[FFmpeg-cvslog] r16365 - in trunk/libavformat: rm.h rmdec.c

Ronald S. Bultje rsbultje
Sun Dec 28 15:21:00 CET 2008


Hi,

On Sun, Dec 28, 2008 at 5:40 AM, Vitor Sessak <vitor1001 at gmail.com> wrote:
> This commit or the next one broke real-14_4 and real-rv40 FATE tests on
> Linux (see, for example,
> http://fate.multimedia.cx/index.php?test_result=6416893 ).

For the 14k4: the MD5 is the same [1], and output sounds good. So how
do I know what made the test fail? Is it stderr text?
For the video file, the sample's short and truncated and causes an
abort, so I'm assuming it's because of the double free on stream end
which I referred to in another mailing thread. Let's see after I
commit that...

Thanks for noticing,
Ronald

[1] http://fate.multimedia.cx/index.php?test_result=6415205
Actual stdout text:
37e64339cf728911991f4b045ef452e8
Expected stdout text:
37e64339cf728911991f4b045ef452e8




More information about the ffmpeg-cvslog mailing list