diff mbox series

[v3,3/7] ARM: dts: bcm2711: Enable PCIe controller

Message ID 20191126091946.7970-4-nsaenzjulienne@suse.de (mailing list archive)
State Superseded, archived
Headers show
Series Raspberry Pi 4 PCIe support | expand

Commit Message

Nicolas Saenz Julienne Nov. 26, 2019, 9:19 a.m. UTC
This enables bcm2711's PCIe bus, which is hardwired to a VIA
Technologies XHCI USB 3.0 controller.

Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>

---

This will likely need a rebase once the RPi GENET patches land.

Changes since v2:
  - Remove unused interrupt-map
  - correct dma-ranges to it's full size, non power of 2 bus DMA
    constraints now supported in linux-next[1]
  - add device_type
  - rename alias from pcie_0 to pcie0

Changes since v1:
  - remove linux,pci-domain

[1] https://lkml.org/lkml/2019/11/21/235

 arch/arm/boot/dts/bcm2711.dtsi | 41 ++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

Comments

Phil Elwell Nov. 26, 2019, 9:37 a.m. UTC | #1
Hi Nicolas,

On 26/11/2019 09:19, Nicolas Saenz Julienne wrote:
> This enables bcm2711's PCIe bus, which is hardwired to a VIA
> Technologies XHCI USB 3.0 controller.
> 
> Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
> 
> ---
> 
> This will likely need a rebase once the RPi GENET patches land.
> 
> Changes since v2:
>    - Remove unused interrupt-map
>    - correct dma-ranges to it's full size, non power of 2 bus DMA
>      constraints now supported in linux-next[1]
>    - add device_type
>    - rename alias from pcie_0 to pcie0
> 
> Changes since v1:
>    - remove linux,pci-domain
> 
> [1] https://lkml.org/lkml/2019/11/21/235
> 
>   arch/arm/boot/dts/bcm2711.dtsi | 41 ++++++++++++++++++++++++++++++++++
>   1 file changed, 41 insertions(+)
> 
> diff --git a/arch/arm/boot/dts/bcm2711.dtsi b/arch/arm/boot/dts/bcm2711.dtsi
> index 667658497898..2e121fc8b3d0 100644
> --- a/arch/arm/boot/dts/bcm2711.dtsi
> +++ b/arch/arm/boot/dts/bcm2711.dtsi
> @@ -288,6 +288,47 @@ IRQ_TYPE_LEVEL_LOW)>,
>   		arm,cpu-registers-not-fw-configured;
>   	};
>   
> +	scb {
> +		compatible = "simple-bus";
> +		#address-cells = <2>;
> +		#size-cells = <1>;
> +
> +		ranges = <0x0 0x7c000000  0x0 0xfc000000  0x03800000>,
> +			 <0x6 0x00000000  0x6 0x00000000  0x40000000>;
> +
> +		pcie0: pcie@7d500000 {
> +			compatible = "brcm,bcm2711-pcie";
> +			reg = <0x0 0x7d500000 0x9310>;
> +			device_type = "pci";
> +			#address-cells = <3>;
> +			#interrupt-cells = <1>;
> +			#size-cells = <2>;
> +			interrupts = <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>,
> +				     <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>;
> +			interrupt-names = "pcie", "msi";
> +			interrupt-map-mask = <0x0 0x0 0x0 0x7>;
> +			interrupt-map = <0 0 0 1 &gicv2 GIC_SPI 143
> +							IRQ_TYPE_LEVEL_HIGH>;
> +			msi-controller;
> +			msi-parent = <&pcie0>;
> +
> +			ranges = <0x02000000 0x0 0xf8000000 0x6 0x00000000
> +				  0x0 0x04000000>;
> +			/*
> +			 * The wrapper around the PCIe block has a bug
> +			 * preventing it from accessing beyond the first 3GB of
> +			 * memory. As the bus DMA mask is rounded up to the
> +			 * closest power of two of the dma-range size, we're
> +			 * forced to set the limit at 2GB. This can be
> +			 * harmlessly changed in the future once the DMA code
> +			 * handles non power of two DMA limits.
> +			 */
> +			dma-ranges = <0x02000000 0x0 0x00000000 0x0 0x00000000
> +				      0x0 0xc0000000>;

The comment doesn't match the data here - I think for now the size field 
needs to be reduced to 2GB to match the comment.

Phil

> +			brcm,enable-ssc;
> +		};
> +	};
> +
>   	cpus: cpus {
>   		#address-cells = <1>;
>   		#size-cells = <0>;
>
Nicolas Saenz Julienne Nov. 26, 2019, 9:43 a.m. UTC | #2
On Tue, 2019-11-26 at 09:37 +0000, Phil Elwell wrote:
> Hi Nicolas,
> 
> On 26/11/2019 09:19, Nicolas Saenz Julienne wrote:
> > This enables bcm2711's PCIe bus, which is hardwired to a VIA
> > Technologies XHCI USB 3.0 controller.
> > 
> > Signed-off-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
> > 
> > ---
> > 
> > This will likely need a rebase once the RPi GENET patches land.
> > 
> > Changes since v2:
> >    - Remove unused interrupt-map
> >    - correct dma-ranges to it's full size, non power of 2 bus DMA
> >      constraints now supported in linux-next[1]
> >    - add device_type
> >    - rename alias from pcie_0 to pcie0
> > 
> > Changes since v1:
> >    - remove linux,pci-domain
> > 
> > [1] https://lkml.org/lkml/2019/11/21/235
> > 
> >   arch/arm/boot/dts/bcm2711.dtsi | 41 ++++++++++++++++++++++++++++++++++
> >   1 file changed, 41 insertions(+)
> > 
> > diff --git a/arch/arm/boot/dts/bcm2711.dtsi b/arch/arm/boot/dts/bcm2711.dtsi
> > index 667658497898..2e121fc8b3d0 100644
> > --- a/arch/arm/boot/dts/bcm2711.dtsi
> > +++ b/arch/arm/boot/dts/bcm2711.dtsi
> > @@ -288,6 +288,47 @@ IRQ_TYPE_LEVEL_LOW)>,
> >   		arm,cpu-registers-not-fw-configured;
> >   	};
> >   
> > +	scb {
> > +		compatible = "simple-bus";
> > +		#address-cells = <2>;
> > +		#size-cells = <1>;
> > +
> > +		ranges = <0x0 0x7c000000  0x0 0xfc000000  0x03800000>,
> > +			 <0x6 0x00000000  0x6 0x00000000  0x40000000>;
> > +
> > +		pcie0: pcie@7d500000 {
> > +			compatible = "brcm,bcm2711-pcie";
> > +			reg = <0x0 0x7d500000 0x9310>;
> > +			device_type = "pci";
> > +			#address-cells = <3>;
> > +			#interrupt-cells = <1>;
> > +			#size-cells = <2>;
> > +			interrupts = <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>,
> > +				     <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>;
> > +			interrupt-names = "pcie", "msi";
> > +			interrupt-map-mask = <0x0 0x0 0x0 0x7>;
> > +			interrupt-map = <0 0 0 1 &gicv2 GIC_SPI 143
> > +							IRQ_TYPE_LEVEL_HIGH>;
> > +			msi-controller;
> > +			msi-parent = <&pcie0>;
> > +
> > +			ranges = <0x02000000 0x0 0xf8000000 0x6 0x00000000
> > +				  0x0 0x04000000>;
> > +			/*
> > +			 * The wrapper around the PCIe block has a bug
> > +			 * preventing it from accessing beyond the first 3GB of
> > +			 * memory. As the bus DMA mask is rounded up to the
> > +			 * closest power of two of the dma-range size, we're
> > +			 * forced to set the limit at 2GB. This can be
> > +			 * harmlessly changed in the future once the DMA code
> > +			 * handles non power of two DMA limits.
> > +			 */
> > +			dma-ranges = <0x02000000 0x0 0x00000000 0x0 0x00000000
> > +				      0x0 0xc0000000>;
> 
> The comment doesn't match the data here - I think for now the size field 
> needs to be reduced to 2GB to match the comment.

You're right, my bad, should've edited it out. The good part is that with this
commit[1], which will soon be in Linus' tree, we don't need to fake dma-ranges
size anymore.

So for the record, the comment should state the following:

	/*
	 * The wrapper around the PCIe block has a bug
	 * preventing it from accessing beyond the first 3GB of
	 * memory.
	 */

Regards,
Nicolas

[1] https://lkml.org/lkml/2019/11/21/235
diff mbox series

Patch

diff --git a/arch/arm/boot/dts/bcm2711.dtsi b/arch/arm/boot/dts/bcm2711.dtsi
index 667658497898..2e121fc8b3d0 100644
--- a/arch/arm/boot/dts/bcm2711.dtsi
+++ b/arch/arm/boot/dts/bcm2711.dtsi
@@ -288,6 +288,47 @@  IRQ_TYPE_LEVEL_LOW)>,
 		arm,cpu-registers-not-fw-configured;
 	};
 
+	scb {
+		compatible = "simple-bus";
+		#address-cells = <2>;
+		#size-cells = <1>;
+
+		ranges = <0x0 0x7c000000  0x0 0xfc000000  0x03800000>,
+			 <0x6 0x00000000  0x6 0x00000000  0x40000000>;
+
+		pcie0: pcie@7d500000 {
+			compatible = "brcm,bcm2711-pcie";
+			reg = <0x0 0x7d500000 0x9310>;
+			device_type = "pci";
+			#address-cells = <3>;
+			#interrupt-cells = <1>;
+			#size-cells = <2>;
+			interrupts = <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>,
+				     <GIC_SPI 148 IRQ_TYPE_LEVEL_HIGH>;
+			interrupt-names = "pcie", "msi";
+			interrupt-map-mask = <0x0 0x0 0x0 0x7>;
+			interrupt-map = <0 0 0 1 &gicv2 GIC_SPI 143
+							IRQ_TYPE_LEVEL_HIGH>;
+			msi-controller;
+			msi-parent = <&pcie0>;
+
+			ranges = <0x02000000 0x0 0xf8000000 0x6 0x00000000
+				  0x0 0x04000000>;
+			/*
+			 * The wrapper around the PCIe block has a bug
+			 * preventing it from accessing beyond the first 3GB of
+			 * memory. As the bus DMA mask is rounded up to the
+			 * closest power of two of the dma-range size, we're
+			 * forced to set the limit at 2GB. This can be
+			 * harmlessly changed in the future once the DMA code
+			 * handles non power of two DMA limits.
+			 */
+			dma-ranges = <0x02000000 0x0 0x00000000 0x0 0x00000000
+				      0x0 0xc0000000>;
+			brcm,enable-ssc;
+		};
+	};
+
 	cpus: cpus {
 		#address-cells = <1>;
 		#size-cells = <0>;