Re: [PATCH] ASoC: codecs: lpass-va-macro: Fix LPASS Codec Version for SC7280

[Srinivas Kandagatla <srini@kernel.org>]

On 6/11/26 11:16 AM, Luca Weiss wrote:
> On Wed Jun 10, 2026 at 12:52 PM CEST, Srinivas Kandagatla wrote:
>>
>> Thanks Luca for the patch,
>>
>> On 5/26/26 4:03 PM, Luca Weiss wrote:
>>> According to both the static definition in downstream...
>>>
>>>   yupik-audio-overlay.dtsi: qcom,bolero-version = <4>;
>>>   #define BOLERO_VERSION_2_0 0x0004)
>>>
>>> and the runtime detection:
>>>
>>>   CDC_VA_TOP_CSR_CORE_ID_0=0x1
>>>   CDC_VA_TOP_CSR_CORE_ID_1=0xf
>>>
>>> SC7280 has LPASS Codec Version 2.0 and not, as declared with
>>> sm8250_va_data LPASS_CODEC_VERSION_1_0.
>>>
>>> Create new va_macro_data with .version not set to use the runtime
>>> detection and correctly get .version = LPASS_CODEC_VERSION_2_0.
>>>
>>> Fixes: 77212f300bfd ("ASoC: codecs: lpass-va-macro: set the default codec version for sm8250")
>>
>> Can you help me understand if this change was fixing any issue, if so
>> can you pl add CC stable
> 
> I'm pretty sure that just came around because I was looking at the
> different parts of the audio stack and trying to figure out if there's
> something wrong somewhere, and how the different versions floating
> around downstream & upstream & documentation match up.
> 
> I'm not aware of a situation where this change makes a difference -
> apart from the dev_dbg() print.
> 
>>
>>> Signed-off-by: Luca Weiss <luca.weiss@fairphone.com>
>>> ---
>>> I'm fairly confident this is correct, but please someone double check
>>> this if they can.
>>
>> Yes, you are correct, this is actually Bolero 1.1.0 Version where things
>> have changed in codec. if you read major number it will be 1
>> This has been artificially tagged as 2.0 in the driver.
>> So the change that you are doing is correct.
>>
>> Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
> 
> Thanks a lot for checking! Opens even more questions in my mind, why
> 1.1.0 == 2.0 but somebody somewhere decided that at some point I guess.

I have no idea why this was done, Looking at the downstream drivers, am
guessing that there were two different variants of lpass codec drivers
in downstream one is less than v1 and other one is v2. 1.1.0 is
supported in v2 driver.

--srini
> 
> Regards
> Luca