scl-gpio = <>;
sda-gpio = <>;

Are not enough for some SoCs to have a working recovery.
Some need:

scl-gpio = <>;
sda-gpio = <>;
pinctrl-names = "default", "recovery";
pinctrl-0 = <&i2c_pins_hw>;
pinctrl-1 = <&i2c_pins_gpio>;

The driver was not filling rinfo->pinctrl with the device node
pinctrl data which is needed by generic recovery code.

Signed-off-by: Yann Sionneau <ysionneau@kalray.eu>
---
V3 -> V4:
* Replace `else if` by simply `if`.

V2 -> V3:
* put back 'if (!rinfo->pinctrl)' test since devm_pinctrl_get()
can return NULL if CONFIG_PINCTRL is not set.
* print an error msg when devm_pinctrl_get() returns an error that
is not -EPROBE_DEFER.
* print a dbg msg if devm_pinctrl_get() return NULL. 

V1 -> V2:
* remove the unnecessary 'if (!rinfo->pinctrl)' test
* test if return is -EPROBE_DEFER, in that case, return it.
* Reword the commit message according to review

 drivers/i2c/busses/i2c-designware-master.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/i2c/busses/i2c-designware-master.c b/drivers/i2c/busses/i2c-designware-master.c
index b720fcc5c10a..4a9fd49c283d 100644
--- a/drivers/i2c/busses/i2c-designware-master.c
+++ b/drivers/i2c/busses/i2c-designware-master.c
@@ -17,6 +17,7 @@
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/module.h>
+#include <linux/pinctrl/consumer.h>
 #include <linux/pm_runtime.h>
 #include <linux/regmap.h>
 #include <linux/reset.h>
@@ -855,6 +856,17 @@ static int i2c_dw_init_recovery_info(struct dw_i2c_dev *dev)
 		return PTR_ERR(gpio);
 	rinfo->sda_gpiod = gpio;
 
+	rinfo->pinctrl = devm_pinctrl_get(dev->dev);
+	if (IS_ERR(rinfo->pinctrl)) {
+		if (PTR_ERR(rinfo->pinctrl) == -EPROBE_DEFER)
+			return PTR_ERR(rinfo->pinctrl);
+
+		rinfo->pinctrl = NULL;
+		dev_err(dev->dev, "getting pinctrl info failed: bus recovery might not work\n");
+	}
+	if (!rinfo->pinctrl)
+		dev_dbg(dev->dev, "pinctrl is disabled, bus recovery might not work\n");
+
 	rinfo->recover_bus = i2c_generic_scl_recovery;
 	rinfo->prepare_recovery = i2c_dw_prepare_recovery;
 	rinfo->unprepare_recovery = i2c_dw_unprepare_recovery;
-- 
2.17.1

On Tue, Aug 22, 2023 at 03:11:23PM +0200, Yann Sionneau wrote:
> scl-gpio = <>;
> sda-gpio = <>;
> 
> Are not enough for some SoCs to have a working recovery.
> Some need:
> 
> scl-gpio = <>;
> sda-gpio = <>;
> pinctrl-names = "default", "recovery";
> pinctrl-0 = <&i2c_pins_hw>;
> pinctrl-1 = <&i2c_pins_gpio>;
> 
> The driver was not filling rinfo->pinctrl with the device node
> pinctrl data which is needed by generic recovery code.
> 
> Signed-off-by: Yann Sionneau <ysionneau@kalray.eu>
> ---
> V3 -> V4:
> * Replace `else if` by simply `if`.

You forgot my tag. Why?
Also I think this will generate more code and more noise in debug case.
So, I admit I gave a bad suggestion in previous round.

Please, go for v3 with my tag (as v5).

-- 
With Best Regards,
Andy Shevchenko

Hi,

On 8/22/23 15:26, Andy Shevchenko wrote:
> On Tue, Aug 22, 2023 at 03:11:23PM +0200, Yann Sionneau wrote:
>> scl-gpio = <>;
>> sda-gpio = <>;
>>
>> Are not enough for some SoCs to have a working recovery.
>> Some need:
>>
>> scl-gpio = <>;
>> sda-gpio = <>;
>> pinctrl-names = "default", "recovery";
>> pinctrl-0 = <&i2c_pins_hw>;
>> pinctrl-1 = <&i2c_pins_gpio>;
>>
>> The driver was not filling rinfo->pinctrl with the device node
>> pinctrl data which is needed by generic recovery code.
>>
>> Signed-off-by: Yann Sionneau <ysionneau@kalray.eu>
>> ---
>> V3 -> V4:
>> * Replace `else if` by simply `if`.
> You forgot my tag. Why?
Woops sorry, I am not used to sending patches upstream. I didn't know I 
had to add the ack tag. Will do it!
> Also I think this will generate more code and more noise in debug case.
> So, I admit I gave a bad suggestion in previous round.
Ah yes you are right in case of dbg+CONFIG_PINCTRL not set it will print 
both messages but just one is enough.
>
> Please, go for v3 with my tag (as v5).
>
Ok, I'll also add `i2c: designware: ` to $subject as Mika Westerberg 
suggested.

Thanks,

-- 

Yann

On Tue, Aug 22, 2023 at 04:15:25PM +0200, Yann Sionneau wrote:
> On 8/22/23 15:26, Andy Shevchenko wrote:
> > On Tue, Aug 22, 2023 at 03:11:23PM +0200, Yann Sionneau wrote:

...

> > You forgot my tag. Why?
> Woops sorry, I am not used to sending patches upstream. I didn't know I had
> to add the ack tag. Will do it!

Any tags, Acked-by, Reviewed-by, ...

> > Also I think this will generate more code and more noise in debug case.
> > So, I admit I gave a bad suggestion in previous round.
> Ah yes you are right in case of dbg+CONFIG_PINCTRL not set it will print
> both messages but just one is enough.
> > 
> > Please, go for v3 with my tag (as v5).
> > 
> Ok, I'll also add `i2c: designware: ` to $subject as Mika Westerberg
> suggested.

It seems that Subject was completely missing.
In the commit message it's a first line followed by a blank line.

-- 
With Best Regards,
Andy Shevchenko

On Tue, Aug 22, 2023 at 04:26:08PM +0300, Andy Shevchenko wrote:
> On Tue, Aug 22, 2023 at 03:11:23PM +0200, Yann Sionneau wrote:
> > scl-gpio = <>;
> > sda-gpio = <>;
> > 
> > Are not enough for some SoCs to have a working recovery.
> > Some need:
> > 
> > scl-gpio = <>;
> > sda-gpio = <>;
> > pinctrl-names = "default", "recovery";
> > pinctrl-0 = <&i2c_pins_hw>;
> > pinctrl-1 = <&i2c_pins_gpio>;
> > 
> > The driver was not filling rinfo->pinctrl with the device node
> > pinctrl data which is needed by generic recovery code.
> > 
> > Signed-off-by: Yann Sionneau <ysionneau@kalray.eu>
> > ---
> > V3 -> V4:
> > * Replace `else if` by simply `if`.
> 
> You forgot my tag. Why?
> Also I think this will generate more code and more noise in debug case.
> So, I admit I gave a bad suggestion in previous round.
> 
> Please, go for v3 with my tag (as v5).

While there fix $subject to follow the convention used in the driver.