From: "Christian König" <christian.koenig@amd.com>
To: Andrey Grodzovsky <andrey.grodzovsky@amd.com>,
dri-devel@lists.freedesktop.org, amd-gfx@lists.freedesktop.org,
linux-pci@vger.kernel.org, ckoenig.leichtzumerken@gmail.com,
daniel.vetter@ffwll.ch, Harry.Wentland@amd.com
Cc: Alexander.Deucher@amd.com, gregkh@linuxfoundation.org,
helgaas@kernel.org, Felix.Kuehling@amd.com
Subject: Re: [PATCH v5 15/27] drm/scheduler: Fix hang when sched_entity released
Date: Thu, 29 Apr 2021 09:18:13 +0200 [thread overview]
Message-ID: <a8314d77-578f-e0df-5c49-77d5f10c76c7@amd.com> (raw)
In-Reply-To: <20210428151207.1212258-16-andrey.grodzovsky@amd.com>
I need to take another look at this part when I don't have a massive
headache any more.
Maybe split the patch set up into different parts, something like:
1. Adding general infrastructure.
2. Making sure all memory is unpolated.
3. Job and fence handling
Christian.
Am 28.04.21 um 17:11 schrieb Andrey Grodzovsky:
> Problem: If scheduler is already stopped by the time sched_entity
> is released and entity's job_queue not empty I encountred
> a hang in drm_sched_entity_flush. This is because drm_sched_entity_is_idle
> never becomes false.
>
> Fix: In drm_sched_fini detach all sched_entities from the
> scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
> Also wakeup all those processes stuck in sched_entity flushing
> as the scheduler main thread which wakes them up is stopped by now.
>
> v2:
> Reverse order of drm_sched_rq_remove_entity and marking
> s_entity as stopped to prevent reinserion back to rq due
> to race.
>
> v3:
> Drop drm_sched_rq_remove_entity, only modify entity->stopped
> and check for it in drm_sched_entity_is_idle
>
> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
> Reviewed-by: Christian König <christian.koenig@amd.com>
> ---
> drivers/gpu/drm/scheduler/sched_entity.c | 3 ++-
> drivers/gpu/drm/scheduler/sched_main.c | 24 ++++++++++++++++++++++++
> 2 files changed, 26 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
> index f0790e9471d1..cb58f692dad9 100644
> --- a/drivers/gpu/drm/scheduler/sched_entity.c
> +++ b/drivers/gpu/drm/scheduler/sched_entity.c
> @@ -116,7 +116,8 @@ static bool drm_sched_entity_is_idle(struct drm_sched_entity *entity)
> rmb(); /* for list_empty to work without lock */
>
> if (list_empty(&entity->list) ||
> - spsc_queue_count(&entity->job_queue) == 0)
> + spsc_queue_count(&entity->job_queue) == 0 ||
> + entity->stopped)
> return true;
>
> return false;
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> index 908b0b56032d..ba087354d0a8 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -897,9 +897,33 @@ EXPORT_SYMBOL(drm_sched_init);
> */
> void drm_sched_fini(struct drm_gpu_scheduler *sched)
> {
> + struct drm_sched_entity *s_entity;
> + int i;
> +
> if (sched->thread)
> kthread_stop(sched->thread);
>
> + for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
> + struct drm_sched_rq *rq = &sched->sched_rq[i];
> +
> + if (!rq)
> + continue;
> +
> + spin_lock(&rq->lock);
> + list_for_each_entry(s_entity, &rq->entities, list)
> + /*
> + * Prevents reinsertion and marks job_queue as idle,
> + * it will removed from rq in drm_sched_entity_fini
> + * eventually
> + */
> + s_entity->stopped = true;
> + spin_unlock(&rq->lock);
> +
> + }
> +
> + /* Wakeup everyone stuck in drm_sched_entity_flush for this scheduler */
> + wake_up_all(&sched->job_scheduled);
> +
> /* Confirm no work left behind accessing device structures */
> cancel_delayed_work_sync(&sched->work_tdr);
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2021-04-29 7:18 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-28 15:11 [PATCH v5 00/27] RFC Support hot device unplug in amdgpu Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 01/27] drm/ttm: Remap all page faults to per process dummy page Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 02/27] drm/ttm: Expose ttm_tt_unpopulate for driver use Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 03/27] drm/amdgpu: Split amdgpu_device_fini into early and late Andrey Grodzovsky
2021-04-29 7:04 ` Christian König
2021-04-30 3:10 ` Alex Deucher
2021-04-30 5:19 ` Lazar, Lijo
2021-04-30 5:39 ` Andrey Grodzovsky
2021-04-30 5:49 ` Lazar, Lijo
2021-04-28 15:11 ` [PATCH v5 04/27] drm/amdkfd: Split kfd suspend from devie exit Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 05/27] drm/amdgpu: Add early fini callback Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 06/27] drm/amdgpu: Handle IOMMU enabled case Andrey Grodzovsky
2021-04-29 7:08 ` Christian König
2021-05-03 20:43 ` Andrey Grodzovsky
2021-05-04 7:03 ` Christian König
2021-05-04 15:43 ` Andrey Grodzovsky
2021-05-05 14:51 ` Andrey Grodzovsky
2021-04-30 3:13 ` Alex Deucher
2021-05-03 18:00 ` Andrey Grodzovsky
2021-05-04 17:05 ` Felix Kuehling
2021-05-05 14:05 ` Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 07/27] drm/amdgpu: Remap all page faults to per process dummy page Andrey Grodzovsky
2021-04-29 7:09 ` Christian König
2021-04-28 15:11 ` [PATCH v5 08/27] PCI: add support for dev_groups to struct pci_device_driver Andrey Grodzovsky
2021-04-28 16:53 ` Bjorn Helgaas
2021-04-29 16:53 ` Andrey Grodzovsky
2021-04-29 19:23 ` Bjorn Helgaas
2021-04-29 20:36 ` Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 09/27] dmr/amdgpu: Move some sysfs attrs creation to default_attr Andrey Grodzovsky
2021-04-28 17:23 ` Bjorn Helgaas
2021-04-29 7:11 ` Christian König
2021-04-28 15:11 ` [PATCH v5 10/27] drm/amdgpu: Guard against write accesses after device removal Andrey Grodzovsky
2021-04-29 7:14 ` Christian König
2021-04-28 15:11 ` [PATCH v5 11/27] drm/sched: Make timeout timer rearm conditional Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 12/27] drm/amdgpu: Prevent any job recoveries after device is unplugged Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 13/27] drm/amdgpu: When filizing the fence driver. stop scheduler first Andrey Grodzovsky
2021-04-29 7:15 ` Christian König
2021-04-29 17:12 ` Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 14/27] drm/amdgpu: Fix hang on device removal Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 15/27] drm/scheduler: Fix hang when sched_entity released Andrey Grodzovsky
2021-04-29 7:18 ` Christian König [this message]
2021-04-29 17:06 ` Andrey Grodzovsky
2021-04-30 6:47 ` Christian König
2021-04-30 16:10 ` Andrey Grodzovsky
2021-05-05 13:57 ` Andrey Grodzovsky
2021-05-07 16:29 ` Daniel Vetter
2021-05-07 16:32 ` Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 16/27] drm/amdgpu: Unmap all MMIO mappings Andrey Grodzovsky
2021-04-29 7:19 ` Christian König
2021-04-29 16:55 ` Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 17/27] drm/amdgpu: Add rw_sem to pushing job into sched queue Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 18/27] drm/sched: Expose drm_sched_entity_kill_jobs Andrey Grodzovsky
2021-04-28 15:11 ` [PATCH v5 19/27] drm/amdgpu: Finilise device fences on device remove Andrey Grodzovsky
2021-04-28 15:20 ` Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 20/27] drm: Scope all DRM IOCTLs with drm_dev_enter/exit Andrey Grodzovsky
2021-04-29 11:23 ` Daniel Vetter
2021-04-29 11:32 ` Daniel Vetter
2021-04-29 16:04 ` Andrey Grodzovsky
2021-04-29 16:15 ` Felix Kuehling
2021-04-29 16:21 ` Andrey Grodzovsky
2021-04-29 16:29 ` Felix Kuehling
2021-04-29 16:33 ` Andrey Grodzovsky
2021-04-29 19:05 ` Daniel Vetter
2021-04-29 20:34 ` Andrey Grodzovsky
2021-04-30 10:25 ` Daniel Vetter
2021-04-30 17:27 ` Andrey Grodzovsky
2021-05-05 13:57 ` Andrey Grodzovsky
2021-05-06 9:40 ` Daniel Vetter
2021-05-06 16:25 ` Andrey Grodzovsky
2021-05-07 9:11 ` Daniel Vetter
2021-05-07 15:39 ` Andrey Grodzovsky
2021-05-07 16:24 ` Daniel Vetter
2021-05-07 18:00 ` Andrey Grodzovsky
2021-05-10 15:46 ` Daniel Vetter
2021-04-28 15:12 ` [PATCH v5 21/27] drm/amdgpu: Add support for hot-unplug feature at DRM level Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 22/27] drm/amd/display: Scope all DM queued work with drm_dev_enter/exit Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 23/27] drm/amd/powerplay: Scope all PM " Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 24/27] drm/amdkfd: Scope all KFD " Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 25/27] drm/amdgpu: Scope all amdgpu " Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 26/27] drm/amd/display: Remove superflous drm_mode_config_cleanup Andrey Grodzovsky
2021-04-28 15:12 ` [PATCH v5 27/27] drm/amdgpu: Verify DMA opearations from device are done Andrey Grodzovsky
2021-04-28 17:07 ` [PATCH v5 00/27] RFC Support hot device unplug in amdgpu Bjorn Helgaas
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=a8314d77-578f-e0df-5c49-77d5f10c76c7@amd.com \
--to=christian.koenig@amd.com \
--cc=Alexander.Deucher@amd.com \
--cc=Felix.Kuehling@amd.com \
--cc=Harry.Wentland@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=andrey.grodzovsky@amd.com \
--cc=ckoenig.leichtzumerken@gmail.com \
--cc=daniel.vetter@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=gregkh@linuxfoundation.org \
--cc=helgaas@kernel.org \
--cc=linux-pci@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).