Bug 500944 – =media-libs/x264-0.0.20111220 segmentation faults during initialization of x264 encoder in x264_cqm_init () from /usr/lib/libx264.so.120

• Status: RESOLVED FIXED
• Product: Gentoo Linux

[Description Raymond Jennings 2014-02-11 07:02:56 UTC]

vlc causes a segfault while trying to convert an mp4 movie.

Reproducible: Always

Actual Results:  
crash

Expected Results:  
converted movie

[Comment 1 Raymond Jennings 2014-02-11 07:04:11 UTC]

Created attachment 370122 [details]
gdb stack dump

[Comment 2 Jeroen Roovers (RETIRED) 2014-02-11 11:56:19 UTC]

Please post your `emerge --info media-video/vlc media-libs/x264' output in a comment.

[Comment 3 Raymond Jennings 2014-02-11 21:16:52 UTC]

Created attachment 370178 [details]
emerge --info for vlc and x264

[Comment 4 Tom Wijsman (TomWij) (RETIRED) 2014-02-19 16:41:02 UTC]

> #0  0xb75b7222 in ?? () from /lib/libc.so.6
> #0  0xb75b7222 in ?? () from /lib/libc.so.6

Can you build sys-libs/glibc with debugging symbols and then try again?

[Comment 5 Raymond Jennings 2014-02-19 19:12:45 UTC]

Uh...I did.

[Comment 6 Raymond Jennings 2014-02-19 19:14:20 UTC]

to be precise I put -ggdb3 on the CFLAGS for glibc

[Comment 7 Tom Wijsman (TomWij) (RETIRED) 2014-02-21 14:51:31 UTC]

What have you put in CFLAGS for media-libs/x264 and media-video/vlc?

[Comment 8 Tom Wijsman (TomWij) (RETIRED) 2014-02-22 14:11:45 UTC]

While researching bug #499996, I came across a solution for this bug (See Als); it thus seems GCC 4.8 is taking more toll than we think, I'm reassigning this as this is x264's terrain given that it causes more reverse dependencies to fail.

[Comment 9 Raymond Jennings 2014-02-22 20:59:18 UTC]

For the record, my make.conf cflags is "-ggdb3 -O2" plus the pipe stuff

[Comment 10 Raymond Jennings 2014-02-22 21:02:20 UTC]

Also for upstream, wouldn't it be better to fix x264 to remove the undefined behavior than to bandaid the build process?

[Comment 11 Alexander Tsoy 2014-10-14 11:08:32 UTC]

Isn't this already fixed in current stable x264-0.0.20140308 ?

[Comment 12 Anthony Basile 2014-10-18 17:05:42 UTC]

(In reply to Alexander Tsoy from comment #11)
> Isn't this already fixed in current stable x264-0.0.20140308 ?

0.0.20140308 has been marked stable.  Can someone test on an i686 and see if the original segfault is still reproduceable.  I'm not in a position to test this.

[Comment 13 Anthony Basile 2014-10-18 21:21:37 UTC]

(In reply to Anthony Basile from comment #12)
> (In reply to Alexander Tsoy from comment #11)
> > Isn't this already fixed in current stable x264-0.0.20140308 ?
> 
> 0.0.20140308 has been marked stable.  Can someone test on an i686 and see if
> the original segfault is still reproduceable.  I'm not in a position to test
> this.

Okay I've answered my own question.  This is fixed.  The patch in the URL has been applied to x264-0.0.20140308.

If I'm wrong, please go ahead and reopen this.