From: Ilija Hadzic <ihadzic@research.bell-labs.com>
To: airlied@gmail.com, dri-devel@lists.freedesktop.org
Subject: fix for CRTC mutex corruption
Date: Tue, 29 Oct 2013 11:09:40 -0400 [thread overview]
Message-ID: <1383059386-20917-1-git-send-email-ihadzic@research.bell-labs.com> (raw)
The following patch series fixes the mutex corruption problem
due to bit-copying of drm_crtc structure that happens when
drm_crtc_helper_set_config function takes the error-recovery
path. The patch series is the alternative solution for the
patch that was proposed and NACK-ed two weeks ago [1].
The actual fix is implemented in patch #6; preceding
5 patches are necessary prerequisites.
A couple of my own remarks:
1) Someone (including myself) may be tempted to eliminate the
bit-copy for encoder and connector structures as well. I would, however,
prefer to leave that improvement for a different patch series.
The primary reason is that this patch series addresses an acute
(and serious) problem (mutex corruption), while doing the equivalent
rework for connector and encoder structure would be only for the sake
of improving the code quality.
2) The problem exists in old (stable@...) kernels and my original intention
(in the patch that was NACK-ed [1]) was to make the fix simple enough
to be eligible for stable@.... This patch series is now probably more complex
than what stable@... may be willing to accept. So the question in my mind
is what we should do with the old kernels.
3) This patch series does not include the fix for incidental finding
described in earlier discussions about this problem [2] because it's not
the part of the fix that this patch series is targeting, so I'd leave
it for later.
regards,
Ilija
[1] http://lists.freedesktop.org/archives/dri-devel/2013-October/047479.html
[2] http://lists.freedesktop.org/archives/dri-devel/2013-October/047557.html
next reply other threads:[~2013-10-29 15:15 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-29 15:09 Ilija Hadzic [this message]
2013-10-29 15:09 ` [PATCH 1/6] drm: remove redundant if statement Ilija Hadzic
2013-10-29 15:09 ` [PATCH 2/6] drm: eliminate old_fb from drm_crtc_helper_set_config Ilija Hadzic
2013-10-29 15:09 ` [PATCH 3/6] drm: restore crtc origin if mode_set_base fails Ilija Hadzic
2013-10-29 15:09 ` [PATCH 4/6] drm: fix error recovery path in drm_crtc_helper_set_mode Ilija Hadzic
2013-10-29 15:09 ` [PATCH 5/6] drm: do not set crtc enabled field twice Ilija Hadzic
2013-10-29 15:09 ` [PATCH 6/6] drm: eliminate bit-copy restoration of crtc Ilija Hadzic
2013-10-29 18:40 ` fix for CRTC mutex corruption Daniel Vetter
2013-10-29 19:22 ` Ilija Hadzic
2013-10-29 20:11 ` Daniel Vetter
2013-10-30 13:45 ` Ilija Hadzic
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=1383059386-20917-1-git-send-email-ihadzic@research.bell-labs.com \
--to=ihadzic@research.bell-labs.com \
--cc=airlied@gmail.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 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).