[Bug 657221] Bug with clock dv

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=657221
  GStreamer | gst-plugins-good | git

--- Comment #10 from anthony <anthony.violo at ubicast.eu> 2011-11-04 15:48:08 UTC ---
I try many test but i still can't figure out what is the problem. I think it
comes from my device.
I add your patch, but when i have the bug  raw1394_read_cycle_timer return the
same value than when it works well.

I dont know if the value returned is the good one. I noticed that :

1. When the bug arrived with alsasink, i have this warn :
"baseaudiosink
gstbaseaudiosink.c:1088:gst_base_audio_sink_skew_slaving:<alsasink0> correct
clock skew 20002520 > 20000000"
so when i add slave-method=2 et provide-clock=true properties in alsasink, the
pipeline works.

2. when i want to make a streaming i have other warm :
"x264enc :0::<vencoder_2> VBV underflow (frame 308, -7488 bits)"

and

i try to add printf to have more informations.

Here is the result obtained:

{{{
/GstPipeline:pipeline0/GstIdentity:identity0: last-message = "chain   *******
(identity0:sink)i (7680 bytes, timestamp: 0:00:36.288375000, duration:
0:00:00.040000000, offset: 1741842, offset_end:  1743762, flags: 0) 0x9086060"
----------------------------------------------------------------
cycle_timer : -1266673534, local_time : 1320417289863809
cycle : -1266673534, _1394clock->cycle_timer_lo : -1266823081
calcul second : 0, cycle_timer ; -1266673534
----------------------------------------------------------------
cycle_timer : -1266670707, local_time : 1320417289863883
cycle : -1266670707, _1394clock->cycle_timer_lo : -1266673534
calcul second : 0, cycle_timer ; -1266670707
----------------------------------------------------------------
cycle_timer : -1266227878, local_time : 1320417289877400
cycle : -1266227878, _1394clock->cycle_timer_lo : -1266670707
calcul second : 0, cycle_timer ; -1266227878
----------------------------------------------------------------
cycle_timer : -1266227131, local_time : 1320417289877430
cycle : -1266227131, _1394clock->cycle_timer_lo : -1266227878
calcul second : 0, cycle_timer ; -1266227131
----------------------------------------------------------------
cycle_timer : -1265977228, local_time : 1320417289885056
cycle : -1265977228, _1394clock->cycle_timer_lo : -1266227131
calcul second : 0, cycle_timer ; -1265977228
----------------------------------------------------------------
cycle_timer : -1265979391, local_time : 1320417289885093
cycle : -1265979391, _1394clock->cycle_timer_lo : -1265977228
pass here!!
calcul second : 1, cycle_timer ; -1265979391
----------------------------------------------------------------
cycle_timer : -1265798285, local_time : 1320417289890503
cycle : -1265798285, _1394clock->cycle_timer_lo : -1265979391
calcul second : 1, cycle_timer ; -1265798285
----------------------------------------------------------------
/GstPipeline:pipeline0/GstIdentity:identity0: last-message = "chain   *******
(identity0:sink)i (192000 bytes, timestamp: 0:00:36.328375000, duration:
0:00:01.000000000, offset: 1743762, offset_end:  1791762, flags: 128)
0xb4c06620"
}}}

We can noticed that when _1394clock->cycle_timer_lo seems smaller than cycle
and when this arrive, i have a overflow log message and
_1394clock->cycle_timer_hi is incremented. So result is incorrect and that why
i have duration which is modified.

Maybe the better solution is to add property to use clock hardware.

i hope this will help to correct the bug or to add property provide-clock like
alsasink

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.