Re: [PATCH v4 1/8] soc: renesas: rz-sysc: Add syscon/regmap support

[Geert Uytterhoeven]

Hi Claudiu,

On Fri, 8 Aug 2025 at 12:32, Claudiu Beznea <claudiu.beznea@xxxxxxxxx> wrote:
> On 08.08.2025 12:29, Geert Uytterhoeven wrote:
> > On Fri, 8 Aug 2025 at 08:18, Claudiu <claudiu.beznea@xxxxxxxxx> wrote:
> >> From: John Madieu <john.madieu.xa@xxxxxxxxxxxxxx>
> >>
> >> The RZ/G3E system controller has various registers that control or report
> >> some properties specific to individual IPs. The regmap is registered as a
> >> syscon device to allow these IP drivers to access the registers through the
> >> regmap API.
> >>
> >> As other RZ SoCs might have custom read/write callbacks or max-offsets,
> >> register a custom regmap configuration.
> >>
> >> Signed-off-by: John Madieu <john.madieu.xa@xxxxxxxxxxxxxx>
> >> [claudiu.beznea:
> >>  - do not check the match->data validity in rz_sysc_probe() as it is
> >>    always valid
> >>  - dinamically allocate regmap_cfg]
> >> Signed-off-by: Claudiu Beznea <claudiu.beznea.uj@xxxxxxxxxxxxxx>
> >> ---
> >>
> >> Changes in v4:
> >> - adjusted the patch description by dropping "add" from
> >>   "add register a custom regmap configuration"
> >> - updated the list of changes from Claudiu Beznea
> >> - dynamically allocate the regmap_config as proposed at [2]
> >> - this patch is needed for proper function of USB (as proposed in this
> >>   series) that being the reason it is introduced here, as well
> >>
> >> [2] https://lore.kernel.org/all/CAMuHMdVyf3Xtpw=LWHrnD2CVQX4xYm=FBHvY_dx9OesHDz5zNg@xxxxxxxxxxxxxx/
> >
> >> --- a/drivers/soc/renesas/rz-sysc.c
> >> +++ b/drivers/soc/renesas/rz-sysc.c
> > =
> >> @@ -117,7 +125,26 @@ static int rz_sysc_probe(struct platform_device *pdev)
> >>                 return PTR_ERR(sysc->base);
> >>
> >>         sysc->dev = dev;
> >> -       return rz_sysc_soc_init(sysc, match);
> >> +       ret = rz_sysc_soc_init(sysc, match);
> >> +       if (ret)
> >> +               return ret;
> >> +
> >> +       regmap_cfg = devm_kzalloc(dev, sizeof(*regmap_cfg), GFP_KERNEL);
> >> +       if (!regmap_cfg)
> >> +               return -ENOMEM;
> >
> > Is there any specific reason you decided to allocate regmap_cfg
> > separately, instead of embedding it into struct rz_sysc?
>
> Sorry, I missed to mention.
>
> I chose to have it like this as the regmap_cfg is not used anywhere else
> (through rz_sysc) except in probe.

OK.  Upon closer look, devm_regmap_init_mmio() does not save the
regmap_cfg pointer for later use, so it can be allocated using kzalloc()
instead, and freed immediately after calling devm_regmap_init_mmio().

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds