[PATCH 0/4] ARM: sun8i: Fix A23 and A33 clock gates indices

[PATCH 0/4] ARM: sun8i: Fix A23 and A33 clock gates indices

[Maxime Ripard]

Hi everyone,

Here is a patch set that adds the missing clocks for the message box
in the A33 and A23 SoCs.

In order to support that properly, the addition of a new clock driver
for the A33 has been needed, and we split the gates definition that
was previously shared to each DTSI.

Let me know what you think,
Maxime

Maxime Ripard (4):
  clk: sunxi: Add A33 gates support
  ARM: sun8i: Add the A33 AHB1 gates clock driver
  ARM: sun8i: Move A23 AHB1 gates out of common DTSI
  ARM: sun8i: A23: Add missing msgbox gate

 arch/arm/boot/dts/sun8i-a23-a33.dtsi | 25 -------------------------
 arch/arm/boot/dts/sun8i-a23.dtsi     | 25 +++++++++++++++++++++++++
 arch/arm/boot/dts/sun8i-a33.dtsi     | 27 +++++++++++++++++++++++++++
 drivers/clk/sunxi/clk-simple-gates.c |  2 ++
 4 files changed, 54 insertions(+), 25 deletions(-)

-- 
2.5.1

[PATCH 1/4] clk: sunxi: Add A33 gates support

[Maxime Ripard]

The A33 gates are different from the A23 ones, add a new hook to simple
gates to handle this clock

Reported-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
 drivers/clk/sunxi/clk-simple-gates.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/clk/sunxi/clk-simple-gates.c b/drivers/clk/sunxi/clk-simple-gates.c
index 6ce91180da1b..0214c6548afd 100644
--- a/drivers/clk/sunxi/clk-simple-gates.c
+++ b/drivers/clk/sunxi/clk-simple-gates.c
@@ -128,6 +128,8 @@ CLK_OF_DECLARE(sun8i_a23_apb1, "allwinner,sun8i-a23-apb1-gates-clk",
 	       sunxi_simple_gates_init);
 CLK_OF_DECLARE(sun8i_a23_apb2, "allwinner,sun8i-a23-apb2-gates-clk",
 	       sunxi_simple_gates_init);
+CLK_OF_DECLARE(sun8i_a33_ahb1, "allwinner,sun8i-a33-ahb1-gates-clk",
+	       sunxi_simple_gates_init);
 CLK_OF_DECLARE(sun9i_a80_ahb0, "allwinner,sun9i-a80-ahb0-gates-clk",
 	       sunxi_simple_gates_init);
 CLK_OF_DECLARE(sun9i_a80_ahb1, "allwinner,sun9i-a80-ahb1-gates-clk",
-- 
2.5.1

[PATCH 2/4] ARM: sun8i: Add the A33 AHB1 gates clock driver

[Maxime Ripard]

The A33 has different a different gates array than the A23, add the node to
the DT.

Reported-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
 arch/arm/boot/dts/sun8i-a33.dtsi | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/arch/arm/boot/dts/sun8i-a33.dtsi b/arch/arm/boot/dts/sun8i-a33.dtsi
index faa7d3c1fcea..3457edb3bf50 100644
--- a/arch/arm/boot/dts/sun8i-a33.dtsi
+++ b/arch/arm/boot/dts/sun8i-a33.dtsi
@@ -72,6 +72,33 @@
 			clock-output-names = "pll11";
 		};
 
+		ahb1_gates: clk@01c20060 {
+			#clock-cells = <1>;
+			compatible = "allwinner,sun8i-a33-ahb1-gates-clk";
+			reg = <0x01c20060 0x8>;
+			clocks = <&ahb1>;
+			clock-indices = <1>, <5>,
+				        <6>, <8>, <9>,
+				        <10>, <13>, <14>,
+					<19>, <20>,
+					<21>, <24>, <26>,
+					<29>, <32>, <36>,
+					<40>, <44>, <46>,
+					<52>, <53>,
+					<54>, <57>,
+					<58>;
+			clock-output-names = "ahb1_mipidsi", "ahb1_ss",
+					"ahb1_dma","ahb1_mmc0", "ahb1_mmc1",
+					"ahb1_mmc2", "ahb1_nand", "ahb1_sdram",
+					"ahb1_hstimer", "ahb1_spi0",
+					"ahb1_spi1", "ahb1_otg", "ahb1_ehci",
+					"ahb1_ohci", "ahb1_ve", "ahb1_lcd",
+					"ahb1_csi", "ahb1_be",	"ahb1_fe",
+					"ahb1_gpu", "ahb1_msgbox",
+					"ahb1_spinlock", "ahb1_drc",
+					"ahb1_sat";
+		};
+
 		mbus_clk: clk@01c2015c {
 			#clock-cells = <0>;
 			compatible = "allwinner,sun8i-a23-mbus-clk";
-- 
2.5.1

Re: [PATCH 2/4] ARM: sun8i: Add the A33 AHB1 gates clock driver

[Peter Korsgaard]

>>>>> "Maxime" == Maxime Ripard <maxime.ripard@free-electrons.com> writes:

 > The A33 has different a different gates array than the A23, add the node to
 > the DT.

NIT: 2x different.

Shouldn't the a23 one then move from sun8i-a23-a33.dtsi to
sun8i-a23.dtsi?

-- 
Bye, Peter Korsgaard

Re: [PATCH 2/4] ARM: sun8i: Add the A33 AHB1 gates clock driver

[Peter Korsgaard]

>>>>> "Peter" == Peter Korsgaard <peter@korsgaard.com> writes:

>>>>> "Maxime" == Maxime Ripard <maxime.ripard@free-electrons.com> writes:
 >> The A33 has different a different gates array than the A23, add the node to
 >> the DT.

 > NIT: 2x different.

 > Shouldn't the a23 one then move from sun8i-a23-a33.dtsi to
 > sun8i-a23.dtsi?

Ahh, I didn't notice patch 3.

-- 
Bye, Peter Korsgaard

Re: [PATCH 2/4] ARM: sun8i: Add the A33 AHB1 gates clock driver

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 697 bytes --]

Hi Peter,

On Sun, Sep 13, 2015 at 05:13:12PM +0200, Peter Korsgaard wrote:
> >>>>> "Maxime" == Maxime Ripard <maxime.ripard@free-electrons.com> writes:
> 
>  > The A33 has different a different gates array than the A23, add the node to
>  > the DT.
> 
> NIT: 2x different.

Ah, good catch, thanks!

> Shouldn't the a23 one then move from sun8i-a23-a33.dtsi to
> sun8i-a23.dtsi?

It is, but in order to preserve bisectability, I declared the A33 node
first, that has precedence over the A23 still in the common DTSI, and
removed the A23 in the next patch.

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

[PATCH 3/4] ARM: sun8i: Move A23 AHB1 gates out of common DTSI

[Maxime Ripard]

The AHB1 gates were assumed to be identical between the A23 and the A33,
which turned out to be wrong. Move the A23 gates definition to the A23
DTSI.

Reported-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
 arch/arm/boot/dts/sun8i-a23-a33.dtsi | 25 -------------------------
 arch/arm/boot/dts/sun8i-a23.dtsi     | 25 +++++++++++++++++++++++++
 2 files changed, 25 insertions(+), 25 deletions(-)

diff --git a/arch/arm/boot/dts/sun8i-a23-a33.dtsi b/arch/arm/boot/dts/sun8i-a23-a33.dtsi
index 27a925ec17d2..a34bd5f49c2a 100644
--- a/arch/arm/boot/dts/sun8i-a23-a33.dtsi
+++ b/arch/arm/boot/dts/sun8i-a23-a33.dtsi
@@ -175,31 +175,6 @@
 			clock-output-names = "apb1";
 		};
 
-		ahb1_gates: clk@01c20060 {
-			#clock-cells = <1>;
-			compatible = "allwinner,sun8i-a23-ahb1-gates-clk";
-			reg = <0x01c20060 0x8>;
-			clocks = <&ahb1>;
-			clock-indices = <1>, <6>,
-					<8>, <9>, <10>,
-					<13>, <14>,
-					<19>, <20>,
-					<21>, <24>, <26>,
-					<29>, <32>, <36>,
-					<40>, <44>, <46>,
-					<52>, <54>,
-					<57>;
-			clock-output-names = "ahb1_mipidsi", "ahb1_dma",
-					"ahb1_mmc0", "ahb1_mmc1", "ahb1_mmc2",
-					"ahb1_nand", "ahb1_sdram",
-					"ahb1_hstimer", "ahb1_spi0",
-					"ahb1_spi1", "ahb1_otg", "ahb1_ehci",
-					"ahb1_ohci", "ahb1_ve", "ahb1_lcd",
-					"ahb1_csi", "ahb1_be",	"ahb1_fe",
-					"ahb1_gpu", "ahb1_spinlock",
-					"ahb1_drc";
-		};
-
 		apb1_gates: clk@01c20068 {
 			#clock-cells = <1>;
 			compatible = "allwinner,sun8i-a23-apb1-gates-clk";
diff --git a/arch/arm/boot/dts/sun8i-a23.dtsi b/arch/arm/boot/dts/sun8i-a23.dtsi
index 2cc27c7a59dc..4a85e213843b 100644
--- a/arch/arm/boot/dts/sun8i-a23.dtsi
+++ b/arch/arm/boot/dts/sun8i-a23.dtsi
@@ -50,6 +50,31 @@
 	};
 
 	clocks {
+		ahb1_gates: clk@01c20060 {
+			#clock-cells = <1>;
+			compatible = "allwinner,sun8i-a23-ahb1-gates-clk";
+			reg = <0x01c20060 0x8>;
+			clocks = <&ahb1>;
+			clock-indices = <1>, <6>,
+					<8>, <9>, <10>,
+					<13>, <14>,
+					<19>, <20>,
+					<21>, <24>, <26>,
+					<29>, <32>, <36>,
+					<40>, <44>, <46>,
+					<52>, <54>,
+					<57>;
+			clock-output-names = "ahb1_mipidsi", "ahb1_dma",
+					"ahb1_mmc0", "ahb1_mmc1", "ahb1_mmc2",
+					"ahb1_nand", "ahb1_sdram",
+					"ahb1_hstimer", "ahb1_spi0",
+					"ahb1_spi1", "ahb1_otg", "ahb1_ehci",
+					"ahb1_ohci", "ahb1_ve", "ahb1_lcd",
+					"ahb1_csi", "ahb1_be",	"ahb1_fe",
+					"ahb1_gpu", "ahb1_spinlock",
+					"ahb1_drc";
+		};
+
 		mbus_clk: clk@01c2015c {
 			#clock-cells = <0>;
 			compatible = "allwinner,sun8i-a23-mbus-clk";
-- 
2.5.1

[PATCH 4/4] ARM: sun8i: A23: Add missing msgbox gate

[Maxime Ripard]

Even though it's not mentionned in the A23 user manual, the A23 has a gate
for the AHB1 clock to the msgbox IP. Add it to the clock-indices.

Reported-by: Chen-Yu Tsai <wens@csie.org>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
 arch/arm/boot/dts/sun8i-a23.dtsi | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/arm/boot/dts/sun8i-a23.dtsi b/arch/arm/boot/dts/sun8i-a23.dtsi
index 4a85e213843b..92e6616979ea 100644
--- a/arch/arm/boot/dts/sun8i-a23.dtsi
+++ b/arch/arm/boot/dts/sun8i-a23.dtsi
@@ -62,8 +62,8 @@
 					<21>, <24>, <26>,
 					<29>, <32>, <36>,
 					<40>, <44>, <46>,
-					<52>, <54>,
-					<57>;
+					<52>, <53>,
+					<54>, <57>;
 			clock-output-names = "ahb1_mipidsi", "ahb1_dma",
 					"ahb1_mmc0", "ahb1_mmc1", "ahb1_mmc2",
 					"ahb1_nand", "ahb1_sdram",
@@ -71,8 +71,8 @@
 					"ahb1_spi1", "ahb1_otg", "ahb1_ehci",
 					"ahb1_ohci", "ahb1_ve", "ahb1_lcd",
 					"ahb1_csi", "ahb1_be",	"ahb1_fe",
-					"ahb1_gpu", "ahb1_spinlock",
-					"ahb1_drc";
+					"ahb1_gpu", "ahb1_msgbox",
+					"ahb1_spinlock", "ahb1_drc";
 		};
 
 		mbus_clk: clk@01c2015c {
-- 
2.5.1

Re: [PATCH 0/4] ARM: sun8i: Fix A23 and A33 clock gates indices

[Chen-Yu Tsai]

On Fri, Sep 11, 2015 at 9:26 PM, Maxime Ripard
<maxime.ripard@free-electrons.com> wrote:
> Hi everyone,
>
> Here is a patch set that adds the missing clocks for the message box
> in the A33 and A23 SoCs.
>
> In order to support that properly, the addition of a new clock driver
> for the A33 has been needed, and we split the gates definition that
> was previously shared to each DTSI.
>
> Let me know what you think,
> Maxime
>
> Maxime Ripard (4):
>   clk: sunxi: Add A33 gates support
>   ARM: sun8i: Add the A33 AHB1 gates clock driver
>   ARM: sun8i: Move A23 AHB1 gates out of common DTSI
>   ARM: sun8i: A23: Add missing msgbox gate
>
>  arch/arm/boot/dts/sun8i-a23-a33.dtsi | 25 -------------------------
>  arch/arm/boot/dts/sun8i-a23.dtsi     | 25 +++++++++++++++++++++++++
>  arch/arm/boot/dts/sun8i-a33.dtsi     | 27 +++++++++++++++++++++++++++
>  drivers/clk/sunxi/clk-simple-gates.c |  2 ++
>  4 files changed, 54 insertions(+), 25 deletions(-)

Apart from the small issue with the second patch's commit message,
this series looks good.

Reviewed-by: Chen-Yu Tsai <wens@csie.org>

Re: [PATCH 0/4] ARM: sun8i: Fix A23 and A33 clock gates indices

[Maxime Ripard]

[-- Attachment #1: Type: text/plain, Size: 570 bytes --]

On Fri, Sep 11, 2015 at 03:26:33PM +0200, Maxime Ripard wrote:
> Hi everyone,
> 
> Here is a patch set that adds the missing clocks for the message box
> in the A33 and A23 SoCs.
> 
> In order to support that properly, the addition of a new clock driver
> for the A33 has been needed, and we split the gates definition that
> was previously shared to each DTSI.
> 
> Let me know what you think,
> Maxime

Fixed the typo and applied.

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]