[PATCH v2 1/7] clk: Add a basic factor clock

[Michael Turquette]

Quoting Maxime Ripard (2015-07-23 23:50:17)
> Hi Mike,
> 
> On Thu, Jul 23, 2015 at 05:00:01PM -0700, Michael Turquette wrote:
> > Quoting Maxime Ripard (2015-05-23 00:49:30)
> > > On Fri, May 22, 2015 at 12:35:35PM +0800, Chen-Yu Tsai wrote:
> > > > Hi,
> > > > 
> > > > On Fri, May 22, 2015 at 4:54 AM, Maxime Ripard
> > > > <maxime.ripard at free-electrons.com> wrote:
> > > > > Some clocks are using a factor component, however, unlike their mux, gate
> > > > > or divider counterpart, these factors don't have a basic clock
> > > > > implementation.
> > > > 
> > > > I think "multiplier" would be a better name here, considering it is the
> > > > counterpart of "divider". "factor" implies you can multiply and/or divide
> > > > the clock rate.
> > > 
> > > You're probably right, I didn't though of it that way :)
> > 
> > I also prefer multiplier over factor.
> 
> Ok. Note that I re-submitted it some time this week and forgot to
> change it, but I'm perfectly fine with the name.

Hi Maxime,

Yes, I saw your V2 after the fact.

> 
> > Jim Quinlan submitted a similar patch. See here:
> > 
> > http://www.spinics.net/lists/linux-clk/msg00691.html
> 
> It looks very similar indeed.
> 
> > I nacked that patch because Stephen and I are trying to figure out how
> > the basic clock types should work going forward. Code reuse is good, but
> > they are not very maintainable.
> 
> What are the issues with maintaining them? The only drawback I'm
> seeing with introducing such a driver is that you can't really have a
> clock that is both a divider and a multiplier, but that can be solved
> by splitting it into two sub-clocks.

There are a bunch of problems with the basic clock types. First is that
there is some feature creep every merge window that subtly breaks an
existing user (e.g. the round_rate stuff in clk-divider.c), and then
there are the growing number of flags to handle corner cases that are
one-off quirks for a single chip.

These make it harder to maintain, but it is possible.

The real problem with these basic clock types is that they are an
abstraction layer at the wrong level. Each clock type implements both
the policy of a given clock, as well as the machine-specific details.
For example clk-divider.c has made some assumptions in the past about
rounding the rate, or how to calculate the best divider; this is a
matter of policy and is useful on its own. But additionally that same
policy is glued to a specific implementation: memory-mapped register
controls for a clock divider.

The I/O accessor stuff needs to be addressed at some point.  Currently
the basic clock types assume specific patterns of access to
memory-mapped clock registers. There are lots of other clock controls
out there that talk to firmware, or over i2c, or whatever. The amount of
code that has to be copy/pasted for each different type of access is
100%; i.e. we do not have abstractions at the right level such as
.get_best_div(struct clk_hw *hw, unsigned long rate).

What I would like to see in time is a re-usable layer for clock policy
(e.g. common rules for how dividers or multipliers should behave), and
then have that sit on top of the machine-specific callbacks that
directly touch the hardware, such as the .get_best_div callback above.

For this reason I like to limit the number of new basic clock types. If
there is a single user then I'm inclined to have the author put it with
the machine-specific code. But in this case since there are two users, I
see the value in making a new basic clock type. It will just be more
code to update in the far away future when we get around to implementing
the changes above.

> 
> From a pure maintainance point of view, refusing it would mean that
> each and every platform would have to come up with its own
> implementation. For example, we do have clk-factors.c for sunxi that
> does just that, and implies some cooperation from each clock driver
> that have to provide some code to determine the various components of
> the output formula. This can prove to be very challenging (and bug
> prone) for clocks like the audio one we have where we have 1
> multiplier and 2 dividers that needs some adjusting.
> 
> Splitting it into sub-clocks for each of these components would allow
> to have less bugs, while keeping the whole thing very simple, and the
> implementation on the driver side very trivial.
> 
> Overall, the clk-factors code we have (client side) is approximately a
> thousand lines of code logic that could be replaced by (less) trivial
> probing code for such a driver.
> 
> > Since there are two potential users of this code, I should reconsider.
> 
> Like I said, eventually, I'd like to leverage that code a lot more
> than for the single clock alone, and I think it could benefit other
> platforms too (like Jim has proven).
> 
> > Can you two come up with a common implementation that works for both of
> > you?
> 
> Yes, sure. Jim, how do you want to go with this? Do you want to
> resubmit your patch on top of 4.2-rc something so that I could give it
> a try? Otherwise, I posted such a patch this week
> 
> https://lkml.kernel.org/r/1437235304-2208-1-git-send-email-maxime.ripard@free-electrons.com
> 
> Can you give it a try and your feedback?
> 
> > Please do not put any CLK_OF_DECLARE stuff in there (there isn't
> > any now, but I wanted to be clear).
> 
> That's never been my intention ;)

OK, sounds great.

Thanks,
Mike

> 
> Thanks!
> Maxime
> 
> -- 
> Maxime Ripard, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel