Hi Paul, Thanks for the review.
Are you saying that, in the event that any memory allocation fails within the demuxer, the demuxer is responsible for cleaning-up already previously allocated memory so that other parts of the pipeline can continue operating? In other words, ffmpeg is not expected to immediately die if memory allocation fails in one demuxer/decoder/muxer? I just want to make sure I understand the feedback. Best, -- Pierre On Mon, Nov 22, 2021 at 3:19 AM Paul B Mahol <[email protected]> wrote: > > > The realloc calls still leaks on error. _______________________________________________ ffmpeg-devel mailing list [email protected] https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
