After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 138666 - [playbin] cannot handle files with large initial delay in one stream
[playbin] cannot handle files with large initial delay in one stream
Status: RESOLVED FIXED
Product: GStreamer
Classification: Platform
Component: gst-plugins
git master
Other Linux
: Normal normal
: 0.8.8
Assigned To: GStreamer Maintainers
GStreamer Maintainers
Depends on:
Blocks: 138435
 
 
Reported: 2004-03-31 19:27 UTC by Stephane Loeuillet
Modified: 2005-01-27 13:57 UTC
See Also:
GNOME target: ---
GNOME version: 2.9/2.10



Description Stephane Loeuillet 2004-03-31 19:27:08 UTC
gst-player --gst-basicgthread _bad_/div3/lego_troopers_divx.avi

console :

ERROR        GST_PADS(13951) gstpad.c(3105):gst_pad_pull:<src_0> (null)
ERROR        GST_PADS(13951) gstpad.c(3105):gst_pad_pull:<src_0> pull on pad
src_0:sink but the peer pad avidemux0:audio_00 has no gethandler

file says :
lego_troopers_divx.avi: RIFF (little-endian) data, AVI, 368 x 276, 10.00 fps,
video: DivX 3 Low-Motion, audio: uncompressed PCM (mono, 22050 Hz)

mplayer says :
VIDEO:  [DIV3]  368x276  24bpp  10,000 fps   30,2 kbps ( 3,7 kbyte/s)
==========================================================================
Opening audio decoder: [pcm] Uncompressed PCM audio decoder
AUDIO: 22050 Hz, 1 ch, 8 bit (0x8), ratio: 22050->22050 (176,4 kbit)
Selected audio codec: [pcm] afm:pcm (Uncompressed PCM)
==========================================================================
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family
Selected video codec: [ffdivx] vfm:ffmpeg (FFmpeg DivX ;-) (MS MPEG-4 v3))
==========================================================================
Checking audio filter chain for 22050Hz/1ch/8bit -> 22050Hz/2ch/8bit...
AF_pre: af format: 1 bps, 1 ch, 22050 hz, little endian unsigned int
AF_pre: 22050Hz 1ch Unsigned 8-bit
AO: [oss] 22050Hz 2ch Unsigned 8-bit (1 bps)
Building audio filter chain for 22050Hz/1ch/8bit -> 22050Hz/2ch/8bit...
Comment 1 Stephane Loeuillet 2004-03-31 20:10:01 UTC
gst-player --gst-scheduler=basicgthread 2_sado\ maso\ breast\ piercing\ sick.avi

does the same error :
ERROR        GST_PADS(15098) gstpad.c(3105):gst_pad_pull:<src_0> (null)
ERROR        GST_PADS(15098) gstpad.c(3105):gst_pad_pull:<src_0> pull on pad
src_0:sink but the peer pad avidemux0:audio_00 has no gethandler

this one is a iv32/uncompressed PCM
2_sado maso breast piercing sick.avi: RIFF (little-endian) data, AVI, 240 x 180,
12.00 fps, video: Indeo 3.2, audio: uncompressed PCM (mono, 11025 Hz)

the common point between those 2 is the audio stream format

mplayer says this on the audio part :
Opening audio decoder: [pcm] Uncompressed PCM audio decoder
AUDIO: 11025 Hz, 1 ch, 8 bit (0x8), ratio: 11025->11025 (88,2 kbit)
Selected audio codec: [pcm] afm:pcm (Uncompressed PCM)
Comment 2 Stephane Loeuillet 2004-03-31 20:21:03 UTC
same thing for _bad_/iv32/quelle.avi :
quelle.avi: RIFF (little-endian) data, AVI, 240 x 180, ~15 fps, video: Indeo
3.2, audio: uncompressed PCM (mono, 22050 Hz)

ERROR        GST_PADS(15202) gstpad.c(3105):gst_pad_pull:<src_0> (null)
ERROR        GST_PADS(15202) gstpad.c(3105):gst_pad_pull:<src_0> pull on pad
src_0:sink but the peer pad avidemux0:audio_00 has no gethandler
------------------------------------------
_bad_/iv32/sado-maso\ \(piss\).mpg (which is an avi)
_bad_/iv50/chienne_de_maria...avi
_bad_/iv50/Q_The_White_Stripes...avi
Comment 3 Stephane Loeuillet 2004-03-31 20:45:44 UTC
_bad_/mp42/Sado\ Maso\ Porno\ -\ porn\ sex\ sm\ slave\ \(2mins\)*.avi
_bad_/_rgb/DIB_showattachment.avi

Comment 4 Stephane Loeuillet 2004-03-31 21:45:28 UTC
ERROR        GST_PADS(25281) gstpad.c(3105):gst_pad_pull:<src_0> (null)
ERROR        GST_PADS(25281) gstpad.c(3105):gst_pad_pull:<src_0> pull on pad
src_0:sink but the peer pad qtdemux0:audio_00 has no gethandler

it's with the file that Uraeus tested in 
http://bugzilla.gnome.org/show_bug.cgi?id=138321
Comment 5 Ronald Bultje 2004-04-11 17:19:51 UTC
I see the same bug. Interestingly, other schedulers work fine here, which makes
me believe it's a scheduler bug in basic (yay! even basic is imperfect). Entry
plays it fine.

The lego_troopers avi plays fine, but with hickups. Xine has similar hickups,
but less severe. This might be related to the large chunks that the file
contains (audio). I don't know how to fix that as of now.
Comment 6 Ronald Bultje 2004-04-11 19:40:24 UTC
Benjamin committed a fix to basic to make this work. Plays fine now, still has
stuttering playback. Changing topic to reflect new bug.
Comment 7 Ronald Bultje 2004-09-15 19:42:19 UTC
I just looked at this again, and the issue really is that timestamps are so far
apart. When the audio is at 55s, the video is still around 46s. So there may be
some A/V sync issue in this file or whatsoever.

mplayer works around it by not using the scale/rate of files with such headers.
Instead, they divide the total length of the audio stream (bytes) by the length
of the video stream (time) and use that as av_bps value for timestamp
calculation. Might be interesting to use for us as well, although I'm not
convinced that it's correct.
Comment 8 Ronald Bultje 2004-10-01 11:05:57 UTC
Ok, so current CVS of $today plays this almost correctly. There's one small item
remaining: dynamic queue resizing in your media player. The audio starts at 8s,
and our queue is 1s (or 3s in playbin), so it hangs. Apart from that, if you
manually resize queues, it plays perfectly.
Comment 9 Christian Fredrik Kalager Schaller 2005-01-13 22:17:31 UTC
Gst-player is no longer maintained, and have been deprecated in favour of Totem.
gst-player will tommorow be removed from bugzilla. Closing as wontfix.
Comment 10 Ronald Bultje 2005-01-13 22:45:20 UTC
Dude, this is a playbin bug.
Comment 11 Ronald Bultje 2005-01-27 13:57:16 UTC
Hm, my initial comment is wrong, init_delay isn't large at all. A/V sync is
good, it's just a bit choppy. From what I can see, this is just the movie, it
repeats frames quite a lot. Therefore, I no longer think this needs further
fixing - marking fixed.