Re: [PATCH v2 4/6] ASoC: renesas: fsi: refactor clock initialization

[Bui Duc Phuc]

Hi Morimoto-san,

> Ah, OK.
> So the patch 1) moves fsi_clk_init() and 2) update it.
> It is including many features in 1 patch. Please separate it.
>
> One note here is that /* clock function */ is for all fsi_clk_xxx(),
> so don't move it.

Thank you for your guidance.
I will split this into two separate patches accordingly.

Best regards,
Phuc

On Wed, Apr 15, 2026 at 11:55 AM Kuninori Morimoto
<kuninori.morimoto.gx@xxxxxxxxxxx> wrote:
>
>
> Hi Bui
>
> > > I have mentioned in previous mail to just move fsi_clk_init(), but why do
> > > you need to move it ? It works without any issue without moving function,
> > > I guess ?
> >
> > I moved fsi_clk_init() below the two functions fsi_clk_set_rate_cpg
> > and fsi_clk_set_rate_external because, inside fsi_clk_init(),
> > I assign these functions to clock->set_rate. Moving the function was
> > necessary to avoid compilation errors.
>
> Ah, OK.
> So the patch 1) moves fsi_clk_init() and 2) update it.
> It is including many features in 1 patch. Please separate it.
>
> One note here is that /* clock function */ is for all fsi_clk_xxx(),
> so don't move it.
>
> > > And why you need to call fsi_clk_init() twice ?
> > The FSI controller has two independent ports (Port A and Port B).
> > Each port requires its own clock resource initialization and configuration.
>
> Ah, yes indeed.
>
>
> Thank you for your help !!
>
> Best regards
> ---
> Kuninori Morimoto