf01ee5ba28 ("drm/imagination: Implement job submission and scheduling")
Signed-off-by: Brendan King <[email protected]>
Reviewed-by: Matt Coster <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Signed-off-by: Matt Coster <[email protected]>

show more ...

drm/sched: Use struct for drm_sched_init() params

drm_sched_init() has a great many parameters and upcoming new
functionality for the scheduler might add even more. Generally, the
great number of pa

drm/sched: Use struct for drm_sched_init() params

drm_sched_init() has a great many parameters and upcoming new
functionality for the scheduler might add even more. Generally, the
great number of parameters reduces readability and has already caused
one missnaming, addressed in:

commit 6f1cacf4eba7 ("drm/nouveau: Improve variable name in
nouveau_sched_init()").

Introduce a new struct for the scheduler init parameters and port all
users.

Reviewed-by: Liviu Dudau <[email protected]>
Acked-by: Matthew Brost <[email protected]> # for Xe
Reviewed-by: Boris Brezillon <[email protected]> # for Panfrost and Panthor
Reviewed-by: Christian Gmeiner <[email protected]> # for Etnaviv
Reviewed-by: Frank Binns <[email protected]> # for Imagination
Reviewed-by: Tvrtko Ursulin <[email protected]> # for Sched
Reviewed-by: Maíra Canal <[email protected]> # for v3d
Reviewed-by: Danilo Krummrich <[email protected]>
Reviewed-by: Lizhi Hou <[email protected]> # for amdxdna
Signed-off-by: Philipp Stanner <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

show more ...

drm/sched: add optional errno to drm_sched_start()

The current implementation of drm_sched_start uses a hardcoded
-ECANCELED to dispose of a job when the parent/hw fence is NULL.
This results in drm

drm/sched: add optional errno to drm_sched_start()

The current implementation of drm_sched_start uses a hardcoded
-ECANCELED to dispose of a job when the parent/hw fence is NULL.
This results in drm_sched_job_done being called with -ECANCELED for
each job with a NULL parent in the pending list, making it difficult
to distinguish between recovery methods, whether a queue reset or a
full GPU reset was used.

To improve this, we first try a soft recovery for timeout jobs and
use the error code -ENODATA. If soft recovery fails, we proceed with
a queue reset, where the error code remains -ENODATA for the job.
Finally, for a full GPU reset, we use error codes -ECANCELED or
-ETIME. This patch adds an error code parameter to drm_sched_start,
allowing us to differentiate between queue reset and GPU reset
failures. This enables user mode and test applications to validate
the expected correctness of the requested operation. After a
successful queue reset, the only way to continue normal operation is
to call drm_sched_job_done with the specific error code -ENODATA.

v1: Initial implementation by Jesse utilized amdgpu_device_lock_reset_domain
and amdgpu_device_unlock_reset_domain to allow user mode to track
the queue reset status and distinguish between queue reset and
GPU reset.
v2: Christian suggested using the error codes -ENODATA for queue reset
and -ECANCELED or -ETIME for GPU reset, returned to
amdgpu_cs_wait_ioctl.
v3: To meet the requirements, we introduce a new function
drm_sched_start_ex with an additional parameter to set
dma_fence_set_error, allowing us to handle the specific error
codes appropriately and dispose of bad jobs with the selected
error code depending on whether it was a queue reset or GPU reset.
v4: Alex suggested using a new name, drm_sched_start_with_recovery_error,
which more accurately describes the function's purpose.
Additionally, it was recommended to add documentation details
about the new method.
v5: Fixed declaration of new function drm_sched_start_with_recovery_error.(Alex)
v6 (chk): rebase on upstream changes, cleanup the commit message,
drop the new function again and update all callers,
apply the errno also to scheduler fences with hw fences
v7 (chk): rebased

Signed-off-by: Jesse Zhang <[email protected]>
Signed-off-by: Vitaly Prosyak <[email protected]>
Signed-off-by: Christian König <[email protected]>
Acked-by: Daniel Vetter <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

show more ...

drm/scheduler: remove full_recover from drm_sched_start

This was basically just another one of amdgpus hacks. The parameter
allowed to restart the scheduler without turning fence signaling on
again.

drm/scheduler: remove full_recover from drm_sched_start

This was basically just another one of amdgpus hacks. The parameter
allowed to restart the scheduler without turning fence signaling on
again.

That this is absolutely not a good idea should be obvious by now since
the fences will then just sit there and never signal.

While at it cleanup the code a bit.

Signed-off-by: Christian König <[email protected]>
Reviewed-by: Matthew Brost <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]

show more ...

drm/sched: Fix compilation issues with DRM priority rename

Fix compilation issues with DRM scheduler priority rename MIN to LOW.

Signed-off-by: Luben Tuikov <[email protected]>
Reported-by: kerne

drm/sched: Fix compilation issues with DRM priority rename

Fix compilation issues with DRM scheduler priority rename MIN to LOW.

Signed-off-by: Luben Tuikov <[email protected]>
Reported-by: kernel test robot <[email protected]>
Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Cc: Danilo Krummrich <[email protected]>
Cc: Frank Binns <[email protected]>
Cc: Donald Robson <[email protected]>
Cc: Matt Coster <[email protected]>
Cc: Direct Rendering Infrastructure - Development <[email protected]>
Fixes: fe375c74806dbd ("drm/sched: Rename priority MIN to LOW")
Fixes: 38f922a563aac3 ("drm/sched: Reverse run-queue priority enumeration")
Fixes: 5f03a507b29e44 ("drm/nouveau: implement 1:1 scheduler - entity relationship")
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
Reviewed-by: Christian König <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

show more ...

drm/imagination: Implement job submission and scheduling

Implement job submission ioctl. Job scheduling is implemented using
drm_sched.

Jobs are submitted in a stream format. This is intended to al

drm/imagination: Implement job submission and scheduling

Implement job submission ioctl. Job scheduling is implemented using
drm_sched.

Jobs are submitted in a stream format. This is intended to allow the UAPI
data format to be independent of the actual FWIF structures in use, which
vary depending on the GPU in use.

The stream formats are documented at:
https://gitlab.freedesktop.org/mesa/mesa/-/blob/f8d2b42ae65c2f16f36a43e0ae39d288431e4263/src/imagination/csbgen/rogue_kmd_stream.xml

Changes since v8:
- Updated for upstreamed DRM scheduler changes
- Removed workaround code for the pending_list previously being updated
after run_job() returned
- Fixed null deref in pvr_queue_cleanup_fw_context() for bad stream ptr
given to create_context ioctl
- Corrected license identifiers

Changes since v7:
- Updated for v8 "DRM scheduler changes for XE" patchset

Changes since v6:
- Fix fence handling in pvr_sync_signal_array_add()
- Add handling for SUBMIT_JOB_FRAG_CMD_DISABLE_PIXELMERGE flag
- Fix missing dma_resv locking in job submit path

Changes since v5:
- Fix leak in job creation error path

Changes since v4:
- Use a regular workqueue for job scheduling

Changes since v3:
- Support partial render jobs
- Add job timeout handler
- Split sync handling out of job code
- Use drm_dev_{enter,exit}

Changes since v2:
- Use drm_sched for job scheduling

Co-developed-by: Boris Brezillon <[email protected]>
Signed-off-by: Boris Brezillon <[email protected]>
Co-developed-by: Donald Robson <[email protected]>
Signed-off-by: Donald Robson <[email protected]>
Signed-off-by: Sarah Walker <[email protected]>
Link: https://lore.kernel.org/r/c98dab7a5f5fb891fbed7e4990d19b5d13964365.1700668843.git.donald.robson@imgtec.com
Signed-off-by: Maxime Ripard <[email protected]>

show more ...