Add support for Agilex5 SoCFPGA platform
Subsystems (6)
cover Add support for Agilex5 SoCFPGA platform
From: Niravkumar L Rabara <[email protected]>
This patch set introduce the changes required for Agilx5 platform.
patch [1/4] - Introduced compatible string for Agilex5 board patch [2/4] - Add reset and clock header and yaml file. patch [3/4] - Add clock driver for Agilex5 platform. This patch depends on patch 2.
patch [4/4] - Add device tree files, socfpga_agilex5_socdk_swvp.dts is used for Virtual Platform (SIMICS) and socfpga_agilex5_socdk_nand.dts is used for NAND Flash based board. This patch depends on patch 3.
Niravkumar L Rabara (4):
dt-bindings: intel: Add Intel Agilex5 compatible
dt-bindings: clock: Add Intel Agilex5 clocks and resets
clk: socfpga: agilex5: Add clock driver for Agilex5 platform
arm64: dts: agilex5: Add initial support for Intel's Agilex5 SoCFPGA
.../bindings/arm/intel,socfpga.yaml | 1 + .../bindings/clock/intel,agilex5.yaml | 42 + arch/arm64/boot/dts/intel/Makefile | 3 + .../arm64/boot/dts/intel/socfpga_agilex5.dtsi | 641 +++++++++++++ .../boot/dts/intel/socfpga_agilex5_socdk.dts | 184 ++++ .../dts/intel/socfpga_agilex5_socdk_nand.dts | 131 +++ .../dts/intel/socfpga_agilex5_socdk_swvp.dts | 248 ++++++ drivers/clk/socfpga/Kconfig | 4 +- drivers/clk/socfpga/Makefile | 2 +- drivers/clk/socfpga/clk-agilex5.c | 843 ++++++++++++++++++ drivers/clk/socfpga/clk-pll-s10.c | 48 + drivers/clk/socfpga/stratix10-clk.h | 2 + include/dt-bindings/clock/agilex5-clock.h | 100 +++ .../dt-bindings/reset/altr,rst-mgr-agilex5.h | 79 ++ 14 files changed, 2325 insertions(+), 3 deletions(-) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5.yaml create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_nand.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_swvp.dts create mode 100644 drivers/clk/socfpga/clk-agilex5.c create mode 100644 include/dt-bindings/clock/agilex5-clock.h create mode 100644 include/dt-bindings/reset/altr,rst-mgr-agilex5.h
Signature · 1 line
2.25.1
PATCH 1/4 dt-bindings: intel: Add Intel Agilex5 compatible
From: Niravkumar L Rabara <[email protected]>
Add new compatible for Intel Agilex5 based boards.
--- Documentation/devicetree/bindings/arm/intel,socfpga.yaml | 1 + 1 file changed, 1 insertion(+)
Documentation/devicetree/bindings/arm/intel,socfpga.yaml +1 -0
@@ -20,6 +20,7 @@ properties: - intel,n5x-socdk - intel,socfpga-agilex-n6000 - intel,socfpga-agilex-socdk + - intel,socfpga-agilex5-socdk - const: intel,socfpga-agilex additionalProperties: true
Signature · 1 line
2.25.1
Re: [PATCH 1/4] dt-bindings: intel: Add Intel Agilex5 compatible
Quoted 14 lines · On 18/06/2023 15:22, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add new compatible for Intel Agilex5 based boards.
Signed-off-by Niravkumar L Rabara <[email protected]>--- Documentation/devicetree/bindings/arm/intel,socfpga.yaml | 1 + 1 file changed, 1 insertion(+)Documentation/devicetree/bindings/arm/intel,socfpga.yaml +0
-0
Documentation/devicetree/bindings/arm/intel,socfpga.yaml properties: · line 24
Quoted 5 lines
Documentation/devicetree/bindings/arm/intel,socfpga.yaml +1
-0@@ -20,6 +20,7 @@ properties: - intel,n5x-socdk - intel,socfpga-agilex-n6000 - intel,socfpga-agilex-socdk + - intel,socfpga-agilex5-socdk - const: intel,socfpga-agilex
This is agilex5, not agilex. Why are you using the same SoC compatible? You have entire commit msg to explain your hardware and avoid such questions...
Best regards,
Krzysztof
Re: [PATCH 1/4] dt-bindings: intel: Add Intel Agilex5 compatible
Quoted 6 lines
Quoted 6 lines
Documentation/devicetree/bindings/arm/intel,socfpga.yaml +0
-0@@ -20,6 +20,7 @@ properties:
Quoted 8 lines
Documentation/devicetree/bindings/arm/intel,socfpga.yaml properties: · line 24
Quoted 5 lines
Documentation/devicetree/bindings/arm/intel,socfpga.yaml +1
-0- intel,n5x-socdk - intel,socfpga-agilex-n6000 - intel,socfpga-agilex-socdk + - intel,socfpga-agilex5-socdk - const: intel,socfpga-agilexThis is agilex5, not agilex. Why are you using the same SoC compatible? You have entire commit msg to explain your hardware and avoid such questions...
Yes, better to use different SoC compatible for Agilex5, will add new description and item for Agilex5 board in v2 patch.
PATCH 2/4 dt-bindings: clock: Add Intel Agilex5 clocks and resets
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add clock and reset ID definitions for Intel Agilex5 SoCFPGA
--- .../bindings/clock/intel,agilex5.yaml | 42 ++++++++ include/dt-bindings/clock/agilex5-clock.h | 100 ++++++++++++++++++ .../dt-bindings/reset/altr,rst-mgr-agilex5.h | 79 ++++++++++++++ 3 files changed, 221 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5.yaml create mode 100644 include/dt-bindings/clock/agilex5-clock.h create mode 100644 include/dt-bindings/reset/altr,rst-mgr-agilex5.h
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +42 -0
@@ -0,0 +1,42 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 platform clock controller binding + +maintainers: + - Teh Wen Ping <[email protected]> + - Niravkumar L Rabara <[email protected]> + +description: + The Intel Agilex5 Clock controller is an integrated clock controller, which + generates and supplies to all modules. + +properties: + compatible: + const: intel,agilex5-clkmgr + + '#clock-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - '#clock-cells' + +additionalProperties: false + +examples: + # Clock controller node: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...
include/dt-bindings/clock/agilex5-clock.h new file +100 -0
100 lines not on this page · raw message
include/dt-bindings/reset/altr,rst-mgr-agilex5.h new file +79 -0
79 lines not on this page · raw message
Signature · 1 line
2.25.1
Re: [PATCH 2/4] dt-bindings: clock: Add Intel Agilex5 clocks and resets
Documentation/devicetree/bindings/clock/intel,agilex5.yaml
Quoted 21 lines · On 18/06/2023 15:22, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock and reset ID definitions for Intel Agilex5 SoCFPGA
Co-developed-by Teh Wen Ping <[email protected]>Signed-off-by Teh Wen Ping <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- .../bindings/clock/intel,agilex5.yaml | 42 ++++++++ include/dt-bindings/clock/agilex5-clock.h | 100 ++++++++++++++++++ .../dt-bindings/reset/altr,rst-mgr-agilex5.h | 79 ++++++++++++++ 3 files changed, 221 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5.yaml create mode 100644 include/dt-bindings/clock/agilex5-clock.h create mode 100644 include/dt-bindings/reset/altr,rst-mgr-agilex5.hDocumentation/devicetree/bindings/clock/intel,agilex5.yaml new file +0
-0
Filename matching compatible, so missing "clk"
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 7
Quoted 8 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +7
-0@@ -0,0 +1,42 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 platform clock controller binding
Drop "binding"
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 15
Quoted 8 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +8
-0+ +maintainers: + - Teh Wen Ping <[email protected]> + - Niravkumar L Rabara <[email protected]> + +description: + The Intel Agilex5 Clock controller is an integrated clock controller, which + generates and supplies to all modules.
"generates and supplies" what?
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 19
Quoted 4 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +4
-0+ +properties: + compatible: + const: intel,agilex5-clkmgr
Why "clkmgr", not "clk"? You did not call it Clock manager anywhere in the description or title.
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 30
Quoted 11 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +11
-0+ + '#clock-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - '#clock-cells'
Keep the same order as in properties:
include/dt-bindings/clock/agilex5-clock.h
Quoted 17 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +12
-0+ +additionalProperties: false + +examples: + # Clock controller node: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...include/dt-bindings/clock/agilex5-clock.h new file +0
-0
Filename the same as binding. Missing vendor prefix, entirely different device name.
Quoted 5 lines
include/dt-bindings/clock/agilex5-clock.h new file +4
-0@@ -0,0 +1,100 @@ +/* SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause */ +/* +* Copyright (C) 2022, Intel Corporation +*/
...
include/dt-bindings/reset/altr,rst-mgr-agilex5.h
Quoted 7 lines
include/dt-bindings/clock/agilex5-clock.h new file +2
-0+ +#endif /* __AGILEX5_CLOCK_H */ include/dt-bindings/reset/altr,rst-mgr-agilex5.h new file +0
-0
Same filename as binding.
But why do you need this file? Your device is not a reset controller.
Best regards,
Krzysztof
Re: [PATCH 2/4] dt-bindings: clock: Add Intel Agilex5 clocks and resets
Quoted 17 lines
From: Krzysztof Kozlowski <[email protected]>
Quoted 16 lines
From: Niravkumar L Rabara <[email protected]>
Add clock and reset ID definitions for Intel Agilex5 SoCFPGA
Co-developed-by Teh Wen Ping <[email protected]>Signed-off-by Teh Wen Ping <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- .../bindings/clock/intel,agilex5.yaml | 42 ++++++++ include/dt-bindings/clock/agilex5-clock.h | 100 ++++++++++++++++++ .../dt-bindings/reset/altr,rst-mgr-agilex5.h | 79 ++++++++++++++ 3 files changed, 221 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5.yaml create mode 100644 include/dt-bindings/clock/agilex5-clock.h create mode 100644 include/dt-bindings/reset/altr,rst-mgr-agilex5.h
Quoted 8 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml
Quoted 5 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +0
-0Filename matching compatible, so missing "clk"
Will update in V2 patch, rename file to intel,agilex5-clk.yaml
Quoted 11 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 7
Quoted 8 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +7
-0@@ -0,0 +1,42 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 platform clock controller bindingDrop "binding"
Will update in V2 patch.
Quoted 10 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 15
Quoted 8 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +8
-0+ +maintainers: + - Teh Wen Ping <[email protected]> + - Niravkumar L Rabara <[email protected]> + +description: + The Intel Agilex5 Clock controller is an integrated clock controller, which + generates and supplies to all modules."generates and supplies" what?
Will change to "generates and supplies clock" in V2 patch.
Quoted 10 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 19
Quoted 5 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +4
-0+ +properties: + compatible: + const: intel,agilex5-clkmgrWhy "clkmgr", not "clk"? You did not call it Clock manager anywhere in the description or title.
The register in Agilex5 handling the clock is named clock_mgr. Previous IntelSocFPGA, Agilex and Stratix10, are also named clkmgr.
Quoted 5 lines
Quoted 5 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +5
-0+ + '#clock-cells': + const: 1 + + reg:
Quoted 9 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 30
Quoted 6 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +6
-0+ maxItems: 1 + +required: + - compatible + - reg + - '#clock-cells'Keep the same order as in properties:
Will update in V2 patch.
Quoted 12 lines
Quoted 12 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +12
-0+ +additionalProperties: false + +examples: + # Clock controller node: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...
Quoted 9 lines
include/dt-bindings/clock/agilex5-clock.h
Quoted 5 lines
include/dt-bindings/clock/agilex5-clock.h new file +0
-0Filename the same as binding. Missing vendor prefix, entirely different device name.
Will change filename to intel,agilex5-clock.h in V2.
Quoted 18 lines
Quoted 5 lines
include/dt-bindings/clock/agilex5-clock.h new file +4
-0@@ -0,0 +1,100 @@ +/* SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause */ +/* +* Copyright (C) 2022, Intel Corporation +*/ ...
include/dt-bindings/reset/altr,rst-mgr-agilex5.h
Quoted 7 lines
include/dt-bindings/clock/agilex5-clock.h new file +2
-0+ +#endif /* __AGILEX5_CLOCK_H */ include/dt-bindings/reset/altr,rst-mgr-agilex5.h new file +0
-0Same filename as binding.
Quoted 1 line
But why do you need this file? Your device is not a reset controller.
Because Agilex5 device tree uses the reset definition from this file.
Best Regards,
Wen Ping
Re: [PATCH 2/4] dt-bindings: clock: Add Intel Agilex5 clocks and resets
Quoted 11 lines · On 20/06/2023 12:39, [email protected] wrote:
Quoted 10 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 19
Quoted 5 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +4
-0+ +properties: + compatible: + const: intel,agilex5-clkmgrWhy "clkmgr", not "clk"? You did not call it Clock manager anywhere in the description or title.
Quoted 2 lines
The register in Agilex5 handling the clock is named
clock_mgr. Previous IntelSocFPGA, Agilex and Stratix10, are also named clkmgr.
So use it in description.
Quoted 40 lines
Quoted 15 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 30
Quoted 12 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +11
-0+ + '#clock-cells': + const: 1 + + reg: + maxItems: 1 + +required: + - compatible + - reg + - '#clock-cells'Keep the same order as in properties:
Will update in V2 patch.
Quoted 12 lines
Quoted 12 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +12
-0+ +additionalProperties: false + +examples: + # Clock controller node: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...Quoted 9 lines
include/dt-bindings/clock/agilex5-clock.h
Quoted 5 lines
include/dt-bindings/clock/agilex5-clock.h new file +0
-0Filename the same as binding. Missing vendor prefix, entirely different device name.
Quoted 1 line
Will change filename to intel,agilex5-clock.h in V2.
Read the comment - same as binding. You did not call binding that way... unless you rename the binding.
Quoted 11 lines
Quoted 11 lines
include/dt-bindings/reset/altr,rst-mgr-agilex5.h
Quoted 8 lines
include/dt-bindings/clock/agilex5-clock.h new file +2
-0+ +#endif /* __AGILEX5_CLOCK_H */ include/dt-bindings/reset/altr,rst-mgr-agilex5.h new file +0
-0Same filename as binding.
Quoted 3 lines
Quoted 1 line
But why do you need this file? Your device is not a reset controller.
Because Agilex5 device tree uses the reset definition from this file.
That's not the correct reason. The binding header has nothing to do with this device. You miss another patch adding support for your device (compatible) with this header.
Best regards,
Krzysztof
Re: [PATCH 2/4] dt-bindings: clock: Add Intel Agilex5 clocks and resets
Quoted 12 lines
From: Krzysztof Kozlowski @ 2023-06-20 11:06 UTC (permalink / raw)
Quoted 11 lines
Quoted 10 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml line 19
Quoted 5 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +4
-0+ +properties: + compatible: + const: intel,agilex5-clkmgrWhy "clkmgr", not "clk"? You did not call it Clock manager anywhere in the description or title.
Quoted 4 lines
Quoted 2 lines
The register in Agilex5 handling the clock is named
clock_mgr. Previous IntelSocFPGA, Agilex and Stratix10, are also named clkmgr.So use it in description.
Noted. Will update the description in V2.
Quoted 22 lines
Quoted 22 lines
Quoted 21 lines
include/dt-bindings/clock/agilex5-clock.h
Quoted 17 lines
Documentation/devicetree/bindings/clock/intel,agilex5.yaml new file +12
-0+ +additionalProperties: false + +examples: + # Clock controller node: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...include/dt-bindings/clock/agilex5-clock.h new file +0
-0Filename the same as binding. Missing vendor prefix, entirely different device name.
Quoted 4 lines
Quoted 1 line
Will change filename to intel,agilex5-clock.h in V2.
Read the comment - same as binding. You did not call binding that way... unless you rename the binding.
Just to confirm, the binding name you are referring to is "intel,agilex5-clkmgr"? I will change the filename to intel,agilex5-clkmgr.h in V2.
Best Regards,
Wen Ping
Rob Herring
maintainerRe: [PATCH 2/4] dt-bindings: clock: Add Intel Agilex5 clocks and resets
In reply to [email protected]Quoted 7 lines · On Sun, 18 Jun 2023 21:22:33 +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock and reset ID definitions for Intel Agilex5 SoCFPGA
Co-developed-by Teh Wen Ping <[email protected]>Signed-off-by Teh Wen Ping <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>
Quoted 9 lines
--- .../bindings/clock/intel,agilex5.yaml | 42 ++++++++ include/dt-bindings/clock/agilex5-clock.h | 100 ++++++++++++++++++ .../dt-bindings/reset/altr,rst-mgr-agilex5.h | 79 ++++++++++++++ 3 files changed, 221 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5.yaml create mode 100644 include/dt-bindings/clock/agilex5-clock.h create mode 100644 include/dt-bindings/reset/altr,rst-mgr-agilex5.h
My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):
yamllint warnings/errors:
dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/clock/intel,agilex5.yaml: title: 'Intel SoCFPGA Agilex5 platform clock controller binding' should not be valid under {'pattern': '([Bb]inding| [Ss]chema)'}
hint: Everything is a binding/schema, no need to say it. Describe what hardware the binding is for.
from schema $id: http://devicetree.org/meta-schemas/core.yaml#
doc reference errors (make refcheckdocs):
See https://patchwork.ozlabs.org/project/devicetree-bindings/patch/[email protected]
The base for the series is generally the latest rc1. A different dependency should be noted in *this* patch.
If you already ran 'make dt_binding_check' and didn't see the above error(s), then make sure 'yamllint' is installed and dt-schema is up to date:
pip3 install dtschema --upgrade
Please check and re-submit after running the above command yourself. Note that DT_SCHEMA_FILES can be set to your schema file to speed up checking your schema. However, it must be unset to test all examples with your schema.
PATCH 3/4 clk: socfpga: agilex5: Add clock driver for Agilex5 platform
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
The clock manager driver for Agilex5 is very similar to the Agilex platform. This patch makes the necessary changes for the driver to differentiate between the Agilex and the Agilex5 platforms.
--- drivers/clk/socfpga/Kconfig | 4 +- drivers/clk/socfpga/Makefile | 2 +- drivers/clk/socfpga/clk-agilex5.c | 843 ++++++++++++++++++++++++++++ drivers/clk/socfpga/clk-pll-s10.c | 48 ++ drivers/clk/socfpga/stratix10-clk.h | 2 + 5 files changed, 896 insertions(+), 3 deletions(-) create mode 100644 drivers/clk/socfpga/clk-agilex5.c
drivers/clk/socfpga/Kconfig +2 -2
@@ -4,7 +4,7 @@ config CLK_INTEL_SOCFPGA default ARCH_INTEL_SOCFPGA help Support for the clock controllers present on Intel SoCFPGA and eASIC - devices like Aria, Cyclone, Stratix 10, Agilex and N5X eASIC. + devices like Aria, Cyclone, Stratix 10, Agilex, N5X eASIC and Agilex5. if CLK_INTEL_SOCFPGA
@@ -13,7 +13,7 @@ config CLK_INTEL_SOCFPGA32 default ARM && ARCH_INTEL_SOCFPGA config CLK_INTEL_SOCFPGA64 - bool "Intel Stratix / Agilex / N5X clock controller support" if COMPILE_TEST && (!ARM64 || !ARCH_INTEL_SOCFPGA) + bool "Intel Stratix / Agilex / N5X clock / Agilex5 controller support" if COMPILE_TEST && (!ARM64 || !ARCH_INTEL_SOCFPGA) default ARM64 && ARCH_INTEL_SOCFPGA endif # CLK_INTEL_SOCFPGA
drivers/clk/socfpga/Makefile +1 -1
@@ -3,4 +3,4 @@ obj-$(CONFIG_CLK_INTEL_SOCFPGA32) += clk.o clk-gate.o clk-pll.o clk-periph.o \ clk-pll-a10.o clk-periph-a10.o clk-gate-a10.o obj-$(CONFIG_CLK_INTEL_SOCFPGA64) += clk-s10.o \ clk-pll-s10.o clk-periph-s10.o clk-gate-s10.o \ - clk-agilex.o + clk-agilex.o clk-agilex5.o
drivers/clk/socfpga/clk-agilex5.c new file +843 -0
843 lines not on this page · raw message
drivers/clk/socfpga/clk-pll-s10.c +48 -0
57 lines not on this page · raw message
drivers/clk/socfpga/stratix10-clk.h +2 -0
@@ -77,6 +77,8 @@ struct clk_hw *agilex_register_pll(const struct stratix10_pll_clock *clks, void __iomem *reg);struct clk_hw *n5x_register_pll (const struct stratix10_pll_clock *clks,void __iomem *reg); +struct clk_hw *agilex5_register_pll (const struct stratix10_pll_clock *clks, +void __iomem *reg);struct clk_hw *s10_register_periph (const struct stratix10_perip_c_clock *clks,void __iomem *reg); struct clk_hw *n5x_register_periph(const struct n5x_perip_c_clock *clks,
Signature · 1 line
2.25.1
Dinh Nguyen
maintainerRe: [PATCH 3/4] clk: socfpga: agilex5: Add clock driver for Agilex5 platform
Quoted 3 lines · On 6/18/23 08:22, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
The clock manager driver for Agilex5 is very similar to the Agilex
Then why create a whole new driver? Surely there's alot of re-use you can do?
Quoted 52 lines
platform. This patch makes the necessary changes for the driver to differentiate between the Agilex and the Agilex5 platforms.
Signed-off-by Teh Wen Ping <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- drivers/clk/socfpga/Kconfig | 4 +- drivers/clk/socfpga/Makefile | 2 +- drivers/clk/socfpga/clk-agilex5.c | 843 ++++++++++++++++++++++++++++ drivers/clk/socfpga/clk-pll-s10.c | 48 ++ drivers/clk/socfpga/stratix10-clk.h | 2 + 5 files changed, 896 insertions(+), 3 deletions(-)create mode 100644
drivers/clk/socfpga/clk-agilex5.cdrivers/clk/socfpga/Kconfig +2
-2@@ -4,7 +4,7 @@ config CLK_INTEL_SOCFPGA default ARCH_INTEL_SOCFPGA help Support for the clock controllers present on Intel SoCFPGA and eASIC - devices like Aria, Cyclone, Stratix 10, Agilex and N5X eASIC. + devices like Aria, Cyclone, Stratix 10, Agilex, N5X eASIC and Agilex5. if CLK_INTEL_SOCFPGA@@ -13,7 +13,7 @@ config CLK_INTEL_SOCFPGA32 default ARM && ARCH_INTEL_SOCFPGA config CLK_INTEL_SOCFPGA64 - bool "Intel Stratix / Agilex / N5X clock controller support" if COMPILE_TEST && (!ARM64 || !ARCH_INTEL_SOCFPGA) + bool "Intel Stratix / Agilex / N5X clock / Agilex5 controller support" if COMPILE_TEST && (!ARM64 || !ARCH_INTEL_SOCFPGA) default ARM64 && ARCH_INTEL_SOCFPGA endif # CLK_INTEL_SOCFPGAdrivers/clk/socfpga/Makefile +1
-1@@ -3,4 +3,4 @@ obj-$(CONFIG_CLK_INTEL_SOCFPGA32) += clk.o clk-gate.o clk-pll.o clk-periph.o \ clk-pll-a10.o clk-periph-a10.o clk-gate-a10.o obj-$(CONFIG_CLK_INTEL_SOCFPGA64) += clk-s10.o \ clk-pll-s10.o clk-periph-s10.o clk-gate-s10.o \ - clk-agilex.o + clk-agilex.o clk-agilex5.odrivers/clk/socfpga/clk-agilex5.c new file +0
-0
drivers/clk/socfpga/clk-agilex5.c line 3
Quoted 3 lines
drivers/clk/socfpga/clk-agilex5.c new file +3
-0@@ -0,0 +1,843 @@ +// SPDX-License-Identifier: GPL-2.0 +/* +* Copyright (C) 2022, Intel Corporation
It's 2023 now!
Quoted 490 lines
drivers/clk/socfpga/clk-agilex5.c new file +490
-0+*/ +#include <linux/slab.h> +#include <linux/clk-provider.h> +#include <linux/of_device.h> +#include <linux/of_address.h> +#include <linux/platform_device.h> + +#include <dt-bindings/clock/agilex5-clock.h> + +#include "stratix10-clk.h" + +static const struct clk_parent_data pll_mux[] = { + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data boot_mux[] = { + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, +}; + +static const struct clk_parent_data core0_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c0" , + .name ="peri_pll_c0" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data core1_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c0" , + .name ="peri_pll_c0" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data core2_free_mux[] = { + { + .fw_name ="main_pll_c0" , + .name ="main_pll_c0" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data core3_free_mux[] = { + { + .fw_name ="main_pll_c0" , + .name ="main_pll_c0" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data dsu_free_mux[] = { + { + .fw_name ="main_pll_c2" , + .name ="main_pll_c2" , + }, + { + .fw_name ="peri_pll_c0" , + .name ="peri_pll_c0" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data noc_free_mux[] = { + { + .fw_name ="main_pll_c3" , + .name ="main_pll_c3" , + }, + { + .fw_name ="peri_pll_c1" , + .name ="peri_pll_c1" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data emaca_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data emacb_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data emac_ptp_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data gpio_db_free_mux[] = { + { + .fw_name ="main_pll_c3" , + .name ="main_pll_c3" , + }, + { + .fw_name ="peri_pll_c1" , + .name ="peri_pll_c1" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data psi_ref_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data usb31_free_mux[] = { + { + .fw_name ="main_pll_c3" , + .name ="main_pll_c3" , + }, + { + .fw_name ="peri_pll_c2" , + .name ="peri_pll_c2" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data s2f_usr0_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data s2f_usr1_free_mux[] = { + { + .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , + }, + { + .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , + }, + { + .fw_name ="osc1" , + .name ="osc1" , + }, + { + .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , + }, + { + .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , + }, +}; + +static const struct clk_parent_data core0_mux[] = { + { + .fw_name ="core0_free_clk" , + .name ="core0_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data core1_mux[] = { + { + .fw_name ="core1_free_clk" , + .name ="core1_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data core2_mux[] = { + { + .fw_name ="core2_free_clk" , + .name ="core2_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data core3_mux[] = { + { + .fw_name ="core3_free_clk" , + .name ="core3_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data dsu_mux[] = { + { + .fw_name ="dsu_free_clk" , + .name ="dsu_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data emac_mux[] = { + { + .fw_name ="emaca_free_clk" , + .name ="emaca_free_clk" , + }, + { + .fw_name ="emacb_free_clk" , + .name ="emacb_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data noc_mux[] = { + { + .fw_name ="noc_free_clk" , + .name ="noc_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data s2f_user0_mux[] = { + { + .fw_name ="s2f_user0_free_clk" , + .name ="s2f_user0_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data s2f_user1_mux[] = { + { + .fw_name ="s2f_user1_free_clk" , + .name ="s2f_user1_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data psi_mux[] = { + { + .fw_name ="psi_ref_free_clk" , + .name ="psi_ref_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data gpio_db_mux[] = { + { + .fw_name ="gpio_db_free_clk" , + .name ="gpio_db_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data emac_ptp_mux[] = { + { + .fw_name ="emac_ptp_free_clk" , + .name ="emac_ptp_free_clk" , + }, + { + .fw_name ="boot_clk" , + .name ="boot_clk" , + }, +}; + +static const struct clk_parent_data usb31_mux[] = { + { + .fw_name ="usb31_free_clk" ,
drivers/clk/socfpga/clk-agilex5.c line 503
Quoted 10 lines
drivers/clk/socfpga/clk-agilex5.c new file +10
-0+ .name = "usb31_free_clk", + }, + { + .fw_name = "boot_clk", + .name = "boot_clk", + }, +}; + +/* +* TODO - Clocks in AO (always on) controller
Remove your TODO, so did you do it already?
Quoted 173 lines
drivers/clk/socfpga/clk-agilex5.c new file +173
-0+* 2 main PLLs only +*/ +static const struct stratix10_pll_clock agilex5_pll_clks[] = { + {AGILEX5_BOOT_CLK ,"boot_clk" , boot_mux,ARRAY_SIZE (boot_mux),0 , +0x0 }, + {AGILEX5_MAIN_PLL_CLK ,"main_pll" , pll_mux,ARRAY_SIZE (pll_mux),0 , +0x48 }, + {AGILEX5_PERIPH_PLL_CLK ,"periph_pll" , pll_mux,ARRAY_SIZE (pll_mux),0 , +0x9C }, +}; + +static const struct stratix10_perip_c_clock agilex5_main_perip_c_clks[] = { + {AGILEX5_MAIN_PLL_C0_CLK ,"main_pll_c0" ,"main_pll" ,NULL ,1 ,0 , +0x5C }, + {AGILEX5_MAIN_PLL_C1_CLK ,"main_pll_c1" ,"main_pll" ,NULL ,1 ,0 , +0x60 }, + {AGILEX5_MAIN_PLL_C2_CLK ,"main_pll_c2" ,"main_pll" ,NULL ,1 ,0 , +0x64 }, + {AGILEX5_MAIN_PLL_C3_CLK ,"main_pll_c3" ,"main_pll" ,NULL ,1 ,0 , +0x68 }, + {AGILEX5_PERIPH_PLL_C0_CLK ,"peri_pll_c0" ,"periph_pll" ,NULL ,1 ,0 , +0xB0 }, + {AGILEX5_PERIPH_PLL_C1_CLK ,"peri_pll_c1" ,"periph_pll" ,NULL ,1 ,0 , +0xB4 }, + {AGILEX5_PERIPH_PLL_C2_CLK ,"peri_pll_c2" ,"periph_pll" ,NULL ,1 ,0 , +0xB8 }, + {AGILEX5_PERIPH_PLL_C3_CLK ,"peri_pll_c3" ,"periph_pll" ,NULL ,1 ,0 , +0xBC }, +}; + +/* Non-SW clock-gated enabled clocks */ +static const struct stratix10_perip_cnt_clock agilex5_main_perip_cnt_clks[] = { + {AGILEX5_CORE0_FREE_CLK ,"core0_free_clk" ,NULL , core0_free_mux, +ARRAY_SIZE (core0_free_mux),0 ,0x0104 ,0 ,0 ,0 }, + {AGILEX5_CORE1_FREE_CLK ,"core1_free_clk" ,NULL , core1_free_mux, +ARRAY_SIZE (core1_free_mux),0 ,0x0104 ,0 ,0 ,0 }, + {AGILEX5_CORE2_FREE_CLK ,"core2_free_clk" ,NULL , core2_free_mux, +ARRAY_SIZE (core2_free_mux),0 ,0x010C ,0 ,0 ,0 }, + {AGILEX5_CORE3_FREE_CLK ,"core3_free_clk" ,NULL , core3_free_mux, +ARRAY_SIZE (core3_free_mux),0 ,0x0110 ,0 ,0 ,0 }, + {AGILEX5_DSU_FREE_CLK ,"dsu_free_clk" ,NULL , dsu_free_mux, +ARRAY_SIZE (dsu_free_mux),0 ,0x0100 ,0 ,0 ,0 }, + {AGILEX5_NOC_FREE_CLK ,"noc_free_clk" ,NULL , noc_free_mux, +ARRAY_SIZE (noc_free_mux),0 ,0x40 ,0 ,0 ,0 }, + {AGILEX5_EMAC_A_FREE_CLK ,"emaca_free_clk" ,NULL , emaca_free_mux, +ARRAY_SIZE (emaca_free_mux),0 ,0xD4 ,0 ,0x88 ,0 }, + {AGILEX5_EMAC_B_FREE_CLK ,"emacb_free_clk" ,NULL , emacb_free_mux, +ARRAY_SIZE (emacb_free_mux),0 ,0xD8 ,0 ,0x88 ,1 }, + {AGILEX5_EMAC_PTP_FREE_CLK ,"emac_ptp_free_clk" ,NULL , + emac_ptp_free_mux,ARRAY_SIZE (emac_ptp_free_mux),0 ,0xDC ,0 ,0x88 , +2 }, + {AGILEX5_GPIO_DB_FREE_CLK ,"gpio_db_free_clk" ,NULL , gpio_db_free_mux, +ARRAY_SIZE (gpio_db_free_mux),0 ,0xE0 ,0 ,0x88 ,3 }, + {AGILEX5_S2F_USER0_FREE_CLK ,"s2f_user0_free_clk" ,NULL , + s2f_usr0_free_mux,ARRAY_SIZE (s2f_usr0_free_mux),0 ,0xE8 ,0 ,0x30 , +2 }, + {AGILEX5_S2F_USER1_FREE_CLK ,"s2f_user1_free_clk" ,NULL , + s2f_usr1_free_mux,ARRAY_SIZE (s2f_usr1_free_mux),0 ,0xEC ,0 ,0x88 , +5 }, + {AGILEX5_PSI_REF_FREE_CLK ,"psi_ref_free_clk" ,NULL , psi_ref_free_mux, +ARRAY_SIZE (psi_ref_free_mux),0 ,0xF0 ,0 ,0x88 ,6 }, + {AGILEX5_USB31_FREE_CLK ,"usb31_free_clk" ,NULL , usb31_free_mux, +ARRAY_SIZE (usb31_free_mux),0 ,0xF8 ,0 ,0x88 ,7 }, +}; + +/* SW Clock gate enabled clocks */ +static const struct stratix10_gate_clock agilex5_gate_clks[] = { +/* Main PLL0 Begin */ +/* MPU clocks */ + {AGILEX5_CORE0_CLK ,"core0_clk" ,NULL , core0_mux, +ARRAY_SIZE (core0_mux),0 ,0x24 ,8 ,0 ,0 ,0 ,0x30 ,5 ,0 }, + {AGILEX5_CORE1_CLK ,"core1_clk" ,NULL , core1_mux, +ARRAY_SIZE (core1_mux),0 ,0x24 ,9 ,0 ,0 ,0 ,0x30 ,5 ,0 }, + {AGILEX5_CORE2_CLK ,"core2_clk" ,NULL , core2_mux, +ARRAY_SIZE (core2_mux),0 ,0x24 ,10 ,0 ,0 ,0 ,0x30 ,6 ,0 }, + {AGILEX5_CORE3_CLK ,"core3_clk" ,NULL , core3_mux, +ARRAY_SIZE (core3_mux),0 ,0x24 ,11 ,0 ,0 ,0 ,0x30 ,7 ,0 }, + {AGILEX5_MPU_CLK ,"dsu_clk" ,NULL , dsu_mux,ARRAY_SIZE (dsu_mux),0 ,0 , +0 ,0 ,0 ,0 ,0x34 ,4 ,0 }, + {AGILEX5_MPU_PERIPH_CLK ,"mpu_periph_clk" ,NULL , dsu_mux, +ARRAY_SIZE (dsu_mux),0 ,0 ,0 ,0x44 ,20 ,2 ,0x34 ,4 ,0 }, + {AGILEX5_MPU_CCU_CLK ,"mpu_ccu_clk" ,NULL , dsu_mux, +ARRAY_SIZE (dsu_mux),0 ,0 ,0 ,0x44 ,18 ,2 ,0x34 ,4 ,0 }, + +/* l4 main clk has no divider now */ + {AGILEX5_L4_MAIN_CLK ,"l4_main_clk" ,NULL , noc_mux, +ARRAY_SIZE (noc_mux),0 ,0x24 ,1 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_L4_MP_CLK ,"l4_mp_clk" ,NULL , noc_mux,ARRAY_SIZE (noc_mux),0 , +0x24 ,2 ,0x44 ,4 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_L4_SYS_FREE_CLK ,"l4_sys_free_clk" ,NULL , noc_mux, +ARRAY_SIZE (noc_mux),0 ,0 ,0 ,0x44 ,2 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_L4_SP_CLK ,"l4_sp_clk" ,NULL , noc_mux,ARRAY_SIZE (noc_mux), +CLK_IS_CRITICAL ,0x24 ,3 ,0x44 ,6 ,2 ,0x30 ,1 ,0 }, + +/* Core sight clocks*/ + {AGILEX5_CS_AT_CLK ,"cs_at_clk" ,NULL , noc_mux,ARRAY_SIZE (noc_mux),0 , +0x24 ,4 ,0x44 ,24 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_CS_TRACE_CLK ,"cs_trace_clk" ,NULL , noc_mux, +ARRAY_SIZE (noc_mux),0 ,0x24 ,4 ,0x44 ,26 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_CS_PDBG_CLK ,"cs_pdbg_clk" ,"cs_at_clk" ,NULL ,1 ,0 ,0x24 ,4 , +0x44 ,28 ,1 ,0 ,0 ,0 }, +/* Main PLL0 End */ + +/* Main Peripheral PLL1 Begin */ + {AGILEX5_EMAC0_CLK ,"emac0_clk" ,NULL , emac_mux,ARRAY_SIZE (emac_mux), +0 ,0x7C ,0 ,0 ,0 ,0 ,0x94 ,26 ,0 }, + {AGILEX5_EMAC1_CLK ,"emac1_clk" ,NULL , emac_mux,ARRAY_SIZE (emac_mux), +0 ,0x7C ,1 ,0 ,0 ,0 ,0x94 ,27 ,0 }, + {AGILEX5_EMAC2_CLK ,"emac2_clk" ,NULL , emac_mux,ARRAY_SIZE (emac_mux), +0 ,0x7C ,2 ,0 ,0 ,0 ,0x94 ,28 ,0 }, + {AGILEX5_EMAC_PTP_CLK ,"emac_ptp_clk" ,NULL , emac_ptp_mux, +ARRAY_SIZE (emac_ptp_mux),0 ,0x7C ,3 ,0 ,0 ,0 ,0x88 ,2 ,0 }, + {AGILEX5_GPIO_DB_CLK ,"gpio_db_clk" ,NULL , gpio_db_mux, +ARRAY_SIZE (gpio_db_mux),0 ,0x7C ,4 ,0x98 ,0 ,16 ,0x88 ,3 ,0 }, +/* Main Peripheral PLL1 End */ + +/* Peripheral clocks */ + {AGILEX5_S2F_USER0_CLK ,"s2f_user0_clk" ,NULL , s2f_user0_mux, +ARRAY_SIZE (s2f_user0_mux),0 ,0x24 ,6 ,0 ,0 ,0 ,0x30 ,2 ,0 }, + {AGILEX5_S2F_USER1_CLK ,"s2f_user1_clk" ,NULL , s2f_user1_mux, +ARRAY_SIZE (s2f_user1_mux),0 ,0x7C ,6 ,0 ,0 ,0 ,0x88 ,5 ,0 }, + {AGILEX5_PSI_REF_CLK ,"psi_ref_clk" ,NULL , psi_mux, +ARRAY_SIZE (psi_mux),0 ,0x7C ,7 ,0 ,0 ,0 ,0x88 ,6 ,0 }, + {AGILEX5_USB31_SUSPEND_CLK ,"usb31_suspend_clk" ,NULL , usb31_mux, +ARRAY_SIZE (usb31_mux),0 ,0x7C ,25 ,0 ,0 ,0 ,0x88 ,7 ,0 }, + {AGILEX5_USB31_BUS_CLK_EARLY ,"usb31_bus_clk_early" ,"l4_main_clk" , +NULL ,1 ,0 ,0x7C ,25 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_USB2OTG_HCLK ,"usb2otg_hclk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C , +8 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIM_0_CLK ,"spim_0_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C ,9 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIM_1_CLK ,"spim_1_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C ,11 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIS_0_CLK ,"spis_0_clk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,12 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIS_1_CLK ,"spis_1_clk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,13 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_DMA_CORE_CLK ,"dma_core_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C , +14 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_DMA_HS_CLK ,"dma_hs_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C ,14 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I3C_0_CORE_CLK ,"i3c_0_core_clk" ,"l4_mp_clk" ,NULL ,1 ,0 , +0x7C ,18 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I3C_1_CORE_CLK ,"i3c_1_core_clk" ,"l4_mp_clk" ,NULL ,1 ,0 , +0x7C ,19 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_0_PCLK ,"i2c_0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,15 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_1_PCLK ,"i2c_1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,16 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_EMAC0_PCLK ,"i2c_emac0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,17 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_EMAC1_PCLK ,"i2c_emac1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,22 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_EMAC2_PCLK ,"i2c_emac2_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,27 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_UART_0_PCLK ,"uart_0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,20 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_UART_1_PCLK ,"uart_1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,21 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPTIMER_0_PCLK ,"sptimer_0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,23 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPTIMER_1_PCLK ,"sptimer_1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,24 ,0 ,0 ,0 ,0 ,0 ,0 }, + +/*NAND, SD/MMC and SoftPHY overall clocking*/ + {AGILEX5_DFI_CLK ,"dfi_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0 ,0 ,0x44 ,16 , +2 ,0 ,0 ,0 }, + {AGILEX5_NAND_NF_CLK ,"nand_nf_clk" ,"dfi_clk" ,NULL ,1 ,0 ,0x7C ,10 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_NAND_BCH_CLK ,"nand_bch_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C , +10 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SDMMC_SDPHY_REG_CLK ,"sdmmc_sdphy_reg_clk" ,"l4_mp_clk" ,NULL , +1 ,0 ,0x7C ,5 ,0 ,0 ,0 ,0 ,0 ,0 },
drivers/clk/socfpga/clk-agilex5.c line 686
Quoted 10 lines
drivers/clk/socfpga/clk-agilex5.c new file +10
-0+ {AGILEX5_SDMCLK ,"sdmclk" ,"dfi_clk" ,NULL ,1 ,0 ,0x7C ,5 ,0 ,0 ,0 ,0 , + 0, 0 }, + { AGILEX5_SOFTPHY_REG_PCLK, "softphy_reg_pclk", "l4_mp_clk", NULL, 1, 0, + 0x7C, 26, 0, 0, 0, 0, 0, 0 }, + { AGILEX5_SOFTPHY_PHY_CLK, "softphy_phy_clk", "l4_mp_clk", NULL, 1, 0, + 0x7C, 26, 0x44, 16, 2, 0, 0, 0 }, + { AGILEX5_SOFTPHY_CTRL_CLK, "softphy_ctrl_clk", "dfi_clk", NULL, 1, 0, + 0x7C, 26, 0, 0, 0, 0, 0, 0 }, +}; +
As far as I can tell, there are very little differences between this and Agilex! Please reuse!
Quoted 209 lines
drivers/clk/socfpga/clk-agilex5.c new file +157
-0+static int +agilex5_clk_register_c_perip (const struct stratix10_perip_c_clock *clks, +int nums,struct stratix10_clock_data *data) +{ +struct clk_hw *hw_clk; +void __iomem *base = data->base; +int i; + +for (i =0 ; i < nums; i++) { + hw_clk =s10_register_periph (&clks[i], base); +if (IS_ERR (hw_clk)) { +pr_err ("%s: failed to register clock %s\n" , __func__, + clks[i].name); +continue ; + } + data->clk_data.hws[clks[i].id] = hw_clk; + } +return 0 ; +} + +static int +agilex5_clk_register_cnt_perip (const struct stratix10_perip_cnt_clock *clks, +int nums,struct stratix10_clock_data *data) +{ +struct clk_hw *hw_clk; +void __iomem *base = data->base; +int i; + +for (i =0 ; i < nums; i++) { + hw_clk =s10_register_cnt_periph (&clks[i], base); +if (IS_ERR (hw_clk)) { +pr_err ("%s: failed to register clock %s\n" , __func__, + clks[i].name); +continue ; + } + data->clk_data.hws[clks[i].id] = hw_clk; + } + +return 0 ; +} + +static int agilex5_clk_register_gate (const struct stratix10_gate_clock *clks, +int nums, +struct stratix10_clock_data *data) +{ +struct clk_hw *hw_clk; +void __iomem *base = data->base; +int i; + +for (i =0 ; i < nums; i++) { + hw_clk =agilex_register_gate (&clks[i], base); +if (IS_ERR (hw_clk)) { +pr_err ("%s: failed to register clock %s\n" , __func__, + clks[i].name); +continue ; + } + data->clk_data.hws[clks[i].id] = hw_clk; + } + +return 0 ; +} + +static int agilex5_clk_register_pll (const struct stratix10_pll_clock *clks, +int nums,struct stratix10_clock_data *data) +{ +struct clk_hw *hw_clk; +void __iomem *base = data->base; +int i; + +for (i =0 ; i < nums; i++) { + hw_clk =agilex5_register_pll (&clks[i], base); +if (IS_ERR (hw_clk)) { +pr_err ("%s: failed to register clock %s\n" , __func__, + clks[i].name); +continue ; + } + data->clk_data.hws[clks[i].id] = hw_clk; + } + +return 0 ; +} + +static int agilex5_clkmgr_init (struct platform_device *pdev) +{ +struct device_node *np = pdev->dev.of_node; +struct device *dev = &pdev->dev; +struct stratix10_clock_data *clk_data; +struct resource *res; +void __iomem *base; +int i, num_clks; + + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); + base =devm_ioremap_resource (dev, res); +if (IS_ERR (base)) +return PTR_ERR (base); + + num_clks =AGILEX5_NUM_CLKS ; + + clk_data =devm_kzalloc (dev, +struct_size (clk_data, clk_data.hws, num_clks), +GFP_KERNEL ); +if (!clk_data) +return -ENOMEM ; + +for (i =0 ; i < num_clks; i++) + clk_data->clk_data.hws[i] =ERR_PTR (-ENOENT ); + + clk_data->base = base; + clk_data->clk_data.num = num_clks; + +agilex5_clk_register_pll (agilex5_pll_clks,ARRAY_SIZE (agilex5_pll_clks), + clk_data); + +agilex5_clk_register_c_perip (agilex5_main_perip_c_clks, +ARRAY_SIZE (agilex5_main_perip_c_clks), + clk_data); + +agilex5_clk_register_cnt_perip (agilex5_main_perip_cnt_clks, +ARRAY_SIZE (agilex5_main_perip_cnt_clks), + clk_data); + +agilex5_clk_register_gate (agilex5_gate_clks, +ARRAY_SIZE (agilex5_gate_clks), clk_data); + +of_clk_add_hw_provider (np, of_clk_hw_onecell_get, &clk_data->clk_data); +return 0 ; +} + +static int agilex5_clkmgr_probe (struct platform_device *pdev) +{ +int (*probe_func)(struct platform_device *init_func); + + probe_func =of_device_get_match_data (&pdev->dev); +if (!probe_func) +return -ENODEV ; +return probe_func (pdev); +} + +static const struct of_device_id agilex5_clkmgr_match_table[] = { + { .compatible ="intel,agilex5-clkmgr" , .data = agilex5_clkmgr_init }, + {} +}; + +static struct platform_driver agilex5_clkmgr_driver = { + .probe = agilex5_clkmgr_probe, + .driver = { + .name ="agilex5-clkmgr" , + .suppress_bind_attrs = true, + .of_match_table = agilex5_clkmgr_match_table, + }, +}; + +static int __init agilex5_clk_init (void ) +{ +return platform_driver_register (&agilex5_clkmgr_driver); +} +core_initcall (agilex5_clk_init);drivers/clk/socfpga/clk-pll-s10.c +37
-0@@ -175,6 +175,14 @@ static const struct clk_ops agilex_clk_pll_ops = { .prepare = clk_pll_prepare, }; +/* TODO need to fix, Agilex5 SM requires change */ +static const struct clk_ops agilex5_clk_pll_ops = { +/* TODO This may require a custom Agilex5 implementation */ + .recalc_rate = agilex_clk_pll_recalc_rate, + .get_parent = clk_pll_get_parent, + .prepare = clk_pll_prepare, +}; +static const struct clk_ops clk_pll_ops = { .recalc_rate = clk_pll_recalc_rate, .get_parent = clk_pll_get_parent,@@ -304,3 +312,43 @@ struct clk_hw *n5x_register_pll(const struct stratix10_pll_clock *clks, }return hw_clk; } + +struct clk_hw *agilex5_register_pll (const struct stratix10_pll_clock *clks, +void __iomem *reg) +{ +struct clk_hw *hw_clk; +struct socfpga_pll *pll_clk; +struct clk_init_data init; +const char *name = clks->name; +int ret; + + pll_clk =kzalloc (sizeof (*pll_clk),GFP_KERNEL ); +if (WARN_ON (!pll_clk)) +return NULL ; + + pll_clk->hw.reg = reg + clks->offset; + +if (streq (name,SOCFPGA_BOOT_CLK )) + init.ops = &clk_boot_ops; +else + init.ops = &agilex5_clk_pll_ops; + + init.name = name; + init.flags = clks->flags; + + init.num_parents = clks->num_parents; + init.parent_names =NULL ; + init.parent_data = clks->parent_data; + pll_clk->hw.hw.init = &init; +
drivers/clk/socfpga/clk-pll-s10.c struct clk_hw *n5x_register_pll(const struct stratix10_pll_clock *clks, · line 353
Quoted 10 lines
drivers/clk/socfpga/clk-pll-s10.c +10
-0+ pll_clk->hw.bit_idx =SOCFPGA_PLL_POWER ; + hw_clk = &pll_clk->hw.hw; + + ret =clk_hw_register (NULL , hw_clk); +if (ret) { +kfree (pll_clk); +return ERR_PTR (ret); + } +return hw_clk; +}
Both functions are identical to Agilex, so why the need to recreate?
Quoted 6 lines
drivers/clk/socfpga/clk-pll-s10.c +1
-0+drivers/clk/socfpga/stratix10-clk.h +0
-0
drivers/clk/socfpga/stratix10-clk.h struct clk_hw *agilex_register_pll(const struct stratix10_pll_clock *clks, · line 84
Quoted 8 lines
drivers/clk/socfpga/stratix10-clk.h +2
-0@@ -77,6 +77,8 @@ struct clk_hw *agilex_register_pll(const struct stratix10_pll_clock *clks, void __iomem *reg);struct clk_hw *n5x_register_pll (const struct stratix10_pll_clock *clks,void __iomem *reg); +struct clk_hw *agilex5_register_pll (const struct stratix10_pll_clock *clks, +void __iomem *reg);struct clk_hw *s10_register_periph (const struct stratix10_perip_c_clock *clks,void __iomem *reg); struct clk_hw *n5x_register_periph(const struct n5x_perip_c_clock *clks,
I'd like for you to send this whole patchset for my internal review before you send out a V2!
Dinh
PATCH 4/4 arm64: dts: agilex5: Add initial support for Intel's Agilex5 SoCFPGA
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add the initial device tree files for Intel's Agilex5 SoCFPGA platform.
--- arch/arm64/boot/dts/intel/Makefile | 3 + .../arm64/boot/dts/intel/socfpga_agilex5.dtsi | 641 ++++++++++++++++++ .../boot/dts/intel/socfpga_agilex5_socdk.dts | 184 +++++ .../dts/intel/socfpga_agilex5_socdk_nand.dts | 131 ++++ .../dts/intel/socfpga_agilex5_socdk_swvp.dts | 248 +++++++ 5 files changed, 1207 insertions(+) create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_nand.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_swvp.dts
arch/arm64/boot/dts/intel/Makefile +3 -0
@@ -2,5 +2,8 @@ dtb-$(CONFIG_ARCH_INTEL_SOCFPGA) += socfpga_agilex_n6000.dtb \ socfpga_agilex_socdk.dtb \ socfpga_agilex_socdk_nand.dtb \ + socfpga_agilex5_socdk.dtb \ + socfpga_agilex5_socdk_nand.dtb \ + socfpga_agilex5_socdk_swvp.dtb \ socfpga_n5x_socdk.dtb dtb-$(CONFIG_ARCH_KEEMBAY) += keembay-evm.dtb
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +641 -0
641 lines not on this page · raw message
arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts new file +184 -0
184 lines not on this page · raw message
arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_nand.dts new file +131 -0
131 lines not on this page · raw message
arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_swvp.dts new file +248 -0
248 lines not on this page · raw message
Signature · 1 line
2.25.1
Re: [PATCH 4/4] arm64: dts: agilex5: Add initial support for Intel's Agilex5 SoCFPGA
Quoted 37 lines · On 18/06/2023 15:22, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add the initial device tree files for Intel's Agilex5 SoCFPGA platform.
Signed-off-by Adrian Ng Ho Yin <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- arch/arm64/boot/dts/intel/Makefile | 3 + .../arm64/boot/dts/intel/socfpga_agilex5.dtsi | 641 ++++++++++++++++++ .../boot/dts/intel/socfpga_agilex5_socdk.dts | 184 +++++ .../dts/intel/socfpga_agilex5_socdk_nand.dts | 131 ++++ .../dts/intel/socfpga_agilex5_socdk_swvp.dts | 248 +++++++ 5 files changed, 1207 insertions(+) create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_nand.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_swvp.dtsarch/arm64/boot/dts/intel/Makefile +3
-0@@ -2,5 +2,8 @@ dtb-$(CONFIG_ARCH_INTEL_SOCFPGA) += socfpga_agilex_n6000.dtb \ socfpga_agilex_socdk.dtb \ socfpga_agilex_socdk_nand.dtb \ + socfpga_agilex5_socdk.dtb \ + socfpga_agilex5_socdk_nand.dtb \ + socfpga_agilex5_socdk_swvp.dtb \ socfpga_n5x_socdk.dtb dtb-$(CONFIG_ARCH_KEEMBAY) += keembay-evm.dtbarch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +0
-0
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 1
Quoted 1 line
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +1
-0@@ -0,0 +1,641 @@ +// SPDX-License-Identifier: GPL-2.0
Drop indent before license.
Quoted 59 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +59
-0+/* + * Copyright (C) 2023, Intel Corporation + */ + +/dts-v1/; +#include <dt-bindings/reset/altr,rst-mgr-agilex5.h> +#include <dt-bindings/gpio/gpio.h> +#include <dt-bindings/interrupt-controller/arm-gic.h> +#include <dt-bindings/interrupt-controller/irq.h> +#include <dt-bindings/clock/agilex5-clock.h> + +/ { + compatible = "intel,socfpga-agilex"; + #address-cells = <2>; + #size-cells = <2>; + + reserved-memory { + #address-cells = <2>; + #size-cells = <2>; + ranges; + + service_reserved: svcbuffer@0 { + compatible = "shared-dma-pool"; + reg = <0x0 0x80000000 0x0 0x2000000>; + alignment = <0x1000>; + no-map; + }; + }; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu0: cpu@0 { + compatible = "arm,cortex-a55"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x0>; + }; + + cpu1: cpu@1 { + compatible = "arm,cortex-a55"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x100>; + }; + + cpu2: cpu@2 { + compatible = "arm,cortex-a76"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x200>; + }; + + cpu3: cpu@3 { + compatible = "arm,cortex-a76"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x300>;
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 70
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ }; + }; + + psci { + compatible = "arm,psci-0.2"; + method = "smc"; + }; + + intc: interrupt-controller@1d000000 { + compatible = "arm,gic-v3", "arm,cortex-a15-gic";
reg is always after compatible. Then ranges, if applicable.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 76
Quoted 6 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +6
-0+ #interrupt-cells = <3>; + #address-cells = <2>; + #size-cells =<2>; + interrupt-controller; + #redistributor-regions = <1>; + label = "GIC";
It does not look like you tested the DTS against bindings. Please run make dtbs_check (see Documentation/devicetree/bindings/writing-schema.rst or
https://www.linaro.org/blog/tips-and-tricks-for-validating-devicetree-sources-with-the-devicetree-schema/
for instructions).
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 77
Quoted 1 line
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +1
-0+ status = "okay";
Drop, you don't need status.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 88
Quoted 11 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +11
-0+ ranges; + redistributor-stride = <0x0 0x20000>; + reg = <0x0 0x1d000000 0 0x10000>, + <0x0 0x1d060000 0 0x100000>; + + its: msi-controller@1d040000 { + compatible = "arm,gic-v3-its"; + reg = <0x0 0x1d040000 0x0 0x20000>; + label = "ITS"; + msi-controller; + status = "okay";
Drop
Anyway, entire node should be in soc. You clearly did not test it with dtbs W=1. Neither with dtbs_check.
Quoted 57 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +57
-0+ }; + }; + + /* Clock tree 5 main sources*/ + clocks { + cb_intosc_hs_div2_clk: cb-intosc-hs-div2-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + cb_intosc_ls_clk: cb-intosc-ls-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + f2s_free_clk: f2s-free-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + osc1: osc1 { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + qspi_clk: qspi-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + clock-frequency = <200000000>; + }; + }; + + timer { + compatible = "arm,armv8-timer"; + interrupt-parent = <&intc>; + interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>; + }; + + usbphy0: usbphy { + #phy-cells = <0>; + compatible = "usb-nop-xceiv"; + }; + + soc { + #address-cells = <1>; + #size-cells = <1>; + compatible = "simple-bus"; + device_type = "soc"; + interrupt-parent = <&intc>; + ranges = <0 0 0 0xffffffff>; + + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>;
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 155
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ #clock-cells = <1>; + }; + + stmmac_axi_setup: stmmac-axi-config { + snps,wr_osr_lmt = <31>; + snps,rd_osr_lmt = <31>; + snps,blen = <0 0 0 32 16 8 4>; + }; + + mtl_rx_setup: rx-queues-config {
These two nodes do not belong to SoC. Soc is for MMIO-based nodes.
Quoted 27 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +27
-0+ snps,rx-queues-to-use = <8>; + snps,rx-sched-sp; + queue0 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x0>; + }; + queue1 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x1>; + }; + queue2 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x2>; + }; + queue3 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x3>; + }; + queue4 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x4>; + }; + queue5 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x5>; + }; + queue6 {
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 192
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ snps,dcb-algorithm; + snps,map-to-dma-channel = <0x6>; + }; + queue7 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x7>; + }; + }; + + mtl_tx_setup: tx-queues-config {
This as well
Quoted 30 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +30
-0+ snps,tx-queues-to-use = <8>; + snps,tx-sched-wrr; + queue0 { + snps,weight = <0x09>; + snps,dcb-algorithm; + }; + queue1 { + snps,weight = <0x0A>; + snps,dcb-algorithm; + }; + queue2 { + snps,weight = <0x0B>; + snps,dcb-algorithm; + }; + queue3 { + snps,weight = <0x0C>; + snps,dcb-algorithm; + }; + queue4 { + snps,weight = <0x0D>; + snps,dcb-algorithm; + }; + queue5 { + snps,weight = <0x0E>; + snps,dcb-algorithm; + }; + queue6 { + snps,weight = <0x0F>; + snps,dcb-algorithm; + };
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 232
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ queue7 { + snps,weight = <0x10>; + snps,dcb-algorithm; + }; + }; + + gmac0: ethernet@10810000 { + compatible = "altr,socfpga-stmmac-a10-s10", + "snps,dwxgmac-2.10", + "snps,dwxgmac";
You have broken alignment.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 236
Quoted 4 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +4
-0+ reg = <0x10810000 0x3500>; + interrupts = <GIC_SPI 190 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 191 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 192 IRQ_TYPE_LEVEL_HIGH>,
In multiple places. Really.
Quoted 6 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +6
-0+ <GIC_SPI 193 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 194 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 195 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 196 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 197 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 198 IRQ_TYPE_LEVEL_HIGH>,
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 252
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ <GIC_SPI 199 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 200 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 201 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 202 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 203 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 204 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 205 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 206 IRQ_TYPE_LEVEL_HIGH>; + interrupt-names = "macirq", + "macirq_tx0",
This is ugly.
Quoted 6 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +6
-0+ "macirq_tx1", + "macirq_tx2", + "macirq_tx3", + "macirq_tx4", + "macirq_tx5", + "macirq_tx6",
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 268
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ "macirq_tx7", + "macirq_rx0", + "macirq_rx1", + "macirq_rx2", + "macirq_rx3", + "macirq_rx4", + "macirq_rx5", + "macirq_rx6", + "macirq_rx7"; + mac-address = [00 00 00 00 00 00];
Drop, it's SoC file.
Quoted 14 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +14
-0+ resets = <&rst EMAC0_RESET>, <&rst EMAC0_OCP_RESET>; + reset-names = "stmmaceth", "stmmaceth-ocp"; + tx-fifo-depth = <32768>; + rx-fifo-depth = <16384>; + snps,multicast-filter-bins = <64>; + snps,perfect-filter-entries = <64>; + snps,axi-config = <&stmmac_axi_setup>; + snps,mtl-rx-config = <&mtl_rx_setup>; + snps,mtl-tx-config = <&mtl_tx_setup>; + snps,pbl = <32>; + snps,pblx8; + snps,multi-irq-en; + snps,tso; + altr,sysmgr-syscon = <&sysmgr 0x44 0>;
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 292
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ altr,smtg-hub; + snps,rx-vlan-offload; + clocks = <&clkmgr AGILEX5_EMAC0_CLK>, <&clkmgr AGILEX5_EMAC_PTP_CLK>; + clock-names = "stmmaceth", "ptp_ref"; + status = "disabled"; + }; + + i2c0: i2c@10c02800 { + #address-cells = <1>; + #size-cells = <0>;
compatible first, reg second.
Quoted 85 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +85
-0+ compatible = "snps,designware-i2c"; + reg = <0x10c02800 0x100>; + interrupts = <GIC_SPI 103 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C0_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c1: i2c@10c02900 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02900 0x100>; + interrupts = <GIC_SPI 104 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C1_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c2: i2c@10c02a00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02a00 0x100>; + interrupts = <GIC_SPI 105 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C2_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c3: i2c@10c02b00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02b00 0x100>; + interrupts = <GIC_SPI 106 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C3_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c4: i2c@10c02c00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02c00 0x100>; + interrupts = <GIC_SPI 107 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C4_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i3c0: i3c@10da0000 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,dw-i3c-master-1.00a"; + reg = <0x10da0000 0x1000>; + interrupts = <GIC_SPI 164 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I3C0_RESET>; + clocks = <&clkmgr AGILEX5_L4_MP_CLK>; + status = "disabled"; + }; + + i3c1: i3c@10da1000 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,dw-i3c-master-1.00a"; + reg = <0x10da1000 0x1000>; + interrupts = <GIC_SPI 165 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I3C1_RESET>; + clocks = <&clkmgr AGILEX5_L4_MP_CLK>; + status = "disabled"; + }; + + gpio1: gpio@10C03300 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,dw-apb-gpio"; + reg = <0x10C03300 0x100>; + resets = <&rst GPIO1_RESET>; + status = "disabled"; + + portb: gpio-controller@0 { + compatible = "snps,dw-apb-gpio-port"; + gpio-controller;
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 387
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ #gpio-cells = <2>; + snps,nr-gpios = <24>; + reg = <0>; + interrupt-controller; + #interrupt-cells = <2>; + interrupts = <GIC_SPI 111 IRQ_TYPE_LEVEL_HIGH>; + }; + }; + + mmc: mmc0@10808000 {
It does not look like you tested the DTS against bindings. Please run make dtbs_check (see Documentation/devicetree/bindings/writing-schema.rst or
https://www.linaro.org/blog/tips-and-tricks-for-validating-devicetree-sources-with-the-devicetree-schema/
for instructions).
Sorry, but your DTS lacks any basic (internal!) review and basic tests with automated tools.
..
Quoted 7 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +7
-0+ spi0: spi@10da4000 { + compatible = "snps,dw-apb-ssi"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0x10da4000 0x1000>; + interrupts = <GIC_SPI 99 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst SPIM0_RESET>;
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 479
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ reset-names = "spi"; + reg-io-width = <4>; + num-cs = <4>; + clocks = <&clkmgr AGILEX5_L4_MAIN_CLK>; + dmas = <&dmac0 2>, <&dmac0 3>; + dma-names ="tx", "rx"; + + status = "disabled"; + + flash: m25p128@0 {
It's getting worse...
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 405
Quoted 1 line
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +1
-0+ status = "okay";
and worse
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 491
Quoted 11 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +11
-0+ compatible = "st,m25p80"; + spi-max-frequency = <25000000>; + m25p,fast-read; + reg = <0>; + + #address-cells = <1>; + #size-cells = <1>; + + partition@0 { + label = "spi_flash_part0"; + reg = <0x0 0x100000>;
and less conforming to style
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 30
Quoted 3 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +3
-0+ }; + }; +
... and less.
...
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 426
Quoted 7 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +7
-0+ + chosen { + stdout-path = "serial0:115200n8"; + bootargs = "console=uart8250,mmio32,0x10c02000,115200n8 \ + root=/dev/ram0 rw initrd=0x10000000 init=/sbin/init \ + ramdisk_size=10000000 earlycon=uart8250,mmio32,0x10c02000,115200n8 \ + panic=-1 nosmp rootfstype=ext3";
NAK. Drop entire bootags.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 431
Quoted 5 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +5
-0+ }; + + leds { + compatible = "gpio-leds"; + hps0 {
No, srsly? So all our cleanups for few years mean nothing?
Best regards,
Krzysztof
Re: [PATCH 4/4] arm64: dts: agilex5: Add initial support for Intel's Agilex5 SoCFPGA
Quoted 339 lines · On 18/06/2023 15:22, [email protected] wrote:
Quoted 42 lines
From: Niravkumar L Rabara <[email protected]>
Add the initial device tree files for Intel's Agilex5 SoCFPGA platform.
Signed-off-by Adrian Ng Ho Yin <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- arch/arm64/boot/dts/intel/Makefile | 3 + .../arm64/boot/dts/intel/socfpga_agilex5.dtsi | 641 ++++++++++++++++++ .../boot/dts/intel/socfpga_agilex5_socdk.dts | 184 +++++ .../dts/intel/socfpga_agilex5_socdk_nand.dts | 131 ++++ .../dts/intel/socfpga_agilex5_socdk_swvp.dts | 248 +++++++ 5 files changed, 1207 insertions(+) create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_nand.dts create mode 100644
arch/arm64/boot/dts/intel/socfpga_agilex5_socdk_swvp.dtsarch/arm64/boot/dts/intel/Makefile +3
-0@@ -2,5 +2,8 @@ dtb-$(CONFIG_ARCH_INTEL_SOCFPGA) += socfpga_agilex_n6000.dtb \ socfpga_agilex_socdk.dtb \ socfpga_agilex_socdk_nand.dtb \ + socfpga_agilex5_socdk.dtb \ + socfpga_agilex5_socdk_nand.dtb \ + socfpga_agilex5_socdk_swvp.dtb \ socfpga_n5x_socdk.dtb dtb-$(CONFIG_ARCH_KEEMBAY) += keembay-evm.dtb diff --git
a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsib/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
new file mode 100644
index 000000000000..9454d88d6457arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 1
Quoted 1 line
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +1
-0@@ -0,0 +1,641 @@ +// SPDX-License-Identifier: GPL-2.0Drop indent before license.
Quoted 58 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +58
-0+/* + * Copyright (C) 2023, Intel Corporation */ + +/dts-v1/; +#include <dt-bindings/reset/altr,rst-mgr-agilex5.h> +#include <dt-bindings/gpio/gpio.h> +#include <dt-bindings/interrupt-controller/arm-gic.h> +#include <dt-bindings/interrupt-controller/irq.h> +#include <dt-bindings/clock/agilex5-clock.h> + +/ { + compatible = "intel,socfpga-agilex"; + #address-cells = <2>; + #size-cells = <2>; + + reserved-memory { + #address-cells = <2>; + #size-cells = <2>; + ranges; + + service_reserved: svcbuffer@0 { + compatible = "shared-dma-pool"; + reg = <0x0 0x80000000 0x0 0x2000000>; + alignment = <0x1000>; + no-map; + }; + }; + + cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu0: cpu@0 { + compatible = "arm,cortex-a55"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x0>; + }; + + cpu1: cpu@1 { + compatible = "arm,cortex-a55"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x100>; + }; + + cpu2: cpu@2 { + compatible = "arm,cortex-a76"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x200>; + }; + + cpu3: cpu@3 { + compatible = "arm,cortex-a76"; + device_type = "cpu"; + enable-method = "psci"; + reg = <0x300>;arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 70
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ }; + }; + + psci { + compatible = "arm,psci-0.2"; + method = "smc"; + }; + + intc: interrupt-controller@1d000000 { + compatible = "arm,gic-v3", "arm,cortex-a15-gic";reg is always after compatible. Then ranges, if applicable.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 76
Quoted 6 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +6
-0+ #interrupt-cells = <3>; + #address-cells = <2>; + #size-cells =<2>; + interrupt-controller; + #redistributor-regions = <1>; + label = "GIC";It does not look like you tested the DTS against bindings. Please run
make dtbs_check(seeDocumentation/devicetree/bindings/writing-schema.rstor https://www.linaro.org/blog/tips-and-tricks-for-validating-devicetree-sources-with-the-devicetree-schema/
for instructions).arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 76
Quoted 1 line
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +1
-0+ status = "okay";Drop, you don't need status.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 88
Quoted 11 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +11
-0+ ranges; + redistributor-stride = <0x0 0x20000>; + reg = <0x0 0x1d000000 0 0x10000>, + <0x0 0x1d060000 0 0x100000>; + + its: msi-controller@1d040000 { + compatible = "arm,gic-v3-its"; + reg = <0x0 0x1d040000 0x0 0x20000>; + label = "ITS"; + msi-controller; + status = "okay";Drop
Anyway, entire node should be in soc. You clearly did not test it with dtbs W=1. Neither with
dtbs_check.Quoted 57 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +57
-0+ }; + }; + + /* Clock tree 5 main sources*/ + clocks { + cb_intosc_hs_div2_clk: cb-intosc-hs-div2-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + cb_intosc_ls_clk: cb-intosc-ls-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + f2s_free_clk: f2s-free-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + osc1: osc1 { + #clock-cells = <0>; + compatible = "fixed-clock"; + }; + + qspi_clk: qspi-clk { + #clock-cells = <0>; + compatible = "fixed-clock"; + clock-frequency = <200000000>; + }; + }; + + timer { + compatible = "arm,armv8-timer"; + interrupt-parent = <&intc>; + interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>, + <GIC_PPI 10 (GIC_CPU_MASK_SIMPLE(4) | IRQ_TYPE_LEVEL_LOW)>; + }; + + usbphy0: usbphy { + #phy-cells = <0>; + compatible = "usb-nop-xceiv"; + }; + + soc { + #address-cells = <1>; + #size-cells = <1>; + compatible = "simple-bus"; + device_type = "soc"; + interrupt-parent = <&intc>; + ranges = <0 0 0 0xffffffff>; + + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>;arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 155
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ #clock-cells = <1>; + }; + + stmmac_axi_setup: stmmac-axi-config { + snps,wr_osr_lmt = <31>; + snps,rd_osr_lmt = <31>; + snps,blen = <0 0 0 32 16 8 4>; + }; + + mtl_rx_setup: rx-queues-config {These two nodes do not belong to SoC. Soc is for MMIO-based nodes.
Quoted 27 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +27
-0+ snps,rx-queues-to-use = <8>; + snps,rx-sched-sp; + queue0 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x0>; + }; + queue1 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x1>; + }; + queue2 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x2>; + }; + queue3 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x3>; + }; + queue4 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x4>; + }; + queue5 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x5>; + }; + queue6 {arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 192
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ snps,dcb-algorithm; + snps,map-to-dma-channel = <0x6>; + }; + queue7 { + snps,dcb-algorithm; + snps,map-to-dma-channel = <0x7>; + }; + }; + + mtl_tx_setup: tx-queues-config {This as well
Quoted 30 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +30
-0+ snps,tx-queues-to-use = <8>; + snps,tx-sched-wrr; + queue0 { + snps,weight = <0x09>; + snps,dcb-algorithm; + }; + queue1 { + snps,weight = <0x0A>; + snps,dcb-algorithm; + }; + queue2 { + snps,weight = <0x0B>; + snps,dcb-algorithm; + }; + queue3 { + snps,weight = <0x0C>; + snps,dcb-algorithm; + }; + queue4 { + snps,weight = <0x0D>; + snps,dcb-algorithm; + }; + queue5 { + snps,weight = <0x0E>; + snps,dcb-algorithm; + }; + queue6 { + snps,weight = <0x0F>; + snps,dcb-algorithm; + };arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 232
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ queue7 { + snps,weight = <0x10>; + snps,dcb-algorithm; + }; + }; + + gmac0: ethernet@10810000 { + compatible = "altr,socfpga-stmmac-a10-s10", + "snps,dwxgmac-2.10", + "snps,dwxgmac";You have broken alignment.
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 236
Quoted 4 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +4
-0+ reg = <0x10810000 0x3500>; + interrupts = <GIC_SPI 190 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 191 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 192 IRQ_TYPE_LEVEL_HIGH>,In multiple places. Really.
Quoted 6 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +6
-0+ <GIC_SPI 193 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 194 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 195 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 196 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 197 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 198 IRQ_TYPE_LEVEL_HIGH>,arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 252
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ <GIC_SPI 199 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 200 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 201 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 202 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 203 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 204 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 205 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 206 IRQ_TYPE_LEVEL_HIGH>; + interrupt-names = "macirq", + "macirq_tx0",This is ugly.
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ "macirq_tx1", + "macirq_tx2", + "macirq_tx3", + "macirq_tx4", + "macirq_tx5", + "macirq_tx6", + "macirq_tx7", + "macirq_rx0", + "macirq_rx1", + "macirq_rx2",
Quoted 8 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 268
Quoted 6 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +6
-0+ "macirq_rx3", + "macirq_rx4", + "macirq_rx5", + "macirq_rx6", + "macirq_rx7"; + mac-address = [00 00 00 00 00 00];Drop, it's SoC file.
Will remove these nodes, ethernet related changes will be submitted in separate patch later.
Quoted 186 lines
Quoted 15 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +14
-0+ resets = <&rst EMAC0_RESET>, <&rst EMAC0_OCP_RESET>; + reset-names = "stmmaceth", "stmmaceth-ocp"; + tx-fifo-depth = <32768>; + rx-fifo-depth = <16384>; + snps,multicast-filter-bins = <64>; + snps,perfect-filter-entries = <64>; + snps,axi-config = <&stmmac_axi_setup>; + snps,mtl-rx-config = <&mtl_rx_setup>; + snps,mtl-tx-config = <&mtl_tx_setup>; + snps,pbl = <32>; + snps,pblx8; + snps,multi-irq-en; + snps,tso; + altr,sysmgr-syscon = <&sysmgr 0x44 0>;arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 292
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ altr,smtg-hub; + snps,rx-vlan-offload; + clocks = <&clkmgr AGILEX5_EMAC0_CLK>, <&clkmgr AGILEX5_EMAC_PTP_CLK>; + clock-names = "stmmaceth", "ptp_ref"; + status = "disabled"; + }; + + i2c0: i2c@10c02800 { + #address-cells = <1>; + #size-cells = <0>;compatible first, reg second.
Quoted 85 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +85
-0+ compatible = "snps,designware-i2c"; + reg = <0x10c02800 0x100>; + interrupts = <GIC_SPI 103 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C0_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c1: i2c@10c02900 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02900 0x100>; + interrupts = <GIC_SPI 104 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C1_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c2: i2c@10c02a00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02a00 0x100>; + interrupts = <GIC_SPI 105 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C2_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c3: i2c@10c02b00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02b00 0x100>; + interrupts = <GIC_SPI 106 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C3_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i2c4: i2c@10c02c00 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,designware-i2c"; + reg = <0x10c02c00 0x100>; + interrupts = <GIC_SPI 107 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I2C4_RESET>; + clocks = <&clkmgr AGILEX5_L4_SP_CLK>; + status = "disabled"; + }; + + i3c0: i3c@10da0000 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,dw-i3c-master-1.00a"; + reg = <0x10da0000 0x1000>; + interrupts = <GIC_SPI 164 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I3C0_RESET>; + clocks = <&clkmgr AGILEX5_L4_MP_CLK>; + status = "disabled"; + }; + + i3c1: i3c@10da1000 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,dw-i3c-master-1.00a"; + reg = <0x10da1000 0x1000>; + interrupts = <GIC_SPI 165 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst I3C1_RESET>; + clocks = <&clkmgr AGILEX5_L4_MP_CLK>; + status = "disabled"; + }; + + gpio1: gpio@10C03300 { + #address-cells = <1>; + #size-cells = <0>; + compatible = "snps,dw-apb-gpio"; + reg = <0x10C03300 0x100>; + resets = <&rst GPIO1_RESET>; + status = "disabled"; + + portb: gpio-controller@0 { + compatible = "snps,dw-apb-gpio-port"; + gpio-controller;arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 387
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ #gpio-cells = <2>; + snps,nr-gpios = <24>; + reg = <0>; + interrupt-controller; + #interrupt-cells = <2>; + interrupts = <GIC_SPI 111 IRQ_TYPE_LEVEL_HIGH>; + }; + }; + + mmc: mmc0@10808000 {It does not look like you tested the DTS against bindings. Please run
make dtbs_check(seeDocumentation/devicetree/bindings/writing-schema.rstor https://www.linaro.org/blog/tips-and-tricks-for-validating-devicetree-sources-with-the-devicetree-schema/
for instructions).Sorry, but your DTS lacks any basic (internal!) review and basic tests with automated tools.
..
Quoted 7 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +7
-0+ spi0: spi@10da4000 { + compatible = "snps,dw-apb-ssi"; + #address-cells = <1>; + #size-cells = <0>; + reg = <0x10da4000 0x1000>; + interrupts = <GIC_SPI 99 IRQ_TYPE_LEVEL_HIGH>; + resets = <&rst SPIM0_RESET>;arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 479
Quoted 10 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +10
-0+ reset-names = "spi"; + reg-io-width = <4>; + num-cs = <4>; + clocks = <&clkmgr AGILEX5_L4_MAIN_CLK>; + dmas = <&dmac0 2>, <&dmac0 3>; + dma-names ="tx", "rx"; + + status = "disabled"; + + flash: m25p128@0 {It's getting worse...
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 404
Quoted 1 line
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +1
-0+ status = "okay";and worse
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 491
Quoted 11 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +11
-0+ compatible = "st,m25p80"; + spi-max-frequency = <25000000>; + m25p,fast-read; + reg = <0>; + + #address-cells = <1>; + #size-cells = <1>; + + partition@0 { + label = "spi_flash_part0"; + reg = <0x0 0x100000>;and less conforming to style
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 30
Quoted 3 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +3
-0+ }; + }; +... and less.
...arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 425
Quoted 7 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +7
-0+ + chosen { + stdout-path = "serial0:115200n8"; + bootargs = "console=uart8250,mmio32,0x10c02000,115200n8 \ + root=/dev/ram0 rw initrd=0x10000000 init=/sbin/init \ + ramdisk_size=10000000 earlycon=uart8250,mmio32,0x10c02000,115200n8 \ + panic=-1 nosmp rootfstype=ext3";NAK. Drop entire bootags.
Quoted 8 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi line 430
Quoted 5 lines
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +5
-0+ }; + + leds { + compatible = "gpio-leds"; + hps0 {No, srsly? So all our cleanups for few years mean nothing?
Will fix all the review comments in v2 patch.
Thanks,
Nirav
v2 cover Add support for Agilex5 SoCFPGA platform
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
patch [1/5] - Introduced compatible string for Agilex5 board. patch [2/5] - Add Agilex5 reset ID definitions.
patch [3/5] - Add Agilex5 clock manager header and yaml file. patch [4/5] - Reused and modified Agilex clock manager driver for Agilex5 to avoid code duplication. This patch depends on patch 4. patch [5/5] - Add device tree files for Agilex5 platform. This patch depends on patch 1,2,3 & 4.
patch v2 changes:-
- Add separate discription and const for Agilex5 board in yaml file.
- Add reset ID definitions required for Agilex5 and reused altr,rst-mgr-s10 bindings similar to Agilex.
- Instead of creating separate clock manager driver, re-use agilex clock manager driver and modified it for agilex5 changes to avoid code duplicate.
- Fixed device tree alignment issues and other build warnings. Removed ethernet nodes as it will be included in a separate patch.
Niravkumar L Rabara (5):
dt-bindings: intel: Add Intel Agilex5 compatible
dt-bindings: reset: add reset IDs for Agilex5
dt-bindings: clock: add Intel Agilex5 clock manager
clk: socfpga: agilex: add clock driver for the Agilex5
arm64: dts: agilex5: add initial support for Intel Agilex5 SoCFPGA
.../bindings/arm/intel,socfpga.yaml | 5 + .../bindings/clock/intel,agilex5-clkmgr.yaml | 41 ++ arch/arm64/boot/dts/intel/Makefile | 1 + .../arm64/boot/dts/intel/socfpga_agilex5.dtsi | 468 ++++++++++++++++++ .../boot/dts/intel/socfpga_agilex5_socdk.dts | 39 ++ drivers/clk/socfpga/clk-agilex.c | 433 +++++++++++++++- .../dt-bindings/clock/intel,agilex5-clkmgr.h | 100 ++++ include/dt-bindings/reset/altr,rst-mgr-s10.h | 5 +- 8 files changed, 1089 insertions(+), 3 deletions(-) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts create mode 100644 include/dt-bindings/clock/intel,agilex5-clkmgr.h
Signature · 1 line
2.25.1
PATCH v2 1/5 dt-bindings: intel: Add Intel Agilex5 compatible
From: Niravkumar L Rabara <[email protected]>
Agilex5 is a new SoCFPGA in Intel Agilex SoCFPGA Family, include compatible string for Agilex5 SoCFPGA board.
--- Documentation/devicetree/bindings/arm/intel,socfpga.yaml | 5 +++++ 1 file changed, 5 insertions(+)
Documentation/devicetree/bindings/arm/intel,socfpga.yaml +5 -0
@@ -21,6 +21,11 @@ properties: - intel,socfpga-agilex-n6000 - intel,socfpga-agilex-socdk - const: intel,socfpga-agilex + - description: Agilex5 boards + items: + - enum: + - intel,socfpga-agilex5-socdk + - const: intel,socfpga-agilex5 additionalProperties: true
Signature · 1 line
2.25.1
Re: [PATCH v2 1/5] dt-bindings: intel: Add Intel Agilex5 compatible
Quoted 7 lines · On Tue, Aug 01, 2023 at 09:02:30AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Agilex5 is a new SoCFPGA in Intel Agilex SoCFPGA Family, include compatible string for Agilex5 SoCFPGA board.
Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>
Cheers,
Conor.
Quoted 23 lines
--- Documentation/devicetree/bindings/arm/intel,socfpga.yaml | 5 +++++ 1 file changed, 5 insertions(+)Documentation/devicetree/bindings/arm/intel,socfpga.yaml +5
-0@@ -21,6 +21,11 @@ properties: - intel,socfpga-agilex-n6000 - intel,socfpga-agilex-socdk - const: intel,socfpga-agilex + - description: Agilex5 boards + items: + - enum: + - intel,socfpga-agilex5-socdk + - const: intel,socfpga-agilex5 additionalProperties: trueSignature · 1 line
2.25.1
PATCH v2 2/5 dt-bindings: reset: add reset IDs for Agilex5
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add reset ID definitions required for Intel Agilex5 SoCFPGA, re-use altr,rst-mgr-s10.h as common header file similar S10 & Agilex.
--- include/dt-bindings/reset/altr,rst-mgr-s10.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
include/dt-bindings/reset/altr,rst-mgr-s10.h +4 -1
@@ -63,12 +63,15 @@#define I2C2_RESET 74#define I2C3_RESET 75#define I2C4_RESET 76 -/* 77-79 is empty */ +#define I3C0_RESET 77 +#define I3C1_RESET 78 +/* 79 is empty */ #define UART0_RESET 80#define UART1_RESET 81/* 82-87 is empty */ #define GPIO0_RESET 88#define GPIO1_RESET 89 +#define WATCHDOG4_RESET 90/* BRGMODRST */ #define SOC2FPGA_RESET 96
Signature · 1 line
2.25.1
Re: [PATCH v2 2/5] dt-bindings: reset: add reset IDs for Agilex5
Quoted 7 lines · On Tue, Aug 01, 2023 at 09:02:31AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add reset ID definitions required for Intel Agilex5 SoCFPGA, re-use altr,rst-mgr-s10.h as common header file similar S10 & Agilex.
Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>
Cheers,
Conor.
Quoted 28 lines
--- include/dt-bindings/reset/altr,rst-mgr-s10.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)include/dt-bindings/reset/altr,rst-mgr-s10.h +4
-1@@ -63,12 +63,15 @@#define I2C2_RESET 74#define I2C3_RESET 75#define I2C4_RESET 76 -/* 77-79 is empty */ +#define I3C0_RESET 77 +#define I3C1_RESET 78 +/* 79 is empty */ #define UART0_RESET 80#define UART1_RESET 81/* 82-87 is empty */ #define GPIO0_RESET 88#define GPIO1_RESET 89 +#define WATCHDOG4_RESET 90/* BRGMODRST */ #define SOC2FPGA_RESET 96Signature · 1 line
2.25.1
PATCH v2 3/5 dt-bindings: clock: add Intel Agilex5 clock manager
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.
--- .../bindings/clock/intel,agilex5-clkmgr.yaml | 41 +++++++ .../dt-bindings/clock/intel,agilex5-clkmgr.h | 100 ++++++++++++++++++ 2 files changed, 141 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml create mode 100644 include/dt-bindings/clock/intel,agilex5-clkmgr.h
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml new file +41 -0
@@ -0,0 +1,41 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5-clkmgr.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 clock manager + +maintainers: + - Dinh Nguyen <[email protected]> + +description: + The Intel Agilex5 Clock Manager is an integrated clock controller, which + generates and supplies clock to all the modules. + +properties: + compatible: + const: intel,agilex5-clkmgr + + reg: + maxItems: 1 + + '#clock-cells': + const: 1 + +required: + - compatible + - reg + - '#clock-cells' + +additionalProperties: false + +examples: + # Clock controller node: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...
include/dt-bindings/clock/intel,agilex5-clkmgr.h new file +100 -0
100 lines not on this page · raw message
Signature · 1 line
2.25.1
Re: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 46 lines · On Tue, Aug 01, 2023 at 09:02:32AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- .../bindings/clock/intel,agilex5-clkmgr.yaml | 41 +++++++ .../dt-bindings/clock/intel,agilex5-clkmgr.h | 100 ++++++++++++++++++ 2 files changed, 141 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml create mode 100644 include/dt-bindings/clock/intel,agilex5-clkmgr.hDocumentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml new file +25
-0@@ -0,0 +1,41 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5-clkmgr.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 clock manager + +maintainers: + - Dinh Nguyen <[email protected]> + +description: + The Intel Agilex5 Clock Manager is an integrated clock controller, which + generates and supplies clock to all the modules. + +properties: + compatible: + const: intel,agilex5-clkmgr + + reg: + maxItems: 1 + + '#clock-cells': + const: 1 +
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml line 34
Quoted 10 lines
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml new file +9
-0+required: + - compatible + - reg + - '#clock-cells' + +additionalProperties: false + +examples: + # Clock controller node:
This comment seems utterly pointless.
Otherwise this looks okay to me.
Thanks,
Conor.
Quoted 7 lines
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml new file +7
-0+ - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...
RE: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 77 lines
-----Original Message-----
From: Conor Dooley <[email protected]>
Sent: Wednesday, 2 August, 2023 4:58 AM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock managerQuoted 50 lines · On Tue, Aug 01, 2023 at 09:02:32AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- .../bindings/clock/intel,agilex5-clkmgr.yaml | 41 +++++++ .../dt-bindings/clock/intel,agilex5-clkmgr.h | 100 ++++++++++++++++++ 2 files changed, 141 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml create mode 100644 include/dt-bindings/clock/intel,agilex5-clkmgr.hdiff --git
a/Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yamlb/Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yamlnew file mode 100644
index 000000000000..60e57a9fb939Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yam new file +26
-0+++ l@@ -0,0 +1,41 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) %YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5-clkmgr.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 clock manager + +maintainers: + - Dinh Nguyen <[email protected]> + +description: + The Intel Agilex5 Clock Manager is an integrated clock controller, +which + generates and supplies clock to all the modules. + +properties: + compatible: + const: intel,agilex5-clkmgr + + reg: + maxItems: 1 + + '#clock-cells': + const: 1 +Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yam
Quoted 10 lines
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yam new file +9
-0+required: + - compatible + - reg + - '#clock-cells' + +additionalProperties: false + +examples: + # Clock controller node:
Quoted 8 lines
This comment seems utterly pointless.
Otherwise this looks okay to me.Reviewed-by Conor Dooley <[email protected]>Thanks,
Conor.
Removed in [PATCH v3 3/5].
Thanks,
Nirav
Quoted 7 lines
Quoted 7 lines
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yam new file +7
-0+ - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...
Re: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 11 lines · Niravkumar L
Quoted 11 lines
From: Conor Dooley <[email protected]>
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml line 34
Quoted 3 lines · On Tue, Aug 01, 2023 at 09:02:32AM +0800, [email protected] wrote:
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml new file +2
-0+examples: + # Clock controller node:This comment seems utterly pointless.
Otherwise this looks okay to me.
Quoted 7 lines
Quoted 5 lines
Reviewed-by Conor Dooley <[email protected]>Thanks,
Conor.Removed in [PATCH v3 3/5].
To be clear, you don't need to send a v3 just for that - I gave you the reviewed-by after all.
PATCH v3 3/5 dt-bindings: clock: add Intel Agilex5 clock manager
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.
--- .../bindings/clock/intel,agilex5-clkmgr.yaml | 40 +++++++ .../dt-bindings/clock/intel,agilex5-clkmgr.h | 100 ++++++++++++++++++ 2 files changed, 140 insertions(+) create mode 100644 Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml create mode 100644 include/dt-bindings/clock/intel,agilex5-clkmgr.h
Documentation/devicetree/bindings/clock/intel,agilex5-clkmgr.yaml new file +40 -0
@@ -0,0 +1,40 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/clock/intel,agilex5-clkmgr.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Intel SoCFPGA Agilex5 clock manager + +maintainers: + - Dinh Nguyen <[email protected]> + +description: + The Intel Agilex5 Clock Manager is an integrated clock controller, which + generates and supplies clock to all the modules. + +properties: + compatible: + const: intel,agilex5-clkmgr + + reg: + maxItems: 1 + + '#clock-cells': + const: 1 + +required: + - compatible + - reg + - '#clock-cells' + +additionalProperties: false + +examples: + - | + clkmgr: clock-controller@10d10000 { + compatible = "intel,agilex5-clkmgr"; + reg = <0x10d10000 0x1000>; + #clock-cells = <1>; + }; +...
include/dt-bindings/clock/intel,agilex5-clkmgr.h new file +100 -0
100 lines not on this page · raw message
Signature · 1 line
2.25.1
Re: [PATCH v3 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 9 lines · On Wed, Aug 02, 2023 at 10:58:42AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Reviewed-by Conor Dooley <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>
Damn, I was too late - you already sent a v3 :/
However, there only seems to be a v3 of this one patch and it was sent in reply to the v2 series? The normal thing to do is resend the entire series, not just one patch, as a new thread. Not using a new thread may make it harder to apply & will also bury the email in people's mailboxes that use things like mutt. A single patch as a reply is also confusing, as the rest of the v3 looks like it is missing!
Thanks,
Conor.
RE: [PATCH v3 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 37 lines
-----Original Message-----
From: Conor Dooley <[email protected]>
Sent: Wednesday, 2 August, 2023 3:02 PM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v3 3/5] dt-bindings: clock: add Intel Agilex5 clock managerQuoted 9 lines · On Wed, Aug 02, 2023 at 10:58:42AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Reviewed-by Conor Dooley <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>Damn, I was too late - you already sent a v3 :/
However, there only seems to be a v3 of this one patch and it was sent in reply to the v2 series? The normal thing to do is resend the entire series, not just one patch, as a new thread. Not using a new thread may make it harder to apply & will also bury the email in people's mailboxes that use things like mutt. A single patch as a reply is also confusing, as the rest of the v3 looks like it is missing!
Quoted 2 lines
Thanks,
Conor.
Sorry I made a mistake.
Should I send out entire series with PATCH v3 subject? Or should I wait for review comment on remaining patches and then send entire series with rework and subject prefix PATCH v3?
Thanks,
Nirav
Dinh Nguyen
maintainerRe: [PATCH v3 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 37 lines · Niravkumar L
Quoted 37 lines
-----Original Message-----
From: Conor Dooley <[email protected]>
Sent: Wednesday, 2 August, 2023 3:02 PM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v3 3/5] dt-bindings: clock: add Intel Agilex5 clock managerQuoted 9 lines · On Wed, Aug 02, 2023 at 10:58:42AM +0800, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Reviewed-by Conor Dooley <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>Damn, I was too late - you already sent a v3 :/
However, there only seems to be a v3 of this one patch and it was sent in reply to the v2 series? The normal thing to do is resend the entire series, not just one patch, as a new thread. Not using a new thread may make it harder to apply & will also bury the email in people's mailboxes that use things like mutt. A single patch as a reply is also confusing, as the rest of the v3 looks like it is missing!
Quoted 6 lines
Quoted 2 lines
Thanks,
Conor.Sorry I made a mistake.
Should I send out entire series with PATCH v3 subject? Or should I wait for review comment on remaining patches and then send entire series with rework and subject prefix PATCH v3?
No need to send out a V3. I've applied patches 1-3 and 5. Will give a little more time for the clk patch.
Dinh
Re: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
In reply to [email protected]Quoted 9 lines · On 01/08/2023 03:02, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>---
Do not attach (thread) your patchsets to some other threads (unrelated or older versions). This buries them deep in the mailbox and might interfere with applying entire sets.
Best regards,
Krzysztof
RE: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock manager
Quoted 27 lines
-----Original Message-----
From: Krzysztof Kozlowski <[email protected]>
Sent: Monday, 7 August, 2023 3:35 AM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2 3/5] dt-bindings: clock: add Intel Agilex5 clock managerQuoted 9 lines · On 01/08/2023 03:02, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add clock ID definitions for Intel Agilex5 SoCFPGA.
The registers in Agilex5 handling the clock is named as clock manager.Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>---
Quoted 6 lines
Do not attach (thread) your patchsets to some other threads (unrelated or older versions). This buries them deep in the mailbox and might interfere with applying entire sets.
Best regards,
Krzysztof
Sorry it was a mistake.
Will be careful now onwards.
Thanks,
Nirav
PATCH v2 4/5 clk: socfpga: agilex: add clock driver for the Agilex5
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add support for Intel's SoCFPGA Agilex5 platform. The clock manager driver for the Agilex5 is very similar to the Agilex platform,we can re-use most of the Agilex clock driver.
--- drivers/clk/socfpga/clk-agilex.c | 433 ++++++++++++++++++++++++++++++- 1 file changed, 431 insertions(+), 2 deletions(-)
drivers/clk/socfpga/clk-agilex.c +431 -2
@@ -1,6 +1,6 @@// SPDX-License-Identifier: GPL-2.0 /* -* Copyright (C) 2019, Intel Corporation +* Copyright (C) 2019-2023, Intel Corporation */ #include <linux/slab.h> #include <linux/clk-provider.h>
@@ -9,6 +9,7 @@#include <linux/platform_device.h> #include <dt-bindings/clock/agilex-clock.h> +#include <dt-bindings/clock/intel,agilex5-clkmgr.h> #include "stratix10-clk.h"
@@ -41,6 +42,67 @@ static const struct clk_parent_data mpu_free_mux[] = { .name = "f2s-free-clk", }, }; +static const struct clk_parent_data core0_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c0" , + .name ="peri_pll_c0" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data core1_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c0" , + .name ="peri_pll_c0" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data core2_free_mux[] = { + { .fw_name ="main_pll_c0" , + .name ="main_pll_c0" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data core3_free_mux[] = { + { .fw_name ="main_pll_c0" , + .name ="main_pll_c0" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data dsu_free_mux[] = { + { .fw_name ="main_pll_c2" , + .name ="main_pll_c2" , }, + { .fw_name ="peri_pll_c0" , + .name ="peri_pll_c0" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; +static const struct clk_parent_data noc_free_mux[] = { { .fw_name ="main_pll_c1" , .name = "main_pll_c1", },
@@ -53,7 +115,6 @@ static const struct clk_parent_data noc_free_mux[] = { { .fw_name = "f2s-free-clk", .name = "f2s-free-clk", }, }; -static const struct clk_parent_data emaca_free_mux[] = { { .fw_name ="main_pll_c2" , .name ="main_pll_c2" , },
@@ -158,6 +219,110 @@ static const struct clk_parent_data s2f_usr1_free_mux[] = { .name = "f2s-free-clk", }, }; +static const struct clk_parent_data agilex5_emaca_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_emacb_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_emac_ptp_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_gpio_db_free_mux[] = { + { .fw_name ="main_pll_c3" , + .name ="main_pll_c3" , }, + { .fw_name ="peri_pll_c1" , + .name ="peri_pll_c1" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_psi_ref_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_usb31_free_mux[] = { + { .fw_name ="main_pll_c3" , + .name ="main_pll_c3" , }, + { .fw_name ="peri_pll_c2" , + .name ="peri_pll_c2" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_s2f_usr0_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; + +static const struct clk_parent_data agilex5_s2f_usr1_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , }, + { .fw_name ="peri_pll_c3" , + .name ="peri_pll_c3" , }, + { .fw_name ="osc1" , + .name ="osc1" , }, + { .fw_name ="cb-intosc-hs-div2-clk" , + .name ="cb-intosc-hs-div2-clk" , }, + { .fw_name ="f2s-free-clk" , + .name ="f2s-free-clk" , }, +}; +static const struct clk_parent_data mpu_mux[] = { { .fw_name ="mpu_free_clk" , .name = "mpu_free_clk", },
@@ -165,6 +330,41 @@ static const struct clk_parent_data mpu_mux[] = { .name = "boot_clk", }, }; +static const struct clk_parent_data core0_mux[] = { + { .fw_name ="core0_free_clk" , + .name ="core0_free_clk" , }, + { .fw_name ="boot_clk" , + .name ="boot_clk" , }, +}; + +static const struct clk_parent_data core1_mux[] = { + { .fw_name ="core1_free_clk" , + .name ="core1_free_clk" , }, + { .fw_name ="boot_clk" , + .name ="boot_clk" , }, +}; + +static const struct clk_parent_data core2_mux[] = { + { .fw_name ="core2_free_clk" , + .name ="core2_free_clk" , }, + { .fw_name ="boot_clk" , + .name ="boot_clk" , }, +}; + +static const struct clk_parent_data core3_mux[] = { + { .fw_name ="core3_free_clk" , + .name ="core3_free_clk" , }, + { .fw_name ="boot_clk" , + .name ="boot_clk" , }, +}; + +static const struct clk_parent_data dsu_mux[] = { + { .fw_name ="dsu_free_clk" , + .name ="dsu_free_clk" , }, + { .fw_name ="boot_clk" , + .name ="boot_clk" , }, +}; +static const struct clk_parent_data emac_mux[] = { { .fw_name ="emaca_free_clk" , .name = "emaca_free_clk", },
@@ -223,6 +423,13 @@ static const struct clk_parent_data emac_ptp_mux[] = { .name = "boot_clk", }, }; +static const struct clk_parent_data usb31_mux[] = { + { .fw_name ="usb31_free_clk" , + .name ="usb31_free_clk" , }, + { .fw_name ="boot_clk" , + .name ="boot_clk" , }, +}; +/* clocks in AO (always on) controller */ static const struct stratix10_pll_clock agilex_pll_clks[] = { {AGILEX_BOOT_CLK ,"boot_clk" , boot_mux,ARRAY_SIZE (boot_mux),0 ,
@@ -255,6 +462,25 @@ static const struct stratix10_perip_c_clock agilex_main_perip_c_clks[] = { { AGILEX_PERIPH_PLL_C3_CLK, "peri_pll_c3", "periph_pll", NULL, 1, 0, 0xBC}, }; +static const struct stratix10_perip_c_clock agilex5_main_perip_c_clks[] = { + {AGILEX5_MAIN_PLL_C0_CLK ,"main_pll_c0" ,"main_pll" ,NULL ,1 ,0 , +0x5C }, + {AGILEX5_MAIN_PLL_C1_CLK ,"main_pll_c1" ,"main_pll" ,NULL ,1 ,0 , +0x60 }, + {AGILEX5_MAIN_PLL_C2_CLK ,"main_pll_c2" ,"main_pll" ,NULL ,1 ,0 , +0x64 }, + {AGILEX5_MAIN_PLL_C3_CLK ,"main_pll_c3" ,"main_pll" ,NULL ,1 ,0 , +0x68 }, + {AGILEX5_PERIPH_PLL_C0_CLK ,"peri_pll_c0" ,"periph_pll" ,NULL ,1 ,0 , +0xB0 }, + {AGILEX5_PERIPH_PLL_C1_CLK ,"peri_pll_c1" ,"periph_pll" ,NULL ,1 ,0 , +0xB4 }, + {AGILEX5_PERIPH_PLL_C2_CLK ,"peri_pll_c2" ,"periph_pll" ,NULL ,1 ,0 , +0xB8 }, + {AGILEX5_PERIPH_PLL_C3_CLK ,"peri_pll_c3" ,"periph_pll" ,NULL ,1 ,0 , +0xBC }, +}; +static const struct stratix10_perip_cnt_clock agilex_main_perip_cnt_clks[] = { {AGILEX_MPU_FREE_CLK ,"mpu_free_clk" ,NULL , mpu_free_mux,ARRAY_SIZE (mpu_free_mux), 0, 0x3C, 0, 0, 0},
@@ -280,6 +506,46 @@ static const struct stratix10_perip_cnt_clock agilex_main_perip_cnt_clks[] = { ARRAY_SIZE(psi_ref_free_mux), 0, 0xF0, 0, 0x88, 6}, }; +/* Non-SW clock-gated enabled clocks */ +static const struct stratix10_perip_cnt_clock agilex5_main_perip_cnt_clks[] = { + {AGILEX5_CORE0_FREE_CLK ,"core0_free_clk" ,NULL , core0_free_mux, +ARRAY_SIZE (core0_free_mux),0 ,0x0104 ,0 ,0 ,0 }, + {AGILEX5_CORE1_FREE_CLK ,"core1_free_clk" ,NULL , core1_free_mux, +ARRAY_SIZE (core1_free_mux),0 ,0x0104 ,0 ,0 ,0 }, + {AGILEX5_CORE2_FREE_CLK ,"core2_free_clk" ,NULL , core2_free_mux, +ARRAY_SIZE (core2_free_mux),0 ,0x010C ,0 ,0 ,0 }, + {AGILEX5_CORE3_FREE_CLK ,"core3_free_clk" ,NULL , core3_free_mux, +ARRAY_SIZE (core3_free_mux),0 ,0x0110 ,0 ,0 ,0 }, + {AGILEX5_DSU_FREE_CLK ,"dsu_free_clk" ,NULL , dsu_free_mux, +ARRAY_SIZE (dsu_free_mux),0 ,0x0100 ,0 ,0 ,0 }, + {AGILEX5_NOC_FREE_CLK ,"noc_free_clk" ,NULL , noc_free_mux, +ARRAY_SIZE (noc_free_mux),0 ,0x40 ,0 ,0 ,0 }, + {AGILEX5_EMAC_A_FREE_CLK ,"emaca_free_clk" ,NULL , + agilex5_emaca_free_mux,ARRAY_SIZE (agilex5_emaca_free_mux),0 ,0xD4 , +0 ,0x88 ,0 }, + {AGILEX5_EMAC_B_FREE_CLK ,"emacb_free_clk" ,NULL , + agilex5_emacb_free_mux,ARRAY_SIZE (agilex5_emacb_free_mux),0 ,0xD8 , +0 ,0x88 ,1 }, + {AGILEX5_EMAC_PTP_FREE_CLK ,"emac_ptp_free_clk" ,NULL , + agilex5_emac_ptp_free_mux,ARRAY_SIZE (agilex5_emac_ptp_free_mux),0 , +0xDC ,0 ,0x88 ,2 }, + {AGILEX5_GPIO_DB_FREE_CLK ,"gpio_db_free_clk" ,NULL , + agilex5_gpio_db_free_mux,ARRAY_SIZE (agilex5_gpio_db_free_mux),0 , +0xE0 ,0 ,0x88 ,3 }, + {AGILEX5_S2F_USER0_FREE_CLK ,"s2f_user0_free_clk" ,NULL , + agilex5_s2f_usr0_free_mux,ARRAY_SIZE (agilex5_s2f_usr0_free_mux),0 , +0xE8 ,0 ,0x30 ,2 }, + {AGILEX5_S2F_USER1_FREE_CLK ,"s2f_user1_free_clk" ,NULL , + agilex5_s2f_usr1_free_mux,ARRAY_SIZE (agilex5_s2f_usr1_free_mux),0 , +0xEC ,0 ,0x88 ,5 }, + {AGILEX5_PSI_REF_FREE_CLK ,"psi_ref_free_clk" ,NULL , + agilex5_psi_ref_free_mux,ARRAY_SIZE (agilex5_psi_ref_free_mux),0 , +0xF0 ,0 ,0x88 ,6 }, + {AGILEX5_USB31_FREE_CLK ,"usb31_free_clk" ,NULL , + agilex5_usb31_free_mux,ARRAY_SIZE (agilex5_usb31_free_mux),0 ,0xF8 , +0 ,0x88 ,7 }, +}; +static const struct stratix10_gate_clock agilex_gate_clks[] = { {AGILEX_MPU_CLK ,"mpu_clk" ,NULL , mpu_mux,ARRAY_SIZE (mpu_mux),0 ,0x24 ,0 ,0 ,0 ,0 ,0x30 ,0 ,0 },
@@ -335,6 +601,122 @@ static const struct stratix10_gate_clock agilex_gate_clks[] = { 10, 0, 0, 0, 0, 0, 4}, }; +/* SW Clock gate enabled clocks */ +static const struct stratix10_gate_clock agilex5_gate_clks[] = { +/* Main PLL0 Begin */ +/* MPU clocks */ + {AGILEX5_CORE0_CLK ,"core0_clk" ,NULL , core0_mux, +ARRAY_SIZE (core0_mux),0 ,0x24 ,8 ,0 ,0 ,0 ,0x30 ,5 ,0 }, + {AGILEX5_CORE1_CLK ,"core1_clk" ,NULL , core1_mux, +ARRAY_SIZE (core1_mux),0 ,0x24 ,9 ,0 ,0 ,0 ,0x30 ,5 ,0 }, + {AGILEX5_CORE2_CLK ,"core2_clk" ,NULL , core2_mux, +ARRAY_SIZE (core2_mux),0 ,0x24 ,10 ,0 ,0 ,0 ,0x30 ,6 ,0 }, + {AGILEX5_CORE3_CLK ,"core3_clk" ,NULL , core3_mux, +ARRAY_SIZE (core3_mux),0 ,0x24 ,11 ,0 ,0 ,0 ,0x30 ,7 ,0 }, + {AGILEX5_MPU_CLK ,"dsu_clk" ,NULL , dsu_mux,ARRAY_SIZE (dsu_mux),0 ,0 , +0 ,0 ,0 ,0 ,0x34 ,4 ,0 }, + {AGILEX5_MPU_PERIPH_CLK ,"mpu_periph_clk" ,NULL , dsu_mux, +ARRAY_SIZE (dsu_mux),0 ,0 ,0 ,0x44 ,20 ,2 ,0x34 ,4 ,0 }, + {AGILEX5_MPU_CCU_CLK ,"mpu_ccu_clk" ,NULL , dsu_mux, +ARRAY_SIZE (dsu_mux),0 ,0 ,0 ,0x44 ,18 ,2 ,0x34 ,4 ,0 }, + {AGILEX5_L4_MAIN_CLK ,"l4_main_clk" ,NULL , noc_mux, +ARRAY_SIZE (noc_mux),0 ,0x24 ,1 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_L4_MP_CLK ,"l4_mp_clk" ,NULL , noc_mux,ARRAY_SIZE (noc_mux),0 , +0x24 ,2 ,0x44 ,4 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_L4_SYS_FREE_CLK ,"l4_sys_free_clk" ,NULL , noc_mux, +ARRAY_SIZE (noc_mux),0 ,0 ,0 ,0x44 ,2 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_L4_SP_CLK ,"l4_sp_clk" ,NULL , noc_mux,ARRAY_SIZE (noc_mux), +CLK_IS_CRITICAL ,0x24 ,3 ,0x44 ,6 ,2 ,0x30 ,1 ,0 }, + +/* Core sight clocks*/ + {AGILEX5_CS_AT_CLK ,"cs_at_clk" ,NULL , noc_mux,ARRAY_SIZE (noc_mux),0 , +0x24 ,4 ,0x44 ,24 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_CS_TRACE_CLK ,"cs_trace_clk" ,NULL , noc_mux, +ARRAY_SIZE (noc_mux),0 ,0x24 ,4 ,0x44 ,26 ,2 ,0x30 ,1 ,0 }, + {AGILEX5_CS_PDBG_CLK ,"cs_pdbg_clk" ,"cs_at_clk" ,NULL ,1 ,0 ,0x24 ,4 , +0x44 ,28 ,1 ,0 ,0 ,0 }, +/* Main PLL0 End */ + +/* Main Peripheral PLL1 Begin */ + {AGILEX5_EMAC0_CLK ,"emac0_clk" ,NULL , emac_mux,ARRAY_SIZE (emac_mux), +0 ,0x7C ,0 ,0 ,0 ,0 ,0x94 ,26 ,0 }, + {AGILEX5_EMAC1_CLK ,"emac1_clk" ,NULL , emac_mux,ARRAY_SIZE (emac_mux), +0 ,0x7C ,1 ,0 ,0 ,0 ,0x94 ,27 ,0 }, + {AGILEX5_EMAC2_CLK ,"emac2_clk" ,NULL , emac_mux,ARRAY_SIZE (emac_mux), +0 ,0x7C ,2 ,0 ,0 ,0 ,0x94 ,28 ,0 }, + {AGILEX5_EMAC_PTP_CLK ,"emac_ptp_clk" ,NULL , emac_ptp_mux, +ARRAY_SIZE (emac_ptp_mux),0 ,0x7C ,3 ,0 ,0 ,0 ,0x88 ,2 ,0 }, + {AGILEX5_GPIO_DB_CLK ,"gpio_db_clk" ,NULL , gpio_db_mux, +ARRAY_SIZE (gpio_db_mux),0 ,0x7C ,4 ,0x98 ,0 ,16 ,0x88 ,3 ,1 }, +/* Main Peripheral PLL1 End */ + +/* Peripheral clocks */ + {AGILEX5_S2F_USER0_CLK ,"s2f_user0_clk" ,NULL , s2f_user0_mux, +ARRAY_SIZE (s2f_user0_mux),0 ,0x24 ,6 ,0 ,0 ,0 ,0x30 ,2 ,0 }, + {AGILEX5_S2F_USER1_CLK ,"s2f_user1_clk" ,NULL , s2f_user1_mux, +ARRAY_SIZE (s2f_user1_mux),0 ,0x7C ,6 ,0 ,0 ,0 ,0x88 ,5 ,0 }, + {AGILEX5_PSI_REF_CLK ,"psi_ref_clk" ,NULL , psi_mux, +ARRAY_SIZE (psi_mux),0 ,0x7C ,7 ,0 ,0 ,0 ,0x88 ,6 ,0 }, + {AGILEX5_USB31_SUSPEND_CLK ,"usb31_suspend_clk" ,NULL , usb31_mux, +ARRAY_SIZE (usb31_mux),0 ,0x7C ,25 ,0 ,0 ,0 ,0x88 ,7 ,0 }, + {AGILEX5_USB31_BUS_CLK_EARLY ,"usb31_bus_clk_early" ,"l4_main_clk" , +NULL ,1 ,0 ,0x7C ,25 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_USB2OTG_HCLK ,"usb2otg_hclk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C , +8 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIM_0_CLK ,"spim_0_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C ,9 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIM_1_CLK ,"spim_1_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C ,11 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIS_0_CLK ,"spis_0_clk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,12 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPIS_1_CLK ,"spis_1_clk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,13 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_DMA_CORE_CLK ,"dma_core_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C , +14 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_DMA_HS_CLK ,"dma_hs_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C ,14 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I3C_0_CORE_CLK ,"i3c_0_core_clk" ,"l4_mp_clk" ,NULL ,1 ,0 , +0x7C ,18 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I3C_1_CORE_CLK ,"i3c_1_core_clk" ,"l4_mp_clk" ,NULL ,1 ,0 , +0x7C ,19 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_0_PCLK ,"i2c_0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,15 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_1_PCLK ,"i2c_1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,16 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_EMAC0_PCLK ,"i2c_emac0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,17 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_EMAC1_PCLK ,"i2c_emac1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,22 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_I2C_EMAC2_PCLK ,"i2c_emac2_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,27 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_UART_0_PCLK ,"uart_0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,20 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_UART_1_PCLK ,"uart_1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 ,0x7C ,21 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPTIMER_0_PCLK ,"sptimer_0_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,23 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SPTIMER_1_PCLK ,"sptimer_1_pclk" ,"l4_sp_clk" ,NULL ,1 ,0 , +0x7C ,24 ,0 ,0 ,0 ,0 ,0 ,0 }, + +/* NAND, SD/MMC and SoftPHY overall clocking */ + {AGILEX5_DFI_CLK ,"dfi_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0 ,0 ,0x44 ,16 , +2 ,0 ,0 ,0 }, + {AGILEX5_NAND_NF_CLK ,"nand_nf_clk" ,"dfi_clk" ,NULL ,1 ,0 ,0x7C ,10 , +0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_NAND_BCH_CLK ,"nand_bch_clk" ,"l4_mp_clk" ,NULL ,1 ,0 ,0x7C , +10 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SDMMC_SDPHY_REG_CLK ,"sdmmc_sdphy_reg_clk" ,"l4_mp_clk" ,NULL , +1 ,0 ,0x7C ,5 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SDMCLK ,"sdmclk" ,"dfi_clk" ,NULL ,1 ,0 ,0x7C ,5 ,0 ,0 ,0 ,0 , +0 ,0 }, + {AGILEX5_SOFTPHY_REG_PCLK ,"softphy_reg_pclk" ,"l4_mp_clk" ,NULL ,1 ,0 , +0x7C ,26 ,0 ,0 ,0 ,0 ,0 ,0 }, + {AGILEX5_SOFTPHY_PHY_CLK ,"softphy_phy_clk" ,"l4_mp_clk" ,NULL ,1 ,0 , +0x7C ,26 ,0x44 ,16 ,2 ,0 ,0 ,0 }, + {AGILEX5_SOFTPHY_CTRL_CLK ,"softphy_ctrl_clk" ,"dfi_clk" ,NULL ,1 ,0 , +0x7C ,26 ,0 ,0 ,0 ,0 ,0 ,0 }, +}; +static int n5x_clk_register_c_perip (const struct n5x_perip_c_clock *clks,int nums,struct stratix10_clock_data *data) {
@@ -535,6 +917,51 @@ static int n5x_clkmgr_init(struct platform_device *pdev)return 0 ; } +static int agilex5_clkmgr_init (struct platform_device *pdev) +{ +struct device_node *np = pdev->dev.of_node; +struct device *dev = &pdev->dev; +struct stratix10_clock_data *clk_data; +struct resource *res; +void __iomem *base; +int i, num_clks; + + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); + base =devm_ioremap_resource (dev, res); +if (IS_ERR (base)) +return PTR_ERR (base); + + num_clks =AGILEX5_NUM_CLKS ; + + clk_data =devm_kzalloc (dev,struct_size (clk_data, clk_data.hws, + num_clks),GFP_KERNEL ); +if (!clk_data) +return -ENOMEM ; + +for (i =0 ; i < num_clks; i++) + clk_data->clk_data.hws[i] =ERR_PTR (-ENOENT ); + + clk_data->base = base; + clk_data->clk_data.num = num_clks; + +agilex_clk_register_pll (agilex_pll_clks,ARRAY_SIZE (agilex_pll_clks), + clk_data); + +agilex_clk_register_c_perip (agilex5_main_perip_c_clks, +ARRAY_SIZE (agilex5_main_perip_c_clks), + clk_data); + +agilex_clk_register_cnt_perip (agilex5_main_perip_cnt_clks, +ARRAY_SIZE (agilex5_main_perip_cnt_clks), + clk_data); + +agilex_clk_register_gate (agilex5_gate_clks, +ARRAY_SIZE (agilex5_gate_clks), clk_data); + +of_clk_add_hw_provider (np, of_clk_hw_onecell_get, &clk_data->clk_data); +return 0 ; +} +static int agilex_clkmgr_probe (struct platform_device *pdev) {int (*probe_func)(struct platform_device *init_func);
@@ -550,6 +977,8 @@ static const struct of_device_id agilex_clkmgr_match_table[] = { .data = agilex_clkmgr_init }, { .compatible = "intel,easic-n5x-clkmgr", .data = n5x_clkmgr_init }, + { .compatible = "intel,agilex5-clkmgr", + .data = agilex5_clkmgr_init }, { } };
Signature · 1 line
2.25.1
Dinh Nguyen
maintainerRe: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Hi Stephen/Mike,
Quoted 6 lines · On 7/31/23 20:02, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add support for Intel's SoCFPGA Agilex5 platform. The clock manager driver for the Agilex5 is very similar to the Agilex platform,we can re-use most of the Agilex clock driver.
Quoted 7 lines
Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- drivers/clk/socfpga/clk-agilex.c | 433 ++++++++++++++++++++++++++++++- 1 file changed, 431 insertions(+), 2 deletions(-)
If you're ok with this patch, can I take this through armsoc?
Dinh
Stephen Boyd
maintainerRe: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Quoting Dinh Nguyen (2023-08-08 04:03:47)
Quoted 17 lines
Hi Stephen/Mike,
Quoted 6 lines · On 7/31/23 20:02, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add support for Intel's SoCFPGA Agilex5 platform. The clock manager driver for the Agilex5 is very similar to the Agilex platform,we can re-use most of the Agilex clock driver.
Quoted 7 lines
Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- drivers/clk/socfpga/clk-agilex.c | 433 ++++++++++++++++++++++++++++++- 1 file changed, 431 insertions(+), 2 deletions(-)
Quoted 2 lines
If you're ok with this patch, can I take this through armsoc?
Usually any binding files go through arm-soc and clk tree but the driver only goes through clk tree via a PR. Is that possible here?
Dinh Nguyen
maintainerRe: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Quoted 18 lines · Stephen Boyd
Quoting Dinh Nguyen (2023-08-08 04:03:47)
Quoted 17 lines
Hi Stephen/Mike,
Quoted 13 lines · On 7/31/23 20:02, [email protected] wrote:
From: Niravkumar L Rabara <[email protected]>
Add support for Intel's SoCFPGA Agilex5 platform. The clock manager driver for the Agilex5 is very similar to the Agilex platform,we can re-use most of the Agilex clock driver.
Signed-off-by Teh Wen Ping <[email protected]>Reviewed-by Dinh Nguyen <[email protected]>Signed-off-by Niravkumar L Rabara <[email protected]>--- drivers/clk/socfpga/clk-agilex.c | 433 ++++++++++++++++++++++++++++++- 1 file changed, 431 insertions(+), 2 deletions(-)
Quoted 5 lines
Quoted 2 lines
If you're ok with this patch, can I take this through armsoc?
Usually any binding files go through arm-soc and clk tree but the driver only goes through clk tree via a PR. Is that possible here?
Ok. Should be fine in this case.
Thanks,
Dinh
Stephen Boyd
maintainerRe: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
In reply to [email protected]Quoting [email protected] (2023-07-31 18:02:33)
Quoted 21 lines
drivers/clk/socfpga/clk-agilex.c +2
-1@@ -1,6 +1,6 @@// SPDX-License-Identifier: GPL-2.0 /* -* Copyright (C) 2019, Intel Corporation +* Copyright (C) 2019-2023, Intel Corporation */ #include <linux/slab.h> #include <linux/clk-provider.h> @@ -9,6 +9,7 @@#include <linux/platform_device.h> #include <dt-bindings/clock/agilex-clock.h> +#include <dt-bindings/clock/intel,agilex5-clkmgr.h> #include "stratix10-clk.h"
drivers/clk/socfpga/clk-agilex.c static const struct clk_parent_data mpu_free_mux[] = { · line 47
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +3
-0@@ -41,6 +42,67 @@ static const struct clk_parent_data mpu_free_mux[] = { .name = "f2s-free-clk", }, }; +static const struct clk_parent_data core0_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name = "main_pll_c1", },
We're adding support for something new? Only set .fw_name in that case, as .name will never be used. To do even better, set only .index so that we don't do any string comparisons.
Quoted 10 lines
drivers/clk/socfpga/clk-agilex.c +10
-0+ { .fw_name = "peri_pll_c0", + .name = "peri_pll_c0", }, + { .fw_name = "osc1", + .name = "osc1", }, + { .fw_name = "cb-intosc-hs-div2-clk", + .name = "cb-intosc-hs-div2-clk", }, + { .fw_name = "f2s-free-clk", + .name = "f2s-free-clk", }, +}; +
[...]
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 924
Quoted 13 lines
drivers/clk/socfpga/clk-agilex.c +6
-0+static int n5x_clk_register_c_perip (const struct n5x_perip_c_clock *clks,int nums,struct stratix10_clock_data *data) {@@ -535,6 +917,51 @@ static int n5x_clkmgr_init(struct platform_device *pdev)return 0 ; } +static int agilex5_clkmgr_init (struct platform_device *pdev) +{ +struct device_node *np = pdev->dev.of_node; +struct device *dev = &pdev->dev; +struct stratix10_clock_data *clk_data;
Maybe call this stratix_data so that clk_data.clk_data is stratix_data.clk_data.
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 930
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +6
-0+struct resource *res; +void __iomem *base; +int i, num_clks; + + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); + base =devm_ioremap_resource (dev, res);
This is a single function call, devm_platform_ioremap_resource().
Quoted 21 lines
drivers/clk/socfpga/clk-agilex.c +21
-0+if (IS_ERR (base)) +return PTR_ERR (base); + + num_clks =AGILEX5_NUM_CLKS ; + + clk_data =devm_kzalloc (dev,struct_size (clk_data, clk_data.hws, + num_clks),GFP_KERNEL ); +if (!clk_data) +return -ENOMEM ; + +for (i =0 ; i < num_clks; i++) + clk_data->clk_data.hws[i] =ERR_PTR (-ENOENT ); + + clk_data->base = base; + clk_data->clk_data.num = num_clks; + +agilex_clk_register_pll (agilex_pll_clks,ARRAY_SIZE (agilex_pll_clks), + clk_data); + + agilex_clk_register_c_perip(agilex5_main_perip_c_clks, + ARRAY_SIZE(agilex5_main_perip_c_clks),
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 961
Quoted 10 lines
drivers/clk/socfpga/clk-agilex.c +10
-0+ clk_data); + +agilex_clk_register_cnt_perip (agilex5_main_perip_cnt_clks, +ARRAY_SIZE (agilex5_main_perip_cnt_clks), + clk_data); + +agilex_clk_register_gate (agilex5_gate_clks, +ARRAY_SIZE (agilex5_gate_clks), clk_data); + +of_clk_add_hw_provider (np, of_clk_hw_onecell_get, &clk_data->clk_data);
devm? Or when is this provider removed?
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +3
-0+return 0 ; +} +static int agilex_clkmgr_probe (struct platform_device *pdev) {int (*probe_func)(struct platform_device *init_func);
RE: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Quoted 40 lines
-----Original Message-----
From: Stephen Boyd <[email protected]>
Sent: Thursday, 10 August, 2023 5:27 AM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5Quoting [email protected] (2023-07-31 18:02:33)
Quoted 25 lines
drivers/clk/socfpga/clk-agilex.c +2
-1@@ -1,6 +1,6 @@// SPDX-License-Identifier: GPL-2.0 /* -* Copyright (C) 2019, Intel Corporation +* Copyright (C) 2019-2023, Intel Corporation */ #include <linux/slab.h> #include <linux/clk-provider.h> @@ -9,6 +9,7 @@#include <linux/platform_device.h> #include <dt-bindings/clock/agilex-clock.h> +#include <dt-bindings/clock/intel,agilex5-clkmgr.h> #include "stratix10-clk.h" @@ -41,6 +42,67 @@ static const struct clk_parent_data mpu_free_mux[] = { .name = "f2s-free-clk", }, };
Quoted 8 lines
drivers/clk/socfpga/clk-agilex.c static const struct clk_parent_data mpu_free_mux[] = { · line 47
Quoted 3 lines
drivers/clk/socfpga/clk-agilex.c +3
-0+static const struct clk_parent_data core0_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , },We're adding support for something new? Only set .fw_name in that case, as .name will never be used. To do even better, set only .index so that we don't do any string comparisons.
Yes we are adding Agilex5 SoCFPGA platform specific clocks.
I will remove .name and only keep .fw_name in next version of this patch.
Quoted 19 lines
Quoted 10 lines
drivers/clk/socfpga/clk-agilex.c +10
-0+ { .fw_name = "peri_pll_c0", + .name = "peri_pll_c0", }, + { .fw_name = "osc1", + .name = "osc1", }, + { .fw_name = "cb-intosc-hs-div2-clk", + .name = "cb-intosc-hs-div2-clk", }, + { .fw_name = "f2s-free-clk", + .name = "f2s-free-clk", }, +}; +[...]
Quoted 8 lines
+
static int n5x_clk_register_c_perip(const structn5x_perip_c_clock*clks,
int nums, structstratix10_clock_data*data) { @@ -535,6 +917,51 @@ static intn5x_clkmgr_init(struct platform_device *pdev)return 0;}
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 924
Quoted 4 lines
drivers/clk/socfpga/clk-agilex.c +4
-0+static int agilex5_clkmgr_init (struct platform_device *pdev) { +struct device_node *np = pdev->dev.of_node; +struct device *dev = &pdev->dev; +struct stratix10_clock_data *clk_data;Maybe call this
stratix_dataso that clk_data.clk_data is stratix_data.clk_data.
Will fix this in next version.
Quoted 9 lines
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 930
Quoted 7 lines
drivers/clk/socfpga/clk-agilex.c +6
-0+struct resource *res; +void __iomem *base; +int i, num_clks; + + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); + base =devm_ioremap_resource (dev, res);This is a single function call, devm_platform_ioremap_resource().i
Noted. Will fix in next version.
Quoted 28 lines
Quoted 28 lines
drivers/clk/socfpga/clk-agilex.c +27
-0+if (IS_ERR (base)) +return PTR_ERR (base); + + num_clks =AGILEX5_NUM_CLKS ; + + clk_data =devm_kzalloc (dev,struct_size (clk_data, clk_data.hws, + num_clks),GFP_KERNEL ); +if (!clk_data) +return -ENOMEM ; + +for (i =0 ; i < num_clks; i++) + clk_data->clk_data.hws[i] =ERR_PTR (-ENOENT ); + + clk_data->base = base; + clk_data->clk_data.num = num_clks; + +agilex_clk_register_pll (agilex_pll_clks,ARRAY_SIZE (agilex_pll_clks), + clk_data); + +agilex_clk_register_c_perip (agilex5_main_perip_c_clks, +ARRAY_SIZE (agilex5_main_perip_c_clks), + clk_data); + +agilex_clk_register_cnt_perip (agilex5_main_perip_cnt_clks, +ARRAY_SIZE (agilex5_main_perip_cnt_clks), + clk_data); +
Quoted 8 lines
drivers/clk/socfpga/clk-agilex.c
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +6
-0+agilex_clk_register_gate (agilex5_gate_clks, +ARRAY_SIZE (agilex5_gate_clks), + clk_data); + +of_clk_add_hw_provider (np, of_clk_hw_onecell_get, + &clk_data->clk_data);devm? Or when is this provider removed?
Will fix in next version.
Quoted 6 lines
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +3
-0+return 0 ; +} +static int agilex_clkmgr_probe (struct platform_device *pdev) {int (*probe_func)(struct platform_device *init_func);
Dinh Nguyen
maintainerRe: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Quoted 87 lines · Niravkumar L
Quoted 40 lines
-----Original Message-----
From: Stephen Boyd <[email protected]>
Sent: Thursday, 10 August, 2023 5:27 AM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5Quoting [email protected] (2023-07-31 18:02:33)
Quoted 25 lines
drivers/clk/socfpga/clk-agilex.c +2
-1@@ -1,6 +1,6 @@// SPDX-License-Identifier: GPL-2.0 /* -* Copyright (C) 2019, Intel Corporation +* Copyright (C) 2019-2023, Intel Corporation */ #include <linux/slab.h> #include <linux/clk-provider.h> @@ -9,6 +9,7 @@#include <linux/platform_device.h> #include <dt-bindings/clock/agilex-clock.h> +#include <dt-bindings/clock/intel,agilex5-clkmgr.h> #include "stratix10-clk.h" @@ -41,6 +42,67 @@ static const struct clk_parent_data mpu_free_mux[] = { .name = "f2s-free-clk", }, };Quoted 8 lines
drivers/clk/socfpga/clk-agilex.c static const struct clk_parent_data mpu_free_mux[] = { · line 47
Quoted 3 lines
drivers/clk/socfpga/clk-agilex.c +3
-0+static const struct clk_parent_data core0_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name ="main_pll_c1" , },We're adding support for something new? Only set .fw_name in that case, as .name will never be used. To do even better, set only .index so that we don't do any string comparisons.
Yes we are adding Agilex5 SoCFPGA platform specific clocks.
I will remove .name and only keep .fw_name in next version of this patch.Quoted 19 lines
Quoted 10 lines
drivers/clk/socfpga/clk-agilex.c +10
-0+ { .fw_name = "peri_pll_c0", + .name = "peri_pll_c0", }, + { .fw_name = "osc1", + .name = "osc1", }, + { .fw_name = "cb-intosc-hs-div2-clk", + .name = "cb-intosc-hs-div2-clk", }, + { .fw_name = "f2s-free-clk", + .name = "f2s-free-clk", }, +}; +[...]
Quoted 8 lines
+
static int n5x_clk_register_c_perip(const structn5x_perip_c_clock*clks,
int nums, structstratix10_clock_data*data) { @@ -535,6 +917,51 @@ static intn5x_clkmgr_init(struct platform_device *pdev)return 0;}
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 924
Quoted 4 lines
drivers/clk/socfpga/clk-agilex.c +4
-0+static int agilex5_clkmgr_init (struct platform_device *pdev) { +struct device_node *np = pdev->dev.of_node; +struct device *dev = &pdev->dev; +struct stratix10_clock_data *clk_data;Maybe call this
stratix_dataso that clk_data.clk_data is stratix_data.clk_data.Will fix this in next version.
Quoted 8 lines
Quoted 7 lines
drivers/clk/socfpga/clk-agilex.c +6
-0+struct resource *res; +void __iomem *base; +int i, num_clks; + + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); + base =devm_ioremap_resource (dev, res);
Quoted 4 lines
Quoted 1 line
This is a single function call, devm_platform_ioremap_resource().i
Noted. Will fix in next version.
When you resend a V3, just send this patch. I've already applied the other 4 patches.
Dinh
RE: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5
Quoted 113 lines
-----Original Message-----
From: Dinh Nguyen <[email protected]>
Sent: Monday, 14 August, 2023 10:48 AM
To: Rabara, Niravkumar L <[email protected]>; Stephen Boyd <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5Quoted 8 lines · Niravkumar L
Quoted 8 lines
-----Original Message-----
From: Stephen Boyd <[email protected]>
Sent: Thursday, 10 August, 2023 5:27 AM
To: Rabara, Niravkumar L <[email protected]>
Cc: Ng, Adrian Ho Yin <[email protected]>; [email protected]; [email protected]; [email protected];Quoted 25 lines
Quoted 25 lines
[email protected]; [email protected]; linux- [email protected]; Turquette, Mike <[email protected]>; [email protected]; [email protected]; [email protected];
[email protected]; [email protected]
Subject: Re: [PATCH v2 4/5] clk: socfpga: agilex: add clock driver for the Agilex5Quoting [email protected] (2023-07-31 18:02:33)
Quoted 16 lines
drivers/clk/socfpga/clk-agilex.c +1
-1@@ -1,6 +1,6 @@// SPDX-License-Identifier: GPL-2.0 /* -* Copyright (C) 2019, Intel Corporation +* Copyright (C) 2019-2023, Intel Corporation */ #include <linux/slab.h> #include <linux/clk-provider.h> @@ -9,6 +9,7 @@#include <linux/platform_device.h> Quoted 6 lines
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c static const struct clk_parent_data
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +1
-0#include <dt-bindings/clock/agilex-clock.h> +#include <dt-bindings/clock/intel,agilex5-clkmgr.h> #include "stratix10-clk.h" @@ -41,6 +42,67 @@ static const struct clk_parent_datampu_free_mux[] = {
Quoted 7 lines
Quoted 7 lines
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c +3
-0.name = "f2s-free-clk", }, }; +static const struct clk_parent_data core0_free_mux[] = { + { .fw_name ="main_pll_c1" , + .name = "main_pll_c1", },Quoted 19 lines
Quoted 4 lines
We're adding support for something new? Only set .fw_name in that case, as .name will never be used. To do even better, set only .index so that we don't do any string comparisons.
Yes we are adding Agilex5 SoCFPGA platform specific clocks.
I will remove .name and only keep .fw_name in next version of this patch.Quoted 12 lines
Quoted 9 lines
drivers/clk/socfpga/clk-agilex.c +9
-0+ { .fw_name = "peri_pll_c0", + .name = "peri_pll_c0", }, + { .fw_name = "osc1", + .name = "osc1", }, + { .fw_name = "cb-intosc-hs-div2-clk", + .name = "cb-intosc-hs-div2-clk", }, + { .fw_name = "f2s-free-clk", + .name = "f2s-free-clk", }, }; +[...]
Quoted 2 lines
+
static int n5x_clk_register_c_perip(const structn5x_perip_c_clock*clks,
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
int nums, struct
stratix10_clock_data*data) { @@ -535,6 +917,51 @@ static intn5x_clkmgr_init(struct platform_device *pdev)return 0;}
Quoted 6 lines
Quoted 6 lines
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 924
Quoted 4 lines
drivers/clk/socfpga/clk-agilex.c +4
-0+static int agilex5_clkmgr_init (struct platform_device *pdev) { +struct device_node *np = pdev->dev.of_node; +struct device *dev = &pdev->dev; +struct stratix10_clock_data *clk_data;Maybe call this
stratix_dataso that clk_data.clk_data isstratix_data.clk_data.
Quoted 13 lines
Will fix this in next version.
Quoted 9 lines
drivers/clk/socfpga/clk-agilex.c static int n5x_clkmgr_init(struct platform_device *pdev) · line 930
Quoted 7 lines
drivers/clk/socfpga/clk-agilex.c +6
-0+struct resource *res; +void __iomem *base; +int i, num_clks; + + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); + base =devm_ioremap_resource (dev, res);This is a single function call, devm_platform_ioremap_resource().i
Quoted 7 lines
Quoted 2 lines
Noted. Will fix in next version.
When you resend a V3, just send this patch. I've already applied the other 4 patches.
Dinh
Noted Dinh.
Thanks,
Nirav
PATCH v2 5/5 arm64: dts: agilex5: add initial support for Intel Agilex5 SoCFPGA
In reply to [email protected]From: Niravkumar L Rabara <[email protected]>
Add the initial device tree files for Intel Agilex5 SoCFPGA platform.
--- arch/arm64/boot/dts/intel/Makefile | 1 + .../arm64/boot/dts/intel/socfpga_agilex5.dtsi | 468 ++++++++++++++++++ .../boot/dts/intel/socfpga_agilex5_socdk.dts | 39 ++ 3 files changed, 508 insertions(+) create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi create mode 100644 arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts
arch/arm64/boot/dts/intel/Makefile +1 -0
@@ -2,5 +2,6 @@ dtb-$(CONFIG_ARCH_INTEL_SOCFPGA) += socfpga_agilex_n6000.dtb \ socfpga_agilex_socdk.dtb \ socfpga_agilex_socdk_nand.dtb \ + socfpga_agilex5_socdk.dtb \ socfpga_n5x_socdk.dtb dtb-$(CONFIG_ARCH_KEEMBAY) += keembay-evm.dtb
arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi new file +468 -0
468 lines not on this page · raw message
arch/arm64/boot/dts/intel/socfpga_agilex5_socdk.dts new file +39 -0
@@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2023, Intel Corporation + */ +#include "socfpga_agilex5.dtsi" + +/ { + model = "SoCFPGA Agilex5 SoCDK"; + compatible = "intel,socfpga-agilex5-socdk", "intel,socfpga-agilex5"; + + aliases { + serial0 = &uart0; + }; + + chosen { + stdout-path = "serial0:115200n8"; + }; +}; + +&gpio1 { + status = "okay"; +}; + +&osc1 { + clock-frequency = <25000000>; +}; + +&uart0 { + status = "okay"; +}; + +&usb0 { + status = "okay"; + disable-over-current; +}; + +&watchdog0 { + status = "okay"; +};
Signature · 1 line
2.25.1