[vls-devel] Re: dvbinput.cpp CVS - broken

Alexander Afanasyev alex at icn.bmstu.ru
Sun Jul 27 15:44:07 CEST 2003


Privet!

what logs i can send more???

cawka

Sunday, July 27, 2003, 5:39:07 PM, you wrote:

> I think I need more logging. Time to add some to the code.

> Alexander Afanasyev wrote:
>> Hi!
>> 
>> If i start program under valgrind it says, that
>> such program cannot be founded :(
>> 
>> -- cut --
>> *** Exception *** in constructor 1 (0xbffff050)
>> *** Exception *** in copy constructor (0x81a2c50, copy of 0xbffff050)
>> *** Exception *** in destructor (0xbffff050)
>> *** Exception *** in copy constructor (0xbffff160, copy of 0x81a2c50)
>> 2003-07-27 17:12:13 [DEBUG/ss1]  In ts2ts.cpp line 134 -> Starting to read program "Input DVB ss1"
>> 2003-07-27 17:12:13 [ERROR/ss1]  Unable to start program TET
>> 2003-07-27 17:12:13 [ERROR/ss1]  Error: 123Program "TET" doesn't exist
>> *** Exception *** in destructor (0xbffff160)
>> *** Exception *** in destructor (0x81a2c50)
>> 2003-07-27 17:12:13 [DEBUG/Vls]  In src/server/manager.cpp line 486 -> Manager unlocked
>> libdvbpsi error (PSI decoder): TS discontinuity (received 6, expected 0)
>> 
>> New PAT
>>   transport_stream_id : 12
>>   version_number      : 0
>>     | program_number @ [NIT|PMT]_PID
>>     |              0 @ 0x10 (16)
>>     |           6100 @ 0x17d4 (6100)
>> -- end cut --
>> 
>> i tried several channels...
>> 
>> 
>> Alexander Afanasyev
>> 
>> Sunday, July 27, 2003, 4:57:23 PM, you wrote:
>> 
>> 
>>>Alexander Afanasyev wrote:
>>>
>>>>Hi!
>>>>
>>>>As i wrote, now for me work vls 0.5.3/libdvb 0.2.2/
>>>>linuxtv-1.0.0pre2 (later drivers i try too)
>>>>
>>>>I have Technotrend DVB-S (skystar1) with hardware decoder.
>>>>p4 1.6/256 ram.
>>>>
>>>
>>>That means vls has to repackage PS into TS, but that should not give 
>>>such a high load. Could you run it under valgrind (same as previous) and 
>>>send me the valgrind log and the cachegrind logs? I especially need the 
>>>latter for timing/cpu consumption!
>> 
>> 
>>>>maybe some threads eats a lot of processor time, maybe
>>>>somwhere needs usleeps ?
>>>>
>>>
>>>No sleep, because it messes with your timing at send time, which will 
>>>result in choppy images.
>> 
>> 
>>>Grtz,
>>>Jean-Paul Saman.
>> 
>> 
>>>>Alexander Afanasyev
>>>>
>>>>Sunday, July 27, 2003, 4:28:48 PM, you wrote:
>>>>
>>>>
>>>>
>>>>>Alexander Afanasyev wrote:
>>>>>
>>>>>
>>>>>>What can be wrong???
>>>>>>
>>>>>>--- cut from top utility --
>>>>>>15:44:09  up 17:51,  3 users,  load average: 0,81, 0,29, 0,09
>>>>>>49 processes: 46 sleeping, 3 running, 0 zombie, 0 stopped
>>>>>>CPU states: 20,3% user, 42,7% system,  0,0% nice,  0,0% iowait, 36,9% idle
>>>>>>Mem:   255744k av,  215252k used,   40492k free,       0k shrd,   45664k buff
>>>>>>       86772k active,              72524k inactive
>>>>>>Swap:       0k av,       0k used,       0k free                   94660k cached
>>>>>>
>>>>>> PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
>>>>>>15850 root       9   0  8968 8968  1716 S    55,4  3,5   0:36 vlsd
>>>>>
>>>>>hmm that is a lot for vls!
>>>>
>>>>
>>>>>What is the vls version you are using?
>>>>>What is the cards type/model number?
>>>>
>>>>
>>>>>>15864 root       9   0     0    0     0 SW    1,9  0,0   0:01 kdvb-fe-0:0
>>>>>>15874 root      10   0  1048 1048   836 R     0,3  0,4   0:00 top
>>>>>>   1 root       9   0   480  480   420 S     0,0  0,1   0:04 init
>>>>>>   2 root       9   0     0    0     0 SW    0,0  0,0   0:00 keventd
>>>>>>   3 root       9   0     0    0     0 SW    0,0  0,0   0:02 kapmd
>>>>>>   4 root      19  19     0    0     0 SWN   0,0  0,0   0:00 ksoftirqd_CPU0
>>>>>>-------- end cut ----------
>>>>>>
>>>>>>Alexander Afanasyev
>>>>>>
>>>>>>Sunday, July 27, 2003, 3:30:44 AM, you wrote:
>>>>>>
>>>>>
>>>>>Grtz,
>>>>>Jean-Paul Saman.
>>>>
>>>>
>> 

-- 
This is the vls-devel mailing-list, see http://www.videolan.org/streaming/
To unsubscribe, please read http://developers.videolan.org/lists.html
If you are in trouble, please contact <postmaster at videolan.org>



More information about the vls-devel mailing list