diff mbox series

[v2,1/2] gpiolib: of: Prepare of_gpiochip_add() / of_gpiochip_remove() for fwnode

Message ID 20221108133853.61884-1-andriy.shevchenko@linux.intel.com
State Accepted
Commit 8afe82550240640617abfb3d6ba2c7579261e7fa
Headers show
Series [v2,1/2] gpiolib: of: Prepare of_gpiochip_add() / of_gpiochip_remove() for fwnode | expand

Commit Message

Andy Shevchenko Nov. 8, 2022, 1:38 p.m. UTC
GPIO library is getting rid of of_node, fwnode should be utilized instead.
Prepare of_gpiochip_add() / of_gpiochip_remove() for fwnode.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
---
v2: added tag (Dmitry)
 drivers/gpio/gpiolib-of.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)


base-commit: 80280df758c1498485988b30cf6887fde7796056

Comments

Andy Shevchenko Nov. 10, 2022, 1:53 p.m. UTC | #1
On Thu, Nov 10, 2022 at 02:22:40PM +0100, Thierry Reding wrote:
> On Tue, Nov 08, 2022 at 03:38:52PM +0200, Andy Shevchenko wrote:

...

> > +	np = to_of_node(chip->fwnode);
> 
> This breaks a number of GPIO controllers on Tegra where chip->fwnode
> ends up never getting set. I also see this break drivers like the MFD-
> based gpio-max77620, so I don't think this is anything specific to the
> Tegra drivers.
> 
> Looking at how fwnode handling works, it seems like we're checking the
> wrong value here, since chip->fwnode is only for explicit overrides of
> the fwnode value.
> 
> The below patch fixes the regression for me:

Thank you! Can you submit it as a formal fix? (Also see below)
Of if Bart prefers I can respin fixed verison. Bart?

...

> -       np = to_of_node(chip->fwnode);
> +       np = to_of_node(chip->gpiodev->dev.fwnode);

dev_fwnode(&chip->gpiodev->dev)

...


Your report makes me wonder if I can Cc you the patch that changes that logic,
so you can help with a testing on OF platforms.
Marek Szyprowski Nov. 10, 2022, 3:07 p.m. UTC | #2
Hi Andy,

On 10.11.2022 14:53, Andy Shevchenko wrote:
> On Thu, Nov 10, 2022 at 02:22:40PM +0100, Thierry Reding wrote:
>> On Tue, Nov 08, 2022 at 03:38:52PM +0200, Andy Shevchenko wrote:
> ...
>>> +	np = to_of_node(chip->fwnode);
>> This breaks a number of GPIO controllers on Tegra where chip->fwnode
>> ends up never getting set. I also see this break drivers like the MFD-
>> based gpio-max77620, so I don't think this is anything specific to the
>> Tegra drivers.
>>
>> Looking at how fwnode handling works, it seems like we're checking the
>> wrong value here, since chip->fwnode is only for explicit overrides of
>> the fwnode value.
>>
>> The below patch fixes the regression for me:
> Thank you! Can you submit it as a formal fix? (Also see below)
> Of if Bart prefers I can respin fixed verison. Bart?
>
> ...
>> -       np = to_of_node(chip->fwnode);
>> +       np = to_of_node(chip->gpiodev->dev.fwnode);
> dev_fwnode(&chip->gpiodev->dev)
>
> ...
>
>
> Your report makes me wonder if I can Cc you the patch that changes that logic,
> so you can help with a testing on OF platforms.

I've also found this issue with today's linux-next and bisected to this 
patch. I confirm that the above change fixes the boot issue on Raspberry 
Pi 4B and Odroid-M1 boards. Feel free to add:

Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>


Best regards
Bartosz Golaszewski Nov. 10, 2022, 3:29 p.m. UTC | #3
On Thu, Nov 10, 2022 at 2:53 PM Andy Shevchenko
<andriy.shevchenko@linux.intel.com> wrote:
>
> On Thu, Nov 10, 2022 at 02:22:40PM +0100, Thierry Reding wrote:
> > On Tue, Nov 08, 2022 at 03:38:52PM +0200, Andy Shevchenko wrote:
>
> ...
>
> > > +   np = to_of_node(chip->fwnode);
> >
> > This breaks a number of GPIO controllers on Tegra where chip->fwnode
> > ends up never getting set. I also see this break drivers like the MFD-
> > based gpio-max77620, so I don't think this is anything specific to the
> > Tegra drivers.
> >
> > Looking at how fwnode handling works, it seems like we're checking the
> > wrong value here, since chip->fwnode is only for explicit overrides of
> > the fwnode value.
> >
> > The below patch fixes the regression for me:
>
> Thank you! Can you submit it as a formal fix? (Also see below)
> Of if Bart prefers I can respin fixed verison. Bart?
>

Let's have a fix on top of your changes. Thierry: can you send the
patch to the list?

Bart
diff mbox series

Patch

diff --git a/drivers/gpio/gpiolib-of.c b/drivers/gpio/gpiolib-of.c
index be9c34cca322..000020eb78d8 100644
--- a/drivers/gpio/gpiolib-of.c
+++ b/drivers/gpio/gpiolib-of.c
@@ -1104,9 +1104,11 @@  static int of_gpiochip_add_pin_range(struct gpio_chip *chip) { return 0; }
 
 int of_gpiochip_add(struct gpio_chip *chip)
 {
+	struct device_node *np;
 	int ret;
 
-	if (!chip->of_node)
+	np = to_of_node(chip->fwnode);
+	if (!np)
 		return 0;
 
 	if (!chip->of_xlate) {
@@ -1123,18 +1125,18 @@  int of_gpiochip_add(struct gpio_chip *chip)
 	if (ret)
 		return ret;
 
-	of_node_get(chip->of_node);
+	fwnode_handle_get(chip->fwnode);
 
 	ret = of_gpiochip_scan_gpios(chip);
 	if (ret)
-		of_node_put(chip->of_node);
+		fwnode_handle_put(chip->fwnode);
 
 	return ret;
 }
 
 void of_gpiochip_remove(struct gpio_chip *chip)
 {
-	of_node_put(chip->of_node);
+	fwnode_handle_put(chip->fwnode);
 }
 
 void of_gpio_dev_init(struct gpio_chip *gc, struct gpio_device *gdev)