All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
To: Dan Carpenter <dan.carpenter@oracle.com>
Cc: dri-devel@lists.freedesktop.org
Subject: Re: drm: fix blob pointer check
Date: Fri, 11 Mar 2016 11:52:44 +0000	[thread overview]
Message-ID: <56E2B18C.8060005@intel.com> (raw)
In-Reply-To: <20160311113903.GA12675@mwanda>

Hi Dan,

Thanks a lot for pointing this out. I saw you previous comment but 
didn't realize the issue.
I'll set the pointer to NULL.

I don't if there is an agreement on this, but do you think the 
unref/free functions should check for error pointers?

-
Lionel

On 11/03/16 11:39, Dan Carpenter wrote:
> Hi Lionel Landwerlin,
>
> The patch 562c5b4d8986: "drm: fix blob pointer check" from Mar 10,
> 2016, has a problem.
>
> drivers/gpu/drm/drm_atomic_helper.c
>    2924          blob = drm_property_create_blob(dev,
>    2925                                          sizeof(struct drm_color_lut) * size,
>    2926                                          NULL);
>    2927          if (IS_ERR(blob)) {
>    2928                  ret = PTR_ERR(blob);
>    2929                  goto fail;
>
> These types of goto fails are a trap for the unwary.
>
> I deliberately reported the bug instead of fixing it because I am a jerk
> and because last time when I did this jerky thing, people were not
> convinced that they would fall for the trap every single time.
>
> http://www.spinics.net/lists/cgroups/msg15262.html
>
>    2930          }
>    2931
>
> [ snip ]
>
>    2973  fail:
>    2974          if (ret == -EDEADLK)
>    2975                  goto backoff;
>    2976
>    2977          drm_atomic_state_free(state);
>    2978          drm_property_unreference_blob(blob);
>                                                ^^^^
> Blob is an error pointer here so it will oops inside the function call.
>
> The better way to write this is to unwind in the reverse order from the
> allocations.  So since we allocated state first then we free it last.
> Use explicit names based on what the goto does.  err_unreference:
> err_free_state:.  Don't free things that haven't been allocated.
>
> Or you could set "blob = NULL;" before the goto.
>
>    2979
>    2980          return;
>    2981  backoff:
>    2982          drm_atomic_state_clear(state);
>    2983          drm_atomic_legacy_backoff(state);
>    2984
>    2985          goto retry;
>    2986  }
>
> regards,
> dan carpenter
>

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

  reply	other threads:[~2016-03-11 11:52 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-11 11:39 drm: fix blob pointer check Dan Carpenter
2016-03-11 11:52 ` Lionel Landwerlin [this message]
2016-03-11 12:01   ` David Herrmann
2016-03-14  9:06   ` 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=56E2B18C.8060005@intel.com \
    --to=lionel.g.landwerlin@intel.com \
    --cc=dan.carpenter@oracle.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.