[PULL] clk: at91: sama5d2 materials

[PULL] clk: at91: sama5d2 materials

[Boris Brezillon]

Stephen, Mike,

Here is a PR for the sama5d2 materials.
I know we're already late in the release cycle and I'm not sure you can
take this PR into account for 4.3, so if you can't could you please
queue it for 4.4?

Thanks,

Boris

The following changes since commit c42743f11f58646a50bc4b96da363429dca0c5e6:

  clk: add bindings for the Ux500 clocks (2015-08-13 11:29:45 -0700)

are available in the git repository at:

  https://github.com/bbrezillon/linux-at91.git clk/atmel-clk-next

for you to fetch changes up to 3be0527c0f8bbbb744a6a34603d6a83e511c6741:

  clk: at91: add generated clock driver (2015-08-18 10:11:57 +0200)

----------------------------------------------------------------
Nicolas Ferre (4):
      clk: at91: cleanup PMC header file for PCR register fields
      clk: at91: modify PMC peripheral clock to deal with newer register layout
      clk: at91: add PMC sama5d2 support
      clk: at91: add generated clock driver

 Documentation/devicetree/bindings/clock/at91-clock.txt |  35 +++++++++++++++++++++
 arch/arm/mach-at91/Kconfig                             |   3 ++
 drivers/clk/at91/Makefile                              |   1 +
 drivers/clk/at91/clk-generated.c                       | 306 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 drivers/clk/at91/clk-peripheral.c                      |  24 +++++++++-----
 drivers/clk/at91/pmc.c                                 |  21 +++++++++++++
 drivers/clk/at91/pmc.h                                 |   3 ++
 include/dt-bindings/clock/at91.h                       |   1 +
 include/linux/clk/at91_pmc.h                           |  22 ++++++++-----
 9 files changed, 400 insertions(+), 16 deletions(-)
 create mode 100644 drivers/clk/at91/clk-generated.c

[PULL] clk: at91: sama5d2 materials

[Michael Turquette]

Quoting Boris Brezillon (2015-08-18 01:24:01)
> Stephen, Mike,
> 
> Here is a PR for the sama5d2 materials.
> I know we're already late in the release cycle and I'm not sure you can
> take this PR into account for 4.3, so if you can't could you please
> queue it for 4.4?

Queued for 4.4.

Thanks,
Mike

> 
> Thanks,
> 
> Boris
> 
> The following changes since commit c42743f11f58646a50bc4b96da363429dca0c5e6:
> 
>   clk: add bindings for the Ux500 clocks (2015-08-13 11:29:45 -0700)
> 
> are available in the git repository at:
> 
>   https://github.com/bbrezillon/linux-at91.git clk/atmel-clk-next
> 
> for you to fetch changes up to 3be0527c0f8bbbb744a6a34603d6a83e511c6741:
> 
>   clk: at91: add generated clock driver (2015-08-18 10:11:57 +0200)
> 
> ----------------------------------------------------------------
> Nicolas Ferre (4):
>       clk: at91: cleanup PMC header file for PCR register fields
>       clk: at91: modify PMC peripheral clock to deal with newer register layout
>       clk: at91: add PMC sama5d2 support
>       clk: at91: add generated clock driver
> 
>  Documentation/devicetree/bindings/clock/at91-clock.txt |  35 +++++++++++++++++++++
>  arch/arm/mach-at91/Kconfig                             |   3 ++
>  drivers/clk/at91/Makefile                              |   1 +
>  drivers/clk/at91/clk-generated.c                       | 306 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  drivers/clk/at91/clk-peripheral.c                      |  24 +++++++++-----
>  drivers/clk/at91/pmc.c                                 |  21 +++++++++++++
>  drivers/clk/at91/pmc.h                                 |   3 ++
>  include/dt-bindings/clock/at91.h                       |   1 +
>  include/linux/clk/at91_pmc.h                           |  22 ++++++++-----
>  9 files changed, 400 insertions(+), 16 deletions(-)
>  create mode 100644 drivers/clk/at91/clk-generated.c
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PULL] clk: at91: sama5d2 materials

[Stephen Boyd]

On 08/25, Michael Turquette wrote:
> Quoting Boris Brezillon (2015-08-18 01:24:01)
> > Stephen, Mike,
> > 
> > Here is a PR for the sama5d2 materials.
> > I know we're already late in the release cycle and I'm not sure you can
> > take this PR into account for 4.3, so if you can't could you please
> > queue it for 4.4?
> 
> Queued for 4.4.
> 
> Thanks,
> Mike
> 
> > 
> > Thanks,
> > 
> > Boris
> > 
> > The following changes since commit c42743f11f58646a50bc4b96da363429dca0c5e6:
> > 

This commit was in clk-next and then got rebased. clk-next isn't
stable, please don't base pull requests on it. If you have some
dependency on the patches in there then we need to know so that
we can provide some branch.

I cherry-picked the 4 patches to clk-next. I had to squash in this fix
to get it to compile. Please check.

---8<----
diff --git a/drivers/clk/at91/clk-generated.c b/drivers/clk/at91/clk-generated.c
index 631123ca6f85..abc80949e1dd 100644
--- a/drivers/clk/at91/clk-generated.c
+++ b/drivers/clk/at91/clk-generated.c
@@ -99,22 +99,22 @@ static int clk_generated_determine_rate(struct clk_hw *hw,
 					struct clk_rate_request *req)
 {
 	struct clk_generated *gck = to_clk_generated(hw);
-	struct clk *parent = NULL;
+	struct clk_hw *parent = NULL;
 	long best_rate = -EINVAL;
 	unsigned long tmp_rate, min_rate;
 	int best_diff = -1;
 	int tmp_diff;
 	int i;
 
-	for (i = 0; i < __clk_get_num_parents(hw->clk); i++) {
+	for (i = 0; i < clk_hw_get_num_parents(hw); i++) {
 		u32 div;
 		unsigned long parent_rate;
 
-		parent = clk_get_parent_by_index(hw->clk, i);
+		parent = clk_hw_get_parent_by_index(hw, i);
 		if (!parent)
 			continue;
 
-		parent_rate = __clk_get_rate(parent);
+		parent_rate = clk_hw_get_rate(parent);
 		min_rate = DIV_ROUND_CLOSEST(parent_rate, GENERATED_MAX_DIV + 1);
 		if (!parent_rate ||
 		    (gck->range.max && min_rate > gck->range.max))
@@ -128,7 +128,7 @@ static int clk_generated_determine_rate(struct clk_hw *hw,
 				best_rate = tmp_rate;
 				best_diff = tmp_diff;
 				req->best_parent_rate = parent_rate;
-				req->best_parent_hw = __clk_get_hw(parent);
+				req->best_parent_hw = parent;
 			}
 
 			if (!best_diff || tmp_rate < req->rate)
@@ -156,7 +156,7 @@ static int clk_generated_set_parent(struct clk_hw *hw, u8 index)
 {
 	struct clk_generated *gck = to_clk_generated(hw);
 
-	if (index >= __clk_get_num_parents(hw->clk))
+	if (index >= clk_hw_get_num_parents(hw))
 		return -EINVAL;
 
 	gck->parent_id = index;

-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum,
a Linux Foundation Collaborative Project

[PULL] clk: at91: sama5d2 materials

[Boris Brezillon]

Hi Stephen,

On Thu, 1 Oct 2015 12:47:48 -0700
Stephen Boyd <sboyd@codeaurora.org> wrote:

> On 08/25, Michael Turquette wrote:
> > Quoting Boris Brezillon (2015-08-18 01:24:01)
> > > Stephen, Mike,
> > > 
> > > Here is a PR for the sama5d2 materials.
> > > I know we're already late in the release cycle and I'm not sure you can
> > > take this PR into account for 4.3, so if you can't could you please
> > > queue it for 4.4?
> > 
> > Queued for 4.4.
> > 
> > Thanks,
> > Mike
> > 
> > > 
> > > Thanks,
> > > 
> > > Boris
> > > 
> > > The following changes since commit c42743f11f58646a50bc4b96da363429dca0c5e6:
> > > 
> 
> This commit was in clk-next and then got rebased. clk-next isn't
> stable, please don't base pull requests on it. If you have some
> dependency on the patches in there then we need to know so that
> we can provide some branch.

Yes, sorry about that, I'll base my future PR on linus' master branch.

> 
> I cherry-picked the 4 patches to clk-next. I had to squash in this fix
> to get it to compile. Please check.

Thanks, looks good to me.
Nicolas, can you test it on a real platform?

> 
> ---8<----
> diff --git a/drivers/clk/at91/clk-generated.c b/drivers/clk/at91/clk-generated.c
> index 631123ca6f85..abc80949e1dd 100644
> --- a/drivers/clk/at91/clk-generated.c
> +++ b/drivers/clk/at91/clk-generated.c
> @@ -99,22 +99,22 @@ static int clk_generated_determine_rate(struct clk_hw *hw,
>  					struct clk_rate_request *req)
>  {
>  	struct clk_generated *gck = to_clk_generated(hw);
> -	struct clk *parent = NULL;
> +	struct clk_hw *parent = NULL;
>  	long best_rate = -EINVAL;
>  	unsigned long tmp_rate, min_rate;
>  	int best_diff = -1;
>  	int tmp_diff;
>  	int i;
>  
> -	for (i = 0; i < __clk_get_num_parents(hw->clk); i++) {
> +	for (i = 0; i < clk_hw_get_num_parents(hw); i++) {
>  		u32 div;
>  		unsigned long parent_rate;
>  
> -		parent = clk_get_parent_by_index(hw->clk, i);
> +		parent = clk_hw_get_parent_by_index(hw, i);
>  		if (!parent)
>  			continue;
>  
> -		parent_rate = __clk_get_rate(parent);
> +		parent_rate = clk_hw_get_rate(parent);
>  		min_rate = DIV_ROUND_CLOSEST(parent_rate, GENERATED_MAX_DIV + 1);
>  		if (!parent_rate ||
>  		    (gck->range.max && min_rate > gck->range.max))
> @@ -128,7 +128,7 @@ static int clk_generated_determine_rate(struct clk_hw *hw,
>  				best_rate = tmp_rate;
>  				best_diff = tmp_diff;
>  				req->best_parent_rate = parent_rate;
> -				req->best_parent_hw = __clk_get_hw(parent);
> +				req->best_parent_hw = parent;
>  			}
>  
>  			if (!best_diff || tmp_rate < req->rate)
> @@ -156,7 +156,7 @@ static int clk_generated_set_parent(struct clk_hw *hw, u8 index)
>  {
>  	struct clk_generated *gck = to_clk_generated(hw);
>  
> -	if (index >= __clk_get_num_parents(hw->clk))
> +	if (index >= clk_hw_get_num_parents(hw))
>  		return -EINVAL;
>  
>  	gck->parent_id = index;
> 



-- 
Boris Brezillon, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com

[PULL] clk: at91: sama5d2 materials

[Nicolas Ferre]

Le 01/10/2015 22:38, Boris Brezillon a ?crit :
> Hi Stephen,
> 
> On Thu, 1 Oct 2015 12:47:48 -0700
> Stephen Boyd <sboyd@codeaurora.org> wrote:
> 
>> On 08/25, Michael Turquette wrote:
>>> Quoting Boris Brezillon (2015-08-18 01:24:01)
>>>> Stephen, Mike,
>>>>
>>>> Here is a PR for the sama5d2 materials.
>>>> I know we're already late in the release cycle and I'm not sure you can
>>>> take this PR into account for 4.3, so if you can't could you please
>>>> queue it for 4.4?
>>>
>>> Queued for 4.4.
>>>
>>> Thanks,
>>> Mike
>>>
>>>>
>>>> Thanks,
>>>>
>>>> Boris
>>>>
>>>> The following changes since commit c42743f11f58646a50bc4b96da363429dca0c5e6:
>>>>
>>
>> This commit was in clk-next and then got rebased. clk-next isn't
>> stable, please don't base pull requests on it. If you have some
>> dependency on the patches in there then we need to know so that
>> we can provide some branch.
> 
> Yes, sorry about that, I'll base my future PR on linus' master branch.
> 
>>
>> I cherry-picked the 4 patches to clk-next. I had to squash in this fix
>> to get it to compile. Please check.
> 
> Thanks, looks good to me.
> Nicolas, can you test it on a real platform?

Yep, tested and works okay. So for your changes (and for the record
actually):
Tested-by: Nicolas Ferre <nicolas.ferre@atmel.com>


>> ---8<----
>> diff --git a/drivers/clk/at91/clk-generated.c b/drivers/clk/at91/clk-generated.c
>> index 631123ca6f85..abc80949e1dd 100644
>> --- a/drivers/clk/at91/clk-generated.c
>> +++ b/drivers/clk/at91/clk-generated.c
>> @@ -99,22 +99,22 @@ static int clk_generated_determine_rate(struct clk_hw *hw,
>>  					struct clk_rate_request *req)
>>  {
>>  	struct clk_generated *gck = to_clk_generated(hw);
>> -	struct clk *parent = NULL;
>> +	struct clk_hw *parent = NULL;
>>  	long best_rate = -EINVAL;
>>  	unsigned long tmp_rate, min_rate;
>>  	int best_diff = -1;
>>  	int tmp_diff;
>>  	int i;
>>  
>> -	for (i = 0; i < __clk_get_num_parents(hw->clk); i++) {
>> +	for (i = 0; i < clk_hw_get_num_parents(hw); i++) {
>>  		u32 div;
>>  		unsigned long parent_rate;
>>  
>> -		parent = clk_get_parent_by_index(hw->clk, i);
>> +		parent = clk_hw_get_parent_by_index(hw, i);
>>  		if (!parent)
>>  			continue;
>>  
>> -		parent_rate = __clk_get_rate(parent);
>> +		parent_rate = clk_hw_get_rate(parent);
>>  		min_rate = DIV_ROUND_CLOSEST(parent_rate, GENERATED_MAX_DIV + 1);
>>  		if (!parent_rate ||
>>  		    (gck->range.max && min_rate > gck->range.max))
>> @@ -128,7 +128,7 @@ static int clk_generated_determine_rate(struct clk_hw *hw,
>>  				best_rate = tmp_rate;
>>  				best_diff = tmp_diff;
>>  				req->best_parent_rate = parent_rate;
>> -				req->best_parent_hw = __clk_get_hw(parent);
>> +				req->best_parent_hw = parent;
>>  			}
>>  
>>  			if (!best_diff || tmp_rate < req->rate)
>> @@ -156,7 +156,7 @@ static int clk_generated_set_parent(struct clk_hw *hw, u8 index)
>>  {
>>  	struct clk_generated *gck = to_clk_generated(hw);
>>  
>> -	if (index >= __clk_get_num_parents(hw->clk))
>> +	if (index >= clk_hw_get_num_parents(hw))
>>  		return -EINVAL;
>>  
>>  	gck->parent_id = index;
>>
> 
> 
> 


-- 
Nicolas Ferre