From: Chester Lin <hidden> Date: 2021-08-05 06:55:00
Hello,
Here I'd like to propose a patchset, which is initial upstream support for NXP
S32G2. S32G is a processor family developed by NXP for automotive solutions,
such as vehicle networking and automotive high-performance processing. This
series focuses on S32G2, which is the latest generation we can find at the
moment. As the first round to support S32G2, this patchset only enables basic
components and interfaces the SoC must have while kernel booting, which aims
to have minimum hardware enablement for these two boards, S32G-VNP-EVB and
S32G-VNP-RDB2. The concepts of how these boards work are originated from the
downstream kernel tree[1] developed by NXP, which provides lots of details
about the SoC S32G274A and its integrated boards. This series has been
verified with downstream ATF[2] & U-Boot[3] based on the ATF boot flow.
Thanks,
Chester
[1] https://source.codeaurora.org/external/autobsps32/linux/
[2] https://source.codeaurora.org/external/autobsps32/arm-trusted-firmware/
[3] https://source.codeaurora.org/external/autobsps32/u-boot/
Chester Lin (8):
dt-bindings: arm: fsl: add NXP S32G2 boards
dt-bindings: serial: fsl-linflexuart: convert to json-schema format
dt-bindings: serial: fsl-linflexuart: Add compatible for S32G2
arm64: dts: add NXP S32G2 support
arm64: dts: s32g2: add serial/uart support
arm64: dts: s32g2: add VNP-EVB and VNP-RDB2 support
arm64: dts: s32g2: add memory nodes for evb and rdb2
MAINTAINERS: Add an entry for NXP S32G2 boards
.../devicetree/bindings/arm/fsl.yaml | 7 +
.../bindings/serial/fsl,s32-linflexuart.txt | 22 ---
.../bindings/serial/fsl,s32-linflexuart.yaml | 66 +++++++++
MAINTAINERS | 6 +
arch/arm64/boot/dts/freescale/Makefile | 2 +
arch/arm64/boot/dts/freescale/s32g2.dtsi | 129 ++++++++++++++++++
.../arm64/boot/dts/freescale/s32g274a-evb.dts | 29 ++++
.../boot/dts/freescale/s32g274a-rdb2.dts | 33 +++++
8 files changed, 272 insertions(+), 22 deletions(-)
delete mode 100644 Documentation/devicetree/bindings/serial/fsl,s32-linflexuart.txt
create mode 100644 Documentation/devicetree/bindings/serial/fsl,s32-linflexuart.yaml
create mode 100644 arch/arm64/boot/dts/freescale/s32g2.dtsi
create mode 100644 arch/arm64/boot/dts/freescale/s32g274a-evb.dts
create mode 100644 arch/arm64/boot/dts/freescale/s32g274a-rdb2.dts
--
2.30.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2+-description:S32V234 based Boardsitems:-enum:
--
2.30.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)+%YAML1.2+---+$id:http://devicetree.org/schemas/serial/fsl,s32-linflexuart.yaml#+$schema:http://devicetree.org/meta-schemas/core.yaml#++title:Freescale LINFlexD UART++description:|+The LINFlexD controller implements several LIN protocol versions, as well+as support for full-duplex UART communication through 8-bit and 9-bit+frames. See chapter 47 ("LINFlexD") in the reference manual+https://www.nxp.com/webapp/Download?colCode=S32V234RM.++maintainers:+-Greg Kroah-Hartman <gregkh@linuxfoundation.org>+-Rob Herring <robh@kernel.org>++allOf:+-$ref:"serial.yaml"++properties:+compatible:+description:The LINFlexD controller on S32V234 SoC, which can be+configured in UART mode.+items:+-const:fsl,s32v234-linflexuart++reg:+maxItems:1++interrupts:+maxItems:1++required:+-compatible+-reg+-interrupts++unevaluatedProperties:false++examples:+-|+serial@40053000 {+compatible = "fsl,s32v234-linflexuart";+reg = <0x40053000 0x1000>;+interrupts = <0 59 4>;+};
--
2.30.0
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Chester Lin <hidden> Date: 2021-08-05 06:55:37
Add a compatible string for the uart binding of NXP S32G2 platforms. Here
we use "s32v234-linflexuart" as fallback since the current linflexuart
driver still works on S32G2.
Signed-off-by: Chester Lin <redacted>
---
.../bindings/serial/fsl,s32-linflexuart.yaml | 26 ++++++++++++++++---
1 file changed, 22 insertions(+), 4 deletions(-)
@@ -21,10 +21,20 @@ allOf:properties:compatible:-description:The LINFlexD controller on S32V234 SoC, which can be-configured in UART mode.-items:--const:fsl,s32v234-linflexuart+minItems:1+maxItems:2+oneOf:+-description:The LINFlexD controller on S32G2 SoC, which can be+configured in UART mode.+items:+-enum:+-fsl,s32g2-linflexuart+-const:fsl,s32v234-linflexuart++-description:The LINFlexD controller on S32V234 SoC, which can be+configured in UART mode.+items:+-const:fsl,s32v234-linflexuartreg:maxItems:1
From: Chester Lin <hidden> Date: 2021-08-05 06:56:19
Add a new entry for the maintenance of NXP S32G2 DT files.
Signed-off-by: Chester Lin <redacted>
---
MAINTAINERS | 6 ++++++
1 file changed, 6 insertions(+)
From: Krzysztof Kozlowski <krzk@kernel.org> Date: 2021-08-05 07:50:00
On 05/08/2021 08:54, Chester Lin wrote:
quoted hunk
Add a new entry for the maintenance of NXP S32G2 DT files.
Signed-off-by: Chester Lin <redacted>
---
MAINTAINERS | 6 ++++++
1 file changed, 6 insertions(+)
On Thu, Aug 05, 2021 at 09:49:51AM +0200, Krzysztof Kozlowski wrote:
On 05/08/2021 08:54, Chester Lin wrote:
quoted
Add a new entry for the maintenance of NXP S32G2 DT files.
Signed-off-by: Chester Lin <redacted>
---
MAINTAINERS | 6 ++++++
1 file changed, 6 insertions(+)
Thanks for reminding that the patch didn't land. I just resent it with
your Reviewed-by tag added. Thanks!
Shawn
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
On Thu, Aug 05, 2021 at 02:54:21PM +0800, Chester Lin wrote:
Hello,
Here I'd like to propose a patchset, which is initial upstream support for NXP
S32G2. S32G is a processor family developed by NXP for automotive solutions,
such as vehicle networking and automotive high-performance processing. This
series focuses on S32G2, which is the latest generation we can find at the
moment. As the first round to support S32G2, this patchset only enables basic
components and interfaces the SoC must have while kernel booting, which aims
to have minimum hardware enablement for these two boards, S32G-VNP-EVB and
S32G-VNP-RDB2. The concepts of how these boards work are originated from the
downstream kernel tree[1] developed by NXP, which provides lots of details
about the SoC S32G274A and its integrated boards. This series has been
verified with downstream ATF[2] & U-Boot[3] based on the ATF boot flow.
Thanks,
Chester
[1] https://source.codeaurora.org/external/autobsps32/linux/
[2] https://source.codeaurora.org/external/autobsps32/arm-trusted-firmware/
[3] https://source.codeaurora.org/external/autobsps32/u-boot/
Chester Lin (8):
dt-bindings: arm: fsl: add NXP S32G2 boards
dt-bindings: serial: fsl-linflexuart: convert to json-schema format
dt-bindings: serial: fsl-linflexuart: Add compatible for S32G2
arm64: dts: add NXP S32G2 support
arm64: dts: s32g2: add serial/uart support
arm64: dts: s32g2: add VNP-EVB and VNP-RDB2 support
arm64: dts: s32g2: add memory nodes for evb and rdb2
The dts changes look good to me. I will pick up the series once
bindings gets acked by Rob.
Shawn
From: Andreas Färber <afaerber@suse.de> Date: 2021-08-12 15:30:43
Hello Shawn and Krzysztof,
On 09.08.21 10:03, Shawn Guo wrote:
On Thu, Aug 05, 2021 at 09:49:51AM +0200, Krzysztof Kozlowski wrote:
quoted
On 05/08/2021 08:54, Chester Lin wrote:
quoted
Add a new entry for the maintenance of NXP S32G2 DT files.
Signed-off-by: Chester Lin <redacted>
---
MAINTAINERS | 6 ++++++
1 file changed, 6 insertions(+)
I had specifically asked Chester to add a MAINTAINERS section.
Is your apparent suggestion of not accepting this MAINTAINERS patch
based on the assumption that we're dealing with only one email address
in three files? What do you see as the threshold to warrant a section?
From my point of view, above MAINTAINERS entry is incomplete, as we
should CC the full team working on S32G for patch review, not just
Chester himself.
So that would in my mind have been additional R: and L: entries in that
MAINTAINERS section.
Thanks for reminding that the patch didn't land. I just resent it with
your Reviewed-by tag added. Thanks!
Your above patch does not make clear to me what syntax we should use for
adding email addresses to .dts[i] files now:
https://lore.kernel.org/lkml/20210809081033.GQ30984@dragon/
Especially when not dealing with file authors.
I get the impression it is not a replacement for an F: wildcard used in
MAINTAINERS, but rather a complement?
Please understand that this is not about a single .dts file, as your
patch suggests, but about a complete SoC family consisting of s32g*.dts*
as well as in the future drivers specific to this platform. It seems way
easier to specify the list of maintainers/reviewers in MAINTAINERS once
with suitable wildcard paths, than copying them into each and every
.dtsi and .dts file and driver .c/.h and later needing to sync multiple
places. If a bot or user has fixes or cleanups for the S32G code, we
want to know about it, so that NXP can consider it for their BSP
branches and SUSE for our SLE branches, and obviously for follow-up
patch series that are already in the works and waiting on this one.
Once merged, I would expect Chester or someone from NXP to set up an
S32G tree on kernel.org that gets integrated into linux-next and sends
pull requests to the SoC tree maintainers without bothering i.MX and
Layerscape maintainers. Did you handle that differently for S32V?
Thanks,
Andreas
P.S. Have you checked or considered whether your script change might
start to CC non-existing email addresses, since we wouldn't be allowed
to remove them from copyright or authorship statements to prevent that?
--
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer
HRB 36809 (AG Nürnberg)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2
@Rob: Should for new boards the description: syntax be used also for
enums? Or just at SoC level, and for board enums still traditional #
comments?
+ - const: fsl,s32g2
@NXP: Is it sufficient here to have s32g2, or should we call this
s32g274a and adjust the description above to S32G274A?
Related, is the trailing A for Arm, like for the Layerscape chips? I.e.,
not for Alpha or rev.A or something that will change for non-eval chips?
+
- description: S32V234 based Boards
items:
- enum:
Otherwise,
Reviewed-by: Andreas Färber <afaerber@suse.de>
Thanks,
Andreas
--
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer
HRB 36809 (AG Nürnberg)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Krzysztof Kozlowski <krzk@kernel.org> Date: 2021-08-12 15:54:33
On 12/08/2021 17:30, Andreas Färber wrote:
Hello Shawn and Krzysztof,
On 09.08.21 10:03, Shawn Guo wrote:
quoted
On Thu, Aug 05, 2021 at 09:49:51AM +0200, Krzysztof Kozlowski wrote:
quoted
On 05/08/2021 08:54, Chester Lin wrote:
quoted
Add a new entry for the maintenance of NXP S32G2 DT files.
Signed-off-by: Chester Lin <redacted>
---
MAINTAINERS | 6 ++++++
1 file changed, 6 insertions(+)
I had specifically asked Chester to add a MAINTAINERS section.
Is your apparent suggestion of not accepting this MAINTAINERS patch
based on the assumption that we're dealing with only one email address
in three files? What do you see as the threshold to warrant a section?
From my point of view, above MAINTAINERS entry is incomplete, as we
should CC the full team working on S32G for patch review, not just
Chester himself.
So that would in my mind have been additional R: and L: entries in that
MAINTAINERS section.
I assumed this is a sub-sub-architecture (something coming from
Layerscape or i.MX) but it seems it's not, therefore I don't mind having
separate entry in MAINTAINERS. The idea of in-file maintainers was for
specific boards and SoCs belonging to existing sub-architectures.
I agree with your following reasons.
quoted
Thanks for reminding that the patch didn't land. I just resent it with
your Reviewed-by tag added. Thanks!
Your above patch does not make clear to me what syntax we should use for
adding email addresses to .dts[i] files now:
https://lore.kernel.org/lkml/20210809081033.GQ30984@dragon/
Especially when not dealing with file authors.
I get the impression it is not a replacement for an F: wildcard used in
MAINTAINERS, but rather a complement?
Please understand that this is not about a single .dts file, as your
patch suggests, but about a complete SoC family consisting of s32g*.dts*
as well as in the future drivers specific to this platform. It seems way
easier to specify the list of maintainers/reviewers in MAINTAINERS once
with suitable wildcard paths, than copying them into each and every
.dtsi and .dts file and driver .c/.h and later needing to sync multiple
places. If a bot or user has fixes or cleanups for the S32G code, we
want to know about it, so that NXP can consider it for their BSP
branches and SUSE for our SLE branches, and obviously for follow-up
patch series that are already in the works and waiting on this one.
Once merged, I would expect Chester or someone from NXP to set up an
S32G tree on kernel.org that gets integrated into linux-next and sends
pull requests to the SoC tree maintainers without bothering i.MX and
Layerscape maintainers. Did you handle that differently for S32V?
Thanks,
Andreas
P.S. Have you checked or considered whether your script change might
start to CC non-existing email addresses, since we wouldn't be allowed
to remove them from copyright or authorship statements to prevent that?
The same can happen for DT bindings maintainers.
Best regards,
Krzysztof
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
Since this is dual-licensed and BSD-2-Clause would seem compatible with
GPLv3, this could probably be s/GPL-2.0-only/GPL-2.0-or-later/ ? Not a
blocker.
From: Andreas Färber <afaerber@suse.de> Date: 2021-08-12 16:28:05
On 05.08.21 08:54, Chester Lin wrote:
quoted hunk
Add a compatible string for the uart binding of NXP S32G2 platforms. Here
we use "s32v234-linflexuart" as fallback since the current linflexuart
driver still works on S32G2.
Signed-off-by: Chester Lin <redacted>
---
.../bindings/serial/fsl,s32-linflexuart.yaml | 26 ++++++++++++++++---
1 file changed, 22 insertions(+), 4 deletions(-)
@@ -21,10 +21,20 @@ allOf:properties:compatible:-description:The LINFlexD controller on S32V234 SoC, which can be-configured in UART mode.-items:--const:fsl,s32v234-linflexuart+minItems:1+maxItems:2
Are these necessary for oneOf?
+ oneOf:
+ - description: The LINFlexD controller on S32G2 SoC, which can be
+ configured in UART mode.
+ items:
+ - enum:
+ - fsl,s32g2-linflexuart
+ - const: fsl,s32v234-linflexuart
This reads inconsistent to me: Either this oneOf is for S32G2 only, then
please turn the enum into a const. Or change the description to "on SoCs
compatible with S32V234" if we expect the enum list to grow.
I believe the idea here was to avoid unnecessary driver compatible and
earlycon compatible additions, while preparing for eventual quirks
specific to S32G2.
@NXP: Should this be s32g2- like above or s32g274a- specifically? Do you
agree this is a useful thing to prepare here, as opposed to using only
s32v234- in the s32g2* DT?
I assume the ordering is done alphabetically as S32G < S32V;
alternatively we might order S32V234 first and then the compatible ones.
+
+ - description: The LINFlexD controller on S32V234 SoC, which can be
+ configured in UART mode.
+ items:
+ - const: fsl,s32v234-linflexuart
To minimize this S32G2 patch, would it be valid to do oneOf for the
single S32V in the preceding patch already? Then we would avoid the text
movement and re-indentation above and more easily see the lines newly
getting added for S32G2.
@@ -0,0 +1,98 @@+// SPDX-License-Identifier: GPL-2.0-or-later OR MIT+/*
* NXP S32G2 SoC family
*
?
@NXP: Are any models other than 274A in the queue that we should
distinguish between s32g2.dtsi and s32g274a.dtsi here already?
Why? Does it have any peripherals that go beyond 32-bit space?
For 64-bit Realtek platforms Rob had asked me to use 1, if possible.
I do understand that for /memory nodes we do have high-memory addresses,
so 2 for the root node looks correct.
Please order reg after compatible by convention, and sort
interrupt-controller or at least #interrupt-cells (applying to
consumers) last, after the below one applying to this device itself.
From: Andreas Färber <afaerber@suse.de> Date: 2021-08-12 17:42:46
Hi Chester et al.,
On 05.08.21 08:54, Chester Lin wrote:
Add serial/uart support for NXP S32G2.
You might mention here that (following our initial stub) this commit is
now apparently based on the CodeAurora BSP branch foo (and therefore
adding its last-year copyright below and separate from 4/8).
@NXP: If there are downstream Signed-off-bys that you would like to see
included for this portion here, please speak up.
Note: In the past there had been controversies as to whether to define
aliases globally for a SoC or in a .dts specific to a board's usage.
In this case it does not seem to matter much, as uart0 is being used as
console on the reference boards.
Is there any marking on the PCB that a /* ... */ comment could reference
here? Right now uart0 is close to stdout-path above, but that will
change once more device nodes get added and enabled alphabetically.
What is uart1? Please add a comment. Does it actually work without
clocks property?
Thanks,
Andreas
--
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer
HRB 36809 (AG Nürnberg)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Andreas Färber <afaerber@suse.de> Date: 2021-08-12 18:25:06
Hi Chester et al.,
On 05.08.21 08:54, Chester Lin wrote:
Add memory nodes for S32G-VNP-EVB and S32G-VNP-RDB2 since they have fixed
RAM size.
You can drop "since they have fixed RAM size" - if they didn't, you
would simply choose the lowest size and rely on the bootloader (U-Boot)
to overwrite it with the actually detected size.
Please expand why this patch is separate - BSP based, I assume?
This looks strange to me - either put /* 4 GiB RAM */ before the node,
three lines above, and/or append comment /* 2 GiB */ on each line below.
Note the space, and suggest to be precise about factor 1024 vs. 1000.
+ reg = <0 0x80000000 0 0x80000000>,
Note that this gives you the range to use for the .dtsi /soc node:
Address 0x0 with size 0x80000000 gets mapped to 0x0 0x0, excluding the
upper 0x80000000 for the RAM here. Or address 0x0 0x0 for two /soc cells
if there are high-memory peripherals.
+ <8 0x80000000 0 0x80000000>;
Maybe use 0x8 here and 0x0 above? (second 0 stays same, so don't mind)
Actually I traced the pmu_parse_irqs() and found that this SoC never falls into
the pmu_has_irq_affinity() check because it's a percpu IRQ so the flow ends at
pmu_parse_percpu_irq(). But it looks good to me to have an interrupt-affinity
to indicate that each core has an associated PPI for PMU events as the binding
file has suggested.
Why? Does it have any peripherals that go beyond 32-bit space?
For 64-bit Realtek platforms Rob had asked me to use 1, if possible.
I do understand that for /memory nodes we do have high-memory addresses,
so 2 for the root node looks correct.
Actually it's a limitation due to [PATCH 7/8] "arm64: dts: s32g2: add memory
nodes for evb and rdb2", which adds memory nodes to indicate maximum system
RAM size combined by two separated memory banks, and the second bank starts
from the offset 0x880000000 so that's why we need 64-bit address space here.
Please feel free to let me know if any suggestion.
I added memory nodes in [PATCH 7/8] "arm64: dts: s32g2: add memory nodes for
evb and rdb2" to describe maximum RAM areas, which are based on the information
we found in NXP BSP and boards.
@NXP: Please feel free to correct me if anything wrong, thanks.
Please order reg after compatible by convention, and sort
interrupt-controller or at least #interrupt-cells (applying to
consumers) last, after the below one applying to this device itself.
Actually I traced the pmu_parse_irqs() and found that this SoC never falls into
the pmu_has_irq_affinity() check because it's a percpu IRQ so the flow ends at
pmu_parse_percpu_irq(). But it looks good to me to have an interrupt-affinity
to indicate that each core has an associated PPI for PMU events as the binding
file has suggested.
My comment was based on the DT binding requesting it also for PPI and
previous review requests for me to add it elsewhere, so it's quite
possible that your code analysis is correct.
FWIW also the cache nodes were not evaluated last time I checked,
although the specs do show those nodes and properties.
https://github.com/devicetree-org/devicetree-specification/blob/v0.3/source/devicenodes.rst
quoted
quoted
+ };
[...]
quoted
quoted
+
+ soc {
+ compatible = "simple-bus";
[...]
quoted
quoted
+ #address-cells = <2>;
+ #size-cells = <2>;
Why? Does it have any peripherals that go beyond 32-bit space?
For 64-bit Realtek platforms Rob had asked me to use 1, if possible.
I do understand that for /memory nodes we do have high-memory addresses,
so 2 for the root node looks correct.
Actually it's a limitation due to [PATCH 7/8] "arm64: dts: s32g2: add memory
nodes for evb and rdb2", which adds memory nodes to indicate maximum system
RAM size combined by two separated memory banks, and the second bank starts
from the offset 0x880000000 so that's why we need 64-bit address space here.
Please feel free to let me know if any suggestion.
The /memory nodes are on the root node and thus only use #address-cells
and #size-cells from /, not from /soc here. Thus the only criterium is
whether something within /soc needs it.
PCI controllers may have 3 address cells, SPI/I2C controllers 1, so you
can have multiple sizes within one DT, you just need suitable ranges
wherever mappings to the parent address space are needed.
(I thought you had that distinction in an earlier draft I reviewed.)
Similarly, /reserved-memory would be outside the /soc node and use the
root node's #(address|size)-cells.
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/reserved-memory/reserved-memory.txt
Note: I'm assuming that if someone were to add a DT for the Cortex-M7s,
the address space mapping of any shared peripherals will likely differ,
as would the interrupts obviously, so I assume no direct .dtsi sharing
of /soc sub-nodes could sensibly be applied here.
@NXP: Please correct me if I'm wrong in either M7 not having access to
DDR RAM or large enough SRAM for Linux or another DT-consuming OS (and
this line of thinking not being applicable then) or you having use cases
to factor out some of the /soc sub-nodes for sharing between the two
(which would then give another reason for #address-/size-cells = <1> for
32-bit Cortex-M7).
[...]
I added memory nodes in [PATCH 7/8] "arm64: dts: s32g2: add memory nodes for
evb and rdb2" to describe maximum RAM areas, which are based on the information
we found in NXP BSP and boards.
Please order reg after compatible by convention, and sort
interrupt-controller or at least #interrupt-cells (applying to
consumers) last, after the below one applying to this device itself.
Is there any marking on the PCB that a /* ... */ comment could reference
here? Right now uart0 is close to stdout-path above, but that will
change once more device nodes get added and enabled alphabetically.
Yes, it's marked as "UART J58". Will add it in v2.
quoted
+&uart0 {
+ status = "okay";
No clock-frequency or clocks property needed?
As a minimum support patch, it can still reply on TF-A's clock settings so no
clock property is required at the moment.
From: Radu Nicolae Pirea (NXP OSS) <hidden> Date: 2021-08-13 09:55:09
Hi Andreas
On Thu, 2021-08-12 at 19:42 +0200, Andreas Färber wrote:
Hi Chester et al.,
On 05.08.21 08:54, Chester Lin wrote:
quoted
Add serial/uart support for NXP S32G2.
You might mention here that (following our initial stub) this commit
is
now apparently based on the CodeAurora BSP branch foo (and therefore
adding its last-year copyright below and separate from 4/8).
quoted
@NXP: If there are downstream Signed-off-bys that you would like to
see
included for this portion here, please speak up.
Larisa signed-off should be added.
Signed-off-by: Larisa Grigore <redacted>
Note: In the past there had been controversies as to whether to
define
aliases globally for a SoC or in a .dts specific to a board's usage.
In this case it does not seem to matter much, as uart0 is being used
as
console on the reference boards.
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
Since this is dual-licensed and BSD-2-Clause would seem compatible with
GPLv3, this could probably be s/GPL-2.0-only/GPL-2.0-or-later/ ? Not a
blocker.
There's no license identifier in the original file so it's not a problem to
choose "GPL-2.0-or-later".
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
Since this is dual-licensed and BSD-2-Clause would seem compatible with
GPLv3, this could probably be s/GPL-2.0-only/GPL-2.0-or-later/ ? Not a
blocker.
There's no license identifier in the original file so it's not a problem to
choose "GPL-2.0-or-later".
That is not entirely correct. If there is no explicit license in the
file, it's kernel's default so GPL-2.0-only. You cannot relicense
derivative work without getting permission from authors and copyright
holders.
However if you did not copy the text/description from original bindings,
your work won't be derivative so you can relicense it.
Best regards,
Krzysztof
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
Since this is dual-licensed and BSD-2-Clause would seem compatible with
GPLv3, this could probably be s/GPL-2.0-only/GPL-2.0-or-later/ ? Not a
blocker.
There's no license identifier in the original file so it's not a problem to
choose "GPL-2.0-or-later".
That is not entirely correct. If there is no explicit license in the
file, it's kernel's default so GPL-2.0-only. You cannot relicense
derivative work without getting permission from authors and copyright
holders.
However if you did not copy the text/description from original bindings,
your work won't be derivative so you can relicense it.
Sorry for my misunderstanding and thanks for the reminder.
Regards,
Chester
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Radu Nicolae Pirea (NXP OSS) <hidden> Date: 2021-08-13 14:27:58
On Thu, 2021-08-12 at 18:27 +0200, Andreas Färber wrote:
On 05.08.21 08:54, Chester Lin wrote:
quoted
Add a compatible string for the uart binding of NXP S32G2
platforms. Here
we use "s32v234-linflexuart" as fallback since the current
linflexuart
driver still works on S32G2.
Signed-off-by: Chester Lin <redacted>
---
.../bindings/serial/fsl,s32-linflexuart.yaml | 26
++++++++++++++++---
1 file changed, 22 insertions(+), 4 deletions(-)
properties:
compatible:
- description: The LINFlexD controller on S32V234 SoC, which can
be
- configured in UART mode.
- items:
- - const: fsl,s32v234-linflexuart
+ minItems: 1
+ maxItems: 2
Are these necessary for oneOf?
quoted
+ oneOf:
+ - description: The LINFlexD controller on S32G2 SoC, which
can be
+ configured in UART mode.
+ items:
+ - enum:
+ - fsl,s32g2-linflexuart
+ - const: fsl,s32v234-linflexuart
This reads inconsistent to me: Either this oneOf is for S32G2 only,
then
please turn the enum into a const. Or change the description to "on
SoCs
compatible with S32V234" if we expect the enum list to grow.
I believe the idea here was to avoid unnecessary driver compatible
and
earlycon compatible additions, while preparing for eventual quirks
specific to S32G2.
@NXP: Should this be s32g2- like above or s32g274a- specifically? Do
you
agree this is a useful thing to prepare here, as opposed to using
only
s32v234- in the s32g2* DT?
s32g2- is fine, but the vendor should be nxp, not fsl.
nxp,s32g2-linflexuart
I assume the ordering is done alphabetically as S32G < S32V;
alternatively we might order S32V234 first and then the compatible
ones.
quoted
+
+ - description: The LINFlexD controller on S32V234 SoC, which
can be
+ configured in UART mode.
+ items:
+ - const: fsl,s32v234-linflexuart
To minimize this S32G2 patch, would it be valid to do oneOf for the
single S32V in the preceding patch already? Then we would avoid the
text
movement and re-indentation above and more easily see the lines newly
getting added for S32G2.
From: Chester Lin <hidden> Date: 2021-08-13 14:59:00
On Thu, Aug 12, 2021 at 08:25:00PM +0200, Andreas Färber wrote:
Hi Chester et al.,
On 05.08.21 08:54, Chester Lin wrote:
quoted
Add memory nodes for S32G-VNP-EVB and S32G-VNP-RDB2 since they have fixed
RAM size.
You can drop "since they have fixed RAM size" - if they didn't, you
would simply choose the lowest size and rely on the bootloader (U-Boot)
to overwrite it with the actually detected size.
Please expand why this patch is separate - BSP based, I assume?
Yes, the information of memory banks is from s32 downstream kernel, which is
listed in the board DTs of older releases [bsp27] although newer releases
[bsp28 & bsp29] have moved it into s32 downstream u-boot as runtime fdt-fixup.
I think we should still reveal this information in kernel DTs in order to have
better understanding of system memory ranges that each board can have.
@NXP: Please don't hesitate to let me know if any better idea, thanks!
This looks strange to me - either put /* 4 GiB RAM */ before the node,
three lines above, and/or append comment /* 2 GiB */ on each line below.
Note the space, and suggest to be precise about factor 1024 vs. 1000.
Thank you for the suggestion.
quoted
+ reg = <0 0x80000000 0 0x80000000>,
Note that this gives you the range to use for the .dtsi /soc node:
Address 0x0 with size 0x80000000 gets mapped to 0x0 0x0, excluding the
upper 0x80000000 for the RAM here. Or address 0x0 0x0 for two /soc cells
if there are high-memory peripherals.
I don't know if memory ranges might be changed for new boards or CPU revisions
in the future, which means the first memory range might be expanded as well
[e.g. 0~4GB]. Based this assumption, I think the size should also be changed
accordingly. Not sure if overlays can still work with this case but overwriting
all reg properties under /soc could be awful.
However if we only have to think of current hardware spec, it's good to declare
"range = <0 0 0 0x80000000>".
Please feel free to let me know if any suggestions, thanks.
quoted
+ <8 0x80000000 0 0x80000000>;
Maybe use 0x8 here and 0x0 above? (second 0 stays same, so don't mind)
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2
Given this is an entirely different family from i.MX and new?, shouldn't
it use 'nxp' instead of 'fsl'? Either way,
Acked-by: Rob Herring <robh@kernel.org>
Rob
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
Since this is dual-licensed and BSD-2-Clause would seem compatible with
GPLv3, this could probably be s/GPL-2.0-only/GPL-2.0-or-later/ ? Not a
blocker.
There's no license identifier in the original file so it's not a problem to
choose "GPL-2.0-or-later".
That is not entirely correct. If there is no explicit license in the
file, it's kernel's default so GPL-2.0-only. You cannot relicense
derivative work without getting permission from authors and copyright
holders.
However if you did not copy the text/description from original bindings,
your work won't be derivative so you can relicense it.
Sorry for my misunderstanding and thanks for the reminder.
TBC, please try and get permission to relicense. The single author was
from NXP so changing the license here just needs the okay from any NXP
employee.
But use 'GPL-2.0-only' not 'GPL-2.0-or-later' please. As gregkh says,
are you okay with GPLv4, v5, etc.?
Rob
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -1,22 +0,0 @@-* Freescale LINFlexD UART--The LINFlexD controller implements several LIN protocol versions, as well as-support for full-duplex UART communication through 8-bit and 9-bit frames.--See chapter 47 ("LINFlexD") in the reference manual[1].--Required properties:-- compatible :- - "fsl,s32v234-linflexuart" for LINFlexD configured in UART mode, which- is compatible with the one integrated on S32V234 SoC-- reg : Address and length of the register set for the device-- interrupts : Should contain uart interrupt--Example:-uart0: serial@40053000 {- compatible = "fsl,s32v234-linflexuart";- reg = <0x0 0x40053000 0x0 0x1000>;- interrupts = <0 59 4>;-};--[1] https://www.nxp.com/webapp/Download?colCode=S32V234RM
@@ -0,0 +1,48 @@+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)+%YAML1.2+---+$id:http://devicetree.org/schemas/serial/fsl,s32-linflexuart.yaml#+$schema:http://devicetree.org/meta-schemas/core.yaml#++title:Freescale LINFlexD UART++description:|+The LINFlexD controller implements several LIN protocol versions, as well+as support for full-duplex UART communication through 8-bit and 9-bit+frames. See chapter 47 ("LINFlexD") in the reference manual+https://www.nxp.com/webapp/Download?colCode=S32V234RM.++maintainers:+-Greg Kroah-Hartman <gregkh@linuxfoundation.org>+-Rob Herring <robh@kernel.org>
Someone that cares about this h/w, not who applies patches.
+
+allOf:
+ - $ref: "serial.yaml"
+
+properties:
+ compatible:
+ description: The LINFlexD controller on S32V234 SoC, which can be
+ configured in UART mode.
Drop 'on S32V234 SoC' so we're not editting for every new SoC.
From: Rob Herring <robh@kernel.org> Date: 2021-08-13 18:09:23
On Thu, Aug 05, 2021 at 02:54:24PM +0800, Chester Lin wrote:
quoted hunk
Add a compatible string for the uart binding of NXP S32G2 platforms. Here
we use "s32v234-linflexuart" as fallback since the current linflexuart
driver still works on S32G2.
Signed-off-by: Chester Lin <redacted>
---
.../bindings/serial/fsl,s32-linflexuart.yaml | 26 ++++++++++++++++---
1 file changed, 22 insertions(+), 4 deletions(-)
@@ -21,10 +21,20 @@ allOf:properties:compatible:-description:The LINFlexD controller on S32V234 SoC, which can be-configured in UART mode.-items:--const:fsl,s32v234-linflexuart+minItems:1+maxItems:2
minItems/maxItems not needed here.
quoted hunk
+ oneOf:
+ - description: The LINFlexD controller on S32G2 SoC, which can be
+ configured in UART mode.
+ items:
+ - enum:
+ - fsl,s32g2-linflexuart
+ - const: fsl,s32v234-linflexuart
+
+ - description: The LINFlexD controller on S32V234 SoC, which can be
+ configured in UART mode.
+ items:
+ - const: fsl,s32v234-linflexuart
reg:
maxItems: 1
This doesn't really warrant another example just for a new compatible
string.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Rob Herring <robh@kernel.org> Date: 2021-08-13 18:11:18
On Thu, Aug 12, 2021 at 06:27:57PM +0200, Andreas Färber wrote:
On 05.08.21 08:54, Chester Lin wrote:
quoted
Add a compatible string for the uart binding of NXP S32G2 platforms. Here
we use "s32v234-linflexuart" as fallback since the current linflexuart
driver still works on S32G2.
Signed-off-by: Chester Lin <redacted>
---
.../bindings/serial/fsl,s32-linflexuart.yaml | 26 ++++++++++++++++---
1 file changed, 22 insertions(+), 4 deletions(-)
@@ -21,10 +21,20 @@ allOf:properties:compatible:-description:The LINFlexD controller on S32V234 SoC, which can be-configured in UART mode.-items:--const:fsl,s32v234-linflexuart+minItems:1+maxItems:2
Are these necessary for oneOf?
quoted
+ oneOf:
+ - description: The LINFlexD controller on S32G2 SoC, which can be
+ configured in UART mode.
+ items:
+ - enum:
+ - fsl,s32g2-linflexuart
+ - const: fsl,s32v234-linflexuart
This reads inconsistent to me: Either this oneOf is for S32G2 only, then
please turn the enum into a const. Or change the description to "on SoCs
compatible with S32V234" if we expect the enum list to grow.
I believe the idea here was to avoid unnecessary driver compatible and
earlycon compatible additions, while preparing for eventual quirks
specific to S32G2.
@NXP: Should this be s32g2- like above or s32g274a- specifically? Do you
agree this is a useful thing to prepare here, as opposed to using only
s32v234- in the s32g2* DT?
I assume the ordering is done alphabetically as S32G < S32V;
alternatively we might order S32V234 first and then the compatible ones.
quoted
+
+ - description: The LINFlexD controller on S32V234 SoC, which can be
+ configured in UART mode.
+ items:
+ - const: fsl,s32v234-linflexuart
To minimize this S32G2 patch, would it be valid to do oneOf for the
single S32V in the preceding patch already? Then we would avoid the text
movement and re-indentation above and more easily see the lines newly
getting added for S32G2.
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2
Given this is an entirely different family from i.MX and new?, shouldn't
it use 'nxp' instead of 'fsl'? Either way,
It sounds good and Radu from NXP has mentioned a similar idea for the
compatible string of linflexuart. To keep the naming consistency, should we
change all 'fsl' to 'nxp' as well? For example, we could rename the fsl.yaml
to nxp.yaml. However, changing all of them would cause some impacts, which will
need more verifications on new strings. Otherwise we would have to tolerate the
naming differences only used by s32g2.
Thanks,
Chester
That's enough redistributor space for 16 CPUs. However, you only
describe 4. Either the number of CPUs is wrong, the size is wrong, or
the GIC has been configured for more cores than the SoC has.
Please order reg after compatible by convention, and sort
interrupt-controller or at least #interrupt-cells (applying to
consumers) last, after the below one applying to this device itself.
CC'ing Marc for additional GIC scrutiny, often the sizes are wrong.
There is more than just sizes. The interrupt specifier for the
maintenance interrupt is also wrong.
M.
--
Without deviation from the norm, progress is not possible.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
That's enough redistributor space for 16 CPUs. However, you only
describe 4. Either the number of CPUs is wrong, the size is wrong, or
the GIC has been configured for more cores than the SoC has.
Confirmed the SoC can only find 4 redistributors:
localhost:~ # dmesg | grep CPU
[ 0.000000] Booting Linux on physical CPU 0x0000000000 [0x410fd034]
[ 0.000000] Detected VIPT I-cache on CPU0
[ 0.000000] CPU features: detected: GIC system register CPU interface
[ 0.000000] CPU features: detected: ARM erratum 845719
[ 0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=4, Nodes=1
[ 0.000000] rcu: RCU restricting CPUs from NR_CPUS=480 to nr_cpu_ids=4.
[ 0.000000] GICv3: CPU0: found redistributor 0 region 0:0x0000000050880000
[ 0.063865] smp: Bringing up secondary CPUs ...
[ 0.068852] Detected VIPT I-cache on CPU1
[ 0.068894] GICv3: CPU1: found redistributor 1 region 0:0x00000000508a0000
[ 0.068963] CPU1: Booted secondary processor 0x0000000001 [0x410fd034]
[ 0.069809] Detected VIPT I-cache on CPU2
[ 0.069851] GICv3: CPU2: found redistributor 100 region 0:0x00000000508c0000
[ 0.069903] CPU2: Booted secondary processor 0x0000000100 [0x410fd034]
[ 0.070698] Detected VIPT I-cache on CPU3
[ 0.070722] GICv3: CPU3: found redistributor 101 region 0:0x00000000508e0000
[ 0.070749] CPU3: Booted secondary processor 0x0000000101 [0x410fd034]
[ 0.070847] smp: Brought up 1 node, 4 CPUs
<..snip..>
I will correct the size to 0x80000, thanks!
Please order reg after compatible by convention, and sort
interrupt-controller or at least #interrupt-cells (applying to
consumers) last, after the below one applying to this device itself.
That's enough redistributor space for 16 CPUs. However, you only
describe 4. Either the number of CPUs is wrong, the size is wrong, or
the GIC has been configured for more cores than the SoC has.
Confirmed the SoC can only find 4 redistributors:
localhost:~ # dmesg | grep CPU
[ 0.000000] Booting Linux on physical CPU 0x0000000000 [0x410fd034]
[ 0.000000] Detected VIPT I-cache on CPU0
[ 0.000000] CPU features: detected: GIC system register CPU interface
[ 0.000000] CPU features: detected: ARM erratum 845719
[ 0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=4, Nodes=1
[ 0.000000] rcu: RCU restricting CPUs from NR_CPUS=480 to nr_cpu_ids=4.
[ 0.000000] GICv3: CPU0: found redistributor 0 region 0:0x0000000050880000
[ 0.063865] smp: Bringing up secondary CPUs ...
[ 0.068852] Detected VIPT I-cache on CPU1
[ 0.068894] GICv3: CPU1: found redistributor 1 region 0:0x00000000508a0000
[ 0.068963] CPU1: Booted secondary processor 0x0000000001 [0x410fd034]
[ 0.069809] Detected VIPT I-cache on CPU2
[ 0.069851] GICv3: CPU2: found redistributor 100 region 0:0x00000000508c0000
[ 0.069903] CPU2: Booted secondary processor 0x0000000100 [0x410fd034]
[ 0.070698] Detected VIPT I-cache on CPU3
[ 0.070722] GICv3: CPU3: found redistributor 101 region 0:0x00000000508e0000
[ 0.070749] CPU3: Booted secondary processor 0x0000000101 [0x410fd034]
[ 0.070847] smp: Brought up 1 node, 4 CPUs
<..snip..>
That's not the correct way to find out. Each CPU tries to find its
matching RD in the region. This doesn't mean there aren't more RDs
present in the GIC.
You need to iterate over all the RDs in the region until you find one
that has GICR_TYPER.Last == 1. This will give you the actual count.
Alternatively, you can check whether the RD at 508e0000 has that bit
set. If it doesn't, then you know there are more RDs than CPUs.
M.
--
Without deviation from the norm, progress is not possible.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
That's enough redistributor space for 16 CPUs. However, you only
describe 4. Either the number of CPUs is wrong, the size is wrong, or
the GIC has been configured for more cores than the SoC has.
Confirmed the SoC can only find 4 redistributors:
localhost:~ # dmesg | grep CPU
[ 0.000000] Booting Linux on physical CPU 0x0000000000 [0x410fd034]
[ 0.000000] Detected VIPT I-cache on CPU0
[ 0.000000] CPU features: detected: GIC system register CPU interface
[ 0.000000] CPU features: detected: ARM erratum 845719
[ 0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=4, Nodes=1
[ 0.000000] rcu: RCU restricting CPUs from NR_CPUS=480 to nr_cpu_ids=4.
[ 0.000000] GICv3: CPU0: found redistributor 0 region 0:0x0000000050880000
[ 0.063865] smp: Bringing up secondary CPUs ...
[ 0.068852] Detected VIPT I-cache on CPU1
[ 0.068894] GICv3: CPU1: found redistributor 1 region 0:0x00000000508a0000
[ 0.068963] CPU1: Booted secondary processor 0x0000000001 [0x410fd034]
[ 0.069809] Detected VIPT I-cache on CPU2
[ 0.069851] GICv3: CPU2: found redistributor 100 region 0:0x00000000508c0000
[ 0.069903] CPU2: Booted secondary processor 0x0000000100 [0x410fd034]
[ 0.070698] Detected VIPT I-cache on CPU3
[ 0.070722] GICv3: CPU3: found redistributor 101 region 0:0x00000000508e0000
[ 0.070749] CPU3: Booted secondary processor 0x0000000101 [0x410fd034]
[ 0.070847] smp: Brought up 1 node, 4 CPUs
<..snip..>
That's not the correct way to find out. Each CPU tries to find its
matching RD in the region. This doesn't mean there aren't more RDs
present in the GIC.
You need to iterate over all the RDs in the region until you find one
that has GICR_TYPER.Last == 1. This will give you the actual count.
Alternatively, you can check whether the RD at 508e0000 has that bit
set. If it doesn't, then you know there are more RDs than CPUs.
M.
Thanks for your guidance. Not sure if any debug log can be enabled for this
check so I temporarily add an ugly message as below:
@@ -866,10 +866,11 @@ static int __gic_populate_rdist(struct redist_region *region, void __iomem *ptr)gic_data_rdist_rd_base()=ptr;gic_data_rdist()->phys_base=region->phys_base+offset;-pr_info("CPU%d: found redistributor %lx region %d:%pa\n",+pr_info("CPU%d: found redistributor %lx region %d:%pa last: %d\n",smp_processor_id(),mpidr,(int)(region-gic_data.redist_regions),-&gic_data_rdist()->phys_base);+&gic_data_rdist()->phys_base,+(typer&GICR_TYPER_LAST)?1:0);return0;}
The following log shows that the "Last" bit (GICR_TYPER[4]) of RD at
508e0000 has been set.
localhost:~ # dmesg | grep GIC
[ 0.000000] CPU features: detected: GIC system register CPU interface
[ 0.000000] GICv3: 544 SPIs implemented
[ 0.000000] GICv3: 0 Extended SPIs implemented
[ 0.000000] GICv3: Distributor has no Range Selector support
[ 0.000000] GICv3: 16 PPIs implemented
[ 0.000000] GICv3: CPU0: found redistributor 0 region 0:0x0000000050880000 last: 0
[ 0.078745] GICv3: CPU1: found redistributor 1 region 0:0x00000000508a0000 last: 0
[ 0.089598] GICv3: CPU2: found redistributor 100 region 0:0x00000000508c0000 last: 0
[ 0.100395] GICv3: CPU3: found redistributor 101 region 0:0x00000000508e0000 last: 1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
From: Marc Zyngier <maz@kernel.org> Date: 2021-08-21 14:20:35
On Sat, 21 Aug 2021 13:39:04 +0100,
Chester Lin [off-list ref] wrote:
The following log shows that the "Last" bit (GICR_TYPER[4]) of RD at
508e0000 has been set.
localhost:~ # dmesg | grep GIC
[ 0.000000] CPU features: detected: GIC system register CPU interface
[ 0.000000] GICv3: 544 SPIs implemented
[ 0.000000] GICv3: 0 Extended SPIs implemented
[ 0.000000] GICv3: Distributor has no Range Selector support
[ 0.000000] GICv3: 16 PPIs implemented
[ 0.000000] GICv3: CPU0: found redistributor 0 region 0:0x0000000050880000 last: 0
[ 0.078745] GICv3: CPU1: found redistributor 1 region 0:0x00000000508a0000 last: 0
[ 0.089598] GICv3: CPU2: found redistributor 100 region 0:0x00000000508c0000 last: 0
[ 0.100395] GICv3: CPU3: found redistributor 101 region 0:0x00000000508e0000 last: 1
Looks convincing enough. Trimming the RD range to 512kB is then the
right thing to do.
Thanks,
M.
--
Without deviation from the norm, progress is not possible.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2
Given this is an entirely different family from i.MX and new?, shouldn't
it use 'nxp' instead of 'fsl'? Either way,
It sounds good and Radu from NXP has mentioned a similar idea for the
compatible string of linflexuart. To keep the naming consistency, should we
change all 'fsl' to 'nxp' as well?
I assume that question was just unclearly phrased, so for the record:
ABI stability rules forbid us from changing "all 'fsl'" in compatible
strings or property names.
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/ABI.rst
Deployed firmware providing mainline-merged platforms with DTBs using
fsl prefix (e.g., the quoted LX2160A) needs to continue working with
newer drivers, and deployed mainline Linux should continue working after
firmware updates that modify the DTB provided to Linux.
This is a new platform/SoC therefore there is no ABI. There is no
requirement in the kernel that a new ABI (which you define in this
patchset in the bindings) should be compatible with something
somewhere. It's some misunderstanding of stable ABI. Therefore all new
compatibles are allowed to be nxp, not fsl.
No one here proposed renaming existing compatibles from fsl tro nxp.
We talk about new ones.
Different question of course whether you want to be nice to some
existing out-of-tree users... but then have in mind that we don't care
about out of tree. :) Anyway being nice to out-of-tree is not part of
ABI. It's just being nice and useful.
Best regards,
Krzysztof
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2
Given this is an entirely different family from i.MX and new?, shouldn't
it use 'nxp' instead of 'fsl'? Either way,
It sounds good and Radu from NXP has mentioned a similar idea for the
compatible string of linflexuart. To keep the naming consistency, should we
change all 'fsl' to 'nxp' as well?
I assume that question was just unclearly phrased, so for the record:
ABI stability rules forbid us from changing "all 'fsl'" in compatible
strings or property names.
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/ABI.rst
Deployed firmware providing mainline-merged platforms with DTBs using
fsl prefix (e.g., the quoted LX2160A) needs to continue working with
newer drivers, and deployed mainline Linux should continue working after
firmware updates that modify the DTB provided to Linux.
This is a new platform/SoC therefore there is no ABI. There is no
requirement in the kernel that a new ABI (which you define in this
patchset in the bindings) should be compatible with something
somewhere. It's some misunderstanding of stable ABI. Therefore all new
compatibles are allowed to be nxp, not fsl.
No one here proposed renaming existing compatibles from fsl tro nxp.
We talk about new ones.
Chester seemingly did: "all 'fsl' ... as well", not "all new 'fsl'"
ones, in the patch context of existing fsl.yaml. Like I said, it may
just have been unluckily worded.
Therefore my saying that it does contain tons of non-new SoC/platform
bindings that he's not allowed to break by changing them.
Different question of course whether you want to be nice to some
existing out-of-tree users... but then have in mind that we don't care
about out of tree. :) Anyway being nice to out-of-tree is not part of
ABI. It's just being nice and useful.
Nobody is suggesting new S32G ABI be compatible with downstream BSPs.
These patches and changes we're discussing already differ from the BSP.
My point was that as soon as we merge S32G into mainline, it will become
ABI and shouldn't be changed incompatibly anymore once in a release.
These automotive platforms don't run off-the-shelf distros yet and will
need to get their bootloaders upstreamed, too. In particular we'll need
mainline TF-A to merge the SCMI implementation before we can rely on it
here in the kernel for a clk driver; that's holding up MMC and Ethernet.
Best regards,
Andreas
--
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer
HRB 36809 (AG Nürnberg)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2
Given this is an entirely different family from i.MX and new?, shouldn't
it use 'nxp' instead of 'fsl'?
S32V also still used fsl prefix, despite the company name long being NXP
(same for several Layerscape and i.MX models).
If, as Radu indicated on 3/8, NXP wants to make that switch now for S32G
then I see no reason against nxp. I verified that it's already defined:
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/vendor-prefixes.yaml
However, should the matching .dts[i] files using nxp prefix (4-6/8) then
still go under dts/freescale/, or should they go to a new dts/nxp/ then?
That would separate it from S32V. Intel did do a switch from dts/altera/
to dts/intel/ at some point, so there's precedence for either, I guess.
No idea whether anything might break if we moved S32V alongside S32G.
Similarly, the easiest and most merge-friendly would be to leave
arm/fsl.yaml and add the nxp-prefixed S32G2 there, as done here. If NXP
want to rename fsl.yaml to nxp.yaml in a general housekeeping effort,
that could be done independently, outside Chester's patchset.
Either way,
Acked-by: Rob Herring <robh@kernel.org>
Thanks,
Andreas
--
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer
HRB 36809 (AG Nürnberg)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
@@ -983,6 +983,13 @@ properties:-const:solidrun,lx2160a-cex7-const:fsl,lx2160a+-description:S32G2 based Boards+items:+-enum:+-fsl,s32g274a-evb+-fsl,s32g274a-rdb2+-const:fsl,s32g2
Given this is an entirely different family from i.MX and new?, shouldn't
it use 'nxp' instead of 'fsl'? Either way,
It sounds good and Radu from NXP has mentioned a similar idea for the
compatible string of linflexuart. To keep the naming consistency, should we
change all 'fsl' to 'nxp' as well?
I assume that question was just unclearly phrased, so for the record:
ABI stability rules forbid us from changing "all 'fsl'" in compatible
strings or property names.
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/ABI.rst
Deployed firmware providing mainline-merged platforms with DTBs using
fsl prefix (e.g., the quoted LX2160A) needs to continue working with
newer drivers, and deployed mainline Linux should continue working after
firmware updates that modify the DTB provided to Linux.
So, if NXP wants to use nxp prefix for new S32G bindings, you can do
that for your additions only, but for LINFlexD UART (3/8) you will still
need to use fsl for the "historical" S32V binding used as fallback.
Please keep S32G consistent with itself - so if we decide on nxp here,
we should apply it to SoC, boards, LINFlexD and any future peripherals.
For example, we could rename the fsl.yaml
to nxp.yaml.
Since other people might be contributing i.MX boards etc. to that file,
better not make your patch series conflict with other people's patches,
so that it can get merged and we can move on to the next patchsets.
The schema filename is not ABI, so it can be renamed later.
The .dtb path may become ABI (e.g., U-Boot $fdtfile), thus my comment
about consciously deciding between freescale/ vs. nxp/ subdirectory.
However, changing all of them would cause some impacts, which will
need more verifications on new strings. Otherwise we would have to tolerate the
naming differences only used by s32g2.
I fear tolerating the mess one way or another is the only viable way.
Otherwise both bindings and drivers would need duplication for backwards
compatibility, for no good reason - Freescale was acquired back in 2015.
Cheers,
Andreas
--
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer
HRB 36809 (AG Nürnberg)
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel