[FFmpeg-devel] [PATCH 1/2] lavfi/video: ensure that filter_frame is called.

[Michael Niedermayer]

On Thu, Dec 27, 2012 at 08:32:12AM +0100, Nicolas George wrote:
> Le septidi 7 nivôse, an CCXXI, Michael Niedermayer a écrit :
> > breaks fate:
> > 
> > --- ./tests/ref/lavfi/tile  2012-12-23 16:33:21.097039227 +0100
> > +++ tests/data/fate/lavfi-tile  2012-12-27 01:01:32.011142221 +0100
> > @@ -1 +1 @@
> > -tile                bfd98570151154d078b14398f867f0e8
> > +tile
> > \ No newline at end of file
> > Test lavfi-tile failed. Look at tests/data/fate/lavfi-tile.err for details.
> > make: *** [fate-lavfi-tile] Error 139
> > make: *** Waiting for unfinished jobs....
> 
> I can not reproduce this here: fate-lavfi-tile passes (both after patch 1/2
> and 2/2), as IIRC does all FATE with both patches.

after 1/2: (same happens with or without 2/2, trace is from without)
make fate-lavfi-tile
==27913== Invalid read of size 8
==27913==    at 0x4BE036: ff_copy_rectangle2 (drawutils.c:209)
==27913==    by 0x4A86CE: filter_frame (vf_tile.c:196)
==27913==    by 0x4ACA9E: ff_end_frame (video.c:350)
==27913==    by 0x488553: ff_filter_frame (avfilter.c:645)
==27913==    by 0x48B091: request_frame (buffersrc.c:372)
==27913==    by 0x48B2C5: av_buffersrc_add_ref (buffersrc.c:150)
==27913==    by 0x4733BF: output_packet (ffmpeg.c:1672)
==27913==    by 0x474260: process_input (ffmpeg.c:2886)
==27913==    by 0x464AD6: main (ffmpeg.c:2982)
==27913==  Address 0x8 is not stack'd, malloc'd or (recently) free'd


[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Complexity theory is the science of finding the exact solution to an
approximation. Benchmarking OTOH is finding an approximation of the exact
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20121227/fcac4058/attachment.asc>