This is very old, but I'm still interested in using I2S to directly drive 
an audio device (software defined radio module).
Could you post your write-up if available? Thanks.

On Tuesday, May 16, 2017 at 11:21:30 AM UTC-7, Graham wrote:
>
> It will be a day or so, but I will post the write-up here.
>
> You will not be able to control the IC independently, since the kernel 
> claims the device.
> The Linux audio system expects to be able to control the audio controls 
> such as volume, etc.
> by reaching into the CODEC through the control bus.  The kernel blocks 
> direct user
> space access to the device.
>
> In fact one way to make sure the device tree loaded correctly is that the 
> "UU" symbol
> appears at the devices' I2C address.
>
> --- Graham
>
> ==
>
> On Tuesday, May 16, 2017 at 10:51:19 AM UTC-5, ags wrote:
>>
>> I would appreciate the writeup, thank you.
>>
>> My project requires interfacing to an IC with I2S input. I planned on 
>> using (something like) aplay to write audio out to I2S/McASP channel (using 
>> built-in driver support) and hoped I could also use the built-in support 
>> (drivers) for SPI and/or I2C using the /dev/spidev<x.y> or /dev/i2c-<n> 
>> devices to control the IC.
>>
>> On Monday, May 15, 2017 at 6:46:50 PM UTC-7, Graham wrote:
>>>
>>> Yes, I was able to get a CircuitCo Rev.B Audio cape running, using the 
>>> I2S/McASP interface.
>>> I'll write it up for you, if you are interested.
>>>
>>> If you are going to use a different CODEC or other device on the 
>>> I2S/McASP interface,
>>> you will need to see if a driver already exists for it, or if it fits a 
>>> generalized I2S audio interface that
>>> is already inside the kernel can be invoked and controlled from a device 
>>> tree.
>>>
>>> If it is unique, then you will have to write your own Linux driver and 
>>> recompile the kernel.
>>>
>>> If you just want audio, it is a lot easier to just use a USB CODEC.
>>>
>>> --- Graham
>>>
>>> ==
>>>
>>> On Monday, May 15, 2017 at 6:11:11 PM UTC-5, ags wrote:
>>>>
>>>> I'm also interested as I have a project where I will interface directly 
>>>> to the I2S/McASP interface. Did you figure it out?
>>>>
>>>>
>>>> On Tuesday, January 3, 2017 at 8:57:29 AM UTC-8, Graham wrote:
>>>>>
>>>>> I spent most of the weekend down in the rabbit-hole, trying to get a
>>>>> CircuitCo Rev_B Audio cape to work, (unsuccessfully.)
>>>>>
>>>>> Is this cape compatible-with / supported-by Debian 8.6/kernel 4.4 ?
>>>>>
>>>>> Does the BB-BONE-AUDI-02-00A0.dtbo overlay that comes with the current 
>>>>> distribution work?
>>>>>
>>>>> How can you tell if an overlay actually loaded, with 4.4? 
>>>>>     /sys/devices/platform/bone_capemgr/slots 
>>>>> as well as the boot log, only shows the first physical four, and no 
>>>>> longer shows the higher numbered "pseudo-capes" and overlay status.
>>>>>
>>>>> I understand that
>>>>> The CircuitCo cape does not have an EEPROM, so everything needs to be 
>>>>> configured explicitly.
>>>>> I need to use a base .dts with HDMI audio disabled, then load the 
>>>>> overlay for the CircuitCo card.
>>>>> I also need to load the asound.state file.
>>>>>
>>>>> Am I approaching this correctly?
>>>>>
>>>>> I can not use a USB-soundcard for audio. I have several applications
>>>>> that need McASP/I2S running for several other codecs, but I thought I 
>>>>> would start 
>>>>> with the CircuitCo cape as a starting point.
>>>>>
>>>>> Thanks,
>>>>> --- Graham
>>>>>
>>>>> ==
>>>>>
>>>>>

-- 
For more options, visit http://beagleboard.org/discuss
--- 
You received this message because you are subscribed to the Google Groups 
"BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/beagleboard/3b1bb63a-326a-4822-9835-6bbf3439db63%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to