| Message ID | 201609241528.36266.cehoyos@ag.or.at |
|---|---|
| State | Rejected |
| Headers | show |
On Sat, Sep 24, 2016 at 03:28:36PM +0200, Carl Eugen Hoyos wrote: > Hi! > > Attached patch hopefully fixes ticket #5136. > > Please review, Carl Eugen > utils.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > 519f209901429efaf9dffec381290a6a83573deb 0001-lavf-utils-Do-not-overflow-in-update_initial_timesta.patch > From 09a31f8f6584f0c97443fd6edc7fcf7142b6a6ff Mon Sep 17 00:00:00 2001 > From: Carl Eugen Hoyos <cehoyos@ag.or.at> > Date: Sat, 24 Sep 2016 15:26:21 +0200 > Subject: [PATCH] lavf/utils: Do not overflow in update_initial_timestamps(). > > Fixes ticket #5136. > --- > libavformat/utils.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) probably ok it might be ultimately easier to reject demuxer output that has timestamps close to INT64 MAX/MIN instead of protecting all computations one by one though thx [...]
2016-09-25 20:59 GMT+02:00 Michael Niedermayer <michael@niedermayer.cc>: > On Sat, Sep 24, 2016 at 03:28:36PM +0200, Carl Eugen Hoyos wrote: >> Hi! >> >> Attached patch hopefully fixes ticket #5136. >> >> Please review, Carl Eugen > >> utils.c | 4 +++- >> 1 file changed, 3 insertions(+), 1 deletion(-) >> 519f209901429efaf9dffec381290a6a83573deb >> 0001-lavf-utils-Do-not-overflow-in-update_initial_timesta.patch >> From 09a31f8f6584f0c97443fd6edc7fcf7142b6a6ff Mon Sep 17 00:00:00 2001 >> From: Carl Eugen Hoyos <cehoyos@ag.or.at> >> Date: Sat, 24 Sep 2016 15:26:21 +0200 >> Subject: [PATCH] lavf/utils: Do not overflow in >> update_initial_timestamps(). >> >> Fixes ticket #5136. >> --- >> libavformat/utils.c | 4 +++- >> 1 file changed, 3 insertions(+), 1 deletion(-) > > probably ok > it might be ultimately easier to reject demuxer output that has > timestamps close to INT64 MAX/MIN instead of protecting all > computations one by one though I don't disagree but I am probably unable to improve the patch. Do you want me to apply? Carl Eugen
On Sun, Sep 25, 2016 at 09:04:30PM +0200, Carl Eugen Hoyos wrote: > 2016-09-25 20:59 GMT+02:00 Michael Niedermayer <michael@niedermayer.cc>: > > On Sat, Sep 24, 2016 at 03:28:36PM +0200, Carl Eugen Hoyos wrote: > >> Hi! > >> > >> Attached patch hopefully fixes ticket #5136. > >> > >> Please review, Carl Eugen > > > >> utils.c | 4 +++- > >> 1 file changed, 3 insertions(+), 1 deletion(-) > >> 519f209901429efaf9dffec381290a6a83573deb > >> 0001-lavf-utils-Do-not-overflow-in-update_initial_timesta.patch > >> From 09a31f8f6584f0c97443fd6edc7fcf7142b6a6ff Mon Sep 17 00:00:00 2001 > >> From: Carl Eugen Hoyos <cehoyos@ag.or.at> > >> Date: Sat, 24 Sep 2016 15:26:21 +0200 > >> Subject: [PATCH] lavf/utils: Do not overflow in > >> update_initial_timestamps(). > >> > >> Fixes ticket #5136. > >> --- > >> libavformat/utils.c | 4 +++- > >> 1 file changed, 3 insertions(+), 1 deletion(-) > > > > probably ok > > it might be ultimately easier to reject demuxer output that has > > timestamps close to INT64 MAX/MIN instead of protecting all > > computations one by one though > > I don't disagree but I am probably unable to improve the patch. > Do you want me to apply? ive posted an alternative [...]
2016-09-25 23:55 GMT+02:00 Michael Niedermayer <michael@niedermayer.cc>: >> > probably ok >> > it might be ultimately easier to reject demuxer output that has >> > timestamps close to INT64 MAX/MIN instead of protecting all >> > computations one by one though >> >> I don't disagree but I am probably unable to improve the patch. >> Do you want me to apply? > > ive posted an alternative Looks much simpler, thank you! Carl Eugen
diff --git a/libavformat/utils.c b/libavformat/utils.c index 93ea6ff..4362a12 100644 --- a/libavformat/utils.c +++ b/libavformat/utils.c @@ -1069,7 +1069,9 @@ static void update_initial_timestamps(AVFormatContext *s, int stream_index, if (st->first_dts != AV_NOPTS_VALUE || dts == AV_NOPTS_VALUE || st->cur_dts == AV_NOPTS_VALUE || - is_relative(dts)) + is_relative(dts) || + INT64_MIN + RELATIVE_TS_BASE > st->cur_dts || + (dts > 0 ? INT64_MAX - dts < st->cur_dts - RELATIVE_TS_BASE : INT64_MIN - dts > st->cur_dts - RELATIVE_TS_BASE)) return; st->first_dts = dts - (st->cur_dts - RELATIVE_TS_BASE);
Hi! Attached patch hopefully fixes ticket #5136. Please review, Carl Eugen From 09a31f8f6584f0c97443fd6edc7fcf7142b6a6ff Mon Sep 17 00:00:00 2001 From: Carl Eugen Hoyos <cehoyos@ag.or.at> Date: Sat, 24 Sep 2016 15:26:21 +0200 Subject: [PATCH] lavf/utils: Do not overflow in update_initial_timestamps(). Fixes ticket #5136. --- libavformat/utils.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)