[PATCH] pinctrl: improve warning messages

[PATCH] pinctrl: improve warning messages

[Linus Walleij]

Print out the affected group name on activation of pin mux
settings, and warn if you cannot free a pin that should have
been part of a certain setting.

Cc: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
 drivers/pinctrl/pinmux.c | 35 +++++++++++++++++++++++++++++------
 1 file changed, 29 insertions(+), 6 deletions(-)

diff --git a/drivers/pinctrl/pinmux.c b/drivers/pinctrl/pinmux.c
index 5f51588..008f965 100644
--- a/drivers/pinctrl/pinmux.c
+++ b/drivers/pinctrl/pinmux.c
@@ -400,10 +400,14 @@ int pinmux_enable_setting(struct pinctrl_setting const *setting)
 	ret = pctlops->get_group_pins(pctldev, setting->data.mux.group,
 				      &pins, &num_pins);
 	if (ret) {
+		const char *gname;
+
 		/* errors only affect debug data, so just warn */
+		gname = pctlops->get_group_name(pctldev,
+						setting->data.mux.group);
 		dev_warn(pctldev->dev,
-			 "could not get pins for group selector %d\n",
-			 setting->data.mux.group);
+			 "could not get pins for group %s\n",
+			 gname);
 		num_pins = 0;
 	}
 
@@ -411,9 +415,14 @@ int pinmux_enable_setting(struct pinctrl_setting const *setting)
 	for (i = 0; i < num_pins; i++) {
 		ret = pin_request(pctldev, pins[i], setting->dev_name, NULL);
 		if (ret) {
+			const char *gname;
+
+			gname = pctlops->get_group_name(pctldev,
+						setting->data.mux.group);
 			dev_err(pctldev->dev,
-				"could not request pin %d on device %s\n",
-				pins[i], pinctrl_dev_get_name(pctldev));
+				"could not request pin %d from group %s on "
+				"device %s\n",
+				pins[i], gname, pinctrl_dev_get_name(pctldev));
 			goto err_pin_request;
 		}
 	}
@@ -466,10 +475,14 @@ void pinmux_disable_setting(struct pinctrl_setting const *setting)
 	ret = pctlops->get_group_pins(pctldev, setting->data.mux.group,
 				      &pins, &num_pins);
 	if (ret) {
+		const char *gname;
+
 		/* errors only affect debug data, so just warn */
+		gname = pctlops->get_group_name(pctldev,
+						setting->data.mux.group);
 		dev_warn(pctldev->dev,
-			 "could not get pins for group selector %d\n",
-			 setting->data.mux.group);
+			 "could not get pins for group %s\n",
+			 gname);
 		num_pins = 0;
 	}
 
@@ -486,6 +499,16 @@ void pinmux_disable_setting(struct pinctrl_setting const *setting)
 			desc->mux_setting = NULL;
 			/* And release the pin */
 			pin_free(pctldev, pins[i], NULL);
+		} else {
+			const char *gname;
+
+			gname = pctlops->get_group_name(pctldev,
+						setting->data.mux.group);
+			dev_warn(pctldev->dev,
+				 "not freeing pin %d as part of deactivating "
+				 "group %s - it is already used for some other "
+				 "setting",
+				 pins[i], gname);
 		}
 	}
 
-- 
1.8.1.4

Re: [PATCH] pinctrl: improve warning messages

[Stephen Warren]

On 08/14/2013 10:25 AM, Linus Walleij wrote:
> Print out the affected group name on activation of pin mux
> settings, and warn if you cannot free a pin that should have
> been part of a certain setting.

> diff --git a/drivers/pinctrl/pinmux.c b/drivers/pinctrl/pinmux.c

> @@ -411,9 +415,14 @@ int pinmux_enable_setting(struct pinctrl_setting const *setting)
>  	for (i = 0; i < num_pins; i++) {
>  		ret = pin_request(pctldev, pins[i], setting->dev_name, NULL);
>  		if (ret) {
> +			const char *gname;
> +
> +			gname = pctlops->get_group_name(pctldev,
> +						setting->data.mux.group);
>  			dev_err(pctldev->dev,
> -				"could not request pin %d on device %s\n",
> -				pins[i], pinctrl_dev_get_name(pctldev));
> +				"could not request pin %d from group %s on "
> +				"device %s\n",
> +				pins[i], gname, pinctrl_dev_get_name(pctldev));

I think it'd be useful to print the pin name rather than pin number here.

> @@ -486,6 +499,16 @@ void pinmux_disable_setting(struct pinctrl_setting const *setting)
>  			desc->mux_setting = NULL;
>  			/* And release the pin */
>  			pin_free(pctldev, pins[i], NULL);
> +		} else {
> +			const char *gname;
> +
> +			gname = pctlops->get_group_name(pctldev,
> +						setting->data.mux.group);
> +			dev_warn(pctldev->dev,
> +				 "not freeing pin %d as part of deactivating "
> +				 "group %s - it is already used for some other "
> +				 "setting",
> +				 pins[i], gname);
>  		}
>  	}

I think that will only happen if one of the pinmux_enabling_setting
prints already happened. Is it worth adding this one?

Re: [PATCH] pinctrl: improve warning messages

[Linus Walleij]

On Wed, Aug 14, 2013 at 6:34 PM, Stephen Warren <swarren@wwwdotorg.org> wrote:
> On 08/14/2013 10:25 AM, Linus Walleij wrote:

>>                       dev_err(pctldev->dev,
>> -                             "could not request pin %d on device %s\n",
>> -                             pins[i], pinctrl_dev_get_name(pctldev));
>> +                             "could not request pin %d from group %s on "
>> +                             "device %s\n",
>> +                             pins[i], gname, pinctrl_dev_get_name(pctldev));
>
> I think it'd be useful to print the pin name rather than pin number here.

Fixed it.

>> @@ -486,6 +499,16 @@ void pinmux_disable_setting(struct pinctrl_setting const *setting)
>>                       desc->mux_setting = NULL;
>>                       /* And release the pin */
>>                       pin_free(pctldev, pins[i], NULL);
>> +             } else {
>> +                     const char *gname;
>> +
>> +                     gname = pctlops->get_group_name(pctldev,
>> +                                             setting->data.mux.group);
>> +                     dev_warn(pctldev->dev,
>> +                              "not freeing pin %d as part of deactivating "
>> +                              "group %s - it is already used for some other "
>> +                              "setting",
>> +                              pins[i], gname);
>>               }
>>       }
>
> I think that will only happen if one of the pinmux_enabling_setting
> prints already happened. Is it worth adding this one?

It happens at two very distinct places in the run path, disabling a setting
may appear at a totally different place in the dmesg so I think so.

It's just below the just added code that will avoid free:ing a pin if some
other setting is muxing it, so it's this print I was talking about
with Sonic.

Yours,
Linus Walleij