From: "Christian König" <ckoenig.leichtzumerken@gmail.com>
To: Daniel Vetter <daniel.vetter@ffwll.ch>,
DRI Development <dri-devel@lists.freedesktop.org>
Cc: linux-rdma@vger.kernel.org, intel-gfx@lists.freedesktop.org,
"Maarten Lankhorst" <maarten.lankhorst@linux.intel.com>,
LKML <linux-kernel@vger.kernel.org>,
amd-gfx@lists.freedesktop.org,
"Chris Wilson" <chris@chris-wilson.co.uk>,
linaro-mm-sig@lists.linaro.org,
"Daniel Vetter" <daniel.vetter@intel.com>,
"Christian König" <christian.koenig@amd.com>,
linux-media@vger.kernel.org
Subject: Re: [RFC 09/17] drm/amdgpu: use dma-fence annotations in cs_submit()
Date: Wed, 13 May 2020 09:02:38 +0200 [thread overview]
Message-ID: <6cfd324e-0443-3a12-6a2c-25a546c68643@gmail.com> (raw)
In-Reply-To: <20200512085944.222637-10-daniel.vetter@ffwll.ch>
Am 12.05.20 um 10:59 schrieb Daniel Vetter:
> This is a bit tricky, since ->notifier_lock is held while calling
> dma_fence_wait we must ensure that also the read side (i.e.
> dma_fence_begin_signalling) is on the same side. If we mix this up
> lockdep complaints, and that's again why we want to have these
> annotations.
>
> A nice side effect of this is that because of the fs_reclaim priming
> for dma_fence_enable lockdep now automatically checks for us that
> nothing in here allocates memory, without even running any userptr
> workloads.
>
> Cc: linux-media@vger.kernel.org
> Cc: linaro-mm-sig@lists.linaro.org
> Cc: linux-rdma@vger.kernel.org
> Cc: amd-gfx@lists.freedesktop.org
> Cc: intel-gfx@lists.freedesktop.org
> Cc: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Cc: Christian König <christian.koenig@amd.com>
> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> index 7653f62b1b2d..6db3f3c629b0 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> @@ -1213,6 +1213,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
> struct amdgpu_job *job;
> uint64_t seq;
> int r;
> + bool fence_cookie;
>
> job = p->job;
> p->job = NULL;
> @@ -1227,6 +1228,8 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
> */
> mutex_lock(&p->adev->notifier_lock);
>
> + fence_cookie = dma_fence_begin_signalling();
> +
> /* If userptr are invalidated after amdgpu_cs_parser_bos(), return
> * -EAGAIN, drmIoctl in libdrm will restart the amdgpu_cs_ioctl.
> */
> @@ -1264,12 +1267,14 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
> amdgpu_vm_move_to_lru_tail(p->adev, &fpriv->vm);
>
> ttm_eu_fence_buffer_objects(&p->ticket, &p->validated, p->fence);
> + dma_fence_end_signalling(fence_cookie);
Mhm, this could come earlier in theory. E.g. after pushing the job to
the scheduler.
Christian.
> mutex_unlock(&p->adev->notifier_lock);
>
> return 0;
>
> error_abort:
> drm_sched_job_cleanup(&job->base);
> + dma_fence_end_signalling(fence_cookie);
> mutex_unlock(&p->adev->notifier_lock);
>
> error_unlock:
next prev parent reply other threads:[~2020-05-13 7:02 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-12 8:59 [RFC 00/17] dma-fence lockdep annotations Daniel Vetter
2020-05-12 8:59 ` [RFC 01/17] dma-fence: add might_sleep annotation to _wait() Daniel Vetter
2020-05-12 9:08 ` Christian König
2020-06-02 9:45 ` Maarten Lankhorst
2020-05-12 8:59 ` [RFC 02/17] dma-fence: basic lockdep annotations Daniel Vetter
2020-05-12 12:09 ` Jason Gunthorpe
2020-05-12 12:57 ` Daniel Vetter
[not found] ` <158927426244.15653.14406159524439944950@build.alporthouse.com>
2020-05-12 9:08 ` Daniel Vetter
[not found] ` <158927519651.15653.17392305363363808831@build.alporthouse.com>
2020-05-13 8:30 ` Daniel Vetter
2020-05-25 15:41 ` Daniel Vetter
2020-05-26 10:00 ` Maarten Lankhorst
2020-05-28 13:36 ` Thomas Hellström (Intel)
2020-05-28 14:22 ` Daniel Vetter
2020-05-28 21:54 ` Luben Tuikov
2020-05-29 5:49 ` Daniel Vetter
2020-05-12 8:59 ` [RFC 03/17] dma-fence: prime " Daniel Vetter
2020-05-12 8:59 ` [RFC 04/17] drm/vkms: Annotate vblank timer Daniel Vetter
2020-05-12 8:59 ` [RFC 05/17] drm/vblank: Annotate with dma-fence signalling section Daniel Vetter
2020-05-12 8:59 ` [RFC 06/17] drm/atomic-helper: Add dma-fence annotations Daniel Vetter
2020-05-12 8:59 ` [RFC 07/17] drm/amdgpu: add dma-fence annotations to atomic commit path Daniel Vetter
2020-05-12 8:59 ` [RFC 08/17] drm/scheduler: use dma-fence annotations in main thread Daniel Vetter
2020-05-25 15:30 ` Daniel Vetter
2020-05-12 8:59 ` [RFC 09/17] drm/amdgpu: use dma-fence annotations in cs_submit() Daniel Vetter
2020-05-13 7:02 ` Christian König [this message]
2020-05-13 7:07 ` Daniel Vetter
2020-05-12 8:59 ` [RFC 10/17] drm/amdgpu: s/GFP_KERNEL/GFP_ATOMIC in scheduler code Daniel Vetter
2020-05-12 15:56 ` Christian König
2020-05-12 16:20 ` Daniel Vetter
2020-05-12 16:27 ` Daniel Vetter
2020-05-12 17:31 ` Christian König
2020-05-12 18:34 ` Daniel Vetter
2020-05-12 8:59 ` [RFC 11/17] drm/amdgpu: DC also loves to allocate stuff where it shouldn't Daniel Vetter
2020-05-12 8:59 ` [RFC 12/17] drm/amdgpu/dc: Stop dma_resv_lock inversion in commit_tail Daniel Vetter
2020-05-12 8:59 ` [RFC 13/17] drm/scheduler: use dma-fence annotations in tdr work Daniel Vetter
2020-05-12 8:59 ` [RFC 14/17] drm/amdgpu: use dma-fence annotations for gpu reset code Daniel Vetter
2020-05-12 8:59 ` [RFC 15/17] Revert "drm/amdgpu: add fbdev suspend/resume on gpu reset" Daniel Vetter
2020-05-12 8:59 ` [RFC 16/17] drm/amdgpu: gpu recovery does full modesets Daniel Vetter
2020-05-12 12:54 ` Alex Deucher
2020-05-12 12:58 ` Daniel Vetter
2020-05-12 13:12 ` Alex Deucher
2020-05-12 13:17 ` Daniel Vetter
2020-05-12 13:29 ` Alex Deucher
2020-05-12 13:45 ` Daniel Vetter
2020-05-12 14:24 ` Alex Deucher
2020-05-12 16:12 ` Daniel Vetter
2020-05-12 20:10 ` Kazlauskas, Nicholas
2020-05-13 6:02 ` Daniel Vetter
2020-05-12 8:59 ` [RFC 17/17] drm/i915: Annotate dma_fence_work Daniel Vetter
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=6cfd324e-0443-3a12-6a2c-25a546c68643@gmail.com \
--to=ckoenig.leichtzumerken@gmail.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=chris@chris-wilson.co.uk \
--cc=christian.koenig@amd.com \
--cc=daniel.vetter@ffwll.ch \
--cc=daniel.vetter@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=maarten.lankhorst@linux.intel.com \
/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).