Re: [FFmpeg-devel] [PATCH]libavformat/http: fix http error eof

2017-05-23 10:47 GMT+08:00 raymond zheng : > 1. An exception occurred in the CDN edge server, that will lead to close > the http connection. > > 2. Because http is disconnected, so ffurl_read will return 0 > > 3. Avformat will consider I/O is eof > > 4. Right now http

Re: [FFmpeg-devel] [PATCH]libavformat/http: fix http error eof

1. An exception occurred in the CDN edge server, that will lead to close the http connection. 2. Because http is disconnected, so ffurl_read will return 0 3. Avformat will consider I/O is eof 4. Right now http is actually disconnect abnormally, it should return to ERROR, rather than return to

Re: [FFmpeg-devel] [PATCH] avcodec/takdec: Fix multiple runtime error: signed integer overflow: -512 * 4563386 cannot be represented in type 'int'

On Sat, May 20, 2017 at 11:06:45PM +0200, Michael Niedermayer wrote: > On Sat, May 20, 2017 at 07:26:50PM +0200, Marton Balint wrote: > > > > On Sat, 20 May 2017, Michael Niedermayer wrote: > > > > >Found-by: continuous fuzzing process > >

Re: [FFmpeg-devel] [PATCH] avfilter: Enable in MP4 container both AMR-NB and AMR-WB

I resolved our issue with the MOV changes and now have H.264 / AMR-NB and H.264 / AMR-WB recording to MOV using non-patched FFmpeg. The former plays in both QuickTime and the Windows 10 'Movies & TV' app, video and audio. The MOV containing AMR-WB plays video only in QuickTime and neither in the

Re: [FFmpeg-devel] [PATCH 2/2] lavc/ffjni: add missing '\n'

On Sun, May 21, 2017 at 06:56:42PM +0200, Matthieu Bouron wrote: > On Sun, May 21, 2017 at 06:36:10PM +0200, Carl Eugen Hoyos wrote: > > 2017-05-21 18:17 GMT+02:00 Matthieu Bouron : > > > > > -av_log(log_ctx, AV_LOG_ERROR, "Failed to get the JNI environment > >

Re: [FFmpeg-devel] [PATCH 2/3] avcodec/mlpdec: Fix runtime error: shift exponent -5 is negative

On Sat, May 20, 2017 at 11:01:03PM +0200, Michael Niedermayer wrote: > Fixes part of 1708/clusterfuzz-testcase-minimized-5035111957397504 > > Found-by: continuous fuzzing process > https://github.com/google/oss-fuzz/tree/master/targets/ffmpeg > Signed-off-by: Michael Niedermayer

Re: [FFmpeg-devel] [PATCH] lavf/concatdec: do not transfer custom IO flag

On Mon, May 22, 2017 at 05:11:34PM +0200, wm4 wrote: > On Mon, 22 May 2017 16:54:31 +0200 > Clément Bœsch wrote: > > > From: Clément Bœsch > > > > If the source is using a custom IO, setting this flag causes heavy leaks > > since the segments will not have their

Re: [FFmpeg-devel] [PATCH] avfilter: Enable in MP4 container both AMR-NB and AMR-WB

The MOV issue I am working on is likely an issue with our framework. I hope to have this resolved soon. Assuming MOV works, we still would like to use MP4 with AMR NB/WB. What is the difficulty with enabling this? Seems it has been updated to accept Opus recently. Bob On Fri, May 19, 2017 at

Re: [FFmpeg-devel] [PATCH 1/3] avcodec/escape124: Check depth against num_superblocks

On Sat, May 20, 2017 at 11:01:02PM +0200, Michael Niedermayer wrote: > Fixes: runtime error: left shift of 66184 by 15 places cannot be represented > in type 'int' > Fixes: 1707/clusterfuzz-testcase-minimized-6502767008940032 > > Found-by: continuous fuzzing process >

Re: [FFmpeg-devel] [PATCH 3/3] avcodec/mlpdec: Check quant_step_size against huff_lsbs

On Sun, May 21, 2017 at 01:42:18PM +0200, wm4 wrote: > On Sat, 20 May 2017 23:01:04 +0200 > Michael Niedermayer wrote: > > > This reorders the operations so as to avoid computations with the above > > arguments > > before they have been initialized. > > Fixes part of

Re: [FFmpeg-devel] [PATCH] lavf/concatdec: do not transfer custom IO flag

On Mon, 22 May 2017 16:54:31 +0200 Clément Bœsch wrote: > From: Clément Bœsch > > If the source is using a custom IO, setting this flag causes heavy leaks > since the segments will not have their avio context closed. > > Regression since

Re: [FFmpeg-devel] [PATCH]libavformat/http: fix http error eof

2017-05-22 22:36 GMT+08:00 raymond zheng : > I don't think it need a timeout event to disconnect the link, because > when ffurl_read > return to 0, it means the link disconnect. If s->off < target_end, it > means AVERROR, > otherwise, it's normal eof. > I don't use

[FFmpeg-devel] [PATCH] lavf/concatdec: do not transfer custom IO flag

From: Clément Bœsch If the source is using a custom IO, setting this flag causes heavy leaks since the segments will not have their avio context closed. Regression since f5da453b068f55d335ca403d2e2b4dd2ac3d4331. --- libavformat/concatdec.c | 2 +- 1 file changed, 1

Re: [FFmpeg-devel] [PATCH]libavformat/http: fix http error eof

I don't think it need a timeout event to disconnect the link, because when ffurl_read return to 0, it means the link disconnect. If s->off < target_end, it means AVERROR, otherwise, it's normal eof. I don't use chunked in HTTP, so s->chunksize should be initial value, and shouldn't be changed or

Re: [FFmpeg-devel] [PATCH] Use AVOnce as a static variable consistently

2017-05-22 18:06 GMT+08:00 Hendrik Leppkes : > Using AVOnce as a stack variable makes no sense as the state is lost > when the function exists. > > This fixes repeated calls to av(filter/device)_register_all > --- > libavdevice/alldevices.c | 2 +- > libavfilter/allfilters.c

Re: [FFmpeg-devel] [PATCH]libavformat/http: fix http error eof

2017-05-18 15:19 GMT+08:00 raymond zheng : > Hi: > I find an issue about http. I don't use chunked, so s->chunksize will > be set as UINT64_MAX when http open, but because of "if (s->chunksize > 0) > s->chunksize -= len;" then chunksize will not be UINT64_MAX. > >

Re: [FFmpeg-devel] [PATCH] Use AVOnce as a static variable consistently

On Mon, May 22, 2017 at 12:25 PM, wm4 wrote: > On Mon, 22 May 2017 12:06:19 +0200 > Hendrik Leppkes wrote: > >> Using AVOnce as a stack variable makes no sense as the state is lost >> when the function exists. >> >> This fixes repeated calls to

Re: [FFmpeg-devel] [PATCH] Use AVOnce as a static variable consistently

On Mon, 22 May 2017 12:06:19 +0200 Hendrik Leppkes wrote: > Using AVOnce as a stack variable makes no sense as the state is lost > when the function exists. > > This fixes repeated calls to av(filter/device)_register_all > --- > libavdevice/alldevices.c | 2 +- >

[FFmpeg-devel] [PATCH] Use AVOnce as a static variable consistently

Using AVOnce as a stack variable makes no sense as the state is lost when the function exists. This fixes repeated calls to av(filter/device)_register_all --- libavdevice/alldevices.c | 2 +- libavfilter/allfilters.c | 2 +- libavformat/allformats.c | 2 +- 3 files changed, 3 insertions(+), 3