[MPlayer-dev-eng] [PATCH] dvb fix

[David Kuehling]

Sorry for possibly crossed emails.  I'm still receiving mplayer-dev in
digest mode.  Could you please CC me when answering?

>>>>> "Nico" == Nico  <nsabbi at libero.it> writes:

> David Kuehling wrote:

>> mplayer dvb://<channel> -vfm ffmpeg -nosound -cache 4096 <after
>> zapping>
>> 
>> dvb_streaming_read, attempt N. 5 failed with errno 11 when reading
>> 1048 bytes dvb_streaming_read, attempt N. 4 failed with errno 11 when
>> reading 1048 bytes dvb_streaming_read, attempt N. 3 failed with errno
>> 11 when reading 1048 bytes dvb_streaming_read, attempt N. 2 failed
>> with errno 11 when reading 1048 bytes dvb_streaming_read, attempt
>> N. 1 failed with errno 11 when reading 1048 bytes dvb_streaming_read,
>> attempt N. 5 failed with errno 11 when reading 2048 bytes
>> dvb_streaming_read, attempt N. 4 failed with errno 11 when reading
>> 2048 bytes dvb_streaming_read, attempt N. 3 failed with errno 11 when
>> reading 2048 bytes dvb_streaming_read, attempt N. 2 failed with errno
>> 11 when reading 2048 bytes dvb_streaming_read, attempt N. 1 failed
>> with errno 11 when reading 2048 bytes dvb_streaming_read, return 0
>> bytes Successfully enabled DPMS UNINIT COMPLETE
>> 

> here it has finished to empty the cache, and it takes longer because
> you use the cache, after that you see the picture, right?

Yes, but the output (dvb_streaming_read...) indicates, that it tries to
read from the DVB-device *after* pressing the step_channel button, but
*before* uninit was complete.  Why should it do that?  Why must it wait
for those enormous timeouts of 5*0.5 seconds?  What is the relation
between those dvb_streaming_reads and the actual emptying of the cache?

>> TS file format detected.  PROBING UP TO 2000000, PROG: 0 ...
>> 
>> mplayer dvb://<channel> -vfm ffmpeg -nosound <after zapping>
>> 
>> dvb_streaming_read, attempt N. 1 failed with errno 11 when reading
>> 716 bytes dvb_streaming_read, attempt N. 1 failed with errno 11 when
>> reading 2048 bytes dvb_streaming_read, return 0 bytes Successfully
>> enabled DPMS UNINIT COMPLETE
>> 
> here it has finished to empty the cache, and it takes shorter because
> you DON'T use the cache, after that you see the picture, right?

I still don't understand why it should take longer than some
microseconds to empty the cache...

David
-- 
GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg
Fingerprint: B17A DC95 D293 657B 4205  D016 7DEF 5323 C174 7D40