* [PATCH 0/2] DT: correctly describe GIC400 virtualisation regions @ 2013-11-01 10:12 Ian Campbell 2013-11-01 10:13 ` [PATCH 1/2] arm: correct GIC cpu interface range for Cortex A15/GIC 400 Ian Campbell 2013-11-01 10:13 ` [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers Ian Campbell 0 siblings, 2 replies; 7+ messages in thread From: Ian Campbell @ 2013-11-01 10:12 UTC (permalink / raw) To: devicetree-u79uwXL29TY76Z2rM5mHXA Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r Several existing platforms which sport an A15 compatible GIC to not include the extended GIC cpu interface regisiters in their device tree. Fix these. Also extend the bindings to cope with cpu intefface registers which are not in contiguous pages. Ian. -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH 1/2] arm: correct GIC cpu interface range for Cortex A15/GIC 400 2013-11-01 10:12 [PATCH 0/2] DT: correctly describe GIC400 virtualisation regions Ian Campbell @ 2013-11-01 10:13 ` Ian Campbell [not found] ` <1383300809-7527-1-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> 2013-11-01 10:13 ` [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers Ian Campbell 1 sibling, 1 reply; 7+ messages in thread From: Ian Campbell @ 2013-11-01 10:13 UTC (permalink / raw) To: devicetree; +Cc: Ian Campbell, linux-arm-kernel GICv2 (Cortex A15/GIC 400) have CPU interface registers up to offset 0x1004 (the 32-bit GICC_DIR register a 0x1000). The GIC 400 documentation specifies the CPU interface region as being 0x2000 in size. Update all DTS entries claiming "arm,cortex-a15-gic" compatibility. Of these I only have personal experience with the vexpress a Calxeda (ecx) platforms and annecdotal evidence for omap and exynos. For the others I'm just assuming. Also update the example a15 binding in the documentation. The "Texas Instruments Keystone 2 SoC" platform which claim to have an a15 compatible GIC but does not include the GICH/GICV registers. I've not touched that here. Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Cc: devicetree@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org --- Documentation/devicetree/bindings/arm/gic.txt | 2 +- arch/arm/boot/dts/ecx-2000.dts | 2 +- arch/arm/boot/dts/exynos5.dtsi | 2 +- arch/arm/boot/dts/exynos5440.dtsi | 2 +- arch/arm/boot/dts/omap5.dtsi | 2 +- arch/arm/boot/dts/tegra114.dtsi | 2 +- arch/arm/boot/dts/vexpress-v2p-ca15-tc1.dts | 2 +- arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts | 2 +- arch/arm/boot/dts/xenvm-4.2.dts | 2 +- arch/arm64/boot/dts/foundation-v8.dts | 2 +- arch/arm64/boot/dts/rtsm_ve-aemv8a.dts | 2 +- 11 files changed, 11 insertions(+), 11 deletions(-) diff --git a/Documentation/devicetree/bindings/arm/gic.txt b/Documentation/devicetree/bindings/arm/gic.txt index 3dfb0c0..ea215e8 100644 --- a/Documentation/devicetree/bindings/arm/gic.txt +++ b/Documentation/devicetree/bindings/arm/gic.txt @@ -83,7 +83,7 @@ Example: #interrupt-cells = <3>; interrupt-controller; reg = <0x2c001000 0x1000>, - <0x2c002000 0x1000>, + <0x2c002000 0x2000>, <0x2c004000 0x2000>, <0x2c006000 0x2000>; interrupts = <1 9 0xf04>; diff --git a/arch/arm/boot/dts/ecx-2000.dts b/arch/arm/boot/dts/ecx-2000.dts index 139b40c..61927de 100644 --- a/arch/arm/boot/dts/ecx-2000.dts +++ b/arch/arm/boot/dts/ecx-2000.dts @@ -93,7 +93,7 @@ interrupt-controller; interrupts = <1 9 0xf04>; reg = <0xfff11000 0x1000>, - <0xfff12000 0x1000>, + <0xfff12000 0x2000>, <0xfff14000 0x2000>, <0xfff16000 0x2000>; }; diff --git a/arch/arm/boot/dts/exynos5.dtsi b/arch/arm/boot/dts/exynos5.dtsi index f65e124..c72782f 100644 --- a/arch/arm/boot/dts/exynos5.dtsi +++ b/arch/arm/boot/dts/exynos5.dtsi @@ -44,7 +44,7 @@ #interrupt-cells = <3>; interrupt-controller; reg = <0x10481000 0x1000>, - <0x10482000 0x1000>, + <0x10482000 0x2000>, <0x10484000 0x2000>, <0x10486000 0x2000>; interrupts = <1 9 0xf04>; diff --git a/arch/arm/boot/dts/exynos5440.dtsi b/arch/arm/boot/dts/exynos5440.dtsi index ff7f5d8..19c9636 100644 --- a/arch/arm/boot/dts/exynos5440.dtsi +++ b/arch/arm/boot/dts/exynos5440.dtsi @@ -31,7 +31,7 @@ #interrupt-cells = <3>; interrupt-controller; reg = <0x2E1000 0x1000>, - <0x2E2000 0x1000>, + <0x2E2000 0x2000>, <0x2E4000 0x2000>, <0x2E6000 0x2000>; interrupts = <1 9 0xf04>; diff --git a/arch/arm/boot/dts/omap5.dtsi b/arch/arm/boot/dts/omap5.dtsi index e643620..6c4fe08 100644 --- a/arch/arm/boot/dts/omap5.dtsi +++ b/arch/arm/boot/dts/omap5.dtsi @@ -60,7 +60,7 @@ interrupt-controller; #interrupt-cells = <3>; reg = <0x48211000 0x1000>, - <0x48212000 0x1000>, + <0x48212000 0x2000>, <0x48214000 0x2000>, <0x48216000 0x2000>; }; diff --git a/arch/arm/boot/dts/tegra114.dtsi b/arch/arm/boot/dts/tegra114.dtsi index abf6c40..eb7a0c4 100644 --- a/arch/arm/boot/dts/tegra114.dtsi +++ b/arch/arm/boot/dts/tegra114.dtsi @@ -20,7 +20,7 @@ #interrupt-cells = <3>; interrupt-controller; reg = <0x50041000 0x1000>, - <0x50042000 0x1000>, + <0x50042000 0x2000>, <0x50044000 0x2000>, <0x50046000 0x2000>; interrupts = <GIC_PPI 9 diff --git a/arch/arm/boot/dts/vexpress-v2p-ca15-tc1.dts b/arch/arm/boot/dts/vexpress-v2p-ca15-tc1.dts index 9420053..f0ec2be 100644 --- a/arch/arm/boot/dts/vexpress-v2p-ca15-tc1.dts +++ b/arch/arm/boot/dts/vexpress-v2p-ca15-tc1.dts @@ -81,7 +81,7 @@ #address-cells = <0>; interrupt-controller; reg = <0 0x2c001000 0 0x1000>, - <0 0x2c002000 0 0x1000>, + <0 0x2c002000 0 0x2000>, <0 0x2c004000 0 0x2000>, <0 0x2c006000 0 0x2000>; interrupts = <1 9 0xf04>; diff --git a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts index d2803be..c2db05e 100644 --- a/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts +++ b/arch/arm/boot/dts/vexpress-v2p-ca15_a7.dts @@ -98,7 +98,7 @@ #address-cells = <0>; interrupt-controller; reg = <0 0x2c001000 0 0x1000>, - <0 0x2c002000 0 0x1000>, + <0 0x2c002000 0 0x2000>, <0 0x2c004000 0 0x2000>, <0 0x2c006000 0 0x2000>; interrupts = <1 9 0xf04>; diff --git a/arch/arm/boot/dts/xenvm-4.2.dts b/arch/arm/boot/dts/xenvm-4.2.dts index 3369151..92e63eb 100644 --- a/arch/arm/boot/dts/xenvm-4.2.dts +++ b/arch/arm/boot/dts/xenvm-4.2.dts @@ -56,7 +56,7 @@ #address-cells = <0>; interrupt-controller; reg = <0 0x2c001000 0 0x1000>, - <0 0x2c002000 0 0x100>; + <0 0x2c002000 0 0x2000>; }; timer { diff --git a/arch/arm64/boot/dts/foundation-v8.dts b/arch/arm64/boot/dts/foundation-v8.dts index 84fcc50..3e63def 100644 --- a/arch/arm64/boot/dts/foundation-v8.dts +++ b/arch/arm64/boot/dts/foundation-v8.dts @@ -68,7 +68,7 @@ #address-cells = <0>; interrupt-controller; reg = <0x0 0x2c001000 0 0x1000>, - <0x0 0x2c002000 0 0x1000>, + <0x0 0x2c002000 0 0x2000>, <0x0 0x2c004000 0 0x2000>, <0x0 0x2c006000 0 0x2000>; interrupts = <1 9 0xf04>; diff --git a/arch/arm64/boot/dts/rtsm_ve-aemv8a.dts b/arch/arm64/boot/dts/rtsm_ve-aemv8a.dts index 572005e..0d271ce 100644 --- a/arch/arm64/boot/dts/rtsm_ve-aemv8a.dts +++ b/arch/arm64/boot/dts/rtsm_ve-aemv8a.dts @@ -73,7 +73,7 @@ #address-cells = <0>; interrupt-controller; reg = <0x0 0x2c001000 0 0x1000>, - <0x0 0x2c002000 0 0x1000>, + <0x0 0x2c002000 0 0x2000>, <0x0 0x2c004000 0 0x2000>, <0x0 0x2c006000 0 0x2000>; interrupts = <1 9 0xf04>; -- 1.8.4.rc3 ^ permalink raw reply related [flat|nested] 7+ messages in thread
[parent not found: <1383300809-7527-1-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org>]
* Re: [PATCH 1/2] arm: correct GIC cpu interface range for Cortex A15/GIC 400 [not found] ` <1383300809-7527-1-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> @ 2013-11-01 22:43 ` Rob Herring 0 siblings, 0 replies; 7+ messages in thread From: Rob Herring @ 2013-11-01 22:43 UTC (permalink / raw) To: Ian Campbell Cc: devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org On Fri, Nov 1, 2013 at 5:13 AM, Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> wrote: > GICv2 (Cortex A15/GIC 400) have CPU interface registers up to offset 0x1004 > (the 32-bit GICC_DIR register a 0x1000). The GIC 400 documentation specifies > the CPU interface region as being 0x2000 in size. > > Update all DTS entries claiming "arm,cortex-a15-gic" compatibility. Of these I > only have personal experience with the vexpress a Calxeda (ecx) platforms and > annecdotal evidence for omap and exynos. For the others I'm just assuming. > > Also update the example a15 binding in the documentation. > > The "Texas Instruments Keystone 2 SoC" platform which claim to have an a15 > compatible GIC but does not include the GICH/GICV registers. I've not touched > that here. > > Signed-off-by: Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> > Cc: devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org > Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org Acked-by: Rob Herring <rob.herring-bsGFqQB8/DxBDgjK7y7TUQ@public.gmane.org> -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers. 2013-11-01 10:12 [PATCH 0/2] DT: correctly describe GIC400 virtualisation regions Ian Campbell 2013-11-01 10:13 ` [PATCH 1/2] arm: correct GIC cpu interface range for Cortex A15/GIC 400 Ian Campbell @ 2013-11-01 10:13 ` Ian Campbell [not found] ` <1383300809-7527-2-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> 1 sibling, 1 reply; 7+ messages in thread From: Ian Campbell @ 2013-11-01 10:13 UTC (permalink / raw) To: devicetree Cc: Feng Kan, Kumar Sankaran, Ian Campbell, linux-arm-kernel, Loc Ho At least one platform (APM Storm) places the two pages of the GIC cpu interface (and the vcpu side) at non-contiguous locations. Document two additional regions to cover this split and update the corresponding dtsi. Note that Linux (including KVM) does not use any registers in the second page so there is no associated code change here. Xen will use these new regions, although I've not written the corresponding code yet. The ordering of these new regs is slightly counter intuitive but is inteded to be backward compatible. It is also assumed that all such systems will implement the GIC virtualisation extensions. Add comments to all of the reg examples to help clarify what is going on. Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Cc: devicetree@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org Cc: Kumar Sankaran <ksankaran@apm.com> Cc: Loc Ho <lho@apm.com> Cc: Feng Kan <fkan@apm.com> --- Documentation/devicetree/bindings/arm/gic.txt | 57 +++++++++++++++++++++------ arch/arm64/boot/dts/apm-storm.dtsi | 6 ++- 2 files changed, 50 insertions(+), 13 deletions(-) diff --git a/Documentation/devicetree/bindings/arm/gic.txt b/Documentation/devicetree/bindings/arm/gic.txt index ea215e8..00836d9 100644 --- a/Documentation/devicetree/bindings/arm/gic.txt +++ b/Documentation/devicetree/bindings/arm/gic.txt @@ -38,7 +38,8 @@ Main node required properties: - reg : Specifies base physical address(s) and size of the GIC registers. The first region is the GIC distributor register base and size. The 2nd region is - the GIC cpu interface register base and size. + the GIC cpu interface register base and size (See also "Split GIC + cpu interface", below). Optional - interrupts : Interrupt source of the parent interrupt controller on @@ -56,8 +57,8 @@ Example: #interrupt-cells = <3>; #address-cells = <1>; interrupt-controller; - reg = <0xfff11000 0x1000>, - <0xfff10100 0x100>; + reg = <0xfff11000 0x1000>, /* GIC Dist */ + <0xfff10100 0x100>; /* GIC CPU */ }; @@ -70,9 +71,11 @@ primary interrupt controller). Required properties: - reg : Additional regions specifying the base physical address and - size of the VGIC registers. The first additional region is the GIC - virtual interface control register base and size. The 2nd additional - region is the GIC virtual cpu interface register base and size. + size of the VGIC registers. The first additional region (i.e. third + overall) is the GIC virtual interface control register base and + size. The 2nd additional region (i.e. forth overall) is the GIC + virtual cpu interface register base and size (See also "Split GIC + cpu interface", below). - interrupts : VGIC maintenance interrupt. @@ -82,9 +85,41 @@ Example: compatible = "arm,cortex-a15-gic"; #interrupt-cells = <3>; interrupt-controller; - reg = <0x2c001000 0x1000>, - <0x2c002000 0x2000>, - <0x2c004000 0x2000>, - <0x2c006000 0x2000>; - interrupts = <1 9 0xf04>; + reg = <0x2c001000 0x1000>, /* GIC Dist */ + <0x2c002000 0x2000>, /* GIC CPU */ + <0x2c004000 0x2000>, /* GIC VCPU Control */ + <0x2c006000 0x2000>; /* GIC VCPU1 */ + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ + }; + +* Split GIC cpu interface + + The cpu interfaces (bare-metal in region 2 and virtual in region 4) + may be spread over two pages, with the GICC_DIR (Deactivate + Interrupt Register) register falling at the start of the second + page. If these pages are contiguous then this is described via the + size of the second and fourth entries as described above + (e.g. 0x2000 rather than 0x1000). + + However if the two pages are not contiguous then two additional + regions are present (5th and 6th) describing the location of the the + second half of the GIC cpu interface and GIC virtual cpu interface + respectively. + + It is assumed that all such systems will implement the GIC + virtualisation extensions. + +Example: + + gic: interrupt-controller@78010000 { + compatible = "arm,cortex-a15-gic"; + #interrupt-cells = <3>; + interrupt-controller; + reg = <0x78010000 0x1000>, /* GIC Dist */ + <0x78020000 0x1000>, /* GIC CPU1 */ + <0x78040000 0x2000>, /* GIC VCPU Control */ + <0x78060000 0x1000>, /* GIC VCPU1 */ + <0x78030000 0x1000>, /* GIC CPU2 */ + <0x78070000 0x1000>; /* GIC VCPU2 */ + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ }; diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi index bfdc578..b5d141a 100644 --- a/arch/arm64/boot/dts/apm-storm.dtsi +++ b/arch/arm64/boot/dts/apm-storm.dtsi @@ -82,9 +82,11 @@ #interrupt-cells = <3>; interrupt-controller; reg = <0x0 0x78010000 0x0 0x1000>, /* GIC Dist */ - <0x0 0x78020000 0x0 0x1000>, /* GIC CPU */ + <0x0 0x78020000 0x0 0x1000>, /* GIC CPU1 */ <0x0 0x78040000 0x0 0x2000>, /* GIC VCPU Control */ - <0x0 0x78060000 0x0 0x2000>; /* GIC VCPU */ + <0x0 0x78060000 0x0 0x1000>, /* GIC VCPU1 */ + <0x0 0x78030000 0x0 0x1000>, /* GIC CPU2 */ + <0x0 0x78070000 0x0 0x1000>; /* GIC VCPU2 */ interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ }; -- 1.8.4.rc3 ^ permalink raw reply related [flat|nested] 7+ messages in thread
[parent not found: <1383300809-7527-2-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org>]
* Re: [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers. [not found] ` <1383300809-7527-2-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> @ 2013-11-22 6:54 ` Anup Patel [not found] ` <CAAhSdy0+AFrS8ZVwmM6Sx=qScx6d1DkQtc4+2SB3=PpsJiJDog-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> 0 siblings, 1 reply; 7+ messages in thread From: Anup Patel @ 2013-11-22 6:54 UTC (permalink / raw) To: Ian Campbell Cc: devicetree-u79uwXL29TY76Z2rM5mHXA, Feng Kan, Kumar Sankaran, linux-arm-kernel, Loc Ho, Catalin Marinas, Marc Zyngier, Will Deacon On Fri, Nov 1, 2013 at 3:43 PM, Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> wrote: > At least one platform (APM Storm) places the two pages of the GIC cpu interface > (and the vcpu side) at non-contiguous locations. Document two additional > regions to cover this split and update the corresponding dtsi. Note that Linux > (including KVM) does not use any registers in the second page so there is no > associated code change here. Xen will use these new regions, although I've not > written the corresponding code yet. > > The ordering of these new regs is slightly counter intuitive but is inteded to > be backward compatible. It is also assumed that all such systems will implement > the GIC virtualisation extensions. > > Add comments to all of the reg examples to help clarify what is going on. Adding Catalin, Marc Z, and Will D. I looked up the GIC-400 TRM (more specifically page 33 of DDI0471B_gic400_r0p1_trm.pdf) but could not find any info on additional page for CPU interface. Am I looking at the wrong doc? -- Anup > > Signed-off-by: Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> > Cc: devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org > Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org > Cc: Kumar Sankaran <ksankaran-qTEPVZfXA3Y@public.gmane.org> > Cc: Loc Ho <lho-qTEPVZfXA3Y@public.gmane.org> > Cc: Feng Kan <fkan-qTEPVZfXA3Y@public.gmane.org> > --- > Documentation/devicetree/bindings/arm/gic.txt | 57 +++++++++++++++++++++------ > arch/arm64/boot/dts/apm-storm.dtsi | 6 ++- > 2 files changed, 50 insertions(+), 13 deletions(-) > > diff --git a/Documentation/devicetree/bindings/arm/gic.txt b/Documentation/devicetree/bindings/arm/gic.txt > index ea215e8..00836d9 100644 > --- a/Documentation/devicetree/bindings/arm/gic.txt > +++ b/Documentation/devicetree/bindings/arm/gic.txt > @@ -38,7 +38,8 @@ Main node required properties: > > - reg : Specifies base physical address(s) and size of the GIC registers. The > first region is the GIC distributor register base and size. The 2nd region is > - the GIC cpu interface register base and size. > + the GIC cpu interface register base and size (See also "Split GIC > + cpu interface", below). > > Optional > - interrupts : Interrupt source of the parent interrupt controller on > @@ -56,8 +57,8 @@ Example: > #interrupt-cells = <3>; > #address-cells = <1>; > interrupt-controller; > - reg = <0xfff11000 0x1000>, > - <0xfff10100 0x100>; > + reg = <0xfff11000 0x1000>, /* GIC Dist */ > + <0xfff10100 0x100>; /* GIC CPU */ > }; > > > @@ -70,9 +71,11 @@ primary interrupt controller). > Required properties: > > - reg : Additional regions specifying the base physical address and > - size of the VGIC registers. The first additional region is the GIC > - virtual interface control register base and size. The 2nd additional > - region is the GIC virtual cpu interface register base and size. > + size of the VGIC registers. The first additional region (i.e. third > + overall) is the GIC virtual interface control register base and > + size. The 2nd additional region (i.e. forth overall) is the GIC > + virtual cpu interface register base and size (See also "Split GIC > + cpu interface", below). > > - interrupts : VGIC maintenance interrupt. > > @@ -82,9 +85,41 @@ Example: > compatible = "arm,cortex-a15-gic"; > #interrupt-cells = <3>; > interrupt-controller; > - reg = <0x2c001000 0x1000>, > - <0x2c002000 0x2000>, > - <0x2c004000 0x2000>, > - <0x2c006000 0x2000>; > - interrupts = <1 9 0xf04>; > + reg = <0x2c001000 0x1000>, /* GIC Dist */ > + <0x2c002000 0x2000>, /* GIC CPU */ > + <0x2c004000 0x2000>, /* GIC VCPU Control */ > + <0x2c006000 0x2000>; /* GIC VCPU1 */ > + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ > + }; > + > +* Split GIC cpu interface > + > + The cpu interfaces (bare-metal in region 2 and virtual in region 4) > + may be spread over two pages, with the GICC_DIR (Deactivate > + Interrupt Register) register falling at the start of the second > + page. If these pages are contiguous then this is described via the > + size of the second and fourth entries as described above > + (e.g. 0x2000 rather than 0x1000). > + > + However if the two pages are not contiguous then two additional > + regions are present (5th and 6th) describing the location of the the > + second half of the GIC cpu interface and GIC virtual cpu interface > + respectively. > + > + It is assumed that all such systems will implement the GIC > + virtualisation extensions. > + > +Example: > + > + gic: interrupt-controller@78010000 { > + compatible = "arm,cortex-a15-gic"; > + #interrupt-cells = <3>; > + interrupt-controller; > + reg = <0x78010000 0x1000>, /* GIC Dist */ > + <0x78020000 0x1000>, /* GIC CPU1 */ > + <0x78040000 0x2000>, /* GIC VCPU Control */ > + <0x78060000 0x1000>, /* GIC VCPU1 */ > + <0x78030000 0x1000>, /* GIC CPU2 */ > + <0x78070000 0x1000>; /* GIC VCPU2 */ > + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ > }; > diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi > index bfdc578..b5d141a 100644 > --- a/arch/arm64/boot/dts/apm-storm.dtsi > +++ b/arch/arm64/boot/dts/apm-storm.dtsi > @@ -82,9 +82,11 @@ > #interrupt-cells = <3>; > interrupt-controller; > reg = <0x0 0x78010000 0x0 0x1000>, /* GIC Dist */ > - <0x0 0x78020000 0x0 0x1000>, /* GIC CPU */ > + <0x0 0x78020000 0x0 0x1000>, /* GIC CPU1 */ > <0x0 0x78040000 0x0 0x2000>, /* GIC VCPU Control */ > - <0x0 0x78060000 0x0 0x2000>; /* GIC VCPU */ > + <0x0 0x78060000 0x0 0x1000>, /* GIC VCPU1 */ > + <0x0 0x78030000 0x0 0x1000>, /* GIC CPU2 */ > + <0x0 0x78070000 0x0 0x1000>; /* GIC VCPU2 */ > interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ > }; > > -- > 1.8.4.rc3 > > > _______________________________________________ > linux-arm-kernel mailing list > linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
[parent not found: <CAAhSdy0+AFrS8ZVwmM6Sx=qScx6d1DkQtc4+2SB3=PpsJiJDog-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>]
* Re: [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers. [not found] ` <CAAhSdy0+AFrS8ZVwmM6Sx=qScx6d1DkQtc4+2SB3=PpsJiJDog-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> @ 2013-11-22 7:04 ` Anup Patel [not found] ` <CAAhSdy28OS+4QvFfEJ9o2MOGb6jS9Tjus_v7tsZTH9ZA3h=nRA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> 0 siblings, 1 reply; 7+ messages in thread From: Anup Patel @ 2013-11-22 7:04 UTC (permalink / raw) To: Ian Campbell Cc: devicetree-u79uwXL29TY76Z2rM5mHXA, Feng Kan, Kumar Sankaran, linux-arm-kernel, Loc Ho, Catalin Marinas, Marc Zyngier, Will Deacon On Fri, Nov 22, 2013 at 12:24 PM, Anup Patel <anup-aa5cPYPs4nSSZfiimGR9Ow@public.gmane.org> wrote: > On Fri, Nov 1, 2013 at 3:43 PM, Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> wrote: >> At least one platform (APM Storm) places the two pages of the GIC cpu interface >> (and the vcpu side) at non-contiguous locations. Document two additional >> regions to cover this split and update the corresponding dtsi. Note that Linux >> (including KVM) does not use any registers in the second page so there is no >> associated code change here. Xen will use these new regions, although I've not >> written the corresponding code yet. >> >> The ordering of these new regs is slightly counter intuitive but is inteded to >> be backward compatible. It is also assumed that all such systems will implement >> the GIC virtualisation extensions. >> >> Add comments to all of the reg examples to help clarify what is going on. > > Adding Catalin, Marc Z, and Will D. > > I looked up the GIC-400 TRM (more specifically page 33 of > DDI0471B_gic400_r0p1_trm.pdf) but could not find any info > on additional page for CPU interface. > > Am I looking at the wrong doc? Never mind, found the GICC_DIR (offset 0x1000) register. -- Anup > > -- > Anup > >> >> Signed-off-by: Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> >> Cc: devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org >> Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org >> Cc: Kumar Sankaran <ksankaran-qTEPVZfXA3Y@public.gmane.org> >> Cc: Loc Ho <lho-qTEPVZfXA3Y@public.gmane.org> >> Cc: Feng Kan <fkan-qTEPVZfXA3Y@public.gmane.org> >> --- >> Documentation/devicetree/bindings/arm/gic.txt | 57 +++++++++++++++++++++------ >> arch/arm64/boot/dts/apm-storm.dtsi | 6 ++- >> 2 files changed, 50 insertions(+), 13 deletions(-) >> >> diff --git a/Documentation/devicetree/bindings/arm/gic.txt b/Documentation/devicetree/bindings/arm/gic.txt >> index ea215e8..00836d9 100644 >> --- a/Documentation/devicetree/bindings/arm/gic.txt >> +++ b/Documentation/devicetree/bindings/arm/gic.txt >> @@ -38,7 +38,8 @@ Main node required properties: >> >> - reg : Specifies base physical address(s) and size of the GIC registers. The >> first region is the GIC distributor register base and size. The 2nd region is >> - the GIC cpu interface register base and size. >> + the GIC cpu interface register base and size (See also "Split GIC >> + cpu interface", below). >> >> Optional >> - interrupts : Interrupt source of the parent interrupt controller on >> @@ -56,8 +57,8 @@ Example: >> #interrupt-cells = <3>; >> #address-cells = <1>; >> interrupt-controller; >> - reg = <0xfff11000 0x1000>, >> - <0xfff10100 0x100>; >> + reg = <0xfff11000 0x1000>, /* GIC Dist */ >> + <0xfff10100 0x100>; /* GIC CPU */ >> }; >> >> >> @@ -70,9 +71,11 @@ primary interrupt controller). >> Required properties: >> >> - reg : Additional regions specifying the base physical address and >> - size of the VGIC registers. The first additional region is the GIC >> - virtual interface control register base and size. The 2nd additional >> - region is the GIC virtual cpu interface register base and size. >> + size of the VGIC registers. The first additional region (i.e. third >> + overall) is the GIC virtual interface control register base and >> + size. The 2nd additional region (i.e. forth overall) is the GIC >> + virtual cpu interface register base and size (See also "Split GIC >> + cpu interface", below). >> >> - interrupts : VGIC maintenance interrupt. >> >> @@ -82,9 +85,41 @@ Example: >> compatible = "arm,cortex-a15-gic"; >> #interrupt-cells = <3>; >> interrupt-controller; >> - reg = <0x2c001000 0x1000>, >> - <0x2c002000 0x2000>, >> - <0x2c004000 0x2000>, >> - <0x2c006000 0x2000>; >> - interrupts = <1 9 0xf04>; >> + reg = <0x2c001000 0x1000>, /* GIC Dist */ >> + <0x2c002000 0x2000>, /* GIC CPU */ >> + <0x2c004000 0x2000>, /* GIC VCPU Control */ >> + <0x2c006000 0x2000>; /* GIC VCPU1 */ >> + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ >> + }; >> + >> +* Split GIC cpu interface >> + >> + The cpu interfaces (bare-metal in region 2 and virtual in region 4) >> + may be spread over two pages, with the GICC_DIR (Deactivate >> + Interrupt Register) register falling at the start of the second >> + page. If these pages are contiguous then this is described via the >> + size of the second and fourth entries as described above >> + (e.g. 0x2000 rather than 0x1000). >> + >> + However if the two pages are not contiguous then two additional >> + regions are present (5th and 6th) describing the location of the the >> + second half of the GIC cpu interface and GIC virtual cpu interface >> + respectively. >> + >> + It is assumed that all such systems will implement the GIC >> + virtualisation extensions. >> + >> +Example: >> + >> + gic: interrupt-controller@78010000 { >> + compatible = "arm,cortex-a15-gic"; >> + #interrupt-cells = <3>; >> + interrupt-controller; >> + reg = <0x78010000 0x1000>, /* GIC Dist */ >> + <0x78020000 0x1000>, /* GIC CPU1 */ >> + <0x78040000 0x2000>, /* GIC VCPU Control */ >> + <0x78060000 0x1000>, /* GIC VCPU1 */ >> + <0x78030000 0x1000>, /* GIC CPU2 */ >> + <0x78070000 0x1000>; /* GIC VCPU2 */ >> + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ >> }; >> diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi >> index bfdc578..b5d141a 100644 >> --- a/arch/arm64/boot/dts/apm-storm.dtsi >> +++ b/arch/arm64/boot/dts/apm-storm.dtsi >> @@ -82,9 +82,11 @@ >> #interrupt-cells = <3>; >> interrupt-controller; >> reg = <0x0 0x78010000 0x0 0x1000>, /* GIC Dist */ >> - <0x0 0x78020000 0x0 0x1000>, /* GIC CPU */ >> + <0x0 0x78020000 0x0 0x1000>, /* GIC CPU1 */ >> <0x0 0x78040000 0x0 0x2000>, /* GIC VCPU Control */ >> - <0x0 0x78060000 0x0 0x2000>; /* GIC VCPU */ >> + <0x0 0x78060000 0x0 0x1000>, /* GIC VCPU1 */ >> + <0x0 0x78030000 0x0 0x1000>, /* GIC CPU2 */ >> + <0x0 0x78070000 0x0 0x1000>; /* GIC VCPU2 */ >> interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ >> }; >> >> -- >> 1.8.4.rc3 >> >> >> _______________________________________________ >> linux-arm-kernel mailing list >> linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org >> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
[parent not found: <CAAhSdy28OS+4QvFfEJ9o2MOGb6jS9Tjus_v7tsZTH9ZA3h=nRA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>]
* Re: [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers. [not found] ` <CAAhSdy28OS+4QvFfEJ9o2MOGb6jS9Tjus_v7tsZTH9ZA3h=nRA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org> @ 2013-11-22 10:36 ` Anup Patel 0 siblings, 0 replies; 7+ messages in thread From: Anup Patel @ 2013-11-22 10:36 UTC (permalink / raw) To: Anup Patel Cc: Ian Campbell, devicetree-u79uwXL29TY76Z2rM5mHXA, Feng Kan, Marc Zyngier, Catalin Marinas, Will Deacon, Loc Ho, Kumar Sankaran, linux-arm-kernel, vkale-qTEPVZfXA3Y On 22 November 2013 12:34, Anup Patel <anup-aa5cPYPs4nSSZfiimGR9Ow@public.gmane.org> wrote: > On Fri, Nov 22, 2013 at 12:24 PM, Anup Patel <anup-aa5cPYPs4nSSZfiimGR9Ow@public.gmane.org> wrote: >> On Fri, Nov 1, 2013 at 3:43 PM, Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> wrote: >>> At least one platform (APM Storm) places the two pages of the GIC cpu interface >>> (and the vcpu side) at non-contiguous locations. Document two additional >>> regions to cover this split and update the corresponding dtsi. Note that Linux >>> (including KVM) does not use any registers in the second page so there is no >>> associated code change here. Xen will use these new regions, although I've not >>> written the corresponding code yet. >>> >>> The ordering of these new regs is slightly counter intuitive but is inteded to >>> be backward compatible. It is also assumed that all such systems will implement >>> the GIC virtualisation extensions. >>> >>> Add comments to all of the reg examples to help clarify what is going on. >> >> Adding Catalin, Marc Z, and Will D. >> >> I looked up the GIC-400 TRM (more specifically page 33 of >> DDI0471B_gic400_r0p1_trm.pdf) but could not find any info >> on additional page for CPU interface. >> >> Am I looking at the wrong doc? > > Never mind, found the GICC_DIR (offset 0x1000) register. > > -- > Anup > >> >> -- >> Anup >> >>> >>> Signed-off-by: Ian Campbell <ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org> >>> Cc: devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org >>> Cc: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org >>> Cc: Kumar Sankaran <ksankaran-qTEPVZfXA3Y@public.gmane.org> >>> Cc: Loc Ho <lho-qTEPVZfXA3Y@public.gmane.org> >>> Cc: Feng Kan <fkan-qTEPVZfXA3Y@public.gmane.org> >>> --- >>> Documentation/devicetree/bindings/arm/gic.txt | 57 +++++++++++++++++++++------ >>> arch/arm64/boot/dts/apm-storm.dtsi | 6 ++- >>> 2 files changed, 50 insertions(+), 13 deletions(-) >>> >>> diff --git a/Documentation/devicetree/bindings/arm/gic.txt b/Documentation/devicetree/bindings/arm/gic.txt >>> index ea215e8..00836d9 100644 >>> --- a/Documentation/devicetree/bindings/arm/gic.txt >>> +++ b/Documentation/devicetree/bindings/arm/gic.txt >>> @@ -38,7 +38,8 @@ Main node required properties: >>> >>> - reg : Specifies base physical address(s) and size of the GIC registers. The >>> first region is the GIC distributor register base and size. The 2nd region is >>> - the GIC cpu interface register base and size. >>> + the GIC cpu interface register base and size (See also "Split GIC >>> + cpu interface", below). >>> >>> Optional >>> - interrupts : Interrupt source of the parent interrupt controller on >>> @@ -56,8 +57,8 @@ Example: >>> #interrupt-cells = <3>; >>> #address-cells = <1>; >>> interrupt-controller; >>> - reg = <0xfff11000 0x1000>, >>> - <0xfff10100 0x100>; >>> + reg = <0xfff11000 0x1000>, /* GIC Dist */ >>> + <0xfff10100 0x100>; /* GIC CPU */ >>> }; >>> >>> >>> @@ -70,9 +71,11 @@ primary interrupt controller). >>> Required properties: >>> >>> - reg : Additional regions specifying the base physical address and >>> - size of the VGIC registers. The first additional region is the GIC >>> - virtual interface control register base and size. The 2nd additional >>> - region is the GIC virtual cpu interface register base and size. >>> + size of the VGIC registers. The first additional region (i.e. third >>> + overall) is the GIC virtual interface control register base and >>> + size. The 2nd additional region (i.e. forth overall) is the GIC >>> + virtual cpu interface register base and size (See also "Split GIC >>> + cpu interface", below). >>> >>> - interrupts : VGIC maintenance interrupt. >>> >>> @@ -82,9 +85,41 @@ Example: >>> compatible = "arm,cortex-a15-gic"; >>> #interrupt-cells = <3>; >>> interrupt-controller; >>> - reg = <0x2c001000 0x1000>, >>> - <0x2c002000 0x2000>, >>> - <0x2c004000 0x2000>, >>> - <0x2c006000 0x2000>; >>> - interrupts = <1 9 0xf04>; >>> + reg = <0x2c001000 0x1000>, /* GIC Dist */ >>> + <0x2c002000 0x2000>, /* GIC CPU */ >>> + <0x2c004000 0x2000>, /* GIC VCPU Control */ >>> + <0x2c006000 0x2000>; /* GIC VCPU1 */ >>> + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ >>> + }; >>> + >>> +* Split GIC cpu interface >>> + >>> + The cpu interfaces (bare-metal in region 2 and virtual in region 4) >>> + may be spread over two pages, with the GICC_DIR (Deactivate >>> + Interrupt Register) register falling at the start of the second >>> + page. If these pages are contiguous then this is described via the >>> + size of the second and fourth entries as described above >>> + (e.g. 0x2000 rather than 0x1000). >>> + >>> + However if the two pages are not contiguous then two additional >>> + regions are present (5th and 6th) describing the location of the the >>> + second half of the GIC cpu interface and GIC virtual cpu interface >>> + respectively. >>> + >>> + It is assumed that all such systems will implement the GIC >>> + virtualisation extensions. >>> + >>> +Example: >>> + >>> + gic: interrupt-controller@78010000 { >>> + compatible = "arm,cortex-a15-gic"; >>> + #interrupt-cells = <3>; >>> + interrupt-controller; >>> + reg = <0x78010000 0x1000>, /* GIC Dist */ >>> + <0x78020000 0x1000>, /* GIC CPU1 */ >>> + <0x78040000 0x2000>, /* GIC VCPU Control */ >>> + <0x78060000 0x1000>, /* GIC VCPU1 */ >>> + <0x78030000 0x1000>, /* GIC CPU2 */ >>> + <0x78070000 0x1000>; /* GIC VCPU2 */ >>> + interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ >>> }; >>> diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi >>> index bfdc578..b5d141a 100644 >>> --- a/arch/arm64/boot/dts/apm-storm.dtsi >>> +++ b/arch/arm64/boot/dts/apm-storm.dtsi >>> @@ -82,9 +82,11 @@ >>> #interrupt-cells = <3>; >>> interrupt-controller; >>> reg = <0x0 0x78010000 0x0 0x1000>, /* GIC Dist */ >>> - <0x0 0x78020000 0x0 0x1000>, /* GIC CPU */ >>> + <0x0 0x78020000 0x0 0x1000>, /* GIC CPU1 */ >>> <0x0 0x78040000 0x0 0x2000>, /* GIC VCPU Control */ >>> - <0x0 0x78060000 0x0 0x2000>; /* GIC VCPU */ >>> + <0x0 0x78060000 0x0 0x1000>, /* GIC VCPU1 */ >>> + <0x0 0x78030000 0x0 0x1000>, /* GIC CPU2 */ >>> + <0x0 0x78070000 0x0 0x1000>; /* GIC VCPU2 */ >>> interrupts = <1 9 0xf04>; /* GIC Maintenence IRQ */ >>> }; >>> >>> -- >>> 1.8.4.rc3 >>> I have looked at the proposed device tree binding change from APM X-Gene perspective. This change looks good to me. (Vinayak ??) Regards, Anup >>> >>> _______________________________________________ >>> linux-arm-kernel mailing list >>> linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org >>> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel > > _______________________________________________ > linux-arm-kernel mailing list > linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2013-11-22 10:36 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2013-11-01 10:12 [PATCH 0/2] DT: correctly describe GIC400 virtualisation regions Ian Campbell
2013-11-01 10:13 ` [PATCH 1/2] arm: correct GIC cpu interface range for Cortex A15/GIC 400 Ian Campbell
[not found] ` <1383300809-7527-1-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org>
2013-11-01 22:43 ` Rob Herring
2013-11-01 10:13 ` [PATCH 2/2] arm: Support for platforms with split GIC cpu interface registers Ian Campbell
[not found] ` <1383300809-7527-2-git-send-email-ian.campbell-Sxgqhf6Nn4DQT0dZR+AlfA@public.gmane.org>
2013-11-22 6:54 ` Anup Patel
[not found] ` <CAAhSdy0+AFrS8ZVwmM6Sx=qScx6d1DkQtc4+2SB3=PpsJiJDog-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-11-22 7:04 ` Anup Patel
[not found] ` <CAAhSdy28OS+4QvFfEJ9o2MOGb6jS9Tjus_v7tsZTH9ZA3h=nRA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-11-22 10:36 ` Anup Patel
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).