[PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Kuninori Morimoto]

Hi Paul, Simon

These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
This is 1st step of FSI-DT support.
My plan is...

 1st : sh/ARM: add FSI-DIV clock control functions
 2nd : FSI driver supports audio clock calculation method
 3rd : sh/ARM: remove old style .set_rate callback from platform
 4th : FSI support DT
 5th : platform use DT for FSI

Kuninori Morimoto (3):
      sh: clkfwk: add sh_clk_fsidiv_register()
      ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
      ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks

 arch/arm/mach-shmobile/board-ap4evb.c        |    2 +-
 arch/arm/mach-shmobile/board-mackerel.c      |    2 +-
 arch/arm/mach-shmobile/clock-sh7372.c        |   92 +++-----------------------
 arch/arm/mach-shmobile/include/mach/sh7372.h |    2 -
 drivers/sh/clk/cpg.c                         |   86 ++++++++++++++++++++++++
 include/linux/sh_clk.h                       |    9 +++
 6 files changed, 107 insertions(+), 86 deletions(-)


Best regards
---
Kuninori Morimoto

Re: [PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Simon Horman]

On Tue, Oct 30, 2012 at 08:05:59PM -0700, Kuninori Morimoto wrote:
> 
> Hi Paul, Simon
> 
> These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
> This is 1st step of FSI-DT support.
> My plan is...
> 
>  1st : sh/ARM: add FSI-DIV clock control functions
>  2nd : FSI driver supports audio clock calculation method
>  3rd : sh/ARM: remove old style .set_rate callback from platform
>  4th : FSI support DT
>  5th : platform use DT for FSI
> 
> Kuninori Morimoto (3):
>       sh: clkfwk: add sh_clk_fsidiv_register()
>       ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
>       ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks

I am slightly confused about why 4 patches, but its not a big deal.

Paul, how do you want to handle merging this?  The 1st patch looks like its
appropriate for your tree while the rest seem to belong in my tree. But I'm
happy for all the changes to go through one tree if that makes sense.

Re: [PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Paul Mundt]

On Wed, Oct 31, 2012 at 03:08:26PM +0800, Simon Horman wrote:
> On Tue, Oct 30, 2012 at 08:05:59PM -0700, Kuninori Morimoto wrote:
> > 
> > Hi Paul, Simon
> > 
> > These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
> > This is 1st step of FSI-DT support.
> > My plan is...
> > 
> >  1st : sh/ARM: add FSI-DIV clock control functions
> >  2nd : FSI driver supports audio clock calculation method
> >  3rd : sh/ARM: remove old style .set_rate callback from platform
> >  4th : FSI support DT
> >  5th : platform use DT for FSI
> > 
> > Kuninori Morimoto (3):
> >       sh: clkfwk: add sh_clk_fsidiv_register()
> >       ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> >       ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> 
> I am slightly confused about why 4 patches, but its not a big deal.
> 
> Paul, how do you want to handle merging this?  The 1st patch looks like its
> appropriate for your tree while the rest seem to belong in my tree. But I'm
> happy for all the changes to go through one tree if that makes sense.

It's not such a big deal either way. Once the first patch has been
corrected with the appropriate I/O accessors you're welcome to take it
with the rest with my Acked-by. I don't believe we have anything on the
SH side that depends on it, so there's not a lot of point in decoupling
it from the rest of the patch set.

Re: [PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Kuninori Morimoto]

Hi Simon

> > > These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
> > > This is 1st step of FSI-DT support.
> > > My plan is...
> > > 
> > >  1st : sh/ARM: add FSI-DIV clock control functions
> > >  2nd : FSI driver supports audio clock calculation method
> > >  3rd : sh/ARM: remove old style .set_rate callback from platform
> > >  4th : FSI support DT
> > >  5th : platform use DT for FSI
> > > 
> > > Kuninori Morimoto (3):
> > >       sh: clkfwk: add sh_clk_fsidiv_register()
> > >       ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> > >       ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> > 
> > I am slightly confused about why 4 patches, but its not a big deal.
> > 
> > Paul, how do you want to handle merging this?  The 1st patch looks like its
> > appropriate for your tree while the rest seem to belong in my tree. But I'm
> > happy for all the changes to go through one tree if that makes sense.
> 
> It's not such a big deal either way. Once the first patch has been
> corrected with the appropriate I/O accessors you're welcome to take it
> with the rest with my Acked-by. I don't believe we have anything on the
> SH side that depends on it, so there's not a lot of point in decoupling
> it from the rest of the patch set.

Now ALSA SoC maintainer accepted above 2nd step patch-set of FSI driver,
So, I would like to send 3rd step patches to SH-ML.
But it needs above (1st step) patch-set.
Where can I find ?

Best regards
---
Kuninori Morimoto

Re: [PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Simon Horman]

On Wed, Nov 07, 2012 at 04:09:12PM -0800, Kuninori Morimoto wrote:
> 
> Hi Simon
> 
> > > > These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
> > > > This is 1st step of FSI-DT support.
> > > > My plan is...
> > > > 
> > > >  1st : sh/ARM: add FSI-DIV clock control functions
> > > >  2nd : FSI driver supports audio clock calculation method
> > > >  3rd : sh/ARM: remove old style .set_rate callback from platform
> > > >  4th : FSI support DT
> > > >  5th : platform use DT for FSI
> > > > 
> > > > Kuninori Morimoto (3):
> > > >       sh: clkfwk: add sh_clk_fsidiv_register()
> > > >       ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> > > >       ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> > > 
> > > I am slightly confused about why 4 patches, but its not a big deal.
> > > 
> > > Paul, how do you want to handle merging this?  The 1st patch looks like its
> > > appropriate for your tree while the rest seem to belong in my tree. But I'm
> > > happy for all the changes to go through one tree if that makes sense.
> > 
> > It's not such a big deal either way. Once the first patch has been
> > corrected with the appropriate I/O accessors you're welcome to take it
> > with the rest with my Acked-by. I don't believe we have anything on the
> > SH side that depends on it, so there's not a lot of point in decoupling
> > it from the rest of the patch set.
> 
> Now ALSA SoC maintainer accepted above 2nd step patch-set of FSI driver,
> So, I would like to send 3rd step patches to SH-ML.
> But it needs above (1st step) patch-set.
> Where can I find ?

Hi Morimoto-san,

sorry for being a bit slow. There seem to be four patches in this series:

[PATCH 1/3] sh: clkfwk: add sh_clk_fsidiv_register()
[PATCH 2/3] ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
[PATCH 3/3] ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
[PATCH 4/3] ARM: shmobile: r8a7740: add FSI-DVI clocks


It seems that the first patch for drivers/sh/clk/cpg.c and
include/linux/sh_clk.h and thus would normally go through Paul's tree.
And that it is a dependency of the other three patches.

Paul, I wonder if it would be easiest to try and push all of these changes
through my tree.

Re: [PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Paul Mundt]

On Thu, Nov 08, 2012 at 10:06:31AM +0900, Simon Horman wrote:
> On Wed, Nov 07, 2012 at 04:09:12PM -0800, Kuninori Morimoto wrote:
> > 
> > Hi Simon
> > 
> > > > > These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
> > > > > This is 1st step of FSI-DT support.
> > > > > My plan is...
> > > > > 
> > > > >  1st : sh/ARM: add FSI-DIV clock control functions
> > > > >  2nd : FSI driver supports audio clock calculation method
> > > > >  3rd : sh/ARM: remove old style .set_rate callback from platform
> > > > >  4th : FSI support DT
> > > > >  5th : platform use DT for FSI
> > > > > 
> > > > > Kuninori Morimoto (3):
> > > > >       sh: clkfwk: add sh_clk_fsidiv_register()
> > > > >       ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> > > > >       ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> > > > 
> > > > I am slightly confused about why 4 patches, but its not a big deal.
> > > > 
> > > > Paul, how do you want to handle merging this?  The 1st patch looks like its
> > > > appropriate for your tree while the rest seem to belong in my tree. But I'm
> > > > happy for all the changes to go through one tree if that makes sense.
> > > 
> > > It's not such a big deal either way. Once the first patch has been
> > > corrected with the appropriate I/O accessors you're welcome to take it
> > > with the rest with my Acked-by. I don't believe we have anything on the
> > > SH side that depends on it, so there's not a lot of point in decoupling
> > > it from the rest of the patch set.
> > 
> > Now ALSA SoC maintainer accepted above 2nd step patch-set of FSI driver,
> > So, I would like to send 3rd step patches to SH-ML.
> > But it needs above (1st step) patch-set.
> > Where can I find ?
> 
> Hi Morimoto-san,
> 
> sorry for being a bit slow. There seem to be four patches in this series:
> 
> [PATCH 1/3] sh: clkfwk: add sh_clk_fsidiv_register()
> [PATCH 2/3] ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> [PATCH 3/3] ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> [PATCH 4/3] ARM: shmobile: r8a7740: add FSI-DVI clocks
> 
> 
> It seems that the first patch for drivers/sh/clk/cpg.c and
> include/linux/sh_clk.h and thus would normally go through Paul's tree.
> And that it is a dependency of the other three patches.
> 
> Paul, I wonder if it would be easiest to try and push all of these changes
> through my tree.
> 
Yes, it's fine going through your tree. You can add my acked-by if you
wish.

Re: [PATCH 0/3] sh/ARM: shmobile: add sh_clk_fsidiv_register()

[Simon Horman]

On Thu, Nov 08, 2012 at 02:28:29PM +0900, Paul Mundt wrote:
> On Thu, Nov 08, 2012 at 10:06:31AM +0900, Simon Horman wrote:
> > On Wed, Nov 07, 2012 at 04:09:12PM -0800, Kuninori Morimoto wrote:
> > > 
> > > Hi Simon
> > > 
> > > > > > These patches add sh_clk_fsidiv_register() for FSI-DIV clock control.
> > > > > > This is 1st step of FSI-DT support.
> > > > > > My plan is...
> > > > > > 
> > > > > >  1st : sh/ARM: add FSI-DIV clock control functions
> > > > > >  2nd : FSI driver supports audio clock calculation method
> > > > > >  3rd : sh/ARM: remove old style .set_rate callback from platform
> > > > > >  4th : FSI support DT
> > > > > >  5th : platform use DT for FSI
> > > > > > 
> > > > > > Kuninori Morimoto (3):
> > > > > >       sh: clkfwk: add sh_clk_fsidiv_register()
> > > > > >       ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> > > > > >       ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> > > > > 
> > > > > I am slightly confused about why 4 patches, but its not a big deal.
> > > > > 
> > > > > Paul, how do you want to handle merging this?  The 1st patch looks like its
> > > > > appropriate for your tree while the rest seem to belong in my tree. But I'm
> > > > > happy for all the changes to go through one tree if that makes sense.
> > > > 
> > > > It's not such a big deal either way. Once the first patch has been
> > > > corrected with the appropriate I/O accessors you're welcome to take it
> > > > with the rest with my Acked-by. I don't believe we have anything on the
> > > > SH side that depends on it, so there's not a lot of point in decoupling
> > > > it from the rest of the patch set.
> > > 
> > > Now ALSA SoC maintainer accepted above 2nd step patch-set of FSI driver,
> > > So, I would like to send 3rd step patches to SH-ML.
> > > But it needs above (1st step) patch-set.
> > > Where can I find ?
> > 
> > Hi Morimoto-san,
> > 
> > sorry for being a bit slow. There seem to be four patches in this series:
> > 
> > [PATCH 1/3] sh: clkfwk: add sh_clk_fsidiv_register()
> > [PATCH 2/3] ARM: shmobile: sh7372: sh7372_fsidivX_clk become non-global
> > [PATCH 3/3] ARM: shmobile: sh7372: use sh_clk_fsidiv_register() for FSI-DIV clocks
> > [PATCH 4/3] ARM: shmobile: r8a7740: add FSI-DVI clocks
> > 
> > 
> > It seems that the first patch for drivers/sh/clk/cpg.c and
> > include/linux/sh_clk.h and thus would normally go through Paul's tree.
> > And that it is a dependency of the other three patches.
> > 
> > Paul, I wonder if it would be easiest to try and push all of these changes
> > through my tree.
> > 
> Yes, it's fine going through your tree. You can add my acked-by if you
> wish.

Thanks. I already sent a pull request, but I expect to send a revised one.
I will add your ack.