accel/amdxdna: Fix deadlock between context destroy and job timeout
authorLizhi Hou <lizhi.hou@amd.com>
Fri, 7 Nov 2025 18:10:50 +0000 (10:10 -0800)
committerLizhi Hou <lizhi.hou@amd.com>
Thu, 13 Nov 2025 17:10:43 +0000 (09:10 -0800)
commitca2583412306ceda9304a7c4302fd9efbf43e963
tree9f0e7a10a0651d1e4f68dcbec9f49ccfd9b898be
parent6ff9385c07aa311f01f87307e6256231be7d8675
accel/amdxdna: Fix deadlock between context destroy and job timeout

Hardware context destroy function holds dev_lock while waiting for all jobs
to complete. The timeout job also needs to acquire dev_lock, this leads to
a deadlock.

Fix the issue by temporarily releasing dev_lock before waiting for all
jobs to finish, and reacquiring it afterward.

Fixes: 4fd6ca90fc7f ("accel/amdxdna: Refactor hardware context destroy routine")
Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://patch.msgid.link/20251107181050.1293125-1-lizhi.hou@amd.com
drivers/accel/amdxdna/aie2_ctx.c