[PATCH 1/2] pinctrl: at91-pio4: Use irq_set_handler_locked()

[PATCH 1/2] pinctrl: at91-pio4: Use irq_set_handler_locked()

[Nicolas Ferre]

Use irq_set_handler_locked() as it avoids a redundant lookup of the irq
descriptor.
This adaptation landed on the pinctrl-at91.c driver while this one was being
developped: synchronize with this enhancement and avoid the following
compilation error:

../drivers/pinctrl/pinctrl-at91-pio4.c: In function 'atmel_gpio_irq_set_type':
../drivers/pinctrl/pinctrl-at91-pio4.c:173:3: error: implicit declaration of
function '__irq_set_handler_locked' [-Werror=implicit-function-declaration]
   __irq_set_handler_locked(d->irq, handle_edge_irq);
      ^

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
---
 drivers/pinctrl/pinctrl-at91-pio4.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
index 5e2189f17fe1..1f1a64ef3559 100644
--- a/drivers/pinctrl/pinctrl-at91-pio4.c
+++ b/drivers/pinctrl/pinctrl-at91-pio4.c
@@ -170,23 +170,23 @@ static int atmel_gpio_irq_set_type(struct irq_data *d, unsigned type)
 
 	switch (type) {
 	case IRQ_TYPE_EDGE_RISING:
-		__irq_set_handler_locked(d->irq, handle_edge_irq);
+		irq_set_handler_locked(d, handle_edge_irq);
 		reg |= ATMEL_PIO_CFGR_EVTSEL_RISING;
 		break;
 	case IRQ_TYPE_EDGE_FALLING:
-		__irq_set_handler_locked(d->irq, handle_edge_irq);
+		irq_set_handler_locked(d, handle_edge_irq);
 		reg |= ATMEL_PIO_CFGR_EVTSEL_FALLING;
 		break;
 	case IRQ_TYPE_EDGE_BOTH:
-		__irq_set_handler_locked(d->irq, handle_edge_irq);
+		irq_set_handler_locked(d, handle_edge_irq);
 		reg |= ATMEL_PIO_CFGR_EVTSEL_BOTH;
 		break;
 	case IRQ_TYPE_LEVEL_LOW:
-		__irq_set_handler_locked(d->irq, handle_level_irq);
+		irq_set_handler_locked(d, handle_level_irq);
 		reg |= ATMEL_PIO_CFGR_EVTSEL_LOW;
 		break;
 	case IRQ_TYPE_LEVEL_HIGH:
-		__irq_set_handler_locked(d->irq, handle_level_irq);
+		irq_set_handler_locked(d, handle_level_irq);
 		reg |= ATMEL_PIO_CFGR_EVTSEL_HIGH;
 		break;
 	case IRQ_TYPE_NONE:
-- 
2.1.3

[PATCH 2/2] pinctrl: at91-pio4: Adapt to new irq flow handler prototype

[Nicolas Ferre]

Adapt atmel_gpio_irq_handler() function as the "irq" argument is now removed
from the irq flow handlers prototype in commit
bd0b9ac405e1794d72533c3d487aa65b6b955a0c (genirq: Remove irq argument from irq
flow handlers).

This allows to remove this warning:
../drivers/pinctrl/pinctrl-at91-pio4.c: In function 'atmel_pinctrl_probe':
../drivers/pinctrl/pinctrl-at91-pio4.c:1000:3: warning: passing argument 2 of
'irq_set_chained_handler' from incompatible pointer type [enabled by default]
   irq_set_chained_handler(res->start, atmel_gpio_irq_handler);
   ^
In file included from ../include/linux/gpio/driver.h:6:0,
                 from ../include/asm-generic/gpio.h:12,
                 from ../arch/arm/include/asm/gpio.h:9,
                 from ../include/linux/gpio.h:48,
                 from ../drivers/pinctrl/pinctrl-at91-pio4.c:18:
../include/linux/irq.h:548:1: note: expected 'irq_flow_handler_t' but argument
is of type 'void (*)(unsigned int,  struct irq_desc *)'
 irq_set_chained_handler(unsigned int irq, irq_flow_handler_t handle)

This change appeared while the driver was being reviewed and was being included
in pinctrl tree.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
---
 drivers/pinctrl/pinctrl-at91-pio4.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
index 1f1a64ef3559..cae0748aca25 100644
--- a/drivers/pinctrl/pinctrl-at91-pio4.c
+++ b/drivers/pinctrl/pinctrl-at91-pio4.c
@@ -248,10 +248,11 @@ static struct irq_chip atmel_gpio_irq_chip = {
 	.irq_set_wake	= atmel_gpio_irq_set_wake,
 };
 
-static void atmel_gpio_irq_handler(unsigned int irq, struct irq_desc *desc)
+static void atmel_gpio_irq_handler(struct irq_desc *desc)
 {
-	struct atmel_pioctrl *atmel_pioctrl = irq_get_handler_data(irq);
+	struct atmel_pioctrl *atmel_pioctrl = irq_desc_get_handler_data(desc);
 	struct irq_chip *chip = irq_desc_get_chip(desc);
+	unsigned int irq = irq_desc_get_irq(desc);
 	unsigned long isr;
 	int n, bank = -1;
 
-- 
2.1.3

Re: [PATCH 1/2] pinctrl: at91-pio4: Use irq_set_handler_locked()

[Nicolas Ferre]

Hi all,

Well it seems my cover letter is missing. So, here it is:

---8<----
pinctrl: at91-pio4: fixes for current linux-next (irq)

While testing the sama5d2 on today's linux-next I discovered that some
adaptations were missing. So, I don't know if you were aware but here
are my fixes in case you don't have them already.

+Cc: Thomas and Stephen in case these errors happen while building
linux-next...

Bye,

Re: [PATCH 1/2] pinctrl: at91-pio4: Use irq_set_handler_locked()

[Nicolas Ferre]

Le 02/10/2015 15:57, Nicolas Ferre a écrit :
> Hi all,
> 
> Well it seems my cover letter is missing. So, here it is:
> 
> ---8<----
> pinctrl: at91-pio4: fixes for current linux-next (irq)
> 
> While testing the sama5d2 on today's linux-next I discovered that some
> adaptations were missing. So, I don't know if you were aware but here
> are my fixes in case you don't have them already.
> 
> +Cc: Thomas and Stephen in case these errors happen while building
> linux-next...

After seeing that Linus applied 2 patches by Ludovic, I realized that as
I had foreseen correctly: it was known and corrected.

Sorry for the noise. Bye,
-- 
Nicolas Ferre
--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] pinctrl: at91-pio4: Use irq_set_handler_locked()

[Ludovic Desroches]

Hi Nicolas,

On Fri, Oct 02, 2015 at 03:57:32PM +0200, Nicolas Ferre wrote:
> Hi all,
> 
> Well it seems my cover letter is missing. So, here it is:
> 
> ---8<----
> pinctrl: at91-pio4: fixes for current linux-next (irq)
> 
> While testing the sama5d2 on today's linux-next I discovered that some
> adaptations were missing. So, I don't know if you were aware but here
> are my fixes in case you don't have them already.
> 
> +Cc: Thomas and Stephen in case these errors happen while building
> linux-next...

Yes it happens, fixes sent and taken by Linus.

> 
> Bye,

Ludovic