[mythtv] Re: ivtv buffer overflow

Andy Hopper shopper at cs.unt.edu
Tue Dec 13 16:11:58 EST 2005


I was thinking that this was an IVTV issue and posted a message over on
ivtv-dev:

http://www.gossamer-threads.com/lists/ivtv/devel/26052

And then saw your last post over here. I tried using the ivtvdecoder.cpp
version that you link to (though it did not build out of the box due to
the fact that RingBuffer doesn't have a GetTotalReadPosition() method; I
just called GetReadPosition() which seems to work based upon my testing so
far).

Anyhow, I was wondering if this approach is the right way to go for now,
or whether there's some deeper fix that's in the works (i.e., if LiveTV is
broken as Niels commented, I was wondering if it is just broken for IVTV,
etc.)

Thanks,
Andy Hopper

>Isaac Richards wrote:
>>>If I quit LiveTV just before the recording starts all goes well. Is
>>>there anything different in the path LiveTV->recording and
>>>LiveTV->none->recording that could cause this?
>>
>>
>> Livetv -> recording is currently broken.
>
>Ah, that might explain. I assume that the ivtv overflow messages are
>also triggered by this.
>
>On the other hand, I also just reverted libs/libmythtv/ivtvdecoder.cpp
>to a revision just before the LiveTV changes (I used
>http://svn.mythtv.org/trac/browser/trunk/mythtv/libs/libmythtv/ivtvdecoder.cpp?rev=6959&format=txt)
>and I now succeeded to do two planned recordings that were started while
>watching LiveTV. The frontend didn't go to the recording automatically
>(it goes back to menu), but at least the recording itself goes well and
>can be watched when navigating to it.
>
>Not sure it is related, but these times I didn't see the 'scheduler:
>Last message repeated 1 times' (see my previous e-mail) after the
>'scheduler: Started recording' message:




More information about the mythtv-dev mailing list