[PATCH] ARM: dts: fix PWM clock found on RK3288 Socs

[PATCH] ARM: dts: fix PWM clock found on RK3288 Socs

[Caesar Wang]

We use the new PWM IP on RK3288,but the PWM's clock indeed incorrect.

Signed-off-by: Caesar Wang <caesar.wang@rock-chips.com>
---
 arch/arm/boot/dts/rk3288.dtsi | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/arm/boot/dts/rk3288.dtsi b/arch/arm/boot/dts/rk3288.dtsi
index cb18bb4..76bff02 100644
--- a/arch/arm/boot/dts/rk3288.dtsi
+++ b/arch/arm/boot/dts/rk3288.dtsi
@@ -418,7 +418,7 @@
 		#pwm-cells = <3>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&pwm0_pin>;
-		clocks = <&cru PCLK_PWM>;
+		clocks = <&cru PCLK_RKPWM>;
 		clock-names = "pwm";
 		status = "disabled";
 	};
@@ -429,7 +429,7 @@
 		#pwm-cells = <3>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&pwm1_pin>;
-		clocks = <&cru PCLK_PWM>;
+		clocks = <&cru PCLK_RKPWM>;
 		clock-names = "pwm";
 		status = "disabled";
 	};
@@ -440,7 +440,7 @@
 		#pwm-cells = <3>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&pwm2_pin>;
-		clocks = <&cru PCLK_PWM>;
+		clocks = <&cru PCLK_RKPWM>;
 		clock-names = "pwm";
 		status = "disabled";
 	};
@@ -451,7 +451,7 @@
 		#pwm-cells = <2>;
 		pinctrl-names = "default";
 		pinctrl-0 = <&pwm3_pin>;
-		clocks = <&cru PCLK_PWM>;
+		clocks = <&cru PCLK_RKPWM>;
 		clock-names = "pwm";
 		status = "disabled";
 	};
-- 
1.9.1

Re: [PATCH] ARM: dts: fix PWM clock found on RK3288 Socs

[Doug Anderson]

Caesar,

On Tue, Nov 18, 2014 at 7:25 PM, Caesar Wang <caesar.wang@rock-chips.com> wrote:
> We use the new PWM IP on RK3288,but the PWM's clock indeed incorrect.
>
> Signed-off-by: Caesar Wang <caesar.wang@rock-chips.com>
> ---
>  arch/arm/boot/dts/rk3288.dtsi | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)

Right, I reported this when Kever was doing his "unused clocks" patch.
That's why this clock is currently "ignore unused".

...but though your change is right, I'd rather your change doesn't
land quite yet.  It totally wrecks havoc with the PWM regulator used
on many rk3288 boards and will glitch the main "logic" voltage at
bootup.  I have some patches that are MASSIVELY WIP:

https://chromium-review.googlesource.com/230640
WIP: regulator: pwm: Tell regulator framework that we can't turn off

https://chromium-review.googlesource.com/230641
WIP: PWM: handle the fact that firmware might have left us enabled

https://chromium-review.googlesource.com/230642
WIP: pwm-regulator stuff

I think that last one might blow up into several patches.  We need to
solve the problems from those patches before yours can really land.


It's pretty high on my list to finish the above patches, but right now
suspend/resume issues are higher priority so it's a bit stalled...  If
someone wants to hijack any of the above patches please let me know--I
won't be offended.

-Doug

Re: [PATCH] ARM: dts: fix PWM clock found on RK3288 Socs

[Caesar Wang]

Doug,

在 2014/11/19 13:14, Doug Anderson 写道:
> Caesar,
>
> On Tue, Nov 18, 2014 at 7:25 PM, Caesar Wang <caesar.wang@rock-chips.com> wrote:
>> We use the new PWM IP on RK3288,but the PWM's clock indeed incorrect.
>>
>> Signed-off-by: Caesar Wang <caesar.wang@rock-chips.com>
>> ---
>>   arch/arm/boot/dts/rk3288.dtsi | 8 ++++----
>>   1 file changed, 4 insertions(+), 4 deletions(-)
> Right, I reported this when Kever was doing his "unused clocks" patch.
> That's why this clock is currently "ignore unused".
>
> ...but though your change is right, I'd rather your change doesn't
> land quite yet.  It totally wrecks havoc with the PWM regulator used
> on many rk3288 boards and will glitch the main "logic" voltage at
> bootup.  I have some patches that are MASSIVELY WIP:
>
> https://chromium-review.googlesource.com/230640
> WIP: regulator: pwm: Tell regulator framework that we can't turn off
>
> https://chromium-review.googlesource.com/230641
> WIP: PWM: handle the fact that firmware might have left us enabled
>
> https://chromium-review.googlesource.com/230642
> WIP: pwm-regulator stuff
>
> I think that last one might blow up into several patches.  We need to
> solve the problems from those patches before yours can really land.
>
>
> It's pretty high on my list to finish the above patches, but right now
> suspend/resume issues are higher priority so it's a bit stalled...  If
> someone wants to hijack any of the above patches please let me know--I
> won't be offended.

OK,I see.


> -Doug
>
>
>

-- 
Best regards,
Caesar