RE: [PATCH] clk: clk-fixed-factor: Use new macro CLK_OF_DECLARE_DRIVER

From: Rajan Vaja
Date: Fri Mar 16 2018 - 07:49:47 EST


Hi Stephen,

Thanks for the comment.

> -----Original Message-----
> From: Stephen Boyd [mailto:sboyd@xxxxxxxxxx]
> Sent: 16 March 2018 12:17 AM
> To: Rajan Vaja <RAJANV@xxxxxxxxxx>
> Cc: linux-clk@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Jolly Shah
> <JOLLYS@xxxxxxxxxx>; Michal Simek <michals@xxxxxxxxxx>;
> mturquette@xxxxxxxxxxxx
> Subject: RE: [PATCH] clk: clk-fixed-factor: Use new macro
> CLK_OF_DECLARE_DRIVER
>
> Quoting Rajan Vaja (2018-03-09 11:27:40)
> > > From: Stephen Boyd [mailto:sboyd@xxxxxxxxxx]
> > >
> > > Is the intent to register the clk twice? I believe things are working as
> > > intended without this patch, so maybe you can explain a little more what
> > > you're trying to fix.
> > [Rajan] Yes. During of_clk_init() if some DT fixed factor clock has
> > parent which is neither mentioned in output-clock-names of clock
> > controller nor registered as clock provider, of_clk_init() will try to
> > forcefully register in second loop.
> >
> > if (force || parent_ready(clk_provider->np)) {
> >
> > /* Don't populate platform devices */
> > of_node_set_flag(clk_provider->np,
> > OF_POPULATED);
> >
> > So registration of this DT fixed-factor clock would fail as parent
> > would be NULL as below (called from _of_fixed_factor_clk_setup()):
> > parent_name = of_clk_get_parent_name(node, 0);
> >
> > On the other hand, even if registration failed, that node will be
> > marked as OF_POPULATED, so probe of clk-fixed-factor.c will also not
> > be called and that DT fixed-factor clock would never be registered.
> >
> > Same thing is discussed at https://lkml.org/lkml/2017/6/5/681 .
>
> Ok. I believe the answer is to fix the DT to describe the parent chain
> properly with clock-output-names. Otherwise, we have no good way of
> figuring out what the name should be.
[Rajan] clock DT binding doc says that clock-output-names property
is optional and sometimes not recommended.
I think this patch fixes the issue we have which mandates to add clock-output-names
property (for this particular case). Also, IIUC platform driver probe in clk-fixed-factor.c
will never be called unless we use CLK_OF_DECLARE_DRIVER.
I completely agree that proper solution would be to stop using strings to
describe clock topology.

>
> The alternative is to start working on a solution for having the clk
> framework stop using strings to describe clk topology. My plan there is
> to allow clk registration to indicate that another parent names array
> should be used with clk_get() of the device or node pointer that's
> associated with the clk during registration to find the parent . If we
> had that, then we could hook up clks into the tree by calling clk_get()
> in the framework and map through the clock-names property. This also
> gets us to a point where clk names don't have to be globally unique,
> which would be nice.