Em 23-09-2010 13:45, Dmitri Belimov escreveu:
> Hi
>
>> Em 20-09-2010 17:07, Dmitri Belimov escreveu:
>>> Hi
>>>
>>> I rework my last patch for audio and now audio works well. This
>>> patch can be submited to GIT tree Quality of audio now is good for
>>> SECAM-DK. For other standard I set some value from datasheet need
>>> some tests.
>>>
>>> 1. Fix pcm buffer overflow
>>> 2. Rework pcm buffer fill method
>>> 3. Swap bytes in audio stream
>>> 4. Change some registers value for TM6010
>>> 5. Change pcm buffer size
>>> --- a/drivers/staging/tm6000/tm6000-stds.c
>>> +++ b/drivers/staging/tm6000/tm6000-stds.c
>>> @@ -96,6 +96,7 @@ static struct tm6000_std_tv_settings tv_stds[] = {
>>>
>>> {TM6010_REQ07_R04_LUMA_HAGC_CONTROL, 0xdc},
>>> {TM6010_REQ07_R0D_CHROMA_KILL_LEVEL, 0x07},
>>> + {TM6010_REQ08_R05_A_STANDARD_MOD,
>>> 0x21}, /* FIXME */
>>
>> This didn't seem to work for PAL-M. Probably, the right value for it
>> is 0x22, to follow NTSC/M, since both uses the same audio standard.
>>
>> On some tests, I was able to receive some audio there, at the proper
>> rate, with a tm6010-based device. It died when I tried to change the
>> channel, so I didn't rear yet the real audio, but I suspect it will
>> work on my next tests.
>>
>> Yet, is being hard to test, as the driver has a some spinlock logic
>> broken. I'm enclosing the logs.
>
> Yes. I have some as crash from mplayer and arecord.
>
>> I was able to test only when using a monitor on the same machine. All
>> trials of using vnc and X11 export ended by not receiving any audio
>> and hanging the machine.
>>
>> I suspect that we need to fix the spinlock issue, in order to better
>> test it.
>
> Who can fix it?
Well, any of us ;)
I did a BKL lock fix series of patches, and hverkuil is improving them.
They'll make easier to avoid problems inside tm6000. We just need to make
sure that we'll hold/release the proper locks at tm6000-alsa, after applying
it at the mainstream.
Cheers,
Mauro
--
To unsubscribe from this list: send the line "unsubscribe linux-media" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
