| Message ID | 20170919183204.12751-1-daniel.thompson@linaro.org |
|---|---|
| State | Accepted |
| Commit | bc3d3447b66a9eb398c7cce96f05b7c78d725abc |
| Headers | show |
| Series | [v2] arm64: dts: foundation-v8: Enable PSCI mode | expand |
On 19/09/17 19:32, Daniel Thompson wrote: > Currently if the Foundation model is running ARM Trusted Firmware then > the kernel, which is configured to use spin tables, cannot start secondary > processors or "power off" the simulation. > > After adding a couple of labels to the include file and splitting out the > spin-table configuration into a header, we add a couple of new headers > together with two new DTs (GICv2+PSCI and GICv3+PSCI). > > The new GICv3+PSCI DT has been boot tested, the remaining three (two of > which existed prior to this patch) have been "tested" by decompiling the > blobs and comparing them against a reference. > How different are these from the ones hosted in [1] ? On argument is that we want to take the DTS out of device tree as firmware is responsible for generating them. Alternatively, we may be duplicating resulting in discrepancies over time by coping it into kernel. Since users of ARM TF must be able to access these, I am not sure if it makes sense to merge these. Or we remove it from ARM TF to avoid any conflicts/discrepancies. Thoughts ? -- Regards, Sudeep [1] https://github.com/ARM-software/arm-trusted-firmware/tree/master/fdts
On 20/09/17 10:42, Sudeep Holla wrote: > > > On 19/09/17 19:32, Daniel Thompson wrote: >> Currently if the Foundation model is running ARM Trusted Firmware then >> the kernel, which is configured to use spin tables, cannot start secondary >> processors or "power off" the simulation. >> >> After adding a couple of labels to the include file and splitting out the >> spin-table configuration into a header, we add a couple of new headers >> together with two new DTs (GICv2+PSCI and GICv3+PSCI). >> >> The new GICv3+PSCI DT has been boot tested, the remaining three (two of >> which existed prior to this patch) have been "tested" by decompiling the >> blobs and comparing them against a reference. >> > > How different are these from the ones hosted in [1] ? They look like they were either independently written or diverged a long time ago. The existing kernel DTs describe hardware absent from the ARM TF ones and vice versa. With specific reference to PSCI it looks like my patches could perhaps be improved by adding idle-state support. > On argument is that we want to take the DTS out of device tree as > firmware is responsible for generating them. Alternatively, we may be > duplicating resulting in discrepancies over time by coping it into kernel. The general problem is copying from where? The kernel DTs are a well maintained centralized repository which is *really* useful. git grep across the kernel DTs is a hugely powerful tool when trying to better understand an ecosystem as sprawling and diverse as ARMs. In fact I've even seen those sort of searchs used as a basis to clean up unused code. Seeing that centralized repository splinter into separate per-vendor silos would be a huge loss for kernel developers. > Since users of ARM TF must be able to access these, I am not sure if it > makes sense to merge these. Or we remove it from ARM TF to avoid any > conflicts/discrepancies. > > Thoughts ? I kind of agree that maintaining DTs and DT documentation in the kernel is a little odd given that the kernel is not the only player here (FreeBSD, u-boot, etc). However it is sufficiently well maintained that projects are content(ish) to regard the kernel as the canonical source for these things (u-boot, for example, seeks to shadow kernel DTs without modifying them). However regardless of the above I'd say they should be removed from ARM TF. ARM TF does not use, modify, pass on or in any way consume DT... it has no skin in the game here. Why does it want to own a few of blobs for a small subset of the platforms it supports? I'm afraid that makes no sense to me, to the extent that it didn't even occur to me to *look* in the ARM TF sources to find any DTs for FVP until you pointed them out. In other words, whilst people could discuss alternative ways to manage DTs[1], I can't see any universe where ARM TF would be a logical place to keep them. Daniel. [1] ... and I'd further suggest that only perhaps people who are prepared to put resources into fixing it should convene such a discussion.
Sorry for late response, I thought I had sent this mail out long back but was sitting in my draft :( On 20/09/17 12:17, Daniel Thompson wrote: > On 20/09/17 10:42, Sudeep Holla wrote: >> >> >> On 19/09/17 19:32, Daniel Thompson wrote: >>> Currently if the Foundation model is running ARM Trusted Firmware then >>> the kernel, which is configured to use spin tables, cannot start >>> secondary >>> processors or "power off" the simulation. >>> >>> After adding a couple of labels to the include file and splitting out >>> the >>> spin-table configuration into a header, we add a couple of new headers >>> together with two new DTs (GICv2+PSCI and GICv3+PSCI). >>> >>> The new GICv3+PSCI DT has been boot tested, the remaining three (two of >>> which existed prior to this patch) have been "tested" by decompiling the >>> blobs and comparing them against a reference. >>> >> >> How different are these from the ones hosted in [1] ? > > They look like they were either independently written or diverged a long > time ago. The existing kernel DTs describe hardware absent from the ARM > TF ones and vice versa. > OK. > With specific reference to PSCI it looks like my patches could perhaps > be improved by adding idle-state support. > > Yes I know. >> On argument is that we want to take the DTS out of device tree as >> firmware is responsible for generating them. Alternatively, we may be >> duplicating resulting in discrepancies over time by coping it into >> kernel. > > The general problem is copying from where? > > The kernel DTs are a well maintained centralized repository which is > *really* useful. git grep across the kernel DTs is a hugely powerful > tool when trying to better understand an ecosystem as sprawling and > diverse as ARMs. In fact I've even seen those sort of searchs used as a > basis to clean up unused code. Seeing that centralized repository > splinter into separate per-vendor silos would be a huge loss for kernel > developers. > Agreed. But models are configurable and last time this discussion came up, some argued that the DTs must be modified based on the configuration automatically by models or some external scripts. > >> Since users of ARM TF must be able to access these, I am not sure if it >> makes sense to merge these. Or we remove it from ARM TF to avoid any >> conflicts/discrepancies. > >> Thoughts ? > > I kind of agree that maintaining DTs and DT documentation in the kernel > is a little odd given that the kernel is not the only player here > (FreeBSD, u-boot, etc). However it is sufficiently well maintained that > projects are content(ish) to regard the kernel as the canonical source > for these things (u-boot, for example, seeks to shadow kernel DTs > without modifying them). > No argument there. > However regardless of the above I'd say they should be removed from ARM > TF. ARM TF does not use, modify, pass on or in any way consume DT... it > has no skin in the game here. Why does it want to own a few of blobs for > a small subset of the platforms it supports? I'm afraid that makes no > sense to me, to the extent that it didn't even occur to me to *look* in > the ARM TF sources to find any DTs for FVP until you pointed them out. > Agreed, I can talk to TF guys if required. > In other words, whilst people could discuss alternative ways to manage > DTs[1], I can't see any universe where ARM TF would be a logical place > to keep them. > > [1] ... and I'd further suggest that only perhaps people who are > prepared to put resources into fixing it should convene such a > discussion. While I agree with you, I am worried on the scalability. Models provide tons of options(may not be foundation platform but for sure the base AEM ones). It may so get unmanageable if we keep adding one DTS per configuration. If we are OK restricting the set of configurations to what you have proposed, then it seems fine. I will try to push this for v4.15, but I still want to hear more opinions on the scalability part here. -- Regards, Sudeep
On 02/10/17 18:26, Sudeep Holla wrote: > Sorry for late response, I thought I had sent this mail out long back > but was sitting in my draft :( No worries. I've been at Linaro connect this last week anyway. > On 20/09/17 12:17, Daniel Thompson wrote: >> On 20/09/17 10:42, Sudeep Holla wrote: >>> >>> >>> On 19/09/17 19:32, Daniel Thompson wrote: >>>> Currently if the Foundation model is running ARM Trusted Firmware then >>>> the kernel, which is configured to use spin tables, cannot start >>>> secondary >>>> processors or "power off" the simulation. >>>> >>>> After adding a couple of labels to the include file and splitting out >>>> the >>>> spin-table configuration into a header, we add a couple of new headers >>>> together with two new DTs (GICv2+PSCI and GICv3+PSCI). >>>> >>>> The new GICv3+PSCI DT has been boot tested, the remaining three (two of >>>> which existed prior to this patch) have been "tested" by decompiling the >>>> blobs and comparing them against a reference. >>>> >>> >>> How different are these from the ones hosted in [1] ? >> >> They look like they were either independently written or diverged a long >> time ago. The existing kernel DTs describe hardware absent from the ARM >> TF ones and vice versa. >> > > OK. > >> With specific reference to PSCI it looks like my patches could perhaps >> be improved by adding idle-state support. > > Yes I know. You want a v3 with it added? >>> On argument is that we want to take the DTS out of device tree as >>> firmware is responsible for generating them. Alternatively, we may be >>> duplicating resulting in discrepancies over time by coping it into >>> kernel. >> >> The general problem is copying from where? >> >> The kernel DTs are a well maintained centralized repository which is >> *really* useful. git grep across the kernel DTs is a hugely powerful >> tool when trying to better understand an ecosystem as sprawling and >> diverse as ARMs. In fact I've even seen those sort of searchs used as a >> basis to clean up unused code. Seeing that centralized repository >> splinter into separate per-vendor silos would be a huge loss for kernel >> developers. >> > > Agreed. But models are configurable and last time this discussion came > up, some argued that the DTs must be modified based on the configuration > automatically by models or some external scripts. Indeed. I can definitely understand why the *models* might want to be bundled with DTs (or a DT generator, or a --just-give-me-a-working-dt-for-my-command-line-options-option). >> <snip> >> In other words, whilst people could discuss alternative ways to manage >> DTs[1], I can't see any universe where ARM TF would be a logical place >> to keep them. >> >> [1] ... and I'd further suggest that only perhaps people who are >> prepared to put resources into fixing it should convene such a >> discussion. > > While I agree with you, I am worried on the scalability. Models provide > tons of options(may not be foundation platform but for sure the base AEM > ones). It may so get unmanageable if we keep adding one DTS per > configuration. > > If we are OK restricting the set of configurations to what you have > proposed, then it seems fine. > > I will try to push this for v4.15, but I still want to hear more > opinions on the scalability part here. I'm sure you mean other peoples opinions ;-) (especially so, since after reading things back, I realized I expressed mine rather more forefully than intended) but I think there is a reason why PSCI is special... basically its to do with the docs. I figured out how to use foundation model by reading the documentation. That documentation says "to run Linux go read the Linaro stuff". As it happens the Linaro "stuff" loops you through to a different ARM doc which tells you how to download ARM TF and bootloader binaries from Linaro. Thus if you follow the docs (and are patient enough) you come out the other side with a model running ARM TF and ready for PSCI. At that point you're ready for mainline kernel development... and you wonder why there is only one CPU running ;-). Daniel.
On 03/10/17 10:12, Daniel Thompson wrote: > On 02/10/17 18:26, Sudeep Holla wrote: >> Sorry for late response, I thought I had sent this mail out long back >> but was sitting in my draft :( > > No worries. I've been at Linaro connect this last week anyway. > > >> On 20/09/17 12:17, Daniel Thompson wrote: >>> On 20/09/17 10:42, Sudeep Holla wrote: >>>> >>>> >>>> On 19/09/17 19:32, Daniel Thompson wrote: >>>>> Currently if the Foundation model is running ARM Trusted Firmware then >>>>> the kernel, which is configured to use spin tables, cannot start >>>>> secondary >>>>> processors or "power off" the simulation. >>>>> >>>>> After adding a couple of labels to the include file and splitting out >>>>> the >>>>> spin-table configuration into a header, we add a couple of new headers >>>>> together with two new DTs (GICv2+PSCI and GICv3+PSCI). >>>>> >>>>> The new GICv3+PSCI DT has been boot tested, the remaining three >>>>> (two of >>>>> which existed prior to this patch) have been "tested" by >>>>> decompiling the >>>>> blobs and comparing them against a reference. >>>>> >>>> >>>> How different are these from the ones hosted in [1] ? >>> >>> They look like they were either independently written or diverged a long >>> time ago. The existing kernel DTs describe hardware absent from the ARM >>> TF ones and vice versa. >>> >> >> OK. >> >>> With specific reference to PSCI it looks like my patches could perhaps >>> be improved by adding idle-state support. >> >> Yes I know. > > You want a v3 with it added? > No, that's fine. I have pushed this already [1] with Mark's ack for now. We can add it later after some testing, not urgent. -- Regards, Sudeep [1] https://git.kernel.org/sudeep.holla/linux/h/for-next/juno
diff --git a/arch/arm64/boot/dts/arm/Makefile b/arch/arm64/boot/dts/arm/Makefile index 75cc2aa10101..25f82c377f67 100644 --- a/arch/arm64/boot/dts/arm/Makefile +++ b/arch/arm64/boot/dts/arm/Makefile @@ -1,4 +1,6 @@ -dtb-$(CONFIG_ARCH_VEXPRESS) += foundation-v8.dtb foundation-v8-gicv3.dtb +dtb-$(CONFIG_ARCH_VEXPRESS) += \ + foundation-v8.dtb foundation-v8-psci.dtb \ + foundation-v8-gicv3.dtb foundation-v8-gicv3-psci.dtb dtb-$(CONFIG_ARCH_VEXPRESS) += juno.dtb juno-r1.dtb juno-r2.dtb dtb-$(CONFIG_ARCH_VEXPRESS) += rtsm_ve-aemv8a.dtb dtb-$(CONFIG_ARCH_VEXPRESS) += vexpress-v2f-1xv7-ca53x2.dtb diff --git a/arch/arm64/boot/dts/arm/foundation-v8-gicv2.dtsi b/arch/arm64/boot/dts/arm/foundation-v8-gicv2.dtsi new file mode 100644 index 000000000000..851abf34fc80 --- /dev/null +++ b/arch/arm64/boot/dts/arm/foundation-v8-gicv2.dtsi @@ -0,0 +1,19 @@ +/* + * ARM Ltd. + * + * ARMv8 Foundation model DTS (GICv2 configuration) + */ + +/ { + gic: interrupt-controller@2c001000 { + compatible = "arm,cortex-a15-gic", "arm,cortex-a9-gic"; + #interrupt-cells = <3>; + #address-cells = <2>; + interrupt-controller; + reg = <0x0 0x2c001000 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/arm/foundation-v8-gicv3-psci.dts b/arch/arm64/boot/dts/arm/foundation-v8-gicv3-psci.dts new file mode 100644 index 000000000000..e096e670bec3 --- /dev/null +++ b/arch/arm64/boot/dts/arm/foundation-v8-gicv3-psci.dts @@ -0,0 +1,9 @@ +/* + * ARM Ltd. + * + * ARMv8 Foundation model DTS (GICv3+PSCI configuration) + */ + +#include "foundation-v8.dtsi" +#include "foundation-v8-gicv3.dtsi" +#include "foundation-v8-psci.dtsi" diff --git a/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dts b/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dts index 35588dfa095c..c5d834d7d0ba 100644 --- a/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dts +++ b/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dts @@ -5,26 +5,5 @@ */ #include "foundation-v8.dtsi" - -/ { - gic: interrupt-controller@2f000000 { - compatible = "arm,gic-v3"; - #interrupt-cells = <3>; - #address-cells = <2>; - #size-cells = <2>; - ranges; - interrupt-controller; - reg = <0x0 0x2f000000 0x0 0x10000>, - <0x0 0x2f100000 0x0 0x200000>, - <0x0 0x2c000000 0x0 0x2000>, - <0x0 0x2c010000 0x0 0x2000>, - <0x0 0x2c02f000 0x0 0x2000>; - interrupts = <1 9 4>; - - its: its@2f020000 { - compatible = "arm,gic-v3-its"; - msi-controller; - reg = <0x0 0x2f020000 0x0 0x20000>; - }; - }; -}; +#include "foundation-v8-gicv3.dtsi" +#include "foundation-v8-spin-table.dtsi" diff --git a/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dtsi b/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dtsi new file mode 100644 index 000000000000..91fc5c60d88b --- /dev/null +++ b/arch/arm64/boot/dts/arm/foundation-v8-gicv3.dtsi @@ -0,0 +1,28 @@ +/* + * ARM Ltd. + * + * ARMv8 Foundation model DTS (GICv3 configuration) + */ + +/ { + gic: interrupt-controller@2f000000 { + compatible = "arm,gic-v3"; + #interrupt-cells = <3>; + #address-cells = <2>; + #size-cells = <2>; + ranges; + interrupt-controller; + reg = <0x0 0x2f000000 0x0 0x10000>, + <0x0 0x2f100000 0x0 0x200000>, + <0x0 0x2c000000 0x0 0x2000>, + <0x0 0x2c010000 0x0 0x2000>, + <0x0 0x2c02f000 0x0 0x2000>; + interrupts = <1 9 4>; + + its: its@2f020000 { + compatible = "arm,gic-v3-its"; + msi-controller; + reg = <0x0 0x2f020000 0x0 0x20000>; + }; + }; +}; diff --git a/arch/arm64/boot/dts/arm/foundation-v8-psci.dts b/arch/arm64/boot/dts/arm/foundation-v8-psci.dts new file mode 100644 index 000000000000..723f23c7cd31 --- /dev/null +++ b/arch/arm64/boot/dts/arm/foundation-v8-psci.dts @@ -0,0 +1,9 @@ +/* + * ARM Ltd. + * + * ARMv8 Foundation model DTS (GICv2+PSCI configuration) + */ + +#include "foundation-v8.dtsi" +#include "foundation-v8-gicv2.dtsi" +#include "foundation-v8-psci.dtsi" diff --git a/arch/arm64/boot/dts/arm/foundation-v8-psci.dtsi b/arch/arm64/boot/dts/arm/foundation-v8-psci.dtsi new file mode 100644 index 000000000000..16cdf395728b --- /dev/null +++ b/arch/arm64/boot/dts/arm/foundation-v8-psci.dtsi @@ -0,0 +1,28 @@ +/* + * ARM Ltd. + * + * ARMv8 Foundation model DTS (PSCI configuration) + */ + +/ { + psci { + compatible = "arm,psci-1.0"; + method = "smc"; + }; +}; + +&cpu0 { + enable-method = "psci"; +}; + +&cpu1 { + enable-method = "psci"; +}; + +&cpu2 { + enable-method = "psci"; +}; + +&cpu3 { + enable-method = "psci"; +}; diff --git a/arch/arm64/boot/dts/arm/foundation-v8-spin-table.dtsi b/arch/arm64/boot/dts/arm/foundation-v8-spin-table.dtsi new file mode 100644 index 000000000000..4d4186ba0e8c --- /dev/null +++ b/arch/arm64/boot/dts/arm/foundation-v8-spin-table.dtsi @@ -0,0 +1,25 @@ +/* + * ARM Ltd. + * + * ARMv8 Foundation model DTS (spin table configuration) + */ + +&cpu0 { + enable-method = "spin-table"; + cpu-release-addr = <0x0 0x8000fff8>; +}; + +&cpu1 { + enable-method = "spin-table"; + cpu-release-addr = <0x0 0x8000fff8>; +}; + +&cpu2 { + enable-method = "spin-table"; + cpu-release-addr = <0x0 0x8000fff8>; +}; + +&cpu3 { + enable-method = "spin-table"; + cpu-release-addr = <0x0 0x8000fff8>; +}; diff --git a/arch/arm64/boot/dts/arm/foundation-v8.dts b/arch/arm64/boot/dts/arm/foundation-v8.dts index 71168077312d..8ff7c86fc929 100644 --- a/arch/arm64/boot/dts/arm/foundation-v8.dts +++ b/arch/arm64/boot/dts/arm/foundation-v8.dts @@ -5,17 +5,5 @@ */ #include "foundation-v8.dtsi" - -/ { - gic: interrupt-controller@2c001000 { - compatible = "arm,cortex-a15-gic", "arm,cortex-a9-gic"; - #interrupt-cells = <3>; - #address-cells = <2>; - interrupt-controller; - reg = <0x0 0x2c001000 0 0x1000>, - <0x0 0x2c002000 0 0x2000>, - <0x0 0x2c004000 0 0x2000>, - <0x0 0x2c006000 0 0x2000>; - interrupts = <1 9 0xf04>; - }; -}; +#include "foundation-v8-gicv2.dtsi" +#include "foundation-v8-spin-table.dtsi" diff --git a/arch/arm64/boot/dts/arm/foundation-v8.dtsi b/arch/arm64/boot/dts/arm/foundation-v8.dtsi index 8ecdd4331980..60f6ab920743 100644 --- a/arch/arm64/boot/dts/arm/foundation-v8.dtsi +++ b/arch/arm64/boot/dts/arm/foundation-v8.dtsi @@ -28,36 +28,28 @@ #address-cells = <2>; #size-cells = <0>; - cpu@0 { + cpu0: cpu@0 { device_type = "cpu"; compatible = "arm,armv8"; reg = <0x0 0x0>; - enable-method = "spin-table"; - cpu-release-addr = <0x0 0x8000fff8>; next-level-cache = <&L2_0>; }; - cpu@1 { + cpu1: cpu@1 { device_type = "cpu"; compatible = "arm,armv8"; reg = <0x0 0x1>; - enable-method = "spin-table"; - cpu-release-addr = <0x0 0x8000fff8>; next-level-cache = <&L2_0>; }; - cpu@2 { + cpu2: cpu@2 { device_type = "cpu"; compatible = "arm,armv8"; reg = <0x0 0x2>; - enable-method = "spin-table"; - cpu-release-addr = <0x0 0x8000fff8>; next-level-cache = <&L2_0>; }; - cpu@3 { + cpu3: cpu@3 { device_type = "cpu"; compatible = "arm,armv8"; reg = <0x0 0x3>; - enable-method = "spin-table"; - cpu-release-addr = <0x0 0x8000fff8>; next-level-cache = <&L2_0>; };