[issue1644] Regression: Broken FLV probe on big-endian system
New submission from ami_stuff ami_st...@o2.pl: It looks like below patch broken FLV format probe on my system: http://lists.mplayerhq.hu/pipermail/ffmpeg-cvslog/2009-September/024362.html I get Unknown format error message with every FLV file. Example: http://samples.mplayerhq.hu/FLV/zelda.flv When I remove the patch, everything works correctly. -- messages: 8498 priority: normal status: new substatus: new title: Regression: Broken FLV probe on big-endian system type: bug _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1642] MKV crashing in mplayer and handbrake
Carl Eugen Hoyos ceho...@rainbow.studorg.tuwien.ac.at added the comment: This is not a valid bug-report. Did you read http://ffmpeg.org/bugreports.html ? -- status: new - closed substatus: new - invalid _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1642 _
[issue1644] Regression: Broken FLV probe on big-endian system
Carl Eugen Hoyos ceho...@rainbow.studorg.tuwien.ac.at added the comment: Please add complete, uncut output and the exact version number that broke detection (links on ffmpeg-* changed before). Output of svn log would be nice, too. -- status: new - open substatus: new - needs_more_info _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1644] Regression: Broken FLV probe on big-endian system
ami_stuff ami_st...@o2.pl added the comment:
r19858 broke detection on my config.
FFmpeg version SVN-r20900-snapshot, Copyright (c) 2000-2009 Fabrice Bellard, et
al.
built on Dec 20 2009 06:11:22 with gcc 4.5.0 20090827 (experimental)
libavutil 50. 7. 0 / 50. 7. 0
libavcodec52.43. 0 / 52.43. 0
libavformat 52.44. 0 / 52.44. 0
libavdevice 52. 2. 0 / 52. 2. 0
libavfilter1.12. 0 / 1.12. 0
libswscale 0. 7. 2 / 0. 7. 2
zelda.flv: Unknown format
Author: michael
Date: Tue Sep 15 16:56:09 2009
New Revision: 19858
Log:
Check offset in flv_probe()
Modified:
trunk/libavformat/flvdec.c
Modified: trunk/libavformat/flvdec.c
==
--- trunk/libavformat/flvdec.c Tue Sep 15 15:28:14 2009(r19857)
+++ trunk/libavformat/flvdec.c Tue Sep 15 16:56:09 2009(r19858)
@@ -39,7 +39,7 @@ static int flv_probe(AVProbeData *p)
const uint8_t *d;
d = p-buf;
-if (d[0] == 'F' d[1] == 'L' d[2] == 'V' d[3] 5 d[5]==0) {
+if (d[0] == 'F' d[1] == 'L' d[2] == 'V' d[3] 5 d[5]==0
AV_RB32(d+5)8) {
return AVPROBE_SCORE_MAX;
}
return 0;
_
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644
_
[issue1644] Regression: Broken FLV probe on big-endian system
ami_stuff ami_st...@o2.pl added the comment:
BTW. I verified AV_RB32() if it work correctly with .sox file - I decoded it
without problems, so I guess this function works correctly?
from libavformat/soxdec.c:
static int sox_probe(AVProbeData *p)
{
/*if (AV_RL32(p-buf) == SOX_TAG || AV_RB32(p-buf) == SOX_TAG)*/
if (AV_RB32(p-buf) == SOX_TAG)
return AVPROBE_SCORE_MAX;
return 0;
}
_
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644
_
[issue1644] Regression: Broken FLV probe on big-endian system
Carl Eugen Hoyos ceho...@rainbow.studorg.tuwien.ac.at added the comment: Thanks for the missing Information. Do I understand correctly that detection works fine on LE? -- substatus: needs_more_info - open _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1645] WMV3 decoding fail
New submission from Michael N. Pritula prit...@yandex-team.ru: There is a file with wmv3 stream, that couldn't be decoded by latest FFmpeg. $ ffmpeg -i wmv3-wmav2.trunc -vcodec huffyuv wmv3-wmav2.avi FFmpeg version SVN-r20918, Copyright (c) 2000-2009 Fabrice Bellard, et al. built on Dec 24 2009 16:11:30 with gcc 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) configuration: --cc='ccache gcc' --prefix=/home/pritula/wNewFfmpeg/package-root --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-version3 --enable-libfaad --enable-libfaac --enable-libvorbis --enable-libtheora --enable-libxvid --enable-libx264 --enable-libschroedinger --disable-encoder=libschroedinger --enable-libgsm --enable-libnut --enable-gpl --enable-nonfree --enable-postproc --enable-pthreads --enable-avfilter --enable-shared --extra-cflags='-I/home/pritula/wNewFfmpeg/package-root/include -g' --extra-ldflags=-L/home/pritula/wNewFfmpeg/package-root/lib libavutil 50. 7. 0 / 50. 7. 0 libavcodec52.45. 0 / 52.45. 0 libavformat 52.44. 0 / 52.44. 0 libavdevice 52. 2. 0 / 52. 2. 0 libavfilter1.12. 0 / 1.12. 0 libswscale 0. 7. 2 / 0. 7. 2 libpostproc 51. 2. 0 / 51. 2. 0 [wmv3 @ 0x8078970]Reserved RES_SM=2 is forbidden Seems stream 1 codec frame rate differs from container frame rate: 1000.00 (1000/1) - 29.97 (3/1001) Input #0, asf, from 'wmv3-wmav2.trunc': Metadata: ToolName: TMPGEnc 4.0 XPress Version. 4.2.3.193 WMFSDKVersion : 9.00.00.3268 WMFSDKNeeded: 0.0.0. IsVBR : 0 Duration: 00:04:47.05, start: 3.00, bitrate: 285 kb/s Stream #0.0(rus): Audio: wmav2, 48000 Hz, 2 channels, s16, 192 kb/s Stream #0.1(rus): Video: wmv3, yuv420p, 720x480, 1000 kb/s, 29.97 tbr, 1k tbn, 1k tbc [huffyuv @ 0x807a070]using huffyuv 2.2.0 or newer interlacing flag [wmv3 @ 0x8078970]Reserved RES_SM=2 is forbidden Output #0, avi, to 'wmv3-wmav2.avi': Stream #0.0(rus): Video: huffyuv, yuv422p, 720x480, q=2-31, 200 kb/s, 90k tbn, 29.97 tbc Stream #0.1(rus): Audio: mp2, 48000 Hz, 2 channels, s16, 64 kb/s Stream mapping: Stream #0.1 - #0.0 Stream #0.0 - #0.1 Error while opening decoder for input stream #0.1 When using -vn option all is ok (sounds well): $ ffmpeg -i wmv3-wmav2.trunc -vn wmv3-wmav2.avi FFmpeg version SVN-r20918, Copyright (c) 2000-2009 Fabrice Bellard, et al. built on Dec 24 2009 16:11:30 with gcc 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) configuration: --cc='ccache gcc' --prefix=/home/pritula/wNewFfmpeg/package-root --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-version3 --enable-libfaad --enable-libfaac --enable-libvorbis --enable-libtheora --enable-libxvid --enable-libx264 --enable-libschroedinger --disable-encoder=libschroedinger --enable-libgsm --enable-libnut --enable-gpl --enable-nonfree --enable-postproc --enable-pthreads --enable-avfilter --enable-shared --extra-cflags='-I/home/pritula/wNewFfmpeg/package-root/include -g' --extra-ldflags=-L/home/pritula/wNewFfmpeg/package-root/lib libavutil 50. 7. 0 / 50. 7. 0 libavcodec52.45. 0 / 52.45. 0 libavformat 52.44. 0 / 52.44. 0 libavdevice 52. 2. 0 / 52. 2. 0 libavfilter1.12. 0 / 1.12. 0 libswscale 0. 7. 2 / 0. 7. 2 libpostproc 51. 2. 0 / 51. 2. 0 [wmv3 @ 0x8078970]Reserved RES_SM=2 is forbidden Seems stream 1 codec frame rate differs from container frame rate: 1000.00 (1000/1) - 29.97 (3/1001) Input #0, asf, from 'wmv3-wmav2.trunc': Metadata: ToolName: TMPGEnc 4.0 XPress Version. 4.2.3.193 WMFSDKVersion : 9.00.00.3268 WMFSDKNeeded: 0.0.0. IsVBR : 0 Duration: 00:04:47.05, start: 3.00, bitrate: 285 kb/s Stream #0.0(rus): Audio: wmav2, 48000 Hz, 2 channels, s16, 192 kb/s Stream #0.1(rus): Video: wmv3, yuv420p, 720x480, 1000 kb/s, 29.97 tbr, 1k tbn, 1k tbc Output #0, avi, to 'wmv3-wmav2.avi': Stream #0.0(rus): Audio: mp2, 48000 Hz, 2 channels, s16, 64 kb/s Stream mapping: Stream #0.0 - #0.0 Press [q] to stop encoding size= 581kB time=65.47 bitrate= 72.7kbits/s video:0kB audio:512kB global headers:0kB muxing overhead 13.584433% When using -vcodec copy option all is ok too (and plays well): $ ffmpeg -i wmv3-wmav2.trunc -vcodec copy wmv3-wmav2.avi FFmpeg version SVN-r20918, Copyright (c) 2000-2009 Fabrice Bellard, et al. built on Dec 24 2009 16:11:30 with gcc 4.1.2 20061115 (prerelease) (Debian 4.1.1-21) configuration: --cc='ccache gcc' --prefix=/home/pritula/wNewFfmpeg/package-root --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-version3 --enable-libfaad --enable-libfaac --enable-libvorbis --enable-libtheora --enable-libxvid --enable-libx264 --enable-libschroedinger --disable-encoder=libschroedinger --enable-libgsm --enable-libnut --enable-gpl --enable-nonfree --enable-postproc --enable-pthreads --enable-avfilter --enable-shared
[issue1644] Regression: Broken FLV probe on big-endian system
ami_stuff ami_st...@o2.pl added the comment: Thanks for the missing Information. Do I understand correctly that detection works fine on LE? Yes, I downloaded the same binary revision (SVN-r20900) for x86 and this works fine. _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1643] cannot use multiple cores to decode VC1 video
Diego Biurrun di...@biurrun.de added the comment: Please stop spamming us with useless bug reports, thank you. -- status: new - closed substatus: new - invalid _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1643 _
[issue1645] WMV3 decoding fail
Carl Eugen Hoyos ceho...@rainbow.studorg.tuwien.ac.at added the comment: Note that it also looks corrupt with wmvdmo and wmv9dmo if colour space YV12 is used (and not YUY2 or UYVY). -- status: new - open substatus: new - open type: bug - feature_request _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1645 _
[issue1291] wmv3 not known in mov container
compn te...@twmi.rr.com added the comment: dup -- status: open - closed substatus: open - duplicate superseder: +wmv3 and wma2 in mov container _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1291 _
[issue1644] Regression: Broken FLV probe on big-endian system
Michael Niedermayer michae...@gmx.at added the comment: On Thu, Dec 24, 2009 at 03:57:14PM +, ami_stuff wrote: Thanks for the missing Information. Do I understand correctly that detection works fine on LE? Yes, I downloaded the same binary revision (SVN-r20900) for x86 and this works fine. binary? are you using some kind of binary packages? if so this is closed/invalid We do have a flv regression tests and it does pass on fate ppc ... [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Thouse who are best at talking, realize last or never when they are wrong. _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1646] cdg seeking with ffplay?
New submission from compn te...@twmi.rr.com: seeking does not seem to work with cdg http://samples.mplayerhq.hu/karaoke/SC8932- 15%20Gorillaz%20-%20Feel%20Good%20Inc%20(Radio%20Version).cdg ffplay and use arrow keys = no seek. ffmpeg -ss 15 works. -- messages: 8510 priority: normal status: open substatus: open title: cdg seeking with ffplay? type: feature_request _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1646 _
[issue1644] Regression: Broken FLV probe on big-endian system
ami_stuff ami_st...@o2.pl added the comment: I downloaded the same binary revision (SVN-r20900) for x86 and this works fine. binary? are you using some kind of binary packages? if so this is closed/invalid Yes, I have no time and it's too much work to compile m68k x86 biulds together, so I download x86 build for comparision form the net. We do have a flv regression tests and it does pass on fate ppc ... Ok then, I will not report more big-endian issues. Maybe it's a problem with my dev environment. _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1644] Regression: Broken FLV probe on big-endian system
Michael Niedermayer michae...@gmx.at added the comment: On Thu, Dec 24, 2009 at 08:29:01PM +, ami_stuff wrote: [...] We do have a flv regression tests and it does pass on fate ppc ... Ok then, I will not report more big-endian issues. Maybe it's a problem with my dev environment. hmm, if you can find out where the bug was that information is welcome of course. Maybe try a different compiler, try -O0 make sure the AV_ macro is defined to the proper thing ... [..] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB It is dangerous to be right in matters on which the established authorities are wrong. -- Voltaire _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1644 _
[issue1645] WMV3 decoding fail
Kostya Shishkov kostya.shish...@gmail.com added the comment: This encoded stream uses unsupported feature nobody cares about, so it has not been REd and implemented. IIRC, RES_SM=2 indicates that there is a sprite coded in extradata. Also it may be interlaced YUV411. -- substatus: open - analyzed _ FFmpeg issue tracker iss...@roundup.ffmpeg.org https://roundup.ffmpeg.org/roundup/ffmpeg/issue1645 _
[issue1568] codec_tag not correct when use stream_copy
regits stiger1...@gmail.com added the comment:
can you check the audiocodecid in the flv header tag?
it's not correct.
this is because when do stream copy, codec-codec_tag
used icodec-codec_tag.
see the codes below:(ffmpeg.c)
if (ost-st-stream_copy) {
/* if stream_copy is selected, no need to
decode or encode */
codec-codec_id = icodec-codec_id;
codec-codec_type = icodec-codec_type;
if(!codec-codec_tag){
if( !os-oformat-codec_tag
|| av_codec_get_id (os-oformat-
codec_tag, icodec-codec_tag) 0
|| av_codec_get_tag(os-oformat-
codec_tag, icodec-codec_id) = 0)
codec-codec_tag = icodec-
codec_tag;
}
in this case, if av_codec_get_id (os-oformat-
codec_tag, icodec-codec_tag) 0, then codec-
codec_tag will use icodec-codec_tag(it's ||, not ).
it means, when stream_copy, if icodec-codec_tag is in
oformat-codec_tag, then it will use icodec-codec_tag.
but in fact, the same icodec-codec_tag in oformat-
codec_tag maybe map to other codec_id.
see the file attached, s16le.wav is the original wav
file,
test.flv is the flv file use the original ffmpeg.
test1.flv is the flv file which i disabled codec-
codec_tag = icodec-codec_tag.
i'm sure this error will not occur only to me, when we
used stream_copy, and icodec-codec_tag is also in
oformat-codec_tag but map to different codec_id, this
error will occur.
--logs---
$ file s16le.wav
s16le.wav: RIFF (little-endian) data, WAVE audio,
Microsoft PCM, 16 bit, mono 11025 Hz
$ ./ffmpeg -i s16le.wav -acodec copy test.flv
$ ./flvparser test.flv | more
INFO: FLV file version:1
INFO: FLV file contain /Audio
DATA: Script Data Tag
onMetaData[ECMA array]: count 7
duration[DOUBLE]:20 seconds
audiodatarate[DOUBLE]:172
audiosamplerate[DOUBLE]:11025 Hz
audiosamplesize[DOUBLE]:16 bits
stereo[BOOL]:0 ,(1-stereo, 0-mono)
audiocodecid[DOUBLE]:1 ,(0/3-PCM,1-ADPCM,2/14-
MP3,10-AAC,11-Speex)
filesize[DOUBLE]:444292 bytes
DATA: Audio Tag 1,00:00:00.000, len:4097
Audio codec:Linear PCM, little endian, 11 kHz, 16
bits, Mono
DATA: Audio Tag 2,00:00:00.186, len:4097
Audio codec:Linear PCM, little endian, 11 kHz, 16
bits, Mono
$ ./ffmpeg.new -i s16le.wav -acodec copy test1.flv
[sti...@socsupport ffmpeg-export-2009-11-25]
$ ./flvparser test1.flv | more
INFO: FLV file version:1
INFO: FLV file contain /Audio
DATA: Script Data Tag
onMetaData[ECMA array]: count 7
duration[DOUBLE]:20 seconds
audiodatarate[DOUBLE]:172
audiosamplerate[DOUBLE]:11025 Hz
audiosamplesize[DOUBLE]:16 bits
stereo[BOOL]:0 ,(1-stereo, 0-mono)
audiocodecid[DOUBLE]:3 ,(0/3-PCM,1-ADPCM,2/14-
MP3,10-AAC,11-Speex)
filesize[DOUBLE]:444292 bytes
DATA: Audio Tag 1,00:00:00.000, len:4097
Audio codec:Linear PCM, little endian, 11 kHz, 16
bits, Mono
DATA: Audio Tag 2,00:00:00.186, len:4097
Audio codec:Linear PCM, little endian, 11 kHz, 16
bits, Mono
DATA: Audio Tag 3,00:00:00.372, len:4097
File 's16le.zip' not attached - you can download it from
https://roundup.ffmpeg.org/roundup/ffmpeg/file688.
--
status: closed - open
substatus: works_for_me - open
_
FFmpeg issue tracker iss...@roundup.ffmpeg.org
https://roundup.ffmpeg.org/roundup/ffmpeg/issue1568
_
