[PATCH 0/6 v2] MERAM support for LCDC

[PATCH 0/6 v2] MERAM support for LCDC

[Damian Hobson-Garcia]

This patch series adds support for read-ahead cache-access from the 
LCDC to the main memory.  These patches enabled the LCDC to access
the MERAM memory via ICB.  The specific ICBs to use are chip dependent
and defined in the platform data.

The MERAM clock will enable when the LCDC clocks do, and are allowed to 
be disbled when the LCDC clocks are.

Changes from V1
---------------
* Specified meram info data (struct sh_mobile_meram_info *meram_dev) as
  part of the LCDC device data, not the LCDC channel data, since the info
  data corresponds to the whole device, not just one configuration.

* Added dynamic clock enable/disable via runtime_pm.

* Changed .id from -1 to 0 in platform data

Damian Hobson-Garcia (6):
  sh_mobile: Add MSTP clock for MERAM
  sh_mobile_meram: MERAM framework for LCDC
  sh_mobile_meram: Add support for NV24 framebuffers
  sh_mobile_meram: MERAM platform data for LCDC
  fbdev: sh_mobile_meram: Add clock enable/disble via runtime PM
  fbdev: sh_mobile_meram: Enable/disable MERAM clocks along with LCDC
    clocks

 arch/arm/mach-shmobile/board-ap4evb.c   |   73 ++++
 arch/arm/mach-shmobile/board-mackerel.c |   74 ++++
 arch/arm/mach-shmobile/clock-sh7372.c   |    4 +-
 drivers/video/Kconfig                   |   12 +
 drivers/video/Makefile                  |    1 +
 drivers/video/sh_mobile_lcdcfb.c        |  117 ++++++-
 drivers/video/sh_mobile_lcdcfb.h        |    1 +
 drivers/video/sh_mobile_meram.c         |  594 +++++++++++++++++++++++++++++++
 drivers/video/sh_mobile_meram.h         |   41 +++
 include/video/sh_mobile_lcdc.h          |    3 +
 include/video/sh_mobile_meram.h         |   74 ++++
 11 files changed, 984 insertions(+), 10 deletions(-)
 create mode 100644 drivers/video/sh_mobile_meram.c
 create mode 100644 drivers/video/sh_mobile_meram.h
 create mode 100644 include/video/sh_mobile_meram.h

Re: [PATCH 0/6 v2] MERAM support for LCDC

[Paul Mundt]

On Wed, May 18, 2011 at 08:10:05PM +0900, Damian Hobson-Garcia wrote:
> This patch series adds support for read-ahead cache-access from the 
> LCDC to the main memory.  These patches enabled the LCDC to access
> the MERAM memory via ICB.  The specific ICBs to use are chip dependent
> and defined in the platform data.
> 
> The MERAM clock will enable when the LCDC clocks do, and are allowed to 
> be disbled when the LCDC clocks are.
> 
> Changes from V1
> ---------------
> * Specified meram info data (struct sh_mobile_meram_info *meram_dev) as
>   part of the LCDC device data, not the LCDC channel data, since the info
>   data corresponds to the whole device, not just one configuration.
> 
> * Added dynamic clock enable/disable via runtime_pm.
> 
> * Changed .id from -1 to 0 in platform data
> 
> Damian Hobson-Garcia (6):
>   sh_mobile: Add MSTP clock for MERAM
>   sh_mobile_meram: MERAM framework for LCDC
>   sh_mobile_meram: Add support for NV24 framebuffers

These basically seem to be fine, and make up for the bulk of the new
code, so I'll take them as-is. I'll let you and Magnus hash out the
remaining bits after this.

Re: [PATCH 0/6 v2] MERAM support for LCDC

[Damian Hobson-Garcia]

Hi Paul,

On 2011/05/23 15:51, Paul Mundt wrote:
> On Wed, May 18, 2011 at 08:10:05PM +0900, Damian Hobson-Garcia wrote:
>> This patch series adds support for read-ahead cache-access from the 
>> LCDC to the main memory.  These patches enabled the LCDC to access
>> the MERAM memory via ICB.  The specific ICBs to use are chip dependent
>> and defined in the platform data.
>>
>> The MERAM clock will enable when the LCDC clocks do, and are allowed to 
>> be disbled when the LCDC clocks are.
>>
>> Changes from V1
>> ---------------
>> * Specified meram info data (struct sh_mobile_meram_info *meram_dev) as
>>   part of the LCDC device data, not the LCDC channel data, since the info
>>   data corresponds to the whole device, not just one configuration.
>>
>> * Added dynamic clock enable/disable via runtime_pm.
>>
>> * Changed .id from -1 to 0 in platform data
>>
>> Damian Hobson-Garcia (6):
>>   sh_mobile: Add MSTP clock for MERAM
>>   sh_mobile_meram: MERAM framework for LCDC
>>   sh_mobile_meram: Add support for NV24 framebuffers
> 
> These basically seem to be fine, and make up for the bulk of the new
> code, so I'll take them as-is. I'll let you and Magnus hash out the
> remaining bits after this.

Ok great, thanks a lot. I'll submit any further patches to apply on top
of these three.

Damian

Re: [PATCH 0/6 v2] MERAM support for LCDC

[Magnus Damm]

Hi Damian,

On Tue, May 24, 2011 at 11:53 AM, Damian Hobson-Garcia
<dhobsong@igel.co.jp> wrote:
> Hi Paul,
>
> On 2011/05/23 15:51, Paul Mundt wrote:
>> On Wed, May 18, 2011 at 08:10:05PM +0900, Damian Hobson-Garcia wrote:
>>> This patch series adds support for read-ahead cache-access from the
>>> LCDC to the main memory.  These patches enabled the LCDC to access
>>> the MERAM memory via ICB.  The specific ICBs to use are chip dependent
>>> and defined in the platform data.
>>>
>>> The MERAM clock will enable when the LCDC clocks do, and are allowed to
>>> be disbled when the LCDC clocks are.
>>>
>>> Changes from V1
>>> ---------------
>>> * Specified meram info data (struct sh_mobile_meram_info *meram_dev) as
>>>   part of the LCDC device data, not the LCDC channel data, since the info
>>>   data corresponds to the whole device, not just one configuration.
>>>
>>> * Added dynamic clock enable/disable via runtime_pm.
>>>
>>> * Changed .id from -1 to 0 in platform data
>>>
>>> Damian Hobson-Garcia (6):
>>>   sh_mobile: Add MSTP clock for MERAM
>>>   sh_mobile_meram: MERAM framework for LCDC
>>>   sh_mobile_meram: Add support for NV24 framebuffers
>>
>> These basically seem to be fine, and make up for the bulk of the new
>> code, so I'll take them as-is. I'll let you and Magnus hash out the
>> remaining bits after this.
>
> Ok great, thanks a lot. I'll submit any further patches to apply on top
> of these three.

The LCDC driver does no longer compile:

linux-2.6 git 14d74e0cab7a7779a7ff0c3863c04c8a8e507106

  CC      drivers/video/sh_mobile_meram.o
  CC      drivers/video/sh_mobile_lcdcfb.o
drivers/video/sh_mobile_lcdcfb.c: In function 'sh_mobile_lcdc_start':
drivers/video/sh_mobile_lcdcfb.c:640:4: error: 'ret' undeclared (first
use in this function)
drivers/video/sh_mobile_lcdcfb.c:640:4: note: each undeclared
identifier is reported only once for each function it appears in
make[2]: *** [drivers/video/sh_mobile_lcdcfb.o] Error 1
make[1]: *** [drivers/video] Error 2
make: *** [drivers] Error 2

I suspect this change may be related to the MERAM patch, but I'm not sure.

Can you please have a look?

Thanks,

/ magnus

Re: [PATCH 0/6 v2] MERAM support for LCDC

[Damian Hobson-Garcia]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 2041 bytes --]

Hi Magnus,

On Thu, 26 May 2011, Magnus Damm wrote:

> Hi Damian,
>
> On Tue, May 24, 2011 at 11:53 AM, Damian Hobson-Garcia
> <dhobsong@igel.co.jp> wrote:
>> Hi Paul,
>>
>> On 2011/05/23 15:51, Paul Mundt wrote:
>>> On Wed, May 18, 2011 at 08:10:05PM +0900, Damian Hobson-Garcia wrote:
>>>>
>>>> Damian Hobson-Garcia (6):
>>>>   sh_mobile: Add MSTP clock for MERAM
>>>>   sh_mobile_meram: MERAM framework for LCDC
>>>>   sh_mobile_meram: Add support for NV24 framebuffers
>>>
>>> These basically seem to be fine, and make up for the bulk of the new
>>> code, so I'll take them as-is. I'll let you and Magnus hash out the
>>> remaining bits after this.
>>
>> Ok great, thanks a lot. I'll submit any further patches to apply on top
>> of these three.
>
> The LCDC driver does no longer compile:
>
> linux-2.6 git 14d74e0cab7a7779a7ff0c3863c04c8a8e507106
>
>  CC      drivers/video/sh_mobile_meram.o
>  CC      drivers/video/sh_mobile_lcdcfb.o
> drivers/video/sh_mobile_lcdcfb.c: In function 'sh_mobile_lcdc_start':
> drivers/video/sh_mobile_lcdcfb.c:640:4: error: 'ret' undeclared (first
> use in this function)
> drivers/video/sh_mobile_lcdcfb.c:640:4: note: each undeclared
> identifier is reported only once for each function it appears in
> make[2]: *** [drivers/video/sh_mobile_lcdcfb.o] Error 1
> make[1]: *** [drivers/video] Error 2
> make: *** [drivers] Error 2
>
> I suspect this change may be related to the MERAM patch, but I'm not sure.
>
> Can you please have a look?

It seems that when this patch

69843ba7f24950f8ef5dadacfbfbd08f53e3455b
fbdev: sh_mobile_lcdc: reduce scope of a variable

and the meram stuff was merged together there was a problem.

The definition of ret got stashed away into an if-clause, making it
unavaliable for the call on line 640 that the compiler complains about.

I'll submit another pactch to the list that will resolve the conflict, but
as it looks like a merge issue, I don't know how exactly that should be 
dealt with.

Thanks,
Damian