[GIT PULL] imx6q: fix for v3.2-rc

[GIT PULL] imx6q: fix for v3.2-rc

[Shawn Guo]

Hi Arnd,

Please pull the imx6q fix below for v3.2-rc.

Regards,
Shawn

The following changes since commit cfcfc9eca2bcbd26a8e206baeb005b055dbf8e37:

  Linux 3.2-rc2 (2011-11-15 15:02:59 -0200)

are available in the git repository at:
  git://git.linaro.org/people/shawnguo/linux-2.6.git imx6q/fixes

Richard Zhao (1):
      ARM: imx6q: move clock register map to machine_desc.map_io

 arch/arm/mach-imx/clock-imx6q.c         |    7 +++++--
 arch/arm/mach-imx/mach-imx6q.c          |    1 +
 arch/arm/plat-mxc/include/mach/common.h |    1 +
 3 files changed, 7 insertions(+), 2 deletions(-)

[GIT PULL] imx6q: fix for v3.2-rc

[Arnd Bergmann]

On Thursday 17 November 2011, Shawn Guo wrote:
> Please pull the imx6q fix below for v3.2-rc.
> 
> Regards,
> Shawn
> 
> The following changes since commit cfcfc9eca2bcbd26a8e206baeb005b055dbf8e37:
> 
>   Linux 3.2-rc2 (2011-11-15 15:02:59 -0200)
> 
> are available in the git repository at:
>   git://git.linaro.org/people/shawnguo/linux-2.6.git imx6q/fixes
> 
> Richard Zhao (1):
>       ARM: imx6q: move clock register map to machine_desc.map_io
> 

Hi Shawn and Richard,

I've applied the patch to the 'fixes' branch since it seems correct,
but please explain to me what bug this actually fixes. The changelog
does not actually mention why this is broken or why the fix is needed
in 3.2.

	Arnd

[GIT PULL] imx6q: fix for v3.2-rc

[Richard Zhao]

On 24 November 2011 05:15, Arnd Bergmann <arnd@arndb.de> wrote:
> On Thursday 17 November 2011, Shawn Guo wrote:
>> Please pull the imx6q fix below for v3.2-rc.
>>
>> Regards,
>> Shawn
>>
>> The following changes since commit cfcfc9eca2bcbd26a8e206baeb005b055dbf8e37:
>>
>> ? Linux 3.2-rc2 (2011-11-15 15:02:59 -0200)
>>
>> are available in the git repository at:
>> ? git://git.linaro.org/people/shawnguo/linux-2.6.git imx6q/fixes
>>
>> Richard Zhao (1):
>> ? ? ? ARM: imx6q: move clock register map to machine_desc.map_io
>>
>
> Hi Shawn and Richard,
>
> I've applied the patch to the 'fixes' branch since it seems correct,
> but please explain to me what bug this actually fixes. The changelog
> does not actually mention why this is broken or why the fix is needed
> in 3.2.
The patch fixes bug: clock register map corrupt after system boot up.
It's because the map was done in machine_desc.timer.init. It's too late to
call iotable_init. memblock_alloc can not be used.

You can also find the discuss mail thread in
http://comments.gmane.org/gmane.linux.ports.arm.kernel/140544

Sorry for insufficient commit log.

Thanks
Richard
>
> ? ? ? ?Arnd
>