| Message ID | 1457116769-8567-2-git-send-email-eric@anholt.net (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Fri, Mar 04, 2016 at 10:39:29AM -0800, Eric Anholt wrote: > For now this doesn't support the new hardware present on the Pi 3 (BT, > wifi, GPIO expander). Since the GPIO expander isn't supported, we > also don't have the LEDs like the other board files do. > > Signed-off-by: Eric Anholt <eric@anholt.net> > --- > arch/arm/boot/dts/Makefile | 3 +- > arch/arm/boot/dts/bcm2837-rpi-3-b.dts | 22 ++++++++++++ > arch/arm/boot/dts/bcm2837.dtsi | 68 +++++++++++++++++++++++++++++++++++ > 3 files changed, 92 insertions(+), 1 deletion(-) > create mode 100644 arch/arm/boot/dts/bcm2837-rpi-3-b.dts > create mode 100644 arch/arm/boot/dts/bcm2837.dtsi > + timer { > + compatible = "arm,armv7-timer"; > + interrupt-parent = <&local_intc>; > + interrupts = <0>, // PHYS_SECURE_PPI > + <1>, // PHYS_NONSECURE_PPI > + <3>, // VIRT_PPI > + <2>; // HYP_PPI > + always-on; > + }; Are the CPUs in an always-on power domain? Or is it jsut that the kernel doesn't perform power management of CPUs? The always-on proeprty is only intended for the former. Mark.
Mark Rutland <mark.rutland@arm.com> writes: > On Fri, Mar 04, 2016 at 10:39:29AM -0800, Eric Anholt wrote: >> For now this doesn't support the new hardware present on the Pi 3 (BT, >> wifi, GPIO expander). Since the GPIO expander isn't supported, we >> also don't have the LEDs like the other board files do. >> >> Signed-off-by: Eric Anholt <eric@anholt.net> >> --- >> arch/arm/boot/dts/Makefile | 3 +- >> arch/arm/boot/dts/bcm2837-rpi-3-b.dts | 22 ++++++++++++ >> arch/arm/boot/dts/bcm2837.dtsi | 68 +++++++++++++++++++++++++++++++++++ >> 3 files changed, 92 insertions(+), 1 deletion(-) >> create mode 100644 arch/arm/boot/dts/bcm2837-rpi-3-b.dts >> create mode 100644 arch/arm/boot/dts/bcm2837.dtsi > >> + timer { >> + compatible = "arm,armv7-timer"; >> + interrupt-parent = <&local_intc>; >> + interrupts = <0>, // PHYS_SECURE_PPI >> + <1>, // PHYS_NONSECURE_PPI >> + <3>, // VIRT_PPI >> + <2>; // HYP_PPI >> + always-on; >> + }; > > Are the CPUs in an always-on power domain? Or is it jsut that the kernel > doesn't perform power management of CPUs? > > The always-on proeprty is only intended for the former. The kernel doesn't get to do power management of CPUs. We only have control of power domains through the firmware, and the firmware's keeping the CPU domain on.
On Fri, Mar 04, 2016 at 10:55:03AM -0800, Eric Anholt wrote: > Mark Rutland <mark.rutland@arm.com> writes: > > > On Fri, Mar 04, 2016 at 10:39:29AM -0800, Eric Anholt wrote: > >> For now this doesn't support the new hardware present on the Pi 3 (BT, > >> wifi, GPIO expander). Since the GPIO expander isn't supported, we > >> also don't have the LEDs like the other board files do. > >> > >> Signed-off-by: Eric Anholt <eric@anholt.net> > >> --- > >> arch/arm/boot/dts/Makefile | 3 +- > >> arch/arm/boot/dts/bcm2837-rpi-3-b.dts | 22 ++++++++++++ > >> arch/arm/boot/dts/bcm2837.dtsi | 68 +++++++++++++++++++++++++++++++++++ > >> 3 files changed, 92 insertions(+), 1 deletion(-) > >> create mode 100644 arch/arm/boot/dts/bcm2837-rpi-3-b.dts > >> create mode 100644 arch/arm/boot/dts/bcm2837.dtsi > > > >> + timer { > >> + compatible = "arm,armv7-timer"; > >> + interrupt-parent = <&local_intc>; > >> + interrupts = <0>, // PHYS_SECURE_PPI > >> + <1>, // PHYS_NONSECURE_PPI > >> + <3>, // VIRT_PPI > >> + <2>; // HYP_PPI > >> + always-on; > >> + }; > > > > Are the CPUs in an always-on power domain? Or is it jsut that the kernel > > doesn't perform power management of CPUs? > > > > The always-on proeprty is only intended for the former. > > The kernel doesn't get to do power management of CPUs. We only have > control of power domains through the firmware, and the firmware's > keeping the CPU domain on. So there is no way that the CPUs could request for the firmare to place them in a state where the timer would lose context (but other events coukd wake them up), even if they don't do that today? Mark
Mark Rutland <mark.rutland@arm.com> writes: > On Fri, Mar 04, 2016 at 10:55:03AM -0800, Eric Anholt wrote: >> Mark Rutland <mark.rutland@arm.com> writes: >> >> > On Fri, Mar 04, 2016 at 10:39:29AM -0800, Eric Anholt wrote: >> >> For now this doesn't support the new hardware present on the Pi 3 (BT, >> >> wifi, GPIO expander). Since the GPIO expander isn't supported, we >> >> also don't have the LEDs like the other board files do. >> >> >> >> Signed-off-by: Eric Anholt <eric@anholt.net> >> >> --- >> >> arch/arm/boot/dts/Makefile | 3 +- >> >> arch/arm/boot/dts/bcm2837-rpi-3-b.dts | 22 ++++++++++++ >> >> arch/arm/boot/dts/bcm2837.dtsi | 68 +++++++++++++++++++++++++++++++++++ >> >> 3 files changed, 92 insertions(+), 1 deletion(-) >> >> create mode 100644 arch/arm/boot/dts/bcm2837-rpi-3-b.dts >> >> create mode 100644 arch/arm/boot/dts/bcm2837.dtsi >> > >> >> + timer { >> >> + compatible = "arm,armv7-timer"; >> >> + interrupt-parent = <&local_intc>; >> >> + interrupts = <0>, // PHYS_SECURE_PPI >> >> + <1>, // PHYS_NONSECURE_PPI >> >> + <3>, // VIRT_PPI >> >> + <2>; // HYP_PPI >> >> + always-on; >> >> + }; >> > >> > Are the CPUs in an always-on power domain? Or is it jsut that the kernel >> > doesn't perform power management of CPUs? >> > >> > The always-on proeprty is only intended for the former. >> >> The kernel doesn't get to do power management of CPUs. We only have >> control of power domains through the firmware, and the firmware's >> keeping the CPU domain on. > > So there is no way that the CPUs could request for the firmare to place > them in a state where the timer would lose context (but other events > coukd wake them up), even if they don't do that today? I don't see any mechanism in the RPi firmware for the ARM to request suspend or anything like that. Enabling something like that would definitely require an updated DT, to enable a driver that talks to the firmware to make requests like that. (Note that this is the same timer setup we have in bcm2836.dtsi)
On 03/04/2016 11:39 AM, Eric Anholt wrote: > For now this doesn't support the new hardware present on the Pi 3 (BT, > wifi, GPIO expander). Since the GPIO expander isn't supported, we > also don't have the LEDs like the other board files do. The series, Acked-by: Stephen Warren <swarren@wwwdotorg.org> ... although I'd be slightly surprised if this was accepted; when AArch64 was first announced, I thought the mainline ARM kernel community decided that only 64-bit kernels would be supported on 64-bit HW? Still, I don't immediately see that in any commit log or in the kernel doc directory, so perhaps I'm misremembering. Hopefully; life is simpler that way:-)
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile index a4a6d70..71082a3 100644 --- a/arch/arm/boot/dts/Makefile +++ b/arch/arm/boot/dts/Makefile @@ -63,7 +63,8 @@ dtb-$(CONFIG_ARCH_BCM2835) += \ bcm2835-rpi-b-rev2.dtb \ bcm2835-rpi-b-plus.dtb \ bcm2835-rpi-a-plus.dtb \ - bcm2836-rpi-2-b.dtb + bcm2836-rpi-2-b.dtb \ + bcm2837-rpi-3-b.dtb dtb-$(CONFIG_ARCH_BCM_5301X) += \ bcm4708-asus-rt-ac56u.dtb \ bcm4708-asus-rt-ac68u.dtb \ diff --git a/arch/arm/boot/dts/bcm2837-rpi-3-b.dts b/arch/arm/boot/dts/bcm2837-rpi-3-b.dts new file mode 100644 index 0000000..5e8eafd --- /dev/null +++ b/arch/arm/boot/dts/bcm2837-rpi-3-b.dts @@ -0,0 +1,22 @@ +/dts-v1/; +#include "bcm2837.dtsi" +#include "bcm2835-rpi.dtsi" + +/ { + compatible = "raspberrypi,3-model-b", "brcm,bcm2837"; + model = "Raspberry Pi 3 Model B"; + + memory { + reg = <0 0x40000000>; + }; +}; + +&gpio { + pinctrl-0 = <&gpioout &alt0 &i2s_alt0 &alt3>; + + /* I2S interface */ + i2s_alt0: i2s_alt0 { + brcm,pins = <28 29 30 31>; + brcm,function = <BCM2835_FSEL_ALT2>; + }; +}; diff --git a/arch/arm/boot/dts/bcm2837.dtsi b/arch/arm/boot/dts/bcm2837.dtsi new file mode 100644 index 0000000..2f36722 --- /dev/null +++ b/arch/arm/boot/dts/bcm2837.dtsi @@ -0,0 +1,68 @@ +#include "bcm283x.dtsi" + +/ { + compatible = "brcm,bcm2836"; + + soc { + ranges = <0x7e000000 0x3f000000 0x1000000>, + <0x40000000 0x40000000 0x00001000>; + dma-ranges = <0xc0000000 0x00000000 0x3f000000>; + + local_intc: local_intc { + compatible = "brcm,bcm2836-l1-intc"; + reg = <0x40000000 0x100>; + interrupt-controller; + #interrupt-cells = <1>; + interrupt-parent = <&local_intc>; + }; + }; + + timer { + compatible = "arm,armv7-timer"; + interrupt-parent = <&local_intc>; + interrupts = <0>, // PHYS_SECURE_PPI + <1>, // PHYS_NONSECURE_PPI + <3>, // VIRT_PPI + <2>; // HYP_PPI + always-on; + }; + + cpus: cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu0: cpu@0 { + device_type = "cpu"; + compatible = "arm,cortex-a53"; + reg = <0>; + }; + + cpu1: cpu@1 { + device_type = "cpu"; + compatible = "arm,cortex-a53"; + reg = <1>; + }; + + cpu2: cpu@2 { + device_type = "cpu"; + compatible = "arm,cortex-a53"; + reg = <2>; + }; + + cpu3: cpu@3 { + device_type = "cpu"; + compatible = "arm,cortex-a53"; + reg = <3>; + }; + }; +}; + +/* Make the BCM2835-style global interrupt controller be a child of the + * CPU-local interrupt controller. + */ +&intc { + compatible = "brcm,bcm2836-armctrl-ic"; + reg = <0x7e00b200 0x200>; + interrupt-parent = <&local_intc>; + interrupts = <8>; +};
For now this doesn't support the new hardware present on the Pi 3 (BT, wifi, GPIO expander). Since the GPIO expander isn't supported, we also don't have the LEDs like the other board files do. Signed-off-by: Eric Anholt <eric@anholt.net> --- arch/arm/boot/dts/Makefile | 3 +- arch/arm/boot/dts/bcm2837-rpi-3-b.dts | 22 ++++++++++++ arch/arm/boot/dts/bcm2837.dtsi | 68 +++++++++++++++++++++++++++++++++++ 3 files changed, 92 insertions(+), 1 deletion(-) create mode 100644 arch/arm/boot/dts/bcm2837-rpi-3-b.dts create mode 100644 arch/arm/boot/dts/bcm2837.dtsi