Builder ffmpegsos-solaris10-sparc Build #14235

Results:

Failed shell_2 shell_3 shell_4 shell_5

SourceStamp:

BuildSlave:

Reason:

The SingleBranchScheduler scheduler named 'schedule-ffmpegsos-solaris10-sparc' triggered this build

Steps and Logfiles:

1. git update ( 13 secs )
   1. stdio
2. shell 'gsed -i ...' ( 0 secs )
   1. stdio
3. shell_1 'gsed -i ...' ( 0 secs )
   1. stdio
4. shell_2 'gsed -i ...' failed ( 0 secs )
   1. stdio
5. shell_3 './configure --samples="../../../ffmpeg/fate-suite" ...' failed ( 12 secs )
   1. stdio
   2. config.log
6. shell_4 'gmake fate-rsync' failed ( 1 secs )
   1. stdio
7. shell_5 '../../../ffmpeg/fate.sh ../../../ffmpeg/fate_config_sos.sh' failed ( 2 secs )
   1. stdio
   2. configure.log
   3. compile.log
   4. test.log

Build Properties:

Forced Build Properties:

Responsible Users:

1. Michael Niedermayer

   michaelohnoyoudont@niedermayer.cc

Timing:

All Changes:

:

1. Change #285365

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:46:54
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	0260f625ba51ba9f02b189da757862fd96fb1da0

   Comments

   avformat/mov: make room in tts_data by the index entry count in mov_read_trun()
   Move the same range as for index_entries, tts_count is set to
   nb_index_entries right after.
   
   Fixes: out of array access
   Fixes: crash.mp4 / make-crash.py
   Fixes: srZp1wXh4CXQ
   Found-by: OpenAI Security Research
   Replicated through UnModified FFmpeg
   
   Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

   Changed files

   • libavformat/mov.c

2. Change #285366

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:46:54
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	5647739dddb2679ebfe488d6745f0fa1f405bc5d

   Comments

   avformat/mov: keep frag_index.current on its fragment in update_frag_index()
   The assertion failure was found during triage of the security report
   srZp1wXh4CXQ.
   
   Fixes: out of array access
   Fixes: Assertion index_entry_pos <= sti->nb_index_entries failed
   Fixes: crash.mp4 / make-crash.py
   Fixes: variant-assert.mp4 / make-variant-assert.py
   Fixes: srZp1wXh4CXQ
   Found-by: OpenAI Security Research
   Replicated through UnModified FFmpeg
   
   Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

   Changed files

   • libavformat/mov.c

3. Change #285367

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:46:54
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	5e8f97c467ca203ff01b34db7cee0e3496f4e99b

   Comments

   avformat/mov: zero tts_data between tts_count and nb_index_entries in mov_read_trun()
   Fixes: read of uninitialized memory
   Fixes: uninit.mp4 / gen-uninit.py
   Fixes: QcOY0fqViB2h
   Found during triage of the security report srZp1wXh4CXQ
   Replicated through UnModified FFmpeg
   
   Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

   Changed files

   • libavformat/mov.c

4. Change #285368

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:47:39
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	e3f5f17548bc7b773aefd420747863bc35741e37

   Comments

   avformat/mov: bound ipco and ipma by the iprp size
   This is hardening, no memory safety issue is known to depend on it.
   
   Found-by: OpenAI Security Research
   Reported in the security report srZp1wXh4CXQ
   
   Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

   Changed files

   • libavformat/mov.c

5. Change #285369

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:47:39
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	466ea98db4a040491d0fc248d55f8c44a86477e2

   Comments

   avformat/mov: do not seek back to cache ipco properties
   This could use ffio_ensure_seekback() but simply storing the 2 values
   is simpler
   
   no memory safety issue is known to depend on it.
   
   Found-by: OpenAI Security Research
   Reported in the security report srZp1wXh4CXQ
   
   Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

   Changed files

   • libavformat/mov.c

6. Change #285370

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:47:39
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	860267bd6288dfe8ec6c662dc75af955504257c8

   Comments

   avformat/mov: accept ipma associations with property_index 0
   ISO/IEC 23008-12 defines a property_index of 0 as no property being
   associated, but mov_read_iprp() rejected it, and with it the whole file.
   
   Found during triage of the security report srZp1wXh4CXQ
   
   Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>

   Changed files

   • libavformat/mov.c

7. Change #285371

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:53:46
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	8ba5dc011b304781eade0a1ed057f1af80fceeb7

   Comments

   avcodec/dcaenc: compute the LFE FIR table in double precision
   0xffffff800000 is not representable as a float, so the float product
   rounds differently with SSE arithmetic than under x87 excess precision
   and the encoded LFE channel differed between x86-64 and x86-32. The
   product is exact in double, which makes the table identical everywhere.

   Changed files

   • libavcodec/dcaenc.c

8. Change #285372

   Category	ffmpeg
   Changed by	Michael Niedermayer <michaelohnoyoudont@niedermayer.cc>
   Changed at	Sun 11 Oct 2026 00:53:46
   Repository	https://git.ffmpeg.org/ffmpeg.git
   Project	ffmpeg
   Branch	master
   Revision	8d8c75f2b0c33adf968b54223dd8dc1c3d94c87f

   Comments

   fate/acodec: add a 5.1 DCA encoding test covering the LFE path
   The existing dca encoder tests only use a stereo source, so the LFE
   history handling in shift_history() was not covered. Add a test
   encoding tests/data/asynth-44100-6.wav (5.1), its output differs
   without f391d256fd.
   
   Assisted-by: Fairy

   Changed files

   • tests/fate/acodec.mak