[Openchrome-users] Re: [mythtv-users] MythTV Hanging on Video Exit via xvmc-vld

[Thomas Hellström]

Paul Bender wrote:

> Thomas Hellstr?m wrote:
>
>> Paul Bender wrote:
>>
>>> Sam Davies wrote:
>>>
>>>> Hi there
>>>>
>>>> I have a EPIA SP13000 which was running FC4 with the RPMs from
>>>> http://washington.kelkoo.net/epia/FC4/RPMS/ with mythTV using via
>>>> xvmc-vld mpeg2 decoding, and despite a few glitches, all was fine.
>>>>
>>>> I then needed to update X, and did so with yum update xorg-x11\* (with
>>>> the epia repositories set up as detailed). Although xine still seems
>>>> to work fine with the mpeg acceleration, mythtv plays video fine with
>>>> acceleration, however, when you try to exit watching a video, a half
>>>> blue, half black screen is shown for a couple of seconds. The system
>>>> then returns to a half-rendered myth screen, but the system seems to
>>>> have hung.
>>>>
>>>> I have tried getting the latest version of mythtv from svn and
>>>> recompiling, but the same problem occurs.
>>>>
>>>> I seem to remember reading about something similar about this on one
>>>> of the mailing lists a while ago, but I can't seem to find it.
>>>>
>>>> My X log doesn't seem register any errors at the point of crashing.
>>>>
>>>> Thanks in advance
>>>>
>>>> sam
>>>>
>>>> kernel: 2.6.12-1.1447 (with drm module)
>>>> X: 6.8.99.903-0
>>>> myth: from svn revision 8525
>>>
>>>
>>>
>>>
>>> In developing MiniMyth <http://linpvr.org>, I encountered the same 
>>> thing on the CN400 based motherboards. It appears that the problem 
>>> started after the big Xv update in the unichrome drivers. As a 
>>> result, have not upgraded X beyond 6.8.2 and I have not upgraded the 
>>> unichrome drivers beyond the 20050815 snapshot.
>>>
>> Hmm, You guys are not talking about this that got fixed in November, 
>> are you?
>>
>> ------------------------------------------------------------------------
>> r77 | totte | 2005-11-15 22:01:54 +0100 (Tue, 15 Nov 2005) | 3 lines
>>
>> - Fix Xv surface destruction and re-allocation.
>>  (Reported by Cedric Berger).
>>
>> ------------------------------------------------------------------------
>
>
> I do not believe so. When I so that show up in the change logs, I 
> updated and tried it. I still had the same problem.
>
> For me, since the earlier version works fine, it has not been a high 
> priority to track down the problem. Currently my SP800E is not hooked 
> up. However, since it appears that someone else is having the same 
> problem, I guess I will hook up my SP800E and generate some debug logs.
>
Great.
If possible, use latest openchrome svn. I just added some stuff that 
protects against XvMC surface- and context freeing issues. I doubt that 
this is the cause, however.

/Thomas


> _______________________________________________
> openchrome-users mailing list
> openchrome-users at openchrome.org
> http://wiki.openchrome.org/mailman/listinfo/openchrome-users