All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH 00/11] irqchip: stm32: add exti support for stm32mp157c
@ 2018-04-26 16:18 ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

Exti controller has been differently integrated on stm32mp1 SoC.
A parent irq has only one external interrupt Vs stm32f4: one parent irq
can have some external interrupts. On stm32mp1 hierachy domain could be used.
Handlers are call by parent, each parent interrupt could be masked and
unmasked according to the needs.

Introduces chips/host/driver data structure to support different
stm32 exti controllers variant and regroup common functions which could
be reused by variants.

Ludovic Barre (10):
  irqchip: stm32: checkpatch fix
  irqchip: stm32: add falling pending register support
  irqchip: stm32: add suspend support
  irqchip: stm32: add host and driver data structures
  irqchip: stm32: prepare common functions
  irqchip: stm32: add stm32mp1 support with hierarchy domain
  irqchip: stm32: add suspend/resume support for hierarchy domain
  pinctrl: stm32: add irq_eoi for stm32gpio irqchip
  ARM: dts: stm32: add exti support for stm32mp157c
  ARM: dts: stm32: add exti support to stm32mp157 pinctrl

radek (1):
  irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain

 .../interrupt-controller/st,stm32-exti.txt         |   3 +
 arch/arm/boot/dts/stm32mp157-pinctrl.dtsi          |   4 +
 arch/arm/boot/dts/stm32mp157c.dtsi                 |   7 +
 drivers/irqchip/irq-stm32-exti.c                   | 683 ++++++++++++++++++---
 drivers/pinctrl/stm32/pinctrl-stm32.c              |  13 +-
 5 files changed, 612 insertions(+), 98 deletions(-)

-- 
2.7.4

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 00/11] irqchip: stm32: add exti support for stm32mp157c
@ 2018-04-26 16:18 ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

Exti controller has been differently integrated on stm32mp1 SoC.
A parent irq has only one external interrupt Vs stm32f4: one parent irq
can have some external interrupts. On stm32mp1 hierachy domain could be used.
Handlers are call by parent, each parent interrupt could be masked and
unmasked according to the needs.

Introduces chips/host/driver data structure to support different
stm32 exti controllers variant and regroup common functions which could
be reused by variants.

Ludovic Barre (10):
  irqchip: stm32: checkpatch fix
  irqchip: stm32: add falling pending register support
  irqchip: stm32: add suspend support
  irqchip: stm32: add host and driver data structures
  irqchip: stm32: prepare common functions
  irqchip: stm32: add stm32mp1 support with hierarchy domain
  irqchip: stm32: add suspend/resume support for hierarchy domain
  pinctrl: stm32: add irq_eoi for stm32gpio irqchip
  ARM: dts: stm32: add exti support for stm32mp157c
  ARM: dts: stm32: add exti support to stm32mp157 pinctrl

radek (1):
  irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain

 .../interrupt-controller/st,stm32-exti.txt         |   3 +
 arch/arm/boot/dts/stm32mp157-pinctrl.dtsi          |   4 +
 arch/arm/boot/dts/stm32mp157c.dtsi                 |   7 +
 drivers/irqchip/irq-stm32-exti.c                   | 683 ++++++++++++++++++---
 drivers/pinctrl/stm32/pinctrl-stm32.c              |  13 +-
 5 files changed, 612 insertions(+), 98 deletions(-)

-- 
2.7.4

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 00/11] irqchip: stm32: add exti support for stm32mp157c
@ 2018-04-26 16:18 ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

Exti controller has been differently integrated on stm32mp1 SoC.
A parent irq has only one external interrupt Vs stm32f4: one parent irq
can have some external interrupts. On stm32mp1 hierachy domain could be used.
Handlers are call by parent, each parent interrupt could be masked and
unmasked according to the needs.

Introduces chips/host/driver data structure to support different
stm32 exti controllers variant and regroup common functions which could
be reused by variants.

Ludovic Barre (10):
  irqchip: stm32: checkpatch fix
  irqchip: stm32: add falling pending register support
  irqchip: stm32: add suspend support
  irqchip: stm32: add host and driver data structures
  irqchip: stm32: prepare common functions
  irqchip: stm32: add stm32mp1 support with hierarchy domain
  irqchip: stm32: add suspend/resume support for hierarchy domain
  pinctrl: stm32: add irq_eoi for stm32gpio irqchip
  ARM: dts: stm32: add exti support for stm32mp157c
  ARM: dts: stm32: add exti support to stm32mp157 pinctrl

radek (1):
  irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain

 .../interrupt-controller/st,stm32-exti.txt         |   3 +
 arch/arm/boot/dts/stm32mp157-pinctrl.dtsi          |   4 +
 arch/arm/boot/dts/stm32mp157c.dtsi                 |   7 +
 drivers/irqchip/irq-stm32-exti.c                   | 683 ++++++++++++++++++---
 drivers/pinctrl/stm32/pinctrl-stm32.c              |  13 +-
 5 files changed, 612 insertions(+), 98 deletions(-)

-- 
2.7.4

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, radek, Ludovic Barre

From: radek <radoslaw.pietrzyk@gmail.com>

- In stm32_exti_alloc function, discards irq_domain_set_info
  with handle_simple_irq. This overwrite the setting defined while init
  of generic chips. Exti controller manages edge irq type.
- Removes acking in chained irq handler as this is done by
  irq_chip itself inside handle_edge_irq
- removes unneeded irq_domain_ops.xlate callback

Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
Acked-by: Ludovic Barre <ludovic.barre@st.com>
Tested-by: Ludovic Barre <ludovic.barre@st.com>
Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 36f0fbe..8013a87 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -79,13 +79,6 @@ static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 	return irq_reg_readl(gc, stm32_bank->pr_ofst);
 }
 
-static void stm32_exti_irq_ack(struct irq_chip_generic *gc, u32 mask)
-{
-	const struct stm32_exti_bank *stm32_bank = gc->private;
-
-	irq_reg_writel(gc, mask, stm32_bank->pr_ofst);
-}
-
 static void stm32_irq_handler(struct irq_desc *desc)
 {
 	struct irq_domain *domain = irq_desc_get_handler_data(desc);
@@ -106,7 +99,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 			for_each_set_bit(n, &pending, IRQS_PER_BANK) {
 				virq = irq_find_mapping(domain, irq_base + n);
 				generic_handle_irq(virq);
-				stm32_exti_irq_ack(gc, BIT(n));
 			}
 		}
 	}
@@ -176,16 +168,12 @@ static int stm32_irq_set_wake(struct irq_data *data, unsigned int on)
 static int stm32_exti_alloc(struct irq_domain *d, unsigned int virq,
 			    unsigned int nr_irqs, void *data)
 {
-	struct irq_chip_generic *gc;
 	struct irq_fwspec *fwspec = data;
 	irq_hw_number_t hwirq;
 
 	hwirq = fwspec->param[0];
-	gc = irq_get_domain_generic_chip(d, hwirq);
 
 	irq_map_generic_chip(d, virq, hwirq);
-	irq_domain_set_info(d, virq, hwirq, &gc->chip_types->chip, gc,
-			    handle_simple_irq, NULL, NULL);
 
 	return 0;
 }
@@ -200,7 +188,6 @@ static void stm32_exti_free(struct irq_domain *d, unsigned int virq,
 
 struct irq_domain_ops irq_exti_domain_ops = {
 	.map	= irq_map_generic_chip,
-	.xlate	= irq_domain_xlate_onetwocell,
 	.alloc  = stm32_exti_alloc,
 	.free	= stm32_exti_free,
 };
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, radek, Ludovic Barre

From: radek <radoslaw.pietrzyk@gmail.com>

- In stm32_exti_alloc function, discards irq_domain_set_info
  with handle_simple_irq. This overwrite the setting defined while init
  of generic chips. Exti controller manages edge irq type.
- Removes acking in chained irq handler as this is done by
  irq_chip itself inside handle_edge_irq
- removes unneeded irq_domain_ops.xlate callback

Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
Acked-by: Ludovic Barre <ludovic.barre@st.com>
Tested-by: Ludovic Barre <ludovic.barre@st.com>
Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 36f0fbe..8013a87 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -79,13 +79,6 @@ static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 	return irq_reg_readl(gc, stm32_bank->pr_ofst);
 }
 
-static void stm32_exti_irq_ack(struct irq_chip_generic *gc, u32 mask)
-{
-	const struct stm32_exti_bank *stm32_bank = gc->private;
-
-	irq_reg_writel(gc, mask, stm32_bank->pr_ofst);
-}
-
 static void stm32_irq_handler(struct irq_desc *desc)
 {
 	struct irq_domain *domain = irq_desc_get_handler_data(desc);
@@ -106,7 +99,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 			for_each_set_bit(n, &pending, IRQS_PER_BANK) {
 				virq = irq_find_mapping(domain, irq_base + n);
 				generic_handle_irq(virq);
-				stm32_exti_irq_ack(gc, BIT(n));
 			}
 		}
 	}
@@ -176,16 +168,12 @@ static int stm32_irq_set_wake(struct irq_data *data, unsigned int on)
 static int stm32_exti_alloc(struct irq_domain *d, unsigned int virq,
 			    unsigned int nr_irqs, void *data)
 {
-	struct irq_chip_generic *gc;
 	struct irq_fwspec *fwspec = data;
 	irq_hw_number_t hwirq;
 
 	hwirq = fwspec->param[0];
-	gc = irq_get_domain_generic_chip(d, hwirq);
 
 	irq_map_generic_chip(d, virq, hwirq);
-	irq_domain_set_info(d, virq, hwirq, &gc->chip_types->chip, gc,
-			    handle_simple_irq, NULL, NULL);
 
 	return 0;
 }
@@ -200,7 +188,6 @@ static void stm32_exti_free(struct irq_domain *d, unsigned int virq,
 
 struct irq_domain_ops irq_exti_domain_ops = {
 	.map	= irq_map_generic_chip,
-	.xlate	= irq_domain_xlate_onetwocell,
 	.alloc  = stm32_exti_alloc,
 	.free	= stm32_exti_free,
 };
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: radek <radoslaw.pietrzyk@gmail.com>

- In stm32_exti_alloc function, discards irq_domain_set_info
  with handle_simple_irq. This overwrite the setting defined while init
  of generic chips. Exti controller manages edge irq type.
- Removes acking in chained irq handler as this is done by
  irq_chip itself inside handle_edge_irq
- removes unneeded irq_domain_ops.xlate callback

Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
Acked-by: Ludovic Barre <ludovic.barre@st.com>
Tested-by: Ludovic Barre <ludovic.barre@st.com>
Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 36f0fbe..8013a87 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -79,13 +79,6 @@ static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 	return irq_reg_readl(gc, stm32_bank->pr_ofst);
 }
 
-static void stm32_exti_irq_ack(struct irq_chip_generic *gc, u32 mask)
-{
-	const struct stm32_exti_bank *stm32_bank = gc->private;
-
-	irq_reg_writel(gc, mask, stm32_bank->pr_ofst);
-}
-
 static void stm32_irq_handler(struct irq_desc *desc)
 {
 	struct irq_domain *domain = irq_desc_get_handler_data(desc);
@@ -106,7 +99,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 			for_each_set_bit(n, &pending, IRQS_PER_BANK) {
 				virq = irq_find_mapping(domain, irq_base + n);
 				generic_handle_irq(virq);
-				stm32_exti_irq_ack(gc, BIT(n));
 			}
 		}
 	}
@@ -176,16 +168,12 @@ static int stm32_irq_set_wake(struct irq_data *data, unsigned int on)
 static int stm32_exti_alloc(struct irq_domain *d, unsigned int virq,
 			    unsigned int nr_irqs, void *data)
 {
-	struct irq_chip_generic *gc;
 	struct irq_fwspec *fwspec = data;
 	irq_hw_number_t hwirq;
 
 	hwirq = fwspec->param[0];
-	gc = irq_get_domain_generic_chip(d, hwirq);
 
 	irq_map_generic_chip(d, virq, hwirq);
-	irq_domain_set_info(d, virq, hwirq, &gc->chip_types->chip, gc,
-			    handle_simple_irq, NULL, NULL);
 
 	return 0;
 }
@@ -200,7 +188,6 @@ static void stm32_exti_free(struct irq_domain *d, unsigned int virq,
 
 struct irq_domain_ops irq_exti_domain_ops = {
 	.map	= irq_map_generic_chip,
-	.xlate	= irq_domain_xlate_onetwocell,
 	.alloc  = stm32_exti_alloc,
 	.free	= stm32_exti_free,
 };
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 02/11] irqchip: stm32: checkpatch fix
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

-WARNING: struct irq_domain_ops should normally be const
-CHECK: Alignment should match open parenthesis

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 8013a87..b91a8c2 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -186,7 +186,7 @@ static void stm32_exti_free(struct irq_domain *d, unsigned int virq,
 	irq_domain_reset_irq_data(data);
 }
 
-struct irq_domain_ops irq_exti_domain_ops = {
+static const struct irq_domain_ops irq_exti_domain_ops = {
 	.map	= irq_map_generic_chip,
 	.alloc  = stm32_exti_alloc,
 	.free	= stm32_exti_free,
@@ -221,7 +221,7 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 					     handle_edge_irq, clr, 0, 0);
 	if (ret) {
 		pr_err("%pOF: Could not allocate generic interrupt chip.\n",
-			node);
+		       node);
 		goto out_free_domain;
 	}
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 02/11] irqchip: stm32: checkpatch fix
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

-WARNING: struct irq_domain_ops should normally be const
-CHECK: Alignment should match open parenthesis

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 8013a87..b91a8c2 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -186,7 +186,7 @@ static void stm32_exti_free(struct irq_domain *d, unsigned int virq,
 	irq_domain_reset_irq_data(data);
 }
 
-struct irq_domain_ops irq_exti_domain_ops = {
+static const struct irq_domain_ops irq_exti_domain_ops = {
 	.map	= irq_map_generic_chip,
 	.alloc  = stm32_exti_alloc,
 	.free	= stm32_exti_free,
@@ -221,7 +221,7 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 					     handle_edge_irq, clr, 0, 0);
 	if (ret) {
 		pr_err("%pOF: Could not allocate generic interrupt chip.\n",
-			node);
+		       node);
 		goto out_free_domain;
 	}
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 02/11] irqchip: stm32: checkpatch fix
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

-WARNING: struct irq_domain_ops should normally be const
-CHECK: Alignment should match open parenthesis

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 8013a87..b91a8c2 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -186,7 +186,7 @@ static void stm32_exti_free(struct irq_domain *d, unsigned int virq,
 	irq_domain_reset_irq_data(data);
 }
 
-struct irq_domain_ops irq_exti_domain_ops = {
+static const struct irq_domain_ops irq_exti_domain_ops = {
 	.map	= irq_map_generic_chip,
 	.alloc  = stm32_exti_alloc,
 	.free	= stm32_exti_free,
@@ -221,7 +221,7 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 					     handle_edge_irq, clr, 0, 0);
 	if (ret) {
 		pr_err("%pOF: Could not allocate generic interrupt chip.\n",
-			node);
+		       node);
 		goto out_free_domain;
 	}
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 03/11] irqchip: stm32: add falling pending register support
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds support of rising/falling pending registers.
Falling pending register (fpr) is needed for next revision.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 47 ++++++++++++++++++++++++++++++----------
 1 file changed, 36 insertions(+), 11 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index b91a8c2..69a4453 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -23,16 +23,20 @@ struct stm32_exti_bank {
 	u32 rtsr_ofst;
 	u32 ftsr_ofst;
 	u32 swier_ofst;
-	u32 pr_ofst;
+	u32 rpr_ofst;
+	u32 fpr_ofst;
 };
 
+#define UNDEF_REG ~0
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
 	.rtsr_ofst	= 0x08,
 	.ftsr_ofst	= 0x0C,
 	.swier_ofst	= 0x10,
-	.pr_ofst	= 0x14,
+	.rpr_ofst	= 0x14,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank *stm32f4xx_exti_banks[] = {
@@ -45,7 +49,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b1 = {
 	.rtsr_ofst	= 0x00,
 	.ftsr_ofst	= 0x04,
 	.swier_ofst	= 0x08,
-	.pr_ofst	= 0x88,
+	.rpr_ofst	= 0x88,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank stm32h7xx_exti_b2 = {
@@ -54,7 +59,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b2 = {
 	.rtsr_ofst	= 0x20,
 	.ftsr_ofst	= 0x24,
 	.swier_ofst	= 0x28,
-	.pr_ofst	= 0x98,
+	.rpr_ofst	= 0x98,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank stm32h7xx_exti_b3 = {
@@ -63,7 +69,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b3 = {
 	.rtsr_ofst	= 0x40,
 	.ftsr_ofst	= 0x44,
 	.swier_ofst	= 0x48,
-	.pr_ofst	= 0xA8,
+	.rpr_ofst	= 0xA8,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
@@ -75,8 +82,13 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	const struct stm32_exti_bank *stm32_bank = gc->private;
+	unsigned long pending;
+
+	pending = irq_reg_readl(gc, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		pending |= irq_reg_readl(gc, stm32_bank->fpr_ofst);
 
-	return irq_reg_readl(gc, stm32_bank->pr_ofst);
+	return pending;
 }
 
 static void stm32_irq_handler(struct irq_desc *desc)
@@ -85,7 +97,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 	struct irq_chip *chip = irq_desc_get_chip(desc);
 	unsigned int virq, nbanks = domain->gc->num_chips;
 	struct irq_chip_generic *gc;
-	const struct stm32_exti_bank *stm32_bank;
 	unsigned long pending;
 	int n, i, irq_base = 0;
 
@@ -93,7 +104,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 
 	for (i = 0; i < nbanks; i++, irq_base += IRQS_PER_BANK) {
 		gc = irq_get_domain_generic_chip(domain, irq_base);
-		stm32_bank = gc->private;
 
 		while ((pending = stm32_exti_pending(gc))) {
 			for_each_set_bit(n, &pending, IRQS_PER_BANK) {
@@ -192,6 +202,20 @@ static const struct irq_domain_ops irq_exti_domain_ops = {
 	.free	= stm32_exti_free,
 };
 
+static void stm32_irq_ack(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = gc->private;
+
+	irq_gc_lock(gc);
+
+	irq_reg_writel(gc, d->mask, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		irq_reg_writel(gc, d->mask, stm32_bank->fpr_ofst);
+
+	irq_gc_unlock(gc);
+}
+
 static int
 __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		       int bank_nr, struct device_node *node)
@@ -233,12 +257,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 		gc->reg_base = base;
 		gc->chip_types->type = IRQ_TYPE_EDGE_BOTH;
-		gc->chip_types->chip.irq_ack = irq_gc_ack_set_bit;
+		gc->chip_types->chip.irq_ack = stm32_irq_ack;
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
 		gc->chip_types->chip.irq_unmask = irq_gc_mask_set_bit;
 		gc->chip_types->chip.irq_set_type = stm32_irq_set_type;
 		gc->chip_types->chip.irq_set_wake = stm32_irq_set_wake;
-		gc->chip_types->regs.ack = stm32_bank->pr_ofst;
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
 		gc->private = (void *)stm32_bank;
 
@@ -255,7 +278,9 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		writel_relaxed(0, base + stm32_bank->emr_ofst);
 		writel_relaxed(0, base + stm32_bank->rtsr_ofst);
 		writel_relaxed(0, base + stm32_bank->ftsr_ofst);
-		writel_relaxed(~0UL, base + stm32_bank->pr_ofst);
+		writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
+		if (stm32_bank->fpr_ofst != UNDEF_REG)
+			writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
 
 		pr_info("%s: bank%d, External IRQs available:%#x\n",
 			node->full_name, i, irqs_mask);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 03/11] irqchip: stm32: add falling pending register support
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds support of rising/falling pending registers.
Falling pending register (fpr) is needed for next revision.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 47 ++++++++++++++++++++++++++++++----------
 1 file changed, 36 insertions(+), 11 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index b91a8c2..69a4453 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -23,16 +23,20 @@ struct stm32_exti_bank {
 	u32 rtsr_ofst;
 	u32 ftsr_ofst;
 	u32 swier_ofst;
-	u32 pr_ofst;
+	u32 rpr_ofst;
+	u32 fpr_ofst;
 };
 
+#define UNDEF_REG ~0
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
 	.rtsr_ofst	= 0x08,
 	.ftsr_ofst	= 0x0C,
 	.swier_ofst	= 0x10,
-	.pr_ofst	= 0x14,
+	.rpr_ofst	= 0x14,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank *stm32f4xx_exti_banks[] = {
@@ -45,7 +49,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b1 = {
 	.rtsr_ofst	= 0x00,
 	.ftsr_ofst	= 0x04,
 	.swier_ofst	= 0x08,
-	.pr_ofst	= 0x88,
+	.rpr_ofst	= 0x88,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank stm32h7xx_exti_b2 = {
@@ -54,7 +59,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b2 = {
 	.rtsr_ofst	= 0x20,
 	.ftsr_ofst	= 0x24,
 	.swier_ofst	= 0x28,
-	.pr_ofst	= 0x98,
+	.rpr_ofst	= 0x98,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank stm32h7xx_exti_b3 = {
@@ -63,7 +69,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b3 = {
 	.rtsr_ofst	= 0x40,
 	.ftsr_ofst	= 0x44,
 	.swier_ofst	= 0x48,
-	.pr_ofst	= 0xA8,
+	.rpr_ofst	= 0xA8,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
@@ -75,8 +82,13 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	const struct stm32_exti_bank *stm32_bank = gc->private;
+	unsigned long pending;
+
+	pending = irq_reg_readl(gc, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		pending |= irq_reg_readl(gc, stm32_bank->fpr_ofst);
 
-	return irq_reg_readl(gc, stm32_bank->pr_ofst);
+	return pending;
 }
 
 static void stm32_irq_handler(struct irq_desc *desc)
@@ -85,7 +97,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 	struct irq_chip *chip = irq_desc_get_chip(desc);
 	unsigned int virq, nbanks = domain->gc->num_chips;
 	struct irq_chip_generic *gc;
-	const struct stm32_exti_bank *stm32_bank;
 	unsigned long pending;
 	int n, i, irq_base = 0;
 
@@ -93,7 +104,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 
 	for (i = 0; i < nbanks; i++, irq_base += IRQS_PER_BANK) {
 		gc = irq_get_domain_generic_chip(domain, irq_base);
-		stm32_bank = gc->private;
 
 		while ((pending = stm32_exti_pending(gc))) {
 			for_each_set_bit(n, &pending, IRQS_PER_BANK) {
@@ -192,6 +202,20 @@ static const struct irq_domain_ops irq_exti_domain_ops = {
 	.free	= stm32_exti_free,
 };
 
+static void stm32_irq_ack(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = gc->private;
+
+	irq_gc_lock(gc);
+
+	irq_reg_writel(gc, d->mask, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		irq_reg_writel(gc, d->mask, stm32_bank->fpr_ofst);
+
+	irq_gc_unlock(gc);
+}
+
 static int
 __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		       int bank_nr, struct device_node *node)
@@ -233,12 +257,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 		gc->reg_base = base;
 		gc->chip_types->type = IRQ_TYPE_EDGE_BOTH;
-		gc->chip_types->chip.irq_ack = irq_gc_ack_set_bit;
+		gc->chip_types->chip.irq_ack = stm32_irq_ack;
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
 		gc->chip_types->chip.irq_unmask = irq_gc_mask_set_bit;
 		gc->chip_types->chip.irq_set_type = stm32_irq_set_type;
 		gc->chip_types->chip.irq_set_wake = stm32_irq_set_wake;
-		gc->chip_types->regs.ack = stm32_bank->pr_ofst;
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
 		gc->private = (void *)stm32_bank;
 
@@ -255,7 +278,9 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		writel_relaxed(0, base + stm32_bank->emr_ofst);
 		writel_relaxed(0, base + stm32_bank->rtsr_ofst);
 		writel_relaxed(0, base + stm32_bank->ftsr_ofst);
-		writel_relaxed(~0UL, base + stm32_bank->pr_ofst);
+		writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
+		if (stm32_bank->fpr_ofst != UNDEF_REG)
+			writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
 
 		pr_info("%s: bank%d, External IRQs available:%#x\n",
 			node->full_name, i, irqs_mask);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 03/11] irqchip: stm32: add falling pending register support
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds support of rising/falling pending registers.
Falling pending register (fpr) is needed for next revision.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 47 ++++++++++++++++++++++++++++++----------
 1 file changed, 36 insertions(+), 11 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index b91a8c2..69a4453 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -23,16 +23,20 @@ struct stm32_exti_bank {
 	u32 rtsr_ofst;
 	u32 ftsr_ofst;
 	u32 swier_ofst;
-	u32 pr_ofst;
+	u32 rpr_ofst;
+	u32 fpr_ofst;
 };
 
+#define UNDEF_REG ~0
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
 	.rtsr_ofst	= 0x08,
 	.ftsr_ofst	= 0x0C,
 	.swier_ofst	= 0x10,
-	.pr_ofst	= 0x14,
+	.rpr_ofst	= 0x14,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank *stm32f4xx_exti_banks[] = {
@@ -45,7 +49,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b1 = {
 	.rtsr_ofst	= 0x00,
 	.ftsr_ofst	= 0x04,
 	.swier_ofst	= 0x08,
-	.pr_ofst	= 0x88,
+	.rpr_ofst	= 0x88,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank stm32h7xx_exti_b2 = {
@@ -54,7 +59,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b2 = {
 	.rtsr_ofst	= 0x20,
 	.ftsr_ofst	= 0x24,
 	.swier_ofst	= 0x28,
-	.pr_ofst	= 0x98,
+	.rpr_ofst	= 0x98,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank stm32h7xx_exti_b3 = {
@@ -63,7 +69,8 @@ static const struct stm32_exti_bank stm32h7xx_exti_b3 = {
 	.rtsr_ofst	= 0x40,
 	.ftsr_ofst	= 0x44,
 	.swier_ofst	= 0x48,
-	.pr_ofst	= 0xA8,
+	.rpr_ofst	= 0xA8,
+	.fpr_ofst	= UNDEF_REG,
 };
 
 static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
@@ -75,8 +82,13 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	const struct stm32_exti_bank *stm32_bank = gc->private;
+	unsigned long pending;
+
+	pending = irq_reg_readl(gc, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		pending |= irq_reg_readl(gc, stm32_bank->fpr_ofst);
 
-	return irq_reg_readl(gc, stm32_bank->pr_ofst);
+	return pending;
 }
 
 static void stm32_irq_handler(struct irq_desc *desc)
@@ -85,7 +97,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 	struct irq_chip *chip = irq_desc_get_chip(desc);
 	unsigned int virq, nbanks = domain->gc->num_chips;
 	struct irq_chip_generic *gc;
-	const struct stm32_exti_bank *stm32_bank;
 	unsigned long pending;
 	int n, i, irq_base = 0;
 
@@ -93,7 +104,6 @@ static void stm32_irq_handler(struct irq_desc *desc)
 
 	for (i = 0; i < nbanks; i++, irq_base += IRQS_PER_BANK) {
 		gc = irq_get_domain_generic_chip(domain, irq_base);
-		stm32_bank = gc->private;
 
 		while ((pending = stm32_exti_pending(gc))) {
 			for_each_set_bit(n, &pending, IRQS_PER_BANK) {
@@ -192,6 +202,20 @@ static const struct irq_domain_ops irq_exti_domain_ops = {
 	.free	= stm32_exti_free,
 };
 
+static void stm32_irq_ack(struct irq_data *d)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = gc->private;
+
+	irq_gc_lock(gc);
+
+	irq_reg_writel(gc, d->mask, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		irq_reg_writel(gc, d->mask, stm32_bank->fpr_ofst);
+
+	irq_gc_unlock(gc);
+}
+
 static int
 __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		       int bank_nr, struct device_node *node)
@@ -233,12 +257,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 		gc->reg_base = base;
 		gc->chip_types->type = IRQ_TYPE_EDGE_BOTH;
-		gc->chip_types->chip.irq_ack = irq_gc_ack_set_bit;
+		gc->chip_types->chip.irq_ack = stm32_irq_ack;
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
 		gc->chip_types->chip.irq_unmask = irq_gc_mask_set_bit;
 		gc->chip_types->chip.irq_set_type = stm32_irq_set_type;
 		gc->chip_types->chip.irq_set_wake = stm32_irq_set_wake;
-		gc->chip_types->regs.ack = stm32_bank->pr_ofst;
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
 		gc->private = (void *)stm32_bank;
 
@@ -255,7 +278,9 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		writel_relaxed(0, base + stm32_bank->emr_ofst);
 		writel_relaxed(0, base + stm32_bank->rtsr_ofst);
 		writel_relaxed(0, base + stm32_bank->ftsr_ofst);
-		writel_relaxed(~0UL, base + stm32_bank->pr_ofst);
+		writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
+		if (stm32_bank->fpr_ofst != UNDEF_REG)
+			writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
 
 		pr_info("%s: bank%d, External IRQs available:%#x\n",
 			node->full_name, i, irqs_mask);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 04/11] irqchip: stm32: add suspend support
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds suspend feature.
-Use default irq_set_wake function to store wakeup request.
-Suspend function set wake_active into imr of each bank
 and save rising/falling trigger registers.
-Resume function restore the mask_cache interrupt into
 imr of each bank and restore rising/falling trigger registers.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 69 ++++++++++++++++++++++++++++++----------
 1 file changed, 52 insertions(+), 17 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 69a4453..1e09667 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -29,6 +29,14 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_exti_chip_data {
+	const struct stm32_exti_bank *reg_bank;
+	u32 rtsr_cache;
+	u32 ftsr_cache;
+};
+
+static struct stm32_exti_chip_data *stm32_exti_data;
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
@@ -81,7 +89,8 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 	unsigned long pending;
 
 	pending = irq_reg_readl(gc, stm32_bank->rpr_ofst);
@@ -119,7 +128,8 @@ static void stm32_irq_handler(struct irq_desc *desc)
 static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 {
 	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 	int pin = data->hwirq % IRQS_PER_BANK;
 	u32 rtsr, ftsr;
 
@@ -154,25 +164,36 @@ static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
-static int stm32_irq_set_wake(struct irq_data *data, unsigned int on)
+static void stm32_irq_suspend(struct irq_chip_generic *gc)
 {
-	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
-	int pin = data->hwirq % IRQS_PER_BANK;
-	u32 imr;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
 
-	imr = irq_reg_readl(gc, stm32_bank->imr_ofst);
-	if (on)
-		imr |= BIT(pin);
-	else
-		imr &= ~BIT(pin);
-	irq_reg_writel(gc, imr, stm32_bank->imr_ofst);
+	/* save rtsr, ftsr registers */
+	chip_data->rtsr_cache = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
+	chip_data->ftsr_cache = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+
+	irq_reg_writel(gc, gc->wake_active, stm32_bank->imr_ofst);
 
 	irq_gc_unlock(gc);
+}
 
-	return 0;
+static void stm32_irq_resume(struct irq_chip_generic *gc)
+{
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	irq_gc_lock(gc);
+
+	/* restore rtsr, ftsr registers */
+	irq_reg_writel(gc, chip_data->rtsr_cache, stm32_bank->rtsr_ofst);
+	irq_reg_writel(gc, chip_data->ftsr_cache, stm32_bank->ftsr_ofst);
+
+	irq_reg_writel(gc, gc->mask_cache, stm32_bank->imr_ofst);
+
+	irq_gc_unlock(gc);
 }
 
 static int stm32_exti_alloc(struct irq_domain *d, unsigned int virq,
@@ -205,7 +226,8 @@ static const struct irq_domain_ops irq_exti_domain_ops = {
 static void stm32_irq_ack(struct irq_data *d)
 {
 	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
 
@@ -232,6 +254,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		return -ENOMEM;
 	}
 
+	stm32_exti_data = kcalloc(bank_nr, sizeof(*stm32_exti_data),
+				  GFP_KERNEL);
+	if (!stm32_exti_data)
+		return -ENOMEM;
+
 	domain = irq_domain_add_linear(node, bank_nr * IRQS_PER_BANK,
 				       &irq_exti_domain_ops, NULL);
 	if (!domain) {
@@ -251,8 +278,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 	for (i = 0; i < bank_nr; i++) {
 		const struct stm32_exti_bank *stm32_bank = stm32_exti_banks[i];
+		struct stm32_exti_chip_data *chip_data = &stm32_exti_data[i];
 		u32 irqs_mask;
 
+		chip_data->reg_bank = stm32_bank;
+
 		gc = irq_get_domain_generic_chip(domain, i * IRQS_PER_BANK);
 
 		gc->reg_base = base;
@@ -261,9 +291,13 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
 		gc->chip_types->chip.irq_unmask = irq_gc_mask_set_bit;
 		gc->chip_types->chip.irq_set_type = stm32_irq_set_type;
-		gc->chip_types->chip.irq_set_wake = stm32_irq_set_wake;
+		gc->chip_types->chip.irq_set_wake = irq_gc_set_wake;
+		gc->suspend = stm32_irq_suspend;
+		gc->resume = stm32_irq_resume;
+		gc->wake_enabled = IRQ_MSK(IRQS_PER_BANK);
+
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
-		gc->private = (void *)stm32_bank;
+		gc->private = (void *)chip_data;
 
 		/* Determine number of irqs supported */
 		writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
@@ -300,6 +334,7 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 	irq_domain_remove(domain);
 out_unmap:
 	iounmap(base);
+	kfree(stm32_exti_data);
 	return ret;
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 04/11] irqchip: stm32: add suspend support
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds suspend feature.
-Use default irq_set_wake function to store wakeup request.
-Suspend function set wake_active into imr of each bank
 and save rising/falling trigger registers.
-Resume function restore the mask_cache interrupt into
 imr of each bank and restore rising/falling trigger registers.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 69 ++++++++++++++++++++++++++++++----------
 1 file changed, 52 insertions(+), 17 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 69a4453..1e09667 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -29,6 +29,14 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_exti_chip_data {
+	const struct stm32_exti_bank *reg_bank;
+	u32 rtsr_cache;
+	u32 ftsr_cache;
+};
+
+static struct stm32_exti_chip_data *stm32_exti_data;
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
@@ -81,7 +89,8 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 	unsigned long pending;
 
 	pending = irq_reg_readl(gc, stm32_bank->rpr_ofst);
@@ -119,7 +128,8 @@ static void stm32_irq_handler(struct irq_desc *desc)
 static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 {
 	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 	int pin = data->hwirq % IRQS_PER_BANK;
 	u32 rtsr, ftsr;
 
@@ -154,25 +164,36 @@ static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
-static int stm32_irq_set_wake(struct irq_data *data, unsigned int on)
+static void stm32_irq_suspend(struct irq_chip_generic *gc)
 {
-	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
-	int pin = data->hwirq % IRQS_PER_BANK;
-	u32 imr;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
 
-	imr = irq_reg_readl(gc, stm32_bank->imr_ofst);
-	if (on)
-		imr |= BIT(pin);
-	else
-		imr &= ~BIT(pin);
-	irq_reg_writel(gc, imr, stm32_bank->imr_ofst);
+	/* save rtsr, ftsr registers */
+	chip_data->rtsr_cache = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
+	chip_data->ftsr_cache = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+
+	irq_reg_writel(gc, gc->wake_active, stm32_bank->imr_ofst);
 
 	irq_gc_unlock(gc);
+}
 
-	return 0;
+static void stm32_irq_resume(struct irq_chip_generic *gc)
+{
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	irq_gc_lock(gc);
+
+	/* restore rtsr, ftsr registers */
+	irq_reg_writel(gc, chip_data->rtsr_cache, stm32_bank->rtsr_ofst);
+	irq_reg_writel(gc, chip_data->ftsr_cache, stm32_bank->ftsr_ofst);
+
+	irq_reg_writel(gc, gc->mask_cache, stm32_bank->imr_ofst);
+
+	irq_gc_unlock(gc);
 }
 
 static int stm32_exti_alloc(struct irq_domain *d, unsigned int virq,
@@ -205,7 +226,8 @@ static const struct irq_domain_ops irq_exti_domain_ops = {
 static void stm32_irq_ack(struct irq_data *d)
 {
 	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
 
@@ -232,6 +254,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		return -ENOMEM;
 	}
 
+	stm32_exti_data = kcalloc(bank_nr, sizeof(*stm32_exti_data),
+				  GFP_KERNEL);
+	if (!stm32_exti_data)
+		return -ENOMEM;
+
 	domain = irq_domain_add_linear(node, bank_nr * IRQS_PER_BANK,
 				       &irq_exti_domain_ops, NULL);
 	if (!domain) {
@@ -251,8 +278,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 	for (i = 0; i < bank_nr; i++) {
 		const struct stm32_exti_bank *stm32_bank = stm32_exti_banks[i];
+		struct stm32_exti_chip_data *chip_data = &stm32_exti_data[i];
 		u32 irqs_mask;
 
+		chip_data->reg_bank = stm32_bank;
+
 		gc = irq_get_domain_generic_chip(domain, i * IRQS_PER_BANK);
 
 		gc->reg_base = base;
@@ -261,9 +291,13 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
 		gc->chip_types->chip.irq_unmask = irq_gc_mask_set_bit;
 		gc->chip_types->chip.irq_set_type = stm32_irq_set_type;
-		gc->chip_types->chip.irq_set_wake = stm32_irq_set_wake;
+		gc->chip_types->chip.irq_set_wake = irq_gc_set_wake;
+		gc->suspend = stm32_irq_suspend;
+		gc->resume = stm32_irq_resume;
+		gc->wake_enabled = IRQ_MSK(IRQS_PER_BANK);
+
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
-		gc->private = (void *)stm32_bank;
+		gc->private = (void *)chip_data;
 
 		/* Determine number of irqs supported */
 		writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
@@ -300,6 +334,7 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 	irq_domain_remove(domain);
 out_unmap:
 	iounmap(base);
+	kfree(stm32_exti_data);
 	return ret;
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 04/11] irqchip: stm32: add suspend support
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds suspend feature.
-Use default irq_set_wake function to store wakeup request.
-Suspend function set wake_active into imr of each bank
 and save rising/falling trigger registers.
-Resume function restore the mask_cache interrupt into
 imr of each bank and restore rising/falling trigger registers.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 69 ++++++++++++++++++++++++++++++----------
 1 file changed, 52 insertions(+), 17 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 69a4453..1e09667 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -29,6 +29,14 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_exti_chip_data {
+	const struct stm32_exti_bank *reg_bank;
+	u32 rtsr_cache;
+	u32 ftsr_cache;
+};
+
+static struct stm32_exti_chip_data *stm32_exti_data;
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
@@ -81,7 +89,8 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 	unsigned long pending;
 
 	pending = irq_reg_readl(gc, stm32_bank->rpr_ofst);
@@ -119,7 +128,8 @@ static void stm32_irq_handler(struct irq_desc *desc)
 static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 {
 	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 	int pin = data->hwirq % IRQS_PER_BANK;
 	u32 rtsr, ftsr;
 
@@ -154,25 +164,36 @@ static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
-static int stm32_irq_set_wake(struct irq_data *data, unsigned int on)
+static void stm32_irq_suspend(struct irq_chip_generic *gc)
 {
-	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
-	int pin = data->hwirq % IRQS_PER_BANK;
-	u32 imr;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
 
-	imr = irq_reg_readl(gc, stm32_bank->imr_ofst);
-	if (on)
-		imr |= BIT(pin);
-	else
-		imr &= ~BIT(pin);
-	irq_reg_writel(gc, imr, stm32_bank->imr_ofst);
+	/* save rtsr, ftsr registers */
+	chip_data->rtsr_cache = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
+	chip_data->ftsr_cache = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+
+	irq_reg_writel(gc, gc->wake_active, stm32_bank->imr_ofst);
 
 	irq_gc_unlock(gc);
+}
 
-	return 0;
+static void stm32_irq_resume(struct irq_chip_generic *gc)
+{
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	irq_gc_lock(gc);
+
+	/* restore rtsr, ftsr registers */
+	irq_reg_writel(gc, chip_data->rtsr_cache, stm32_bank->rtsr_ofst);
+	irq_reg_writel(gc, chip_data->ftsr_cache, stm32_bank->ftsr_ofst);
+
+	irq_reg_writel(gc, gc->mask_cache, stm32_bank->imr_ofst);
+
+	irq_gc_unlock(gc);
 }
 
 static int stm32_exti_alloc(struct irq_domain *d, unsigned int virq,
@@ -205,7 +226,8 @@ static const struct irq_domain_ops irq_exti_domain_ops = {
 static void stm32_irq_ack(struct irq_data *d)
 {
 	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
-	const struct stm32_exti_bank *stm32_bank = gc->private;
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
 
@@ -232,6 +254,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		return -ENOMEM;
 	}
 
+	stm32_exti_data = kcalloc(bank_nr, sizeof(*stm32_exti_data),
+				  GFP_KERNEL);
+	if (!stm32_exti_data)
+		return -ENOMEM;
+
 	domain = irq_domain_add_linear(node, bank_nr * IRQS_PER_BANK,
 				       &irq_exti_domain_ops, NULL);
 	if (!domain) {
@@ -251,8 +278,11 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 	for (i = 0; i < bank_nr; i++) {
 		const struct stm32_exti_bank *stm32_bank = stm32_exti_banks[i];
+		struct stm32_exti_chip_data *chip_data = &stm32_exti_data[i];
 		u32 irqs_mask;
 
+		chip_data->reg_bank = stm32_bank;
+
 		gc = irq_get_domain_generic_chip(domain, i * IRQS_PER_BANK);
 
 		gc->reg_base = base;
@@ -261,9 +291,13 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
 		gc->chip_types->chip.irq_unmask = irq_gc_mask_set_bit;
 		gc->chip_types->chip.irq_set_type = stm32_irq_set_type;
-		gc->chip_types->chip.irq_set_wake = stm32_irq_set_wake;
+		gc->chip_types->chip.irq_set_wake = irq_gc_set_wake;
+		gc->suspend = stm32_irq_suspend;
+		gc->resume = stm32_irq_resume;
+		gc->wake_enabled = IRQ_MSK(IRQS_PER_BANK);
+
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
-		gc->private = (void *)stm32_bank;
+		gc->private = (void *)chip_data;
 
 		/* Determine number of irqs supported */
 		writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
@@ -300,6 +334,7 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 	irq_domain_remove(domain);
 out_unmap:
 	iounmap(base);
+	kfree(stm32_exti_data);
 	return ret;
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 05/11] irqchip: stm32: add host and driver data structures
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds host and driver data structures to support
different stm32 exti controllers with variants.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 152 ++++++++++++++++++++++++++-------------
 1 file changed, 104 insertions(+), 48 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 1e09667..9655a57 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -29,13 +29,23 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_exti_drv_data {
+	const struct stm32_exti_bank **exti_banks;
+	u32 bank_nr;
+};
+
 struct stm32_exti_chip_data {
+	struct stm32_exti_host_data *host_data;
 	const struct stm32_exti_bank *reg_bank;
 	u32 rtsr_cache;
 	u32 ftsr_cache;
 };
 
-static struct stm32_exti_chip_data *stm32_exti_data;
+struct stm32_exti_host_data {
+	void __iomem *base;
+	struct stm32_exti_chip_data *chips_data;
+	const struct stm32_exti_drv_data *drv_data;
+};
 
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
@@ -51,6 +61,11 @@ static const struct stm32_exti_bank *stm32f4xx_exti_banks[] = {
 	&stm32f4xx_exti_b1,
 };
 
+static const struct stm32_exti_drv_data stm32f4xx_drv_data = {
+	.exti_banks = stm32f4xx_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32f4xx_exti_banks),
+};
+
 static const struct stm32_exti_bank stm32h7xx_exti_b1 = {
 	.imr_ofst	= 0x80,
 	.emr_ofst	= 0x84,
@@ -87,6 +102,11 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 	&stm32h7xx_exti_b3,
 };
 
+static const struct stm32_exti_drv_data stm32h7xx_drv_data = {
+	.exti_banks = stm32h7xx_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32h7xx_exti_banks),
+};
+
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
@@ -237,29 +257,85 @@ static void stm32_irq_ack(struct irq_data *d)
 
 	irq_gc_unlock(gc);
 }
+static struct
+stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
+					   struct device_node *node)
+{
+	struct stm32_exti_host_data *host_data;
+
+	host_data = kzalloc(sizeof(*host_data), GFP_KERNEL);
+	if (!host_data)
+		return NULL;
+
+	host_data->drv_data = dd;
+	host_data->chips_data = kcalloc(dd->bank_nr,
+					sizeof(struct stm32_exti_chip_data),
+					GFP_KERNEL);
+	if (!host_data->chips_data)
+		return NULL;
+
+	host_data->base = of_iomap(node, 0);
+	if (!host_data->base) {
+		pr_err("%pOF: Unable to map registers\n", node);
+		return NULL;
+	}
 
-static int
-__init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
-		       int bank_nr, struct device_node *node)
+	return host_data;
+}
+
+static struct
+stm32_exti_chip_data *stm32_exti_chip_init(struct stm32_exti_host_data *h_data,
+					   u32 bank_idx,
+					   struct device_node *node)
+{
+	const struct stm32_exti_bank *stm32_bank;
+	struct stm32_exti_chip_data *chip_data;
+	void __iomem *base = h_data->base;
+	u32 irqs_mask;
+
+	stm32_bank = h_data->drv_data->exti_banks[bank_idx];
+	chip_data = &h_data->chips_data[bank_idx];
+	chip_data->host_data = h_data;
+	chip_data->reg_bank = stm32_bank;
+
+	/* Determine number of irqs supported */
+	writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
+	irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
+
+	/*
+	 * This IP has no reset, so after hot reboot we should
+	 * clear registers to avoid residue
+	 */
+	writel_relaxed(0, base + stm32_bank->imr_ofst);
+	writel_relaxed(0, base + stm32_bank->emr_ofst);
+	writel_relaxed(0, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(0, base + stm32_bank->ftsr_ofst);
+	writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
+
+	pr_info("%s: bank%d, External IRQs available:%#x\n",
+		node->full_name, bank_idx, irqs_mask);
+
+	return chip_data;
+}
+
+static int __init stm32_exti_init(const struct stm32_exti_drv_data *drv_data,
+				  struct device_node *node)
 {
+	struct stm32_exti_host_data *host_data;
 	unsigned int clr = IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN;
-	int nr_irqs, nr_exti, ret, i;
+	int nr_irqs, ret, i;
 	struct irq_chip_generic *gc;
 	struct irq_domain *domain;
-	void *base;
 
-	base = of_iomap(node, 0);
-	if (!base) {
-		pr_err("%pOF: Unable to map registers\n", node);
-		return -ENOMEM;
+	host_data = stm32_exti_host_init(drv_data, node);
+	if (!host_data) {
+		ret = -ENOMEM;
+		goto out_free_mem;
 	}
 
-	stm32_exti_data = kcalloc(bank_nr, sizeof(*stm32_exti_data),
-				  GFP_KERNEL);
-	if (!stm32_exti_data)
-		return -ENOMEM;
-
-	domain = irq_domain_add_linear(node, bank_nr * IRQS_PER_BANK,
+	domain = irq_domain_add_linear(node, drv_data->bank_nr * IRQS_PER_BANK,
 				       &irq_exti_domain_ops, NULL);
 	if (!domain) {
 		pr_err("%s: Could not register interrupt domain.\n",
@@ -276,16 +352,16 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		goto out_free_domain;
 	}
 
-	for (i = 0; i < bank_nr; i++) {
-		const struct stm32_exti_bank *stm32_bank = stm32_exti_banks[i];
-		struct stm32_exti_chip_data *chip_data = &stm32_exti_data[i];
-		u32 irqs_mask;
+	for (i = 0; i < drv_data->bank_nr; i++) {
+		const struct stm32_exti_bank *stm32_bank;
+		struct stm32_exti_chip_data *chip_data;
 
-		chip_data->reg_bank = stm32_bank;
+		stm32_bank = drv_data->exti_banks[i];
+		chip_data = stm32_exti_chip_init(host_data, i, node);
 
 		gc = irq_get_domain_generic_chip(domain, i * IRQS_PER_BANK);
 
-		gc->reg_base = base;
+		gc->reg_base = host_data->base;
 		gc->chip_types->type = IRQ_TYPE_EDGE_BOTH;
 		gc->chip_types->chip.irq_ack = stm32_irq_ack;
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
@@ -298,26 +374,6 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
 		gc->private = (void *)chip_data;
-
-		/* Determine number of irqs supported */
-		writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
-		irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
-		nr_exti = fls(readl_relaxed(base + stm32_bank->rtsr_ofst));
-
-		/*
-		 * This IP has no reset, so after hot reboot we should
-		 * clear registers to avoid residue
-		 */
-		writel_relaxed(0, base + stm32_bank->imr_ofst);
-		writel_relaxed(0, base + stm32_bank->emr_ofst);
-		writel_relaxed(0, base + stm32_bank->rtsr_ofst);
-		writel_relaxed(0, base + stm32_bank->ftsr_ofst);
-		writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
-		if (stm32_bank->fpr_ofst != UNDEF_REG)
-			writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
-
-		pr_info("%s: bank%d, External IRQs available:%#x\n",
-			node->full_name, i, irqs_mask);
 	}
 
 	nr_irqs = of_irq_count(node);
@@ -333,16 +389,17 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 out_free_domain:
 	irq_domain_remove(domain);
 out_unmap:
-	iounmap(base);
-	kfree(stm32_exti_data);
+	iounmap(host_data->base);
+out_free_mem:
+	kfree(host_data->chips_data);
+	kfree(host_data);
 	return ret;
 }
 
 static int __init stm32f4_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
-	return stm32_exti_init(stm32f4xx_exti_banks,
-			ARRAY_SIZE(stm32f4xx_exti_banks), np);
+	return stm32_exti_init(&stm32f4xx_drv_data, np);
 }
 
 IRQCHIP_DECLARE(stm32f4_exti, "st,stm32-exti", stm32f4_exti_of_init);
@@ -350,8 +407,7 @@ IRQCHIP_DECLARE(stm32f4_exti, "st,stm32-exti", stm32f4_exti_of_init);
 static int __init stm32h7_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
-	return stm32_exti_init(stm32h7xx_exti_banks,
-			ARRAY_SIZE(stm32h7xx_exti_banks), np);
+	return stm32_exti_init(&stm32h7xx_drv_data, np);
 }
 
 IRQCHIP_DECLARE(stm32h7_exti, "st,stm32h7-exti", stm32h7_exti_of_init);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 05/11] irqchip: stm32: add host and driver data structures
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds host and driver data structures to support
different stm32 exti controllers with variants.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 152 ++++++++++++++++++++++++++-------------
 1 file changed, 104 insertions(+), 48 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 1e09667..9655a57 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -29,13 +29,23 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_exti_drv_data {
+	const struct stm32_exti_bank **exti_banks;
+	u32 bank_nr;
+};
+
 struct stm32_exti_chip_data {
+	struct stm32_exti_host_data *host_data;
 	const struct stm32_exti_bank *reg_bank;
 	u32 rtsr_cache;
 	u32 ftsr_cache;
 };
 
-static struct stm32_exti_chip_data *stm32_exti_data;
+struct stm32_exti_host_data {
+	void __iomem *base;
+	struct stm32_exti_chip_data *chips_data;
+	const struct stm32_exti_drv_data *drv_data;
+};
 
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
@@ -51,6 +61,11 @@ static const struct stm32_exti_bank *stm32f4xx_exti_banks[] = {
 	&stm32f4xx_exti_b1,
 };
 
+static const struct stm32_exti_drv_data stm32f4xx_drv_data = {
+	.exti_banks = stm32f4xx_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32f4xx_exti_banks),
+};
+
 static const struct stm32_exti_bank stm32h7xx_exti_b1 = {
 	.imr_ofst	= 0x80,
 	.emr_ofst	= 0x84,
@@ -87,6 +102,11 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 	&stm32h7xx_exti_b3,
 };
 
+static const struct stm32_exti_drv_data stm32h7xx_drv_data = {
+	.exti_banks = stm32h7xx_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32h7xx_exti_banks),
+};
+
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
@@ -237,29 +257,85 @@ static void stm32_irq_ack(struct irq_data *d)
 
 	irq_gc_unlock(gc);
 }
+static struct
+stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
+					   struct device_node *node)
+{
+	struct stm32_exti_host_data *host_data;
+
+	host_data = kzalloc(sizeof(*host_data), GFP_KERNEL);
+	if (!host_data)
+		return NULL;
+
+	host_data->drv_data = dd;
+	host_data->chips_data = kcalloc(dd->bank_nr,
+					sizeof(struct stm32_exti_chip_data),
+					GFP_KERNEL);
+	if (!host_data->chips_data)
+		return NULL;
+
+	host_data->base = of_iomap(node, 0);
+	if (!host_data->base) {
+		pr_err("%pOF: Unable to map registers\n", node);
+		return NULL;
+	}
 
-static int
-__init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
-		       int bank_nr, struct device_node *node)
+	return host_data;
+}
+
+static struct
+stm32_exti_chip_data *stm32_exti_chip_init(struct stm32_exti_host_data *h_data,
+					   u32 bank_idx,
+					   struct device_node *node)
+{
+	const struct stm32_exti_bank *stm32_bank;
+	struct stm32_exti_chip_data *chip_data;
+	void __iomem *base = h_data->base;
+	u32 irqs_mask;
+
+	stm32_bank = h_data->drv_data->exti_banks[bank_idx];
+	chip_data = &h_data->chips_data[bank_idx];
+	chip_data->host_data = h_data;
+	chip_data->reg_bank = stm32_bank;
+
+	/* Determine number of irqs supported */
+	writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
+	irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
+
+	/*
+	 * This IP has no reset, so after hot reboot we should
+	 * clear registers to avoid residue
+	 */
+	writel_relaxed(0, base + stm32_bank->imr_ofst);
+	writel_relaxed(0, base + stm32_bank->emr_ofst);
+	writel_relaxed(0, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(0, base + stm32_bank->ftsr_ofst);
+	writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
+
+	pr_info("%s: bank%d, External IRQs available:%#x\n",
+		node->full_name, bank_idx, irqs_mask);
+
+	return chip_data;
+}
+
+static int __init stm32_exti_init(const struct stm32_exti_drv_data *drv_data,
+				  struct device_node *node)
 {
+	struct stm32_exti_host_data *host_data;
 	unsigned int clr = IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN;
-	int nr_irqs, nr_exti, ret, i;
+	int nr_irqs, ret, i;
 	struct irq_chip_generic *gc;
 	struct irq_domain *domain;
-	void *base;
 
-	base = of_iomap(node, 0);
-	if (!base) {
-		pr_err("%pOF: Unable to map registers\n", node);
-		return -ENOMEM;
+	host_data = stm32_exti_host_init(drv_data, node);
+	if (!host_data) {
+		ret = -ENOMEM;
+		goto out_free_mem;
 	}
 
-	stm32_exti_data = kcalloc(bank_nr, sizeof(*stm32_exti_data),
-				  GFP_KERNEL);
-	if (!stm32_exti_data)
-		return -ENOMEM;
-
-	domain = irq_domain_add_linear(node, bank_nr * IRQS_PER_BANK,
+	domain = irq_domain_add_linear(node, drv_data->bank_nr * IRQS_PER_BANK,
 				       &irq_exti_domain_ops, NULL);
 	if (!domain) {
 		pr_err("%s: Could not register interrupt domain.\n",
@@ -276,16 +352,16 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		goto out_free_domain;
 	}
 
-	for (i = 0; i < bank_nr; i++) {
-		const struct stm32_exti_bank *stm32_bank = stm32_exti_banks[i];
-		struct stm32_exti_chip_data *chip_data = &stm32_exti_data[i];
-		u32 irqs_mask;
+	for (i = 0; i < drv_data->bank_nr; i++) {
+		const struct stm32_exti_bank *stm32_bank;
+		struct stm32_exti_chip_data *chip_data;
 
-		chip_data->reg_bank = stm32_bank;
+		stm32_bank = drv_data->exti_banks[i];
+		chip_data = stm32_exti_chip_init(host_data, i, node);
 
 		gc = irq_get_domain_generic_chip(domain, i * IRQS_PER_BANK);
 
-		gc->reg_base = base;
+		gc->reg_base = host_data->base;
 		gc->chip_types->type = IRQ_TYPE_EDGE_BOTH;
 		gc->chip_types->chip.irq_ack = stm32_irq_ack;
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
@@ -298,26 +374,6 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
 		gc->private = (void *)chip_data;
-
-		/* Determine number of irqs supported */
-		writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
-		irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
-		nr_exti = fls(readl_relaxed(base + stm32_bank->rtsr_ofst));
-
-		/*
-		 * This IP has no reset, so after hot reboot we should
-		 * clear registers to avoid residue
-		 */
-		writel_relaxed(0, base + stm32_bank->imr_ofst);
-		writel_relaxed(0, base + stm32_bank->emr_ofst);
-		writel_relaxed(0, base + stm32_bank->rtsr_ofst);
-		writel_relaxed(0, base + stm32_bank->ftsr_ofst);
-		writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
-		if (stm32_bank->fpr_ofst != UNDEF_REG)
-			writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
-
-		pr_info("%s: bank%d, External IRQs available:%#x\n",
-			node->full_name, i, irqs_mask);
 	}
 
 	nr_irqs = of_irq_count(node);
@@ -333,16 +389,17 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 out_free_domain:
 	irq_domain_remove(domain);
 out_unmap:
-	iounmap(base);
-	kfree(stm32_exti_data);
+	iounmap(host_data->base);
+out_free_mem:
+	kfree(host_data->chips_data);
+	kfree(host_data);
 	return ret;
 }
 
 static int __init stm32f4_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
-	return stm32_exti_init(stm32f4xx_exti_banks,
-			ARRAY_SIZE(stm32f4xx_exti_banks), np);
+	return stm32_exti_init(&stm32f4xx_drv_data, np);
 }
 
 IRQCHIP_DECLARE(stm32f4_exti, "st,stm32-exti", stm32f4_exti_of_init);
@@ -350,8 +407,7 @@ IRQCHIP_DECLARE(stm32f4_exti, "st,stm32-exti", stm32f4_exti_of_init);
 static int __init stm32h7_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
-	return stm32_exti_init(stm32h7xx_exti_banks,
-			ARRAY_SIZE(stm32h7xx_exti_banks), np);
+	return stm32_exti_init(&stm32h7xx_drv_data, np);
 }
 
 IRQCHIP_DECLARE(stm32h7_exti, "st,stm32h7-exti", stm32h7_exti_of_init);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 05/11] irqchip: stm32: add host and driver data structures
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds host and driver data structures to support
different stm32 exti controllers with variants.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 152 ++++++++++++++++++++++++++-------------
 1 file changed, 104 insertions(+), 48 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 1e09667..9655a57 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -29,13 +29,23 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_exti_drv_data {
+	const struct stm32_exti_bank **exti_banks;
+	u32 bank_nr;
+};
+
 struct stm32_exti_chip_data {
+	struct stm32_exti_host_data *host_data;
 	const struct stm32_exti_bank *reg_bank;
 	u32 rtsr_cache;
 	u32 ftsr_cache;
 };
 
-static struct stm32_exti_chip_data *stm32_exti_data;
+struct stm32_exti_host_data {
+	void __iomem *base;
+	struct stm32_exti_chip_data *chips_data;
+	const struct stm32_exti_drv_data *drv_data;
+};
 
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
@@ -51,6 +61,11 @@ static const struct stm32_exti_bank *stm32f4xx_exti_banks[] = {
 	&stm32f4xx_exti_b1,
 };
 
+static const struct stm32_exti_drv_data stm32f4xx_drv_data = {
+	.exti_banks = stm32f4xx_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32f4xx_exti_banks),
+};
+
 static const struct stm32_exti_bank stm32h7xx_exti_b1 = {
 	.imr_ofst	= 0x80,
 	.emr_ofst	= 0x84,
@@ -87,6 +102,11 @@ static const struct stm32_exti_bank *stm32h7xx_exti_banks[] = {
 	&stm32h7xx_exti_b3,
 };
 
+static const struct stm32_exti_drv_data stm32h7xx_drv_data = {
+	.exti_banks = stm32h7xx_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32h7xx_exti_banks),
+};
+
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
@@ -237,29 +257,85 @@ static void stm32_irq_ack(struct irq_data *d)
 
 	irq_gc_unlock(gc);
 }
+static struct
+stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
+					   struct device_node *node)
+{
+	struct stm32_exti_host_data *host_data;
+
+	host_data = kzalloc(sizeof(*host_data), GFP_KERNEL);
+	if (!host_data)
+		return NULL;
+
+	host_data->drv_data = dd;
+	host_data->chips_data = kcalloc(dd->bank_nr,
+					sizeof(struct stm32_exti_chip_data),
+					GFP_KERNEL);
+	if (!host_data->chips_data)
+		return NULL;
+
+	host_data->base = of_iomap(node, 0);
+	if (!host_data->base) {
+		pr_err("%pOF: Unable to map registers\n", node);
+		return NULL;
+	}
 
-static int
-__init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
-		       int bank_nr, struct device_node *node)
+	return host_data;
+}
+
+static struct
+stm32_exti_chip_data *stm32_exti_chip_init(struct stm32_exti_host_data *h_data,
+					   u32 bank_idx,
+					   struct device_node *node)
+{
+	const struct stm32_exti_bank *stm32_bank;
+	struct stm32_exti_chip_data *chip_data;
+	void __iomem *base = h_data->base;
+	u32 irqs_mask;
+
+	stm32_bank = h_data->drv_data->exti_banks[bank_idx];
+	chip_data = &h_data->chips_data[bank_idx];
+	chip_data->host_data = h_data;
+	chip_data->reg_bank = stm32_bank;
+
+	/* Determine number of irqs supported */
+	writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
+	irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
+
+	/*
+	 * This IP has no reset, so after hot reboot we should
+	 * clear registers to avoid residue
+	 */
+	writel_relaxed(0, base + stm32_bank->imr_ofst);
+	writel_relaxed(0, base + stm32_bank->emr_ofst);
+	writel_relaxed(0, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(0, base + stm32_bank->ftsr_ofst);
+	writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
+
+	pr_info("%s: bank%d, External IRQs available:%#x\n",
+		node->full_name, bank_idx, irqs_mask);
+
+	return chip_data;
+}
+
+static int __init stm32_exti_init(const struct stm32_exti_drv_data *drv_data,
+				  struct device_node *node)
 {
+	struct stm32_exti_host_data *host_data;
 	unsigned int clr = IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN;
-	int nr_irqs, nr_exti, ret, i;
+	int nr_irqs, ret, i;
 	struct irq_chip_generic *gc;
 	struct irq_domain *domain;
-	void *base;
 
-	base = of_iomap(node, 0);
-	if (!base) {
-		pr_err("%pOF: Unable to map registers\n", node);
-		return -ENOMEM;
+	host_data = stm32_exti_host_init(drv_data, node);
+	if (!host_data) {
+		ret = -ENOMEM;
+		goto out_free_mem;
 	}
 
-	stm32_exti_data = kcalloc(bank_nr, sizeof(*stm32_exti_data),
-				  GFP_KERNEL);
-	if (!stm32_exti_data)
-		return -ENOMEM;
-
-	domain = irq_domain_add_linear(node, bank_nr * IRQS_PER_BANK,
+	domain = irq_domain_add_linear(node, drv_data->bank_nr * IRQS_PER_BANK,
 				       &irq_exti_domain_ops, NULL);
 	if (!domain) {
 		pr_err("%s: Could not register interrupt domain.\n",
@@ -276,16 +352,16 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 		goto out_free_domain;
 	}
 
-	for (i = 0; i < bank_nr; i++) {
-		const struct stm32_exti_bank *stm32_bank = stm32_exti_banks[i];
-		struct stm32_exti_chip_data *chip_data = &stm32_exti_data[i];
-		u32 irqs_mask;
+	for (i = 0; i < drv_data->bank_nr; i++) {
+		const struct stm32_exti_bank *stm32_bank;
+		struct stm32_exti_chip_data *chip_data;
 
-		chip_data->reg_bank = stm32_bank;
+		stm32_bank = drv_data->exti_banks[i];
+		chip_data = stm32_exti_chip_init(host_data, i, node);
 
 		gc = irq_get_domain_generic_chip(domain, i * IRQS_PER_BANK);
 
-		gc->reg_base = base;
+		gc->reg_base = host_data->base;
 		gc->chip_types->type = IRQ_TYPE_EDGE_BOTH;
 		gc->chip_types->chip.irq_ack = stm32_irq_ack;
 		gc->chip_types->chip.irq_mask = irq_gc_mask_clr_bit;
@@ -298,26 +374,6 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 
 		gc->chip_types->regs.mask = stm32_bank->imr_ofst;
 		gc->private = (void *)chip_data;
-
-		/* Determine number of irqs supported */
-		writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
-		irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
-		nr_exti = fls(readl_relaxed(base + stm32_bank->rtsr_ofst));
-
-		/*
-		 * This IP has no reset, so after hot reboot we should
-		 * clear registers to avoid residue
-		 */
-		writel_relaxed(0, base + stm32_bank->imr_ofst);
-		writel_relaxed(0, base + stm32_bank->emr_ofst);
-		writel_relaxed(0, base + stm32_bank->rtsr_ofst);
-		writel_relaxed(0, base + stm32_bank->ftsr_ofst);
-		writel_relaxed(~0UL, base + stm32_bank->rpr_ofst);
-		if (stm32_bank->fpr_ofst != UNDEF_REG)
-			writel_relaxed(~0UL, base + stm32_bank->fpr_ofst);
-
-		pr_info("%s: bank%d, External IRQs available:%#x\n",
-			node->full_name, i, irqs_mask);
 	}
 
 	nr_irqs = of_irq_count(node);
@@ -333,16 +389,17 @@ __init stm32_exti_init(const struct stm32_exti_bank **stm32_exti_banks,
 out_free_domain:
 	irq_domain_remove(domain);
 out_unmap:
-	iounmap(base);
-	kfree(stm32_exti_data);
+	iounmap(host_data->base);
+out_free_mem:
+	kfree(host_data->chips_data);
+	kfree(host_data);
 	return ret;
 }
 
 static int __init stm32f4_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
-	return stm32_exti_init(stm32f4xx_exti_banks,
-			ARRAY_SIZE(stm32f4xx_exti_banks), np);
+	return stm32_exti_init(&stm32f4xx_drv_data, np);
 }
 
 IRQCHIP_DECLARE(stm32f4_exti, "st,stm32-exti", stm32f4_exti_of_init);
@@ -350,8 +407,7 @@ IRQCHIP_DECLARE(stm32f4_exti, "st,stm32-exti", stm32f4_exti_of_init);
 static int __init stm32h7_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
-	return stm32_exti_init(stm32h7xx_exti_banks,
-			ARRAY_SIZE(stm32h7xx_exti_banks), np);
+	return stm32_exti_init(&stm32h7xx_drv_data, np);
 }
 
 IRQCHIP_DECLARE(stm32h7_exti, "st,stm32h7-exti", stm32h7_exti_of_init);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 06/11] irqchip: stm32: prepare common functions
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch prepares functions which could be reused by
next variant of stm32 exti controller.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 91 +++++++++++++++++++++++++---------------
 1 file changed, 58 insertions(+), 33 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 9655a57..b38c655 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -145,37 +145,50 @@ static void stm32_irq_handler(struct irq_desc *desc)
 	chained_irq_exit(chip, desc);
 }
 
-static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
+static int stm32_exti_set_type(struct irq_data *d,
+			       unsigned int type, u32 *rtsr, u32 *ftsr)
 {
-	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	struct stm32_exti_chip_data *chip_data = gc->private;
-	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
-	int pin = data->hwirq % IRQS_PER_BANK;
-	u32 rtsr, ftsr;
-
-	irq_gc_lock(gc);
-
-	rtsr = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
-	ftsr = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+	u32 mask = BIT(d->hwirq % IRQS_PER_BANK);
 
 	switch (type) {
 	case IRQ_TYPE_EDGE_RISING:
-		rtsr |= BIT(pin);
-		ftsr &= ~BIT(pin);
+		*rtsr |= mask;
+		*ftsr &= ~mask;
 		break;
 	case IRQ_TYPE_EDGE_FALLING:
-		rtsr &= ~BIT(pin);
-		ftsr |= BIT(pin);
+		*rtsr &= ~mask;
+		*ftsr |= mask;
 		break;
 	case IRQ_TYPE_EDGE_BOTH:
-		rtsr |= BIT(pin);
-		ftsr |= BIT(pin);
+		*rtsr |= mask;
+		*ftsr |= mask;
 		break;
 	default:
-		irq_gc_unlock(gc);
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+static int stm32_irq_set_type(struct irq_data *d, unsigned int type)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	u32 rtsr, ftsr;
+	int err;
+
+	irq_gc_lock(gc);
+
+	rtsr = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
+	ftsr = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+
+	err = stm32_exti_set_type(d, type, &rtsr, &ftsr);
+	if (err) {
+		irq_gc_unlock(gc);
+		return err;
+	}
+
 	irq_reg_writel(gc, rtsr, stm32_bank->rtsr_ofst);
 	irq_reg_writel(gc, ftsr, stm32_bank->ftsr_ofst);
 
@@ -184,35 +197,47 @@ static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
-static void stm32_irq_suspend(struct irq_chip_generic *gc)
+static void stm32_chip_suspend(struct stm32_exti_chip_data *chip_data,
+			       u32 wake_active)
 {
-	struct stm32_exti_chip_data *chip_data = gc->private;
 	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
-
-	irq_gc_lock(gc);
+	void __iomem *base = chip_data->host_data->base;
 
 	/* save rtsr, ftsr registers */
-	chip_data->rtsr_cache = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
-	chip_data->ftsr_cache = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+	chip_data->rtsr_cache = readl_relaxed(base + stm32_bank->rtsr_ofst);
+	chip_data->ftsr_cache = readl_relaxed(base + stm32_bank->ftsr_ofst);
 
-	irq_reg_writel(gc, gc->wake_active, stm32_bank->imr_ofst);
+	writel_relaxed(wake_active, base + stm32_bank->imr_ofst);
+}
 
-	irq_gc_unlock(gc);
+static void stm32_chip_resume(struct stm32_exti_chip_data *chip_data,
+			      u32 mask_cache)
+{
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	void __iomem *base = chip_data->host_data->base;
+
+	/* restore rtsr, ftsr, registers */
+	writel_relaxed(chip_data->rtsr_cache, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(chip_data->ftsr_cache, base + stm32_bank->ftsr_ofst);
+
+	writel_relaxed(mask_cache, base + stm32_bank->imr_ofst);
 }
 
-static void stm32_irq_resume(struct irq_chip_generic *gc)
+static void stm32_irq_suspend(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
-	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
+	stm32_chip_suspend(chip_data, gc->wake_active);
+	irq_gc_unlock(gc);
+}
 
-	/* restore rtsr, ftsr registers */
-	irq_reg_writel(gc, chip_data->rtsr_cache, stm32_bank->rtsr_ofst);
-	irq_reg_writel(gc, chip_data->ftsr_cache, stm32_bank->ftsr_ofst);
-
-	irq_reg_writel(gc, gc->mask_cache, stm32_bank->imr_ofst);
+static void stm32_irq_resume(struct irq_chip_generic *gc)
+{
+	struct stm32_exti_chip_data *chip_data = gc->private;
 
+	irq_gc_lock(gc);
+	stm32_chip_resume(chip_data, gc->mask_cache);
 	irq_gc_unlock(gc);
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 06/11] irqchip: stm32: prepare common functions
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch prepares functions which could be reused by
next variant of stm32 exti controller.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 91 +++++++++++++++++++++++++---------------
 1 file changed, 58 insertions(+), 33 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 9655a57..b38c655 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -145,37 +145,50 @@ static void stm32_irq_handler(struct irq_desc *desc)
 	chained_irq_exit(chip, desc);
 }
 
-static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
+static int stm32_exti_set_type(struct irq_data *d,
+			       unsigned int type, u32 *rtsr, u32 *ftsr)
 {
-	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	struct stm32_exti_chip_data *chip_data = gc->private;
-	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
-	int pin = data->hwirq % IRQS_PER_BANK;
-	u32 rtsr, ftsr;
-
-	irq_gc_lock(gc);
-
-	rtsr = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
-	ftsr = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+	u32 mask = BIT(d->hwirq % IRQS_PER_BANK);
 
 	switch (type) {
 	case IRQ_TYPE_EDGE_RISING:
-		rtsr |= BIT(pin);
-		ftsr &= ~BIT(pin);
+		*rtsr |= mask;
+		*ftsr &= ~mask;
 		break;
 	case IRQ_TYPE_EDGE_FALLING:
-		rtsr &= ~BIT(pin);
-		ftsr |= BIT(pin);
+		*rtsr &= ~mask;
+		*ftsr |= mask;
 		break;
 	case IRQ_TYPE_EDGE_BOTH:
-		rtsr |= BIT(pin);
-		ftsr |= BIT(pin);
+		*rtsr |= mask;
+		*ftsr |= mask;
 		break;
 	default:
-		irq_gc_unlock(gc);
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+static int stm32_irq_set_type(struct irq_data *d, unsigned int type)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	u32 rtsr, ftsr;
+	int err;
+
+	irq_gc_lock(gc);
+
+	rtsr = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
+	ftsr = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+
+	err = stm32_exti_set_type(d, type, &rtsr, &ftsr);
+	if (err) {
+		irq_gc_unlock(gc);
+		return err;
+	}
+
 	irq_reg_writel(gc, rtsr, stm32_bank->rtsr_ofst);
 	irq_reg_writel(gc, ftsr, stm32_bank->ftsr_ofst);
 
@@ -184,35 +197,47 @@ static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
-static void stm32_irq_suspend(struct irq_chip_generic *gc)
+static void stm32_chip_suspend(struct stm32_exti_chip_data *chip_data,
+			       u32 wake_active)
 {
-	struct stm32_exti_chip_data *chip_data = gc->private;
 	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
-
-	irq_gc_lock(gc);
+	void __iomem *base = chip_data->host_data->base;
 
 	/* save rtsr, ftsr registers */
-	chip_data->rtsr_cache = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
-	chip_data->ftsr_cache = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+	chip_data->rtsr_cache = readl_relaxed(base + stm32_bank->rtsr_ofst);
+	chip_data->ftsr_cache = readl_relaxed(base + stm32_bank->ftsr_ofst);
 
-	irq_reg_writel(gc, gc->wake_active, stm32_bank->imr_ofst);
+	writel_relaxed(wake_active, base + stm32_bank->imr_ofst);
+}
 
-	irq_gc_unlock(gc);
+static void stm32_chip_resume(struct stm32_exti_chip_data *chip_data,
+			      u32 mask_cache)
+{
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	void __iomem *base = chip_data->host_data->base;
+
+	/* restore rtsr, ftsr, registers */
+	writel_relaxed(chip_data->rtsr_cache, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(chip_data->ftsr_cache, base + stm32_bank->ftsr_ofst);
+
+	writel_relaxed(mask_cache, base + stm32_bank->imr_ofst);
 }
 
-static void stm32_irq_resume(struct irq_chip_generic *gc)
+static void stm32_irq_suspend(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
-	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
+	stm32_chip_suspend(chip_data, gc->wake_active);
+	irq_gc_unlock(gc);
+}
 
-	/* restore rtsr, ftsr registers */
-	irq_reg_writel(gc, chip_data->rtsr_cache, stm32_bank->rtsr_ofst);
-	irq_reg_writel(gc, chip_data->ftsr_cache, stm32_bank->ftsr_ofst);
-
-	irq_reg_writel(gc, gc->mask_cache, stm32_bank->imr_ofst);
+static void stm32_irq_resume(struct irq_chip_generic *gc)
+{
+	struct stm32_exti_chip_data *chip_data = gc->private;
 
+	irq_gc_lock(gc);
+	stm32_chip_resume(chip_data, gc->mask_cache);
 	irq_gc_unlock(gc);
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 06/11] irqchip: stm32: prepare common functions
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch prepares functions which could be reused by
next variant of stm32 exti controller.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 91 +++++++++++++++++++++++++---------------
 1 file changed, 58 insertions(+), 33 deletions(-)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index 9655a57..b38c655 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -145,37 +145,50 @@ static void stm32_irq_handler(struct irq_desc *desc)
 	chained_irq_exit(chip, desc);
 }
 
-static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
+static int stm32_exti_set_type(struct irq_data *d,
+			       unsigned int type, u32 *rtsr, u32 *ftsr)
 {
-	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(data);
-	struct stm32_exti_chip_data *chip_data = gc->private;
-	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
-	int pin = data->hwirq % IRQS_PER_BANK;
-	u32 rtsr, ftsr;
-
-	irq_gc_lock(gc);
-
-	rtsr = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
-	ftsr = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+	u32 mask = BIT(d->hwirq % IRQS_PER_BANK);
 
 	switch (type) {
 	case IRQ_TYPE_EDGE_RISING:
-		rtsr |= BIT(pin);
-		ftsr &= ~BIT(pin);
+		*rtsr |= mask;
+		*ftsr &= ~mask;
 		break;
 	case IRQ_TYPE_EDGE_FALLING:
-		rtsr &= ~BIT(pin);
-		ftsr |= BIT(pin);
+		*rtsr &= ~mask;
+		*ftsr |= mask;
 		break;
 	case IRQ_TYPE_EDGE_BOTH:
-		rtsr |= BIT(pin);
-		ftsr |= BIT(pin);
+		*rtsr |= mask;
+		*ftsr |= mask;
 		break;
 	default:
-		irq_gc_unlock(gc);
 		return -EINVAL;
 	}
 
+	return 0;
+}
+
+static int stm32_irq_set_type(struct irq_data *d, unsigned int type)
+{
+	struct irq_chip_generic *gc = irq_data_get_irq_chip_data(d);
+	struct stm32_exti_chip_data *chip_data = gc->private;
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	u32 rtsr, ftsr;
+	int err;
+
+	irq_gc_lock(gc);
+
+	rtsr = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
+	ftsr = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+
+	err = stm32_exti_set_type(d, type, &rtsr, &ftsr);
+	if (err) {
+		irq_gc_unlock(gc);
+		return err;
+	}
+
 	irq_reg_writel(gc, rtsr, stm32_bank->rtsr_ofst);
 	irq_reg_writel(gc, ftsr, stm32_bank->ftsr_ofst);
 
@@ -184,35 +197,47 @@ static int stm32_irq_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
-static void stm32_irq_suspend(struct irq_chip_generic *gc)
+static void stm32_chip_suspend(struct stm32_exti_chip_data *chip_data,
+			       u32 wake_active)
 {
-	struct stm32_exti_chip_data *chip_data = gc->private;
 	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
-
-	irq_gc_lock(gc);
+	void __iomem *base = chip_data->host_data->base;
 
 	/* save rtsr, ftsr registers */
-	chip_data->rtsr_cache = irq_reg_readl(gc, stm32_bank->rtsr_ofst);
-	chip_data->ftsr_cache = irq_reg_readl(gc, stm32_bank->ftsr_ofst);
+	chip_data->rtsr_cache = readl_relaxed(base + stm32_bank->rtsr_ofst);
+	chip_data->ftsr_cache = readl_relaxed(base + stm32_bank->ftsr_ofst);
 
-	irq_reg_writel(gc, gc->wake_active, stm32_bank->imr_ofst);
+	writel_relaxed(wake_active, base + stm32_bank->imr_ofst);
+}
 
-	irq_gc_unlock(gc);
+static void stm32_chip_resume(struct stm32_exti_chip_data *chip_data,
+			      u32 mask_cache)
+{
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	void __iomem *base = chip_data->host_data->base;
+
+	/* restore rtsr, ftsr, registers */
+	writel_relaxed(chip_data->rtsr_cache, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(chip_data->ftsr_cache, base + stm32_bank->ftsr_ofst);
+
+	writel_relaxed(mask_cache, base + stm32_bank->imr_ofst);
 }
 
-static void stm32_irq_resume(struct irq_chip_generic *gc)
+static void stm32_irq_suspend(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
-	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
 
 	irq_gc_lock(gc);
+	stm32_chip_suspend(chip_data, gc->wake_active);
+	irq_gc_unlock(gc);
+}
 
-	/* restore rtsr, ftsr registers */
-	irq_reg_writel(gc, chip_data->rtsr_cache, stm32_bank->rtsr_ofst);
-	irq_reg_writel(gc, chip_data->ftsr_cache, stm32_bank->ftsr_ofst);
-
-	irq_reg_writel(gc, gc->mask_cache, stm32_bank->imr_ofst);
+static void stm32_irq_resume(struct irq_chip_generic *gc)
+{
+	struct stm32_exti_chip_data *chip_data = gc->private;
 
+	irq_gc_lock(gc);
+	stm32_chip_resume(chip_data, gc->mask_cache);
 	irq_gc_unlock(gc);
 }
 
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

Exti controller has been differently integrated on stm32mp1 SoC.
A parent irq has only one external interrupt. A hierachy domain could
be used. Handlers are call by parent, each parent interrupt could be
masked and unmasked according to the needs.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 .../interrupt-controller/st,stm32-exti.txt         |   3 +
 drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
 2 files changed, 325 insertions(+)

diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
index edf03f0..136bd61 100644
--- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
+++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
@@ -5,11 +5,14 @@ Required properties:
 - compatible: Should be:
     "st,stm32-exti"
     "st,stm32h7-exti"
+    "st,stm32mp1-exti"
 - reg: Specifies base physical address and size of the registers
 - interrupt-controller: Indentifies the node as an interrupt controller
 - #interrupt-cells: Specifies the number of cells to encode an interrupt
   specifier, shall be 2
 - interrupts: interrupts references to primary interrupt controller
+  (only needed for exti controller with multiple exti under
+  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
 
 Example:
 
diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index b38c655..ebf7146 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -15,6 +15,8 @@
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 
+#include <dt-bindings/interrupt-controller/arm-gic.h>
+
 #define IRQS_PER_BANK 32
 
 struct stm32_exti_bank {
@@ -29,14 +31,24 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_desc_irq {
+	u32 exti;
+	u32 irq_parent;
+};
+
 struct stm32_exti_drv_data {
 	const struct stm32_exti_bank **exti_banks;
+	const struct stm32_desc_irq *desc_irqs;
 	u32 bank_nr;
+	u32 irq_nr;
 };
 
 struct stm32_exti_chip_data {
 	struct stm32_exti_host_data *host_data;
 	const struct stm32_exti_bank *reg_bank;
+	struct raw_spinlock rlock;
+	u32 wake_active;
+	u32 mask_cache;
 	u32 rtsr_cache;
 	u32 ftsr_cache;
 };
@@ -107,6 +119,89 @@ static const struct stm32_exti_drv_data stm32h7xx_drv_data = {
 	.bank_nr = ARRAY_SIZE(stm32h7xx_exti_banks),
 };
 
+static const struct stm32_exti_bank stm32mp1_exti_b1 = {
+	.imr_ofst	= 0x80,
+	.emr_ofst	= 0x84,
+	.rtsr_ofst	= 0x00,
+	.ftsr_ofst	= 0x04,
+	.swier_ofst	= 0x08,
+	.rpr_ofst	= 0x0C,
+	.fpr_ofst	= 0x10,
+};
+
+static const struct stm32_exti_bank stm32mp1_exti_b2 = {
+	.imr_ofst	= 0x90,
+	.emr_ofst	= 0x94,
+	.rtsr_ofst	= 0x20,
+	.ftsr_ofst	= 0x24,
+	.swier_ofst	= 0x28,
+	.rpr_ofst	= 0x2C,
+	.fpr_ofst	= 0x30,
+};
+
+static const struct stm32_exti_bank stm32mp1_exti_b3 = {
+	.imr_ofst	= 0xA0,
+	.emr_ofst	= 0xA4,
+	.rtsr_ofst	= 0x40,
+	.ftsr_ofst	= 0x44,
+	.swier_ofst	= 0x48,
+	.rpr_ofst	= 0x4C,
+	.fpr_ofst	= 0x50,
+};
+
+static const struct stm32_exti_bank *stm32mp1_exti_banks[] = {
+	&stm32mp1_exti_b1,
+	&stm32mp1_exti_b2,
+	&stm32mp1_exti_b3,
+};
+
+static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
+	{ .exti = 1, .irq_parent = 7 },
+	{ .exti = 2, .irq_parent = 8 },
+	{ .exti = 3, .irq_parent = 9 },
+	{ .exti = 4, .irq_parent = 10 },
+	{ .exti = 5, .irq_parent = 23 },
+	{ .exti = 6, .irq_parent = 64 },
+	{ .exti = 7, .irq_parent = 65 },
+	{ .exti = 8, .irq_parent = 66 },
+	{ .exti = 9, .irq_parent = 67 },
+	{ .exti = 10, .irq_parent = 40 },
+	{ .exti = 11, .irq_parent = 42 },
+	{ .exti = 12, .irq_parent = 76 },
+	{ .exti = 13, .irq_parent = 77 },
+	{ .exti = 14, .irq_parent = 121 },
+	{ .exti = 15, .irq_parent = 127 },
+	{ .exti = 16, .irq_parent = 1 },
+	{ .exti = 65, .irq_parent = 144 },
+	{ .exti = 68, .irq_parent = 143 },
+	{ .exti = 73, .irq_parent = 129 },
+};
+
+static const struct stm32_exti_drv_data stm32mp1_drv_data = {
+	.exti_banks = stm32mp1_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32mp1_exti_banks),
+	.desc_irqs = stm32mp1_desc_irq,
+	.irq_nr = ARRAY_SIZE(stm32mp1_desc_irq),
+};
+
+static int stm32_exti_to_irq(const struct stm32_exti_drv_data *drv_data,
+			     irq_hw_number_t hwirq)
+{
+	const struct stm32_desc_irq *desc_irq;
+	int i;
+
+	if (!drv_data->desc_irqs)
+		return -EINVAL;
+
+	for (i = 0; i < drv_data->irq_nr; i++) {
+		desc_irq = &drv_data->desc_irqs[i];
+		if (desc_irq->exti == hwirq)
+			return desc_irq->irq_parent;
+	}
+
+	return -EINVAL;
+}
+
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
@@ -282,6 +377,173 @@ static void stm32_irq_ack(struct irq_data *d)
 
 	irq_gc_unlock(gc);
 }
+
+static inline u32 stm32_exti_set_bit(struct irq_data *d, u32 reg)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	void __iomem *base = chip_data->host_data->base;
+	u32 val;
+
+	val = readl_relaxed(base + reg);
+	val |= BIT(d->hwirq % IRQS_PER_BANK);
+	writel_relaxed(val, base + reg);
+
+	return val;
+}
+
+static inline u32 stm32_exti_clr_bit(struct irq_data *d, u32 reg)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	void __iomem *base = chip_data->host_data->base;
+	u32 val;
+
+	val = readl_relaxed(base + reg);
+	val &= ~BIT(d->hwirq % IRQS_PER_BANK);
+	writel_relaxed(val, base + reg);
+
+	return val;
+}
+
+static void stm32_exti_h_eoi(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+
+	stm32_exti_set_bit(d, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		stm32_exti_set_bit(d, stm32_bank->fpr_ofst);
+
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_eoi_parent(d);
+}
+
+static void stm32_exti_h_mask(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+	chip_data->mask_cache = stm32_exti_clr_bit(d, stm32_bank->imr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_mask_parent(d);
+}
+
+static void stm32_exti_h_unmask(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+	chip_data->mask_cache = stm32_exti_set_bit(d, stm32_bank->imr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_unmask_parent(d);
+}
+
+static int stm32_exti_h_set_type(struct irq_data *d, unsigned int type)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	void __iomem *base = chip_data->host_data->base;
+	u32 rtsr, ftsr;
+	int err;
+
+	raw_spin_lock(&chip_data->rlock);
+	rtsr = readl_relaxed(base + stm32_bank->rtsr_ofst);
+	ftsr = readl_relaxed(base + stm32_bank->ftsr_ofst);
+
+	err = stm32_exti_set_type(d, type, &rtsr, &ftsr);
+	if (err) {
+		raw_spin_unlock(&chip_data->rlock);
+		return err;
+	}
+
+	writel_relaxed(rtsr, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(ftsr, base + stm32_bank->ftsr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	return 0;
+}
+
+static int stm32_exti_h_set_wake(struct irq_data *d, unsigned int on)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	u32 mask = BIT(d->hwirq % IRQS_PER_BANK);
+
+	raw_spin_lock(&chip_data->rlock);
+
+	if (on)
+		chip_data->wake_active |= mask;
+	else
+		chip_data->wake_active &= ~mask;
+
+	raw_spin_unlock(&chip_data->rlock);
+
+	return 0;
+}
+
+static int stm32_exti_h_set_affinity(struct irq_data *d,
+				     const struct cpumask *dest, bool force)
+{
+	if (d->parent_data->chip)
+		return irq_chip_set_affinity_parent(d, dest, force);
+
+	return -EINVAL;
+}
+
+static struct irq_chip stm32_exti_h_chip = {
+	.name			= "stm32-exti-h",
+	.irq_eoi		= stm32_exti_h_eoi,
+	.irq_mask		= stm32_exti_h_mask,
+	.irq_unmask		= stm32_exti_h_unmask,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_set_type		= stm32_exti_h_set_type,
+	.irq_set_wake		= stm32_exti_h_set_wake,
+	.flags			= IRQCHIP_MASK_ON_SUSPEND,
+#ifdef CONFIG_SMP
+	.irq_set_affinity	= stm32_exti_h_set_affinity,
+#endif
+};
+
+static int stm32_exti_h_domain_alloc(struct irq_domain *dm,
+				     unsigned int virq,
+				     unsigned int nr_irqs, void *data)
+{
+	struct stm32_exti_host_data *host_data = dm->host_data;
+	struct stm32_exti_chip_data *chip_data;
+	struct irq_fwspec *fwspec = data;
+	struct irq_fwspec p_fwspec;
+	irq_hw_number_t hwirq;
+	int p_irq, bank;
+
+	hwirq = fwspec->param[0];
+	bank  = hwirq / IRQS_PER_BANK;
+	chip_data = &host_data->chips_data[bank];
+
+	irq_domain_set_hwirq_and_chip(dm, virq, hwirq,
+				      &stm32_exti_h_chip, chip_data);
+
+	p_irq = stm32_exti_to_irq(host_data->drv_data, hwirq);
+	if (p_irq >= 0) {
+		p_fwspec.fwnode = dm->parent->fwnode;
+		p_fwspec.param_count = 3;
+		p_fwspec.param[0] = GIC_SPI;
+		p_fwspec.param[1] = p_irq;
+		p_fwspec.param[2] = IRQ_TYPE_LEVEL_HIGH;
+
+		return irq_domain_alloc_irqs_parent(dm, virq, 1, &p_fwspec);
+	}
+
+	return 0;
+}
+
 static struct
 stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
 					   struct device_node *node)
@@ -323,6 +585,8 @@ stm32_exti_chip_data *stm32_exti_chip_init(struct stm32_exti_host_data *h_data,
 	chip_data->host_data = h_data;
 	chip_data->reg_bank = stm32_bank;
 
+	raw_spin_lock_init(&chip_data->rlock);
+
 	/* Determine number of irqs supported */
 	writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
 	irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
@@ -421,6 +685,56 @@ static int __init stm32_exti_init(const struct stm32_exti_drv_data *drv_data,
 	return ret;
 }
 
+static const struct irq_domain_ops stm32_exti_h_domain_ops = {
+	.alloc	= stm32_exti_h_domain_alloc,
+	.free	= irq_domain_free_irqs_common,
+};
+
+static int
+__init stm32_exti_hierarchy_init(const struct stm32_exti_drv_data *drv_data,
+				 struct device_node *node,
+				 struct device_node *parent)
+{
+	struct irq_domain *parent_domain, *domain;
+	struct stm32_exti_host_data *host_data;
+	int ret, i;
+
+	parent_domain = irq_find_host(parent);
+	if (!parent_domain) {
+		pr_err("interrupt-parent not found\n");
+		return -EINVAL;
+	}
+
+	host_data = stm32_exti_host_init(drv_data, node);
+	if (!host_data) {
+		ret = -ENOMEM;
+		goto out_free_mem;
+	}
+
+	for (i = 0; i < drv_data->bank_nr; i++)
+		stm32_exti_chip_init(host_data, i, node);
+
+	domain = irq_domain_add_hierarchy(parent_domain, 0,
+					  drv_data->bank_nr * IRQS_PER_BANK,
+					  node, &stm32_exti_h_domain_ops,
+					  host_data);
+
+	if (!domain) {
+		pr_err("%s: Could not register exti domain.\n", node->name);
+		ret = -ENOMEM;
+		goto out_unmap;
+	}
+
+	return 0;
+
+out_unmap:
+	iounmap(host_data->base);
+out_free_mem:
+	kfree(host_data->chips_data);
+	kfree(host_data);
+	return ret;
+}
+
 static int __init stm32f4_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
@@ -436,3 +750,11 @@ static int __init stm32h7_exti_of_init(struct device_node *np,
 }
 
 IRQCHIP_DECLARE(stm32h7_exti, "st,stm32h7-exti", stm32h7_exti_of_init);
+
+static int __init stm32mp1_exti_of_init(struct device_node *np,
+					struct device_node *parent)
+{
+	return stm32_exti_hierarchy_init(&stm32mp1_drv_data, np, parent);
+}
+
+IRQCHIP_DECLARE(stm32mp1_exti, "st,stm32mp1-exti", stm32mp1_exti_of_init);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

Exti controller has been differently integrated on stm32mp1 SoC.
A parent irq has only one external interrupt. A hierachy domain could
be used. Handlers are call by parent, each parent interrupt could be
masked and unmasked according to the needs.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 .../interrupt-controller/st,stm32-exti.txt         |   3 +
 drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
 2 files changed, 325 insertions(+)

diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
index edf03f0..136bd61 100644
--- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
+++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
@@ -5,11 +5,14 @@ Required properties:
 - compatible: Should be:
     "st,stm32-exti"
     "st,stm32h7-exti"
+    "st,stm32mp1-exti"
 - reg: Specifies base physical address and size of the registers
 - interrupt-controller: Indentifies the node as an interrupt controller
 - #interrupt-cells: Specifies the number of cells to encode an interrupt
   specifier, shall be 2
 - interrupts: interrupts references to primary interrupt controller
+  (only needed for exti controller with multiple exti under
+  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
 
 Example:
 
diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index b38c655..ebf7146 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -15,6 +15,8 @@
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 
+#include <dt-bindings/interrupt-controller/arm-gic.h>
+
 #define IRQS_PER_BANK 32
 
 struct stm32_exti_bank {
@@ -29,14 +31,24 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_desc_irq {
+	u32 exti;
+	u32 irq_parent;
+};
+
 struct stm32_exti_drv_data {
 	const struct stm32_exti_bank **exti_banks;
+	const struct stm32_desc_irq *desc_irqs;
 	u32 bank_nr;
+	u32 irq_nr;
 };
 
 struct stm32_exti_chip_data {
 	struct stm32_exti_host_data *host_data;
 	const struct stm32_exti_bank *reg_bank;
+	struct raw_spinlock rlock;
+	u32 wake_active;
+	u32 mask_cache;
 	u32 rtsr_cache;
 	u32 ftsr_cache;
 };
@@ -107,6 +119,89 @@ static const struct stm32_exti_drv_data stm32h7xx_drv_data = {
 	.bank_nr = ARRAY_SIZE(stm32h7xx_exti_banks),
 };
 
+static const struct stm32_exti_bank stm32mp1_exti_b1 = {
+	.imr_ofst	= 0x80,
+	.emr_ofst	= 0x84,
+	.rtsr_ofst	= 0x00,
+	.ftsr_ofst	= 0x04,
+	.swier_ofst	= 0x08,
+	.rpr_ofst	= 0x0C,
+	.fpr_ofst	= 0x10,
+};
+
+static const struct stm32_exti_bank stm32mp1_exti_b2 = {
+	.imr_ofst	= 0x90,
+	.emr_ofst	= 0x94,
+	.rtsr_ofst	= 0x20,
+	.ftsr_ofst	= 0x24,
+	.swier_ofst	= 0x28,
+	.rpr_ofst	= 0x2C,
+	.fpr_ofst	= 0x30,
+};
+
+static const struct stm32_exti_bank stm32mp1_exti_b3 = {
+	.imr_ofst	= 0xA0,
+	.emr_ofst	= 0xA4,
+	.rtsr_ofst	= 0x40,
+	.ftsr_ofst	= 0x44,
+	.swier_ofst	= 0x48,
+	.rpr_ofst	= 0x4C,
+	.fpr_ofst	= 0x50,
+};
+
+static const struct stm32_exti_bank *stm32mp1_exti_banks[] = {
+	&stm32mp1_exti_b1,
+	&stm32mp1_exti_b2,
+	&stm32mp1_exti_b3,
+};
+
+static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
+	{ .exti = 1, .irq_parent = 7 },
+	{ .exti = 2, .irq_parent = 8 },
+	{ .exti = 3, .irq_parent = 9 },
+	{ .exti = 4, .irq_parent = 10 },
+	{ .exti = 5, .irq_parent = 23 },
+	{ .exti = 6, .irq_parent = 64 },
+	{ .exti = 7, .irq_parent = 65 },
+	{ .exti = 8, .irq_parent = 66 },
+	{ .exti = 9, .irq_parent = 67 },
+	{ .exti = 10, .irq_parent = 40 },
+	{ .exti = 11, .irq_parent = 42 },
+	{ .exti = 12, .irq_parent = 76 },
+	{ .exti = 13, .irq_parent = 77 },
+	{ .exti = 14, .irq_parent = 121 },
+	{ .exti = 15, .irq_parent = 127 },
+	{ .exti = 16, .irq_parent = 1 },
+	{ .exti = 65, .irq_parent = 144 },
+	{ .exti = 68, .irq_parent = 143 },
+	{ .exti = 73, .irq_parent = 129 },
+};
+
+static const struct stm32_exti_drv_data stm32mp1_drv_data = {
+	.exti_banks = stm32mp1_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32mp1_exti_banks),
+	.desc_irqs = stm32mp1_desc_irq,
+	.irq_nr = ARRAY_SIZE(stm32mp1_desc_irq),
+};
+
+static int stm32_exti_to_irq(const struct stm32_exti_drv_data *drv_data,
+			     irq_hw_number_t hwirq)
+{
+	const struct stm32_desc_irq *desc_irq;
+	int i;
+
+	if (!drv_data->desc_irqs)
+		return -EINVAL;
+
+	for (i = 0; i < drv_data->irq_nr; i++) {
+		desc_irq = &drv_data->desc_irqs[i];
+		if (desc_irq->exti == hwirq)
+			return desc_irq->irq_parent;
+	}
+
+	return -EINVAL;
+}
+
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
@@ -282,6 +377,173 @@ static void stm32_irq_ack(struct irq_data *d)
 
 	irq_gc_unlock(gc);
 }
+
+static inline u32 stm32_exti_set_bit(struct irq_data *d, u32 reg)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	void __iomem *base = chip_data->host_data->base;
+	u32 val;
+
+	val = readl_relaxed(base + reg);
+	val |= BIT(d->hwirq % IRQS_PER_BANK);
+	writel_relaxed(val, base + reg);
+
+	return val;
+}
+
+static inline u32 stm32_exti_clr_bit(struct irq_data *d, u32 reg)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	void __iomem *base = chip_data->host_data->base;
+	u32 val;
+
+	val = readl_relaxed(base + reg);
+	val &= ~BIT(d->hwirq % IRQS_PER_BANK);
+	writel_relaxed(val, base + reg);
+
+	return val;
+}
+
+static void stm32_exti_h_eoi(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+
+	stm32_exti_set_bit(d, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		stm32_exti_set_bit(d, stm32_bank->fpr_ofst);
+
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_eoi_parent(d);
+}
+
+static void stm32_exti_h_mask(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+	chip_data->mask_cache = stm32_exti_clr_bit(d, stm32_bank->imr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_mask_parent(d);
+}
+
+static void stm32_exti_h_unmask(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+	chip_data->mask_cache = stm32_exti_set_bit(d, stm32_bank->imr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_unmask_parent(d);
+}
+
+static int stm32_exti_h_set_type(struct irq_data *d, unsigned int type)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	void __iomem *base = chip_data->host_data->base;
+	u32 rtsr, ftsr;
+	int err;
+
+	raw_spin_lock(&chip_data->rlock);
+	rtsr = readl_relaxed(base + stm32_bank->rtsr_ofst);
+	ftsr = readl_relaxed(base + stm32_bank->ftsr_ofst);
+
+	err = stm32_exti_set_type(d, type, &rtsr, &ftsr);
+	if (err) {
+		raw_spin_unlock(&chip_data->rlock);
+		return err;
+	}
+
+	writel_relaxed(rtsr, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(ftsr, base + stm32_bank->ftsr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	return 0;
+}
+
+static int stm32_exti_h_set_wake(struct irq_data *d, unsigned int on)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	u32 mask = BIT(d->hwirq % IRQS_PER_BANK);
+
+	raw_spin_lock(&chip_data->rlock);
+
+	if (on)
+		chip_data->wake_active |= mask;
+	else
+		chip_data->wake_active &= ~mask;
+
+	raw_spin_unlock(&chip_data->rlock);
+
+	return 0;
+}
+
+static int stm32_exti_h_set_affinity(struct irq_data *d,
+				     const struct cpumask *dest, bool force)
+{
+	if (d->parent_data->chip)
+		return irq_chip_set_affinity_parent(d, dest, force);
+
+	return -EINVAL;
+}
+
+static struct irq_chip stm32_exti_h_chip = {
+	.name			= "stm32-exti-h",
+	.irq_eoi		= stm32_exti_h_eoi,
+	.irq_mask		= stm32_exti_h_mask,
+	.irq_unmask		= stm32_exti_h_unmask,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_set_type		= stm32_exti_h_set_type,
+	.irq_set_wake		= stm32_exti_h_set_wake,
+	.flags			= IRQCHIP_MASK_ON_SUSPEND,
+#ifdef CONFIG_SMP
+	.irq_set_affinity	= stm32_exti_h_set_affinity,
+#endif
+};
+
+static int stm32_exti_h_domain_alloc(struct irq_domain *dm,
+				     unsigned int virq,
+				     unsigned int nr_irqs, void *data)
+{
+	struct stm32_exti_host_data *host_data = dm->host_data;
+	struct stm32_exti_chip_data *chip_data;
+	struct irq_fwspec *fwspec = data;
+	struct irq_fwspec p_fwspec;
+	irq_hw_number_t hwirq;
+	int p_irq, bank;
+
+	hwirq = fwspec->param[0];
+	bank  = hwirq / IRQS_PER_BANK;
+	chip_data = &host_data->chips_data[bank];
+
+	irq_domain_set_hwirq_and_chip(dm, virq, hwirq,
+				      &stm32_exti_h_chip, chip_data);
+
+	p_irq = stm32_exti_to_irq(host_data->drv_data, hwirq);
+	if (p_irq >= 0) {
+		p_fwspec.fwnode = dm->parent->fwnode;
+		p_fwspec.param_count = 3;
+		p_fwspec.param[0] = GIC_SPI;
+		p_fwspec.param[1] = p_irq;
+		p_fwspec.param[2] = IRQ_TYPE_LEVEL_HIGH;
+
+		return irq_domain_alloc_irqs_parent(dm, virq, 1, &p_fwspec);
+	}
+
+	return 0;
+}
+
 static struct
 stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
 					   struct device_node *node)
@@ -323,6 +585,8 @@ stm32_exti_chip_data *stm32_exti_chip_init(struct stm32_exti_host_data *h_data,
 	chip_data->host_data = h_data;
 	chip_data->reg_bank = stm32_bank;
 
+	raw_spin_lock_init(&chip_data->rlock);
+
 	/* Determine number of irqs supported */
 	writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
 	irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
@@ -421,6 +685,56 @@ static int __init stm32_exti_init(const struct stm32_exti_drv_data *drv_data,
 	return ret;
 }
 
+static const struct irq_domain_ops stm32_exti_h_domain_ops = {
+	.alloc	= stm32_exti_h_domain_alloc,
+	.free	= irq_domain_free_irqs_common,
+};
+
+static int
+__init stm32_exti_hierarchy_init(const struct stm32_exti_drv_data *drv_data,
+				 struct device_node *node,
+				 struct device_node *parent)
+{
+	struct irq_domain *parent_domain, *domain;
+	struct stm32_exti_host_data *host_data;
+	int ret, i;
+
+	parent_domain = irq_find_host(parent);
+	if (!parent_domain) {
+		pr_err("interrupt-parent not found\n");
+		return -EINVAL;
+	}
+
+	host_data = stm32_exti_host_init(drv_data, node);
+	if (!host_data) {
+		ret = -ENOMEM;
+		goto out_free_mem;
+	}
+
+	for (i = 0; i < drv_data->bank_nr; i++)
+		stm32_exti_chip_init(host_data, i, node);
+
+	domain = irq_domain_add_hierarchy(parent_domain, 0,
+					  drv_data->bank_nr * IRQS_PER_BANK,
+					  node, &stm32_exti_h_domain_ops,
+					  host_data);
+
+	if (!domain) {
+		pr_err("%s: Could not register exti domain.\n", node->name);
+		ret = -ENOMEM;
+		goto out_unmap;
+	}
+
+	return 0;
+
+out_unmap:
+	iounmap(host_data->base);
+out_free_mem:
+	kfree(host_data->chips_data);
+	kfree(host_data);
+	return ret;
+}
+
 static int __init stm32f4_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
@@ -436,3 +750,11 @@ static int __init stm32h7_exti_of_init(struct device_node *np,
 }
 
 IRQCHIP_DECLARE(stm32h7_exti, "st,stm32h7-exti", stm32h7_exti_of_init);
+
+static int __init stm32mp1_exti_of_init(struct device_node *np,
+					struct device_node *parent)
+{
+	return stm32_exti_hierarchy_init(&stm32mp1_drv_data, np, parent);
+}
+
+IRQCHIP_DECLARE(stm32mp1_exti, "st,stm32mp1-exti", stm32mp1_exti_of_init);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

Exti controller has been differently integrated on stm32mp1 SoC.
A parent irq has only one external interrupt. A hierachy domain could
be used. Handlers are call by parent, each parent interrupt could be
masked and unmasked according to the needs.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 .../interrupt-controller/st,stm32-exti.txt         |   3 +
 drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
 2 files changed, 325 insertions(+)

diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
index edf03f0..136bd61 100644
--- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
+++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
@@ -5,11 +5,14 @@ Required properties:
 - compatible: Should be:
     "st,stm32-exti"
     "st,stm32h7-exti"
+    "st,stm32mp1-exti"
 - reg: Specifies base physical address and size of the registers
 - interrupt-controller: Indentifies the node as an interrupt controller
 - #interrupt-cells: Specifies the number of cells to encode an interrupt
   specifier, shall be 2
 - interrupts: interrupts references to primary interrupt controller
+  (only needed for exti controller with multiple exti under
+  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
 
 Example:
 
diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index b38c655..ebf7146 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -15,6 +15,8 @@
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
 
+#include <dt-bindings/interrupt-controller/arm-gic.h>
+
 #define IRQS_PER_BANK 32
 
 struct stm32_exti_bank {
@@ -29,14 +31,24 @@ struct stm32_exti_bank {
 
 #define UNDEF_REG ~0
 
+struct stm32_desc_irq {
+	u32 exti;
+	u32 irq_parent;
+};
+
 struct stm32_exti_drv_data {
 	const struct stm32_exti_bank **exti_banks;
+	const struct stm32_desc_irq *desc_irqs;
 	u32 bank_nr;
+	u32 irq_nr;
 };
 
 struct stm32_exti_chip_data {
 	struct stm32_exti_host_data *host_data;
 	const struct stm32_exti_bank *reg_bank;
+	struct raw_spinlock rlock;
+	u32 wake_active;
+	u32 mask_cache;
 	u32 rtsr_cache;
 	u32 ftsr_cache;
 };
@@ -107,6 +119,89 @@ static const struct stm32_exti_drv_data stm32h7xx_drv_data = {
 	.bank_nr = ARRAY_SIZE(stm32h7xx_exti_banks),
 };
 
+static const struct stm32_exti_bank stm32mp1_exti_b1 = {
+	.imr_ofst	= 0x80,
+	.emr_ofst	= 0x84,
+	.rtsr_ofst	= 0x00,
+	.ftsr_ofst	= 0x04,
+	.swier_ofst	= 0x08,
+	.rpr_ofst	= 0x0C,
+	.fpr_ofst	= 0x10,
+};
+
+static const struct stm32_exti_bank stm32mp1_exti_b2 = {
+	.imr_ofst	= 0x90,
+	.emr_ofst	= 0x94,
+	.rtsr_ofst	= 0x20,
+	.ftsr_ofst	= 0x24,
+	.swier_ofst	= 0x28,
+	.rpr_ofst	= 0x2C,
+	.fpr_ofst	= 0x30,
+};
+
+static const struct stm32_exti_bank stm32mp1_exti_b3 = {
+	.imr_ofst	= 0xA0,
+	.emr_ofst	= 0xA4,
+	.rtsr_ofst	= 0x40,
+	.ftsr_ofst	= 0x44,
+	.swier_ofst	= 0x48,
+	.rpr_ofst	= 0x4C,
+	.fpr_ofst	= 0x50,
+};
+
+static const struct stm32_exti_bank *stm32mp1_exti_banks[] = {
+	&stm32mp1_exti_b1,
+	&stm32mp1_exti_b2,
+	&stm32mp1_exti_b3,
+};
+
+static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
+	{ .exti = 1, .irq_parent = 7 },
+	{ .exti = 2, .irq_parent = 8 },
+	{ .exti = 3, .irq_parent = 9 },
+	{ .exti = 4, .irq_parent = 10 },
+	{ .exti = 5, .irq_parent = 23 },
+	{ .exti = 6, .irq_parent = 64 },
+	{ .exti = 7, .irq_parent = 65 },
+	{ .exti = 8, .irq_parent = 66 },
+	{ .exti = 9, .irq_parent = 67 },
+	{ .exti = 10, .irq_parent = 40 },
+	{ .exti = 11, .irq_parent = 42 },
+	{ .exti = 12, .irq_parent = 76 },
+	{ .exti = 13, .irq_parent = 77 },
+	{ .exti = 14, .irq_parent = 121 },
+	{ .exti = 15, .irq_parent = 127 },
+	{ .exti = 16, .irq_parent = 1 },
+	{ .exti = 65, .irq_parent = 144 },
+	{ .exti = 68, .irq_parent = 143 },
+	{ .exti = 73, .irq_parent = 129 },
+};
+
+static const struct stm32_exti_drv_data stm32mp1_drv_data = {
+	.exti_banks = stm32mp1_exti_banks,
+	.bank_nr = ARRAY_SIZE(stm32mp1_exti_banks),
+	.desc_irqs = stm32mp1_desc_irq,
+	.irq_nr = ARRAY_SIZE(stm32mp1_desc_irq),
+};
+
+static int stm32_exti_to_irq(const struct stm32_exti_drv_data *drv_data,
+			     irq_hw_number_t hwirq)
+{
+	const struct stm32_desc_irq *desc_irq;
+	int i;
+
+	if (!drv_data->desc_irqs)
+		return -EINVAL;
+
+	for (i = 0; i < drv_data->irq_nr; i++) {
+		desc_irq = &drv_data->desc_irqs[i];
+		if (desc_irq->exti == hwirq)
+			return desc_irq->irq_parent;
+	}
+
+	return -EINVAL;
+}
+
 static unsigned long stm32_exti_pending(struct irq_chip_generic *gc)
 {
 	struct stm32_exti_chip_data *chip_data = gc->private;
@@ -282,6 +377,173 @@ static void stm32_irq_ack(struct irq_data *d)
 
 	irq_gc_unlock(gc);
 }
+
+static inline u32 stm32_exti_set_bit(struct irq_data *d, u32 reg)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	void __iomem *base = chip_data->host_data->base;
+	u32 val;
+
+	val = readl_relaxed(base + reg);
+	val |= BIT(d->hwirq % IRQS_PER_BANK);
+	writel_relaxed(val, base + reg);
+
+	return val;
+}
+
+static inline u32 stm32_exti_clr_bit(struct irq_data *d, u32 reg)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	void __iomem *base = chip_data->host_data->base;
+	u32 val;
+
+	val = readl_relaxed(base + reg);
+	val &= ~BIT(d->hwirq % IRQS_PER_BANK);
+	writel_relaxed(val, base + reg);
+
+	return val;
+}
+
+static void stm32_exti_h_eoi(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+
+	stm32_exti_set_bit(d, stm32_bank->rpr_ofst);
+	if (stm32_bank->fpr_ofst != UNDEF_REG)
+		stm32_exti_set_bit(d, stm32_bank->fpr_ofst);
+
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_eoi_parent(d);
+}
+
+static void stm32_exti_h_mask(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+	chip_data->mask_cache = stm32_exti_clr_bit(d, stm32_bank->imr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_mask_parent(d);
+}
+
+static void stm32_exti_h_unmask(struct irq_data *d)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+
+	raw_spin_lock(&chip_data->rlock);
+	chip_data->mask_cache = stm32_exti_set_bit(d, stm32_bank->imr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	if (d->parent_data->chip)
+		irq_chip_unmask_parent(d);
+}
+
+static int stm32_exti_h_set_type(struct irq_data *d, unsigned int type)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	const struct stm32_exti_bank *stm32_bank = chip_data->reg_bank;
+	void __iomem *base = chip_data->host_data->base;
+	u32 rtsr, ftsr;
+	int err;
+
+	raw_spin_lock(&chip_data->rlock);
+	rtsr = readl_relaxed(base + stm32_bank->rtsr_ofst);
+	ftsr = readl_relaxed(base + stm32_bank->ftsr_ofst);
+
+	err = stm32_exti_set_type(d, type, &rtsr, &ftsr);
+	if (err) {
+		raw_spin_unlock(&chip_data->rlock);
+		return err;
+	}
+
+	writel_relaxed(rtsr, base + stm32_bank->rtsr_ofst);
+	writel_relaxed(ftsr, base + stm32_bank->ftsr_ofst);
+	raw_spin_unlock(&chip_data->rlock);
+
+	return 0;
+}
+
+static int stm32_exti_h_set_wake(struct irq_data *d, unsigned int on)
+{
+	struct stm32_exti_chip_data *chip_data = irq_data_get_irq_chip_data(d);
+	u32 mask = BIT(d->hwirq % IRQS_PER_BANK);
+
+	raw_spin_lock(&chip_data->rlock);
+
+	if (on)
+		chip_data->wake_active |= mask;
+	else
+		chip_data->wake_active &= ~mask;
+
+	raw_spin_unlock(&chip_data->rlock);
+
+	return 0;
+}
+
+static int stm32_exti_h_set_affinity(struct irq_data *d,
+				     const struct cpumask *dest, bool force)
+{
+	if (d->parent_data->chip)
+		return irq_chip_set_affinity_parent(d, dest, force);
+
+	return -EINVAL;
+}
+
+static struct irq_chip stm32_exti_h_chip = {
+	.name			= "stm32-exti-h",
+	.irq_eoi		= stm32_exti_h_eoi,
+	.irq_mask		= stm32_exti_h_mask,
+	.irq_unmask		= stm32_exti_h_unmask,
+	.irq_retrigger		= irq_chip_retrigger_hierarchy,
+	.irq_set_type		= stm32_exti_h_set_type,
+	.irq_set_wake		= stm32_exti_h_set_wake,
+	.flags			= IRQCHIP_MASK_ON_SUSPEND,
+#ifdef CONFIG_SMP
+	.irq_set_affinity	= stm32_exti_h_set_affinity,
+#endif
+};
+
+static int stm32_exti_h_domain_alloc(struct irq_domain *dm,
+				     unsigned int virq,
+				     unsigned int nr_irqs, void *data)
+{
+	struct stm32_exti_host_data *host_data = dm->host_data;
+	struct stm32_exti_chip_data *chip_data;
+	struct irq_fwspec *fwspec = data;
+	struct irq_fwspec p_fwspec;
+	irq_hw_number_t hwirq;
+	int p_irq, bank;
+
+	hwirq = fwspec->param[0];
+	bank  = hwirq / IRQS_PER_BANK;
+	chip_data = &host_data->chips_data[bank];
+
+	irq_domain_set_hwirq_and_chip(dm, virq, hwirq,
+				      &stm32_exti_h_chip, chip_data);
+
+	p_irq = stm32_exti_to_irq(host_data->drv_data, hwirq);
+	if (p_irq >= 0) {
+		p_fwspec.fwnode = dm->parent->fwnode;
+		p_fwspec.param_count = 3;
+		p_fwspec.param[0] = GIC_SPI;
+		p_fwspec.param[1] = p_irq;
+		p_fwspec.param[2] = IRQ_TYPE_LEVEL_HIGH;
+
+		return irq_domain_alloc_irqs_parent(dm, virq, 1, &p_fwspec);
+	}
+
+	return 0;
+}
+
 static struct
 stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
 					   struct device_node *node)
@@ -323,6 +585,8 @@ stm32_exti_chip_data *stm32_exti_chip_init(struct stm32_exti_host_data *h_data,
 	chip_data->host_data = h_data;
 	chip_data->reg_bank = stm32_bank;
 
+	raw_spin_lock_init(&chip_data->rlock);
+
 	/* Determine number of irqs supported */
 	writel_relaxed(~0UL, base + stm32_bank->rtsr_ofst);
 	irqs_mask = readl_relaxed(base + stm32_bank->rtsr_ofst);
@@ -421,6 +685,56 @@ static int __init stm32_exti_init(const struct stm32_exti_drv_data *drv_data,
 	return ret;
 }
 
+static const struct irq_domain_ops stm32_exti_h_domain_ops = {
+	.alloc	= stm32_exti_h_domain_alloc,
+	.free	= irq_domain_free_irqs_common,
+};
+
+static int
+__init stm32_exti_hierarchy_init(const struct stm32_exti_drv_data *drv_data,
+				 struct device_node *node,
+				 struct device_node *parent)
+{
+	struct irq_domain *parent_domain, *domain;
+	struct stm32_exti_host_data *host_data;
+	int ret, i;
+
+	parent_domain = irq_find_host(parent);
+	if (!parent_domain) {
+		pr_err("interrupt-parent not found\n");
+		return -EINVAL;
+	}
+
+	host_data = stm32_exti_host_init(drv_data, node);
+	if (!host_data) {
+		ret = -ENOMEM;
+		goto out_free_mem;
+	}
+
+	for (i = 0; i < drv_data->bank_nr; i++)
+		stm32_exti_chip_init(host_data, i, node);
+
+	domain = irq_domain_add_hierarchy(parent_domain, 0,
+					  drv_data->bank_nr * IRQS_PER_BANK,
+					  node, &stm32_exti_h_domain_ops,
+					  host_data);
+
+	if (!domain) {
+		pr_err("%s: Could not register exti domain.\n", node->name);
+		ret = -ENOMEM;
+		goto out_unmap;
+	}
+
+	return 0;
+
+out_unmap:
+	iounmap(host_data->base);
+out_free_mem:
+	kfree(host_data->chips_data);
+	kfree(host_data);
+	return ret;
+}
+
 static int __init stm32f4_exti_of_init(struct device_node *np,
 				       struct device_node *parent)
 {
@@ -436,3 +750,11 @@ static int __init stm32h7_exti_of_init(struct device_node *np,
 }
 
 IRQCHIP_DECLARE(stm32h7_exti, "st,stm32h7-exti", stm32h7_exti_of_init);
+
+static int __init stm32mp1_exti_of_init(struct device_node *np,
+					struct device_node *parent)
+{
+	return stm32_exti_hierarchy_init(&stm32mp1_drv_data, np, parent);
+}
+
+IRQCHIP_DECLARE(stm32mp1_exti, "st,stm32mp1-exti", stm32mp1_exti_of_init);
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 08/11] irqchip: stm32: add suspend/resume support for hierarchy domain
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds suspend/resume feature for exti hierarchy domain.
-suspend function sets wake_active into imr of each banks
-resume function restores the mask_cache interrupt into
 imr of each banks

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 49 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index ebf7146..5089c1e 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -14,6 +14,7 @@
 #include <linux/irqdomain.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
+#include <linux/syscore_ops.h>
 
 #include <dt-bindings/interrupt-controller/arm-gic.h>
 
@@ -59,6 +60,8 @@ struct stm32_exti_host_data {
 	const struct stm32_exti_drv_data *drv_data;
 };
 
+static struct stm32_exti_host_data *stm32_host_data;
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
@@ -498,6 +501,48 @@ static int stm32_exti_h_set_affinity(struct irq_data *d,
 	return -EINVAL;
 }
 
+#ifdef CONFIG_PM
+static int stm32_exti_h_suspend(void)
+{
+	struct stm32_exti_chip_data *chip_data;
+	int i;
+
+	for (i = 0; i < stm32_host_data->drv_data->bank_nr; i++) {
+		chip_data = &stm32_host_data->chips_data[i];
+		raw_spin_lock(&chip_data->rlock);
+		stm32_chip_suspend(chip_data, chip_data->wake_active);
+		raw_spin_unlock(&chip_data->rlock);
+	}
+
+	return 0;
+}
+
+static void stm32_exti_h_resume(void)
+{
+	struct stm32_exti_chip_data *chip_data;
+	int i;
+
+	for (i = 0; i < stm32_host_data->drv_data->bank_nr; i++) {
+		chip_data = &stm32_host_data->chips_data[i];
+		raw_spin_lock(&chip_data->rlock);
+		stm32_chip_resume(chip_data, chip_data->mask_cache);
+		raw_spin_unlock(&chip_data->rlock);
+	}
+}
+
+static struct syscore_ops stm32_exti_h_syscore_ops = {
+	.suspend	= stm32_exti_h_suspend,
+	.resume		= stm32_exti_h_resume,
+};
+
+static void stm32_exti_h_syscore_init(void)
+{
+	register_syscore_ops(&stm32_exti_h_syscore_ops);
+}
+#else
+static inline void stm32_exti_h_syscore_init(void) {}
+#endif
+
 static struct irq_chip stm32_exti_h_chip = {
 	.name			= "stm32-exti-h",
 	.irq_eoi		= stm32_exti_h_eoi,
@@ -567,6 +612,8 @@ stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
 		return NULL;
 	}
 
+	stm32_host_data = host_data;
+
 	return host_data;
 }
 
@@ -725,6 +772,8 @@ __init stm32_exti_hierarchy_init(const struct stm32_exti_drv_data *drv_data,
 		goto out_unmap;
 	}
 
+	stm32_exti_h_syscore_init();
+
 	return 0;
 
 out_unmap:
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 08/11] irqchip: stm32: add suspend/resume support for hierarchy domain
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds suspend/resume feature for exti hierarchy domain.
-suspend function sets wake_active into imr of each banks
-resume function restores the mask_cache interrupt into
 imr of each banks

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 49 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index ebf7146..5089c1e 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -14,6 +14,7 @@
 #include <linux/irqdomain.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
+#include <linux/syscore_ops.h>
 
 #include <dt-bindings/interrupt-controller/arm-gic.h>
 
@@ -59,6 +60,8 @@ struct stm32_exti_host_data {
 	const struct stm32_exti_drv_data *drv_data;
 };
 
+static struct stm32_exti_host_data *stm32_host_data;
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
@@ -498,6 +501,48 @@ static int stm32_exti_h_set_affinity(struct irq_data *d,
 	return -EINVAL;
 }
 
+#ifdef CONFIG_PM
+static int stm32_exti_h_suspend(void)
+{
+	struct stm32_exti_chip_data *chip_data;
+	int i;
+
+	for (i = 0; i < stm32_host_data->drv_data->bank_nr; i++) {
+		chip_data = &stm32_host_data->chips_data[i];
+		raw_spin_lock(&chip_data->rlock);
+		stm32_chip_suspend(chip_data, chip_data->wake_active);
+		raw_spin_unlock(&chip_data->rlock);
+	}
+
+	return 0;
+}
+
+static void stm32_exti_h_resume(void)
+{
+	struct stm32_exti_chip_data *chip_data;
+	int i;
+
+	for (i = 0; i < stm32_host_data->drv_data->bank_nr; i++) {
+		chip_data = &stm32_host_data->chips_data[i];
+		raw_spin_lock(&chip_data->rlock);
+		stm32_chip_resume(chip_data, chip_data->mask_cache);
+		raw_spin_unlock(&chip_data->rlock);
+	}
+}
+
+static struct syscore_ops stm32_exti_h_syscore_ops = {
+	.suspend	= stm32_exti_h_suspend,
+	.resume		= stm32_exti_h_resume,
+};
+
+static void stm32_exti_h_syscore_init(void)
+{
+	register_syscore_ops(&stm32_exti_h_syscore_ops);
+}
+#else
+static inline void stm32_exti_h_syscore_init(void) {}
+#endif
+
 static struct irq_chip stm32_exti_h_chip = {
 	.name			= "stm32-exti-h",
 	.irq_eoi		= stm32_exti_h_eoi,
@@ -567,6 +612,8 @@ stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
 		return NULL;
 	}
 
+	stm32_host_data = host_data;
+
 	return host_data;
 }
 
@@ -725,6 +772,8 @@ __init stm32_exti_hierarchy_init(const struct stm32_exti_drv_data *drv_data,
 		goto out_unmap;
 	}
 
+	stm32_exti_h_syscore_init();
+
 	return 0;
 
 out_unmap:
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 08/11] irqchip: stm32: add suspend/resume support for hierarchy domain
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds suspend/resume feature for exti hierarchy domain.
-suspend function sets wake_active into imr of each banks
-resume function restores the mask_cache interrupt into
 imr of each banks

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/irqchip/irq-stm32-exti.c | 49 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)

diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
index ebf7146..5089c1e 100644
--- a/drivers/irqchip/irq-stm32-exti.c
+++ b/drivers/irqchip/irq-stm32-exti.c
@@ -14,6 +14,7 @@
 #include <linux/irqdomain.h>
 #include <linux/of_address.h>
 #include <linux/of_irq.h>
+#include <linux/syscore_ops.h>
 
 #include <dt-bindings/interrupt-controller/arm-gic.h>
 
@@ -59,6 +60,8 @@ struct stm32_exti_host_data {
 	const struct stm32_exti_drv_data *drv_data;
 };
 
+static struct stm32_exti_host_data *stm32_host_data;
+
 static const struct stm32_exti_bank stm32f4xx_exti_b1 = {
 	.imr_ofst	= 0x00,
 	.emr_ofst	= 0x04,
@@ -498,6 +501,48 @@ static int stm32_exti_h_set_affinity(struct irq_data *d,
 	return -EINVAL;
 }
 
+#ifdef CONFIG_PM
+static int stm32_exti_h_suspend(void)
+{
+	struct stm32_exti_chip_data *chip_data;
+	int i;
+
+	for (i = 0; i < stm32_host_data->drv_data->bank_nr; i++) {
+		chip_data = &stm32_host_data->chips_data[i];
+		raw_spin_lock(&chip_data->rlock);
+		stm32_chip_suspend(chip_data, chip_data->wake_active);
+		raw_spin_unlock(&chip_data->rlock);
+	}
+
+	return 0;
+}
+
+static void stm32_exti_h_resume(void)
+{
+	struct stm32_exti_chip_data *chip_data;
+	int i;
+
+	for (i = 0; i < stm32_host_data->drv_data->bank_nr; i++) {
+		chip_data = &stm32_host_data->chips_data[i];
+		raw_spin_lock(&chip_data->rlock);
+		stm32_chip_resume(chip_data, chip_data->mask_cache);
+		raw_spin_unlock(&chip_data->rlock);
+	}
+}
+
+static struct syscore_ops stm32_exti_h_syscore_ops = {
+	.suspend	= stm32_exti_h_suspend,
+	.resume		= stm32_exti_h_resume,
+};
+
+static void stm32_exti_h_syscore_init(void)
+{
+	register_syscore_ops(&stm32_exti_h_syscore_ops);
+}
+#else
+static inline void stm32_exti_h_syscore_init(void) {}
+#endif
+
 static struct irq_chip stm32_exti_h_chip = {
 	.name			= "stm32-exti-h",
 	.irq_eoi		= stm32_exti_h_eoi,
@@ -567,6 +612,8 @@ stm32_exti_host_data *stm32_exti_host_init(const struct stm32_exti_drv_data *dd,
 		return NULL;
 	}
 
+	stm32_host_data = host_data;
+
 	return host_data;
 }
 
@@ -725,6 +772,8 @@ __init stm32_exti_hierarchy_init(const struct stm32_exti_drv_data *drv_data,
 		goto out_unmap;
 	}
 
+	stm32_exti_h_syscore_init();
+
 	return 0;
 
 out_unmap:
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 09/11] pinctrl: stm32: add irq_eoi for stm32gpio irqchip
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

-Parent domain of stm32gpio evolves to hierarchy domain
 and could have a handle_fasteoi_irq. So an irq_eoi parent callback
 is needed for children.
-Replace space by tabulation.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/pinctrl/stm32/pinctrl-stm32.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/pinctrl/stm32/pinctrl-stm32.c b/drivers/pinctrl/stm32/pinctrl-stm32.c
index 6cbcff4..dfed609 100644
--- a/drivers/pinctrl/stm32/pinctrl-stm32.c
+++ b/drivers/pinctrl/stm32/pinctrl-stm32.c
@@ -267,12 +267,13 @@ static void stm32_gpio_irq_release_resources(struct irq_data *irq_data)
 }
 
 static struct irq_chip stm32_gpio_irq_chip = {
-	.name           = "stm32gpio",
-	.irq_ack       = irq_chip_ack_parent,
-	.irq_mask       = irq_chip_mask_parent,
-	.irq_unmask     = irq_chip_unmask_parent,
-	.irq_set_type   = irq_chip_set_type_parent,
-	.irq_set_wake   = irq_chip_set_wake_parent,
+	.name		= "stm32gpio",
+	.irq_eoi	= irq_chip_eoi_parent,
+	.irq_ack	= irq_chip_ack_parent,
+	.irq_mask	= irq_chip_mask_parent,
+	.irq_unmask	= irq_chip_unmask_parent,
+	.irq_set_type	= irq_chip_set_type_parent,
+	.irq_set_wake	= irq_chip_set_wake_parent,
 	.irq_request_resources = stm32_gpio_irq_request_resources,
 	.irq_release_resources = stm32_gpio_irq_release_resources,
 };
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 09/11] pinctrl: stm32: add irq_eoi for stm32gpio irqchip
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

-Parent domain of stm32gpio evolves to hierarchy domain
 and could have a handle_fasteoi_irq. So an irq_eoi parent callback
 is needed for children.
-Replace space by tabulation.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/pinctrl/stm32/pinctrl-stm32.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/pinctrl/stm32/pinctrl-stm32.c b/drivers/pinctrl/stm32/pinctrl-stm32.c
index 6cbcff4..dfed609 100644
--- a/drivers/pinctrl/stm32/pinctrl-stm32.c
+++ b/drivers/pinctrl/stm32/pinctrl-stm32.c
@@ -267,12 +267,13 @@ static void stm32_gpio_irq_release_resources(struct irq_data *irq_data)
 }
 
 static struct irq_chip stm32_gpio_irq_chip = {
-	.name           = "stm32gpio",
-	.irq_ack       = irq_chip_ack_parent,
-	.irq_mask       = irq_chip_mask_parent,
-	.irq_unmask     = irq_chip_unmask_parent,
-	.irq_set_type   = irq_chip_set_type_parent,
-	.irq_set_wake   = irq_chip_set_wake_parent,
+	.name		= "stm32gpio",
+	.irq_eoi	= irq_chip_eoi_parent,
+	.irq_ack	= irq_chip_ack_parent,
+	.irq_mask	= irq_chip_mask_parent,
+	.irq_unmask	= irq_chip_unmask_parent,
+	.irq_set_type	= irq_chip_set_type_parent,
+	.irq_set_wake	= irq_chip_set_wake_parent,
 	.irq_request_resources = stm32_gpio_irq_request_resources,
 	.irq_release_resources = stm32_gpio_irq_release_resources,
 };
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 09/11] pinctrl: stm32: add irq_eoi for stm32gpio irqchip
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

-Parent domain of stm32gpio evolves to hierarchy domain
 and could have a handle_fasteoi_irq. So an irq_eoi parent callback
 is needed for children.
-Replace space by tabulation.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 drivers/pinctrl/stm32/pinctrl-stm32.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/pinctrl/stm32/pinctrl-stm32.c b/drivers/pinctrl/stm32/pinctrl-stm32.c
index 6cbcff4..dfed609 100644
--- a/drivers/pinctrl/stm32/pinctrl-stm32.c
+++ b/drivers/pinctrl/stm32/pinctrl-stm32.c
@@ -267,12 +267,13 @@ static void stm32_gpio_irq_release_resources(struct irq_data *irq_data)
 }
 
 static struct irq_chip stm32_gpio_irq_chip = {
-	.name           = "stm32gpio",
-	.irq_ack       = irq_chip_ack_parent,
-	.irq_mask       = irq_chip_mask_parent,
-	.irq_unmask     = irq_chip_unmask_parent,
-	.irq_set_type   = irq_chip_set_type_parent,
-	.irq_set_wake   = irq_chip_set_wake_parent,
+	.name		= "stm32gpio",
+	.irq_eoi	= irq_chip_eoi_parent,
+	.irq_ack	= irq_chip_ack_parent,
+	.irq_mask	= irq_chip_mask_parent,
+	.irq_unmask	= irq_chip_unmask_parent,
+	.irq_set_type	= irq_chip_set_type_parent,
+	.irq_set_wake	= irq_chip_set_wake_parent,
 	.irq_request_resources = stm32_gpio_irq_request_resources,
 	.irq_release_resources = stm32_gpio_irq_release_resources,
 };
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 10/11] ARM: dts: stm32: add exti support for stm32mp157c
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds external interrupt (exti) support
on stm32mp157c SoC.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 arch/arm/boot/dts/stm32mp157c.dtsi | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157c.dtsi b/arch/arm/boot/dts/stm32mp157c.dtsi
index bfcf84b..e6fcf8f 100644
--- a/arch/arm/boot/dts/stm32mp157c.dtsi
+++ b/arch/arm/boot/dts/stm32mp157c.dtsi
@@ -167,6 +167,13 @@
 			#reset-cells = <1>;
 		};
 
+		exti: interrupt-controller@5000d000 {
+			compatible = "st,stm32mp1-exti", "syscon";
+			interrupt-controller;
+			#interrupt-cells = <2>;
+			reg = <0x5000d000 0x400>;
+		};
+
 		usart1: serial@5c000000 {
 			compatible = "st,stm32h7-uart";
 			reg = <0x5c000000 0x400>;
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 10/11] ARM: dts: stm32: add exti support for stm32mp157c
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds external interrupt (exti) support
on stm32mp157c SoC.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 arch/arm/boot/dts/stm32mp157c.dtsi | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157c.dtsi b/arch/arm/boot/dts/stm32mp157c.dtsi
index bfcf84b..e6fcf8f 100644
--- a/arch/arm/boot/dts/stm32mp157c.dtsi
+++ b/arch/arm/boot/dts/stm32mp157c.dtsi
@@ -167,6 +167,13 @@
 			#reset-cells = <1>;
 		};
 
+		exti: interrupt-controller@5000d000 {
+			compatible = "st,stm32mp1-exti", "syscon";
+			interrupt-controller;
+			#interrupt-cells = <2>;
+			reg = <0x5000d000 0x400>;
+		};
+
 		usart1: serial@5c000000 {
 			compatible = "st,stm32h7-uart";
 			reg = <0x5c000000 0x400>;
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 10/11] ARM: dts: stm32: add exti support for stm32mp157c
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds external interrupt (exti) support
on stm32mp157c SoC.

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 arch/arm/boot/dts/stm32mp157c.dtsi | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157c.dtsi b/arch/arm/boot/dts/stm32mp157c.dtsi
index bfcf84b..e6fcf8f 100644
--- a/arch/arm/boot/dts/stm32mp157c.dtsi
+++ b/arch/arm/boot/dts/stm32mp157c.dtsi
@@ -167,6 +167,13 @@
 			#reset-cells = <1>;
 		};
 
+		exti: interrupt-controller at 5000d000 {
+			compatible = "st,stm32mp1-exti", "syscon";
+			interrupt-controller;
+			#interrupt-cells = <2>;
+			reg = <0x5000d000 0x400>;
+		};
+
 		usart1: serial at 5c000000 {
 			compatible = "st,stm32h7-uart";
 			reg = <0x5c000000 0x400>;
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 11/11] ARM: dts: stm32: add exti support to stm32mp157 pinctrl
  2018-04-26 16:18 ` Ludovic Barre
  (?)
@ 2018-04-26 16:18   ` Ludovic Barre
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds support of external interrupt for
gpio[a..k], gpioz

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 arch/arm/boot/dts/stm32mp157-pinctrl.dtsi | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi b/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
index 6f044100..2f6872b 100644
--- a/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
+++ b/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
@@ -12,6 +12,8 @@
 			#size-cells = <1>;
 			compatible = "st,stm32mp157-pinctrl";
 			ranges = <0 0x50002000 0xa400>;
+			interrupt-parent = <&exti>;
+			st,syscfg = <&exti 0x60 0xff>;
 			pins-are-numbered;
 
 			gpioa: gpio@50002000 {
@@ -166,6 +168,8 @@
 			compatible = "st,stm32mp157-z-pinctrl";
 			ranges = <0 0x54004000 0x400>;
 			pins-are-numbered;
+			interrupt-parent = <&exti>;
+			st,syscfg = <&exti 0x60 0xff>;
 			status = "disabled";
 
 			gpioz: gpio@54004000 {
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 11/11] ARM: dts: stm32: add exti support to stm32mp157 pinctrl
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: Thomas Gleixner, Jason Cooper, Marc Zyngier, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, Ludovic Barre

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds support of external interrupt for
gpio[a..k], gpioz

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 arch/arm/boot/dts/stm32mp157-pinctrl.dtsi | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi b/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
index 6f044100..2f6872b 100644
--- a/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
+++ b/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
@@ -12,6 +12,8 @@
 			#size-cells = <1>;
 			compatible = "st,stm32mp157-pinctrl";
 			ranges = <0 0x50002000 0xa400>;
+			interrupt-parent = <&exti>;
+			st,syscfg = <&exti 0x60 0xff>;
 			pins-are-numbered;
 
 			gpioa: gpio@50002000 {
@@ -166,6 +168,8 @@
 			compatible = "st,stm32mp157-z-pinctrl";
 			ranges = <0 0x54004000 0x400>;
 			pins-are-numbered;
+			interrupt-parent = <&exti>;
+			st,syscfg = <&exti 0x60 0xff>;
 			status = "disabled";
 
 			gpioz: gpio@54004000 {
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* [PATCH 11/11] ARM: dts: stm32: add exti support to stm32mp157 pinctrl
@ 2018-04-26 16:18   ` Ludovic Barre
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic Barre @ 2018-04-26 16:18 UTC (permalink / raw)
  To: linux-arm-kernel

From: Ludovic Barre <ludovic.barre@st.com>

This patch adds support of external interrupt for
gpio[a..k], gpioz

Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
---
 arch/arm/boot/dts/stm32mp157-pinctrl.dtsi | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi b/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
index 6f044100..2f6872b 100644
--- a/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
+++ b/arch/arm/boot/dts/stm32mp157-pinctrl.dtsi
@@ -12,6 +12,8 @@
 			#size-cells = <1>;
 			compatible = "st,stm32mp157-pinctrl";
 			ranges = <0 0x50002000 0xa400>;
+			interrupt-parent = <&exti>;
+			st,syscfg = <&exti 0x60 0xff>;
 			pins-are-numbered;
 
 			gpioa: gpio at 50002000 {
@@ -166,6 +168,8 @@
 			compatible = "st,stm32mp157-z-pinctrl";
 			ranges = <0 0x54004000 0x400>;
 			pins-are-numbered;
+			interrupt-parent = <&exti>;
+			st,syscfg = <&exti 0x60 0xff>;
 			status = "disabled";
 
 			gpioz: gpio at 54004000 {
-- 
2.7.4

^ permalink raw reply related	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-04-26 16:18   ` Ludovic Barre
@ 2018-05-01 14:56     ` Rob Herring
  -1 siblings, 0 replies; 56+ messages in thread
From: Rob Herring @ 2018-05-01 14:56 UTC (permalink / raw)
  To: Ludovic Barre
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	linux-arm-kernel, devicetree

On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
> From: Ludovic Barre <ludovic.barre@st.com>
> 
> Exti controller has been differently integrated on stm32mp1 SoC.
> A parent irq has only one external interrupt. A hierachy domain could
> be used. Handlers are call by parent, each parent interrupt could be
> masked and unmasked according to the needs.
> 
> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
> ---
>  .../interrupt-controller/st,stm32-exti.txt         |   3 +
>  drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
>  2 files changed, 325 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
> index edf03f0..136bd61 100644
> --- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
> +++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
> @@ -5,11 +5,14 @@ Required properties:
>  - compatible: Should be:
>      "st,stm32-exti"
>      "st,stm32h7-exti"
> +    "st,stm32mp1-exti"
>  - reg: Specifies base physical address and size of the registers
>  - interrupt-controller: Indentifies the node as an interrupt controller
>  - #interrupt-cells: Specifies the number of cells to encode an interrupt
>    specifier, shall be 2
>  - interrupts: interrupts references to primary interrupt controller
> +  (only needed for exti controller with multiple exti under
> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>  
>  Example:
>  
> diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
> index b38c655..ebf7146 100644
> --- a/drivers/irqchip/irq-stm32-exti.c
> +++ b/drivers/irqchip/irq-stm32-exti.c

[...]

> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
> +	{ .exti = 1, .irq_parent = 7 },
> +	{ .exti = 2, .irq_parent = 8 },
> +	{ .exti = 3, .irq_parent = 9 },
> +	{ .exti = 4, .irq_parent = 10 },
> +	{ .exti = 5, .irq_parent = 23 },
> +	{ .exti = 6, .irq_parent = 64 },
> +	{ .exti = 7, .irq_parent = 65 },
> +	{ .exti = 8, .irq_parent = 66 },
> +	{ .exti = 9, .irq_parent = 67 },
> +	{ .exti = 10, .irq_parent = 40 },
> +	{ .exti = 11, .irq_parent = 42 },
> +	{ .exti = 12, .irq_parent = 76 },
> +	{ .exti = 13, .irq_parent = 77 },
> +	{ .exti = 14, .irq_parent = 121 },
> +	{ .exti = 15, .irq_parent = 127 },
> +	{ .exti = 16, .irq_parent = 1 },
> +	{ .exti = 65, .irq_parent = 144 },
> +	{ .exti = 68, .irq_parent = 143 },
> +	{ .exti = 73, .irq_parent = 129 },
> +};

You can use an interrupt-map property rather than put this into the 
driver.

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-01 14:56     ` Rob Herring
  0 siblings, 0 replies; 56+ messages in thread
From: Rob Herring @ 2018-05-01 14:56 UTC (permalink / raw)
  To: linux-arm-kernel

On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
> From: Ludovic Barre <ludovic.barre@st.com>
> 
> Exti controller has been differently integrated on stm32mp1 SoC.
> A parent irq has only one external interrupt. A hierachy domain could
> be used. Handlers are call by parent, each parent interrupt could be
> masked and unmasked according to the needs.
> 
> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
> ---
>  .../interrupt-controller/st,stm32-exti.txt         |   3 +
>  drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
>  2 files changed, 325 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
> index edf03f0..136bd61 100644
> --- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
> +++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
> @@ -5,11 +5,14 @@ Required properties:
>  - compatible: Should be:
>      "st,stm32-exti"
>      "st,stm32h7-exti"
> +    "st,stm32mp1-exti"
>  - reg: Specifies base physical address and size of the registers
>  - interrupt-controller: Indentifies the node as an interrupt controller
>  - #interrupt-cells: Specifies the number of cells to encode an interrupt
>    specifier, shall be 2
>  - interrupts: interrupts references to primary interrupt controller
> +  (only needed for exti controller with multiple exti under
> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>  
>  Example:
>  
> diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
> index b38c655..ebf7146 100644
> --- a/drivers/irqchip/irq-stm32-exti.c
> +++ b/drivers/irqchip/irq-stm32-exti.c

[...]

> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
> +	{ .exti = 1, .irq_parent = 7 },
> +	{ .exti = 2, .irq_parent = 8 },
> +	{ .exti = 3, .irq_parent = 9 },
> +	{ .exti = 4, .irq_parent = 10 },
> +	{ .exti = 5, .irq_parent = 23 },
> +	{ .exti = 6, .irq_parent = 64 },
> +	{ .exti = 7, .irq_parent = 65 },
> +	{ .exti = 8, .irq_parent = 66 },
> +	{ .exti = 9, .irq_parent = 67 },
> +	{ .exti = 10, .irq_parent = 40 },
> +	{ .exti = 11, .irq_parent = 42 },
> +	{ .exti = 12, .irq_parent = 76 },
> +	{ .exti = 13, .irq_parent = 77 },
> +	{ .exti = 14, .irq_parent = 121 },
> +	{ .exti = 15, .irq_parent = 127 },
> +	{ .exti = 16, .irq_parent = 1 },
> +	{ .exti = 65, .irq_parent = 144 },
> +	{ .exti = 68, .irq_parent = 143 },
> +	{ .exti = 73, .irq_parent = 129 },
> +};

You can use an interrupt-map property rather than put this into the 
driver.

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-05-01 14:56     ` Rob Herring
  (?)
@ 2018-05-02 16:03       ` Ludovic BARRE
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-02 16:03 UTC (permalink / raw)
  To: Rob Herring
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	linux-arm-kernel, devicetree

Hi Rob


On 05/01/2018 04:56 PM, Rob Herring wrote:
> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>> From: Ludovic Barre <ludovic.barre@st.com>
>>
>> Exti controller has been differently integrated on stm32mp1 SoC.
>> A parent irq has only one external interrupt. A hierachy domain could
>> be used. Handlers are call by parent, each parent interrupt could be
>> masked and unmasked according to the needs.
>>
>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>> ---
>>   .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>   drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
>>   2 files changed, 325 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> index edf03f0..136bd61 100644
>> --- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> +++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> @@ -5,11 +5,14 @@ Required properties:
>>   - compatible: Should be:
>>       "st,stm32-exti"
>>       "st,stm32h7-exti"
>> +    "st,stm32mp1-exti"
>>   - reg: Specifies base physical address and size of the registers
>>   - interrupt-controller: Indentifies the node as an interrupt controller
>>   - #interrupt-cells: Specifies the number of cells to encode an interrupt
>>     specifier, shall be 2
>>   - interrupts: interrupts references to primary interrupt controller
>> +  (only needed for exti controller with multiple exti under
>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>   
>>   Example:
>>   
>> diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
>> index b38c655..ebf7146 100644
>> --- a/drivers/irqchip/irq-stm32-exti.c
>> +++ b/drivers/irqchip/irq-stm32-exti.c
> 
> [...]
> 
>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>> +	{ .exti = 1, .irq_parent = 7 },
>> +	{ .exti = 2, .irq_parent = 8 },
>> +	{ .exti = 3, .irq_parent = 9 },
>> +	{ .exti = 4, .irq_parent = 10 },
>> +	{ .exti = 5, .irq_parent = 23 },
>> +	{ .exti = 6, .irq_parent = 64 },
>> +	{ .exti = 7, .irq_parent = 65 },
>> +	{ .exti = 8, .irq_parent = 66 },
>> +	{ .exti = 9, .irq_parent = 67 },
>> +	{ .exti = 10, .irq_parent = 40 },
>> +	{ .exti = 11, .irq_parent = 42 },
>> +	{ .exti = 12, .irq_parent = 76 },
>> +	{ .exti = 13, .irq_parent = 77 },
>> +	{ .exti = 14, .irq_parent = 121 },
>> +	{ .exti = 15, .irq_parent = 127 },
>> +	{ .exti = 16, .irq_parent = 1 },
>> +	{ .exti = 65, .irq_parent = 144 },
>> +	{ .exti = 68, .irq_parent = 143 },
>> +	{ .exti = 73, .irq_parent = 129 },
>> +};
> 
> You can use an interrupt-map property rather than put this into the
> driver.

interrupt-map seemed interesting and promising like used in pci host.
At first sight this property can't be used into node with
"interrupt-controller" property (see in drivers/of/irq.c function: 
of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
it first, and after lookup the interrupt-map.

Rob, Thomas, Jason, Marc what do you prefers or the right ways...?

> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-02 16:03       ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-02 16:03 UTC (permalink / raw)
  To: Rob Herring
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	linux-arm-kernel, devicetree

Hi Rob


On 05/01/2018 04:56 PM, Rob Herring wrote:
> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>> From: Ludovic Barre <ludovic.barre@st.com>
>>
>> Exti controller has been differently integrated on stm32mp1 SoC.
>> A parent irq has only one external interrupt. A hierachy domain could
>> be used. Handlers are call by parent, each parent interrupt could be
>> masked and unmasked according to the needs.
>>
>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>> ---
>>   .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>   drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
>>   2 files changed, 325 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> index edf03f0..136bd61 100644
>> --- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> +++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> @@ -5,11 +5,14 @@ Required properties:
>>   - compatible: Should be:
>>       "st,stm32-exti"
>>       "st,stm32h7-exti"
>> +    "st,stm32mp1-exti"
>>   - reg: Specifies base physical address and size of the registers
>>   - interrupt-controller: Indentifies the node as an interrupt controller
>>   - #interrupt-cells: Specifies the number of cells to encode an interrupt
>>     specifier, shall be 2
>>   - interrupts: interrupts references to primary interrupt controller
>> +  (only needed for exti controller with multiple exti under
>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>   
>>   Example:
>>   
>> diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
>> index b38c655..ebf7146 100644
>> --- a/drivers/irqchip/irq-stm32-exti.c
>> +++ b/drivers/irqchip/irq-stm32-exti.c
> 
> [...]
> 
>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>> +	{ .exti = 1, .irq_parent = 7 },
>> +	{ .exti = 2, .irq_parent = 8 },
>> +	{ .exti = 3, .irq_parent = 9 },
>> +	{ .exti = 4, .irq_parent = 10 },
>> +	{ .exti = 5, .irq_parent = 23 },
>> +	{ .exti = 6, .irq_parent = 64 },
>> +	{ .exti = 7, .irq_parent = 65 },
>> +	{ .exti = 8, .irq_parent = 66 },
>> +	{ .exti = 9, .irq_parent = 67 },
>> +	{ .exti = 10, .irq_parent = 40 },
>> +	{ .exti = 11, .irq_parent = 42 },
>> +	{ .exti = 12, .irq_parent = 76 },
>> +	{ .exti = 13, .irq_parent = 77 },
>> +	{ .exti = 14, .irq_parent = 121 },
>> +	{ .exti = 15, .irq_parent = 127 },
>> +	{ .exti = 16, .irq_parent = 1 },
>> +	{ .exti = 65, .irq_parent = 144 },
>> +	{ .exti = 68, .irq_parent = 143 },
>> +	{ .exti = 73, .irq_parent = 129 },
>> +};
> 
> You can use an interrupt-map property rather than put this into the
> driver.

interrupt-map seemed interesting and promising like used in pci host.
At first sight this property can't be used into node with
"interrupt-controller" property (see in drivers/of/irq.c function: 
of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
it first, and after lookup the interrupt-map.

Rob, Thomas, Jason, Marc what do you prefers or the right ways...?

> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-02 16:03       ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-02 16:03 UTC (permalink / raw)
  To: linux-arm-kernel

Hi Rob


On 05/01/2018 04:56 PM, Rob Herring wrote:
> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>> From: Ludovic Barre <ludovic.barre@st.com>
>>
>> Exti controller has been differently integrated on stm32mp1 SoC.
>> A parent irq has only one external interrupt. A hierachy domain could
>> be used. Handlers are call by parent, each parent interrupt could be
>> masked and unmasked according to the needs.
>>
>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>> ---
>>   .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>   drivers/irqchip/irq-stm32-exti.c                   | 322 +++++++++++++++++++++
>>   2 files changed, 325 insertions(+)
>>
>> diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> index edf03f0..136bd61 100644
>> --- a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> +++ b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>> @@ -5,11 +5,14 @@ Required properties:
>>   - compatible: Should be:
>>       "st,stm32-exti"
>>       "st,stm32h7-exti"
>> +    "st,stm32mp1-exti"
>>   - reg: Specifies base physical address and size of the registers
>>   - interrupt-controller: Indentifies the node as an interrupt controller
>>   - #interrupt-cells: Specifies the number of cells to encode an interrupt
>>     specifier, shall be 2
>>   - interrupts: interrupts references to primary interrupt controller
>> +  (only needed for exti controller with multiple exti under
>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>   
>>   Example:
>>   
>> diff --git a/drivers/irqchip/irq-stm32-exti.c b/drivers/irqchip/irq-stm32-exti.c
>> index b38c655..ebf7146 100644
>> --- a/drivers/irqchip/irq-stm32-exti.c
>> +++ b/drivers/irqchip/irq-stm32-exti.c
> 
> [...]
> 
>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>> +	{ .exti = 1, .irq_parent = 7 },
>> +	{ .exti = 2, .irq_parent = 8 },
>> +	{ .exti = 3, .irq_parent = 9 },
>> +	{ .exti = 4, .irq_parent = 10 },
>> +	{ .exti = 5, .irq_parent = 23 },
>> +	{ .exti = 6, .irq_parent = 64 },
>> +	{ .exti = 7, .irq_parent = 65 },
>> +	{ .exti = 8, .irq_parent = 66 },
>> +	{ .exti = 9, .irq_parent = 67 },
>> +	{ .exti = 10, .irq_parent = 40 },
>> +	{ .exti = 11, .irq_parent = 42 },
>> +	{ .exti = 12, .irq_parent = 76 },
>> +	{ .exti = 13, .irq_parent = 77 },
>> +	{ .exti = 14, .irq_parent = 121 },
>> +	{ .exti = 15, .irq_parent = 127 },
>> +	{ .exti = 16, .irq_parent = 1 },
>> +	{ .exti = 65, .irq_parent = 144 },
>> +	{ .exti = 68, .irq_parent = 143 },
>> +	{ .exti = 73, .irq_parent = 129 },
>> +};
> 
> You can use an interrupt-map property rather than put this into the
> driver.

interrupt-map seemed interesting and promising like used in pci host.
At first sight this property can't be used into node with
"interrupt-controller" property (see in drivers/of/irq.c function: 
of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
it first, and after lookup the interrupt-map.

Rob, Thomas, Jason, Marc what do you prefers or the right ways...?

> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-05-02 16:03       ` Ludovic BARRE
@ 2018-05-02 17:45         ` Rob Herring
  -1 siblings, 0 replies; 56+ messages in thread
From: Rob Herring @ 2018-05-02 17:45 UTC (permalink / raw)
  To: Ludovic BARRE
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
	devicetree

On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
> Hi Rob
>
>
>
> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>
>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>
>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>
>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>> A parent irq has only one external interrupt. A hierachy domain could
>>> be used. Handlers are call by parent, each parent interrupt could be
>>> masked and unmasked according to the needs.
>>>
>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>> ---
>>>   .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>   drivers/irqchip/irq-stm32-exti.c                   | 322
>>> +++++++++++++++++++++
>>>   2 files changed, 325 insertions(+)
>>>
>>> diff --git
>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> index edf03f0..136bd61 100644
>>> ---
>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> +++
>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> @@ -5,11 +5,14 @@ Required properties:
>>>   - compatible: Should be:
>>>       "st,stm32-exti"
>>>       "st,stm32h7-exti"
>>> +    "st,stm32mp1-exti"
>>>   - reg: Specifies base physical address and size of the registers
>>>   - interrupt-controller: Indentifies the node as an interrupt controller
>>>   - #interrupt-cells: Specifies the number of cells to encode an
>>> interrupt
>>>     specifier, shall be 2
>>>   - interrupts: interrupts references to primary interrupt controller
>>> +  (only needed for exti controller with multiple exti under
>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>     Example:
>>>   diff --git a/drivers/irqchip/irq-stm32-exti.c
>>> b/drivers/irqchip/irq-stm32-exti.c
>>> index b38c655..ebf7146 100644
>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>
>>
>> [...]
>>
>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>> +       { .exti = 1, .irq_parent = 7 },
>>> +       { .exti = 2, .irq_parent = 8 },
>>> +       { .exti = 3, .irq_parent = 9 },
>>> +       { .exti = 4, .irq_parent = 10 },
>>> +       { .exti = 5, .irq_parent = 23 },
>>> +       { .exti = 6, .irq_parent = 64 },
>>> +       { .exti = 7, .irq_parent = 65 },
>>> +       { .exti = 8, .irq_parent = 66 },
>>> +       { .exti = 9, .irq_parent = 67 },
>>> +       { .exti = 10, .irq_parent = 40 },
>>> +       { .exti = 11, .irq_parent = 42 },
>>> +       { .exti = 12, .irq_parent = 76 },
>>> +       { .exti = 13, .irq_parent = 77 },
>>> +       { .exti = 14, .irq_parent = 121 },
>>> +       { .exti = 15, .irq_parent = 127 },
>>> +       { .exti = 16, .irq_parent = 1 },
>>> +       { .exti = 65, .irq_parent = 144 },
>>> +       { .exti = 68, .irq_parent = 143 },
>>> +       { .exti = 73, .irq_parent = 129 },
>>> +};
>>
>>
>> You can use an interrupt-map property rather than put this into the
>> driver.
>
>
> interrupt-map seemed interesting and promising like used in pci host.
> At first sight this property can't be used into node with
> "interrupt-controller" property (see in drivers/of/irq.c function:
> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
> it first, and after lookup the interrupt-map.
>
> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?

I believe the correct thing to do is simply drop "interrupt-controller".

Rob

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-02 17:45         ` Rob Herring
  0 siblings, 0 replies; 56+ messages in thread
From: Rob Herring @ 2018-05-02 17:45 UTC (permalink / raw)
  To: linux-arm-kernel

On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
> Hi Rob
>
>
>
> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>
>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>
>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>
>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>> A parent irq has only one external interrupt. A hierachy domain could
>>> be used. Handlers are call by parent, each parent interrupt could be
>>> masked and unmasked according to the needs.
>>>
>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>> ---
>>>   .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>   drivers/irqchip/irq-stm32-exti.c                   | 322
>>> +++++++++++++++++++++
>>>   2 files changed, 325 insertions(+)
>>>
>>> diff --git
>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> index edf03f0..136bd61 100644
>>> ---
>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> +++
>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>> @@ -5,11 +5,14 @@ Required properties:
>>>   - compatible: Should be:
>>>       "st,stm32-exti"
>>>       "st,stm32h7-exti"
>>> +    "st,stm32mp1-exti"
>>>   - reg: Specifies base physical address and size of the registers
>>>   - interrupt-controller: Indentifies the node as an interrupt controller
>>>   - #interrupt-cells: Specifies the number of cells to encode an
>>> interrupt
>>>     specifier, shall be 2
>>>   - interrupts: interrupts references to primary interrupt controller
>>> +  (only needed for exti controller with multiple exti under
>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>     Example:
>>>   diff --git a/drivers/irqchip/irq-stm32-exti.c
>>> b/drivers/irqchip/irq-stm32-exti.c
>>> index b38c655..ebf7146 100644
>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>
>>
>> [...]
>>
>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>> +       { .exti = 1, .irq_parent = 7 },
>>> +       { .exti = 2, .irq_parent = 8 },
>>> +       { .exti = 3, .irq_parent = 9 },
>>> +       { .exti = 4, .irq_parent = 10 },
>>> +       { .exti = 5, .irq_parent = 23 },
>>> +       { .exti = 6, .irq_parent = 64 },
>>> +       { .exti = 7, .irq_parent = 65 },
>>> +       { .exti = 8, .irq_parent = 66 },
>>> +       { .exti = 9, .irq_parent = 67 },
>>> +       { .exti = 10, .irq_parent = 40 },
>>> +       { .exti = 11, .irq_parent = 42 },
>>> +       { .exti = 12, .irq_parent = 76 },
>>> +       { .exti = 13, .irq_parent = 77 },
>>> +       { .exti = 14, .irq_parent = 121 },
>>> +       { .exti = 15, .irq_parent = 127 },
>>> +       { .exti = 16, .irq_parent = 1 },
>>> +       { .exti = 65, .irq_parent = 144 },
>>> +       { .exti = 68, .irq_parent = 143 },
>>> +       { .exti = 73, .irq_parent = 129 },
>>> +};
>>
>>
>> You can use an interrupt-map property rather than put this into the
>> driver.
>
>
> interrupt-map seemed interesting and promising like used in pci host.
> At first sight this property can't be used into node with
> "interrupt-controller" property (see in drivers/of/irq.c function:
> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
> it first, and after lookup the interrupt-map.
>
> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?

I believe the correct thing to do is simply drop "interrupt-controller".

Rob

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-05-02 17:45         ` Rob Herring
  (?)
@ 2018-05-03  9:55           ` Ludovic BARRE
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-03  9:55 UTC (permalink / raw)
  To: Rob Herring
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
	devicetree



On 05/02/2018 07:45 PM, Rob Herring wrote:
> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>> Hi Rob
>>
>>
>>
>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>
>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>
>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>
>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>> masked and unmasked according to the needs.
>>>>
>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>> ---
>>>>    .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>    drivers/irqchip/irq-stm32-exti.c                   | 322
>>>> +++++++++++++++++++++
>>>>    2 files changed, 325 insertions(+)
>>>>
>>>> diff --git
>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> index edf03f0..136bd61 100644
>>>> ---
>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> +++
>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>    - compatible: Should be:
>>>>        "st,stm32-exti"
>>>>        "st,stm32h7-exti"
>>>> +    "st,stm32mp1-exti"
>>>>    - reg: Specifies base physical address and size of the registers
>>>>    - interrupt-controller: Indentifies the node as an interrupt controller
>>>>    - #interrupt-cells: Specifies the number of cells to encode an
>>>> interrupt
>>>>      specifier, shall be 2
>>>>    - interrupts: interrupts references to primary interrupt controller
>>>> +  (only needed for exti controller with multiple exti under
>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>      Example:
>>>>    diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>> index b38c655..ebf7146 100644
>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>
>>>
>>> [...]
>>>
>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>> +       { .exti = 1, .irq_parent = 7 },
>>>> +       { .exti = 2, .irq_parent = 8 },
>>>> +       { .exti = 3, .irq_parent = 9 },
>>>> +       { .exti = 4, .irq_parent = 10 },
>>>> +       { .exti = 5, .irq_parent = 23 },
>>>> +       { .exti = 6, .irq_parent = 64 },
>>>> +       { .exti = 7, .irq_parent = 65 },
>>>> +       { .exti = 8, .irq_parent = 66 },
>>>> +       { .exti = 9, .irq_parent = 67 },
>>>> +       { .exti = 10, .irq_parent = 40 },
>>>> +       { .exti = 11, .irq_parent = 42 },
>>>> +       { .exti = 12, .irq_parent = 76 },
>>>> +       { .exti = 13, .irq_parent = 77 },
>>>> +       { .exti = 14, .irq_parent = 121 },
>>>> +       { .exti = 15, .irq_parent = 127 },
>>>> +       { .exti = 16, .irq_parent = 1 },
>>>> +       { .exti = 65, .irq_parent = 144 },
>>>> +       { .exti = 68, .irq_parent = 143 },
>>>> +       { .exti = 73, .irq_parent = 129 },
>>>> +};
>>>
>>>
>>> You can use an interrupt-map property rather than put this into the
>>> driver.
>>
>>
>> interrupt-map seemed interesting and promising like used in pci host.
>> At first sight this property can't be used into node with
>> "interrupt-controller" property (see in drivers/of/irq.c function:
>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>> it first, and after lookup the interrupt-map.
>>
>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
> 
> I believe the correct thing to do is simply drop "interrupt-controller".

if I drop "interrupt-controller" of my node, my driver will not be
initialized by "of_irq_init"
(start_kernel->init_IRQ->irqchip_init->of_irq_init).
Probably, we could replace "IRQCHIP_DECLARE" by a
"module_platform_driver" or "postcore_initcall" but I'm not a big fan of
this solution. what do you think ?

> 
> Rob
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-03  9:55           ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-03  9:55 UTC (permalink / raw)
  To: Rob Herring
  Cc: devicetree, Jason Cooper, Loic PALLARDY, Marc Zyngier,
	linux-kernel, Maxime Coquelin, Gerald BAEZA, Thomas Gleixner,
	moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
	Alexandre Torgue



On 05/02/2018 07:45 PM, Rob Herring wrote:
> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>> Hi Rob
>>
>>
>>
>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>
>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>
>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>
>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>> masked and unmasked according to the needs.
>>>>
>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>> ---
>>>>    .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>    drivers/irqchip/irq-stm32-exti.c                   | 322
>>>> +++++++++++++++++++++
>>>>    2 files changed, 325 insertions(+)
>>>>
>>>> diff --git
>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> index edf03f0..136bd61 100644
>>>> ---
>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> +++
>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>    - compatible: Should be:
>>>>        "st,stm32-exti"
>>>>        "st,stm32h7-exti"
>>>> +    "st,stm32mp1-exti"
>>>>    - reg: Specifies base physical address and size of the registers
>>>>    - interrupt-controller: Indentifies the node as an interrupt controller
>>>>    - #interrupt-cells: Specifies the number of cells to encode an
>>>> interrupt
>>>>      specifier, shall be 2
>>>>    - interrupts: interrupts references to primary interrupt controller
>>>> +  (only needed for exti controller with multiple exti under
>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>      Example:
>>>>    diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>> index b38c655..ebf7146 100644
>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>
>>>
>>> [...]
>>>
>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>> +       { .exti = 1, .irq_parent = 7 },
>>>> +       { .exti = 2, .irq_parent = 8 },
>>>> +       { .exti = 3, .irq_parent = 9 },
>>>> +       { .exti = 4, .irq_parent = 10 },
>>>> +       { .exti = 5, .irq_parent = 23 },
>>>> +       { .exti = 6, .irq_parent = 64 },
>>>> +       { .exti = 7, .irq_parent = 65 },
>>>> +       { .exti = 8, .irq_parent = 66 },
>>>> +       { .exti = 9, .irq_parent = 67 },
>>>> +       { .exti = 10, .irq_parent = 40 },
>>>> +       { .exti = 11, .irq_parent = 42 },
>>>> +       { .exti = 12, .irq_parent = 76 },
>>>> +       { .exti = 13, .irq_parent = 77 },
>>>> +       { .exti = 14, .irq_parent = 121 },
>>>> +       { .exti = 15, .irq_parent = 127 },
>>>> +       { .exti = 16, .irq_parent = 1 },
>>>> +       { .exti = 65, .irq_parent = 144 },
>>>> +       { .exti = 68, .irq_parent = 143 },
>>>> +       { .exti = 73, .irq_parent = 129 },
>>>> +};
>>>
>>>
>>> You can use an interrupt-map property rather than put this into the
>>> driver.
>>
>>
>> interrupt-map seemed interesting and promising like used in pci host.
>> At first sight this property can't be used into node with
>> "interrupt-controller" property (see in drivers/of/irq.c function:
>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>> it first, and after lookup the interrupt-map.
>>
>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
> 
> I believe the correct thing to do is simply drop "interrupt-controller".

if I drop "interrupt-controller" of my node, my driver will not be
initialized by "of_irq_init"
(start_kernel->init_IRQ->irqchip_init->of_irq_init).
Probably, we could replace "IRQCHIP_DECLARE" by a
"module_platform_driver" or "postcore_initcall" but I'm not a big fan of
this solution. what do you think ?

> 
> Rob
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-03  9:55           ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-03  9:55 UTC (permalink / raw)
  To: linux-arm-kernel



On 05/02/2018 07:45 PM, Rob Herring wrote:
> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>> Hi Rob
>>
>>
>>
>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>
>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>
>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>
>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>> masked and unmasked according to the needs.
>>>>
>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>> ---
>>>>    .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>    drivers/irqchip/irq-stm32-exti.c                   | 322
>>>> +++++++++++++++++++++
>>>>    2 files changed, 325 insertions(+)
>>>>
>>>> diff --git
>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> index edf03f0..136bd61 100644
>>>> ---
>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> +++
>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>    - compatible: Should be:
>>>>        "st,stm32-exti"
>>>>        "st,stm32h7-exti"
>>>> +    "st,stm32mp1-exti"
>>>>    - reg: Specifies base physical address and size of the registers
>>>>    - interrupt-controller: Indentifies the node as an interrupt controller
>>>>    - #interrupt-cells: Specifies the number of cells to encode an
>>>> interrupt
>>>>      specifier, shall be 2
>>>>    - interrupts: interrupts references to primary interrupt controller
>>>> +  (only needed for exti controller with multiple exti under
>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>      Example:
>>>>    diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>> index b38c655..ebf7146 100644
>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>
>>>
>>> [...]
>>>
>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>> +       { .exti = 1, .irq_parent = 7 },
>>>> +       { .exti = 2, .irq_parent = 8 },
>>>> +       { .exti = 3, .irq_parent = 9 },
>>>> +       { .exti = 4, .irq_parent = 10 },
>>>> +       { .exti = 5, .irq_parent = 23 },
>>>> +       { .exti = 6, .irq_parent = 64 },
>>>> +       { .exti = 7, .irq_parent = 65 },
>>>> +       { .exti = 8, .irq_parent = 66 },
>>>> +       { .exti = 9, .irq_parent = 67 },
>>>> +       { .exti = 10, .irq_parent = 40 },
>>>> +       { .exti = 11, .irq_parent = 42 },
>>>> +       { .exti = 12, .irq_parent = 76 },
>>>> +       { .exti = 13, .irq_parent = 77 },
>>>> +       { .exti = 14, .irq_parent = 121 },
>>>> +       { .exti = 15, .irq_parent = 127 },
>>>> +       { .exti = 16, .irq_parent = 1 },
>>>> +       { .exti = 65, .irq_parent = 144 },
>>>> +       { .exti = 68, .irq_parent = 143 },
>>>> +       { .exti = 73, .irq_parent = 129 },
>>>> +};
>>>
>>>
>>> You can use an interrupt-map property rather than put this into the
>>> driver.
>>
>>
>> interrupt-map seemed interesting and promising like used in pci host.
>> At first sight this property can't be used into node with
>> "interrupt-controller" property (see in drivers/of/irq.c function:
>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>> it first, and after lookup the interrupt-map.
>>
>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
> 
> I believe the correct thing to do is simply drop "interrupt-controller".

if I drop "interrupt-controller" of my node, my driver will not be
initialized by "of_irq_init"
(start_kernel->init_IRQ->irqchip_init->of_irq_init).
Probably, we could replace "IRQCHIP_DECLARE" by a
"module_platform_driver" or "postcore_initcall" but I'm not a big fan of
this solution. what do you think ?

> 
> Rob
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-05-03  9:55           ` Ludovic BARRE
@ 2018-05-04 20:38             ` Rob Herring
  -1 siblings, 0 replies; 56+ messages in thread
From: Rob Herring @ 2018-05-04 20:38 UTC (permalink / raw)
  To: Ludovic BARRE
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
	devicetree

On Thu, May 3, 2018 at 4:55 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>
>
> On 05/02/2018 07:45 PM, Rob Herring wrote:
>>
>> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com>
>> wrote:
>>>
>>> Hi Rob
>>>
>>>
>>>
>>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>>
>>>>
>>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>>
>>>>>
>>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>>
>>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>>> masked and unmasked according to the needs.
>>>>>
>>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>>> ---
>>>>>    .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>>    drivers/irqchip/irq-stm32-exti.c                   | 322
>>>>> +++++++++++++++++++++
>>>>>    2 files changed, 325 insertions(+)
>>>>>
>>>>> diff --git
>>>>>
>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>
>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>> index edf03f0..136bd61 100644
>>>>> ---
>>>>>
>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>> +++
>>>>>
>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>>    - compatible: Should be:
>>>>>        "st,stm32-exti"
>>>>>        "st,stm32h7-exti"
>>>>> +    "st,stm32mp1-exti"
>>>>>    - reg: Specifies base physical address and size of the registers
>>>>>    - interrupt-controller: Indentifies the node as an interrupt
>>>>> controller
>>>>>    - #interrupt-cells: Specifies the number of cells to encode an
>>>>> interrupt
>>>>>      specifier, shall be 2
>>>>>    - interrupts: interrupts references to primary interrupt controller
>>>>> +  (only needed for exti controller with multiple exti under
>>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>>      Example:
>>>>>    diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>>> index b38c655..ebf7146 100644
>>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>>
>>>>
>>>>
>>>> [...]
>>>>
>>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>>> +       { .exti = 1, .irq_parent = 7 },
>>>>> +       { .exti = 2, .irq_parent = 8 },
>>>>> +       { .exti = 3, .irq_parent = 9 },
>>>>> +       { .exti = 4, .irq_parent = 10 },
>>>>> +       { .exti = 5, .irq_parent = 23 },
>>>>> +       { .exti = 6, .irq_parent = 64 },
>>>>> +       { .exti = 7, .irq_parent = 65 },
>>>>> +       { .exti = 8, .irq_parent = 66 },
>>>>> +       { .exti = 9, .irq_parent = 67 },
>>>>> +       { .exti = 10, .irq_parent = 40 },
>>>>> +       { .exti = 11, .irq_parent = 42 },
>>>>> +       { .exti = 12, .irq_parent = 76 },
>>>>> +       { .exti = 13, .irq_parent = 77 },
>>>>> +       { .exti = 14, .irq_parent = 121 },
>>>>> +       { .exti = 15, .irq_parent = 127 },
>>>>> +       { .exti = 16, .irq_parent = 1 },
>>>>> +       { .exti = 65, .irq_parent = 144 },
>>>>> +       { .exti = 68, .irq_parent = 143 },
>>>>> +       { .exti = 73, .irq_parent = 129 },
>>>>> +};
>>>>
>>>>
>>>>
>>>> You can use an interrupt-map property rather than put this into the
>>>> driver.
>>>
>>>
>>>
>>> interrupt-map seemed interesting and promising like used in pci host.
>>> At first sight this property can't be used into node with
>>> "interrupt-controller" property (see in drivers/of/irq.c function:
>>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>>> it first, and after lookup the interrupt-map.
>>>
>>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
>>
>>
>> I believe the correct thing to do is simply drop "interrupt-controller".
>

Actually, that's not right.

> if I drop "interrupt-controller" of my node, my driver will not be
> initialized by "of_irq_init"
> (start_kernel->init_IRQ->irqchip_init->of_irq_init).
> Probably, we could replace "IRQCHIP_DECLARE" by a
> "module_platform_driver" or "postcore_initcall" but I'm not a big fan of
> this solution. what do you think ?

You'd have to parse 'interrupt-map' yourself to extract the data for
this to work because interrupt-map currently only works when the
translation is transparent (i.e. doesn't need s/w handling). So I
guess leave this in the driver. Sorry for the noise.

Rob

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-04 20:38             ` Rob Herring
  0 siblings, 0 replies; 56+ messages in thread
From: Rob Herring @ 2018-05-04 20:38 UTC (permalink / raw)
  To: linux-arm-kernel

On Thu, May 3, 2018 at 4:55 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>
>
> On 05/02/2018 07:45 PM, Rob Herring wrote:
>>
>> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com>
>> wrote:
>>>
>>> Hi Rob
>>>
>>>
>>>
>>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>>
>>>>
>>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>>
>>>>>
>>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>>
>>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>>> masked and unmasked according to the needs.
>>>>>
>>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>>> ---
>>>>>    .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>>    drivers/irqchip/irq-stm32-exti.c                   | 322
>>>>> +++++++++++++++++++++
>>>>>    2 files changed, 325 insertions(+)
>>>>>
>>>>> diff --git
>>>>>
>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>
>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>> index edf03f0..136bd61 100644
>>>>> ---
>>>>>
>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>> +++
>>>>>
>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>>    - compatible: Should be:
>>>>>        "st,stm32-exti"
>>>>>        "st,stm32h7-exti"
>>>>> +    "st,stm32mp1-exti"
>>>>>    - reg: Specifies base physical address and size of the registers
>>>>>    - interrupt-controller: Indentifies the node as an interrupt
>>>>> controller
>>>>>    - #interrupt-cells: Specifies the number of cells to encode an
>>>>> interrupt
>>>>>      specifier, shall be 2
>>>>>    - interrupts: interrupts references to primary interrupt controller
>>>>> +  (only needed for exti controller with multiple exti under
>>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>>      Example:
>>>>>    diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>>> index b38c655..ebf7146 100644
>>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>>
>>>>
>>>>
>>>> [...]
>>>>
>>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>>> +       { .exti = 1, .irq_parent = 7 },
>>>>> +       { .exti = 2, .irq_parent = 8 },
>>>>> +       { .exti = 3, .irq_parent = 9 },
>>>>> +       { .exti = 4, .irq_parent = 10 },
>>>>> +       { .exti = 5, .irq_parent = 23 },
>>>>> +       { .exti = 6, .irq_parent = 64 },
>>>>> +       { .exti = 7, .irq_parent = 65 },
>>>>> +       { .exti = 8, .irq_parent = 66 },
>>>>> +       { .exti = 9, .irq_parent = 67 },
>>>>> +       { .exti = 10, .irq_parent = 40 },
>>>>> +       { .exti = 11, .irq_parent = 42 },
>>>>> +       { .exti = 12, .irq_parent = 76 },
>>>>> +       { .exti = 13, .irq_parent = 77 },
>>>>> +       { .exti = 14, .irq_parent = 121 },
>>>>> +       { .exti = 15, .irq_parent = 127 },
>>>>> +       { .exti = 16, .irq_parent = 1 },
>>>>> +       { .exti = 65, .irq_parent = 144 },
>>>>> +       { .exti = 68, .irq_parent = 143 },
>>>>> +       { .exti = 73, .irq_parent = 129 },
>>>>> +};
>>>>
>>>>
>>>>
>>>> You can use an interrupt-map property rather than put this into the
>>>> driver.
>>>
>>>
>>>
>>> interrupt-map seemed interesting and promising like used in pci host.
>>> At first sight this property can't be used into node with
>>> "interrupt-controller" property (see in drivers/of/irq.c function:
>>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>>> it first, and after lookup the interrupt-map.
>>>
>>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
>>
>>
>> I believe the correct thing to do is simply drop "interrupt-controller".
>

Actually, that's not right.

> if I drop "interrupt-controller" of my node, my driver will not be
> initialized by "of_irq_init"
> (start_kernel->init_IRQ->irqchip_init->of_irq_init).
> Probably, we could replace "IRQCHIP_DECLARE" by a
> "module_platform_driver" or "postcore_initcall" but I'm not a big fan of
> this solution. what do you think ?

You'd have to parse 'interrupt-map' yourself to extract the data for
this to work because interrupt-map currently only works when the
translation is transparent (i.e. doesn't need s/w handling). So I
guess leave this in the driver. Sorry for the noise.

Rob

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
  2018-04-26 16:18   ` Ludovic Barre
@ 2018-05-08 14:47     ` Marc Zyngier
  -1 siblings, 0 replies; 56+ messages in thread
From: Marc Zyngier @ 2018-05-08 14:47 UTC (permalink / raw)
  To: Ludovic Barre, Thomas Gleixner, Jason Cooper, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, radek

On 26/04/18 17:18, Ludovic Barre wrote:
> From: radek <radoslaw.pietrzyk@gmail.com>
> 
> - In stm32_exti_alloc function, discards irq_domain_set_info
>   with handle_simple_irq. This overwrite the setting defined while init
>   of generic chips. Exti controller manages edge irq type.
> - Removes acking in chained irq handler as this is done by
>   irq_chip itself inside handle_edge_irq
> - removes unneeded irq_domain_ops.xlate callback
> 
> Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
> Acked-by: Ludovic Barre <ludovic.barre@st.com>
> Tested-by: Ludovic Barre <ludovic.barre@st.com>
> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>

Nit: the "From:" should match the SoB line (the address does, but not
the name). I can fix that up when I apply the series.

Thanks,

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

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
@ 2018-05-08 14:47     ` Marc Zyngier
  0 siblings, 0 replies; 56+ messages in thread
From: Marc Zyngier @ 2018-05-08 14:47 UTC (permalink / raw)
  To: linux-arm-kernel

On 26/04/18 17:18, Ludovic Barre wrote:
> From: radek <radoslaw.pietrzyk@gmail.com>
> 
> - In stm32_exti_alloc function, discards irq_domain_set_info
>   with handle_simple_irq. This overwrite the setting defined while init
>   of generic chips. Exti controller manages edge irq type.
> - Removes acking in chained irq handler as this is done by
>   irq_chip itself inside handle_edge_irq
> - removes unneeded irq_domain_ops.xlate callback
> 
> Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
> Acked-by: Ludovic Barre <ludovic.barre@st.com>
> Tested-by: Ludovic Barre <ludovic.barre@st.com>
> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>

Nit: the "From:" should match the SoB line (the address does, but not
the name). I can fix that up when I apply the series.

Thanks,

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

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
  2018-05-08 14:47     ` Marc Zyngier
  (?)
@ 2018-05-11  7:47       ` Ludovic BARRE
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-11  7:47 UTC (permalink / raw)
  To: Marc Zyngier, Thomas Gleixner, Jason Cooper, Rob Herring
  Cc: Maxime Coquelin, Alexandre Torgue, Gerald BAEZA, Loic PALLARDY,
	linux-kernel, linux-arm-kernel, devicetree, radek



On 05/08/2018 04:47 PM, Marc Zyngier wrote:
> On 26/04/18 17:18, Ludovic Barre wrote:
>> From: radek <radoslaw.pietrzyk@gmail.com>
>>
>> - In stm32_exti_alloc function, discards irq_domain_set_info
>>    with handle_simple_irq. This overwrite the setting defined while init
>>    of generic chips. Exti controller manages edge irq type.
>> - Removes acking in chained irq handler as this is done by
>>    irq_chip itself inside handle_edge_irq
>> - removes unneeded irq_domain_ops.xlate callback
>>
>> Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
>> Acked-by: Ludovic Barre <ludovic.barre@st.com>
>> Tested-by: Ludovic Barre <ludovic.barre@st.com>
>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
> 
> Nit: the "From:" should match the SoB line (the address does, but not
> the name). I can fix that up when I apply the series.
> 

Thanks

> Thanks,
> 
> 	M.
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
@ 2018-05-11  7:47       ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-11  7:47 UTC (permalink / raw)
  To: Marc Zyngier, Thomas Gleixner, Jason Cooper, Rob Herring
  Cc: devicetree, radek, Alexandre Torgue, Loic PALLARDY, linux-kernel,
	Maxime Coquelin, Gerald BAEZA, linux-arm-kernel



On 05/08/2018 04:47 PM, Marc Zyngier wrote:
> On 26/04/18 17:18, Ludovic Barre wrote:
>> From: radek <radoslaw.pietrzyk@gmail.com>
>>
>> - In stm32_exti_alloc function, discards irq_domain_set_info
>>    with handle_simple_irq. This overwrite the setting defined while init
>>    of generic chips. Exti controller manages edge irq type.
>> - Removes acking in chained irq handler as this is done by
>>    irq_chip itself inside handle_edge_irq
>> - removes unneeded irq_domain_ops.xlate callback
>>
>> Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
>> Acked-by: Ludovic Barre <ludovic.barre@st.com>
>> Tested-by: Ludovic Barre <ludovic.barre@st.com>
>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
> 
> Nit: the "From:" should match the SoB line (the address does, but not
> the name). I can fix that up when I apply the series.
> 

Thanks

> Thanks,
> 
> 	M.
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain
@ 2018-05-11  7:47       ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-11  7:47 UTC (permalink / raw)
  To: linux-arm-kernel



On 05/08/2018 04:47 PM, Marc Zyngier wrote:
> On 26/04/18 17:18, Ludovic Barre wrote:
>> From: radek <radoslaw.pietrzyk@gmail.com>
>>
>> - In stm32_exti_alloc function, discards irq_domain_set_info
>>    with handle_simple_irq. This overwrite the setting defined while init
>>    of generic chips. Exti controller manages edge irq type.
>> - Removes acking in chained irq handler as this is done by
>>    irq_chip itself inside handle_edge_irq
>> - removes unneeded irq_domain_ops.xlate callback
>>
>> Signed-off-by: Radoslaw Pietrzyk <radoslaw.pietrzyk@gmail.com>
>> Acked-by: Ludovic Barre <ludovic.barre@st.com>
>> Tested-by: Ludovic Barre <ludovic.barre@st.com>
>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
> 
> Nit: the "From:" should match the SoB line (the address does, but not
> the name). I can fix that up when I apply the series.
> 

Thanks

> Thanks,
> 
> 	M.
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
  2018-05-04 20:38             ` Rob Herring
  (?)
@ 2018-05-14 12:40               ` Ludovic BARRE
  -1 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-14 12:40 UTC (permalink / raw)
  To: Rob Herring
  Cc: Thomas Gleixner, Jason Cooper, Marc Zyngier, Maxime Coquelin,
	Alexandre Torgue, Gerald BAEZA, Loic PALLARDY, linux-kernel,
	moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
	devicetree



On 05/04/2018 10:38 PM, Rob Herring wrote:
> On Thu, May 3, 2018 at 4:55 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>>
>>
>> On 05/02/2018 07:45 PM, Rob Herring wrote:
>>>
>>> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com>
>>> wrote:
>>>>
>>>> Hi Rob
>>>>
>>>>
>>>>
>>>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>>>
>>>>>
>>>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>>>
>>>>>>
>>>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>>>
>>>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>>>> masked and unmasked according to the needs.
>>>>>>
>>>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>>>> ---
>>>>>>     .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>>>     drivers/irqchip/irq-stm32-exti.c                   | 322
>>>>>> +++++++++++++++++++++
>>>>>>     2 files changed, 325 insertions(+)
>>>>>>
>>>>>> diff --git
>>>>>>
>>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>>
>>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> index edf03f0..136bd61 100644
>>>>>> ---
>>>>>>
>>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> +++
>>>>>>
>>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>>>     - compatible: Should be:
>>>>>>         "st,stm32-exti"
>>>>>>         "st,stm32h7-exti"
>>>>>> +    "st,stm32mp1-exti"
>>>>>>     - reg: Specifies base physical address and size of the registers
>>>>>>     - interrupt-controller: Indentifies the node as an interrupt
>>>>>> controller
>>>>>>     - #interrupt-cells: Specifies the number of cells to encode an
>>>>>> interrupt
>>>>>>       specifier, shall be 2
>>>>>>     - interrupts: interrupts references to primary interrupt controller
>>>>>> +  (only needed for exti controller with multiple exti under
>>>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>>>       Example:
>>>>>>     diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>>>> index b38c655..ebf7146 100644
>>>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>>>
>>>>>
>>>>>
>>>>> [...]
>>>>>
>>>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>>>> +       { .exti = 1, .irq_parent = 7 },
>>>>>> +       { .exti = 2, .irq_parent = 8 },
>>>>>> +       { .exti = 3, .irq_parent = 9 },
>>>>>> +       { .exti = 4, .irq_parent = 10 },
>>>>>> +       { .exti = 5, .irq_parent = 23 },
>>>>>> +       { .exti = 6, .irq_parent = 64 },
>>>>>> +       { .exti = 7, .irq_parent = 65 },
>>>>>> +       { .exti = 8, .irq_parent = 66 },
>>>>>> +       { .exti = 9, .irq_parent = 67 },
>>>>>> +       { .exti = 10, .irq_parent = 40 },
>>>>>> +       { .exti = 11, .irq_parent = 42 },
>>>>>> +       { .exti = 12, .irq_parent = 76 },
>>>>>> +       { .exti = 13, .irq_parent = 77 },
>>>>>> +       { .exti = 14, .irq_parent = 121 },
>>>>>> +       { .exti = 15, .irq_parent = 127 },
>>>>>> +       { .exti = 16, .irq_parent = 1 },
>>>>>> +       { .exti = 65, .irq_parent = 144 },
>>>>>> +       { .exti = 68, .irq_parent = 143 },
>>>>>> +       { .exti = 73, .irq_parent = 129 },
>>>>>> +};
>>>>>
>>>>>
>>>>>
>>>>> You can use an interrupt-map property rather than put this into the
>>>>> driver.
>>>>
>>>>
>>>>
>>>> interrupt-map seemed interesting and promising like used in pci host.
>>>> At first sight this property can't be used into node with
>>>> "interrupt-controller" property (see in drivers/of/irq.c function:
>>>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>>>> it first, and after lookup the interrupt-map.
>>>>
>>>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
>>>
>>>
>>> I believe the correct thing to do is simply drop "interrupt-controller".
>>
> 
> Actually, that's not right.
> 
>> if I drop "interrupt-controller" of my node, my driver will not be
>> initialized by "of_irq_init"
>> (start_kernel->init_IRQ->irqchip_init->of_irq_init).
>> Probably, we could replace "IRQCHIP_DECLARE" by a
>> "module_platform_driver" or "postcore_initcall" but I'm not a big fan of
>> this solution. what do you think ?
> 
> You'd have to parse 'interrupt-map' yourself to extract the data for
> this to work because interrupt-map currently only works when the
> translation is transparent (i.e. doesn't need s/w handling). So I
> guess leave this in the driver. Sorry for the noise.
> 

Sorry, I don't know if I've correctly understand the answer.
I hesitate between:
- keep the code like this, with stm32mp1_desc_irq tab.
- parse interrupt-map property into stm32 exti driver.

Just to be sure, and avoid misunderstand.

BR
Ludo

> Rob
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* Re: [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-14 12:40               ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-14 12:40 UTC (permalink / raw)
  To: Rob Herring
  Cc: devicetree, Jason Cooper, Loic PALLARDY, Marc Zyngier,
	linux-kernel, Maxime Coquelin, Gerald BAEZA, Thomas Gleixner,
	moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
	Alexandre Torgue



On 05/04/2018 10:38 PM, Rob Herring wrote:
> On Thu, May 3, 2018 at 4:55 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>>
>>
>> On 05/02/2018 07:45 PM, Rob Herring wrote:
>>>
>>> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com>
>>> wrote:
>>>>
>>>> Hi Rob
>>>>
>>>>
>>>>
>>>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>>>
>>>>>
>>>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>>>
>>>>>>
>>>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>>>
>>>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>>>> masked and unmasked according to the needs.
>>>>>>
>>>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>>>> ---
>>>>>>     .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>>>     drivers/irqchip/irq-stm32-exti.c                   | 322
>>>>>> +++++++++++++++++++++
>>>>>>     2 files changed, 325 insertions(+)
>>>>>>
>>>>>> diff --git
>>>>>>
>>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>>
>>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> index edf03f0..136bd61 100644
>>>>>> ---
>>>>>>
>>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> +++
>>>>>>
>>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>>>     - compatible: Should be:
>>>>>>         "st,stm32-exti"
>>>>>>         "st,stm32h7-exti"
>>>>>> +    "st,stm32mp1-exti"
>>>>>>     - reg: Specifies base physical address and size of the registers
>>>>>>     - interrupt-controller: Indentifies the node as an interrupt
>>>>>> controller
>>>>>>     - #interrupt-cells: Specifies the number of cells to encode an
>>>>>> interrupt
>>>>>>       specifier, shall be 2
>>>>>>     - interrupts: interrupts references to primary interrupt controller
>>>>>> +  (only needed for exti controller with multiple exti under
>>>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>>>       Example:
>>>>>>     diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>>>> index b38c655..ebf7146 100644
>>>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>>>
>>>>>
>>>>>
>>>>> [...]
>>>>>
>>>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>>>> +       { .exti = 1, .irq_parent = 7 },
>>>>>> +       { .exti = 2, .irq_parent = 8 },
>>>>>> +       { .exti = 3, .irq_parent = 9 },
>>>>>> +       { .exti = 4, .irq_parent = 10 },
>>>>>> +       { .exti = 5, .irq_parent = 23 },
>>>>>> +       { .exti = 6, .irq_parent = 64 },
>>>>>> +       { .exti = 7, .irq_parent = 65 },
>>>>>> +       { .exti = 8, .irq_parent = 66 },
>>>>>> +       { .exti = 9, .irq_parent = 67 },
>>>>>> +       { .exti = 10, .irq_parent = 40 },
>>>>>> +       { .exti = 11, .irq_parent = 42 },
>>>>>> +       { .exti = 12, .irq_parent = 76 },
>>>>>> +       { .exti = 13, .irq_parent = 77 },
>>>>>> +       { .exti = 14, .irq_parent = 121 },
>>>>>> +       { .exti = 15, .irq_parent = 127 },
>>>>>> +       { .exti = 16, .irq_parent = 1 },
>>>>>> +       { .exti = 65, .irq_parent = 144 },
>>>>>> +       { .exti = 68, .irq_parent = 143 },
>>>>>> +       { .exti = 73, .irq_parent = 129 },
>>>>>> +};
>>>>>
>>>>>
>>>>>
>>>>> You can use an interrupt-map property rather than put this into the
>>>>> driver.
>>>>
>>>>
>>>>
>>>> interrupt-map seemed interesting and promising like used in pci host.
>>>> At first sight this property can't be used into node with
>>>> "interrupt-controller" property (see in drivers/of/irq.c function:
>>>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>>>> it first, and after lookup the interrupt-map.
>>>>
>>>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
>>>
>>>
>>> I believe the correct thing to do is simply drop "interrupt-controller".
>>
> 
> Actually, that's not right.
> 
>> if I drop "interrupt-controller" of my node, my driver will not be
>> initialized by "of_irq_init"
>> (start_kernel->init_IRQ->irqchip_init->of_irq_init).
>> Probably, we could replace "IRQCHIP_DECLARE" by a
>> "module_platform_driver" or "postcore_initcall" but I'm not a big fan of
>> this solution. what do you think ?
> 
> You'd have to parse 'interrupt-map' yourself to extract the data for
> this to work because interrupt-map currently only works when the
> translation is transparent (i.e. doesn't need s/w handling). So I
> guess leave this in the driver. Sorry for the noise.
> 

Sorry, I don't know if I've correctly understand the answer.
I hesitate between:
- keep the code like this, with stm32mp1_desc_irq tab.
- parse interrupt-map property into stm32 exti driver.

Just to be sure, and avoid misunderstand.

BR
Ludo

> Rob
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

* [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain
@ 2018-05-14 12:40               ` Ludovic BARRE
  0 siblings, 0 replies; 56+ messages in thread
From: Ludovic BARRE @ 2018-05-14 12:40 UTC (permalink / raw)
  To: linux-arm-kernel



On 05/04/2018 10:38 PM, Rob Herring wrote:
> On Thu, May 3, 2018 at 4:55 AM, Ludovic BARRE <ludovic.barre@st.com> wrote:
>>
>>
>> On 05/02/2018 07:45 PM, Rob Herring wrote:
>>>
>>> On Wed, May 2, 2018 at 11:03 AM, Ludovic BARRE <ludovic.barre@st.com>
>>> wrote:
>>>>
>>>> Hi Rob
>>>>
>>>>
>>>>
>>>> On 05/01/2018 04:56 PM, Rob Herring wrote:
>>>>>
>>>>>
>>>>> On Thu, Apr 26, 2018 at 06:18:30PM +0200, Ludovic Barre wrote:
>>>>>>
>>>>>>
>>>>>> From: Ludovic Barre <ludovic.barre@st.com>
>>>>>>
>>>>>> Exti controller has been differently integrated on stm32mp1 SoC.
>>>>>> A parent irq has only one external interrupt. A hierachy domain could
>>>>>> be used. Handlers are call by parent, each parent interrupt could be
>>>>>> masked and unmasked according to the needs.
>>>>>>
>>>>>> Signed-off-by: Ludovic Barre <ludovic.barre@st.com>
>>>>>> ---
>>>>>>     .../interrupt-controller/st,stm32-exti.txt         |   3 +
>>>>>>     drivers/irqchip/irq-stm32-exti.c                   | 322
>>>>>> +++++++++++++++++++++
>>>>>>     2 files changed, 325 insertions(+)
>>>>>>
>>>>>> diff --git
>>>>>>
>>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>>
>>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> index edf03f0..136bd61 100644
>>>>>> ---
>>>>>>
>>>>>> a/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> +++
>>>>>>
>>>>>> b/Documentation/devicetree/bindings/interrupt-controller/st,stm32-exti.txt
>>>>>> @@ -5,11 +5,14 @@ Required properties:
>>>>>>     - compatible: Should be:
>>>>>>         "st,stm32-exti"
>>>>>>         "st,stm32h7-exti"
>>>>>> +    "st,stm32mp1-exti"
>>>>>>     - reg: Specifies base physical address and size of the registers
>>>>>>     - interrupt-controller: Indentifies the node as an interrupt
>>>>>> controller
>>>>>>     - #interrupt-cells: Specifies the number of cells to encode an
>>>>>> interrupt
>>>>>>       specifier, shall be 2
>>>>>>     - interrupts: interrupts references to primary interrupt controller
>>>>>> +  (only needed for exti controller with multiple exti under
>>>>>> +  same parent interrupt: st,stm32-exti and st,stm32h7-exti")
>>>>>>       Example:
>>>>>>     diff --git a/drivers/irqchip/irq-stm32-exti.c
>>>>>> b/drivers/irqchip/irq-stm32-exti.c
>>>>>> index b38c655..ebf7146 100644
>>>>>> --- a/drivers/irqchip/irq-stm32-exti.c
>>>>>> +++ b/drivers/irqchip/irq-stm32-exti.c
>>>>>
>>>>>
>>>>>
>>>>> [...]
>>>>>
>>>>>> +static const struct stm32_desc_irq stm32mp1_desc_irq[] = {
>>>>>> +       { .exti = 1, .irq_parent = 7 },
>>>>>> +       { .exti = 2, .irq_parent = 8 },
>>>>>> +       { .exti = 3, .irq_parent = 9 },
>>>>>> +       { .exti = 4, .irq_parent = 10 },
>>>>>> +       { .exti = 5, .irq_parent = 23 },
>>>>>> +       { .exti = 6, .irq_parent = 64 },
>>>>>> +       { .exti = 7, .irq_parent = 65 },
>>>>>> +       { .exti = 8, .irq_parent = 66 },
>>>>>> +       { .exti = 9, .irq_parent = 67 },
>>>>>> +       { .exti = 10, .irq_parent = 40 },
>>>>>> +       { .exti = 11, .irq_parent = 42 },
>>>>>> +       { .exti = 12, .irq_parent = 76 },
>>>>>> +       { .exti = 13, .irq_parent = 77 },
>>>>>> +       { .exti = 14, .irq_parent = 121 },
>>>>>> +       { .exti = 15, .irq_parent = 127 },
>>>>>> +       { .exti = 16, .irq_parent = 1 },
>>>>>> +       { .exti = 65, .irq_parent = 144 },
>>>>>> +       { .exti = 68, .irq_parent = 143 },
>>>>>> +       { .exti = 73, .irq_parent = 129 },
>>>>>> +};
>>>>>
>>>>>
>>>>>
>>>>> You can use an interrupt-map property rather than put this into the
>>>>> driver.
>>>>
>>>>
>>>>
>>>> interrupt-map seemed interesting and promising like used in pci host.
>>>> At first sight this property can't be used into node with
>>>> "interrupt-controller" property (see in drivers/of/irq.c function:
>>>> of_irq_parse_raw) because "of_irq_parse_raw" checks if node got
>>>> it first, and after lookup the interrupt-map.
>>>>
>>>> Rob, Thomas, Jason, Marc what do you prefers or the right ways...?
>>>
>>>
>>> I believe the correct thing to do is simply drop "interrupt-controller".
>>
> 
> Actually, that's not right.
> 
>> if I drop "interrupt-controller" of my node, my driver will not be
>> initialized by "of_irq_init"
>> (start_kernel->init_IRQ->irqchip_init->of_irq_init).
>> Probably, we could replace "IRQCHIP_DECLARE" by a
>> "module_platform_driver" or "postcore_initcall" but I'm not a big fan of
>> this solution. what do you think ?
> 
> You'd have to parse 'interrupt-map' yourself to extract the data for
> this to work because interrupt-map currently only works when the
> translation is transparent (i.e. doesn't need s/w handling). So I
> guess leave this in the driver. Sorry for the noise.
> 

Sorry, I don't know if I've correctly understand the answer.
I hesitate between:
- keep the code like this, with stm32mp1_desc_irq tab.
- parse interrupt-map property into stm32 exti driver.

Just to be sure, and avoid misunderstand.

BR
Ludo

> Rob
> 

^ permalink raw reply	[flat|nested] 56+ messages in thread

end of thread, other threads:[~2018-05-14 12:42 UTC | newest]

Thread overview: 56+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-04-26 16:18 [PATCH 00/11] irqchip: stm32: add exti support for stm32mp157c Ludovic Barre
2018-04-26 16:18 ` Ludovic Barre
2018-04-26 16:18 ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 01/11] irqchip: stm32: Optimizes and cleans up stm32-exti irq_domain Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-05-08 14:47   ` Marc Zyngier
2018-05-08 14:47     ` Marc Zyngier
2018-05-11  7:47     ` Ludovic BARRE
2018-05-11  7:47       ` Ludovic BARRE
2018-05-11  7:47       ` Ludovic BARRE
2018-04-26 16:18 ` [PATCH 02/11] irqchip: stm32: checkpatch fix Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 03/11] irqchip: stm32: add falling pending register support Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 04/11] irqchip: stm32: add suspend support Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 05/11] irqchip: stm32: add host and driver data structures Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 06/11] irqchip: stm32: prepare common functions Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 07/11] irqchip: stm32: add stm32mp1 support with hierarchy domain Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-05-01 14:56   ` Rob Herring
2018-05-01 14:56     ` Rob Herring
2018-05-02 16:03     ` Ludovic BARRE
2018-05-02 16:03       ` Ludovic BARRE
2018-05-02 16:03       ` Ludovic BARRE
2018-05-02 17:45       ` Rob Herring
2018-05-02 17:45         ` Rob Herring
2018-05-03  9:55         ` Ludovic BARRE
2018-05-03  9:55           ` Ludovic BARRE
2018-05-03  9:55           ` Ludovic BARRE
2018-05-04 20:38           ` Rob Herring
2018-05-04 20:38             ` Rob Herring
2018-05-14 12:40             ` Ludovic BARRE
2018-05-14 12:40               ` Ludovic BARRE
2018-05-14 12:40               ` Ludovic BARRE
2018-04-26 16:18 ` [PATCH 08/11] irqchip: stm32: add suspend/resume support for " Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 09/11] pinctrl: stm32: add irq_eoi for stm32gpio irqchip Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 10/11] ARM: dts: stm32: add exti support for stm32mp157c Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18 ` [PATCH 11/11] ARM: dts: stm32: add exti support to stm32mp157 pinctrl Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre
2018-04-26 16:18   ` Ludovic Barre

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.