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 8.0 has been released! | Please read ffmpeg.org/developer.html#Code-of-conduct
Traneptora_ has joined #ffmpeg-devel
Traneptora has quit [Read error: Connection reset by peer]
<JEEB>
haasn: fun case with the "this shouldn't require a buffer reset but apparently does for this user"
<wbs>
BtbN: oh, right - I'm not sure if I had the tab open for a long time inbetween - I don't think there was any pushes inbetween, but I can't really say totally for sure
minimal has quit [Remote host closed the connection]
GewoonLeon has quit [Quit: GewoonLeon]
GewoonLeon1 has joined #ffmpeg-devel
minimal has joined #ffmpeg-devel
GewoonLeon1 has quit [Client Quit]
GewoonLeon has joined #ffmpeg-devel
GewoonLeon has quit [Ping timeout: 256 seconds]
<haasn>
Lynne: I have a bizarre bug that only appears to happen on nvidia, where subsequent draws to the same texture appear to be blended instead of overwriting the previous contents; even though blending is disabled in the render pass
<haasn>
have you ever seen something like that?
<Lynne>
no, that sounds nuts
<ramiro>
michaelni: could you please explain where the magical expression from 8d381b57fd9d17fb5c3a851ca46c738b3afc33a2 comes from?
<ramiro>
michaelni: and how can we obtain/reproduce the sample file that triggered the timeout?
<ramiro>
michaelni: thanks. I'm having a hard time reproducing it. what command line did you use?
desmond-netint has joined #ffmpeg-devel
<michaelni>
ramiro, no bash history from 2018 but likely tools/target_dec_jpegls_fuzzer clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEGLS_fuzzer-5665032743419904 -timeout=10 or something like that
<jamrial>
mkver: you should probably poke michaelni about it
ccawley2011 has joined #ffmpeg-devel
iive has joined #ffmpeg-devel
Everything has quit [Read error: Connection reset by peer]
<ramiro>
michaelni: thanks. I managed to reproduce by creating my own main() for that file (I didn't want to have to use the full fuzzer)
<ramiro>
michaelni: but I still don't quite understand the commit. why couldn't we just check against max_pixels instead? the user should specify their own limit.
<michaelni>
ramiro, a check for max_pixels makes sense but a check for buf_size also makes sense because if you have a 50byte file thats not able to contain a 32000x32000 jpeg or sqrt(max_pixels)^2
lexano has quit [Read error: Connection reset by peer]
lexano has joined #ffmpeg-devel
ccawley2011 has quit [Ping timeout: 248 seconds]
<BtbN>
hm, building with msvc in wsl is weird. It selects cl.exe as hostcc, but also sets hostexesuf to empty
<BtbN>
--host-cc=gcc fixes it, but imo configure should figure either that out itself, or set the hostexesuf correctly
<BtbN>
not sure what the correct way to deal with this is. Should the probe_cc function detect that this compiler will for sure produce an exe, and just set the suffic?
<BtbN>
*suffix
^Neo_ has joined #ffmpeg-devel
^Neo has quit [Ping timeout: 256 seconds]
nate-m has quit [Ping timeout: 256 seconds]
Kimapr has joined #ffmpeg-devel
Kimapr_ has quit [Remote host closed the connection]