[alsa-devel] [Fwd: Re: Davinci ASoC support (patch)]

Tue Apr 1 22:08:19 CEST 2008

lokowich wrote:
>
>
> Vova Barinov wrote:
>> Mark Brown wrote:
>>   
>>> On Fri, Mar 28, 2008 at 09:27:04AM -0500, lokowich wrote:
>>>
>>>   
>>>     
>>>> Forwarded from davinci-linux-open-source list:
>>>>     
>>>>       
>>> CCing in Vladimir Barinov who contributed the DaVinci support.
>>>
>>>   
>>>     
>>>> I've built this ASoC patch into my DaVinci git kernel (thanks for the 
>>>> SENSORS tip), and can use aplay and speaker-test, but no sound.  Figured 
>>>> it may be a mixer/volume control issue, but the alsamixer is unable to 
>>>> open for a failed ioctl on USBDEVFS_IOCTL.  Not sure why there is a USB 
>>>> dependency.  Noted that there is no /sys/class/sound/mixer in my sysfs.  
>>>> Any ideas?
>>>>     
>>>>       
>> How do you figure out that you can play sound?
>> Also what platform do you use - is it Davinci-EVM? what line output do 
>> you use to hear the sound?
>> Please, post your dmesg log and cat /proc/interrupts too.
>>   
>
>     Excerpt from dmesg:
>     Advanced Linux Sound Architecture Driver Version 1.0.16.
>     ASoC version 0.13.2
>     AIC3X Audio Codec 0.1
>     asoc: aic3x <-> davinci-i2s mapping ok
>     ALSA device list:
>       #0: DaVinci EVM (aic3x)
>
>     /proc/interrupts:
>                CPU0
>       6:          0       AINTC  dm644xresizer
>       8:      23608       AINTC  davincifb
>      13:       6995       AINTC  EMAC
>      22:        107       AINTC  ide0
>      32:      39436       AINTC  clockevent
>      33:          3       AINTC  free-run counter
>      39:        217       AINTC  i2c_davinci
>      40:        204       AINTC  serial
>     Err:          0
>
Sorry for the late response.
I don't see EDMA interrupts. So it seems that you have no i2s transfers 
between codec and SoC.
>
>
>     This is the DM6446 EVM.  Note, my kernel has been patched with
>     latest ALSA, to no avail.  I hear the  DAC pop during
>     initialization on both the Line Out and Headphone jacks.
>
>>> It might be helpful to post dmesg output showing the sound support
>>> being loaded
>>>   
>>>     
>>>> Bernard Blackham wrote:
>>>>     
>>>>       
>>>>> Behme wrote:
>>>>>   
>>>>>       
>>>>>         
>>>>>> While searching something different, I just found
>>>>>>
>>>>>> http://www.spinics.net/lists/alsa-devel/msg12535.html
>>>>>>
>>>>>> As I'm not subscribed to alsa-devel, I don't have the original post. 
>>>>>> Maybe any of the authors can post it do this list as well?
>>>>>>
>>>>>> The patch mainly applies to our git. I fixed two small rejects in 
>>>>>> Kconfig and Makefile as the patch is for newer kernel than our git. 
>>>>>> Untested patch against our git in attachment. Anybody likes to test?
>>>>>>     
>>>>>>         
>>>>>>           
>>>>> Well, I tried compiling it and it failed to build. It appears that it's 
>>>>> built on a set of patches for Davinci posted to LAK, including this 
>>>>> completely revamped EDMA support[1]:
>>>>>
>>>>> http://www.spinics.net/lists/arm-kernel/msg47450.html
>>>>>       
>>>>>         
>> Yep, it's needed.
>>   
>>>>> So I had to change a few definitions, import a few functions from the new
>>>>> DMA support and managed to get it compiling and loading, but not yet working
>>>>> (tried various sample rates and formats):
>>>>>
>>>>> root at davinci-dvevm:/tst$ aplay test.wav
>>>>> Playing WAVE 'test.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo
>>>>> ALSA lib pcm_params.c:2152:(snd_pcm_hw_refine_slave) Slave PCM not usable
>>>>> aplay: set_params:878: Broken configuration for this PCM: no configurations
>>>>> available
>>>>>
>>>>> root at davinci-dvevm:/tst$ arecord -r 48000 -f S16_LE
>>>>> Recording WAVE 'stdin' : Signed 16 bit Little Endian, Rate 48000 Hz, Mono
>>>>> ALSA lib pcm_params.c:2152:(snd_pcm_hw_refine_slave) Slave PCM not usable
>>>>> arecord: set_params:878: Broken configuration for this PCM: no
>>>>> configurations available
>>>>>
>>>>>
>>>>> And this is the extent of my knowledge of ALSA - I can continue digging, but
>>>>> maybe somebody else will beat me to it =) It's looking promising though.
>>>>>
>>>>> Attached is the patch I have so far that applies against current git. This
>>>>> is not for going upstream, but just for testing. Notes:
>>>>> - Apply patch
>>>>> - Enable SND_SOC, SND_DAVINCI_SOC, SND_DAVINCI_SOC_I2S, SND_DAVINCI_SOC_EVM
>>>>> and SND_SOC_TLV320AIC3X.
>>>>> - Make sure CONFIG_I2C_DAVINCI is enabled, but disable the
>>>>> CONFIG_SENSORS_TLV320* options.
>>>>> - Build.
>>>>> - Load the snd-soc-evm module.
>>>>>
>>>>> Bernard.
>>>>>
>>>>> [1] These patches appear to have gone absolutely nowhere beyond that list. I
>>>>> can't find them in any tree. But they are cleaner and more conforming to the
>>>>> Linux Kernel style than what's currently in git.
>>>>>
>>>>>   
>>>>> ------------------------------------------------------------------------
>>>>>
>>>>> _______________________________________________
>>>>> Davinci-linux-open-source mailing list
>>>>> Davinci-linux-open-source at linux.davincidsp.com
>>>>> http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
>>>>>   
>>>>>       
>>>>>         
>>>> _______________________________________________
>>>> Alsa-devel mailing list
>>>> Alsa-devel at alsa-project.org
>>>> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
>>>>
>>>>     
>>>>       
>>>   
>>>     
>>
>> _______________________________________________
>> Alsa-devel mailing list
>> Alsa-devel at alsa-project.org
>> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
>>
>>
>>