> -----Original Message-----
> From: Pierre-Louis Bossart <pierre-louis.bossart@xxxxxxxxxxxxxxx>
> Sent: Monday, March 13, 2023 7:47 AM
> To: “Ryan <ryan.lee.analog@xxxxxxxxx>; lgirdwood@xxxxxxxxx;
> broonie@xxxxxxxxxx; perex@xxxxxxxx; tiwai@xxxxxxxx;
> krzysztof.kozlowski@xxxxxxxxxx; rf@xxxxxxxxxxxxxxxxxxxxx;
> ckeepax@xxxxxxxxxxxxxxxxxxxxx; herve.codina@xxxxxxxxxxx;
> wangweidong.a@xxxxxxxxxx; james.schulman@xxxxxxxxxx;
> ajye_huang@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx; shumingf@xxxxxxxxxxx;
> povik+lin@xxxxxxxxxxx; flatmax@xxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx;
> alsa-devel@xxxxxxxxxxxxxxxx; robh+dt@xxxxxxxxxx;
> devicetree@xxxxxxxxxxxxxxx; Lee, RyanS <RyanS.Lee@xxxxxxxxxx>
> Subject: Re: [PATCH V2 1/2] ASoC: max98363: add soundwire amplifier driver
>
> [External]
>
>
> > + ret = sdw_stream_add_slave(max98363->slave, &stream_config,
> > + &port_config, 1, stream);
> > + if (ret) {
> > + dev_err(dai->dev, "Unable to configure port\n");
> > + return ret;
> > + }
> > +
> > + if (params_channels(params) > 16) {
> > + dev_err(component->dev, "Unsupported channels %d\n",
> > + params_channels(params));
> > + return -EINVAL;
> > + }
>
> Do you actually support more than 8 channels?
>
> The data port DPnPrepareCtl and DPn_ChannelEn registers expose 8
> channels max. It's always possible to 'cheat' by packing two channels in the
> same sample, but that would require custom signaling between manager and
> peripheral that isn't present.
>
> Could it be a left-over from a TDM implementation?
Thanks for the comment.
I think it is a left-over from a TDM implementation.
I shall modify the code to accept only max number of channels supported by the amp.
>
> The rest of the patch looks fine.
