[GIT PULL] ARM: OMAP2+: hwmod data: cleanup in preparation for link array removal

[GIT PULL] ARM: OMAP2+: hwmod data: cleanup in preparation for link array removal

[Paul Walmsley]

Hi Tony

The following changes since commit b8a633451951d149494076393698880453bf6bfe:

  ARM: OMAP: hwmod: add softreset delay field and OMAP4 data (2012-02-28 03:46:31 -0700)

are available in the git repository at:

  git://git.pwsan.com/linux-2.6 hwmod_data_cleanup_3.4

for you to fetch changes up to a2d4b88dd32b02e8b5088e0d64b3c102c0852210:

  ARM: OMAP2+: hwmod: extend OCP_* register offsets from 16 to 32 bits (2012-02-28 03:49:34 -0700)

(This branch is based on the 'hwmod_code_cleanup_3.4' branch, pull req 
sent previously)


- Paul

----------------------------------------------------------------
Paul Walmsley (9):
      ARM: OMAP2/3: hwmod data: update old names
      ARM: OMAP3: hwmod data: fix interfaces for the MMC hwmods
      ARM: OMAP3: hwmod data: add DSS->L3 interconnect for 3430ES1
      ARM: OMAP3: hwmod data: GPTIMER12 is attached to a separate interconnect
      ARM: OMAP36xx: hwmod data: fix SmartReflex interface data
      ARM: OMAP4: hwmod data: remove bandgap hwmod
      ARM: OMAP4: hwmod data: add OCP_USER_DSP; mark omap44xx_dsp__iva appropriately
      ARM: OMAP4: hwmod data: uncomment some "excluded" hwmods
      ARM: OMAP2+: hwmod: extend OCP_* register offsets from 16 to 32 bits

 arch/arm/mach-omap2/omap_hwmod_2420_data.c   |    4 +-
 arch/arm/mach-omap2/omap_hwmod_2430_data.c   |    6 +-
 arch/arm/mach-omap2/omap_hwmod_3xxx_data.c   |  137 ++++++++++++++++++++------
 arch/arm/mach-omap2/omap_hwmod_44xx_data.c   |   39 +-------
 arch/arm/plat-omap/include/plat/omap_hwmod.h |    7 +-
 5 files changed, 121 insertions(+), 72 deletions(-)

[GIT PULL] ARM: OMAP2+: hwmod data: cleanup in preparation for link array removal

[Tony Lindgren]

Hi Paul,

* Paul Walmsley <paul@pwsan.com> [120307 19:34]:
> Hi Tony
> 
> The following changes since commit b8a633451951d149494076393698880453bf6bfe:
> 
>   ARM: OMAP: hwmod: add softreset delay field and OMAP4 data (2012-02-28 03:46:31 -0700)
> 
> are available in the git repository at:
> 
>   git://git.pwsan.com/linux-2.6 hwmod_data_cleanup_3.4
> 
> for you to fetch changes up to a2d4b88dd32b02e8b5088e0d64b3c102c0852210:
> 
>   ARM: OMAP2+: hwmod: extend OCP_* register offsets from 16 to 32 bits (2012-02-28 03:49:34 -0700)
> 
> (This branch is based on the 'hwmod_code_cleanup_3.4' branch, pull req 
> sent previously)

Got these now merged into l-o master, but ran into a merge
conflict with already queued sr branch. Resolved it by using
your version except left out vdd_name. As the namings have
changed for smartreflex, now I suspect it's broken for the
interrupts..

Care to take a look and see how we should deal with this?
My merge resolution below for reference.

Regards,

Tony


--- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
+++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c
@@@ -2684,9 -2736,10 +2736,9 @@@ static struct omap_hwmod_ocp_if *omap34
  };
  
  static struct omap_hwmod omap34xx_sr1_hwmod = {
- 	.name		= "sr1_hwmod",
+ 	.name		= "sr1",
  	.class		= &omap34xx_smartreflex_hwmod_class,
  	.main_clk	= "sr1_fck",
 -	.vdd_name	= "mpu_iva",
  	.prcm		= {
  		.omap2 = {
  			.prcm_reg_id = 1,
@@@ -2703,10 -2754,15 +2753,14 @@@
  	.flags		= HWMOD_SET_DEFAULT_CLOCKACT,
  };
  
+ static struct omap_hwmod_ocp_if *omap36xx_sr1_slaves[] = {
+ 	&omap36xx_l4_core__sr1,
+ };
+ 
  static struct omap_hwmod omap36xx_sr1_hwmod = {
- 	.name		= "sr1_hwmod",
+ 	.name		= "sr1",
  	.class		= &omap36xx_smartreflex_hwmod_class,
  	.main_clk	= "sr1_fck",
 -	.vdd_name	= "mpu_iva",
  	.prcm		= {
  		.omap2 = {
  			.prcm_reg_id = 1,
@@@ -2732,9 -2782,10 +2780,9 @@@ static struct omap_hwmod_ocp_if *omap34
  };
  
  static struct omap_hwmod omap34xx_sr2_hwmod = {
- 	.name		= "sr2_hwmod",
+ 	.name		= "sr2",
  	.class		= &omap34xx_smartreflex_hwmod_class,
  	.main_clk	= "sr2_fck",
 -	.vdd_name	= "core",
  	.prcm		= {
  		.omap2 = {
  			.prcm_reg_id = 1,
@@@ -2751,10 -2800,15 +2797,14 @@@
  	.flags		= HWMOD_SET_DEFAULT_CLOCKACT,
  };
  
+ static struct omap_hwmod_ocp_if *omap36xx_sr2_slaves[] = {
+ 	&omap36xx_l4_core__sr2,
+ };
+ 
  static struct omap_hwmod omap36xx_sr2_hwmod = {
- 	.name		= "sr2_hwmod",
+ 	.name		= "sr2",
  	.class		= &omap36xx_smartreflex_hwmod_class,
  	.main_clk	= "sr2_fck",
 -	.vdd_name	= "core",
  	.prcm		= {
  		.omap2 = {
  			.prcm_reg_id = 1,

[GIT PULL] ARM: OMAP2+: hwmod data: cleanup in preparation for link array removal

[Paul Walmsley]

Hi

On Thu, 8 Mar 2012, Tony Lindgren wrote:

> Got these now merged into l-o master, but ran into a merge
> conflict with already queued sr branch. Resolved it by using
> your version except left out vdd_name.

Definitely we want the vdd_name fields.  Other than that it looks okay.  

> As the namings have changed for smartreflex, now I suspect it's broken 
> for the interrupts..

Hmm, not sure I understand that part?  AFAIK the SmartReflex code doesn't 
rely on hwmod names.  Or are you talking about the vdd_name?


- Paul

[GIT PULL] ARM: OMAP2+: hwmod data: cleanup in preparation for link array removal

[Paul Walmsley]

Hi

on second thought...

On Thu, 8 Mar 2012, Paul Walmsley wrote:

> On Thu, 8 Mar 2012, Tony Lindgren wrote:
> 
> > Got these now merged into l-o master, but ran into a merge
> > conflict with already queued sr branch. Resolved it by using
> > your version except left out vdd_name.
> 
> Definitely we want the vdd_name fields.  Other than that it looks okay.  

I guess that other SR series that you merged removed the vdd_name fields?

If so then yeah, we'd better leave those out.

This means that I'm going to have to rebase those big hwmod data changes 
:-(

Can you send me a stable commit for that?


- Paul

[GIT PULL] ARM: OMAP2+: hwmod data: cleanup in preparation for link array removal

[Tony Lindgren]

* Paul Walmsley <paul@pwsan.com> [120308 18:07]:
> Hi
> 
> on second thought...
> 
> On Thu, 8 Mar 2012, Paul Walmsley wrote:
> 
> > On Thu, 8 Mar 2012, Tony Lindgren wrote:
> > 
> > > Got these now merged into l-o master, but ran into a merge
> > > conflict with already queued sr branch. Resolved it by using
> > > your version except left out vdd_name.
> > 
> > Definitely we want the vdd_name fields.  Other than that it looks okay.  
> 
> I guess that other SR series that you merged removed the vdd_name fields?

So it seems, commit cea6b942124683305280962969599fed36f0ea7e. Then
the interrupts we added there for smartreflex it seems.
 
> If so then yeah, we'd better leave those out.
> 
> This means that I'm going to have to rebase those big hwmod data changes 
> :-(
> 
> Can you send me a stable commit for that?

Bummer :( What we already have queued is the sr branch at commit
bb887688bccb48d94fa873f0f6cc7e7e3e4dc05e.

Regards,

Tony