Commit c636ae34 authored by Karol Wachowski's avatar Karol Wachowski
Browse files

accel/ivpu: Trigger recovery on TDR with OS scheduling 



With OS scheduling mode the driver cannot determine which context
caused the timeout, so context abort cannot be used. Instead of
queuing context_abort_work, directly trigger full device recovery
when a job timeout (TDR) occurs in OS scheduling mode.

Fixes: ade00a6c ("accel/ivpu: Perform engine reset instead of device recovery on TDR")
Reviewed-by: default avatarJeff Hugo <jeff.hugo@oss.qualcomm.com>
Reviewed-by: default avatarLizhi Hou <lizhi.hou@amd.com>
Signed-off-by: default avatarKarol Wachowski <karol.wachowski@linux.intel.com>
Link: https://patch.msgid.link/20260402125526.845210-1-karol.wachowski@linux.intel.com
parent 6fdfd240

drivers/accel/ivpu/ivpu_pm.c

+6 −0
Original line number Diff line number Diff line
@@ -221,6 +221,12 @@ static void ivpu_job_timeout_work(struct work_struct *work) 


abort:

	atomic_set(&vdev->job_timeout_counter, 0);



	if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) {

		ivpu_pm_trigger_recovery(vdev, "Job timeout");

		return;

	}



	ivpu_jsm_state_dump(vdev);

	ivpu_dev_coredump(vdev);

	queue_work(system_percpu_wq, &vdev->context_abort_work);