diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c index 30e7b7196dab..d0469e944143 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.c +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -25,7 +25,8 @@ struct panfrost_queue_state { struct drm_gpu_scheduler sched; - + bool stopped; + struct mutex lock; u64 fence_context; u64 emit_seqno; }; @@ -369,6 +370,24 @@ void panfrost_job_enable_interrupts(struct panfrost_device *pfdev) job_write(pfdev, JOB_INT_MASK, irq_mask); } +static bool panfrost_scheduler_stop(struct panfrost_queue_state *queue, + struct drm_sched_job *bad) +{ + bool stopped = false; + + mutex_lock(&queue->lock); + if (!queue->stopped) { + drm_sched_stop(&queue->sched, bad); + if (bad) + drm_sched_increase_karma(bad); + queue->stopped = true; + stopped = true; + } + mutex_unlock(&queue->lock); + + return stopped; +} + static void panfrost_job_timedout(struct drm_sched_job *sched_job) { struct panfrost_job *job = to_panfrost_job(sched_job); @@ -392,19 +411,39 @@ static void panfrost_job_timedout(struct drm_sched_job *sched_job) job_read(pfdev, JS_TAIL_LO(js)), sched_job); + /* Scheduler is already stopped, nothing to do. */ + if (!panfrost_scheduler_stop(&pfdev->js->queue[js], sched_job)) + return; + if (!mutex_trylock(&pfdev->reset_lock)) return; for (i = 0; i < NUM_JOB_SLOTS; i++) { struct drm_gpu_scheduler *sched = &pfdev->js->queue[i].sched; - drm_sched_stop(sched, sched_job); - if (js != i) - /* Ensure any timeouts on other slots have finished */ + /* + * If the queue is still active, make sure we wait for any + * pending timeouts. + */ + if (!pfdev->js->queue[i].stopped) cancel_delayed_work_sync(&sched->work_tdr); - } - drm_sched_increase_karma(sched_job); + /* + * If the scheduler was not already stopped, there's a tiny + * chance a timeout has expired just before we stopped it, and + * drm_sched_stop() does not flush pending works. Let's flush + * them now so the timeout handler doesn't get called in the + * middle of a reset. + */ + if (panfrost_scheduler_stop(&pfdev->js->queue[i], NULL)) + cancel_delayed_work_sync(&sched->work_tdr); + + /* + * Now that we cancelled the pending timeouts, we can safely + * reset the stopped state. + */ + pfdev->js->queue[i].stopped = false; + } spin_lock_irqsave(&pfdev->js->job_lock, flags); for (i = 0; i < NUM_JOB_SLOTS; i++) { @@ -421,11 +460,11 @@ static void panfrost_job_timedout(struct drm_sched_job *sched_job) for (i = 0; i < NUM_JOB_SLOTS; i++) drm_sched_resubmit_jobs(&pfdev->js->queue[i].sched); + mutex_unlock(&pfdev->reset_lock); + /* restart scheduler after GPU is usable again */ for (i = 0; i < NUM_JOB_SLOTS; i++) drm_sched_start(&pfdev->js->queue[i].sched, true); - - mutex_unlock(&pfdev->reset_lock); } static const struct drm_sched_backend_ops panfrost_sched_ops = { @@ -558,6 +597,7 @@ int panfrost_job_open(struct panfrost_file_priv *panfrost_priv) int ret, i; for (i = 0; i < NUM_JOB_SLOTS; i++) { + mutex_init(&js->queue[i].lock); sched = &js->queue[i].sched; ret = drm_sched_entity_init(&panfrost_priv->sched_entity[i], DRM_SCHED_PRIORITY_NORMAL, &sched, @@ -570,10 +610,14 @@ int panfrost_job_open(struct panfrost_file_priv *panfrost_priv) void panfrost_job_close(struct panfrost_file_priv *panfrost_priv) { + struct panfrost_device *pfdev = panfrost_priv->pfdev; + struct panfrost_job_slot *js = pfdev->js; int i; - for (i = 0; i < NUM_JOB_SLOTS; i++) + for (i = 0; i < NUM_JOB_SLOTS; i++) { drm_sched_entity_destroy(&panfrost_priv->sched_entity[i]); + mutex_destroy(&js->queue[i].lock); + } } int panfrost_job_is_idle(struct panfrost_device *pfdev)