On Thu, Aug 14, 2014 at 5:07 AM, Tomeu Vizoso
[off-list ref] wrote:

On 08/13/2014 11:46 AM, Boris BREZILLON wrote:

quoted

Hi Tomeu,

Sorry for the late reply.

On Wed,  6 Aug 2014 15:56:03 +0200
Tomeu Vizoso [off-list ref] wrote:

quoted

Hi,

in this v5 of the patchset I have just moved the storage of the clock
constraints to the struct clk, as suggested by Stephen. Follows the original
cover letter blurb:

I'm retaking Rabin's patches [0] for splitting the clk API in two: one
API for
clk consumers and another for providers. The consumer API uses a clk
structure
that just keeps track of the consumer and has a reference to the actual
clk_core struct, which is used internally.

I have kept a patch from Rabin that aims to aid in debugging nested
enable/disable calls, though my personal aim is to allow more than one
consumer
to influence the final, effective frequency rate. For now this is limited
to
setting floor and ceiling constraints, with the short-term aim of
allowing
devfreq and thermal drivers to set floor and ceiling frequencies on the
memory
clock, respectively.

For those functions in the consumer clk API that were called from
providers, I
have added variants to clk-provider.h that are the same only that accept
a
clk_core instead. These functions are prefixed with clk_provider_.

Patch 1/6 just adds a bunch of defines with the goal of having all the
renames
in their own commit while preserving git-bisectability, with patch 3/6
containing the rename itself as generated by the Coccinelle script in
[1].
Patch 2/6 is needed because sound/soc/mxs/mxs-saif.c calls both the
consumer
and the provider API. The actual implementation of the API split comes in
patch
4/6. I will be happy to organize the refactoring differently if anybody
has a
better idea.

Patch 5/6 warns when there's an unbalanced usage of the enable and
disable
APIs, and patch 6/6 adds the API for setting floor and ceiling
frequencies, per
consumer.

I tested your patch series on an at91 platform (sama5d3), and it works
as expected, but I had to fix some conflicts when applying your patches
on clk-next, and then got a few errors at compile time.

Anyway here is my branch with all those conflicts resolved: [1]. The
last commit [2] fixes the build errors (I'll let you squash/split the
changes as you wish).

Thanks a lot, it has saved me quite some time.

Ok, that's great. I also applied v6 against Linus' master and
linux-next from today, August 14. Both of them have the same conflicts
in arch/arm code, namely i.MX, OMAP2+ and Kirkwood. Could this series
be refreshed against -rc1 after it comes out? That should be the last
rebase hurdle for a while.

Regards,
Mike

I have just re-run my coccinelle script and added the new file to the input
file list. Will be sending v6 now.

Regards,

Tomeu

quoted

Best Regards,

Boris

[1]https://github.com/bbrezillon/linux-at91/tree/per-clk-contraints

[2]https://github.com/bbrezillon/linux-at91/commit/d366c37dcfa5f06de3e27fc3c2807017bece9a2f