Re: [PATCH v1] ARM: imx_v6_v7_defconfig: enable SND_SOC_SPDIF

[Stefan Wahren <wahrenst@gmx.net>]

Am 05.11.24 um 14:58 schrieb Elinor Montmasson:
> Hi Stefan,
>
> On Monday, 4 November, 2024 17:20:38, Stefan Eichenberger wrote:
>> Hi Stefan,
>>
>> On Mon, Nov 04, 2024 at 12:39:40PM +0100, Stefan Wahren wrote:
>>> Hi Stefan,
>>>
>>> Am 04.11.24 um 09:05 schrieb Stefan Eichenberger:
>>>> Hi Stefan,
>>>>
>>>> On Sat, Nov 02, 2024 at 04:35:19PM +0100, Stefan Wahren wrote:
>>>>> Hi Stefan,
>>>>>
>>>>> Am 02.11.24 um 15:36 schrieb Stefan Eichenberger:
>>>>>> Hi Shawn,
>>>>>>
>>>>>> On Sat, Nov 02, 2024 at 03:21:58PM +0800, Shawn Guo wrote:
>>>>>>> On Wed, Oct 30, 2024 at 01:21:12PM +0100, Stefan Eichenberger wrote:
>>>>>>>> From: Stefan Eichenberger <stefan.eichenberger@toradex.com>
>>>>>>>>
>>>>>>>> Enable SND_SOC_SPDIF in imx_v6_v7_defconfig to support SPDIF audio. This
>>>>>>>> change will fix commit d469b771afe1 ("ARM: dts: imx6: update spdif sound
>>>>>>>> card node properties") which moves away from the old "spdif-controller"
>>>>>>>> property to the new "audio-codec" property.
>>>>>>>>
>>>>>>>> Fixes: d469b771afe1 ("ARM: dts: imx6: update spdif sound card node properties")
>>>>>>> It doesn't look a fix to me.
>>>>>> I agree somehow, it was just that before the referenced commit our test
>>>>>> succeeds with the imx_v6_v7_defconfig and after that we get the
>>>>>> following error:
>>>>>> [   24.165534] platform sound-spdif: deferred probe pending: fsl-asoc-card:
>>>>>> snd_soc_register_card failed
>>>>> this error should have been in the commit message including the
>>>>> information which platform/board is affected.
>>>> Okay, I will add this information to the next version. We see this error
>>>> on an Apalis iMX6 which has in my variant an NXP i.MX6Q SoC.
>>>>
>>>>>> So maybe it is not a fix in the sense of a bug, but it fixes the error
>>>>>> message. However, I'm also fine with removing the Fixes tag.
>>>>> But this patch doesn't look like the real approach.
>>>>>
>>>>> Could you please clarify the impact of the regression?
>>>> So the problem is that before commit d469b771afe1 ("ARM: dts: imx6:
>>>> update spdif sound card node properties") the audio driver was
>>>> using an implementation of linux,spdif-dit and linux,spdif-dir which was
>>>> directly inside the fsl,imx-audio-spdif compatible driver. Now with the
>>>> referenced commit the idea is to use the more generic linux,spdif-dir
>>>> and linux,spdif-dit compatible drivers. That's why this driver must be
>>>> enabled in the kernel configuration.
>>>>
>>>>> Is it just this error message and audio works fine or is audio also broken?
>>>> It is not just the error message, audio is not working because the
>>>> driver deferes and because it is not enabled it will never succeed to
>>>> load. I don't know if this is called a regression, because the driver is
>>>> there it is just not enabled in the imx6_v7_defconfig. I thought because
>>>> a lot of the i.MX6 based board use the generic driver, it makes sense to
>>>> enable it in the imx_v6_v7_defconfig.
>>> okay, thanks for the clarification. From my understanding
>>> imx6_v7_defconfig is just an example config for testing. All possible
>>> users of these boards might have their own configs and stumble across
>>> the same issue. So I thought it would be better to add the dependency in
>>> the Kconfig of the FSL audio driver.
>>>
>>> I'm not that audio driver expert and don't know how the dependency
>>> handling between the FSL audio driver and the required codecs like
>>> SND_SOC_SPDIF. So it's possible that I'm completely wrong here and your
>>> approach is the best we can do.
>> That might be a good point. I don't know how this is usually handled.
>> @Shawn and @Elinor, do you think this could be an approach to make
>> SND_SOC_FSL_ASOC_CARD select SND_SOC_SPDIF? It already seems to do this
>> for SND_SOC_WM8994 and SND_SOC_FSL_SPDIF.
> SND_SOC_FSL_ASOC_CARD will compile the machine driver fsl-asoc-card,
> SND_SOC_FSL_SPDIF the CPU DAI driver fsl_spdif for the SPDIF
> and SND_SOC_SPDIF the codec drivers spdif-rx and spdif-tx.
>
> In my commit series I made SND_SOC_FSL_ASOC_CARD select SND_SOC_FSL_SPDIF
> because the old machine driver previously compiled with SND_SOC_IMX_SPDIF
> selected SND_SOC_FSL_SPDIF.
> But because fsl-asoc-card is a generic driver, it could be used on a system
> that doesn't have an SPDIF device, and therefore should not require
> SND_SOC_SPDIF nor SND_SOC_FSL_SPDIF.
> So maybe it is not a good idea to automatically select SND_SOC_FSL_SPDIF or SND_SOC_SPDIF.
>
> On the other hand, if every imx6 or imx7 boards have an SPDIF device, then
> I suppose SND_SOC_SPDIF can be put in imx_v6_v7_defconfig.
Okay, I'm fine with the original approach.
>
> Regards,
> Elinor
>