Skip to contentSkip to conversation

irqchip: qcom: Add IRQ combiner driver

PATCH Superseded Agustin Vega-Frias 4 participants 9 messages Oct 18, 2016 – Oct 25, 2016

Subsystems (4)

v5 of 12 Superseded Latest is v12 →

lore
Share & tools
lore
#0 lore raw

v5 cover irqchip: qcom: Add IRQ combiner driver

Add support for IRQ combiners in the Top-level Control and Status Registers (TCSR) hardware block in Qualcomm Technologies chips.

The first patch adds support for ResourceSource/IRQ domain mapping when using Extended IRQ Resources with a specific ResourceSource. The core ACPI resource management code has been changed to lookup the IRQ domain when an IRQ resource indicates a ResourceSource, and register the IRQ on that domain, instead of a GSI.

The second patch takes advantage of the new capabilities to implement the driver for the IRQ combiners.

Changes V1 -> V2:
* Remove use of GPIO library for the combiner
* Refactor to use fwnode/ResourceSource to IRQ domain mapping introduced in ACPI core

Changes V2 -> V3:
* Removed parsing of _PRS to find IRQs
* Removed acpi_irq_domain_create and acpi_irq_domain_remove

Changes V3 -> V4:
* Add a DSDT device probe table that is used to probe DSDT IRQ chips as necessary when converting HW IRQs to Linux IRQs
* Describe IRQ combiner registers as ACPI Register resources

Changes V4 -> V5:
* Fix issues flagged by the 0-DAY build bot
* Fix some minor style issues raised by Timur

Agustin Vega-Frias (2):
ACPI: Add support for ResourceSource/IRQ domain mapping
irqchip: qcom: Add IRQ combiner driver

 drivers/acpi/Makefile               |   1 +
 drivers/acpi/irqdomain.c            | 141 +++++++++++++++
 drivers/acpi/resource.c             |  21 ++-
 drivers/irqchip/Kconfig             |   8 +
 drivers/irqchip/Makefile            |   1 +
 drivers/irqchip/qcom-irq-combiner.c | 332 ++++++++++++++++++++++++++++++++++++
 include/asm-generic/vmlinux.lds.h   |   1 +
 include/linux/acpi.h                |  71 ++++++++
 include/linux/irqchip.h             |  17 +-
 9 files changed, 581 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
 create mode 100644 drivers/irqchip/qcom-irq-combiner.c
Signature · 2 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.
#1 lore raw

PATCH v5 1/2 ACPI: Add support for ResourceSource/IRQ domain mapping

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---
 drivers/acpi/Makefile             |   1 +
 drivers/acpi/irqdomain.c          | 141 ++++++++++++++++++++++++++++++++++++++
 drivers/acpi/resource.c           |  21 +++---
 include/asm-generic/vmlinux.lds.h |   1 +
 include/linux/acpi.h              |  71 +++++++++++++++++++
 include/linux/irqchip.h           |  17 ++++-
 6 files changed, 240 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
drivers/acpi/Makefile +1 -0
index 9ed0878..880401b 100644
@@ -57,6 +57,7 @@ acpi-$(CONFIG_ACPI_PROCFS_POWER) += cm_sbs.o
 acpi-y				+= acpi_lpat.o
 acpi-$(CONFIG_ACPI_GENERIC_GSI) += gsi.o
 acpi-$(CONFIG_ACPI_WATCHDOG)	+= acpi_watchdog.o
+acpi-$(CONFIG_IRQ_DOMAIN)	+= irqdomain.o
 
 # These are (potentially) separate modules
 
drivers/acpi/irqdomain.c new file +141 -0

141 lines not on this page · raw message

drivers/acpi/resource.c +12 -9
index 56241eb..3fb7abf 100644
@@ -381,14 +381,15 @@ static void acpi_dev_irqresource_disabled(struct resource *res, u32 gsi)
 	res->flags = IORESOURCE_IRQ | IORESOURCE_DISABLED | IORESOURCE_UNSET;
 }
 
-static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
+static void acpi_dev_get_irqresource(struct resource *res, u32 hwirq,
+				     const struct acpi_resource_source *source,
 				     u8 triggering, u8 polarity, u8 shareable,
 				     bool legacy)
 {
 	int irq, p, t;
 
-	if (!valid_IRQ(gsi)) {
-		acpi_dev_irqresource_disabled(res, gsi);
+	if ((source->string_length == 0) && !valid_IRQ(hwirq)) {
+		acpi_dev_irqresource_disabled(res, hwirq);
 		return;
 	}
 
@@ -402,25 +403,25 @@ static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
 	 * using extended IRQ descriptors we take the IRQ configuration
 	 * from _CRS directly.
 	 */
-	if (legacy && !acpi_get_override_irq(gsi, &t, &p)) {
+	if (legacy && !acpi_get_override_irq(hwirq, &t, &p)) {
 		u8 trig = t ? ACPI_LEVEL_SENSITIVE : ACPI_EDGE_SENSITIVE;
 		u8 pol = p ? ACPI_ACTIVE_LOW : ACPI_ACTIVE_HIGH;
 
 		if (triggering != trig || polarity != pol) {
-			pr_warning("ACPI: IRQ %d override to %s, %s\n", gsi,
-				   t ? "level" : "edge", p ? "low" : "high");
+			pr_warn("ACPI: IRQ %d override to %s, %s\n", hwirq,
+				t ? "level" : "edge", p ? "low" : "high");
 			triggering = trig;
 			polarity = pol;
 		}
 	}
 
 	res->flags = acpi_dev_irq_flags(triggering, polarity, shareable);
-	irq = acpi_register_gsi(NULL, gsi, triggering, polarity);
+	irq = acpi_irq_domain_register_irq(source, hwirq, triggering, polarity);
 	if (irq >= 0) {
 		res->start = irq;
 		res->end = irq;
 	} else {
-		acpi_dev_irqresource_disabled(res, gsi);
+		acpi_dev_irqresource_disabled(res, hwirq);
 	}
 }
 
@@ -446,6 +447,7 @@ static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
 bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int index,
 				 struct resource *res)
 {
+	const struct acpi_resource_source dummy = { 0, 0, NULL };
 	struct acpi_resource_irq *irq;
 	struct acpi_resource_extended_irq *ext_irq;
 
@@ -460,7 +462,7 @@ bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int index,
 			acpi_dev_irqresource_disabled(res, 0);
 			return false;
 		}
-		acpi_dev_get_irqresource(res, irq->interrupts[index],
+		acpi_dev_get_irqresource(res, irq->interrupts[index], &dummy,
 					 irq->triggering, irq->polarity,
 					 irq->sharable, true);
 		break;
@@ -471,6 +473,7 @@ bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int index,
 			return false;
 		}
 		acpi_dev_get_irqresource(res, ext_irq->interrupts[index],
+					 &ext_irq->resource_source,
 					 ext_irq->triggering, ext_irq->polarity,
 					 ext_irq->sharable, false);
 		break;
include/asm-generic/vmlinux.lds.h +1 -0
index 3074796..f808afdc 100644
@@ -563,6 +563,7 @@
 	IRQCHIP_OF_MATCH_TABLE()					\
 	ACPI_PROBE_TABLE(irqchip)					\
 	ACPI_PROBE_TABLE(clksrc)					\
+	ACPI_PROBE_TABLE(dsdt)						\
 	EARLYCON_TABLE()
 
 #define INIT_TEXT							\
include/linux/acpi.h +71 -0

97 lines not on this page · raw message

include/linux/irqchip.h +14 -3
index 89c34b2..c2d0c12 100644
@@ -29,6 +29,10 @@
 /*
  * This macro must be used by the different irqchip drivers to declare
  * the association between their version and their initialization function.
+ * Two syntaxes are supported depending on the table where the irqchip device
+ * is declared:
+ *
+ * - MADT irqchip syntax, which requires the following five arguments:
  *
  * @name: name that must be unique accross all IRQCHIP_ACPI_DECLARE of the
  * same file.
@@ -37,10 +41,17 @@
  *            Can be NULL.
  * @data: data to be checked by the validate function.
  * @fn: initialization function
+ *
+ * - DSDT irqchip syntax, which requires the following three arguments:
+ *
+ * @name: name that must be unique across all IRQCHIP_ACPI_DECLARE of the
+ * same file.
+ * @hid: _HID of the DSDT device
+ * @fn: initialization function
  */
-#define IRQCHIP_ACPI_DECLARE(name, subtable, validate, data, fn)	\
-	ACPI_DECLARE_PROBE_ENTRY(irqchip, name, ACPI_SIG_MADT, 		\
-				 subtable, validate, data, fn)
+
+#define IRQCHIP_ACPI_DECLARE(...)					\
+	__IRQCHIP_ACPI_DECLARE(__VA_ARGS__, MADT, _unused, DSDT)(__VA_ARGS__)
 
 #ifdef CONFIG_IRQCHIP
 void irqchip_init(void);
Signature · 2 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.
#8 lore raw

PATCH v5 2/2 irqchip: qcom: Add IRQ combiner driver

In reply to Agustin Vega-Frias

Driver for interrupt combiners in the Top-level Control and Status Registers (TCSR) hardware block in Qualcomm Technologies chips.

An interrupt combiner in this block combines a set of interrupts by OR'ing the individual interrupt signals into a summary interrupt signal routed to a parent interrupt controller, and provides read- only, 32-bit registers to query the status of individual interrupts. The status bit for IRQ n is bit (n % 32) within register (n / 32) of the given combiner. Thus, each combiner can be described as a set of register offsets and the number of IRQs managed.

---
 drivers/irqchip/Kconfig             |   8 +
 drivers/irqchip/Makefile            |   1 +
 drivers/irqchip/qcom-irq-combiner.c | 332 ++++++++++++++++++++++++++++++++++++
 3 files changed, 341 insertions(+)
 create mode 100644 drivers/irqchip/qcom-irq-combiner.c
drivers/irqchip/Kconfig +8 -0
index 82b0b5d..8e0cbbb 100644
@@ -279,3 +279,11 @@ config EZNPS_GIC
 config STM32_EXTI
 	bool
 	select IRQ_DOMAIN
+
+config QCOM_IRQ_COMBINER
+	bool "QCOM IRQ combiner support"
+	depends on ARCH_QCOM && ACPI
+	select IRQ_DOMAIN
+	help
+	  Say yes here to add support for the IRQ combiner devices embedded
+	  in Qualcomm Technologies chips.
drivers/irqchip/Makefile +1 -0
index e4dbfc8..1818a0b 100644
@@ -74,3 +74,4 @@ obj-$(CONFIG_LS_SCFG_MSI)		+= irq-ls-scfg-msi.o
 obj-$(CONFIG_EZNPS_GIC)			+= irq-eznps.o
 obj-$(CONFIG_ARCH_ASPEED)		+= irq-aspeed-vic.o
 obj-$(CONFIG_STM32_EXTI) 		+= irq-stm32-exti.o
+obj-$(CONFIG_QCOM_IRQ_COMBINER)		+= qcom-irq-combiner.o
drivers/irqchip/qcom-irq-combiner.c new file +332 -0

332 lines not on this page · raw message

Signature · 2 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.
#2 lore raw

Re: [PATCH V5 1/2] ACPI: Add support for ResourceSource/IRQ domain mapping

In reply to Agustin Vega-Frias

Hi Agustin,

Quoted 49 lines · Agustin Vega-Frias

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---
 drivers/acpi/Makefile             |   1 +
 drivers/acpi/irqdomain.c          | 141 ++++++++++++++++++++++++++++++++++++++
 drivers/acpi/resource.c           |  21 +++---
 include/asm-generic/vmlinux.lds.h |   1 +
 include/linux/acpi.h              |  71 +++++++++++++++++++
 include/linux/irqchip.h           |  17 ++++-
 6 files changed, 240 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
drivers/acpi/Makefile +1 -0
index 9ed0878..880401b 100644
@@ -57,6 +57,7 @@ acpi-$(CONFIG_ACPI_PROCFS_POWER) += cm_sbs.o
 acpi-y				+= acpi_lpat.o
 acpi-$(CONFIG_ACPI_GENERIC_GSI) += gsi.o
 acpi-$(CONFIG_ACPI_WATCHDOG)	+= acpi_watchdog.o
+acpi-$(CONFIG_IRQ_DOMAIN)	+= irqdomain.o
 
 # These are (potentially) separate modules
 
drivers/acpi/irqdomain.c new file +15 -0
new file mode 100644
index 0000000..c53b9f4
@@ -0,0 +1,141 @@
+/*
+ * ACPI ResourceSource/IRQ domain mapping support
+ *
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+#include <linux/acpi.h>

drivers/acpi/irqdomain.c line 25

Quoted 10 lines
drivers/acpi/irqdomain.c new file +10 -0
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+
+/**
+ * acpi_irq_domain_ensure_probed() - Check if the device has registered
+ *                                   an IRQ domain and probe as necessary
+ *
+ * @device: Device to check and probe
+ *
+ * Returns: 0 on success, -ENODEV otherwise

This is not correct (ie it depends on what

struct acpi_dsdt_probe_entry.probe

returns) and I would like to take this nit as an opportunity to take a step back and ask you a question below.

drivers/acpi/irqdomain.c line 37

Quoted 12 lines
drivers/acpi/irqdomain.c new file +12 -0
+ */
+static int acpi_irq_domain_ensure_probed(struct acpi_device *device)
+{
+	struct acpi_dsdt_probe_entry *entry;
+
+	if (irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY) != 0)
+		return 0;
+
+	for (entry = &__dsdt_acpi_probe_table;
+	     entry < &__dsdt_acpi_probe_table_end; entry++)
+		if (strcmp(entry->_hid, acpi_device_hid(device)) == 0)
+			return entry->probe(device);

Through this approch we are forcing an irqchip (that by the way it has a physical node ACPI companion by being a DSDT device object so it could be managed by a platform driver) to be probed. The question is: is there a reason (apart from the current ACPI resource parsing API) why this can't be implemented through deferred probing and the device dependencies framework Rafael is working on:

http://www.mail-archive.com/[email protected]/msg1246897.html

The DT layer, through the of_irq_get() API, supports probe deferral and what I am asking you is if there is any blocking point (again, apart from the current ACPI API) to implement the same mechanism.

I have not reviewed the previous versions so I am certainly missing some of the bits and pieces already discussed, apologies for that.

Thanks,
Lorenzo

Quoted 345 lines
drivers/acpi/irqdomain.c new file +104 -0
+
+	return -ENODEV;
+}
+
+/**
+ * acpi_irq_domain_register_irq() - Register the mapping for an IRQ produced
+ *                                  by the given acpi_resource_source to a
+ *                                  Linux IRQ number
+ * @source: IRQ source
+ * @hwirq: Hardware IRQ number
+ * @trigger: trigger type of the IRQ number to be mapped
+ * @polarity: polarity of the IRQ to be mapped
+ *
+ * Returns: a valid linux IRQ number on success
+ *          -ENODEV if the given acpi_resource_source cannot be found
+ *          -EPROBE_DEFER if the IRQ domain has not been registered
+ *          -EINVAL for all other errors
+ */
+int acpi_irq_domain_register_irq(const struct acpi_resource_source *source,
+				 u32 hwirq, int trigger, int polarity)
+{
+	struct irq_fwspec fwspec;
+	struct acpi_device *device;
+	acpi_handle handle;
+	acpi_status status;
+	int ret;
+
+	/* An empty acpi_resource_source means it is a GSI */
+	if (!source->string_length)
+		return acpi_register_gsi(NULL, hwirq, trigger, polarity);
+
+	status = acpi_get_handle(NULL, source->string_ptr, &handle);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	device = acpi_bus_get_acpi_device(handle);
+	if (!device)
+		return -ENODEV;
+
+	ret = acpi_irq_domain_ensure_probed(device);
+	if (ret)
+		goto out_put_device;
+
+	fwspec.fwnode = &device->fwnode;
+	fwspec.param[0] = hwirq;
+	fwspec.param[1] = acpi_dev_get_irq_type(trigger, polarity);
+	fwspec.param_count = 2;
+
+	ret = irq_create_fwspec_mapping(&fwspec);
+
+out_put_device:
+	acpi_bus_put_acpi_device(device);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(acpi_irq_domain_register_irq);
+
+/**
+ * acpi_irq_domain_unregister_irq() - Delete the mapping for an IRQ produced
+ *                                    by the given acpi_resource_source to a
+ *                                    Linux IRQ number
+ * @source: IRQ source
+ * @hwirq: Hardware IRQ number
+ *
+ * Returns: 0 on success
+ *          -ENODEV if the given acpi_resource_source cannot be found
+ *          -EINVAL for all other errors
+ */
+int acpi_irq_domain_unregister_irq(const struct acpi_resource_source *source,
+				   u32 hwirq)
+{
+	struct irq_domain *domain;
+	struct acpi_device *device;
+	acpi_handle handle;
+	acpi_status status;
+	int ret = 0;
+
+	if (!source->string_length) {
+		acpi_unregister_gsi(hwirq);
+		return 0;
+	}
+
+	status = acpi_get_handle(NULL, source->string_ptr, &handle);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	device = acpi_bus_get_acpi_device(handle);
+	if (!device)
+		return -ENODEV;
+
+	domain = irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY);
+	if (!domain) {
+		ret = -EINVAL;
+		goto out_put_device;
+	}
+
+	irq_dispose_mapping(irq_find_mapping(domain, hwirq));
+
+out_put_device:
+	acpi_bus_put_acpi_device(device);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(acpi_irq_domain_unregister_irq);
drivers/acpi/resource.c +12 -9
index 56241eb..3fb7abf 100644
@@ -381,14 +381,15 @@ static void acpi_dev_irqresource_disabled(struct resource *res, u32 gsi)
 	res->flags = IORESOURCE_IRQ | IORESOURCE_DISABLED | IORESOURCE_UNSET;
 }
 
-static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
+static void acpi_dev_get_irqresource(struct resource *res, u32 hwirq,
+				     const struct acpi_resource_source *source,
 				     u8 triggering, u8 polarity, u8 shareable,
 				     bool legacy)
 {
 	int irq, p, t;
 
-	if (!valid_IRQ(gsi)) {
-		acpi_dev_irqresource_disabled(res, gsi);
+	if ((source->string_length == 0) && !valid_IRQ(hwirq)) {
+		acpi_dev_irqresource_disabled(res, hwirq);
 		return;
 	}
 
@@ -402,25 +403,25 @@ static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
 	 * using extended IRQ descriptors we take the IRQ configuration
 	 * from _CRS directly.
 	 */
-	if (legacy && !acpi_get_override_irq(gsi, &t, &p)) {
+	if (legacy && !acpi_get_override_irq(hwirq, &t, &p)) {
 		u8 trig = t ? ACPI_LEVEL_SENSITIVE : ACPI_EDGE_SENSITIVE;
 		u8 pol = p ? ACPI_ACTIVE_LOW : ACPI_ACTIVE_HIGH;
 
 		if (triggering != trig || polarity != pol) {
-			pr_warning("ACPI: IRQ %d override to %s, %s\n", gsi,
-				   t ? "level" : "edge", p ? "low" : "high");
+			pr_warn("ACPI: IRQ %d override to %s, %s\n", hwirq,
+				t ? "level" : "edge", p ? "low" : "high");
 			triggering = trig;
 			polarity = pol;
 		}
 	}
 
 	res->flags = acpi_dev_irq_flags(triggering, polarity, shareable);
-	irq = acpi_register_gsi(NULL, gsi, triggering, polarity);
+	irq = acpi_irq_domain_register_irq(source, hwirq, triggering, polarity);
 	if (irq >= 0) {
 		res->start = irq;
 		res->end = irq;
 	} else {
-		acpi_dev_irqresource_disabled(res, gsi);
+		acpi_dev_irqresource_disabled(res, hwirq);
 	}
 }
 
@@ -446,6 +447,7 @@ static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
 bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int index,
 				 struct resource *res)
 {
+	const struct acpi_resource_source dummy = { 0, 0, NULL };
 	struct acpi_resource_irq *irq;
 	struct acpi_resource_extended_irq *ext_irq;
 
@@ -460,7 +462,7 @@ bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int index,
 			acpi_dev_irqresource_disabled(res, 0);
 			return false;
 		}
-		acpi_dev_get_irqresource(res, irq->interrupts[index],
+		acpi_dev_get_irqresource(res, irq->interrupts[index], &dummy,
 					 irq->triggering, irq->polarity,
 					 irq->sharable, true);
 		break;
@@ -471,6 +473,7 @@ bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int index,
 			return false;
 		}
 		acpi_dev_get_irqresource(res, ext_irq->interrupts[index],
+					 &ext_irq->resource_source,
 					 ext_irq->triggering, ext_irq->polarity,
 					 ext_irq->sharable, false);
 		break;
include/asm-generic/vmlinux.lds.h +1 -0
index 3074796..f808afdc 100644
@@ -563,6 +563,7 @@
 	IRQCHIP_OF_MATCH_TABLE()					\
 	ACPI_PROBE_TABLE(irqchip)					\
 	ACPI_PROBE_TABLE(clksrc)					\
+	ACPI_PROBE_TABLE(dsdt)						\
 	EARLYCON_TABLE()
 
 #define INIT_TEXT							\
include/linux/acpi.h +71 -0
index ddbeda6..bb1a838 100644
@@ -26,6 +26,7 @@
 #include <linux/resource_ext.h>
 #include <linux/device.h>
 #include <linux/property.h>
+#include <linux/irqdomain.h>
 
 #ifndef _LINUX
 #define _LINUX
@@ -321,6 +322,31 @@ void acpi_set_irq_model(enum acpi_irq_model_id model,
  */
 void acpi_unregister_gsi (u32 gsi);
 
+#ifdef CONFIG_IRQ_DOMAIN
+
+int acpi_irq_domain_register_irq(const struct acpi_resource_source *source,
+				 u32 hwirq, int trigger, int polarity);
+int acpi_irq_domain_unregister_irq(const struct acpi_resource_source *source,
+				   u32 hwirq);
+
+#else
+
+static inline int acpi_irq_domain_register_irq(
+	const struct acpi_resource_source *source, u32 hwirq, int trigger,
+	int polarity)
+{
+	return acpi_register_gsi(NULL, hwirq, trigger, polarity);
+}
+
+static inline int acpi_irq_domain_unregister_irq(
+	const struct acpi_resource_source *source,  u32 hwirq)
+{
+	acpi_unregister_gsi(hwirq);
+	return 0;
+}
+
+#endif /* CONFIG_IRQ_DOMAIN */
+
 struct pci_dev;
 
 int acpi_pci_irq_enable (struct pci_dev *dev);
@@ -1024,6 +1050,34 @@ struct acpi_probe_entry {
 					  (&ACPI_PROBE_TABLE_END(t) -	\
 					   &ACPI_PROBE_TABLE(t)));	\
 	})
+
+/* Length of Hardware ID field in DSDT entries as per ACPI spec */
+#define ACPI_HID_LEN 9
+
+typedef int (*acpi_dsdt_handler)(struct acpi_device *);
+
+/**
+ * struct acpi_probe_dsdt_entry - boot-time probing entry for DSDT devices
+ * @hid:		_HID of the device
+ * @fn:		Callback to the driver being probed
+ * @driver_data:	Sideband data provided back to the driver
+ */
+struct acpi_dsdt_probe_entry {
+	__u8 _hid[ACPI_HID_LEN];
+	acpi_dsdt_handler probe;
+};
+
+#define ACPI_DECLARE_DSDT_PROBE_ENTRY(name, hid, fn)			\
+	static const struct acpi_dsdt_probe_entry __acpi_probe_##name	\
+		__used __section(__dsdt_acpi_probe_table) =		\
+		{							\
+			._hid = hid,					\
+			.probe = fn,					\
+		}
+
+extern struct acpi_dsdt_probe_entry __dsdt_acpi_probe_table;
+extern struct acpi_dsdt_probe_entry __dsdt_acpi_probe_table_end;
+
 #else
 static inline int acpi_dev_get_property(struct acpi_device *adev,
 					const char *name, acpi_object_type type,
@@ -1101,8 +1155,25 @@ static inline struct fwnode_handle *acpi_get_next_subnode(struct device *dev,
 		     (void *) data }
 
 #define acpi_probe_device_table(t)	({ int __r = 0; __r;})
+
+#define ACPI_DECLARE_DSDT_PROBE_ENTRY(name, hid, fn)			\
+	static const void *__acpi_probe_##name[]			\
+		__attribute__((unused))				\
+		 = { (void *) hid,					\
+		     (void *) fn }
+
 #endif
 
+#define MADT_IRQCHIP_ACPI_DECLARE(name, subtable, validate, data, fn)	\
+	ACPI_DECLARE_PROBE_ENTRY(irqchip, name, ACPI_SIG_MADT,		\
+				 subtable, validate, data, fn)
+
+#define DSDT_IRQCHIP_ACPI_DECLARE(name, hid, fn)	\
+	ACPI_DECLARE_DSDT_PROBE_ENTRY(name, hid, fn)
+
+#define __IRQCHIP_ACPI_DECLARE(_a1, _a2, _a3, _a4, _a5, type, ...)	\
+	type##_IRQCHIP_ACPI_DECLARE
+
 #ifdef CONFIG_ACPI_TABLE_UPGRADE
 void acpi_table_upgrade(void);
 #else
include/linux/irqchip.h +14 -3
index 89c34b2..c2d0c12 100644
@@ -29,6 +29,10 @@
 /*
  * This macro must be used by the different irqchip drivers to declare
  * the association between their version and their initialization function.
+ * Two syntaxes are supported depending on the table where the irqchip device
+ * is declared:
+ *
+ * - MADT irqchip syntax, which requires the following five arguments:
  *
  * @name: name that must be unique accross all IRQCHIP_ACPI_DECLARE of the
  * same file.
@@ -37,10 +41,17 @@
  *            Can be NULL.
  * @data: data to be checked by the validate function.
  * @fn: initialization function
+ *
+ * - DSDT irqchip syntax, which requires the following three arguments:
+ *
+ * @name: name that must be unique across all IRQCHIP_ACPI_DECLARE of the
+ * same file.
+ * @hid: _HID of the DSDT device
+ * @fn: initialization function
  */
-#define IRQCHIP_ACPI_DECLARE(name, subtable, validate, data, fn)	\
-	ACPI_DECLARE_PROBE_ENTRY(irqchip, name, ACPI_SIG_MADT, 		\
-				 subtable, validate, data, fn)
+
+#define IRQCHIP_ACPI_DECLARE(...)					\
+	__IRQCHIP_ACPI_DECLARE(__VA_ARGS__, MADT, _unused, DSDT)(__VA_ARGS__)
 
 #ifdef CONFIG_IRQCHIP
 void irqchip_init(void);
Signature · 7 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
#3 lore raw

Re: [PATCH V5 1/2] ACPI: Add support for ResourceSource/IRQ domain mapping

Quoted 90 lines · Lorenzo Pieralisi

Hi Agustin,

Quoted 49 lines · Agustin Vega-Frias

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---
 drivers/acpi/Makefile             |   1 +
 drivers/acpi/irqdomain.c          | 141 ++++++++++++++++++++++++++++++++++++++
 drivers/acpi/resource.c           |  21 +++---
 include/asm-generic/vmlinux.lds.h |   1 +
 include/linux/acpi.h              |  71 +++++++++++++++++++
 include/linux/irqchip.h           |  17 ++++-
 6 files changed, 240 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
drivers/acpi/Makefile +1 -0
index 9ed0878..880401b 100644
@@ -57,6 +57,7 @@ acpi-$(CONFIG_ACPI_PROCFS_POWER) += cm_sbs.o
 acpi-y				+= acpi_lpat.o
 acpi-$(CONFIG_ACPI_GENERIC_GSI) += gsi.o
 acpi-$(CONFIG_ACPI_WATCHDOG)	+= acpi_watchdog.o
+acpi-$(CONFIG_IRQ_DOMAIN)	+= irqdomain.o
 
 # These are (potentially) separate modules
 
drivers/acpi/irqdomain.c new file +15 -0
new file mode 100644
index 0000000..c53b9f4
@@ -0,0 +1,141 @@
+/*
+ * ACPI ResourceSource/IRQ domain mapping support
+ *
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+#include <linux/acpi.h>

drivers/acpi/irqdomain.c line 25

Quoted 10 lines
drivers/acpi/irqdomain.c new file +10 -0
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+
+/**
+ * acpi_irq_domain_ensure_probed() - Check if the device has registered
+ *                                   an IRQ domain and probe as necessary
+ *
+ * @device: Device to check and probe
+ *
+ * Returns: 0 on success, -ENODEV otherwise

This is not correct (ie it depends on what

struct acpi_dsdt_probe_entry.probe

returns) and I would like to take this nit as an opportunity to take a step back and ask you a question below.

drivers/acpi/irqdomain.c line 37

Quoted 12 lines
drivers/acpi/irqdomain.c new file +12 -0
+ */
+static int acpi_irq_domain_ensure_probed(struct acpi_device *device)
+{
+	struct acpi_dsdt_probe_entry *entry;
+
+	if (irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY) != 0)
+		return 0;
+
+	for (entry = &__dsdt_acpi_probe_table;
+	     entry < &__dsdt_acpi_probe_table_end; entry++)
+		if (strcmp(entry->_hid, acpi_device_hid(device)) == 0)
+			return entry->probe(device);

Through this approch we are forcing an irqchip (that by the way it has a physical node ACPI companion by being a DSDT device object so it could be managed by a platform driver) to be probed. The question is: is there a reason (apart from the current ACPI resource parsing API) why this can't be implemented through deferred probing and the device dependencies framework Rafael is working on:

Quoted 8 lines

http://www.mail-archive.com/[email protected]/msg1246897.html

The DT layer, through the of_irq_get() API, supports probe deferral and what I am asking you is if there is any blocking point (again, apart from the current ACPI API) to implement the same mechanism.

I have not reviewed the previous versions so I am certainly missing some of the bits and pieces already discussed, apologies for that.

Also, this function scares me to no end: lack of locking and recursion are the main things that worry me. My vote would be to implement something based on Rafael's approach (which conveniently solves all kind of other issues).

I'll review this patch series in a more in-depth way soon, but I wanted to chime in and add my own weight to Lorenzo's proposal.

Thanks,

	M.
Signature · 1 line
Jazz is not dead. It just smells funny...
#6 lore raw

Re: [PATCH V5 1/2] ACPI: Add support for ResourceSource/IRQ domain mapping

In reply to Lorenzo Pieralisi

Hey Lorenzo,

Quoted 69 lines · Lorenzo Pieralisi

Hi Agustin,

Quoted 56 lines · Agustin Vega-Frias

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---
 drivers/acpi/Makefile             |   1 +
 drivers/acpi/irqdomain.c          | 141 
++++++++++++++++++++++++++++++++++++++
 drivers/acpi/resource.c           |  21 +++---
 include/asm-generic/vmlinux.lds.h |   1 +
 include/linux/acpi.h              |  71 +++++++++++++++++++
 include/linux/irqchip.h           |  17 ++++-
 6 files changed, 240 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
drivers/acpi/Makefile +1 -0
index 9ed0878..880401b 100644
@@ -57,6 +57,7 @@ acpi-$(CONFIG_ACPI_PROCFS_POWER) += cm_sbs.o
 acpi-y				+= acpi_lpat.o
 acpi-$(CONFIG_ACPI_GENERIC_GSI) += gsi.o
 acpi-$(CONFIG_ACPI_WATCHDOG)	+= acpi_watchdog.o
+acpi-$(CONFIG_IRQ_DOMAIN)	+= irqdomain.o
 
 # These are (potentially) separate modules
 
drivers/acpi/irqdomain.c new file +6 -0
new file mode 100644
index 0000000..c53b9f4
@@ -0,0 +1,141 @@
+/*
+ * ACPI ResourceSource/IRQ domain mapping support
+ *
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or 

modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */

drivers/acpi/irqdomain.c new file +6 -0
+#include <linux/acpi.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+
+/**
+ * acpi_irq_domain_ensure_probed() - Check if the device has 
Quoted 7 lines

registered
+ * an IRQ domain and probe as
necessary
+ *
+ * @device: Device to check and probe
+ *
+ * Returns: 0 on success, -ENODEV otherwise

This is not correct (ie it depends on what

Quoted 5 lines

struct acpi_dsdt_probe_entry.probe

returns) and I would like to take this nit as an opportunity to take a step back and ask you a question below.

Will fix this as necessary if this approach is pursued.

Quoted 21 lines

drivers/acpi/irqdomain.c line 37

Quoted 12 lines

+ */

drivers/acpi/irqdomain.c new file +11 -0
+static int acpi_irq_domain_ensure_probed(struct acpi_device *device)
+{
+	struct acpi_dsdt_probe_entry *entry;
+
+	if (irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY) != 0)
+		return 0;
+
+	for (entry = &__dsdt_acpi_probe_table;
+	     entry < &__dsdt_acpi_probe_table_end; entry++)
+		if (strcmp(entry->_hid, acpi_device_hid(device)) == 0)
+			return entry->probe(device);

Through this approch we are forcing an irqchip (that by the way it has a physical node ACPI companion by being a DSDT device object so it could be managed by a platform driver) to be probed. The question is:
is
there a reason (apart from the current ACPI resource parsing API) why this can't be implemented through deferred probing and the device dependencies framework Rafael is working on:

Quoted 9 lines

http://www.mail-archive.com/linux-kernel at vger.kernel.org/msg1246897.html

The DT layer, through the of_irq_get() API, supports probe deferral and what I am asking you is if there is any blocking point (again, apart from the current ACPI API) to implement the same mechanism.

I have not reviewed the previous versions so I am certainly missing some of the bits and pieces already discussed, apologies for that.

I need to review the latest version of Rafael's patchset to see if the issues
around ACPI IRQ resource conversion are addressed.

The problem we found before was that probe deferral didn't work because the ACPI core converts all ACPI IRQ resources to Linux resources for devices
at the same init level at the same time and that causes the conversion to fail unless the supplier driver is promoted to an earlier init level.

Thanks,
Agustin

Quoted 380 lines

Thanks,
Lorenzo

Quoted 377 lines
drivers/acpi/irqdomain.c new file +6 -0
+
+	return -ENODEV;
+}
+
+/**
+ * acpi_irq_domain_register_irq() - Register the mapping for an IRQ 

produced
+ * by the given acpi_resource_source
to a
+ * Linux IRQ number
+ * @source: IRQ source
+ * @hwirq: Hardware IRQ number
+ * @trigger: trigger type of the IRQ number to be mapped
+ * @polarity: polarity of the IRQ to be mapped
+ *
+ * Returns: a valid linux IRQ number on success

+ *          -ENODEV if the given acpi_resource_source cannot be found
+ *          -EPROBE_DEFER if the IRQ domain has not been registered
+ *          -EINVAL for all other errors

+ */
+int acpi_irq_domain_register_irq(const struct acpi_resource_source *source,

drivers/acpi/irqdomain.c new file +40 -0
+				 u32 hwirq, int trigger, int polarity)
+{
+	struct irq_fwspec fwspec;
+	struct acpi_device *device;
+	acpi_handle handle;
+	acpi_status status;
+	int ret;
+
+	/* An empty acpi_resource_source means it is a GSI */
+	if (!source->string_length)
+		return acpi_register_gsi(NULL, hwirq, trigger, polarity);
+
+	status = acpi_get_handle(NULL, source->string_ptr, &handle);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	device = acpi_bus_get_acpi_device(handle);
+	if (!device)
+		return -ENODEV;
+
+	ret = acpi_irq_domain_ensure_probed(device);
+	if (ret)
+		goto out_put_device;
+
+	fwspec.fwnode = &device->fwnode;
+	fwspec.param[0] = hwirq;
+	fwspec.param[1] = acpi_dev_get_irq_type(trigger, polarity);
+	fwspec.param_count = 2;
+
+	ret = irq_create_fwspec_mapping(&fwspec);
+
+out_put_device:
+	acpi_bus_put_acpi_device(device);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(acpi_irq_domain_register_irq);
+
+/**
+ * acpi_irq_domain_unregister_irq() - Delete the mapping for an IRQ 

produced
+ * by the given
acpi_resource_source to a
+ * Linux IRQ number
+ * @source: IRQ source
+ * @hwirq: Hardware IRQ number
+ *
+ * Returns: 0 on success
+ * -ENODEV if the given acpi_resource_source cannot be found
+ * -EINVAL for all other errors
+ */
+int acpi_irq_domain_unregister_irq(const struct acpi_resource_source *source,

drivers/acpi/irqdomain.c new file +35 -0
+				   u32 hwirq)
+{
+	struct irq_domain *domain;
+	struct acpi_device *device;
+	acpi_handle handle;
+	acpi_status status;
+	int ret = 0;
+
+	if (!source->string_length) {
+		acpi_unregister_gsi(hwirq);
+		return 0;
+	}
+
+	status = acpi_get_handle(NULL, source->string_ptr, &handle);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	device = acpi_bus_get_acpi_device(handle);
+	if (!device)
+		return -ENODEV;
+
+	domain = irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY);
+	if (!domain) {
+		ret = -EINVAL;
+		goto out_put_device;
+	}
+
+	irq_dispose_mapping(irq_find_mapping(domain, hwirq));
+
+out_put_device:
+	acpi_bus_put_acpi_device(device);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(acpi_irq_domain_unregister_irq);
drivers/acpi/resource.c +4 -3
index 56241eb..3fb7abf 100644
@@ -381,14 +381,15 @@ static void acpi_dev_irqresource_disabled(struct 
 resource *res, u32 gsi)
 	res->flags = IORESOURCE_IRQ | IORESOURCE_DISABLED | 
 IORESOURCE_UNSET;
 }
 
-static void acpi_dev_get_irqresource(struct resource *res, u32 gsi,
+static void acpi_dev_get_irqresource(struct resource *res, u32 hwirq,
+				     const struct acpi_resource_source *source,
 				     u8 triggering, u8 polarity, u8 shareable,
 				     bool legacy)
 {
 	int irq, p, t;
 
-	if (!valid_IRQ(gsi)) {
-		acpi_dev_irqresource_disabled(res, gsi);
+	if ((source->string_length == 0) && !valid_IRQ(hwirq)) {
+		acpi_dev_irqresource_disabled(res, hwirq);
 		return;
drivers/acpi/resource.c +5 -5
 	}
@@ -402,25 +403,25 @@ static void acpi_dev_get_irqresource(struct 
 resource *res, u32 gsi,
 	 * using extended IRQ descriptors we take the IRQ configuration
 	 * from _CRS directly.
 	 */
-	if (legacy && !acpi_get_override_irq(gsi, &t, &p)) {
+	if (legacy && !acpi_get_override_irq(hwirq, &t, &p)) {
 		u8 trig = t ? ACPI_LEVEL_SENSITIVE : ACPI_EDGE_SENSITIVE;
 		u8 pol = p ? ACPI_ACTIVE_LOW : ACPI_ACTIVE_HIGH;
 
 		if (triggering != trig || polarity != pol) {
-			pr_warning("ACPI: IRQ %d override to %s, %s\n", gsi,
-				   t ? "level" : "edge", p ? "low" : "high");
+			pr_warn("ACPI: IRQ %d override to %s, %s\n", hwirq,
+				t ? "level" : "edge", p ? "low" : "high");
 			triggering = trig;
 			polarity = pol;
 		}
 	}
 
 	res->flags = acpi_dev_irq_flags(triggering, polarity, shareable);
-	irq = acpi_register_gsi(NULL, gsi, triggering, polarity);
+	irq = acpi_irq_domain_register_irq(source, hwirq, triggering, 
 polarity);
 	if (irq >= 0) {
 		res->start = irq;
 		res->end = irq;
 	} else {
-		acpi_dev_irqresource_disabled(res, gsi);
+		acpi_dev_irqresource_disabled(res, hwirq);
 	}

}

drivers/acpi/resource.c +1 -0
@@ -446,6 +447,7 @@ static void acpi_dev_get_irqresource(struct 
 resource *res, u32 gsi,
 bool acpi_dev_resource_interrupt(struct acpi_resource *ares, int 
 index,
 				 struct resource *res)
 {
+	const struct acpi_resource_source dummy = { 0, 0, NULL };
 	struct acpi_resource_irq *irq;
drivers/acpi/resource.c +1 -1
 	struct acpi_resource_extended_irq *ext_irq;
@@ -460,7 +462,7 @@ bool acpi_dev_resource_interrupt(struct 
 acpi_resource *ares, int index,
 			acpi_dev_irqresource_disabled(res, 0);
 			return false;
 		}
-		acpi_dev_get_irqresource(res, irq->interrupts[index],
+		acpi_dev_get_irqresource(res, irq->interrupts[index], &dummy,
 					 irq->triggering, irq->polarity,
 					 irq->sharable, true);
drivers/acpi/resource.c +1 -0
 		break;
@@ -471,6 +473,7 @@ bool acpi_dev_resource_interrupt(struct 
 acpi_resource *ares, int index,
 			return false;
 		}
 		acpi_dev_get_irqresource(res, ext_irq->interrupts[index],
+					 &ext_irq->resource_source,
 					 ext_irq->triggering, ext_irq->polarity,
 					 ext_irq->sharable, false);
drivers/acpi/resource.c +0 -0
 		break;
include/asm-generic/vmlinux.lds.h +1 -0
index 3074796..f808afdc 100644
@@ -563,6 +563,7 @@
 	IRQCHIP_OF_MATCH_TABLE()					\
 	ACPI_PROBE_TABLE(irqchip)					\
 	ACPI_PROBE_TABLE(clksrc)					\
+	ACPI_PROBE_TABLE(dsdt)						\
 	EARLYCON_TABLE()
 
 #define INIT_TEXT							\
include/linux/acpi.h +26 -0
index ddbeda6..bb1a838 100644
@@ -26,6 +26,7 @@
 #include <linux/resource_ext.h>
 #include <linux/device.h>
 #include <linux/property.h>
+#include <linux/irqdomain.h>
 
 #ifndef _LINUX
 #define _LINUX
@@ -321,6 +322,31 @@ void acpi_set_irq_model(enum acpi_irq_model_id 
 model,
  */
 void acpi_unregister_gsi (u32 gsi);
 
+#ifdef CONFIG_IRQ_DOMAIN
+
+int acpi_irq_domain_register_irq(const struct acpi_resource_source 
 *source,
+				 u32 hwirq, int trigger, int polarity);
+int acpi_irq_domain_unregister_irq(const struct acpi_resource_source 
 *source,
+				   u32 hwirq);
+
+#else
+
+static inline int acpi_irq_domain_register_irq(
+	const struct acpi_resource_source *source, u32 hwirq, int trigger,
+	int polarity)
+{
+	return acpi_register_gsi(NULL, hwirq, trigger, polarity);
+}
+
+static inline int acpi_irq_domain_unregister_irq(
+	const struct acpi_resource_source *source,  u32 hwirq)
+{
+	acpi_unregister_gsi(hwirq);
+	return 0;
+}
+
+#endif /* CONFIG_IRQ_DOMAIN */
+

struct pci_dev;

int acpi_pci_irq_enable (struct pci_dev *dev);

include/linux/acpi.h +8 -0
@@ -1024,6 +1050,34 @@ struct acpi_probe_entry {
 					  (&ACPI_PROBE_TABLE_END(t) -	\
 					   &ACPI_PROBE_TABLE(t)));	\
 	})
+
+/* Length of Hardware ID field in DSDT entries as per ACPI spec */
+#define ACPI_HID_LEN 9
+
+typedef int (*acpi_dsdt_handler)(struct acpi_device *);
+
+/**
+ * struct acpi_probe_dsdt_entry - boot-time probing entry for DSDT 

devices
+ * @hid: _HID of the device
+ * @fn: Callback to the driver being probed
+ * @driver_data: Sideband data provided back to the driver
+ */

include/linux/acpi.h +33 -0
+struct acpi_dsdt_probe_entry {
+	__u8 _hid[ACPI_HID_LEN];
+	acpi_dsdt_handler probe;
+};
+
+#define ACPI_DECLARE_DSDT_PROBE_ENTRY(name, hid, fn)			\
+	static const struct acpi_dsdt_probe_entry __acpi_probe_##name	\
+		__used __section(__dsdt_acpi_probe_table) =		\
+		{							\
+			._hid = hid,					\
+			.probe = fn,					\
+		}
+
+extern struct acpi_dsdt_probe_entry __dsdt_acpi_probe_table;
+extern struct acpi_dsdt_probe_entry __dsdt_acpi_probe_table_end;
+
 #else
 static inline int acpi_dev_get_property(struct acpi_device *adev,
 					const char *name, acpi_object_type type,
@@ -1101,8 +1155,25 @@ static inline struct fwnode_handle 
 *acpi_get_next_subnode(struct device *dev,
 		     (void *) data }
 
 #define acpi_probe_device_table(t)	({ int __r = 0; __r;})
+
+#define ACPI_DECLARE_DSDT_PROBE_ENTRY(name, hid, fn)			\
+	static const void *__acpi_probe_##name[]			\
+		__attribute__((unused))				\
+		 = { (void *) hid,					\
+		     (void *) fn }
+
 #endif
 
+#define MADT_IRQCHIP_ACPI_DECLARE(name, subtable, validate, data, 
 fn)	\
+	ACPI_DECLARE_PROBE_ENTRY(irqchip, name, ACPI_SIG_MADT,		\
+				 subtable, validate, data, fn)
+
+#define DSDT_IRQCHIP_ACPI_DECLARE(name, hid, fn)	\
+	ACPI_DECLARE_DSDT_PROBE_ENTRY(name, hid, fn)
+
+#define __IRQCHIP_ACPI_DECLARE(_a1, _a2, _a3, _a4, _a5, type, ...)	\
+	type##_IRQCHIP_ACPI_DECLARE
+
 #ifdef CONFIG_ACPI_TABLE_UPGRADE

void acpi_table_upgrade(void);
#else

include/linux/irqchip.h +0 -0
index 89c34b2..c2d0c12 100644
@@ -29,6 +29,10 @@
 /*
  * This macro must be used by the different irqchip drivers to 

declare
* the association between their version and their initialization
function.
+ * Two syntaxes are supported depending on the table where the irqchip device
+ * is declared:
+ *
+ * - MADT irqchip syntax, which requires the following five arguments:
*
* @name: name that must be unique accross all IRQCHIP_ACPI_DECLARE
of the
* same file.

include/linux/irqchip.h +2 -0
@@ -37,10 +41,17 @@
  *            Can be NULL.
  * @data: data to be checked by the validate function.
  * @fn: initialization function
+ *
+ * - DSDT irqchip syntax, which requires the following three 

arguments:
+ *
+ * @name: name that must be unique across all IRQCHIP_ACPI_DECLARE of the
+ * same file.
+ * @hid: _HID of the DSDT device
+ * @fn: initialization function
*/

include/linux/irqchip.h +3 -3
-#define IRQCHIP_ACPI_DECLARE(name, subtable, validate, data, fn)	\
-	ACPI_DECLARE_PROBE_ENTRY(irqchip, name, ACPI_SIG_MADT, 		\
-				 subtable, validate, data, fn)
+
+#define IRQCHIP_ACPI_DECLARE(...)					\
+	__IRQCHIP_ACPI_DECLARE(__VA_ARGS__, MADT, _unused, 

DSDT)(__VA_ARGS__)

#ifdef CONFIG_IRQCHIP
void irqchip_init(void);
--
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux Foundation Collaborative Project.

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 4 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm 
Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a 
Linux Foundation Collaborative Project.
#4 lore raw

Re: [PATCH V5 1/2] ACPI: Add support for ResourceSource/IRQ domain mapping

In reply to Marc Zyngier

Hey Marc,

Quoted 103 lines · Marc Zyngier
Quoted 102 lines · Lorenzo Pieralisi

Hi Agustin,

Quoted 57 lines · Agustin Vega-Frias

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---
 drivers/acpi/Makefile             |   1 +
 drivers/acpi/irqdomain.c          | 141 
++++++++++++++++++++++++++++++++++++++
 drivers/acpi/resource.c           |  21 +++---
 include/asm-generic/vmlinux.lds.h |   1 +
 include/linux/acpi.h              |  71 +++++++++++++++++++
 include/linux/irqchip.h           |  17 ++++-
 6 files changed, 240 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
drivers/acpi/Makefile +1 -0
index 9ed0878..880401b 100644
@@ -57,6 +57,7 @@ acpi-$(CONFIG_ACPI_PROCFS_POWER) += cm_sbs.o
 acpi-y				+= acpi_lpat.o
 acpi-$(CONFIG_ACPI_GENERIC_GSI) += gsi.o
 acpi-$(CONFIG_ACPI_WATCHDOG)	+= acpi_watchdog.o
+acpi-$(CONFIG_IRQ_DOMAIN)	+= irqdomain.o
 
 # These are (potentially) separate modules
 
drivers/acpi/irqdomain.c new file +6 -0
new file mode 100644
index 0000000..c53b9f4
@@ -0,0 +1,141 @@
+/*
+ * ACPI ResourceSource/IRQ domain mapping support
+ *
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or 

modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */

drivers/acpi/irqdomain.c new file +6 -0
+#include <linux/acpi.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+
+/**
+ * acpi_irq_domain_ensure_probed() - Check if the device has 
Quoted 7 lines

registered
+ * an IRQ domain and probe as
necessary
+ *
+ * @device: Device to check and probe
+ *
+ * Returns: 0 on success, -ENODEV otherwise

This is not correct (ie it depends on what

struct acpi_dsdt_probe_entry.probe

returns) and I would like to take this nit as an opportunity to take a step back and ask you a question below.

drivers/acpi/irqdomain.c line 37

Quoted 12 lines

+ */

drivers/acpi/irqdomain.c new file +11 -0
+static int acpi_irq_domain_ensure_probed(struct acpi_device *device)
+{
+	struct acpi_dsdt_probe_entry *entry;
+
+	if (irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY) != 0)
+		return 0;
+
+	for (entry = &__dsdt_acpi_probe_table;
+	     entry < &__dsdt_acpi_probe_table_end; entry++)
+		if (strcmp(entry->_hid, acpi_device_hid(device)) == 0)
+			return entry->probe(device);

Through this approch we are forcing an irqchip (that by the way it has a physical node ACPI companion by being a DSDT device object so it could be managed by a platform driver) to be probed. The question is:
is
there a reason (apart from the current ACPI resource parsing API) why this can't be implemented through deferred probing and the device dependencies framework Rafael is working on:

http://www.mail-archive.com/linux-kernel at vger.kernel.org/msg1246897.html

The DT layer, through the of_irq_get() API, supports probe deferral and what I am asking you is if there is any blocking point (again, apart from the current ACPI API) to implement the same mechanism.

Quoted 8 lines
Quoted 2 lines

I have not reviewed the previous versions so I am certainly missing some of the bits and pieces already discussed, apologies for that.

Also, this function scares me to no end: lack of locking and recursion are the main things that worry me. My vote would be to implement something based on Rafael's approach (which conveniently solves all kind
of other issues).

I'll review Rafael's latest patchset and comment back on whether it suits
our needs or if we can build on that.

Thanks

Quoted 9 lines

I'll review this patch series in a more in-depth way soon, but I wanted to chime in and add my own weight to Lorenzo's proposal.

Thanks,

	M.

--
Jazz is not dead. It just smells funny...

Signature · 4 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm 
Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a 
Linux Foundation Collaborative Project.
#7 lore raw

Re: [PATCH V5 1/2] ACPI: Add support for ResourceSource/IRQ domain mapping

In reply to Agustin Vega-Frias
Quoted 7 lines · Agustin Vega-Frias

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---

[...]

Quoted 90 lines
drivers/acpi/irqdomain.c new file +90 -0
+/**
+ * acpi_irq_domain_register_irq() - Register the mapping for an IRQ produced
+ *                                  by the given acpi_resource_source to a
+ *                                  Linux IRQ number
+ * @source: IRQ source
+ * @hwirq: Hardware IRQ number
+ * @trigger: trigger type of the IRQ number to be mapped
+ * @polarity: polarity of the IRQ to be mapped
+ *
+ * Returns: a valid linux IRQ number on success
+ *          -ENODEV if the given acpi_resource_source cannot be found
+ *          -EPROBE_DEFER if the IRQ domain has not been registered
+ *          -EINVAL for all other errors
+ */
+int acpi_irq_domain_register_irq(const struct acpi_resource_source *source,
+				 u32 hwirq, int trigger, int polarity)
+{
+	struct irq_fwspec fwspec;
+	struct acpi_device *device;
+	acpi_handle handle;
+	acpi_status status;
+	int ret;
+
+	/* An empty acpi_resource_source means it is a GSI */
+	if (!source->string_length)
+		return acpi_register_gsi(NULL, hwirq, trigger, polarity);
+
+	status = acpi_get_handle(NULL, source->string_ptr, &handle);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	device = acpi_bus_get_acpi_device(handle);
+	if (!device)
+		return -ENODEV;
+
+	ret = acpi_irq_domain_ensure_probed(device);
+	if (ret)
+		goto out_put_device;
+
+	fwspec.fwnode = &device->fwnode;
+	fwspec.param[0] = hwirq;
+	fwspec.param[1] = acpi_dev_get_irq_type(trigger, polarity);
+	fwspec.param_count = 2;
+
+	ret = irq_create_fwspec_mapping(&fwspec);
+
+out_put_device:
+	acpi_bus_put_acpi_device(device);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(acpi_irq_domain_register_irq);
+
+/**
+ * acpi_irq_domain_unregister_irq() - Delete the mapping for an IRQ produced
+ *                                    by the given acpi_resource_source to a
+ *                                    Linux IRQ number
+ * @source: IRQ source
+ * @hwirq: Hardware IRQ number
+ *
+ * Returns: 0 on success
+ *          -ENODEV if the given acpi_resource_source cannot be found
+ *          -EINVAL for all other errors
+ */
+int acpi_irq_domain_unregister_irq(const struct acpi_resource_source *source,
+				   u32 hwirq)
+{
+	struct irq_domain *domain;
+	struct acpi_device *device;
+	acpi_handle handle;
+	acpi_status status;
+	int ret = 0;
+
+	if (!source->string_length) {
+		acpi_unregister_gsi(hwirq);
+		return 0;
+	}
+
+	status = acpi_get_handle(NULL, source->string_ptr, &handle);
+	if (ACPI_FAILURE(status))
+		return -ENODEV;
+
+	device = acpi_bus_get_acpi_device(handle);
+	if (!device)
+		return -ENODEV;
+
+	domain = irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY);
+	if (!domain) {
+		ret = -EINVAL;
+		goto out_put_device;

drivers/acpi/irqdomain.c line 141

Quoted 10 lines
drivers/acpi/irqdomain.c new file +10 -0
+	}
+
+	irq_dispose_mapping(irq_find_mapping(domain, hwirq));
+
+out_put_device:
+	acpi_bus_put_acpi_device(device);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(acpi_irq_domain_unregister_irq);

I think Marc already raised this before, which irqdomain.c is similar to gsi.c in drivers/acpi/, I prepared another approach [1] which avoid this, I'm open for comments.

I think it's pretty important to finalize the directions we are going, then avoid the duplicate work and speedup for upstream, Marc, Rafael, Lorenzo, could you give us suggestions for this?

Thanks
Hanjun

[1]: https://patchwork.kernel.org/patch/9331771/

#5 lore raw

Re: [PATCH V5 1/2] ACPI: Add support for ResourceSource/IRQ domain mapping

In reply to Marc Zyngier

Hi Marc, Lorenzo,

Quoted 103 lines · Marc Zyngier
Quoted 102 lines · Lorenzo Pieralisi

Hi Agustin,

Quoted 57 lines · Agustin Vega-Frias

This allows irqchip drivers to associate an ACPI DSDT device to an IRQ domain and provides support for using the ResourceSource in Extended IRQ Resources to find the domain and map the IRQs specified on that domain.

---
 drivers/acpi/Makefile             |   1 +
 drivers/acpi/irqdomain.c          | 141 
++++++++++++++++++++++++++++++++++++++
 drivers/acpi/resource.c           |  21 +++---
 include/asm-generic/vmlinux.lds.h |   1 +
 include/linux/acpi.h              |  71 +++++++++++++++++++
 include/linux/irqchip.h           |  17 ++++-
 6 files changed, 240 insertions(+), 12 deletions(-)
 create mode 100644 drivers/acpi/irqdomain.c
drivers/acpi/Makefile +1 -0
index 9ed0878..880401b 100644
@@ -57,6 +57,7 @@ acpi-$(CONFIG_ACPI_PROCFS_POWER) += cm_sbs.o
 acpi-y				+= acpi_lpat.o
 acpi-$(CONFIG_ACPI_GENERIC_GSI) += gsi.o
 acpi-$(CONFIG_ACPI_WATCHDOG)	+= acpi_watchdog.o
+acpi-$(CONFIG_IRQ_DOMAIN)	+= irqdomain.o
 
 # These are (potentially) separate modules
 
drivers/acpi/irqdomain.c new file +6 -0
new file mode 100644
index 0000000..c53b9f4
@@ -0,0 +1,141 @@
+/*
+ * ACPI ResourceSource/IRQ domain mapping support
+ *
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or 

modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */

drivers/acpi/irqdomain.c new file +6 -0
+#include <linux/acpi.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+
+/**
+ * acpi_irq_domain_ensure_probed() - Check if the device has 
Quoted 7 lines

registered
+ * an IRQ domain and probe as
necessary
+ *
+ * @device: Device to check and probe
+ *
+ * Returns: 0 on success, -ENODEV otherwise

This is not correct (ie it depends on what

struct acpi_dsdt_probe_entry.probe

returns) and I would like to take this nit as an opportunity to take a step back and ask you a question below.

drivers/acpi/irqdomain.c line 37

Quoted 12 lines

+ */

drivers/acpi/irqdomain.c new file +11 -0
+static int acpi_irq_domain_ensure_probed(struct acpi_device *device)
+{
+	struct acpi_dsdt_probe_entry *entry;
+
+	if (irq_find_matching_fwnode(&device->fwnode, DOMAIN_BUS_ANY) != 0)
+		return 0;
+
+	for (entry = &__dsdt_acpi_probe_table;
+	     entry < &__dsdt_acpi_probe_table_end; entry++)
+		if (strcmp(entry->_hid, acpi_device_hid(device)) == 0)
+			return entry->probe(device);

Through this approch we are forcing an irqchip (that by the way it has a physical node ACPI companion by being a DSDT device object so it could be managed by a platform driver) to be probed. The question is:
is
there a reason (apart from the current ACPI resource parsing API) why this can't be implemented through deferred probing and the device dependencies framework Rafael is working on:

http://www.mail-archive.com/linux-kernel at vger.kernel.org/msg1246897.html

The DT layer, through the of_irq_get() API, supports probe deferral and what I am asking you is if there is any blocking point (again, apart from the current ACPI API) to implement the same mechanism.

Quoted 9 lines
Quoted 2 lines

I have not reviewed the previous versions so I am certainly missing some of the bits and pieces already discussed, apologies for that.

Also, this function scares me to no end: lack of locking and recursion are the main things that worry me. My vote would be to implement something based on Rafael's approach (which conveniently solves all kind
of other issues).

I have looked at the latest version of Rafael's patchset for device functional
dependency tracking and in its current form it does not address the issues
we see with dependencies on irqchips under ACPI boot.

The problem is that the conversion of ACPI IRQ resources to Linux IRQ resources
for most DSDT devices occurs before the device is even added and is not retried
when a device is re-probed after returning -EPROBE_DEFER. The call chain in most
cases is:

     acpi_init()
         acpi_scan_init()
             acpi_bus_scan()
                 acpi_bus_attach()
                     acpi_default_enumeration()
                         acpi_create_platform_device()
                             acpi_dev_get_resources() /* Resources are 

converted here */
platform_device_register_full()
platform_device_add()
device_add()

What I would like to propose, and I would like feedback before I go down that road,
is to handle this in a way similar to DT, meaning that instead of doing on demand
probing of the irqchip devices described in the DSDT probe table I introduce, we do
that in an init function (e.g. acpi_bus_init_irq).

Thoughts?

Thanks.

Quoted 8 lines

I'll review this patch series in a more in-depth way soon, but I wanted to chime in and add my own weight to Lorenzo's proposal.

Thanks,

	M.

--
Jazz is not dead. It just smells funny...

Signature · 4 lines
Qualcomm Datacenter Technologies, Inc. on behalf of the Qualcomm 
Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a 
Linux Foundation Collaborative Project.