* [PATCH v3 0/3] irqchip: Add Sophgo SG2042 MSI controller @ 2025-01-15 6:31 Chen Wang 2025-01-15 6:33 ` [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI Chen Wang ` (2 more replies) 0 siblings, 3 replies; 15+ messages in thread From: Chen Wang @ 2025-01-15 6:31 UTC (permalink / raw) To: u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet From: Chen Wang <unicorn_wang@outlook.com> This controller is on the Sophgo SG2042 SoC to transform interrupts from PCIe MSI to PLIC interrupts. Thanks, Chen --- Changes in v3: The patch series is based on v6.13-rc7. Fixed following issues as per comments from Krzysztof Kozlowski, Samuel Holland, Christophe JAILLET, Inochi Amaoto, thanks. - bindings: use reg for doorbell, fixed wrong usage of additionalProperties and misc. - Improve driver code: - Fixed potentional memory leak issues. - Fixed some build warnings reported by test robot. - Optimize and simplify the code when allocating hwirq. - Use DECLARE_BITMAP instead of kzalloc. - Some other coding style improvements. Changes in v2: The patch series is based on v6.13-rc2. You can simply review or test the patches at the link [2]. Fixed following issues as per comments from Rob Herring, Thomas Gleixner, thanks. - Improve driver binding description, use msi-ranges instread. - Improve driver code: - Improve coding style. - Fixed bug that possible memory leak of bitmap when sg2042_msi_init_domains returns error. - Use guard(mutex). - Use the MSI parent model. Changes in v1: The patch series is based on v6.12-rc7. You can simply review or test the patches at the link [1]. Link: https://lore.kernel.org/linux-riscv/cover.1731296803.git.unicorn_wang@outlook.com/ [1] Link: https://lore.kernel.org/linux-riscv/cover.1733726057.git.unicorn_wang@outlook.com/ [2] --- Chen Wang (3): dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI irqchip: Add the Sophgo SG2042 MSI interrupt controller riscv: sophgo: dts: add msi controller for SG2042 .../sophgo,sg2042-msi.yaml | 58 ++++ arch/riscv/boot/dts/sophgo/sg2042.dtsi | 10 + drivers/irqchip/Kconfig | 12 + drivers/irqchip/Makefile | 1 + drivers/irqchip/irq-sg2042-msi.c | 268 ++++++++++++++++++ 5 files changed, 349 insertions(+) create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml create mode 100644 drivers/irqchip/irq-sg2042-msi.c base-commit: 5bc55a333a2f7316b58edc7573e8e893f7acb532 -- 2.34.1 ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-15 6:31 [PATCH v3 0/3] irqchip: Add Sophgo SG2042 MSI controller Chen Wang @ 2025-01-15 6:33 ` Chen Wang 2025-01-20 2:42 ` Inochi Amaoto 2025-01-23 21:29 ` Rob Herring 2025-01-15 6:33 ` [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller Chen Wang 2025-01-15 6:34 ` [PATCH v3 3/3] riscv: sophgo: dts: add msi controller for SG2042 Chen Wang 2 siblings, 2 replies; 15+ messages in thread From: Chen Wang @ 2025-01-15 6:33 UTC (permalink / raw) To: u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet From: Chen Wang <unicorn_wang@outlook.com> Add binding for Sophgo SG2042 MSI controller. Signed-off-by: Chen Wang <unicorn_wang@outlook.com> --- .../sophgo,sg2042-msi.yaml | 58 +++++++++++++++++++ 1 file changed, 58 insertions(+) create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml diff --git a/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml new file mode 100644 index 000000000000..f641df191787 --- /dev/null +++ b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml @@ -0,0 +1,58 @@ +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/interrupt-controller/sophgo,sg2042-msi.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Sophgo SG2042 MSI Controller + +maintainers: + - Chen Wang <unicorn_wang@outlook.com> + +description: + This interrupt controller is in Sophgo SG2042 for transforming interrupts from + PCIe MSI to PLIC interrupts. + +allOf: + - $ref: /schemas/interrupts.yaml# + - $ref: /schemas/interrupt-controller/msi-controller.yaml# + +properties: + compatible: + const: sophgo,sg2042-msi + + reg: + items: + - description: msi doorbell address + - description: clear register + + reg-names: + items: + - const: doorbell + - const: clr + + msi-controller: true + + msi-ranges: + maxItems: 1 + +required: + - compatible + - reg + - reg-names + - msi-controller + - msi-ranges + +unevaluatedProperties: false + +examples: + - | + #include <dt-bindings/interrupt-controller/irq.h> + msi-controller@30000000 { + compatible = "sophgo,sg2042-msi"; + reg = <0x30000000 0x4>, <0x30000008 0x4>; + reg-names = "doorbell", "clr"; + msi-controller; + msi-ranges = <&plic 64 IRQ_TYPE_LEVEL_HIGH 32>; + interrupt-parent = <&plic>; + }; -- 2.34.1 ^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-15 6:33 ` [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI Chen Wang @ 2025-01-20 2:42 ` Inochi Amaoto 2025-01-22 13:51 ` Chen Wang 2025-01-23 21:29 ` Rob Herring 1 sibling, 1 reply; 15+ messages in thread From: Inochi Amaoto @ 2025-01-20 2:42 UTC (permalink / raw) To: Chen Wang, u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On Wed, Jan 15, 2025 at 02:33:23PM +0800, Chen Wang wrote: > From: Chen Wang <unicorn_wang@outlook.com> > > Add binding for Sophgo SG2042 MSI controller. > > Signed-off-by: Chen Wang <unicorn_wang@outlook.com> > --- > .../sophgo,sg2042-msi.yaml | 58 +++++++++++++++++++ > 1 file changed, 58 insertions(+) > create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml > > diff --git a/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml > new file mode 100644 > index 000000000000..f641df191787 > --- /dev/null > +++ b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml > @@ -0,0 +1,58 @@ > +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) > +%YAML 1.2 > +--- > +$id: http://devicetree.org/schemas/interrupt-controller/sophgo,sg2042-msi.yaml# > +$schema: http://devicetree.org/meta-schemas/core.yaml# > + > +title: Sophgo SG2042 MSI Controller > + > +maintainers: > + - Chen Wang <unicorn_wang@outlook.com> > + > +description: > + This interrupt controller is in Sophgo SG2042 for transforming interrupts from > + PCIe MSI to PLIC interrupts. > + > +allOf: > + - $ref: /schemas/interrupts.yaml# > + - $ref: /schemas/interrupt-controller/msi-controller.yaml# > + > +properties: > + compatible: > + const: sophgo,sg2042-msi > + > + reg: > + items: > + - description: msi doorbell address > + - description: clear register > + > + reg-names: > + items: > + - const: doorbell > + - const: clr please reverse the items order, the clr addr is more suitable as the MMIO device address when writing device node. doorbeel address is just a IO address and can not be seen from CPU. > + > + msi-controller: true > + > + msi-ranges: > + maxItems: 1 > + > +required: > + - compatible > + - reg > + - reg-names > + - msi-controller > + - msi-ranges > + > +unevaluatedProperties: false > + > +examples: > + - | > + #include <dt-bindings/interrupt-controller/irq.h> > + msi-controller@30000000 { > + compatible = "sophgo,sg2042-msi"; > + reg = <0x30000000 0x4>, <0x30000008 0x4>; > + reg-names = "doorbell", "clr"; > + msi-controller; > + msi-ranges = <&plic 64 IRQ_TYPE_LEVEL_HIGH 32>; > + interrupt-parent = <&plic>; > + }; > -- > 2.34.1 > > > _______________________________________________ > linux-riscv mailing list > linux-riscv@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-20 2:42 ` Inochi Amaoto @ 2025-01-22 13:51 ` Chen Wang 2025-01-22 22:42 ` Inochi Amaoto 0 siblings, 1 reply; 15+ messages in thread From: Chen Wang @ 2025-01-22 13:51 UTC (permalink / raw) To: Inochi Amaoto, Chen Wang, u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On 2025/1/20 10:42, Inochi Amaoto wrote: > On Wed, Jan 15, 2025 at 02:33:23PM +0800, Chen Wang wrote: [......] >> + reg: >> + items: >> + - description: msi doorbell address >> + - description: clear register >> + >> + reg-names: >> + items: >> + - const: doorbell >> + - const: clr > please reverse the items order, the clr addr is more suitable > as the MMIO device address when writing device node. doorbeel > address is just a IO address and can not be seen from CPU. I find dtbcheck will report error if order is switched. On SG2042, address of doorbell is ahead of clr. >> + >> + msi-controller: true >> + >> + msi-ranges: >> + maxItems: 1 >> + >> +required: >> + - compatible >> + - reg >> + - reg-names >> + - msi-controller >> + - msi-ranges >> + >> +unevaluatedProperties: false >> + >> +examples: >> + - | >> + #include <dt-bindings/interrupt-controller/irq.h> >> + msi-controller@30000000 { >> + compatible = "sophgo,sg2042-msi"; >> + reg = <0x30000000 0x4>, <0x30000008 0x4>; >> + reg-names = "doorbell", "clr"; >> + msi-controller; >> + msi-ranges = <&plic 64 IRQ_TYPE_LEVEL_HIGH 32>; >> + interrupt-parent = <&plic>; >> + }; >> -- >> 2.34.1 >> >> >> _______________________________________________ >> linux-riscv mailing list >> linux-riscv@lists.infradead.org >> http://lists.infradead.org/mailman/listinfo/linux-riscv ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-22 13:51 ` Chen Wang @ 2025-01-22 22:42 ` Inochi Amaoto 2025-01-23 0:47 ` Chen Wang 0 siblings, 1 reply; 15+ messages in thread From: Inochi Amaoto @ 2025-01-22 22:42 UTC (permalink / raw) To: Chen Wang, Inochi Amaoto, Chen Wang, u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On Wed, Jan 22, 2025 at 09:51:05PM +0800, Chen Wang wrote: > > On 2025/1/20 10:42, Inochi Amaoto wrote: > > On Wed, Jan 15, 2025 at 02:33:23PM +0800, Chen Wang wrote: > [......] > > > + reg: > > > + items: > > > + - description: msi doorbell address > > > + - description: clear register > > > + > > > + reg-names: > > > + items: > > > + - const: doorbell > > > + - const: clr > > please reverse the items order, the clr addr is more suitable > > as the MMIO device address when writing device node. doorbeel > > address is just a IO address and can not be seen from CPU. > > I find dtbcheck will report error if order is switched. > You should also change the unit address to avoid error. I think you forgot it. > On SG2042, address of doorbell is ahead of clr. > It is the same on SG2044, but there is a problem that the doorbell addr is a IO address and it is not suitable to represent the device addr in the dtb. It also lead to a weird unit address on SG2044 which is hard to understand. Regards, Inochi ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-22 22:42 ` Inochi Amaoto @ 2025-01-23 0:47 ` Chen Wang 0 siblings, 0 replies; 15+ messages in thread From: Chen Wang @ 2025-01-23 0:47 UTC (permalink / raw) To: Inochi Amaoto, Chen Wang, u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On 2025/1/23 6:42, Inochi Amaoto wrote: > On Wed, Jan 22, 2025 at 09:51:05PM +0800, Chen Wang wrote: >> On 2025/1/20 10:42, Inochi Amaoto wrote: >>> On Wed, Jan 15, 2025 at 02:33:23PM +0800, Chen Wang wrote: >> [......] >>>> + reg: >>>> + items: >>>> + - description: msi doorbell address >>>> + - description: clear register >>>> + >>>> + reg-names: >>>> + items: >>>> + - const: doorbell >>>> + - const: clr >>> please reverse the items order, the clr addr is more suitable >>> as the MMIO device address when writing device node. doorbeel >>> address is just a IO address and can not be seen from CPU. >> I find dtbcheck will report error if order is switched. >> > You should also change the unit address to avoid error. > I think you forgot it. I forgot it, thanks for your suggestion. >> On SG2042, address of doorbell is ahead of clr. >> > It is the same on SG2044, but there is a problem that the > doorbell addr is a IO address and it is not suitable to > represent the device addr in the dtb. It also lead to a > weird unit address on SG2044 which is hard to understand. > > Regards, > Inochi ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-15 6:33 ` [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI Chen Wang 2025-01-20 2:42 ` Inochi Amaoto @ 2025-01-23 21:29 ` Rob Herring 2025-02-18 1:26 ` Chen Wang 1 sibling, 1 reply; 15+ messages in thread From: Rob Herring @ 2025-01-23 21:29 UTC (permalink / raw) To: Chen Wang Cc: u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On Wed, Jan 15, 2025 at 02:33:23PM +0800, Chen Wang wrote: > From: Chen Wang <unicorn_wang@outlook.com> > > Add binding for Sophgo SG2042 MSI controller. > > Signed-off-by: Chen Wang <unicorn_wang@outlook.com> > --- > .../sophgo,sg2042-msi.yaml | 58 +++++++++++++++++++ > 1 file changed, 58 insertions(+) > create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml > > diff --git a/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml > new file mode 100644 > index 000000000000..f641df191787 > --- /dev/null > +++ b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml > @@ -0,0 +1,58 @@ > +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) > +%YAML 1.2 > +--- > +$id: http://devicetree.org/schemas/interrupt-controller/sophgo,sg2042-msi.yaml# > +$schema: http://devicetree.org/meta-schemas/core.yaml# > + > +title: Sophgo SG2042 MSI Controller > + > +maintainers: > + - Chen Wang <unicorn_wang@outlook.com> > + > +description: > + This interrupt controller is in Sophgo SG2042 for transforming interrupts from > + PCIe MSI to PLIC interrupts. > + > +allOf: > + - $ref: /schemas/interrupts.yaml# Drop this. > + - $ref: /schemas/interrupt-controller/msi-controller.yaml# > + > +properties: > + compatible: > + const: sophgo,sg2042-msi > + > + reg: > + items: > + - description: msi doorbell address > + - description: clear register > + > + reg-names: > + items: > + - const: doorbell > + - const: clr > + > + msi-controller: true > + > + msi-ranges: > + maxItems: 1 You need #msi-cells. > + > +required: > + - compatible > + - reg > + - reg-names > + - msi-controller > + - msi-ranges > + > +unevaluatedProperties: false > + > +examples: > + - | > + #include <dt-bindings/interrupt-controller/irq.h> > + msi-controller@30000000 { > + compatible = "sophgo,sg2042-msi"; > + reg = <0x30000000 0x4>, <0x30000008 0x4>; > + reg-names = "doorbell", "clr"; > + msi-controller; > + msi-ranges = <&plic 64 IRQ_TYPE_LEVEL_HIGH 32>; > + interrupt-parent = <&plic>; > + }; > -- > 2.34.1 > ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI 2025-01-23 21:29 ` Rob Herring @ 2025-02-18 1:26 ` Chen Wang 0 siblings, 0 replies; 15+ messages in thread From: Chen Wang @ 2025-02-18 1:26 UTC (permalink / raw) To: Rob Herring, Chen Wang Cc: u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet hello, Rob, On 2025/1/24 5:29, Rob Herring wrote: > On Wed, Jan 15, 2025 at 02:33:23PM +0800, Chen Wang wrote: >> From: Chen Wang <unicorn_wang@outlook.com> >> >> Add binding for Sophgo SG2042 MSI controller. >> >> Signed-off-by: Chen Wang <unicorn_wang@outlook.com> >> --- >> .../sophgo,sg2042-msi.yaml | 58 +++++++++++++++++++ >> 1 file changed, 58 insertions(+) >> create mode 100644 Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml >> >> diff --git a/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml >> new file mode 100644 >> index 000000000000..f641df191787 >> --- /dev/null >> +++ b/Documentation/devicetree/bindings/interrupt-controller/sophgo,sg2042-msi.yaml >> @@ -0,0 +1,58 @@ >> +# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause) >> +%YAML 1.2 >> +--- >> +$id: http://devicetree.org/schemas/interrupt-controller/sophgo,sg2042-msi.yaml# >> +$schema: http://devicetree.org/meta-schemas/core.yaml# >> + >> +title: Sophgo SG2042 MSI Controller >> + >> +maintainers: >> + - Chen Wang <unicorn_wang@outlook.com> >> + >> +description: >> + This interrupt controller is in Sophgo SG2042 for transforming interrupts from >> + PCIe MSI to PLIC interrupts. >> + >> +allOf: >> + - $ref: /schemas/interrupts.yaml# > Drop this. I find if we drop this line, dtb check will report "Unevaluated properties are not allowed ('interrupt-parent' was unexpected)" Do we still need to keep this? Thanks, Chen >> + - $ref: /schemas/interrupt-controller/msi-controller.yaml# >> + >> +properties: >> + compatible: >> + const: sophgo,sg2042-msi >> + >> + reg: >> + items: >> + - description: msi doorbell address >> + - description: clear register >> + >> + reg-names: >> + items: >> + - const: doorbell >> + - const: clr >> + >> + msi-controller: true >> + >> + msi-ranges: >> + maxItems: 1 > You need #msi-cells. OK, will add this. >> + >> +required: >> + - compatible >> + - reg >> + - reg-names >> + - msi-controller >> + - msi-ranges >> + >> +unevaluatedProperties: false >> + >> +examples: >> + - | >> + #include <dt-bindings/interrupt-controller/irq.h> >> + msi-controller@30000000 { >> + compatible = "sophgo,sg2042-msi"; >> + reg = <0x30000000 0x4>, <0x30000008 0x4>; >> + reg-names = "doorbell", "clr"; >> + msi-controller; >> + msi-ranges = <&plic 64 IRQ_TYPE_LEVEL_HIGH 32>; >> + interrupt-parent = <&plic>; >> + }; >> -- >> 2.34.1 >> ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller 2025-01-15 6:31 [PATCH v3 0/3] irqchip: Add Sophgo SG2042 MSI controller Chen Wang 2025-01-15 6:33 ` [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI Chen Wang @ 2025-01-15 6:33 ` Chen Wang 2025-01-20 2:43 ` Inochi Amaoto 2025-02-15 1:09 ` Inochi Amaoto 2025-01-15 6:34 ` [PATCH v3 3/3] riscv: sophgo: dts: add msi controller for SG2042 Chen Wang 2 siblings, 2 replies; 15+ messages in thread From: Chen Wang @ 2025-01-15 6:33 UTC (permalink / raw) To: u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet From: Chen Wang <unicorn_wang@outlook.com> Add driver for Sophgo SG2042 MSI interrupt controller. Signed-off-by: Chen Wang <unicorn_wang@outlook.com> --- drivers/irqchip/Kconfig | 12 ++ drivers/irqchip/Makefile | 1 + drivers/irqchip/irq-sg2042-msi.c | 268 +++++++++++++++++++++++++++++++ 3 files changed, 281 insertions(+) create mode 100644 drivers/irqchip/irq-sg2042-msi.c diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig index 9bee02db1643..161fb5df857f 100644 --- a/drivers/irqchip/Kconfig +++ b/drivers/irqchip/Kconfig @@ -749,6 +749,18 @@ config MCHP_EIC help Support for Microchip External Interrupt Controller. +config SOPHGO_SG2042_MSI + bool "Sophgo SG2042 MSI Controller" + depends on ARCH_SOPHGO || COMPILE_TEST + depends on PCI + select IRQ_DOMAIN_HIERARCHY + select IRQ_MSI_LIB + select PCI_MSI + help + Support for the Sophgo SG2042 MSI Controller. + This on-chip interrupt controller enables MSI sources to be + routed to the primary PLIC controller on SoC. + config SUNPLUS_SP7021_INTC bool "Sunplus SP7021 interrupt controller" if COMPILE_TEST default SOC_SP7021 diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile index 25e9ad29b8c4..dd60e597491d 100644 --- a/drivers/irqchip/Makefile +++ b/drivers/irqchip/Makefile @@ -128,4 +128,5 @@ obj-$(CONFIG_WPCM450_AIC) += irq-wpcm450-aic.o obj-$(CONFIG_IRQ_IDT3243X) += irq-idt3243x.o obj-$(CONFIG_APPLE_AIC) += irq-apple-aic.o obj-$(CONFIG_MCHP_EIC) += irq-mchp-eic.o +obj-$(CONFIG_SOPHGO_SG2042_MSI) += irq-sg2042-msi.o obj-$(CONFIG_SUNPLUS_SP7021_INTC) += irq-sp7021-intc.o diff --git a/drivers/irqchip/irq-sg2042-msi.c b/drivers/irqchip/irq-sg2042-msi.c new file mode 100644 index 000000000000..1f88e1671e5c --- /dev/null +++ b/drivers/irqchip/irq-sg2042-msi.c @@ -0,0 +1,268 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * SG2042 MSI Controller + * + * Copyright (C) 2024 Sophgo Technology Inc. + * Copyright (C) 2024 Chen Wang <unicorn_wang@outlook.com> + */ + +#include <linux/cleanup.h> +#include <linux/io.h> +#include <linux/irq.h> +#include <linux/irqdomain.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/msi.h> +#include <linux/of.h> +#include <linux/of_irq.h> +#include <linux/of_pci.h> +#include <linux/of_platform.h> +#include <linux/platform_device.h> +#include <linux/slab.h> + +#include "irq-msi-lib.h" + +#define SG2042_MAX_MSI_VECTOR 32 + +struct sg2042_msi_chipdata { + void __iomem *reg_clr; // clear reg, see TRM, 10.1.33, GP_INTR0_CLR + + phys_addr_t doorbell_addr; // see TRM, 10.1.32, GP_INTR0_SET + + u32 irq_first; // The vector number that MSIs starts + u32 num_irqs; // The number of vectors for MSIs + + DECLARE_BITMAP(msi_map, SG2042_MAX_MSI_VECTOR); + struct mutex msi_map_lock; // lock for msi_map +}; + +static int sg2042_msi_allocate_hwirq(struct sg2042_msi_chipdata *data, int num_req) +{ + int first; + + guard(mutex)(&data->msi_map_lock); + first = bitmap_find_free_region(data->msi_map, data->num_irqs, + get_count_order(num_req)); + return first >= 0 ? first : -ENOSPC; +} + +static void sg2042_msi_free_hwirq(struct sg2042_msi_chipdata *data, + int hwirq, int num_req) +{ + guard(mutex)(&data->msi_map_lock); + bitmap_release_region(data->msi_map, hwirq, get_count_order(num_req)); +} + +static void sg2042_msi_irq_ack(struct irq_data *d) +{ + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); + int bit_off = d->hwirq; + + writel(1 << bit_off, data->reg_clr); + + irq_chip_ack_parent(d); +} + +static void sg2042_msi_irq_compose_msi_msg(struct irq_data *d, + struct msi_msg *msg) +{ + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); + + msg->address_hi = upper_32_bits(data->doorbell_addr); + msg->address_lo = lower_32_bits(data->doorbell_addr); + msg->data = 1 << d->hwirq; +} + +static const struct irq_chip sg2042_msi_middle_irq_chip = { + .name = "SG2042 MSI", + .irq_ack = sg2042_msi_irq_ack, + .irq_mask = irq_chip_mask_parent, + .irq_unmask = irq_chip_unmask_parent, +#ifdef CONFIG_SMP + .irq_set_affinity = irq_chip_set_affinity_parent, +#endif + .irq_compose_msi_msg = sg2042_msi_irq_compose_msi_msg, +}; + +static int sg2042_msi_parent_domain_alloc(struct irq_domain *domain, + unsigned int virq, int hwirq) +{ + struct sg2042_msi_chipdata *data = domain->host_data; + struct irq_fwspec fwspec; + struct irq_data *d; + int ret; + + fwspec.fwnode = domain->parent->fwnode; + fwspec.param_count = 2; + fwspec.param[0] = data->irq_first + hwirq; + fwspec.param[1] = IRQ_TYPE_EDGE_RISING; + + ret = irq_domain_alloc_irqs_parent(domain, virq, 1, &fwspec); + if (ret) + return ret; + + d = irq_domain_get_irq_data(domain->parent, virq); + return d->chip->irq_set_type(d, IRQ_TYPE_EDGE_RISING); +} + +static int sg2042_msi_middle_domain_alloc(struct irq_domain *domain, + unsigned int virq, + unsigned int nr_irqs, void *args) +{ + struct sg2042_msi_chipdata *data = domain->host_data; + int hwirq, err, i; + + hwirq = sg2042_msi_allocate_hwirq(data, nr_irqs); + if (hwirq < 0) + return hwirq; + + for (i = 0; i < nr_irqs; i++) { + err = sg2042_msi_parent_domain_alloc(domain, virq + i, hwirq + i); + if (err) + goto err_hwirq; + + irq_domain_set_hwirq_and_chip(domain, virq + i, hwirq + i, + &sg2042_msi_middle_irq_chip, data); + } + + return 0; + +err_hwirq: + sg2042_msi_free_hwirq(data, hwirq, nr_irqs); + irq_domain_free_irqs_parent(domain, virq, i); + + return err; +} + +static void sg2042_msi_middle_domain_free(struct irq_domain *domain, + unsigned int virq, + unsigned int nr_irqs) +{ + struct irq_data *d = irq_domain_get_irq_data(domain, virq); + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); + + irq_domain_free_irqs_parent(domain, virq, nr_irqs); + sg2042_msi_free_hwirq(data, d->hwirq, nr_irqs); +} + +static const struct irq_domain_ops sg2042_msi_middle_domain_ops = { + .alloc = sg2042_msi_middle_domain_alloc, + .free = sg2042_msi_middle_domain_free, + .select = msi_lib_irq_domain_select, +}; + +#define SG2042_MSI_FLAGS_REQUIRED (MSI_FLAG_USE_DEF_DOM_OPS | \ + MSI_FLAG_USE_DEF_CHIP_OPS) + +#define SG2042_MSI_FLAGS_SUPPORTED MSI_GENERIC_FLAGS_MASK + +static struct msi_parent_ops sg2042_msi_parent_ops = { + .required_flags = SG2042_MSI_FLAGS_REQUIRED, + .supported_flags = SG2042_MSI_FLAGS_SUPPORTED, + .bus_select_mask = MATCH_PCI_MSI, + .bus_select_token = DOMAIN_BUS_NEXUS, + .prefix = "SG2042-", + .init_dev_msi_info = msi_lib_init_dev_msi_info, +}; + +static int sg2042_msi_init_domains(struct sg2042_msi_chipdata *data, + struct device_node *node) +{ + struct fwnode_handle *fwnode = of_node_to_fwnode(node); + struct irq_domain *plic_domain, *middle_domain; + struct device_node *plic_node; + + if (!of_find_property(node, "interrupt-parent", NULL)) { + pr_err("Can't find interrupt-parent!\n"); + return -EINVAL; + } + + plic_node = of_irq_find_parent(node); + if (!plic_node) { + pr_err("Failed to find the PLIC node!\n"); + return -ENXIO; + } + + plic_domain = irq_find_host(plic_node); + of_node_put(plic_node); + if (!plic_domain) { + pr_err("Failed to find the PLIC domain\n"); + return -ENXIO; + } + + middle_domain = irq_domain_create_hierarchy(plic_domain, 0, data->num_irqs, + fwnode, + &sg2042_msi_middle_domain_ops, + data); + if (!middle_domain) { + pr_err("Failed to create the MSI middle domain\n"); + return -ENOMEM; + } + + irq_domain_update_bus_token(middle_domain, DOMAIN_BUS_NEXUS); + + middle_domain->flags |= IRQ_DOMAIN_FLAG_MSI_PARENT; + middle_domain->msi_parent_ops = &sg2042_msi_parent_ops; + + return 0; +} + +static int sg2042_msi_probe(struct platform_device *pdev) +{ + struct sg2042_msi_chipdata *data; + struct of_phandle_args args = {}; + struct resource *res; + int ret; + + data = devm_kzalloc(&pdev->dev, sizeof(struct sg2042_msi_chipdata), GFP_KERNEL); + if (!data) + return -ENOMEM; + + data->reg_clr = devm_platform_ioremap_resource_byname(pdev, "clr"); + if (IS_ERR(data->reg_clr)) { + dev_err(&pdev->dev, "Failed to map clear register\n"); + return PTR_ERR(data->reg_clr); + } + + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "doorbell"); + if (!res) { + dev_err(&pdev->dev, "Failed get resource from set\n"); + return -EINVAL; + } + data->doorbell_addr = res->start; + + ret = of_parse_phandle_with_args(pdev->dev.of_node, "msi-ranges", + "#interrupt-cells", 0, &args); + if (ret) { + dev_err(&pdev->dev, "Unable to parse MSI vec base\n"); + return ret; + } + data->irq_first = (u32)args.args[0]; + + ret = of_property_read_u32_index(pdev->dev.of_node, "msi-ranges", + args.args_count + 1, &data->num_irqs); + if (ret) { + dev_err(&pdev->dev, "Unable to parse MSI vec number\n"); + of_node_put(args.np); + return ret; + } + of_node_put(args.np); + + mutex_init(&data->msi_map_lock); + + return sg2042_msi_init_domains(data, pdev->dev.of_node); +} + +static const struct of_device_id sg2042_msi_of_match[] = { + { .compatible = "sophgo,sg2042-msi" }, + {} +}; + +static struct platform_driver sg2042_msi_driver = { + .driver = { + .name = "sg2042-msi", + .of_match_table = sg2042_msi_of_match, + }, + .probe = sg2042_msi_probe, +}; +builtin_platform_driver(sg2042_msi_driver); -- 2.34.1 ^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller 2025-01-15 6:33 ` [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller Chen Wang @ 2025-01-20 2:43 ` Inochi Amaoto 2025-01-22 13:54 ` Chen Wang 2025-02-15 1:09 ` Inochi Amaoto 1 sibling, 1 reply; 15+ messages in thread From: Inochi Amaoto @ 2025-01-20 2:43 UTC (permalink / raw) To: Chen Wang, u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet Cc: Inochi Amaoto On Wed, Jan 15, 2025 at 02:33:45PM +0800, Chen Wang wrote: > From: Chen Wang <unicorn_wang@outlook.com> > > Add driver for Sophgo SG2042 MSI interrupt controller. > > Signed-off-by: Chen Wang <unicorn_wang@outlook.com> > --- > drivers/irqchip/Kconfig | 12 ++ > drivers/irqchip/Makefile | 1 + > drivers/irqchip/irq-sg2042-msi.c | 268 +++++++++++++++++++++++++++++++ > 3 files changed, 281 insertions(+) > create mode 100644 drivers/irqchip/irq-sg2042-msi.c > > diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig > index 9bee02db1643..161fb5df857f 100644 > --- a/drivers/irqchip/Kconfig > +++ b/drivers/irqchip/Kconfig > @@ -749,6 +749,18 @@ config MCHP_EIC > help > Support for Microchip External Interrupt Controller. > > +config SOPHGO_SG2042_MSI > + bool "Sophgo SG2042 MSI Controller" > + depends on ARCH_SOPHGO || COMPILE_TEST > + depends on PCI > + select IRQ_DOMAIN_HIERARCHY > + select IRQ_MSI_LIB > + select PCI_MSI > + help > + Support for the Sophgo SG2042 MSI Controller. > + This on-chip interrupt controller enables MSI sources to be > + routed to the primary PLIC controller on SoC. > + > config SUNPLUS_SP7021_INTC > bool "Sunplus SP7021 interrupt controller" if COMPILE_TEST > default SOC_SP7021 > diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile > index 25e9ad29b8c4..dd60e597491d 100644 > --- a/drivers/irqchip/Makefile > +++ b/drivers/irqchip/Makefile > @@ -128,4 +128,5 @@ obj-$(CONFIG_WPCM450_AIC) += irq-wpcm450-aic.o > obj-$(CONFIG_IRQ_IDT3243X) += irq-idt3243x.o > obj-$(CONFIG_APPLE_AIC) += irq-apple-aic.o > obj-$(CONFIG_MCHP_EIC) += irq-mchp-eic.o > +obj-$(CONFIG_SOPHGO_SG2042_MSI) += irq-sg2042-msi.o > obj-$(CONFIG_SUNPLUS_SP7021_INTC) += irq-sp7021-intc.o > diff --git a/drivers/irqchip/irq-sg2042-msi.c b/drivers/irqchip/irq-sg2042-msi.c > new file mode 100644 > index 000000000000..1f88e1671e5c > --- /dev/null > +++ b/drivers/irqchip/irq-sg2042-msi.c > @@ -0,0 +1,268 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * SG2042 MSI Controller > + * > + * Copyright (C) 2024 Sophgo Technology Inc. > + * Copyright (C) 2024 Chen Wang <unicorn_wang@outlook.com> > + */ > + > +#include <linux/cleanup.h> > +#include <linux/io.h> > +#include <linux/irq.h> > +#include <linux/irqdomain.h> > +#include <linux/kernel.h> > +#include <linux/module.h> > +#include <linux/msi.h> > +#include <linux/of.h> > +#include <linux/of_irq.h> > +#include <linux/of_pci.h> > +#include <linux/of_platform.h> > +#include <linux/platform_device.h> > +#include <linux/slab.h> > + > +#include "irq-msi-lib.h" > + > +#define SG2042_MAX_MSI_VECTOR 32 > + > +struct sg2042_msi_chipdata { > + void __iomem *reg_clr; // clear reg, see TRM, 10.1.33, GP_INTR0_CLR > + > + phys_addr_t doorbell_addr; // see TRM, 10.1.32, GP_INTR0_SET > + > + u32 irq_first; // The vector number that MSIs starts > + u32 num_irqs; // The number of vectors for MSIs > + > + DECLARE_BITMAP(msi_map, SG2042_MAX_MSI_VECTOR); > + struct mutex msi_map_lock; // lock for msi_map > +}; > + > +static int sg2042_msi_allocate_hwirq(struct sg2042_msi_chipdata *data, int num_req) > +{ > + int first; > + > + guard(mutex)(&data->msi_map_lock); > + first = bitmap_find_free_region(data->msi_map, data->num_irqs, > + get_count_order(num_req)); > + return first >= 0 ? first : -ENOSPC; > +} > + > +static void sg2042_msi_free_hwirq(struct sg2042_msi_chipdata *data, > + int hwirq, int num_req) > +{ > + guard(mutex)(&data->msi_map_lock); > + bitmap_release_region(data->msi_map, hwirq, get_count_order(num_req)); > +} > + > +static void sg2042_msi_irq_ack(struct irq_data *d) > +{ > + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); > + int bit_off = d->hwirq; > + > + writel(1 << bit_off, data->reg_clr); > + > + irq_chip_ack_parent(d); > +} > + > +static void sg2042_msi_irq_compose_msi_msg(struct irq_data *d, > + struct msi_msg *msg) > +{ > + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); > + > + msg->address_hi = upper_32_bits(data->doorbell_addr); > + msg->address_lo = lower_32_bits(data->doorbell_addr); > + msg->data = 1 << d->hwirq; > +} > + > +static const struct irq_chip sg2042_msi_middle_irq_chip = { > + .name = "SG2042 MSI", > + .irq_ack = sg2042_msi_irq_ack, > + .irq_mask = irq_chip_mask_parent, > + .irq_unmask = irq_chip_unmask_parent, > +#ifdef CONFIG_SMP > + .irq_set_affinity = irq_chip_set_affinity_parent, > +#endif > + .irq_compose_msi_msg = sg2042_msi_irq_compose_msi_msg, > +}; > + > +static int sg2042_msi_parent_domain_alloc(struct irq_domain *domain, > + unsigned int virq, int hwirq) > +{ > + struct sg2042_msi_chipdata *data = domain->host_data; > + struct irq_fwspec fwspec; > + struct irq_data *d; > + int ret; > + > + fwspec.fwnode = domain->parent->fwnode; > + fwspec.param_count = 2; > + fwspec.param[0] = data->irq_first + hwirq; > + fwspec.param[1] = IRQ_TYPE_EDGE_RISING; > + > + ret = irq_domain_alloc_irqs_parent(domain, virq, 1, &fwspec); > + if (ret) > + return ret; > + > + d = irq_domain_get_irq_data(domain->parent, virq); > + return d->chip->irq_set_type(d, IRQ_TYPE_EDGE_RISING); > +} > + > +static int sg2042_msi_middle_domain_alloc(struct irq_domain *domain, > + unsigned int virq, > + unsigned int nr_irqs, void *args) > +{ > + struct sg2042_msi_chipdata *data = domain->host_data; > + int hwirq, err, i; > + > + hwirq = sg2042_msi_allocate_hwirq(data, nr_irqs); > + if (hwirq < 0) > + return hwirq; > + > + for (i = 0; i < nr_irqs; i++) { > + err = sg2042_msi_parent_domain_alloc(domain, virq + i, hwirq + i); > + if (err) > + goto err_hwirq; > + > + irq_domain_set_hwirq_and_chip(domain, virq + i, hwirq + i, > + &sg2042_msi_middle_irq_chip, data); > + } > + > + return 0; > + > +err_hwirq: > + sg2042_msi_free_hwirq(data, hwirq, nr_irqs); > + irq_domain_free_irqs_parent(domain, virq, i); > + > + return err; > +} > + > +static void sg2042_msi_middle_domain_free(struct irq_domain *domain, > + unsigned int virq, > + unsigned int nr_irqs) > +{ > + struct irq_data *d = irq_domain_get_irq_data(domain, virq); > + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); > + > + irq_domain_free_irqs_parent(domain, virq, nr_irqs); > + sg2042_msi_free_hwirq(data, d->hwirq, nr_irqs); > +} > + > +static const struct irq_domain_ops sg2042_msi_middle_domain_ops = { > + .alloc = sg2042_msi_middle_domain_alloc, > + .free = sg2042_msi_middle_domain_free, > + .select = msi_lib_irq_domain_select, > +}; > + > +#define SG2042_MSI_FLAGS_REQUIRED (MSI_FLAG_USE_DEF_DOM_OPS | \ > + MSI_FLAG_USE_DEF_CHIP_OPS) > + > +#define SG2042_MSI_FLAGS_SUPPORTED MSI_GENERIC_FLAGS_MASK > + I think it is just OK to set the flag directly in the msi_parent_ops. > +static struct msi_parent_ops sg2042_msi_parent_ops = { > + .required_flags = SG2042_MSI_FLAGS_REQUIRED, > + .supported_flags = SG2042_MSI_FLAGS_SUPPORTED, > + .bus_select_mask = MATCH_PCI_MSI, > + .bus_select_token = DOMAIN_BUS_NEXUS, > + .prefix = "SG2042-", > + .init_dev_msi_info = msi_lib_init_dev_msi_info, > +}; This struct should be const. > + > +static int sg2042_msi_init_domains(struct sg2042_msi_chipdata *data, > + struct device_node *node) > +{ > + struct fwnode_handle *fwnode = of_node_to_fwnode(node); > + struct irq_domain *plic_domain, *middle_domain; > + struct device_node *plic_node; > + > + if (!of_find_property(node, "interrupt-parent", NULL)) { > + pr_err("Can't find interrupt-parent!\n"); > + return -EINVAL; > + } > + > + plic_node = of_irq_find_parent(node); > + if (!plic_node) { > + pr_err("Failed to find the PLIC node!\n"); > + return -ENXIO; > + } > + > + plic_domain = irq_find_host(plic_node); > + of_node_put(plic_node); > + if (!plic_domain) { > + pr_err("Failed to find the PLIC domain\n"); > + return -ENXIO; > + } > + > + middle_domain = irq_domain_create_hierarchy(plic_domain, 0, data->num_irqs, > + fwnode, > + &sg2042_msi_middle_domain_ops, > + data); > + if (!middle_domain) { > + pr_err("Failed to create the MSI middle domain\n"); > + return -ENOMEM; > + } > + > + irq_domain_update_bus_token(middle_domain, DOMAIN_BUS_NEXUS); > + > + middle_domain->flags |= IRQ_DOMAIN_FLAG_MSI_PARENT; > + middle_domain->msi_parent_ops = &sg2042_msi_parent_ops; > + > + return 0; > +} > + > +static int sg2042_msi_probe(struct platform_device *pdev) > +{ > + struct sg2042_msi_chipdata *data; > + struct of_phandle_args args = {}; > + struct resource *res; > + int ret; > + > + data = devm_kzalloc(&pdev->dev, sizeof(struct sg2042_msi_chipdata), GFP_KERNEL); > + if (!data) > + return -ENOMEM; > + > + data->reg_clr = devm_platform_ioremap_resource_byname(pdev, "clr"); > + if (IS_ERR(data->reg_clr)) { > + dev_err(&pdev->dev, "Failed to map clear register\n"); > + return PTR_ERR(data->reg_clr); > + } > + > + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "doorbell"); > + if (!res) { > + dev_err(&pdev->dev, "Failed get resource from set\n"); > + return -EINVAL; > + } > + data->doorbell_addr = res->start; > + > + ret = of_parse_phandle_with_args(pdev->dev.of_node, "msi-ranges", > + "#interrupt-cells", 0, &args); > + if (ret) { > + dev_err(&pdev->dev, "Unable to parse MSI vec base\n"); > + return ret; > + } > + data->irq_first = (u32)args.args[0]; > + > + ret = of_property_read_u32_index(pdev->dev.of_node, "msi-ranges", > + args.args_count + 1, &data->num_irqs); > + if (ret) { > + dev_err(&pdev->dev, "Unable to parse MSI vec number\n"); > + of_node_put(args.np); > + return ret; > + } > + of_node_put(args.np); > + > + mutex_init(&data->msi_map_lock); > + > + return sg2042_msi_init_domains(data, pdev->dev.of_node); > +} > + > +static const struct of_device_id sg2042_msi_of_match[] = { > + { .compatible = "sophgo,sg2042-msi" }, > + {} > +}; > + > +static struct platform_driver sg2042_msi_driver = { > + .driver = { > + .name = "sg2042-msi", > + .of_match_table = sg2042_msi_of_match, > + }, > + .probe = sg2042_msi_probe, > +}; > +builtin_platform_driver(sg2042_msi_driver); > -- > 2.34.1 > ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller 2025-01-20 2:43 ` Inochi Amaoto @ 2025-01-22 13:54 ` Chen Wang 0 siblings, 0 replies; 15+ messages in thread From: Chen Wang @ 2025-01-22 13:54 UTC (permalink / raw) To: Inochi Amaoto, Chen Wang, u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On 2025/1/20 10:43, Inochi Amaoto wrote: > On Wed, Jan 15, 2025 at 02:33:45PM +0800, Chen Wang wrote: [......] >> +#define SG2042_MSI_FLAGS_REQUIRED (MSI_FLAG_USE_DEF_DOM_OPS | \ >> + MSI_FLAG_USE_DEF_CHIP_OPS) >> + >> +#define SG2042_MSI_FLAGS_SUPPORTED MSI_GENERIC_FLAGS_MASK >> + > I think it is just OK to set the flag directly in the msi_parent_ops. This shouldn't be a big problem, maybe it will be convenient for future expansion. >> +static struct msi_parent_ops sg2042_msi_parent_ops = { >> + .required_flags = SG2042_MSI_FLAGS_REQUIRED, >> + .supported_flags = SG2042_MSI_FLAGS_SUPPORTED, >> + .bus_select_mask = MATCH_PCI_MSI, >> + .bus_select_token = DOMAIN_BUS_NEXUS, >> + .prefix = "SG2042-", >> + .init_dev_msi_info = msi_lib_init_dev_msi_info, >> +}; > This struct should be const. Yes, will fix this in next version. Thanks, Chen [......] ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller 2025-01-15 6:33 ` [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller Chen Wang 2025-01-20 2:43 ` Inochi Amaoto @ 2025-02-15 1:09 ` Inochi Amaoto 2025-02-15 3:36 ` Chen Wang 1 sibling, 1 reply; 15+ messages in thread From: Inochi Amaoto @ 2025-02-15 1:09 UTC (permalink / raw) To: Chen Wang, u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet Cc: Inochi Amaoto On Wed, Jan 15, 2025 at 02:33:45PM +0800, Chen Wang wrote: > From: Chen Wang <unicorn_wang@outlook.com> > > Add driver for Sophgo SG2042 MSI interrupt controller. > > Signed-off-by: Chen Wang <unicorn_wang@outlook.com> > --- > drivers/irqchip/Kconfig | 12 ++ > drivers/irqchip/Makefile | 1 + > drivers/irqchip/irq-sg2042-msi.c | 268 +++++++++++++++++++++++++++++++ > 3 files changed, 281 insertions(+) > create mode 100644 drivers/irqchip/irq-sg2042-msi.c > > diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig > index 9bee02db1643..161fb5df857f 100644 > --- a/drivers/irqchip/Kconfig > +++ b/drivers/irqchip/Kconfig > @@ -749,6 +749,18 @@ config MCHP_EIC > help > Support for Microchip External Interrupt Controller. > > +config SOPHGO_SG2042_MSI > + bool "Sophgo SG2042 MSI Controller" > + depends on ARCH_SOPHGO || COMPILE_TEST > + depends on PCI > + select IRQ_DOMAIN_HIERARCHY > + select IRQ_MSI_LIB > + select PCI_MSI > + help > + Support for the Sophgo SG2042 MSI Controller. > + This on-chip interrupt controller enables MSI sources to be > + routed to the primary PLIC controller on SoC. > + > config SUNPLUS_SP7021_INTC > bool "Sunplus SP7021 interrupt controller" if COMPILE_TEST > default SOC_SP7021 > diff --git a/drivers/irqchip/Makefile b/drivers/irqchip/Makefile > index 25e9ad29b8c4..dd60e597491d 100644 > --- a/drivers/irqchip/Makefile > +++ b/drivers/irqchip/Makefile > @@ -128,4 +128,5 @@ obj-$(CONFIG_WPCM450_AIC) += irq-wpcm450-aic.o > obj-$(CONFIG_IRQ_IDT3243X) += irq-idt3243x.o > obj-$(CONFIG_APPLE_AIC) += irq-apple-aic.o > obj-$(CONFIG_MCHP_EIC) += irq-mchp-eic.o > +obj-$(CONFIG_SOPHGO_SG2042_MSI) += irq-sg2042-msi.o > obj-$(CONFIG_SUNPLUS_SP7021_INTC) += irq-sp7021-intc.o > diff --git a/drivers/irqchip/irq-sg2042-msi.c b/drivers/irqchip/irq-sg2042-msi.c > new file mode 100644 > index 000000000000..1f88e1671e5c > --- /dev/null > +++ b/drivers/irqchip/irq-sg2042-msi.c > @@ -0,0 +1,268 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * SG2042 MSI Controller > + * > + * Copyright (C) 2024 Sophgo Technology Inc. > + * Copyright (C) 2024 Chen Wang <unicorn_wang@outlook.com> > + */ > + > +#include <linux/cleanup.h> > +#include <linux/io.h> > +#include <linux/irq.h> > +#include <linux/irqdomain.h> > +#include <linux/kernel.h> > +#include <linux/module.h> > +#include <linux/msi.h> > +#include <linux/of.h> > +#include <linux/of_irq.h> > +#include <linux/of_pci.h> > +#include <linux/of_platform.h> > +#include <linux/platform_device.h> > +#include <linux/slab.h> > + > +#include "irq-msi-lib.h" > + > +#define SG2042_MAX_MSI_VECTOR 32 > + > +struct sg2042_msi_chipdata { > + void __iomem *reg_clr; // clear reg, see TRM, 10.1.33, GP_INTR0_CLR > + > + phys_addr_t doorbell_addr; // see TRM, 10.1.32, GP_INTR0_SET > + > + u32 irq_first; // The vector number that MSIs starts > + u32 num_irqs; // The number of vectors for MSIs > + > + DECLARE_BITMAP(msi_map, SG2042_MAX_MSI_VECTOR); > + struct mutex msi_map_lock; // lock for msi_map > +}; > + > +static int sg2042_msi_allocate_hwirq(struct sg2042_msi_chipdata *data, int num_req) > +{ > + int first; > + > + guard(mutex)(&data->msi_map_lock); > + first = bitmap_find_free_region(data->msi_map, data->num_irqs, > + get_count_order(num_req)); > + return first >= 0 ? first : -ENOSPC; > +} > + > +static void sg2042_msi_free_hwirq(struct sg2042_msi_chipdata *data, > + int hwirq, int num_req) > +{ > + guard(mutex)(&data->msi_map_lock); > + bitmap_release_region(data->msi_map, hwirq, get_count_order(num_req)); > +} > + > +static void sg2042_msi_irq_ack(struct irq_data *d) > +{ > + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); > + int bit_off = d->hwirq; > + > + writel(1 << bit_off, data->reg_clr); > + > + irq_chip_ack_parent(d); > +} > + > +static void sg2042_msi_irq_compose_msi_msg(struct irq_data *d, > + struct msi_msg *msg) > +{ > + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); > + > + msg->address_hi = upper_32_bits(data->doorbell_addr); > + msg->address_lo = lower_32_bits(data->doorbell_addr); > + msg->data = 1 << d->hwirq; > +} > + > +static const struct irq_chip sg2042_msi_middle_irq_chip = { > + .name = "SG2042 MSI", > + .irq_ack = sg2042_msi_irq_ack, > + .irq_mask = irq_chip_mask_parent, > + .irq_unmask = irq_chip_unmask_parent, > +#ifdef CONFIG_SMP > + .irq_set_affinity = irq_chip_set_affinity_parent, > +#endif > + .irq_compose_msi_msg = sg2042_msi_irq_compose_msi_msg, > +}; > + > +static int sg2042_msi_parent_domain_alloc(struct irq_domain *domain, > + unsigned int virq, int hwirq) > +{ > + struct sg2042_msi_chipdata *data = domain->host_data; > + struct irq_fwspec fwspec; > + struct irq_data *d; > + int ret; > + > + fwspec.fwnode = domain->parent->fwnode; > + fwspec.param_count = 2; > + fwspec.param[0] = data->irq_first + hwirq; > + fwspec.param[1] = IRQ_TYPE_EDGE_RISING; > + > + ret = irq_domain_alloc_irqs_parent(domain, virq, 1, &fwspec); > + if (ret) > + return ret; > + > + d = irq_domain_get_irq_data(domain->parent, virq); > + return d->chip->irq_set_type(d, IRQ_TYPE_EDGE_RISING); > +} > + > +static int sg2042_msi_middle_domain_alloc(struct irq_domain *domain, > + unsigned int virq, > + unsigned int nr_irqs, void *args) > +{ > + struct sg2042_msi_chipdata *data = domain->host_data; > + int hwirq, err, i; > + > + hwirq = sg2042_msi_allocate_hwirq(data, nr_irqs); > + if (hwirq < 0) > + return hwirq; > + > + for (i = 0; i < nr_irqs; i++) { > + err = sg2042_msi_parent_domain_alloc(domain, virq + i, hwirq + i); > + if (err) > + goto err_hwirq; > + > + irq_domain_set_hwirq_and_chip(domain, virq + i, hwirq + i, > + &sg2042_msi_middle_irq_chip, data); > + } > + > + return 0; > + > +err_hwirq: > + sg2042_msi_free_hwirq(data, hwirq, nr_irqs); > + irq_domain_free_irqs_parent(domain, virq, i); > + > + return err; > +} > + > +static void sg2042_msi_middle_domain_free(struct irq_domain *domain, > + unsigned int virq, > + unsigned int nr_irqs) > +{ > + struct irq_data *d = irq_domain_get_irq_data(domain, virq); > + struct sg2042_msi_chipdata *data = irq_data_get_irq_chip_data(d); > + > + irq_domain_free_irqs_parent(domain, virq, nr_irqs); > + sg2042_msi_free_hwirq(data, d->hwirq, nr_irqs); > +} > + > +static const struct irq_domain_ops sg2042_msi_middle_domain_ops = { > + .alloc = sg2042_msi_middle_domain_alloc, > + .free = sg2042_msi_middle_domain_free, > + .select = msi_lib_irq_domain_select, > +}; > + > +#define SG2042_MSI_FLAGS_REQUIRED (MSI_FLAG_USE_DEF_DOM_OPS | \ > + MSI_FLAG_USE_DEF_CHIP_OPS) > + > +#define SG2042_MSI_FLAGS_SUPPORTED MSI_GENERIC_FLAGS_MASK > + > +static struct msi_parent_ops sg2042_msi_parent_ops = { > + .required_flags = SG2042_MSI_FLAGS_REQUIRED, > + .supported_flags = SG2042_MSI_FLAGS_SUPPORTED, > + .bus_select_mask = MATCH_PCI_MSI, > + .bus_select_token = DOMAIN_BUS_NEXUS, > + .prefix = "SG2042-", > + .init_dev_msi_info = msi_lib_init_dev_msi_info, > +}; > + > +static int sg2042_msi_init_domains(struct sg2042_msi_chipdata *data, > + struct device_node *node) For ACPI support, I prefer to use device instead of device_node The following suggestion is also for ACPI support > +{ > + struct fwnode_handle *fwnode = of_node_to_fwnode(node); > + struct irq_domain *plic_domain, *middle_domain; > + struct device_node *plic_node; > + > + if (!of_find_property(node, "interrupt-parent", NULL)) { > + pr_err("Can't find interrupt-parent!\n"); > + return -EINVAL; > + } > + > + plic_node = of_irq_find_parent(node); > + if (!plic_node) { > + pr_err("Failed to find the PLIC node!\n"); > + return -ENXIO; > + } > + This can be replace with fwnode_find_reference to support ACPI > + plic_domain = irq_find_host(plic_node); > + of_node_put(plic_node); > + if (!plic_domain) { > + pr_err("Failed to find the PLIC domain\n"); > + return -ENXIO; > + } > + Replace irq_find_host with irq_find_matching_fwnode > + middle_domain = irq_domain_create_hierarchy(plic_domain, 0, data->num_irqs, > + fwnode, > + &sg2042_msi_middle_domain_ops, > + data); Use irq_domain_create_hierarchy > + if (!middle_domain) { > + pr_err("Failed to create the MSI middle domain\n"); > + return -ENOMEM; > + } > + > + irq_domain_update_bus_token(middle_domain, DOMAIN_BUS_NEXUS); > + > + middle_domain->flags |= IRQ_DOMAIN_FLAG_MSI_PARENT; > + middle_domain->msi_parent_ops = &sg2042_msi_parent_ops; > + > + return 0; > +} > + > +static int sg2042_msi_probe(struct platform_device *pdev) > +{ add a "struct device *dev = &pdev->dev;" here and replace &pdev->dev with dev. > + struct sg2042_msi_chipdata *data; > + struct of_phandle_args args = {}; > + struct resource *res; > + int ret; > + > + data = devm_kzalloc(&pdev->dev, sizeof(struct sg2042_msi_chipdata), GFP_KERNEL); > + if (!data) > + return -ENOMEM; > + > + data->reg_clr = devm_platform_ioremap_resource_byname(pdev, "clr"); > + if (IS_ERR(data->reg_clr)) { > + dev_err(&pdev->dev, "Failed to map clear register\n"); > + return PTR_ERR(data->reg_clr); > + } > + > + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "doorbell"); > + if (!res) { > + dev_err(&pdev->dev, "Failed get resource from set\n"); > + return -EINVAL; > + } > + data->doorbell_addr = res->start; > + > + ret = of_parse_phandle_with_args(pdev->dev.of_node, "msi-ranges", > + "#interrupt-cells", 0, &args); > + if (ret) { > + dev_err(&pdev->dev, "Unable to parse MSI vec base\n"); > + return ret; > + } > + data->irq_first = (u32)args.args[0]; > + > + ret = of_property_read_u32_index(pdev->dev.of_node, "msi-ranges", > + args.args_count + 1, &data->num_irqs); > + if (ret) { > + dev_err(&pdev->dev, "Unable to parse MSI vec number\n"); > + of_node_put(args.np); > + return ret; > + } > + of_node_put(args.np); this can be replace by calling fwnode_property_get_reference_args twice. I have found a working example: ret = fwnode_property_get_reference_args(dev_fwnode(dev), "msi-ranges", "#interrupt-cells", 0, 0, &args); if (ret) { dev_err(dev, "Unable to parse MSI vec base\n"); return ret; } fwnode_handle_put(args.fwnode); ret = fwnode_property_get_reference_args(dev_fwnode(dev), "msi-ranges", NULL, args.nargs + 1, 0, &args); if (ret) { dev_err(dev, "Unable to parse MSI vec number\n"); return ret; } data->irq_first = (u32)args.args[0]; data->num_irqs = (u32)args.args[args.nargs - 1]; fwnode_handle_put(args.fwnode); Regards, Inochi ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller 2025-02-15 1:09 ` Inochi Amaoto @ 2025-02-15 3:36 ` Chen Wang 2025-02-15 4:03 ` Inochi Amaoto 0 siblings, 1 reply; 15+ messages in thread From: Chen Wang @ 2025-02-15 3:36 UTC (permalink / raw) To: Inochi Amaoto Cc: Chen Wang, u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On 2025/2/15 9:09, Inochi Amaoto wrote: > On Wed, Jan 15, 2025 at 02:33:45PM +0800, Chen Wang wrote: [......] >> + middle_domain = irq_domain_create_hierarchy(plic_domain, 0, data->num_irqs, >> + fwnode, >> + &sg2042_msi_middle_domain_ops, >> + data); > Use irq_domain_create_hierarchy What's the difference? [......] ^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller 2025-02-15 3:36 ` Chen Wang @ 2025-02-15 4:03 ` Inochi Amaoto 0 siblings, 0 replies; 15+ messages in thread From: Inochi Amaoto @ 2025-02-15 4:03 UTC (permalink / raw) To: Chen Wang, Inochi Amaoto Cc: Chen Wang, u.kleine-koenig, aou, arnd, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet On Sat, Feb 15, 2025 at 11:36:18AM +0800, Chen Wang wrote: > > On 2025/2/15 9:09, Inochi Amaoto wrote: > > On Wed, Jan 15, 2025 at 02:33:45PM +0800, Chen Wang wrote: > [......] > > > + middle_domain = irq_domain_create_hierarchy(plic_domain, 0, data->num_irqs, > > > + fwnode, > > > + &sg2042_msi_middle_domain_ops, > > > + data); > > Use irq_domain_create_hierarchy > > What's the difference? > I made a mistake, just ignore it. The original is OK. Regard, Inochi ^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH v3 3/3] riscv: sophgo: dts: add msi controller for SG2042 2025-01-15 6:31 [PATCH v3 0/3] irqchip: Add Sophgo SG2042 MSI controller Chen Wang 2025-01-15 6:33 ` [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI Chen Wang 2025-01-15 6:33 ` [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller Chen Wang @ 2025-01-15 6:34 ` Chen Wang 2 siblings, 0 replies; 15+ messages in thread From: Chen Wang @ 2025-01-15 6:34 UTC (permalink / raw) To: u.kleine-koenig, aou, arnd, unicorn_wang, conor+dt, guoren, inochiama, krzk+dt, palmer, paul.walmsley, robh, tglx, devicetree, linux-kernel, linux-riscv, chao.wei, xiaoguang.xing, fengchun.li, samuel.holland, christophe.jaillet From: Chen Wang <unicorn_wang@outlook.com> Add msi-controller node to dts for SG2042. Signed-off-by: Chen Wang <unicorn_wang@outlook.com> --- arch/riscv/boot/dts/sophgo/sg2042.dtsi | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/riscv/boot/dts/sophgo/sg2042.dtsi b/arch/riscv/boot/dts/sophgo/sg2042.dtsi index e62ac51ac55a..02fbb978973c 100644 --- a/arch/riscv/boot/dts/sophgo/sg2042.dtsi +++ b/arch/riscv/boot/dts/sophgo/sg2042.dtsi @@ -173,6 +173,16 @@ pllclk: clock-controller@70300100c0 { #clock-cells = <1>; }; + msi: msi-controller@7030010300 { + compatible = "sophgo,sg2042-msi"; + reg = <0x70 0x30010300 0x0 0x4>, + <0x70 0x30010304 0x0 0x4>; + reg-names = "doorbell", "clr"; + msi-controller; + msi-ranges = <&intc 64 IRQ_TYPE_LEVEL_HIGH 32>; + interrupt-parent = <&intc>; + }; + rpgate: clock-controller@7030010368 { compatible = "sophgo,sg2042-rpgate"; reg = <0x70 0x30010368 0x0 0x98>; -- 2.34.1 ^ permalink raw reply related [flat|nested] 15+ messages in thread
end of thread, other threads:[~2025-02-18 1:27 UTC | newest] Thread overview: 15+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2025-01-15 6:31 [PATCH v3 0/3] irqchip: Add Sophgo SG2042 MSI controller Chen Wang 2025-01-15 6:33 ` [PATCH v3 1/3] dt-bindings: interrupt-controller: Add Sophgo SG2042 MSI Chen Wang 2025-01-20 2:42 ` Inochi Amaoto 2025-01-22 13:51 ` Chen Wang 2025-01-22 22:42 ` Inochi Amaoto 2025-01-23 0:47 ` Chen Wang 2025-01-23 21:29 ` Rob Herring 2025-02-18 1:26 ` Chen Wang 2025-01-15 6:33 ` [PATCH v3 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller Chen Wang 2025-01-20 2:43 ` Inochi Amaoto 2025-01-22 13:54 ` Chen Wang 2025-02-15 1:09 ` Inochi Amaoto 2025-02-15 3:36 ` Chen Wang 2025-02-15 4:03 ` Inochi Amaoto 2025-01-15 6:34 ` [PATCH v3 3/3] riscv: sophgo: dts: add msi controller for SG2042 Chen Wang
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).