michaelni changed the topic of #ffmpeg-devel to: Welcome to the FFmpeg development channel | Questions about using FFmpeg or developing with libav* libs should be asked in #ffmpeg | This channel is publicly logged | FFmpeg 7.1 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
cone-839 has joined #ffmpeg-devel
<cone-839>
ffmpeg Peter Ross master:86dd15fd0dd0: avformat/mlvdec: demux LJ92 huffman comressed frames
<cone-839>
ffmpeg Peter Ross master:45c30bc51e52: avformat/mlvdec: process VERS block
<cone-839>
ffmpeg Peter Ross master:330470ef247b: avformat/mlvdec: skip over some other known block types
<cone-839>
ffmpeg Peter Ross master:ba22d6a24f84: avformat/wtvenc: do not output negative 'third timestamp' field
Xafg has quit [Quit: Client closed]
<cone-839>
ffmpeg Peter Ross master:2202dcfc0fbe: avformat/iff: ignore FVER tag when not processing DSD/DST files
<fflogger>
[editedticket] pross: Ticket #3659 ([avcodec] Some H.264 WTV files created by ffmpeg do not playback on WMP and MCE) updated https://trac.ffmpeg.org/ticket/3659#comment:6
kepstin has quit [Remote host closed the connection]
kepstin has joined #ffmpeg-devel
<beastd>
BtbN: yeah. many new issues in coverity. if it's not false positive it would be probably good to get it sorted out. only had a very quick look, but some of those look real.
<beastd>
mabye coverity got an update to detect those (in more cirsumstances). didn't seem related to recent changes to me
ccawley2011 has joined #ffmpeg-devel
<beastd>
welder: about ffmpeg assembly course, be careful about the discord link: one expired and was immedieately regrabbed by some crypto scam.
<beastd>
welder: kierank is in charge of that and probably can tell you where to start. i'm not sure about the state of the effort atm.
<beastd>
at least 3 to 4 lessons were mostly completed AFAIR
mkver has joined #ffmpeg-devel
novaphoenix has quit [Quit: i quit]
novaphoenix has joined #ffmpeg-devel
^Neo has joined #ffmpeg-devel
jamrial has joined #ffmpeg-devel
ccawley2011_ has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 248 seconds]
realies has quit [Quit: ~]
realies has joined #ffmpeg-devel
realies has quit [Quit: ~]
^Neo has quit [Ping timeout: 260 seconds]
Traneptora has quit [Ping timeout: 252 seconds]
cone-326 has joined #ffmpeg-devel
<cone-326>
ffmpeg James Almer master:df50370e1b10: avformat/matroska: add support for VVC streams
<cone-326>
ffmpeg James Almer master:5f6e43f77d38: tests/fate/lavf-container: fix dependencies for lavf_container_fate tests
Marth64 has joined #ffmpeg-devel
<mkver>
jamrial: 5f6e43f77d38f was not on the ML. Would you please give a configure-line where the dependencies were wrong?
<mkver>
Wait, this is for the "fate-lavf-fate-*" tests (why are they named that way).
<cone-326>
ffmpeg Andreas Rheinhardt master:7a5d6690fc54: avcodec/aac_ac3_parser: Remove unused variable
<cone-326>
ffmpeg Andreas Rheinhardt master:b239eefeb524: avcodec/sga: Silence -Wunused-but-set-variable warnings
<cone-326>
ffmpeg Andreas Rheinhardt master:0bc1d2fac4d6: avcodec/aac/aacdec: Fix -Wdeclaration-after-statement
<cone-326>
ffmpeg Andreas Rheinhardt master:4973bb661edd: swscale/cms,graph,lut3d: Use ff_-prefix, don't export internal functions
<cone-326>
ffmpeg Andreas Rheinhardt master:4cf445c75d19: tests/ref/fate/png-icc-parse: Add lossless flag to ref file
<cone-326>
ffmpeg Andreas Rheinhardt master:9bbab3255b7d: avfilter/vf_overlay: Remove spec-incompliant ';'
<cone-326>
ffmpeg Andreas Rheinhardt master:7ab7d9c3c06c: avfilter/avfilter: Add FFFilter, hide internals of AVFilter
<cone-326>
ffmpeg Andreas Rheinhardt master:7822beed85f1: avfilter/vf_xpsnr: Fix leaks
<cone-326>
ffmpeg Andreas Rheinhardt master:e540381f9108: avfilter/vf_xpsnr: Avoid array only one of whose elements is used
<cone-326>
ffmpeg Andreas Rheinhardt master:5a72266d49a7: tests/checkasm/sw_rgb: Fix leaks
<jamrial>
mkver: they did not take into account the post muxing demux part of the tests
<jamrial>
also, DEMMUX includes the file protocol
<mkver>
I was initially surprised because 555c31de9da0ef should already have dealt with this; but then I noticed that these are the fate-lavf-fate-* tests (why are they named that way?) which I didn't touch.
<mkver>
Was there a change in the rule allowing commits to bypass the ML again?
<jamrial>
they are named that way because they depend on fate samples
<mkver>
Then I'd expect a "samples" and not a second "fate" in their name.
<jamrial>
yeah, that may have been a better name
<beastd>
my understanding is that in ffmpeg all patches should be posted on ffmpeg-devel first. could be pushed without review, but only after a push warning with reasonable timing. if it's security related, the timing may be shorter. guess we should document it better? or is it already?
<mkver>
beastd: IMO your take on the current rules is correct.
<mkver>
Great: if librsvg is enabled, libavcodec indirectly inherits a -Wl,--export-dynamic from gmodule, so that all non-static symbols (except those marked as private) are exported from the executables.
<jamrial>
mkver: taken from the pkgconfig cflags?
<mkver>
Yes
abdu has joined #ffmpeg-devel
<BtbN>
Well, librsvg is a bit of a catastrophe anyway
<mkver>
Actually, it is from pkgconfig's "Libs", not Cflags.
<BtbN>
I'm not including it in my builds cause it'd single handedly DOUBLE the binary size
<mkver>
My local x86_32 builds (with GCC) are affected.
<beastd>
mkver: i fear the pkgconfig stuff will probably always be a bit scary. does it even make sense that this flag is propagated via pkgconfig libs?
<beastd>
ugh! > On Linux and Unix, this is not necessary; applications should instead be compiled with the -Wl,--export-dynamic argument inside their compiler flags, and linked against gmodule-export-2.0.
<beastd>
standard advise: expose everything than you won't miss anything. what could possibly go wrong?
<jamrial>
mkver: does adding -msse2 -mfpmath=sse help for you?
<mkver>
jamrial: -mfpmath=sse -msse fixes the failing tests; but now filter-afade-esin fails. It didn't fail before.
<jamrial>
great
<jamrial>
the joys of checksum tests for float stuff
<mkver>
The checksum for the packet with pts 86016 is different: 0xf363fac7 vs 0x33f4facd
abdu77 has joined #ffmpeg-devel
abdu has quit [Ping timeout: 240 seconds]
Marth64 has quit [Quit: Leaving]
novaphoenix has quit [Quit: i quit]
ccawley2011_ has quit [Read error: Connection reset by peer]
ccawley2011_ has joined #ffmpeg-devel
novaphoenix has joined #ffmpeg-devel
ccawley2011_ has quit [Read error: Connection reset by peer]