Re: [PATCH] clk: Mark fwnodes when their clock provider is added

[Geert Uytterhoeven <geert@linux-m68k.org>]

Hi Stephen,

On Wed, Mar 31, 2021 at 4:22 AM Stephen Boyd <sboyd@kernel.org> wrote:
> Quoting Geert Uytterhoeven (2021-03-29 23:58:23)
> > On Tue, Mar 30, 2021 at 3:53 AM Stephen Boyd <sboyd@kernel.org> wrote:
> > > Quoting Saravana Kannan (2021-03-29 16:28:20)
> > > > On Mon, Mar 29, 2021 at 2:25 PM Stephen Boyd <sboyd@kernel.org> wrote:
> > > > > Quoting Geert Uytterhoeven (2021-03-26 11:29:55)
> > > > > > On Fri, Mar 26, 2021 at 7:13 PM Stephen Boyd <sboyd@kernel.org> wrote:
> > > > > > > Quoting Nicolas Saenz Julienne (2021-03-25 11:25:24)
> > > > > > > > >
> > > > > > > > > This patch mainly revealed that clk/bcm/clk-raspberrypi.c driver calls
> > > > > > > > > devm_of_clk_add_hw_provider(), with a device pointer, which has a NULL
> > > > > > > > > dev->of_node. I'm not sure if adding a check for a NULL np in
> > > > > > > > > of_clk_add_hw_provider() is a right fix, though.
> > > > > > > >
> > > > > > > > I believe the right fix is not to call 'devm_of_clk_add_hw_provider()' if
> > > > > > > > 'pdev->dev.of_node == NULL'. In such case, which is RPi3's, only the CPU clock
> > > > > > > > is used, and it's defined and queried later through
> > > > > > > > devm_clk_hw_register_clkdev().
> > > > > > > >
> > > > > > > > @Marek, I don't mind taking care of it if it's OK with you.
> > > > > > > >
> > > > > > >
> > > > > > > Ah I see this is related to the patch I just reviewed. Can you reference
> > > > > > > this in the commit text? And instead of putting the change into the clk
> > > > > > > provider let's check for NULL 'np' in of_clk_add_hw_provider() instead
> > > > > > > and return 0 if there's nothing to do. That way we don't visit this
> > > > > > > problem over and over again.
> > > > > >
> > > > > > I'm not sure the latter is what we reall want: shouldn't calling
> > > > > > *of*_clk_add_hw_provider() with a NULL np be a bug in the provider?
> > > > > >
> > > > >
> > > > > I don't have a strong opinion either way. Would it be useful if the
> > > > > function returned an error when 'np' is NULL?
> > > >
> > > > I lean towards returning an error. Not a strong opinion either.
> > >
> > > Does it have any use?
> >
> > of_clk_del_provider() removes the first provider found with node == NULL.
> > If there are two drivers calling of_clk_add_hw_provider(), and one of
> > hem calls of_clk_del_provider() later, the wrong provider may be
> > removed from the list.
> >
>
> So you're saying we shouldn't add a NULL device node pointer to the list
> so that this can't happen? That doesn't mean returning an error from
> of_clk_add_hw_provider() would be useful though.
> of_clk_add_hw_provider() can return 0 if np == NULL and
> of_clk_del_provider() can return early if np == NULL too.

I don't know if I grasp all meanings of the above.

The main question is if it is valid for a driver to call
of_clk_add_hw_provider()
with np == NULL.
  - If yes, should that register the provider?
      - If yes, how to handle two drivers calling of_clk_add_hw_provider()
        with np = NULL, as their unregistration order is not guaranteed to
        be correct.

If no, is that something to ignore (0), or a bug (error)?

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

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