On Mon, 2025-04-07 at 17:22 +0200, Philipp Stanner wrote:
> From: Philipp Stanner <pstan...@redhat.com>
>
> The GPU scheduler currently does not ensure that its pending_list is
> empty before performing various other teardown tasks in
> drm_sched_fini().
>
> If there are still jobs in the pending_list, this is problematic
> because
> after scheduler teardown, no one will call backend_ops.free_job()
> anymore. This would, consequently, result in memory leaks.
>
> One way to solve this is to implement a waitqueue that
> drm_sched_fini()
> blocks on until the pending_list has become empty.
>
> Add a waitqueue to struct drm_gpu_scheduler. Wake up waiters once the
> pending_list becomes empty. Wait in drm_sched_fini() for that to
> happen.
>
> Signed-off-by: Philipp Stanner <pstan...@redhat.com>
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 84 ++++++++++++++++++++----
> --
> include/drm/gpu_scheduler.h | 8 +++
> 2 files changed, 75 insertions(+), 17 deletions(-)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c
> b/drivers/gpu/drm/scheduler/sched_main.c
> index 829579c41c6b..a6a4deeda72b 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -367,7 +367,7 @@ static void drm_sched_run_job_queue(struct
> drm_gpu_scheduler *sched)
> */
> static void __drm_sched_run_free_queue(struct drm_gpu_scheduler
> *sched)
> {
> - if (!READ_ONCE(sched->pause_submit))
> + if (!READ_ONCE(sched->pause_free))
> queue_work(sched->submit_wq, &sched->work_free_job);
> }
>
> @@ -556,6 +556,7 @@ static void drm_sched_job_timedout(struct
> work_struct *work)
> * is parked at which point it's safe.
> */
> list_del_init(&job->list);
> +
> spin_unlock(&sched->job_list_lock);
>
> status = job->sched->ops->timedout_job(job);
> @@ -572,8 +573,10 @@ static void drm_sched_job_timedout(struct
> work_struct *work)
> spin_unlock(&sched->job_list_lock);
> }
>
> - if (status != DRM_GPU_SCHED_STAT_ENODEV)
> - drm_sched_start_timeout_unlocked(sched);
> + if (status != DRM_GPU_SCHED_STAT_ENODEV) {
> + if (!READ_ONCE(sched->cancel_timeout))
Another thing I want to investigate is whether all those booleans we
read with READ_ONCE are actually necessary?
Because I suspect they are not. The cancel_work_sync() functions should
do the trick, and if they're too late we'd just wait one cycle longer.
@Christian, Tvrtko, could be helpful if you can take a look.
> + drm_sched_start_timeout_unlocked(sched);
> + }
> }
>
> /**
> @@ -1119,12 +1122,22 @@ drm_sched_get_finished_job(struct
> drm_gpu_scheduler *sched)
> /* remove job from pending_list */
> list_del_init(&job->list);
>
> + /*
> + * Inform tasks blocking in drm_sched_fini() that
> it's now safe to proceed.
> + */
> + if (list_empty(&sched->pending_list))
> + wake_up(&sched->pending_list_waitque);
> +
> /* cancel this job's TO timer */
> cancel_delayed_work(&sched->work_tdr);
> /* make the scheduled timestamp more accurate */
> next = list_first_entry_or_null(&sched-
> >pending_list,
> typeof(*next),
> list);
>
> + // TODO RFC: above we wake up the waitque which
> relies on the fact
> + // that timeouts have been deactivated. The below
> should never
> + // reactivate them since the list was empty above.
> Still, should
> + // we document that?
Regarding this, I re-read the code a few times and conclude that this
is not a problem.
P.
> if (next) {
> if (test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT,
> &next->s_fence-
> >scheduled.flags))
> @@ -1324,6 +1337,7 @@ int drm_sched_init(struct drm_gpu_scheduler
> *sched, const struct drm_sched_init_
> init_waitqueue_head(&sched->job_scheduled);
> INIT_LIST_HEAD(&sched->pending_list);
> spin_lock_init(&sched->job_list_lock);
> + init_waitqueue_head(&sched->pending_list_waitque);
> atomic_set(&sched->credit_count, 0);
> INIT_DELAYED_WORK(&sched->work_tdr, drm_sched_job_timedout);
> INIT_WORK(&sched->work_run_job, drm_sched_run_job_work);
> @@ -1331,6 +1345,8 @@ int drm_sched_init(struct drm_gpu_scheduler
> *sched, const struct drm_sched_init_
> atomic_set(&sched->_score, 0);
> atomic64_set(&sched->job_id_count, 0);
> sched->pause_submit = false;
> + sched->pause_free = false;
> + sched->cancel_timeout = false;
>
> sched->ready = true;
> return 0;
> @@ -1348,6 +1364,40 @@ int drm_sched_init(struct drm_gpu_scheduler
> *sched, const struct drm_sched_init_
> }
> EXPORT_SYMBOL(drm_sched_init);
>
> +/**
> + * drm_sched_submission_and_timeout_stop - stop everything except
> for free_job()
> + * @sched: scheduler instance
> + *
> + * Only needed to cleanly tear down the scheduler in
> drm_sched_fini().
> + */
> +static inline void
> +drm_sched_submission_and_timeout_stop(struct drm_gpu_scheduler
> *sched)
> +{
> + WRITE_ONCE(sched->pause_submit, true);
> + WRITE_ONCE(sched->cancel_timeout, true);
> + cancel_work_sync(&sched->work_run_job);
> + cancel_delayed_work_sync(&sched->work_tdr);
> +}
> +
> +static inline void
> +drm_sched_free_stop(struct drm_gpu_scheduler *sched)
> +{
> + WRITE_ONCE(sched->pause_free, true);
> + cancel_work_sync(&sched->work_free_job);
> +}
> +
> +static inline bool
> +drm_sched_no_jobs_pending(struct drm_gpu_scheduler *sched)
> +{
> + bool empty;
> +
> + spin_lock(&sched->job_list_lock);
> + empty = list_empty(&sched->pending_list);
> + spin_unlock(&sched->job_list_lock);
> +
> + return empty;
> +}
> +
> /**
> * drm_sched_fini - Destroy a gpu scheduler
> *
> @@ -1355,26 +1405,24 @@ EXPORT_SYMBOL(drm_sched_init);
> *
> * Tears down and cleans up the scheduler.
> *
> - * This stops submission of new jobs to the hardware through
> - * drm_sched_backend_ops.run_job(). Consequently,
> drm_sched_backend_ops.free_job()
> - * will not be called for all jobs still in
> drm_gpu_scheduler.pending_list.
> - * There is no solution for this currently. Thus, it is up to the
> driver to make
> - * sure that:
> - *
> - * a) drm_sched_fini() is only called after for all submitted jobs
> - * drm_sched_backend_ops.free_job() has been called or that
> - * b) the jobs for which drm_sched_backend_ops.free_job() has not
> been called
> - * after drm_sched_fini() ran are freed manually.
> - *
> - * FIXME: Take care of the above problem and prevent this function
> from leaking
> - * the jobs in drm_gpu_scheduler.pending_list under any
> circumstances.
> + * Note that this function blocks until all the fences returned by
> + * &struct drm_sched_backend_ops.run_job have been signalled.
> */
> void drm_sched_fini(struct drm_gpu_scheduler *sched)
> {
> struct drm_sched_entity *s_entity;
> int i;
>
> - drm_sched_wqueue_stop(sched);
> + /*
> + * Jobs that have neither been scheduled or which have timed
> out are
> + * gone by now, but jobs that have been submitted through
> + * backend_ops.run_job() and have not yet terminated are
> still pending.
> + *
> + * Wait for the pending_list to become empty to avoid
> leaking those jobs.
> + */
> + drm_sched_submission_and_timeout_stop(sched);
> + wait_event(sched->pending_list_waitque,
> drm_sched_no_jobs_pending(sched));
> + drm_sched_free_stop(sched);
>
> for (i = DRM_SCHED_PRIORITY_KERNEL; i < sched->num_rqs; i++)
> {
> struct drm_sched_rq *rq = sched->sched_rq[i];
> @@ -1471,6 +1519,7 @@ EXPORT_SYMBOL(drm_sched_wqueue_ready);
> void drm_sched_wqueue_stop(struct drm_gpu_scheduler *sched)
> {
> WRITE_ONCE(sched->pause_submit, true);
> + WRITE_ONCE(sched->pause_free, true);
> cancel_work_sync(&sched->work_run_job);
> cancel_work_sync(&sched->work_free_job);
> }
> @@ -1488,6 +1537,7 @@ EXPORT_SYMBOL(drm_sched_wqueue_stop);
> void drm_sched_wqueue_start(struct drm_gpu_scheduler *sched)
> {
> WRITE_ONCE(sched->pause_submit, false);
> + WRITE_ONCE(sched->pause_free, false);
> queue_work(sched->submit_wq, &sched->work_run_job);
> queue_work(sched->submit_wq, &sched->work_free_job);
> }
> diff --git a/include/drm/gpu_scheduler.h
> b/include/drm/gpu_scheduler.h
> index 1a7e377d4cbb..badcf9ea4501 100644
> --- a/include/drm/gpu_scheduler.h
> +++ b/include/drm/gpu_scheduler.h
> @@ -29,6 +29,7 @@
> #include <linux/completion.h>
> #include <linux/xarray.h>
> #include <linux/workqueue.h>
> +#include <linux/wait.h>
>
> #define MAX_WAIT_SCHED_ENTITY_Q_EMPTY msecs_to_jiffies(1000)
>
> @@ -533,6 +534,8 @@ struct drm_sched_backend_ops {
> * timeout interval is over.
> * @pending_list: the list of jobs which are currently in the job
> queue.
> * @job_list_lock: lock to protect the pending_list.
> + * @pending_list_waitque: a waitqueue for drm_sched_fini() to block
> on until all
> + * pending jobs have been finished.
> * @hang_limit: once the hangs by a job crosses this limit then it
> is marked
> * guilty and it will no longer be considered for
> scheduling.
> * @score: score to help loadbalancer pick a idle sched
> @@ -540,6 +543,8 @@ struct drm_sched_backend_ops {
> * @ready: marks if the underlying HW is ready to work
> * @free_guilty: A hit to time out handler to free the guilty job.
> * @pause_submit: pause queuing of @work_run_job on @submit_wq
> + * @pause_free: pause queuing of @work_free_job on @submit_wq
> + * @cancel_timeout: pause queuing of @work_tdr on @submit_wq
> * @own_submit_wq: scheduler owns allocation of @submit_wq
> * @dev: system &struct device
> *
> @@ -562,12 +567,15 @@ struct drm_gpu_scheduler {
> struct delayed_work work_tdr;
> struct list_head pending_list;
> spinlock_t job_list_lock;
> + wait_queue_head_t pending_list_waitque;
> int hang_limit;
> atomic_t *score;
> atomic_t _score;
> bool ready;
> bool free_guilty;
> bool pause_submit;
> + bool pause_free;
> + bool cancel_timeout;
> bool own_submit_wq;
> struct device *dev;
> };
