Patchwork [GIT,PULL] clk for next

login
register
mail settings
Submitter Sascha Hauer
Date May 4, 2012, 5:05 p.m.
Message ID <20120504170533.GO4141@pengutronix.de>
Download mbox
Permalink /patch/156971/
State New
Headers show

Pull-request

git://git.linaro.org/people/mturquette/linux.git clk-next

Comments

Sascha Hauer - May 4, 2012, 5:05 p.m.
Erm, no, this is not a pull request, more a request for a pull request.

Currently we are still missing the clk patches for next in -next. It
should be fine to have these in arm-soc since probably no non-arm
architecture will convert to the clock framework in the next two weeks.

Then finally we could merge SoC clk code in the arm-soc tree. BTW there
are lready spear clk patches in the arm-soc tree which conflict with
Mikes branch so Viresh might have to fix them (maybe he sent the fixes
already to the list, can't remember)


So Mike, please consider sending the following pull request to Arnd:

The following changes since commit 66f75a5d028beaf67c931435fdc3e7823125730c:

  Linux 3.4-rc4 (2012-04-21 14:47:52 -0700)

are available in the git repository at:

  git://git.linaro.org/people/mturquette/linux.git clk-next

for you to fetch changes up to 0e1c03017549a9df513622b3f15ff38eb8d35a62:

  clk: clk_set_rate() must fail if CLK_SET_RATE_GATE is set and clk is enabled (2012-05-01 18:13:38 -0700)

----------------------------------------------------------------
Mark Brown (2):
      clk: Remove comment for end of CONFIG_COMMON_CLK section
      clk: Constify parent name arrays

Mike Turquette (6):
      clk: core: correct clk_set_rate kerneldoc
      clk: core: remove dead code paths
      clk: core: clk_calc_new_rates handles NULL parents
      clk: core: enforce clk_ops consistency
      clk: core: copy parent_names & return error codes
      clk: basic: improve parent_names & return errors

Rajendra Nayak (2):
      clk: Make clk_get_rate() return 0 on error
      clk: constify parent name arrays in macros

Rob Herring (2):
      clk: select CLKDEV_LOOKUP for COMMON_CLK
      clk: remove trailing whitespace from clk.h

Saravana Kannan (1):
      clk: Use a separate struct for holding init data.

Shawn Guo (7):
      clk: use kzalloc in clk_register_mux
      clk: remove unnecessary EXPORT_SYMBOL_GPL
      clk: add "const" for clk_ops of basic clks
      clk: declare clk_ops of basic clks in clk-provider.h
      clk: always pass parent_rate into .round_rate
      clk: pass parent_rate into .set_rate
      clk: propagate round_rate for CLK_SET_RATE_PARENT case

Viresh Kumar (5):
      clk: Fix typo in comment
      clk: clk-gate: Create clk_gate_endisable()
      clk: clk-private: Add DEFINE_CLK macro
      clk: Don't set clk->new_rate twice
      clk: clk_set_rate() must fail if CLK_SET_RATE_GATE is set and clk is enabled

 drivers/clk/Kconfig          |    1 +
 drivers/clk/clk-divider.c    |   68 +++++------
 drivers/clk/clk-fixed-rate.c |   49 ++++----
 drivers/clk/clk-gate.c       |  104 ++++++++---------
 drivers/clk/clk-mux.c        |   26 +++--
 drivers/clk/clk.c            |  265 +++++++++++++++++++++++++++---------------
 include/linux/clk-private.h  |   79 +++++--------
 include/linux/clk-provider.h |   95 ++++++++-------
 include/linux/clk.h          |    6 +-
 9 files changed, 383 insertions(+), 310 deletions(-)
Mark Brown - May 4, 2012, 5:45 p.m.
On Fri, May 04, 2012 at 07:05:33PM +0200, Sascha Hauer wrote:

> Currently we are still missing the clk patches for next in -next. It
> should be fine to have these in arm-soc since probably no non-arm
> architecture will convert to the clock framework in the next two weeks.

Actually I've a patch which I keep pushing to the x86 maintainers to
enable the clock API there (ideally we'd enable it on all architectures
that don't currently have one of their own straight off).

> So Mike, please consider sending the following pull request to Arnd:

Seconded!

I'd also like to get the wm831x driver in there, I've not been looking
at it since it doesn't seem sensible to put the effort in until we've
got stuff flowing into -next.
Shawn Guo - May 5, 2012, 6:10 a.m.
On Fri, May 04, 2012 at 07:05:33PM +0200, Sascha Hauer wrote:
> Erm, no, this is not a pull request, more a request for a pull request.
> 
> Currently we are still missing the clk patches for next in -next. It
> should be fine to have these in arm-soc since probably no non-arm
> architecture will convert to the clock framework in the next two weeks.
> 
> Then finally we could merge SoC clk code in the arm-soc tree. BTW there
> are lready spear clk patches in the arm-soc tree which conflict with
> Mikes branch so Viresh might have to fix them (maybe he sent the fixes
> already to the list, can't remember)
> 
> 
> So Mike, please consider sending the following pull request to Arnd:
> 

And please do not forget to pick up the patch below, Mike.

Sascha Hauer (1):
	clk: add a fixed factor clock

Since Russell has published his clkdev branch, with above patch picked
up on clk-next, I have everything ready to send my platform porting.

Thanks,
Shawn
Shawn Guo - May 5, 2012, 9:23 a.m.
On Sat, May 05, 2012 at 02:10:18PM +0800, Shawn Guo wrote:
> And please do not forget to pick up the patch below, Mike.
> 
> Sascha Hauer (1):
> 	clk: add a fixed factor clock
> 
> Since Russell has published his clkdev branch, with above patch picked
> up on clk-next, I have everything ready to send my platform porting.
> 
Mike, I haven't seen the following fixup being picked up.  Can you
please do?

http://thread.gmane.org/gmane.linux.ports.arm.msm/2614/focus=2637
Turquette, Mike - May 6, 2012, 11:44 p.m.
On Sat, May 5, 2012 at 2:23 AM, Shawn Guo <shawn.guo@linaro.org> wrote:
> On Sat, May 05, 2012 at 02:10:18PM +0800, Shawn Guo wrote:
>> And please do not forget to pick up the patch below, Mike.
>>
>> Sascha Hauer (1):
>>       clk: add a fixed factor clock
>>
>> Since Russell has published his clkdev branch, with above patch picked
>> up on clk-next, I have everything ready to send my platform porting.
>>
> Mike, I haven't seen the following fixup being picked up.  Can you
> please do?
>
> http://thread.gmane.org/gmane.linux.ports.arm.msm/2614/focus=2637

Will do.

Regards,
Mike
Turquette, Mike - May 7, 2012, 2:20 a.m.
On Fri, May 4, 2012 at 10:45 AM, Mark Brown
<broonie@opensource.wolfsonmicro.com> wrote:
> On Fri, May 04, 2012 at 07:05:33PM +0200, Sascha Hauer wrote:
> I'd also like to get the wm831x driver in there, I've not been looking
> at it since it doesn't seem sensible to put the effort in until we've
> got stuff flowing into -next.

One thing to think about is the fact that CONFIG_COMMON_CLK pulls in
clkdev.  x86 doesn't have an arch/x86/include/asm/clkdev.h last time I
looked.

Regards,
Mike
Mark Brown - May 7, 2012, 9:24 a.m.
On Sun, May 06, 2012 at 07:20:18PM -0700, Turquette, Mike wrote:
> On Fri, May 4, 2012 at 10:45 AM, Mark Brown

> > On Fri, May 04, 2012 at 07:05:33PM +0200, Sascha Hauer wrote:
> > I'd also like to get the wm831x driver in there, I've not been looking
> > at it since it doesn't seem sensible to put the effort in until we've
> > got stuff flowing into -next.

> One thing to think about is the fact that CONFIG_COMMON_CLK pulls in
> clkdev.  x86 doesn't have an arch/x86/include/asm/clkdev.h last time I
> looked.

Everything seemed to manage to build happily enough when I was testing,
this is how I was build testing, though I'm not immediately sure why
looking again.  This does seem like we're stuck in the same failure that
we're in with gpiolib, and obviously the -next issues get in the way too :(