[alsa-devel] Kernel Device Tree entries for simple-audio-card

Rick Mann rmann at latencyzero.com
Tue Sep 29 00:34:04 CEST 2015


> On Sep 28, 2015, at 15:22 , Caleb Crome <caleb at crome.org> wrote:
> 
> On Mon, Sep 28, 2015 at 11:55 AM, Rick Mann <rmann at latencyzero.com> wrote:
>> That's very similar to what I have (I use ax40 & axr2). I also use macsp0_ahclkx, but I'm not seeing a signal on it (or any other pin).
>> 
> 
> 
>> Does this work for you with a 4.1.x kernel?
> 
> Yes, it works for me.  I assume your codec is master, do you see WCLK
> and BCLK running?  How about MCLK?
> 
> Who is supposed to be driving MCLK in your case, the BBB or the codec?

According to someone else, and as you assume above, the BBB should generate the MCLK and the TLV generates WCLK and BCLK. That's counterintuitive to me, but the configuration it's sending to the TLV for register 0x08 is 0xC0, which sets the bits such that WCLK and BCLK are outputs (master mode). I think MCLK is always an input on this device.

(BTW, thanks for asking. In verifying what I just wrote, I learned a lot about how it's supposed to work).

> I see that you have the AHCLKX set to INPUT.  This should probably
> either be output (if the CPU is driving it), or not connected if the
> codec is driving it.
> 
>> 0x1ac (PIN_INPUT | MUX_MODE0) // P9_25 MCLK MCLK mcasp0_ahclkx

Yeah, I changed that the other day:

	https://github.com/JetForMe/podtique/blob/v1/bbb/cape/Podtique1/BB-ENABLE-PRU.dts

pinctrl-single,pins = <                             //  Hdr     I2S     TLV     SoC
                                                    //  -----   ----    ----    -------------
    0x190 (PIN_INPUT_PULLDOWN | MUX_MODE0)          //  P9_31   BCLK    BCLK    mcasp0_aclkx
    0x194 (PIN_INPUT_PULLDOWN | MUX_MODE0)          //  P9_29   WCLK    WCLK    mcasp0_fsx
    0x198 (PIN_INPUT_PULLDOWN | MUX_MODE0)          //  P9_30   RX      SDOUT   mcasp0_axr0
    0x19c (PIN_INPUT_PULLDOWN | MUX_MODE2)          //  P9_28   TX      SDIN    mcasp0_axr2
    0x1ac (PIN_OUTPUT_PULLDOWN | MUX_MODE0)         //  P9_25   MCLK    MCLK    mcasp0_ahclkx
>;

The TRM indicates all the pins should have their receiver enabled, but I decided to make MCLK an output. 

In all cases, NONE of the pins, including MCLK, changes state. BTW, when I run speaker-test, it gets this far and hangs:

$ speaker-test 

speaker-test 1.0.28

Playback device is default
Stream parameters are 48000Hz, S16_LE, 1 channels
Using 16 octaves of pink noise
Rate set to 48000Hz (requested 48000Hz)
Buffer size range from 128 to 32768
Period size range from 8 to 2048
Using max buffer size 32768
Periods = 4
was set period_size = 2048
was set buffer_size = 32768
 0 - Front Left

Which makes me think some internal sound data buffer is filling up, and the write is blocking, and nothing is emptying the buffer because the McASP isn't accepting bytes, and therefore not sending data out.

If I interrupt speaker-test, it usually outputs this:

^CWrite error: -4,Interrupted system call
xrun_recovery failed: -4,Interrupted system call
Transfer failed: Interrupted system call

Further reinforcing my theory. Which suggests something's still not right for McASP0.




> 
> -Caleb
> 
> 
> 
> 
>> 
>>> On Sep 28, 2015, at 09:11 , Caleb Crome <caleb at crome.org> wrote:
>>> 
>>> I used these pins to get the audio cape going.
>>> 
>>> +/*
>>> +   * MCASP pin mapping
>>> +   *                          ball   BBB
>>> +   * MCASP0_ACLKX             A13    P9.31
>>> +   * MCASP0_AFSX              B13    P9.29
>>> +   * MCASP0_AXR2  (mcasp OUT) C12    P9.28
>>> +   * MCASP0_AXR3  (mcasp IN)  A14    P9.25
>>> +*/
>>> 
>>> mcasp_0_pins_default: mcasp_0_pins_default {
>>> pinctrl-single,pins = <
>>>    0x190 ( PIN_INPUT | MUX_MODE0 ) /* (A13) mcasp0_aclkx.mcasp0_aclkx */
>>>   0x194 ( PIN_INPUT | MUX_MODE0 ) /* (B13) mcasp0_fsx.mcasp0_fsx */
>>>   0x19c ( PIN_OUTPUT | MUX_MODE2 ) /* (C12) mcasp0_ahclkr.mcasp0_axr2 */
>>>   0x1ac ( PIN_INPUT | MUX_MODE2 ) /* (A14) mcasp0_ahclkx.mcasp0_axr3 */
>>>> ;
>>> };
>>> 
>>> And importantly, the audio cape needs to use the AXR2 and AXR3 for my use.
>>> 
>>> +&mcasp0 {
>>> + pinctrl-names = "default";
>>> + pinctrl-0 = <&mcasp_0_pins_default>;
>>> + status = "okay";
>>> +
>>> + op-mode = <0>;          /* MCASP_IIS_MODE */
>>> + tdm-slots = <8>;
>>> + num-serializer = <16>;
>>> + serial-dir = <  /* 0: INACTIVE, 1: TX, 2: RX */
>>> + 0 0 1 2
>>> + 0 0 0 0
>>> + 0 0 0 0
>>> + 0 0 0 0
>>> + >;
>>> + tx-num-evt = <1>;
>>> + rx-num-evt = <1>;
>>> };
>>> 
>>> 
>>> Hope that can be of some help.
>>> 
>>> 
>>> -Caleb
>>> 
>>> 
>>> On Sun, Sep 27, 2015 at 10:14 PM, Rick Mann <rmann at latencyzero.com> wrote:
>>>> Yeah, it turned out I needed a -cells = <0> on the mcasp0 peripheral.
>>>> 
>>>> I've got it finding it and creating a sound card, and if I run speaker-test, it sends a bunch of i2c commands to the codec (which aren't quite right). But the McASP/IIS pins never change state, so, I've made progress, but I'm still stuck.
>>>> 
>>>> Latest DTBO:
>>>> 
>>>>       https://github.com/JetForMe/podtique/blob/v1/bbb/cape/Podtique1/BB-ENABLE-PRU.dts
>>>> 
>>>>> On Sep 27, 2015, at 14:52 , Caleb Crome <caleb at crome.org> wrote:
>>>>> 
>>>>> I found that too.  Check out the documentation here:
>>>>> http://lxr.free-electrons.com/source/Documentation/devicetree/bindings/sound/simple-card.txt
>>>>> 
>>>>> You need to put in the #sound-dai-cells cell.   Your value will likely
>>>>> be 0 for the codec, and maybe 1 for the SOC dai, but I"m not sure.
>>>>> 
>>>>> Unfortunately, I don't really understand how it works.
>>>>> 
>>>>> -Caleb
>>>>> 
>>>>> 
>>>>> On Fri, Sep 25, 2015 at 10:57 PM, Rick Mann <rmann at latencyzero.com> wrote:
>>>>>> Hi. I've built a Beaglebone Black Cape that has, among other things, a TI Audio CODEC on it (TLV320AIC3104). This is the same CODEC as the BBB Audio Cape (http://elinux.org/CircuitCo:Audio_Cape_RevB), and it's connected in almost the exact same way.
>>>>>> 
>>>>>> The last few days I've been figuring out the Device Tree overlay needed to get Linux (4.1.4) to recognize it. It's finding the device and instantiating the low-level driver (tlv320aic3x-codec), although I'm not sure at this point if it's actually talking to it (I have verified that low-level i2c commands do work).
>>>>>> 
>>>>>> But when it comes to the sound card setup, I'm having trouble getting the Device Tree right. I'm hoping someone on this list can offer some insight. The DT is here:
>>>>>> 
>>>>>>      https://github.com/JetForMe/podtique/blob/master/bbb/cape/Podtique1/BB-ENABLE-PRU.dts
>>>>>> 
>>>>>> And this is dmesg when I try to load that (note the name of my overlay is BB-ENABLE-PRU, which is an artifact of the experimentation I've been doing; it'll change to something more appropriate):
>>>>>> 
>>>>>> [   65.859275] bone_capemgr bone_capemgr: part_number 'BB-ENABLE-PRU', version 'N/A'
>>>>>> [   65.859318] bone_capemgr bone_capemgr: slot #4: override
>>>>>> [   65.859336] bone_capemgr bone_capemgr: Using override eeprom data at slot 4
>>>>>> [   65.859355] bone_capemgr bone_capemgr: slot #4: 'Override Board Name,00A0,Override Manuf,BB-ENABLE-PRU'
>>>>>> [   65.859488] bone_capemgr: bone_capemgr bone_capemgr: slot #4: Requesting part number/version based 'BB-ENABLE-PRU-00A0.dtbo
>>>>>> [   65.859506] bone_capemgr: bone_capemgr bone_capemgr: slot #4: Requesting firmware 'BB-ENABLE-PRU-00A0.dtbo' for board-name 'Override Board Name', version '00A0'
>>>>>> [   65.861105] bone_capemgr: bone_capemgr bone_capemgr: slot #4: dtbo 'BB-ENABLE-PRU-00A0.dtbo' loaded; converting to live tree
>>>>>> [   65.875618] gpio-of-helper ocp:gpio_helper: ready
>>>>>> [   65.876327] gpiolib_of: of_get_named_gpiod_flags: parsed 'gpio' property of node '//fixedregulator at 1[0]' - status (0)
>>>>>> [   65.876619] core: lz-codec-reg: no parameters
>>>>>> [   65.876713] fixed: reg-fixed-voltage fixedregulator at 1: lz-codec-reg supplying 0uV
>>>>>> [   65.883284] pruss_uio 4a300000.pruss: pins are not configured from the driver
>>>>>> [   65.900080] bone_capemgr bone_capemgr: slot #4: dtbo 'BB-ENABLE-PRU-00A0.dtbo' loaded; overlay id #0
>>>>>> [   65.946254] gpiolib_of: of_get_named_gpiod_flags: parsed 'gpio-reset' property of node '/ocp/i2c at 4819c000/tlv320aic3104 at 0[0]' - status (0)
>>>>>> [   65.946317] core: tlv320aic3x-codec 2-0018: Looking up IOVDD-supply from device tree
>>>>>> [   65.946405] core: tlv320aic3x-codec 2-0018: Looking up DVDD-supply from device tree
>>>>>> [   65.946468] core: tlv320aic3x-codec 2-0018: Looking up AVDD-supply from device tree
>>>>>> [   65.946544] core: tlv320aic3x-codec 2-0018: Looking up DRVDD-supply from device tree
>>>>>> [   65.946606] snd_soc_core: tlv320aic3x-codec 2-0018: codec register 2-0018
>>>>>> [   65.946640] snd_soc_core: tlv320aic3x-codec 2-0018: ASoC: dai register 2-0018 #1
>>>>>> [   65.946655] snd_soc_core: tlv320aic3x-codec 2-0018: ASoC: Registered DAI 'tlv320aic3x-hifi'
>>>>>> [   65.946671] snd_soc_core: tlv320aic3x-codec 2-0018: ASoC: Registered codec 'tlv320aic3x-codec.2-0018'
>>>>>> [   65.956491] //sound/simple-audio-card,cpu: could not get #sound-dai-cells for /ocp/mcasp at 48038000
>>>>>> [   65.978759] snd_soc_core: davinci-mcasp 48038000.mcasp: ASoC: dai register 48038000.mcasp #1
>>>>>> [   65.978795] snd_soc_core: davinci-mcasp 48038000.mcasp: ASoC: Registered DAI '48038000.mcasp'
>>>>>> [   65.978954] snd_soc_core: davinci-mcasp 48038000.mcasp: ASoC: Registered platform '48038000.mcasp'
>>>>>> [   66.012427] asoc-simple-card sound: parse error -22
>>>>>> [   66.042609] asoc-simple-card: probe of sound failed with error -22
>>>>>> 
>>>>>> Note the bit about "#sound-dai-cells". I'm not sure what's wrong here, and it seems like it might be something in the distro-provided DTB, not my own.
>>>>>> 
>>>>>> aplay lists nothing (I think in part because I have yet to put in a proper asound.conf file):
>>>>>> 
>>>>>> # aplay -l
>>>>>> aplay: device_list:268: no soundcards found...
>>>>>> 
>>>>>> Can someone suggest some diagnostics I can do to see what the state of things is? I'm just not sure what to look for. Thanks!
>>>>>> 
>>>>>> --
>>>>>> Rick Mann
>>>>>> rmann at latencyzero.com
>>>>>> 
>>>>>> 
>>>>>> _______________________________________________
>>>>>> 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
>>>> 
>>>> 
>>>> --
>>>> Rick Mann
>>>> rmann at latencyzero.com
>>>> 
>>>> 
>> 
>> 
>> --
>> Rick Mann
>> rmann at latencyzero.com
>> 
>> 
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel at alsa-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel


-- 
Rick Mann
rmann at latencyzero.com




More information about the Alsa-devel mailing list