Quoting Chris Wilson (2018-05-03 07:36:52)
> If inside hangcheck we see that the engine has paused, but there is an
> execlists interrupt still pending, we know that the tasklet did not
> fire. Dump the GEM trace along with the current engine state, and kick
> the tasklet to recovery without having to go through a GPU reset.
>
> Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Cc: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx>
Oh I thought I sent this earlier, but I appear not to have.
Please review as it can go in all by itself...
> drivers/gpu/drm/i915/intel_hangcheck.c | 23 +++++++++++++++++++++++
> 1 file changed, 23 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/intel_hangcheck.c b/drivers/gpu/drm/i915/intel_hangcheck.c
> index 309e38b00e95..2d7f10492e35 100644
> --- a/drivers/gpu/drm/i915/intel_hangcheck.c
> +++ b/drivers/gpu/drm/i915/intel_hangcheck.c
> @@ -267,6 +267,29 @@ engine_stuck(struct intel_engine_cs *engine, u64 acthd)
> }
> }
>
> + if (test_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted)) {
> + struct intel_engine_execlists *execlists = &engine->execlists;
> + enum intel_engine_hangcheck_action ret = ENGINE_WAIT;
> +
> + if (GEM_SHOW_DEBUG()) {
> + struct drm_printer p = drm_debug_printer("hangcheck");
> +
> + GEM_TRACE_DUMP();
> + intel_engine_dump(engine, &p,
> + "%s stuck\n", engine->name);
> + }
> +
> + if (tasklet_trylock(&execlists->tasklet)) {
> + execlists->tasklet.func(execlists->tasklet.data);
> + tasklet_unlock(&execlists->tasklet);
> +
> + ret = ENGINE_WAIT_KICK;
> + }
> +
> + tasklet_hi_schedule(&execlists->tasklet);
> + return ret;
> + }
> +
> return ENGINE_DEAD;
> }
>
> --
> 2.17.0
>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
