From: Oliver Schinagl <[email protected]>
Having the interrupt tables documented in the kernel documentation is useful as
a quick reference.
Oliver Schinagl (1):
ARM: sunxi: doc: Add sun7i (A20) interrupt table
.../interrupt-controller/sunxi/sun7i-a20.txt | 102 +++++++++++++++++++++
1 file changed, 102 insertions(+)
create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
--
1.8.1.5
From: Oliver Schinagl <[email protected]>
This patch adds some documentation about the Allwinner sun7i (A20) using
the GIC.
Signed-off-by: Oliver Schinagl <[email protected]>
---
.../interrupt-controller/sunxi/sun7i-a20.txt | 102 +++++++++++++++++++++
1 file changed, 102 insertions(+)
create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
diff --git a/Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt b/Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
new file mode 100644
index 0000000..e5e59df
--- /dev/null
+++ b/Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
@@ -0,0 +1,102 @@
+Allwinner A20 (sun7i) interrupt sources
+---------------------------------------
+The sun7i interrupt controller is actually a standard arm GIC one. For more
+information see Documentation/devicetree/bindings/arm/gic.txt
+
+Interrupts are offset by 32, skipping the SGI and PPI's. The IRQ numbers here
+is what is expected in the drivers device tree table.
+
+The interrupt sources available for the Allwinner A20 SoC are the
+following ones:
+
+0: ENMI
+1: UART0
+2: UART1
+3: UART2
+4: UART3
+5: IR0
+6: IR1
+7: I2C0
+8: I2C1
+9: I2C2
+10: SPI0
+11: SPI1
+12: SPI2
+13: SPDIF
+14: AC97
+15: TS
+16: I2S0
+17: UART4
+18: UART5
+19: UART6
+20: UART7
+21: KEYPAD
+22: TIMER0
+23: TIMER1
+24: TIMER2
+25: TIMER3
+26: CAN
+27: DMA
+28: PIO
+29: TOUCH_PANEL
+30: AUDIO_CODEC
+31: LRADC
+32: MMC0
+33: MMC1
+34: MMC2
+35: MMC3
+36: MEMSTICK
+37: NAND
+38: USB0
+39: USB1
+40: USB2
+41: SCR
+42: CSI0
+43: CSI1
+44: LCDCTRL0
+45: LCDCTRL1
+46: MP
+47: DEFEBE0
+48: DEFEBE1
+49: PMU
+50: SPI3
+51: TZASC
+52: PATA
+53: VE
+54: SS
+55: EMAC
+56: SATA
+57: GPS
+58: HDMI
+59: TVE
+60: ACE
+61: TVD
+62: PS2_0
+63: PS2_1
+64: USB3
+65: USB4
+66: PLE_PFM
+67: TIMER4
+68: TIMER5
+69: GPU_GP
+70: GPU_GPMMU
+71: GPU_PP0
+72: GPU_PPMMU0
+73: GPU_PMU
+74: GPU_PP1
+75: GPU_PPMMU1
+76: GPU_RSV0
+77: GPU_RSV1
+78: GPU_RSV2
+79: GPU_RSV3
+80: GPU_RSV4
+81: HS_TIMER0
+82: HS_TIMER1
+83: HS_TIMER2
+84: HS_TIMER3
+85: GMAC
+86: HDMI1
+87: I2S1
+88: I2C3
+89: I2C4
+90: I2S2
--
1.8.1.5
On Fri, Sep 20, 2013 at 12:12:40PM +0100, Oliver Schinagl wrote:
> From: Oliver Schinagl <[email protected]>
>
> This patch adds some documentation about the Allwinner sun7i (A20) using
> the GIC.
>
> Signed-off-by: Oliver Schinagl <[email protected]>
NAK.
This information is not a binding, and it's not a property of the
interrupt controller. It's the integration of a full SoC, which should
be described full in a dtsi where it can actually be used.
The only other SoCs that have this are other sunxi variants, and I see
no reason to have them here either.
Thanks,
Mark.
> ---
> .../interrupt-controller/sunxi/sun7i-a20.txt | 102 +++++++++++++++++++++
> 1 file changed, 102 insertions(+)
> create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
>
> diff --git a/Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt b/Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
> new file mode 100644
> index 0000000..e5e59df
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/interrupt-controller/sunxi/sun7i-a20.txt
> @@ -0,0 +1,102 @@
> +Allwinner A20 (sun7i) interrupt sources
> +---------------------------------------
> +The sun7i interrupt controller is actually a standard arm GIC one. For more
> +information see Documentation/devicetree/bindings/arm/gic.txt
> +
> +Interrupts are offset by 32, skipping the SGI and PPI's. The IRQ numbers here
> +is what is expected in the drivers device tree table.
> +
> +The interrupt sources available for the Allwinner A20 SoC are the
> +following ones:
> +
> +0: ENMI
> +1: UART0
> +2: UART1
> +3: UART2
> +4: UART3
> +5: IR0
> +6: IR1
> +7: I2C0
> +8: I2C1
> +9: I2C2
> +10: SPI0
> +11: SPI1
> +12: SPI2
> +13: SPDIF
> +14: AC97
> +15: TS
> +16: I2S0
> +17: UART4
> +18: UART5
> +19: UART6
> +20: UART7
> +21: KEYPAD
> +22: TIMER0
> +23: TIMER1
> +24: TIMER2
> +25: TIMER3
> +26: CAN
> +27: DMA
> +28: PIO
> +29: TOUCH_PANEL
> +30: AUDIO_CODEC
> +31: LRADC
> +32: MMC0
> +33: MMC1
> +34: MMC2
> +35: MMC3
> +36: MEMSTICK
> +37: NAND
> +38: USB0
> +39: USB1
> +40: USB2
> +41: SCR
> +42: CSI0
> +43: CSI1
> +44: LCDCTRL0
> +45: LCDCTRL1
> +46: MP
> +47: DEFEBE0
> +48: DEFEBE1
> +49: PMU
> +50: SPI3
> +51: TZASC
> +52: PATA
> +53: VE
> +54: SS
> +55: EMAC
> +56: SATA
> +57: GPS
> +58: HDMI
> +59: TVE
> +60: ACE
> +61: TVD
> +62: PS2_0
> +63: PS2_1
> +64: USB3
> +65: USB4
> +66: PLE_PFM
> +67: TIMER4
> +68: TIMER5
> +69: GPU_GP
> +70: GPU_GPMMU
> +71: GPU_PP0
> +72: GPU_PPMMU0
> +73: GPU_PMU
> +74: GPU_PP1
> +75: GPU_PPMMU1
> +76: GPU_RSV0
> +77: GPU_RSV1
> +78: GPU_RSV2
> +79: GPU_RSV3
> +80: GPU_RSV4
> +81: HS_TIMER0
> +82: HS_TIMER1
> +83: HS_TIMER2
> +84: HS_TIMER3
> +85: GMAC
> +86: HDMI1
> +87: I2S1
> +88: I2C3
> +89: I2C4
> +90: I2S2
> --
> 1.8.1.5
>
>
Hi Mark,
On Fri, Sep 20, 2013 at 01:39:56PM +0100, Mark Rutland wrote:
> On Fri, Sep 20, 2013 at 12:12:40PM +0100, Oliver Schinagl wrote:
> > From: Oliver Schinagl <[email protected]>
> >
> > This patch adds some documentation about the Allwinner sun7i (A20) using
> > the GIC.
> >
> > Signed-off-by: Oliver Schinagl <[email protected]>
>
> NAK.
>
> This information is not a binding, and it's not a property of the
> interrupt controller. It's the integration of a full SoC, which should
> be described full in a dtsi where it can actually be used.
>
> The only other SoCs that have this are other sunxi variants, and I see
> no reason to have them here either.
This was mostly here because we had a very sparse documentation for
these SoCs when we first submitted the patches, and these informations
were not easy to find out. So a misplaced documentation is still better
than no documentation.
We do have the same documentation for the clocks here, mostly for the
same reason.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com