* [PATCH 0/2] pinctrl: Use platform_get_irq*() variants to fetch IRQ's
@ 2021-12-24 14:57 Lad Prabhakar
2021-12-24 14:57 ` Lad Prabhakar
2021-12-24 14:57 ` Lad Prabhakar
0 siblings, 2 replies; 15+ messages in thread
From: Lad Prabhakar @ 2021-12-24 14:57 UTC (permalink / raw)
To: linux-gpio; +Cc: Rob Herring, linux-kernel, Prabhakar, Lad Prabhakar
Hi All,
This patch series aims to drop using platform_get_resource() for IRQ types
in preparation for removal of static setup of IRQ resource from DT core
code.
Dropping usage of platform_get_resource() was agreed based on
the discussion [0].
[0] https://patchwork.kernel.org/project/linux-renesas-soc/
patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/
Cheers,
Prabhakar
Lad Prabhakar (2):
pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
pinctrl: at91-pio4: Use platform_get_irq_optional() to get the
interrupt
drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
drivers/pinctrl/samsung/pinctrl-samsung.c | 9 +++++----
2 files changed, 11 insertions(+), 12 deletions(-)
--
2.17.1
^ permalink raw reply [flat|nested] 15+ messages in thread
* [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
2021-12-24 14:57 [PATCH 0/2] pinctrl: Use platform_get_irq*() variants to fetch IRQ's Lad Prabhakar
@ 2021-12-24 14:57 ` Lad Prabhakar
2021-12-24 14:57 ` Lad Prabhakar
1 sibling, 0 replies; 15+ messages in thread
From: Lad Prabhakar @ 2021-12-24 14:57 UTC (permalink / raw)
To: linux-gpio, Tomasz Figa, Krzysztof Kozlowski, Sylwester Nawrocki,
Linus Walleij
Cc: Rob Herring, linux-kernel, Prabhakar, Lad Prabhakar,
linux-arm-kernel, linux-samsung-soc
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
allocation of IRQ resources in DT core code, this causes an issue
when using hierarchical interrupt domains using "interrupts" property
in the node as this bypasses the hierarchical setup and messes up the
irq chaining.
In preparation for removal of static setup of IRQ resource from DT core
code use platform_get_irq_optional().
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
drivers/pinctrl/samsung/pinctrl-samsung.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/pinctrl/samsung/pinctrl-samsung.c b/drivers/pinctrl/samsung/pinctrl-samsung.c
index 8941f658e7f1..0f6e9305fec5 100644
--- a/drivers/pinctrl/samsung/pinctrl-samsung.c
+++ b/drivers/pinctrl/samsung/pinctrl-samsung.c
@@ -1095,7 +1095,6 @@ static int samsung_pinctrl_probe(struct platform_device *pdev)
struct samsung_pinctrl_drv_data *drvdata;
const struct samsung_pin_ctrl *ctrl;
struct device *dev = &pdev->dev;
- struct resource *res;
int ret;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
@@ -1109,9 +1108,11 @@ static int samsung_pinctrl_probe(struct platform_device *pdev)
}
drvdata->dev = dev;
- res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
- if (res)
- drvdata->irq = res->start;
+ ret = platform_get_irq_optional(pdev, 0);
+ if (ret < 0 && ret != -ENXIO)
+ return ret;
+ if (ret > 0)
+ drvdata->irq = ret;
if (ctrl->retention_data) {
drvdata->retention_ctrl = ctrl->retention_data->init(drvdata,
--
2.17.1
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-24 14:57 ` Lad Prabhakar
0 siblings, 0 replies; 15+ messages in thread
From: Lad Prabhakar @ 2021-12-24 14:57 UTC (permalink / raw)
To: linux-gpio, Tomasz Figa, Krzysztof Kozlowski, Sylwester Nawrocki,
Linus Walleij
Cc: Rob Herring, linux-kernel, Prabhakar, Lad Prabhakar,
linux-arm-kernel, linux-samsung-soc
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
allocation of IRQ resources in DT core code, this causes an issue
when using hierarchical interrupt domains using "interrupts" property
in the node as this bypasses the hierarchical setup and messes up the
irq chaining.
In preparation for removal of static setup of IRQ resource from DT core
code use platform_get_irq_optional().
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
drivers/pinctrl/samsung/pinctrl-samsung.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/pinctrl/samsung/pinctrl-samsung.c b/drivers/pinctrl/samsung/pinctrl-samsung.c
index 8941f658e7f1..0f6e9305fec5 100644
--- a/drivers/pinctrl/samsung/pinctrl-samsung.c
+++ b/drivers/pinctrl/samsung/pinctrl-samsung.c
@@ -1095,7 +1095,6 @@ static int samsung_pinctrl_probe(struct platform_device *pdev)
struct samsung_pinctrl_drv_data *drvdata;
const struct samsung_pin_ctrl *ctrl;
struct device *dev = &pdev->dev;
- struct resource *res;
int ret;
drvdata = devm_kzalloc(dev, sizeof(*drvdata), GFP_KERNEL);
@@ -1109,9 +1108,11 @@ static int samsung_pinctrl_probe(struct platform_device *pdev)
}
drvdata->dev = dev;
- res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
- if (res)
- drvdata->irq = res->start;
+ ret = platform_get_irq_optional(pdev, 0);
+ if (ret < 0 && ret != -ENXIO)
+ return ret;
+ if (ret > 0)
+ drvdata->irq = ret;
if (ctrl->retention_data) {
drvdata->retention_ctrl = ctrl->retention_data->init(drvdata,
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
2021-12-24 14:57 [PATCH 0/2] pinctrl: Use platform_get_irq*() variants to fetch IRQ's Lad Prabhakar
@ 2021-12-24 14:57 ` Lad Prabhakar
2021-12-24 14:57 ` Lad Prabhakar
1 sibling, 0 replies; 15+ messages in thread
From: Lad Prabhakar @ 2021-12-24 14:57 UTC (permalink / raw)
To: linux-gpio, Ludovic Desroches, Linus Walleij, Nicolas Ferre,
Alexandre Belloni
Cc: Rob Herring, linux-kernel, Prabhakar, Lad Prabhakar, linux-arm-kernel
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
allocation of IRQ resources in DT core code, this causes an issue
when using hierarchical interrupt domains using "interrupts" property
in the node as this bypasses the hierarchical setup and messes up the
irq chaining.
In preparation for removal of static setup of IRQ resource from DT core
code use platform_get_irq_optional().
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
1 file changed, 6 insertions(+), 8 deletions(-)
diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
index fafd1f55cba7..ebfb106be97d 100644
--- a/drivers/pinctrl/pinctrl-at91-pio4.c
+++ b/drivers/pinctrl/pinctrl-at91-pio4.c
@@ -1045,7 +1045,6 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
const char **group_names;
const struct of_device_id *match;
int i, ret;
- struct resource *res;
struct atmel_pioctrl *atmel_pioctrl;
const struct atmel_pioctrl_data *atmel_pioctrl_data;
@@ -1164,16 +1163,15 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
/* There is one controller but each bank has its own irq line. */
for (i = 0; i < atmel_pioctrl->nbanks; i++) {
- res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
- if (!res) {
+ ret = platform_get_irq_optional(pdev, i);
+ if (ret < 0) {
dev_err(dev, "missing irq resource for group %c\n",
'A' + i);
- return -EINVAL;
+ return ret;
}
- atmel_pioctrl->irqs[i] = res->start;
- irq_set_chained_handler_and_data(res->start,
- atmel_gpio_irq_handler, atmel_pioctrl);
- dev_dbg(dev, "bank %i: irq=%pr\n", i, res);
+ atmel_pioctrl->irqs[i] = ret;
+ irq_set_chained_handler_and_data(ret, atmel_gpio_irq_handler, atmel_pioctrl);
+ dev_dbg(dev, "bank %i: irq=%d\n", i, ret);
}
atmel_pioctrl->irq_domain = irq_domain_add_linear(dev->of_node,
--
2.17.1
^ permalink raw reply related [flat|nested] 15+ messages in thread
* [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-24 14:57 ` Lad Prabhakar
0 siblings, 0 replies; 15+ messages in thread
From: Lad Prabhakar @ 2021-12-24 14:57 UTC (permalink / raw)
To: linux-gpio, Ludovic Desroches, Linus Walleij, Nicolas Ferre,
Alexandre Belloni
Cc: Rob Herring, linux-kernel, Prabhakar, Lad Prabhakar, linux-arm-kernel
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
allocation of IRQ resources in DT core code, this causes an issue
when using hierarchical interrupt domains using "interrupts" property
in the node as this bypasses the hierarchical setup and messes up the
irq chaining.
In preparation for removal of static setup of IRQ resource from DT core
code use platform_get_irq_optional().
Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
---
drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
1 file changed, 6 insertions(+), 8 deletions(-)
diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
index fafd1f55cba7..ebfb106be97d 100644
--- a/drivers/pinctrl/pinctrl-at91-pio4.c
+++ b/drivers/pinctrl/pinctrl-at91-pio4.c
@@ -1045,7 +1045,6 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
const char **group_names;
const struct of_device_id *match;
int i, ret;
- struct resource *res;
struct atmel_pioctrl *atmel_pioctrl;
const struct atmel_pioctrl_data *atmel_pioctrl_data;
@@ -1164,16 +1163,15 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
/* There is one controller but each bank has its own irq line. */
for (i = 0; i < atmel_pioctrl->nbanks; i++) {
- res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
- if (!res) {
+ ret = platform_get_irq_optional(pdev, i);
+ if (ret < 0) {
dev_err(dev, "missing irq resource for group %c\n",
'A' + i);
- return -EINVAL;
+ return ret;
}
- atmel_pioctrl->irqs[i] = res->start;
- irq_set_chained_handler_and_data(res->start,
- atmel_gpio_irq_handler, atmel_pioctrl);
- dev_dbg(dev, "bank %i: irq=%pr\n", i, res);
+ atmel_pioctrl->irqs[i] = ret;
+ irq_set_chained_handler_and_data(ret, atmel_gpio_irq_handler, atmel_pioctrl);
+ dev_dbg(dev, "bank %i: irq=%d\n", i, ret);
}
atmel_pioctrl->irq_domain = irq_domain_add_linear(dev->of_node,
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply related [flat|nested] 15+ messages in thread
* Re: [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
2021-12-24 14:57 ` Lad Prabhakar
@ 2021-12-24 15:03 ` Alexandre Belloni
-1 siblings, 0 replies; 15+ messages in thread
From: Alexandre Belloni @ 2021-12-24 15:03 UTC (permalink / raw)
To: Lad Prabhakar
Cc: linux-gpio, Ludovic Desroches, Linus Walleij, Nicolas Ferre,
Rob Herring, linux-kernel, Prabhakar, linux-arm-kernel
On 24/12/2021 14:57:48+0000, Lad Prabhakar wrote:
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
>
> Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> ---
> drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
> 1 file changed, 6 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
> index fafd1f55cba7..ebfb106be97d 100644
> --- a/drivers/pinctrl/pinctrl-at91-pio4.c
> +++ b/drivers/pinctrl/pinctrl-at91-pio4.c
> @@ -1045,7 +1045,6 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
> const char **group_names;
> const struct of_device_id *match;
> int i, ret;
> - struct resource *res;
> struct atmel_pioctrl *atmel_pioctrl;
> const struct atmel_pioctrl_data *atmel_pioctrl_data;
>
> @@ -1164,16 +1163,15 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
>
> /* There is one controller but each bank has its own irq line. */
> for (i = 0; i < atmel_pioctrl->nbanks; i++) {
> - res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
> - if (!res) {
> + ret = platform_get_irq_optional(pdev, i);
I don't think the irq should be optional here.
> + if (ret < 0) {
> dev_err(dev, "missing irq resource for group %c\n",
> 'A' + i);
> - return -EINVAL;
> + return ret;
> }
> - atmel_pioctrl->irqs[i] = res->start;
> - irq_set_chained_handler_and_data(res->start,
> - atmel_gpio_irq_handler, atmel_pioctrl);
> - dev_dbg(dev, "bank %i: irq=%pr\n", i, res);
> + atmel_pioctrl->irqs[i] = ret;
> + irq_set_chained_handler_and_data(ret, atmel_gpio_irq_handler, atmel_pioctrl);
> + dev_dbg(dev, "bank %i: irq=%d\n", i, ret);
> }
>
> atmel_pioctrl->irq_domain = irq_domain_add_linear(dev->of_node,
> --
> 2.17.1
>
--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-24 15:03 ` Alexandre Belloni
0 siblings, 0 replies; 15+ messages in thread
From: Alexandre Belloni @ 2021-12-24 15:03 UTC (permalink / raw)
To: Lad Prabhakar
Cc: Linus Walleij, linux-kernel, linux-gpio, Ludovic Desroches,
Rob Herring, Prabhakar, linux-arm-kernel
On 24/12/2021 14:57:48+0000, Lad Prabhakar wrote:
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
>
> Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> ---
> drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
> 1 file changed, 6 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
> index fafd1f55cba7..ebfb106be97d 100644
> --- a/drivers/pinctrl/pinctrl-at91-pio4.c
> +++ b/drivers/pinctrl/pinctrl-at91-pio4.c
> @@ -1045,7 +1045,6 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
> const char **group_names;
> const struct of_device_id *match;
> int i, ret;
> - struct resource *res;
> struct atmel_pioctrl *atmel_pioctrl;
> const struct atmel_pioctrl_data *atmel_pioctrl_data;
>
> @@ -1164,16 +1163,15 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
>
> /* There is one controller but each bank has its own irq line. */
> for (i = 0; i < atmel_pioctrl->nbanks; i++) {
> - res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
> - if (!res) {
> + ret = platform_get_irq_optional(pdev, i);
I don't think the irq should be optional here.
> + if (ret < 0) {
> dev_err(dev, "missing irq resource for group %c\n",
> 'A' + i);
> - return -EINVAL;
> + return ret;
> }
> - atmel_pioctrl->irqs[i] = res->start;
> - irq_set_chained_handler_and_data(res->start,
> - atmel_gpio_irq_handler, atmel_pioctrl);
> - dev_dbg(dev, "bank %i: irq=%pr\n", i, res);
> + atmel_pioctrl->irqs[i] = ret;
> + irq_set_chained_handler_and_data(ret, atmel_gpio_irq_handler, atmel_pioctrl);
> + dev_dbg(dev, "bank %i: irq=%d\n", i, ret);
> }
>
> atmel_pioctrl->irq_domain = irq_domain_add_linear(dev->of_node,
> --
> 2.17.1
>
--
Alexandre Belloni, co-owner and COO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
2021-12-24 15:03 ` Alexandre Belloni
@ 2021-12-24 15:09 ` Lad, Prabhakar
-1 siblings, 0 replies; 15+ messages in thread
From: Lad, Prabhakar @ 2021-12-24 15:09 UTC (permalink / raw)
To: Alexandre Belloni
Cc: Lad Prabhakar, open list:GPIO SUBSYSTEM, Ludovic Desroches,
Linus Walleij, Nicolas Ferre, Rob Herring, LKML, LAK
Hi Alexandre,
Thank you for the review.
On Fri, Dec 24, 2021 at 3:03 PM Alexandre Belloni
<alexandre.belloni@bootlin.com> wrote:
>
> On 24/12/2021 14:57:48+0000, Lad Prabhakar wrote:
> > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> > allocation of IRQ resources in DT core code, this causes an issue
> > when using hierarchical interrupt domains using "interrupts" property
> > in the node as this bypasses the hierarchical setup and messes up the
> > irq chaining.
> >
> > In preparation for removal of static setup of IRQ resource from DT core
> > code use platform_get_irq_optional().
> >
> > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> > ---
> > drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
> > 1 file changed, 6 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
> > index fafd1f55cba7..ebfb106be97d 100644
> > --- a/drivers/pinctrl/pinctrl-at91-pio4.c
> > +++ b/drivers/pinctrl/pinctrl-at91-pio4.c
> > @@ -1045,7 +1045,6 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
> > const char **group_names;
> > const struct of_device_id *match;
> > int i, ret;
> > - struct resource *res;
> > struct atmel_pioctrl *atmel_pioctrl;
> > const struct atmel_pioctrl_data *atmel_pioctrl_data;
> >
> > @@ -1164,16 +1163,15 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
> >
> > /* There is one controller but each bank has its own irq line. */
> > for (i = 0; i < atmel_pioctrl->nbanks; i++) {
> > - res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
> > - if (!res) {
> > + ret = platform_get_irq_optional(pdev, i);
>
> I don't think the irq should be optional here.
>
The only difference between platform_get_irq() and
platform_get_irq_optional() is that platform_get_irq() prints an error
message in case of error. Since we are already printing an sane error
message "missing irq resource for group %c\n" in the driver I've
chosen platform_get_irq_optional() instead platform_get_irq().
Let me know if you want me to drop platform_get_irq_optional() and use
platform_get_irq() instead, in that case I'll also drop the error
message "missing irq resource for group %c\n" from the driver.
Cheers,
Prabhakar
> > + if (ret < 0) {
> > dev_err(dev, "missing irq resource for group %c\n",
> > 'A' + i);
> > - return -EINVAL;
> > + return ret;
> > }
> > - atmel_pioctrl->irqs[i] = res->start;
> > - irq_set_chained_handler_and_data(res->start,
> > - atmel_gpio_irq_handler, atmel_pioctrl);
> > - dev_dbg(dev, "bank %i: irq=%pr\n", i, res);
> > + atmel_pioctrl->irqs[i] = ret;
> > + irq_set_chained_handler_and_data(ret, atmel_gpio_irq_handler, atmel_pioctrl);
> > + dev_dbg(dev, "bank %i: irq=%d\n", i, ret);
> > }
> >
> > atmel_pioctrl->irq_domain = irq_domain_add_linear(dev->of_node,
> > --
> > 2.17.1
> >
>
> --
> Alexandre Belloni, co-owner and COO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-24 15:09 ` Lad, Prabhakar
0 siblings, 0 replies; 15+ messages in thread
From: Lad, Prabhakar @ 2021-12-24 15:09 UTC (permalink / raw)
To: Alexandre Belloni
Cc: LKML, Linus Walleij, Lad Prabhakar, open list:GPIO SUBSYSTEM,
Ludovic Desroches, Rob Herring, LAK
Hi Alexandre,
Thank you for the review.
On Fri, Dec 24, 2021 at 3:03 PM Alexandre Belloni
<alexandre.belloni@bootlin.com> wrote:
>
> On 24/12/2021 14:57:48+0000, Lad Prabhakar wrote:
> > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> > allocation of IRQ resources in DT core code, this causes an issue
> > when using hierarchical interrupt domains using "interrupts" property
> > in the node as this bypasses the hierarchical setup and messes up the
> > irq chaining.
> >
> > In preparation for removal of static setup of IRQ resource from DT core
> > code use platform_get_irq_optional().
> >
> > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
> > ---
> > drivers/pinctrl/pinctrl-at91-pio4.c | 14 ++++++--------
> > 1 file changed, 6 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/pinctrl/pinctrl-at91-pio4.c b/drivers/pinctrl/pinctrl-at91-pio4.c
> > index fafd1f55cba7..ebfb106be97d 100644
> > --- a/drivers/pinctrl/pinctrl-at91-pio4.c
> > +++ b/drivers/pinctrl/pinctrl-at91-pio4.c
> > @@ -1045,7 +1045,6 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
> > const char **group_names;
> > const struct of_device_id *match;
> > int i, ret;
> > - struct resource *res;
> > struct atmel_pioctrl *atmel_pioctrl;
> > const struct atmel_pioctrl_data *atmel_pioctrl_data;
> >
> > @@ -1164,16 +1163,15 @@ static int atmel_pinctrl_probe(struct platform_device *pdev)
> >
> > /* There is one controller but each bank has its own irq line. */
> > for (i = 0; i < atmel_pioctrl->nbanks; i++) {
> > - res = platform_get_resource(pdev, IORESOURCE_IRQ, i);
> > - if (!res) {
> > + ret = platform_get_irq_optional(pdev, i);
>
> I don't think the irq should be optional here.
>
The only difference between platform_get_irq() and
platform_get_irq_optional() is that platform_get_irq() prints an error
message in case of error. Since we are already printing an sane error
message "missing irq resource for group %c\n" in the driver I've
chosen platform_get_irq_optional() instead platform_get_irq().
Let me know if you want me to drop platform_get_irq_optional() and use
platform_get_irq() instead, in that case I'll also drop the error
message "missing irq resource for group %c\n" from the driver.
Cheers,
Prabhakar
> > + if (ret < 0) {
> > dev_err(dev, "missing irq resource for group %c\n",
> > 'A' + i);
> > - return -EINVAL;
> > + return ret;
> > }
> > - atmel_pioctrl->irqs[i] = res->start;
> > - irq_set_chained_handler_and_data(res->start,
> > - atmel_gpio_irq_handler, atmel_pioctrl);
> > - dev_dbg(dev, "bank %i: irq=%pr\n", i, res);
> > + atmel_pioctrl->irqs[i] = ret;
> > + irq_set_chained_handler_and_data(ret, atmel_gpio_irq_handler, atmel_pioctrl);
> > + dev_dbg(dev, "bank %i: irq=%d\n", i, ret);
> > }
> >
> > atmel_pioctrl->irq_domain = irq_domain_add_linear(dev->of_node,
> > --
> > 2.17.1
> >
>
> --
> Alexandre Belloni, co-owner and COO, Bootlin
> Embedded Linux and Kernel engineering
> https://bootlin.com
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: (subset) [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
2021-12-24 14:57 ` Lad Prabhakar
@ 2021-12-25 10:18 ` Krzysztof Kozlowski
-1 siblings, 0 replies; 15+ messages in thread
From: Krzysztof Kozlowski @ 2021-12-25 10:18 UTC (permalink / raw)
To: Linus Walleij, Sylwester Nawrocki, Tomasz Figa, linux-gpio,
Lad Prabhakar
Cc: Krzysztof Kozlowski, Rob Herring, linux-kernel, Prabhakar,
linux-samsung-soc, linux-arm-kernel
On Fri, 24 Dec 2021 14:57:47 +0000, Lad Prabhakar wrote:
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
>
> [...]
Applied, thanks!
[1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
commit: a382d568f144b9e533ad210117c6c50d8dbdcaf1
Best regards,
--
Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: (subset) [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-25 10:18 ` Krzysztof Kozlowski
0 siblings, 0 replies; 15+ messages in thread
From: Krzysztof Kozlowski @ 2021-12-25 10:18 UTC (permalink / raw)
To: Linus Walleij, Sylwester Nawrocki, Tomasz Figa, linux-gpio,
Lad Prabhakar
Cc: Krzysztof Kozlowski, Rob Herring, linux-kernel, Prabhakar,
linux-samsung-soc, linux-arm-kernel
On Fri, 24 Dec 2021 14:57:47 +0000, Lad Prabhakar wrote:
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
>
> [...]
Applied, thanks!
[1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
commit: a382d568f144b9e533ad210117c6c50d8dbdcaf1
Best regards,
--
Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
2021-12-24 14:57 ` Lad Prabhakar
@ 2021-12-25 15:34 ` Andy Shevchenko
-1 siblings, 0 replies; 15+ messages in thread
From: Andy Shevchenko @ 2021-12-25 15:34 UTC (permalink / raw)
To: Lad Prabhakar
Cc: open list:GPIO SUBSYSTEM, Tomasz Figa, Krzysztof Kozlowski,
Sylwester Nawrocki, Linus Walleij, Rob Herring,
Linux Kernel Mailing List, Prabhakar, linux-arm Mailing List,
Linux Samsung SOC
On Sat, Dec 25, 2021 at 3:59 AM Lad Prabhakar
<prabhakar.mahadev-lad.rj@bp.renesas.com> wrote:
>
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
> + ret = platform_get_irq_optional(pdev, 0);
> + if (ret < 0 && ret != -ENXIO)
> + return ret;
> + if (ret > 0)
> + drvdata->irq = ret;
Yes, this is how platform_get_irq_optional() is expected to be used
right now. We are going to fix it a bit, so this code won't be
affected.
--
With Best Regards,
Andy Shevchenko
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-25 15:34 ` Andy Shevchenko
0 siblings, 0 replies; 15+ messages in thread
From: Andy Shevchenko @ 2021-12-25 15:34 UTC (permalink / raw)
To: Lad Prabhakar
Cc: open list:GPIO SUBSYSTEM, Tomasz Figa, Krzysztof Kozlowski,
Sylwester Nawrocki, Linus Walleij, Rob Herring,
Linux Kernel Mailing List, Prabhakar, linux-arm Mailing List,
Linux Samsung SOC
On Sat, Dec 25, 2021 at 3:59 AM Lad Prabhakar
<prabhakar.mahadev-lad.rj@bp.renesas.com> wrote:
>
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
> + ret = platform_get_irq_optional(pdev, 0);
> + if (ret < 0 && ret != -ENXIO)
> + return ret;
> + if (ret > 0)
> + drvdata->irq = ret;
Yes, this is how platform_get_irq_optional() is expected to be used
right now. We are going to fix it a bit, so this code won't be
affected.
--
With Best Regards,
Andy Shevchenko
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
2021-12-24 14:57 ` Lad Prabhakar
@ 2021-12-25 15:36 ` Andy Shevchenko
-1 siblings, 0 replies; 15+ messages in thread
From: Andy Shevchenko @ 2021-12-25 15:36 UTC (permalink / raw)
To: Lad Prabhakar
Cc: open list:GPIO SUBSYSTEM, Ludovic Desroches, Linus Walleij,
Nicolas Ferre, Alexandre Belloni, Rob Herring,
Linux Kernel Mailing List, Prabhakar, linux-arm Mailing List
On Sat, Dec 25, 2021 at 3:59 AM Lad Prabhakar
<prabhakar.mahadev-lad.rj@bp.renesas.com> wrote:
>
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
> + ret = platform_get_irq_optional(pdev, i);
> + if (ret < 0) {
> dev_err(dev, "missing irq resource for group %c\n",
> 'A' + i);
> - return -EINVAL;
> + return ret;
> }
This is an incorrect use of platform_get_irq_optional() (It's not
related to what Alexandre said, it's an additional comment).
NAK.
--
With Best Regards,
Andy Shevchenko
^ permalink raw reply [flat|nested] 15+ messages in thread
* Re: [PATCH 2/2] pinctrl: at91-pio4: Use platform_get_irq_optional() to get the interrupt
@ 2021-12-25 15:36 ` Andy Shevchenko
0 siblings, 0 replies; 15+ messages in thread
From: Andy Shevchenko @ 2021-12-25 15:36 UTC (permalink / raw)
To: Lad Prabhakar
Cc: Alexandre Belloni, Linus Walleij, Linux Kernel Mailing List,
open list:GPIO SUBSYSTEM, Ludovic Desroches, Rob Herring,
Prabhakar, linux-arm Mailing List
On Sat, Dec 25, 2021 at 3:59 AM Lad Prabhakar
<prabhakar.mahadev-lad.rj@bp.renesas.com> wrote:
>
> platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static
> allocation of IRQ resources in DT core code, this causes an issue
> when using hierarchical interrupt domains using "interrupts" property
> in the node as this bypasses the hierarchical setup and messes up the
> irq chaining.
>
> In preparation for removal of static setup of IRQ resource from DT core
> code use platform_get_irq_optional().
> + ret = platform_get_irq_optional(pdev, i);
> + if (ret < 0) {
> dev_err(dev, "missing irq resource for group %c\n",
> 'A' + i);
> - return -EINVAL;
> + return ret;
> }
This is an incorrect use of platform_get_irq_optional() (It's not
related to what Alexandre said, it's an additional comment).
NAK.
--
With Best Regards,
Andy Shevchenko
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 15+ messages in thread
end of thread, other threads:[~2021-12-25 15:38 UTC | newest]
Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-24 14:57 [PATCH 0/2] pinctrl: Use platform_get_irq*() variants to fetch IRQ's Lad Prabhakar
2021-12-24 14:57 ` [PATCH 1/2] pinctrl: samsung: Use platform_get_irq_optional() to get the interrupt Lad Prabhakar
2021-12-24 14:57 ` Lad Prabhakar
2021-12-25 10:18 ` (subset) " Krzysztof Kozlowski
2021-12-25 10:18 ` Krzysztof Kozlowski
2021-12-25 15:34 ` Andy Shevchenko
2021-12-25 15:34 ` Andy Shevchenko
2021-12-24 14:57 ` [PATCH 2/2] pinctrl: at91-pio4: " Lad Prabhakar
2021-12-24 14:57 ` Lad Prabhakar
2021-12-24 15:03 ` Alexandre Belloni
2021-12-24 15:03 ` Alexandre Belloni
2021-12-24 15:09 ` Lad, Prabhakar
2021-12-24 15:09 ` Lad, Prabhakar
2021-12-25 15:36 ` Andy Shevchenko
2021-12-25 15:36 ` Andy Shevchenko
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.