Hi all This is my last attempt at releasing pre8 before Linuxtag. Changes from previous report: * old release critical bugs: - 3gp with amr audio broken: fixed * new release critical bugs: - b-frames + lavc in mencoder are reported to be broken * other stuff (from the old pre8 thread): - the utf8 stuff: solved, delayed or critical? - vf_ass: commit or wait? - the win32 gui? solved or delayed? iirc the author said to wait for pre9 - anything else? * Target: weekend * Open tasks (help needed): - check and complete changelog, including ffmpeg part (else release as is), or tell me that it's up to date - write a nice news entry (else I'll happily provide a nasty one) - security patch for pre7 (and maybe an advisor?) use the gentoo patches? forget about it? what about cvs? Commit safe pending patches now, don't break cvs, and come visiting us at Linuxtag next wek. Ciao, Roberto
On Thu, 2006-04-27 at 23:01 +0200, Roberto Togni wrote:
* new release critical bugs: - b-frames + lavc in mencoder are reported to be broken
This is caused by lavc encoder rejecting bad timestamps from mencoder. I believe the old behaviour can be restored for now by always setting the timestamps in ve_lavc.c to MP_NOPTS_VALUE, but I haven't taken more than a superficial look.
- vf_ass: commit or wait?
It likely has more bugs to be fixed later, but it is already useful with several files. It might be committable in a state where it at least doesn't break anything if it is not explicitly used. It would need some fixes and more review before that though, so if the release is supposed to be in one or two days then probably not. The issues I know that probably would need to be fixed: - The security check for font save paths (with font name coming from the played file) only checks that the name portion doesn't contain '/', should at least check for '\' under Windows (and maybe more, would need to verify which characters are safe under which OSes). - The fonts are always saved when playing an mkv file containing them, if it's committed as "at least doesn't break anything if not used" this should probably be turned off by default (there's already an option to turn font saving off, so changing the default would be enough) - Maybe version checks for font libraries used so configure doesn't detect an old version and then fail at compile time (the problem recently reported by Jan Knutar).
Hi On Fri, Apr 28, 2006 at 12:37:56AM +0300, Uoti Urpala wrote:
On Thu, 2006-04-27 at 23:01 +0200, Roberto Togni wrote:
* new release critical bugs: - b-frames + lavc in mencoder are reported to be broken
This is caused by lavc encoder rejecting bad timestamps from mencoder. I believe the old behaviour can be restored for now by always setting the timestamps in ve_lavc.c to MP_NOPTS_VALUE, but I haven't taken more than a superficial look.
yes, i just wanted to suggest that too for the release [...] -- Michael In the past you could go to a library and read, borrow or copy any book Today you'd get arrested for mere telling someone where the library is
On Fri, 2006-04-28 at 14:50 +0200, Michael Niedermayer wrote:
On Fri, Apr 28, 2006 at 12:37:56AM +0300, Uoti Urpala wrote:
This is caused by lavc encoder rejecting bad timestamps from mencoder. I believe the old behaviour can be restored for now by always setting the timestamps in ve_lavc.c to MP_NOPTS_VALUE, but I haven't taken more than a superficial look.
yes, i just wanted to suggest that too for the release
Actually, are the timestamps currently useful anywhere, or would it be better to completely disable them? mencoder.c currently has blit_frame=decode_video(sh_video,frame_data.start,frame_data.in_size, skip_flag>0 && (!sh_video->vfilter || ((vf_instance_t *)sh_video->vfilter)->control(sh_video->vfilter, VFCTRL_SKIP_NEXT_FRAME, 0) != CONTROL_TRUE), mux_v->timer); I think the front of the filter chain should be fed the source pts (rather than pts from the muxer) eventually, so unless some part of the code already does something useful with the timestamps, I'd suggest changing "mux_v->timer" to "MP_NOPTS_VALUE" for the release.
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* other stuff (from the old pre8 thread): - anything else?
I found a regression with the updated libfaad: http://www.mplayerhq.hu/MPlayer/samples/mobileVideo_3gp/t7p69_227_196.3g2 This sample crashes, it just used to produce no sound. Diego
Diego Biurrun wrote:
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* other stuff (from the old pre8 thread): - anything else?
I found a regression with the updated libfaad:
http://www.mplayerhq.hu/MPlayer/samples/mobileVideo_3gp/t7p69_227_196.3g2
This sample crashes, it just used to produce no sound.
It still "works" for me: Opening audio decoder: [faad] AAC (MPEG2/4 Advanced Audio Coding) FAAD: Failed to initialize the decoder! ADecoder init failed :( ADecoder init failed :( Cannot find codec for audio format 0x6134706D. -- Jindrich Makovicka
On Fri, Apr 28, 2006 at 10:08:20AM +0200, Jindrich Makovicka wrote:
Diego Biurrun wrote:
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* other stuff (from the old pre8 thread): - anything else?
I found a regression with the updated libfaad:
http://www.mplayerhq.hu/MPlayer/samples/mobileVideo_3gp/t7p69_227_196.3g2
This sample crashes, it just used to produce no sound.
It still "works" for me:
Opening audio decoder: [faad] AAC (MPEG2/4 Advanced Audio Coding) FAAD: Failed to initialize the decoder! ADecoder init failed :( ADecoder init failed :( Cannot find codec for audio format 0x6134706D.
Yeah, seems to be a bigendian/ppc issue... Diego
On Fri, Apr 28, 2006 at 04:22:44AM +0200, Diego Biurrun wrote:
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* other stuff (from the old pre8 thread): - anything else?
I found a regression with the updated libfaad:
http://www.mplayerhq.hu/MPlayer/samples/mobileVideo_3gp/t7p69_227_196.3g2
This sample crashes, it just used to produce no sound.
Forget this, false alarm. I had the native AAC decoder in my local tree, it was not libfaad that was crashing... Diego
Diego Biurrun wrote:
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* other stuff (from the old pre8 thread): - anything else?
I found a regression with the updated libfaad:
http://www.mplayerhq.hu/MPlayer/samples/mobileVideo_3gp/t7p69_227_196.3g2
I have a problem related to another common mobile video format: Mpeg4+aac in a MP4 container. I'm using mencoder /big/Medien/Video/test.mpg \ -oac faac -faacopts br=96 \ -ovc lavc -lavcopts vcodec=mpeg4:mbd=2:vbitrate=300 \ -of lavf -lavfopts i_certify_that_my_video_stream_does_not_use_b_frames:format=mp4 \ -endpos 10 -o test.mp4 ffmpeg: ffmpeg -y -i /big/Medien/Video/test.mpg -b 300 -t 10 -ac 1 -ab 96 -acodec aac -vcodec mpeg4 test-ffmpeg.mp4 The two resulting videos can both be played with mplayer and the player of the mobile phone (Nokia 6280). But the latter one does not play any sound with the mencoder generated file. The ffmpeg file works perfectly with both players. mplayer -v shows some differences, especially the ESDS MPEG4 Descriptors are mission in the mencoder generated file: diff mencoder.txt ffmpeg.txt [...]
ESDS MPEG4 Decoder Specific Descriptor (2Bytes) ESDS MPEG4 Sync Layer Config Descriptor (1Bytes) -> predefined: 2 [...] < Fourcc: mp4v Codec: ''
Fourcc: mp4v Codec: 'mpeg4' [...]
Maybe there is a MP4 containter bug in mencoder? Regards, Matthias
On Sat, 29 Apr 2006 11:53:05 +0200 Matthias Wieser <mwieser@gmx.de> wrote: [...]
I have a problem related to another common mobile video format: Mpeg4+aac in a MP4 container.
I'm using mencoder /big/Medien/Video/test.mpg \ -oac faac -faacopts br=96 \ -ovc lavc -lavcopts vcodec=mpeg4:mbd=2:vbitrate=300 \ -of lavf -lavfopts i_certify_that_my_video_stream_does_not_use_b_frames:format=mp4 \ -endpos 10 -o test.mp4
ffmpeg: ffmpeg -y -i /big/Medien/Video/test.mpg -b 300 -t 10 -ac 1 -ab 96 -acodec aac -vcodec mpeg4 test-ffmpeg.mp4
The two resulting videos can both be played with mplayer and the player of the mobile phone (Nokia 6280). But the latter one does not play any sound with the mencoder generated file. The ffmpeg file works perfectly with both players.
mplayer -v shows some differences, especially the ESDS MPEG4 Descriptors are mission in the mencoder generated file:
diff mencoder.txt ffmpeg.txt [...]
ESDS MPEG4 Decoder Specific Descriptor (2Bytes) ESDS MPEG4 Sync Layer Config Descriptor (1Bytes) -> predefined: 2 [...] < Fourcc: mp4v Codec: ''
Fourcc: mp4v Codec: 'mpeg4' [...]
Maybe there is a MP4 containter bug in mencoder?
Is it related to the faad2 update and/or is it a regression since pre7? If not, unless you have a trivial patch to fix it, I'm not going to delay the release for it, it's just one more bug. Ciao, Roberto
Am Samstag, 29. April 2006 15:35 schrieb Roberto Togni:
On Sat, 29 Apr 2006 11:53:05 +0200 Matthias Wieser <mwieser@gmx.de> wrote:
[...]
I have a problem related to another common mobile video format: Mpeg4+aac in a MP4 container.
I'm using mencoder /big/Medien/Video/test.mpg \ -oac faac -faacopts br=96 \ -ovc lavc -lavcopts vcodec=mpeg4:mbd=2:vbitrate=300 \ -of lavf -lavfopts i_certify_that_my_video_stream_does_not_use_b_frames:format=mp4 \ -endpos 10 -o test.mp4
ffmpeg: ffmpeg -y -i /big/Medien/Video/test.mpg -b 300 -t 10 -ac 1 -ab 96 -acodec aac -vcodec mpeg4 test-ffmpeg.mp4
The two resulting videos can both be played with mplayer and the player of the mobile phone (Nokia 6280). But the latter one does not play any sound with the mencoder generated file. The ffmpeg file works perfectly with both players.
mplayer -v shows some differences, especially the ESDS MPEG4 Descriptors are mission in the mencoder generated file:
diff mencoder.txt ffmpeg.txt [...]
ESDS MPEG4 Decoder Specific Descriptor (2Bytes) ESDS MPEG4 Sync Layer Config Descriptor (1Bytes) -> predefined: 2
[...] < Fourcc: mp4v Codec: '' ---
Fourcc: mp4v Codec: 'mpeg4'
[...]
Maybe there is a MP4 containter bug in mencoder?
Is it related to the faad2 update and/or is it a regression since pre7?
I don't think that it has something to do with faad as the problem is not related to mplayers decoding but to mencoders encoding (faac).
If not, unless you have a trivial patch to fix it, I'm not going to delay the release for it, it's just one more bug.
That's fine with me, I just wanted to mention this bug so that it becomes known. It's very nice that mencoder can create mp4 containers now, but unfortunately it's not working 100%. But with ffmpeg a workaround exists. Regards, Matthias p.s. If you want, I can upload sample videos created by ffmpeg (working) and mencoder (not working on the mobile phone).
Hi On Sat, Apr 29, 2006 at 11:53:05AM +0200, Matthias Wieser wrote:
Diego Biurrun wrote:
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* other stuff (from the old pre8 thread): - anything else?
I found a regression with the updated libfaad:
http://www.mplayerhq.hu/MPlayer/samples/mobileVideo_3gp/t7p69_227_196.3g2
I have a problem related to another common mobile video format: Mpeg4+aac in a MP4 container.
I'm using mencoder /big/Medien/Video/test.mpg \ -oac faac -faacopts br=96 \ -ovc lavc -lavcopts vcodec=mpeg4:mbd=2:vbitrate=300 \ -of lavf -lavfopts i_certify_that_my_video_stream_does_not_use_b_frames:format=mp4 \ -endpos 10 -o test.mp4
ffmpeg: ffmpeg -y -i /big/Medien/Video/test.mpg -b 300 -t 10 -ac 1 -ab 96 -acodec aac -vcodec mpeg4 test-ffmpeg.mp4
The two resulting videos can both be played with mplayer and the player of the mobile phone (Nokia 6280). But the latter one does not play any sound with the mencoder generated file. The ffmpeg file works perfectly with both players.
mplayer -v shows some differences, especially the ESDS MPEG4 Descriptors are mission in the mencoder generated file:
RTFM aglobal / vglobal options (aglobal will need faac encoding over lavc, dunno how to get the faac wraper in mplayer to produce these headers, maybe it was faacopts raw or so) [...] -- Michael In the past you could go to a library and read, borrow or copy any book Today you'd get arrested for mere telling someone where the library is
Michael Niedermayer wrote:
RTFM aglobal / vglobal options (aglobal will need faac encoding over lavc, dunno how to get the faac wraper in mplayer to produce these headers, maybe it was faacopts raw or so)
[...]
-faacopts mpeg=4:raw
Am Samstag, 29. April 2006 23:41 schrieb Nico Sabbi:
Michael Niedermayer wrote:
RTFM aglobal / vglobal options (aglobal will need faac encoding over lavc, dunno how to get the faac wraper in mplayer to produce these headers, maybe it was faacopts raw or so)
[...]
-faacopts mpeg=4:raw
Does not make a difference. :-( Those differences in the "mencoder -v" output (and some probably minor ones) are still there. And and the mobile phone still does not play the sound. diff mencoder.txt ffmpeg.txt [...]
ESDS MPEG4 Decoder Specific Descriptor (2Bytes) ESDS MPEG4 Sync Layer Config Descriptor (1Bytes) -> predefined: 2 [...] < Fourcc: mp4v Codec: ''
Fourcc: mp4v Codec: 'mpeg4' [...]
Regards, Matthias
Hi On Mon, May 08, 2006 at 02:17:32PM +0200, Matthias Wieser wrote:
Am Samstag, 29. April 2006 23:41 schrieb Nico Sabbi:
Michael Niedermayer wrote:
RTFM aglobal / vglobal options (aglobal will need faac encoding over lavc, dunno how to get the faac wraper in mplayer to produce these headers, maybe it was faacopts raw or so)
[...]
-faacopts mpeg=4:raw
Does not make a difference. :-(
Those differences in the "mencoder -v" output (and some probably minor ones) are still there. And and the mobile phone still does not play the sound.
you mean "mplayer -v" right? please 1. ensure that lavc/lavf are upto date (and make distclean) 2. post the used commandlines 3. try faac over lavc i know mencoder can generate .mp4 files which work on psp,ipod and sony erricson phones, i would be quite surprised if they dont work on nokia ... [...] -- Michael In the past you could go to a library and read, borrow or copy any book Today you'd get arrested for mere telling someone where the library is
Am Montag, 8. Mai 2006 17:06 schrieb Michael Niedermayer:
Hi
On Mon, May 08, 2006 at 02:17:32PM +0200, Matthias Wieser wrote:
Am Samstag, 29. April 2006 23:41 schrieb Nico Sabbi:
Michael Niedermayer wrote:
RTFM aglobal / vglobal options (aglobal will need faac encoding over lavc, dunno how to get the faac wraper in mplayer to produce these headers, maybe it was faacopts raw or so)
[...]
-faacopts mpeg=4:raw
Does not make a difference. :-(
Those differences in the "mencoder -v" output (and some probably minor ones) are still there. And and the mobile phone still does not play the sound.
you mean "mplayer -v" right?
Yes :-)
please 1. ensure that lavc/lavf are upto date (and make distclean)
CVS-0060502-16:38 But I will do a new checkout tomorrow.
2. post the used commandlines
See my first email from 04.29.2006
3. try faac over lavc
What do you mean? I'm using "-oac faac".
i know mencoder can generate .mp4 files which work on psp,ipod and sony erricson phones, i would be quite surprised if they dont work on nokia
Thanks, Matthias
On Mon, May 08, 2006 at 11:55:01PM +0200, Matthias Wieser wrote:
Am Montag, 8. Mai 2006 17:06 schrieb Michael Niedermayer:
Hi
On Mon, May 08, 2006 at 02:17:32PM +0200, Matthias Wieser wrote:
Am Samstag, 29. April 2006 23:41 schrieb Nico Sabbi:
Michael Niedermayer wrote:
RTFM aglobal / vglobal options (aglobal will need faac encoding over lavc, dunno how to get the faac wraper in mplayer to produce these headers, maybe it was faacopts raw or so)
[...]
-faacopts mpeg=4:raw
Does not make a difference. :-(
Those differences in the "mencoder -v" output (and some probably minor ones) are still there. And and the mobile phone still does not play the sound.
you mean "mplayer -v" right?
Yes :-)
please 1. ensure that lavc/lavf are upto date (and make distclean)
CVS-0060502-16:38
But I will do a new checkout tomorrow.
2. post the used commandlines
See my first email from 04.29.2006
that should lack "-faacopts mpeg=4:raw" and vglobal=1
3. try faac over lavc
What do you mean? I'm using "-oac faac".
-oac lavc -lavcopts acodec=aac:aglobal=1:vglobal=1 [...] -- Michael In the past you could go to a library and read, borrow or copy any book Today you'd get arrested for mere telling someone where the library is
Am Dienstag, 9. Mai 2006 11:12 schrieb Michael Niedermayer:
On Mon, May 08, 2006 at 11:55:01PM +0200, Matthias Wieser wrote:
that should lack "-faacopts mpeg=4:raw" and vglobal=1
-faacopts mpeg=4:raw" and vglobal=1 or vglobal=2 did not make a difference.
3. try faac over lavc
What do you mean? I'm using "-oac faac".
-oac lavc -lavcopts acodec=aac:aglobal=1:vglobal=1
Thanks, this works! Now I'm using: mencoder $1 \ -oac lavc -ovc lavc -lavcopts \ vcodec=mpeg4:mbd=2:vbitrate=300:vglobal=1:aglobal=1:acodec=aac:abitrate=96 \ -of lavf -lavfopts \ i_certify_that_my_video_stream_does_not_use_b_frames:format=mp4 \ -af pan=1:0.5:0.5 -channels 1\ -endpos 10 -o $2 Now I'm trying to find out how to do mono audio (-af pan=1:0.5:0.5 -channels 1 does not seem to be enough) with lavc/aac. Regards, Matthias
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make full of Diego, do you? btw.. I get occasional segfault at uninit_video, i must take a look at the problem before the release. (it may be because i have contaminated my pure cvs tree)
Hi, On 4/28/06, Ivan Kalvachev <ikalvachev@gmail.com> wrote:
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make full of Diego, do you?
Well, if we release with the current name scheme, then ther's no problem...
btw.. I get occasional segfault at uninit_video, i must take a look at the problem before the release. (it may be because i have contaminated my pure cvs tree)
... And I also have some segfaults when playing DVDs (doesn't seem to be an MPEG2 problem, more a problem in opening/parsing the informations). I'll look into it this week-end. Off course, before the new release, I'd need to update the french man page (there's been quite a bit of work done on the eng version this week). Guillaume -- I am disillusioned enough to know that no man's opinion on any subject is worth a damn unless backed up with enough genuine information to make him really know what he's talking about. -- H. P. Lovecraft (about the flamewars on FFmpeg and MPlayer-dev mailing lists) http://www.brainyquote.com/quotes/quotes/h/hplovecr278144.html
2006/4/28, Guillaume POIRIER <poirierg@gmail.com>:
Hi,
On 4/28/06, Ivan Kalvachev <ikalvachev@gmail.com> wrote:
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make fool of Diego, do you?
Well, if we release with the current name scheme, then ther's no problem...
_THIS_ is the problem!!! I accept anything else but the pre...
Hi On Fri, Apr 28, 2006 at 10:38:02AM +0300, Ivan Kalvachev wrote:
2006/4/28, Guillaume POIRIER <poirierg@gmail.com>:
Hi,
On 4/28/06, Ivan Kalvachev <ikalvachev@gmail.com> wrote:
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make fool of Diego, do you?
Well, if we release with the current name scheme, then ther's no problem...
_THIS_ is the problem!!! I accept anything else but the pre...
i strongly vote for deciding the naming after the release and release with the current silly pre name stuff [...] -- Michael In the past you could go to a library and read, borrow or copy any book Today you'd get arrested for mere telling someone where the library is
On Fri, Apr 28, 2006 at 02:53:00PM +0200, Michael Niedermayer wrote:
Hi
On Fri, Apr 28, 2006 at 10:38:02AM +0300, Ivan Kalvachev wrote:
2006/4/28, Guillaume POIRIER <poirierg@gmail.com>:
Hi,
On 4/28/06, Ivan Kalvachev <ikalvachev@gmail.com> wrote:
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make fool of Diego, do you?
Well, if we release with the current name scheme, then ther's no problem...
_THIS_ is the problem!!! I accept anything else but the pre...
i strongly vote for deciding the naming after the release and release with the current silly pre name stuff
Agree. Rich
Hi, Michael Niedermayer wrote:
On Fri, Apr 28, 2006 at 10:38:02AM +0300, Ivan Kalvachev wrote:
2006/4/28, Guillaume POIRIER <poirierg@gmail.com>:
Hi,
On 4/28/06, Ivan Kalvachev <ikalvachev@gmail.com> wrote:
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make fool of Diego, do you?
Well, if we release with the current name scheme, then there's no problem...
_THIS_ is the problem!!! I accept anything else but the pre...
i strongly vote for deciding the naming after the release and release with the current silly pre name stuff
Gooooood! So that means that you're gonna meet us at LinuxTag to discuss the new naming scheme? Weee!! :-D Guillaume
2006/4/28, Michael Niedermayer <michaelni@gmx.at>:
Hi
On Fri, Apr 28, 2006 at 10:38:02AM +0300, Ivan Kalvachev wrote:
2006/4/28, Guillaume POIRIER <poirierg@gmail.com>:
Hi,
On 4/28/06, Ivan Kalvachev <ikalvachev@gmail.com> wrote:
2006/4/28, Roberto Togni <r_togni@tiscali.it>:
* Target: weekend
This is impossible. Diego promised they will come with new naming scheme for releases at linux-tag. So we cannot release until this is done. You don't want to make fool of Diego, do you?
Well, if we release with the current name scheme, then ther's no problem...
_THIS_ is the problem!!! I accept anything else but the pre...
i strongly vote for deciding the naming after the release and release with the current silly pre name stuff
OK, but only this time and only because you asked for it.
Hi, Guillaume POIRIER a écrit :
btw.. I get occasional segfault at uninit_video, i must take a look at the problem before the release. (it may be because i have contaminated my pure cvs tree)
... And I also have some segfaults when playing DVDs (doesn't seem to be an MPEG2 problem, more a problem in opening/parsing the informations). I'll look into it this week-end.
vanilla CVS doesn't have this problem. I'll try to see what the problem is with my tree anyhow, when I have the time....
Off course, before the new release, I'd need to update the french man page (there's been quite a bit of work done on the eng version this week).
I've received some new patches while I was at the Linux Tag (BTW, the ones who didn't come really missed smth, it's was really good stuff, and lots of laugh). I'd like to review and apply them before pre8 hits the shelves. Guillaume
Hi, On Friday 28 April 2006 15:11, Ivan Kalvachev wrote:
btw.. I get occasional segfault at uninit_video, i must take a look at the problem before the release. (it may be because i have contaminated my pure cvs tree)
I experienced some uninit_vidoe segfault with some real video (only?). But most video is ok. -- Best Regards, LR
btw.. I get occasional segfault at uninit_video, i must take a look at the problem before the release. (it may be because i have contaminated my pure cvs tree)
I cannot reproduce it with fresh cvs checkout and the code breaks in area that could be affected by local changes I have done. So this non issue.
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* new release critical bugs: - b-frames + lavc in mencoder are reported to be broken
This definitely needs to be fixed.
* other stuff (from the old pre8 thread): - the utf8 stuff: solved, delayed or critical? - vf_ass: commit or wait? - the win32 gui? solved or delayed? iirc the author said to wait for pre9
I say delay all these.
* Open tasks (help needed): - check and complete changelog, including ffmpeg part (else release as is), or tell me that it's up to date - write a nice news entry (else I'll happily provide a nasty one) - security patch for pre7 (and maybe an advisor?) use the gentoo patches? forget about it? what about cvs?
The help messages need to be reviewed, after the printf patches there are many many new ones. I'll try to get to this over the weekend. Diego
On Fri, Apr 28, 2006 at 11:53:28AM +0200, Diego Biurrun wrote:
On Thu, Apr 27, 2006 at 11:01:12PM +0200, Roberto Togni wrote:
This is my last attempt at releasing pre8 before Linuxtag.
Changes from previous report:
* Open tasks (help needed): - check and complete changelog, including ffmpeg part (else release as is), or tell me that it's up to date - write a nice news entry (else I'll happily provide a nasty one) - security patch for pre7 (and maybe an advisor?) use the gentoo patches? forget about it? what about cvs?
The help messages need to be reviewed, after the printf patches there are many many new ones. I'll try to get to this over the weekend.
Finished. Now somebody needs to take care of the changelog. It won't be me, I've done my share of tedious tasks now... Diego
On Thursday, 27 April 2006 at 23:01, Roberto Togni wrote:
* new release critical bugs: - b-frames + lavc in mencoder are reported to be broken
- v4l is broken due to float->double changes by uau R. -- MPlayer developer and RPMs maintainer: http://rpm.greysector.net/mplayer/ There should be a science of discontent. People need hard times and oppression to develop psychic muscles. -- from "Collected Sayings of Muad'Dib" by the Princess Irulan
participants (14)
-
Diego Biurrun -
Dominik 'Rathann' Mierzejewski -
Guillaume POIRIER -
Guillaume Poirier -
Guillaume POIRIER -
Ivan Kalvachev -
Jindrich Makovicka -
Matthias Wieser -
Michael Niedermayer -
Nico Sabbi -
Rich Felker -
Roberto Togni -
Uoti Urpala -
陆然