On Thu, Jun 29, 2023 at 06:33:09PM +0100, Srinivas Kandagatla wrote:
> On 29/06/2023 16:43, Mark Brown wrote:
> > On Wed, Jun 28, 2023 at 10:24:04AM +0100, Srinivas Kandagatla wrote:
> > > For some reason we ended up with a setup without this flag.
> > > This resulted in inconsistent sound card devices numbers which
> > > are also not starting as expected at dai_link->id.
> > > (Ex: MultiMedia1 pcm ended up with device number 4 instead of 0)
> > Why is this a problem?
> In existing Qualcomm setup the backend pcm are added first, which results in
> frontend pcms getting pcm numbers after this.
> For example: with 3 backend dailinks in DT we have frontend pcm start at 3.
> Now if we add new backend dai-link in DT we now have frontend pcm start at
> 4.
> This is a bug in qualcomm driver.
Why is this an actual problem rather than just being a bit ugly? What
is the negative consequence of having a PCM with this number?
> > > With this patch patch now the MultiMedia1 PCM ends up with device number 0
> > > as expected.
> > >
> > > Fixes: 9b4fe0f1cd79 ("ASoC: qdsp6: audioreach: add q6apm-dai support")
> > > Cc: <Stable@xxxxxxxxxxxxxxx>
> >
> > Won't this be an ABI change? That seems like it'd disrupt things in
> > stable.
> Yes, but this is a real bug. without fixing this also results in abi(pcm
> number) change when we add new backend dai-link. I have also sent fix for
> UCM to handle this.
I'm still not clear why you believe this to be a bug.
Attachment:
signature.asc
Description: PGP signature
