On Thu, 2 Nov 2017, Sean McGovern wrote:
Hi,
On Nov 2, 2017 02:50, "Martin Storsjö" <[email protected]> wrote:
On Wed, 1 Nov 2017, Sean McGovern wrote:
Regression since 569d18aa9dc989c37bb4d4b968026fe5afa6fff9.
Bug-Id: 1055
Cc: [email protected]
---
libavformat/matroskadec.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavformat/matroskadec.c b/libavformat/matroskadec.c
index 9c523fb..bad750c 100644
--- a/libavformat/matroskadec.c
+++ b/libavformat/matroskadec.c
@@ -1919,7 +1919,7 @@ static int matroska_parse_tracks(AVFormatContext *s)
track->audio.sub_packet_h = avio_rb16(&b);
track->audio.frame_size = avio_rb16(&b);
track->audio.sub_packet_size = avio_rb16(&b);
- if (flavor <= 0 ||
+ if (flavor < 0 ||
track->audio.coded_framesize <= 0 ||
track->audio.sub_packet_h <= 0 ||
track->audio.frame_size <= 0 ||
--
2.7.4
Possibly ok (the commit message of the offending commit doesn't indicate
which sample it was about so I can't easily check whether that sample
needed to check for strictly flavor <= 0 or not).
// Martin
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
I was probably half-asleep when I wrote the commit message for this. It
should really be something like:
matroskadec: allow RealAudio/Cook/Sipro streams of flavor 0
...Or something similar (suggestions?)
The sample used to fix this issue is from BZ #268 [1].
What I meant was; the original commit,
569d18aa9dc989c37bb4d4b968026fe5afa6fff9, tried to fix some crash with
some fuzzed input files. We should ideally make sure those crashes stay
fixed. Since I didn't write out which sample it was about, it's not very
easy to find out now which one it might have been though, to verify we
don't reopen that crash bug.
// Martin
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
