All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dan Carpenter <dan.carpenter@oracle.com>
To: andrey.grodzovsky@amd.com
Cc: dri-devel@lists.freedesktop.org
Subject: [bug report] drm/sched: Set error to s_fence if HW job submission failed.
Date: Tue, 29 Oct 2019 17:39:14 +0300	[thread overview]
Message-ID: <20191029143914.GA19410@mwanda> (raw)

Hello Andrey Grodzovsky,

The patch e91e5f080e03: "drm/sched: Set error to s_fence if HW job
submission failed." from Oct 24, 2019, leads to the following static
checker warning:

	drivers/gpu/drm/scheduler/sched_main.c:500 drm_sched_resubmit_jobs()
	warn: passing zero to 'PTR_ERR'

drivers/gpu/drm/scheduler/sched_main.c
   477  void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched)
   478  {
   479          struct drm_sched_job *s_job, *tmp;
   480          uint64_t guilty_context;
   481          bool found_guilty = false;
   482          struct dma_fence *fence;
   483  
   484          list_for_each_entry_safe(s_job, tmp, &sched->ring_mirror_list, node) {
   485                  struct drm_sched_fence *s_fence = s_job->s_fence;
   486  
   487                  if (!found_guilty && atomic_read(&s_job->karma) > sched->hang_limit) {
   488                          found_guilty = true;
   489                          guilty_context = s_job->s_fence->scheduled.context;
   490                  }
   491  
   492                  if (found_guilty && s_job->s_fence->scheduled.context == guilty_context)
   493                          dma_fence_set_error(&s_fence->finished, -ECANCELED);
   494  
   495                  dma_fence_put(s_job->s_fence->parent);
   496                  fence = sched->ops->run_job(s_job);
   497  
   498                  if (IS_ERR_OR_NULL(fence)) {
                            ^^^^^^^^^^^^^^^^^^^^

   499                          s_job->s_fence->parent = NULL;
   500                          dma_fence_set_error(&s_fence->finished, PTR_ERR(fence));
                                                                        ^^^^^^^^^^^^^^
If "fence" is NULL then it triggers a WARN_ON() inside the
dma_fence_set_error() function.

	WARN_ON(error >= 0 || error < -MAX_ERRNO);
                ^^^^^^^^^^

   501                  } else {
   502                          s_job->s_fence->parent = fence;
   503                  }
   504  
   505  
   506          }
   507  }

regards,
dan carpenter
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

             reply	other threads:[~2019-10-29 14:39 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-29 14:39 Dan Carpenter [this message]
2019-10-29 14:41 ` [bug report] drm/sched: Set error to s_fence if HW job submission failed Dan Carpenter

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=20191029143914.GA19410@mwanda \
    --to=dan.carpenter@oracle.com \
    --cc=andrey.grodzovsky@amd.com \
    --cc=dri-devel@lists.freedesktop.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.