Re: [PATCH 03/15] soc: renesas: Add R-Car M3-N support

[jacopo mondi]

Hi Geert,

On Wed, Feb 14, 2018 at 01:48:27PM +0100, Geert Uytterhoeven wrote:
> Hi Jacopo,
>
> Thanks for your patch!
>
> On Tue, Feb 13, 2018 at 10:45 AM, Jacopo Mondi
> <jacopo+renesas@xxxxxxxxxx> wrote:
> > Add support for R-Car M3-N (r8a77965) power areas and reset.
> > M3-N power areas are identical to M3-W ones, so just copy and rename
> > them.
>
> They are not identical:
>   - M3-N does not have the CA53-related areas,
>   - M3-W does not have A3VP,
>   - M3-N does not have A2VC0 (M3-W also doesn't, according to latest
> datasheet?).
>
> The datasheet also mentions A3SH, without further info about the register
> block. I think we need to bring this up with Renesas.
>
> >  .../bindings/power/renesas,rcar-sysc.txt           |  1 +
> >  .../devicetree/bindings/reset/renesas,rst.txt      |  1 +
> >  drivers/soc/renesas/Kconfig                        |  9 ++++--
> >  drivers/soc/renesas/Makefile                       |  1 +
> >  drivers/soc/renesas/r8a77965-sysc.c                | 37 ++++++++++++++++++++++
> >  drivers/soc/renesas/rcar-rst.c                     |  1 +
> >  drivers/soc/renesas/rcar-sysc.c                    |  3 ++
> >  drivers/soc/renesas/rcar-sysc.h                    |  1 +
> >  drivers/soc/renesas/renesas-soc.c                  |  8 +++++
> >  include/dt-bindings/power/r8a77965-sysc.h          | 31 ++++++++++++++++++
> >  10 files changed, 91 insertions(+), 2 deletions(-)
> >  create mode 100644 drivers/soc/renesas/r8a77965-sysc.c
> >  create mode 100644 include/dt-bindings/power/r8a77965-sysc.h
>
> The maintainer may ask you to split this patch by functionality...
>
> > --- /dev/null
> > +++ b/drivers/soc/renesas/r8a77965-sysc.c
> > @@ -0,0 +1,37 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Renesas R-Car M3-N System Controller
> > + * Copyright (C) 2018 Jacopo Mondi <jacopo+renesas@xxxxxxxxxx>
> > + *
> > + * Based on Renesas R-Car M3-W System Controller
> > + * Copyright (C) 2016 Glider bvba
> > + */
> > +
> > +#include <linux/bug.h>
> > +#include <linux/kernel.h>
> > +
> > +#include <dt-bindings/power/r8a77965-sysc.h>
> > +
> > +#include "rcar-sysc.h"
> > +
> > +static const struct rcar_sysc_area r8a77965_areas[] __initconst = {
> > +       { "always-on",      0, 0, R8A77965_PD_ALWAYS_ON, -1, PD_ALWAYS_ON },
> > +       { "ca57-scu",   0x1c0, 0, R8A77965_PD_CA57_SCU, R8A77965_PD_ALWAYS_ON,
> > +         PD_SCU },
> > +       { "ca57-cpu0",   0x80, 0, R8A77965_PD_CA57_CPU0, R8A77965_PD_CA57_SCU,
> > +         PD_CPU_NOCR },
> > +       { "ca57-cpu1",   0x80, 1, R8A77965_PD_CA57_CPU1, R8A77965_PD_CA57_SCU,
> > +         PD_CPU_NOCR },
> > +       { "cr7",        0x240, 0, R8A77965_PD_CR7,      R8A77965_PD_ALWAYS_ON },
> > +       { "a3vc",       0x380, 0, R8A77965_PD_A3VC,     R8A77965_PD_ALWAYS_ON },
> > +       { "a2vc0",      0x3c0, 0, R8A77965_PD_A2VC0,    R8A77965_PD_A3VC },
>
> M3-N (and M3-W) does not have A2VC0?

Why do I still see that power area listed in latest renesas-drivers for
M3-W? Are there patch pendings for that?

I'll remove it anyway for M3-N.

Thanks
   j

>
> > +       { "a2vc1",      0x3c0, 1, R8A77965_PD_A2VC1,    R8A77965_PD_A3VC },
> > +       { "3dg-a",      0x100, 0, R8A77965_PD_3DG_A,    R8A77965_PD_ALWAYS_ON },
> > +       { "3dg-b",      0x100, 1, R8A77965_PD_3DG_B,    R8A77965_PD_3DG_A },
> > +       { "a3ir",       0x180, 0, R8A77965_PD_A3IR,     R8A77965_PD_ALWAYS_ON },
>
> A3VP is missing?
>
> > +};
> > +
> > +const struct rcar_sysc_info r8a77965_sysc_info __initconst = {
> > +       .areas = r8a77965_areas,
> > +       .num_areas = ARRAY_SIZE(r8a77965_areas),
> > +};
>
> > --- /dev/null
> > +++ b/include/dt-bindings/power/r8a77965-sysc.h
> > @@ -0,0 +1,31 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/*
> > + * Copyright (C) 2018 Jacopo Mondi <jacopo+renesas@xxxxxxxxxx>
> > + * Copyright (C) 2016 Glider bvba
> > + */
> > +
> > +#ifndef __DT_BINDINGS_POWER_R8A77965_SYSC_H__
> > +#define __DT_BINDINGS_POWER_R8A77965_SYSC_H__
> > +
> > +/*
> > + * These power domain indices match the numbers of the interrupt bits
> > + * representing the power areas in the various Interrupt Registers
> > + * (e.g. SYSCISR, Interrupt Status Register)
> > + */
> > +
> > +#define R8A77965_PD_CA57_CPU0           0
> > +#define R8A77965_PD_CA57_CPU1           1
> > +#define R8A77965_PD_A3VP                9
> > +#define R8A77965_PD_CA57_SCU           12
> > +#define R8A77965_PD_CR7                        13
> > +#define R8A77965_PD_A3VC               14
> > +#define R8A77965_PD_3DG_A              17
> > +#define R8A77965_PD_3DG_B              18
> > +#define R8A77965_PD_A3IR               24
> > +#define R8A77965_PD_A2VC0              25
>
> M3-N (and M3-W) does not have A2VC0?
>
> > +#define R8A77965_PD_A2VC1              26
> > +
> > +/* Always-on power area */
> > +#define R8A77965_PD_ALWAYS_ON          32
> > +
> > +#endif /* __DT_BINDINGS_POWER_R8A77965_SYSC_H__ */
>
> 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