On Wed, May 28, 2025 at 02:51:33PM -0700, Matthew Brost wrote:
On Wed, May 28, 2025 at 02:42:22PM -0700, Lucas De Marchi wrote:
xef is only valid if it's a job from userspace. For in-kernel jobs it
causes a NPD like below:
<4> [] RIP: 0010:xe_sched_job_create+0xbd/0x390 [xe]
...
<4> [] Call Trace:
<4> [] <TASK>
<4> [] __xe_bb_create_job+0xa2/0x240 [xe]
<4> [] ? find_held_lock+0x31/0x90
<4> [] ? xa_find_after+0x12c/0x250
<4> [] xe_bb_create_job+0x6e/0x380 [xe]
<4> [] ? xa_find_after+0x136/0x250
<4> [] ? __drm_dev_dbg+0x7d/0xb0
<4> [] xe_gt_record_default_lrcs+0x542/0xb00 [xe]
Since drm_file starts with 1 for the unique id, just use 0 for the
in-kernel jobs.
Fixes: 2956554823ce ("drm/sched: Store the drm client_id in drm_sched_fence")
Cc: Christian König <[email protected]>
Cc: Pierre-Eric Pelloux-Prayer <[email protected]>
Cc: Philipp Stanner <[email protected]>
Signed-off-by: Lucas De Marchi <[email protected]>
Reviewed-by: Matthew Brost <[email protected]>
thanks, pushed to drm-misc-next.
Lucas De Marchi
