Skip to contentSkip to conversation

Realtek RTL838x/RTL839x IRQ driver

PATCH Superseded Bert Vermeulen 4 participants 8 messages Jan 4, 2021 – Feb 2, 2021

Subsystems (2)

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

v2 cover Realtek RTL838x/RTL839x IRQ driver

v2:
- Addressed all comments by Marc Zyngier.
- Moved interrupt routing symbols to device tree interrupt-map. Parsing is done similar to the renesas,rza1-irqc driver.

Bert Vermeulen (2):
dt-bindings: interrupt-controller: Add Realtek RTL838x/RTL839x support
irqchip: Add support for Realtek RTL838x/RTL839x IRQ controller

 .../realtek,rtl-intc.yaml                     |  57 ++++++
 drivers/irqchip/Makefile                      |   1 +
 drivers/irqchip/irq-realtek-rtl.c             | 180 ++++++++++++++++++
 3 files changed, 238 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml
 create mode 100644 drivers/irqchip/irq-realtek-rtl.c
Signature · 1 line
2.25.1
#1 lore raw

PATCH v2 1/2 dt-bindings: interrupt-controller: Add Realtek RTL838x/RTL839x support

---
 .../realtek,rtl-intc.yaml                     | 57 +++++++++++++++++++
 1 file changed, 57 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml
Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml new file +57 -0
new file mode 100644
index 000000000000..e9b18a65cafb
@@ -0,0 +1,57 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/interrupt-controller/realtek,rtl-intc.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Realtek RTL SoC interrupt controller devicetree bindings
+
+maintainers:
+  - Birger Koblitz <[email protected]>
+  - Bert Vermeulen <[email protected]>
+  - John Crispin <[email protected]>
+
+properties:
+  compatible:
+    const: realtek,rtl-intc
+
+  "#interrupt-cells":
+    const: 1
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  interrupt-controller: true
+
+  "#address-cells":
+    const: 0
+
+  interrupt-map:
+    maxItems: 24
+    description: Describes mapping from SoC interrupts to CPU interrupts
+
+required:
+  - compatible
+  - reg
+  - "#interrupt-cells"
+  - interrupt-controller
+  - "#address-cells"
+  - interrupt-map
+
+additionalProperties: false
+
+examples:
+  - |
+    intc: interrupt-controller@3000 {
+      compatible = "realtek,rtl-intc";
+      #interrupt-cells = <1>;
+      interrupt-controller;
+      reg = <0x3000 0x20>;
+      #address-cells = <0>;
+      interrupt-map =
+              <31 &cpuintc 2>,
+              <30 &cpuintc 1>;
+    }
Signature · 1 line
2.25.1
maintainer
#2 lore raw

Re: [PATCH v2 1/2] dt-bindings: interrupt-controller: Add Realtek RTL838x/RTL839x support

Quoted 6 lines · Bert Vermeulen
---
 .../realtek,rtl-intc.yaml                     | 57 +++++++++++++++++++
 1 file changed, 57 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml

My bot found errors running 'make dt_binding_check' on your patch:

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/linux-dt-review/Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml: properties: 'interrupt-map-mask' is a dependency of 'interrupt-map' /builds/robherring/linux-dt-review/Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml: ignoring, error in schema: properties
warning: no schema found in file: ./Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml
Error: Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.example.dts:30.5-6 syntax error
FATAL ERROR: Unable to parse input tree
make[1]: *** [scripts/Makefile.lib:344: Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.example.dt.yaml] Error 1
make: *** [Makefile:1370: dt_binding_check] Error 2

See https://patchwork.ozlabs.org/patch/1422078

This check can fail if there are any dependencies. The base for a patch series is generally the most recent rc1.

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.

#3 lore raw

Re: [PATCH v2 1/2] dt-bindings: interrupt-controller: Add Realtek RTL838x/RTL839x support

Quoted 9 lines · Rob Herring
Quoted 6 lines · Bert Vermeulen
---
 .../realtek,rtl-intc.yaml                     | 57 +++++++++++++++++++
 1 file changed, 57 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml

My bot found errors running 'make dt_binding_check' on your patch:

I fixed the syntax error that caused this to fail. But running the test now produces some errors that are not so obvious:

interrupt-controller@3000: interrupt-map: [[31, 4294967295, 2], [30,
4294967295, 1], [29, 4294967295, 5]] is too short

Not sure where this comes from, but running dt-validate -v gives this:

Failed validating 'minItems' in schema['properties']['interrupt-map']:

    {'maxItems': 24, 'minItems': 24}

Where did that 24 come from? I don't even have 24 initerrupts to map :-)

Another error is this:

interrupt-controller@3000: 'interrupt-map-mask' is a dependency of
'interrupt-map'

I can see this in your dtschema package's schemas/interrupt-controller.yaml where it's defined as a dependency, but why? I don't need that property.

Signature · 2 lines
Bert Vermeulen
[email protected]
maintainer
#4 lore raw

Re: [PATCH v2 1/2] dt-bindings: interrupt-controller: Add Realtek RTL838x/RTL839x support

Quoted 16 lines · Bert Vermeulen
Quoted 9 lines · Rob Herring
Quoted 6 lines · Bert Vermeulen
---
 .../realtek,rtl-intc.yaml                     | 57 +++++++++++++++++++
 1 file changed, 57 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/interrupt-controller/realtek,rtl-intc.yaml

My bot found errors running 'make dt_binding_check' on your patch:

I fixed the syntax error that caused this to fail. But running the test now produces some errors that are not so obvious:

   interrupt-controller@3000: interrupt-map: [[31, 4294967295, 2], [30,
Quoted 8 lines

4294967295, 1], [29, 4294967295, 5]] is too short

Not sure where this comes from, but running dt-validate -v gives this:

Failed validating 'minItems' in schema['properties']['interrupt-map']:

     {'maxItems': 24, 'minItems': 24}

Where did that 24 come from? I don't even have 24 initerrupts to map :-)

If you don't define minItems, then the default is the same as maxItems. So you need to set 'minItems'.

Quoted 8 lines

Another error is this:

interrupt-controller@3000: 'interrupt-map-mask' is a dependency of
'interrupt-map'

I can see this in your dtschema package's schemas/interrupt-controller.yaml where it's defined as a dependency, but why? I don't need that property.

This should be the other way around. I'll fix it.

Rob

#5 lore raw

PATCH v2 2/2 irqchip: Add support for Realtek RTL838x/RTL839x IRQ controller

In reply to Bert Vermeulen

This is a standard IRQ driver with only status and mask registers.

The mapping from SoC interrupts (18-31) to MIPS core interrupts is done via an interrupt-map in device tree.

---
 drivers/irqchip/Makefile          |   1 +
 drivers/irqchip/irq-realtek-rtl.c | 180 ++++++++++++++++++++++++++++++
 2 files changed, 181 insertions(+)
 create mode 100644 drivers/irqchip/irq-realtek-rtl.c
drivers/irqchip/Makefile +1 -0
index 0ac93bfaec61..4fc1086bed7e 100644
@@ -113,3 +113,4 @@ obj-$(CONFIG_LOONGSON_PCH_PIC)		+= irq-loongson-pch-pic.o
 obj-$(CONFIG_LOONGSON_PCH_MSI)		+= irq-loongson-pch-msi.o
 obj-$(CONFIG_MST_IRQ)			+= irq-mst-intc.o
 obj-$(CONFIG_SL28CPLD_INTC)		+= irq-sl28cpld.o
+obj-$(CONFIG_MACH_REALTEK_RTL)		+= irq-realtek-rtl.o
drivers/irqchip/irq-realtek-rtl.c new file +180 -0

180 lines not on this page · raw message

Signature · 1 line
2.25.1
#6 lore raw

Re: [PATCH v2 2/2] irqchip: Add support for Realtek RTL838x/RTL839x IRQ controller

Quoted 6 lines · Bert Vermeulen

This is a standard IRQ driver with only status and mask registers.

The mapping from SoC interrupts (18-31) to MIPS core interrupts is done via an interrupt-map in device tree.

Quoted 201 lines
---
  drivers/irqchip/Makefile          |   1 +
  drivers/irqchip/irq-realtek-rtl.c | 180 ++++++++++++++++++++++++++++++
  2 files changed, 181 insertions(+)

create mode 100644 drivers/irqchip/irq-realtek-rtl.c

drivers/irqchip/Makefile +1 -0
index 0ac93bfaec61..4fc1086bed7e 100644
@@ -113,3 +113,4 @@ obj-$(CONFIG_LOONGSON_PCH_PIC)		+= irq-loongson-pch-pic.o
  obj-$(CONFIG_LOONGSON_PCH_MSI)		+= irq-loongson-pch-msi.o
  obj-$(CONFIG_MST_IRQ)			+= irq-mst-intc.o
  obj-$(CONFIG_SL28CPLD_INTC)		+= irq-sl28cpld.o
+obj-$(CONFIG_MACH_REALTEK_RTL)		+= irq-realtek-rtl.o
drivers/irqchip/irq-realtek-rtl.c new file +180 -0
new file mode 100644
index 000000000000..bafe9ee4a85a
@@ -0,0 +1,180 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2006-2012 Tony Wu <[email protected]>
+ * Copyright (C) 2020 Birger Koblitz <[email protected]>
+ * Copyright (C) 2020 Bert Vermeulen <[email protected]>
+ * Copyright (C) 2020 John Crispin <[email protected]>
+ */
+
+#include <linux/of_irq.h>
+#include <linux/irqchip.h>
+#include <linux/spinlock.h>
+#include <linux/of_address.h>
+#include <linux/irqchip/chained_irq.h>
+
+/* Global Interrupt Mask Register */
+#define RTL_ICTL_GIMR		0x00
+/* Global Interrupt Status Register */
+#define RTL_ICTL_GISR		0x04
+/* Interrupt Routing Registers */
+#define RTL_ICTL_IRR0		0x08
+#define RTL_ICTL_IRR1		0x0c
+#define RTL_ICTL_IRR2		0x10
+#define RTL_ICTL_IRR3		0x14
+
+#define REG(x)		(realtek_ictl_base + x)
+
+static DEFINE_RAW_SPINLOCK(irq_lock);
+static void __iomem *realtek_ictl_base;
+
+static void realtek_ictl_unmask_irq(struct irq_data *i)
+{
+	unsigned long flags;
+	u32 value;
+
+	raw_spin_lock_irqsave(&irq_lock, flags);
+
+	value = readl(REG(RTL_ICTL_GIMR));
+	value |= BIT(i->hwirq);
+	writel(value, REG(RTL_ICTL_GIMR));
+
+	raw_spin_unlock_irqrestore(&irq_lock, flags);
+}
+
+static void realtek_ictl_mask_irq(struct irq_data *i)
+{
+	unsigned long flags;
+	u32 value;
+
+	raw_spin_lock_irqsave(&irq_lock, flags);
+
+	value = readl(REG(RTL_ICTL_GIMR));
+	value &= ~BIT(i->hwirq);
+	writel(value, REG(RTL_ICTL_GIMR));
+
+	raw_spin_unlock_irqrestore(&irq_lock, flags);
+}
+
+static struct irq_chip realtek_ictl_irq = {
+	.name = "realtek-rtl-intc",
+	.irq_mask = realtek_ictl_mask_irq,
+	.irq_unmask = realtek_ictl_unmask_irq,
+};
+
+static int intc_map(struct irq_domain *d, unsigned int irq, irq_hw_number_t hw)
+{
+	irq_set_chip_and_handler(hw, &realtek_ictl_irq, handle_level_irq);
+
+	return 0;
+}
+
+static const struct irq_domain_ops irq_domain_ops = {
+	.map = intc_map,
+	.xlate = irq_domain_xlate_onecell,
+};
+
+static void realtek_irq_dispatch(struct irq_desc *desc)
+{
+	struct irq_chip *chip = irq_desc_get_chip(desc);
+	struct irq_domain *domain;
+	unsigned int pending;
+
+	chained_irq_enter(chip, desc);
+	pending = readl(REG(RTL_ICTL_GIMR)) & readl(REG(RTL_ICTL_GISR));
+	if (unlikely(!pending)) {
+		spurious_interrupt();
+		goto out;
+	}
+	domain = irq_desc_get_handler_data(desc);
+	generic_handle_irq(irq_find_mapping(domain, __ffs(pending)));
+
+out:
+	chained_irq_exit(chip, desc);
+}
+
+/*
+ * SoC interrupts are cascaded to MIPS CPU interrupts according to the
+ * interrupt-map in the device tree. Each SoC interrupt gets 4 bits for
+ * the CPU interrupt in an Interrupt Routing Register. Max 32 SoC interrupts
+ * thus go into 4 IRRs.
+ */
+static int __init map_interrupts(struct device_node *node)
+{
+	struct device_node *cpu_ictl;
+	const __be32 *imap;
+	u32 imaplen, soc_int, cpu_int, tmp, regs[4];
+	int ret, i, irr_regs[] = {
+		RTL_ICTL_IRR3,
+		RTL_ICTL_IRR2,
+		RTL_ICTL_IRR1,
+		RTL_ICTL_IRR0,
+	};
+
+	ret = of_property_read_u32(node, "#address-cells", &tmp);
+	if (ret || tmp)
+		return -EINVAL;
+
+	imap = of_get_property(node, "interrupt-map", &imaplen);
+	if (!imap || imaplen % 3)
+		return -EINVAL;
+
+	memset(regs, 0, sizeof(regs));
+	for (i = 0; i < imaplen; i += 3 * sizeof(u32)) {
+		soc_int = be32_to_cpup(imap);
+		if (soc_int > 31)
+			return -EINVAL;
+
+		cpu_ictl = of_find_node_by_phandle(be32_to_cpup(imap + 1));
+		if (!cpu_ictl)
+			return -EINVAL;
+		ret = of_property_read_u32(cpu_ictl, "#interrupt-cells", &tmp);
+		if (ret || tmp != 1)
+			return -EINVAL;
+		of_node_put(cpu_ictl);
+
+		cpu_int = be32_to_cpup(imap + 2);
+		if (cpu_int > 7)
+			return -EINVAL;
+
+		regs[(soc_int * 4) / 32] |= cpu_int << (soc_int * 4) % 32;
+		imap += 3;
+	}
+
+	for (i = 0; i < 4; i++)
+		writel(regs[i], REG(irr_regs[i]));
+
+	return 0;
+}
+
+static int __init realtek_rtl_of_init(struct device_node *node, struct device_node *parent)
+{
+	struct irq_domain *domain;
+	int ret;
+
+	domain = irq_domain_add_simple(node, 32, 0,
+				       &irq_domain_ops, NULL);
+	irq_set_chained_handler_and_data(2, realtek_irq_dispatch, domain);
+	irq_set_chained_handler_and_data(3, realtek_irq_dispatch, domain);
+	irq_set_chained_handler_and_data(4, realtek_irq_dispatch, domain);
+	irq_set_chained_handler_and_data(5, realtek_irq_dispatch, domain);
+
+	realtek_ictl_base = of_iomap(node, 0);
+	if (!realtek_ictl_base)
+		return -ENXIO;
+
+	/* Disable all cascaded interrupts */
+	writel(0, REG(RTL_ICTL_GIMR));
+
+	ret = map_interrupts(node);
+	if (ret) {
+		pr_err("invalid interrupt map\n");
+		return ret;
+	}
+
+	/* Clear timer interrupt */
+	write_c0_compare(0);
+
+	return 0;
+}
+
+IRQCHIP_DECLARE(realtek_rtl_intc, "realtek,rtl-intc", realtek_rtl_of_init);
#7 lore raw

Re: [PATCH v2 2/2] irqchip: Add support for Realtek RTL838x/RTL839x IRQ controller

John,

Quoted 8 lines · John Crispin
Quoted 6 lines · Bert Vermeulen

This is a standard IRQ driver with only status and mask registers.

The mapping from SoC interrupts (18-31) to MIPS core interrupts is done via an interrupt-map in device tree.

There is already a v4 on the list[1], so you may want to comment on that one.

It would also help if you'd give a reason for your SoB to be added. Or did you mean to Ack the patch?

Thanks,

        M.

[1] https://lore.kernel.ortg/r/[email protected]

Signature · 1 line
Jazz is not dead. It just smells funny...