[PATCH] acpi: fan.c: printk replacement

[PATCH] acpi: fan.c: printk replacement

[Sudip Mukherjee]

printk replaced with corresponding dev_err and dev_info
fixed one broken user-visible string
multiine comment edited for correct commenting style
asm/uaccess.h replaced with linux/uaccess.h

Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
---
 drivers/acpi/fan.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
index 8acf53e..7900d55 100644
--- a/drivers/acpi/fan.c
+++ b/drivers/acpi/fan.c
@@ -27,7 +27,7 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/types.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <linux/thermal.h>
 #include <linux/acpi.h>
 
@@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
 };
 
 /* --------------------------------------------------------------------------
-                                 Driver Interface
-   -------------------------------------------------------------------------- */
+ *                               Driver Interface
+ * --------------------------------------------------------------------------
+*/
 
 static int acpi_fan_add(struct acpi_device *device)
 {
@@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
 
 	result = acpi_bus_update_power(device->handle, NULL);
 	if (result) {
-		printk(KERN_ERR PREFIX "Setting initial power state\n");
+		dev_err(&device->dev, PREFIX "Setting initial power state\n");
 		goto end;
 	}
 
@@ -168,10 +169,10 @@ static int acpi_fan_add(struct acpi_device *device)
 				   &device->dev.kobj,
 				   "device");
 	if (result)
-		dev_err(&device->dev, "Failed to create sysfs link "
-			"'device'\n");
+		dev_err(&device->dev,
+			"Failed to create sysfs link 'device'\n");
 
-	printk(KERN_INFO PREFIX "%s [%s] (%s)\n",
+	dev_info(&device->dev, PREFIX "%s [%s] (%s)\n",
 	       acpi_device_name(device), acpi_device_bid(device),
 	       !device->power.state ? "on" : "off");
 
@@ -214,10 +215,9 @@ static int acpi_fan_resume(struct device *dev)
 
 	if (!dev)
 		return -EINVAL;
-
 	result = acpi_bus_update_power(to_acpi_device(dev)->handle, NULL);
 	if (result)
-		printk(KERN_ERR PREFIX "Error updating fan power state\n");
+		dev_err(dev, PREFIX "Error updating fan power state\n");
 
 	return result;
 }
-- 
1.8.1.2

Re: [PATCH] acpi: fan.c: printk replacement

[Rafael J. Wysocki]

On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> printk replaced with corresponding dev_err and dev_info
> fixed one broken user-visible string
> multiine comment edited for correct commenting style
> asm/uaccess.h replaced with linux/uaccess.h
> 
> Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> ---
>  drivers/acpi/fan.c | 18 +++++++++---------
>  1 file changed, 9 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> index 8acf53e..7900d55 100644
> --- a/drivers/acpi/fan.c
> +++ b/drivers/acpi/fan.c
> @@ -27,7 +27,7 @@
>  #include <linux/module.h>
>  #include <linux/init.h>
>  #include <linux/types.h>
> -#include <asm/uaccess.h>
> +#include <linux/uaccess.h>
>  #include <linux/thermal.h>
>  #include <linux/acpi.h>
>  
> @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
>  };
>  
>  /* --------------------------------------------------------------------------
> -                                 Driver Interface
> -   -------------------------------------------------------------------------- */
> + *                               Driver Interface
> + * --------------------------------------------------------------------------
> +*/
>  
>  static int acpi_fan_add(struct acpi_device *device)
>  {
> @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
>  
>  	result = acpi_bus_update_power(device->handle, NULL);
>  	if (result) {
> -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> +		dev_err(&device->dev, PREFIX "Setting initial power state\n");

While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
too.

Otherwise I don't see a compelling reason to apply this.

>  		goto end;
>  	}
>  
> @@ -168,10 +169,10 @@ static int acpi_fan_add(struct acpi_device *device)
>  				   &device->dev.kobj,
>  				   "device");
>  	if (result)
> -		dev_err(&device->dev, "Failed to create sysfs link "
> -			"'device'\n");
> +		dev_err(&device->dev,
> +			"Failed to create sysfs link 'device'\n");
>  
> -	printk(KERN_INFO PREFIX "%s [%s] (%s)\n",
> +	dev_info(&device->dev, PREFIX "%s [%s] (%s)\n",
>  	       acpi_device_name(device), acpi_device_bid(device),
>  	       !device->power.state ? "on" : "off");
>  
> @@ -214,10 +215,9 @@ static int acpi_fan_resume(struct device *dev)
>  
>  	if (!dev)
>  		return -EINVAL;
> -

That empty line is intentional, please don't remove it.

Moreover, in any case this change doesn't belong to your patch.

>  	result = acpi_bus_update_power(to_acpi_device(dev)->handle, NULL);
>  	if (result)
> -		printk(KERN_ERR PREFIX "Error updating fan power state\n");
> +		dev_err(dev, PREFIX "Error updating fan power state\n");
>  
>  	return result;
>  }
> 

-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.

Re: [PATCH] acpi: fan.c: printk replacement

[Sudip Mukherjee]

On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > printk replaced with corresponding dev_err and dev_info
> > fixed one broken user-visible string
> > multiine comment edited for correct commenting style
> > asm/uaccess.h replaced with linux/uaccess.h
> > 
> > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > ---
> >  drivers/acpi/fan.c | 18 +++++++++---------
> >  1 file changed, 9 insertions(+), 9 deletions(-)
> > 
> > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > index 8acf53e..7900d55 100644
> > --- a/drivers/acpi/fan.c
> > +++ b/drivers/acpi/fan.c
> > @@ -27,7 +27,7 @@
> >  #include <linux/module.h>
> >  #include <linux/init.h>
> >  #include <linux/types.h>
> > -#include <asm/uaccess.h>
> > +#include <linux/uaccess.h>
> >  #include <linux/thermal.h>
> >  #include <linux/acpi.h>
> >  
> > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> >  };
> >  
> >  /* --------------------------------------------------------------------------
> > -                                 Driver Interface
> > -   -------------------------------------------------------------------------- */
> > + *                               Driver Interface
> > + * --------------------------------------------------------------------------
> > +*/
> >  
> >  static int acpi_fan_add(struct acpi_device *device)
> >  {
> > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> >  
> >  	result = acpi_bus_update_power(device->handle, NULL);
> >  	if (result) {
> > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> 
> While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> too.
> 
> Otherwise I don't see a compelling reason to apply this.
> 

Hi,
Since in the patch I am not using any pr_* , so I am unable to understand why you are asking for a proper pr_fmt().
I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
My understanding was dev_* is more preffered than pr_*.
waiting for your suggestion on this.

thanks
sudip


> > 
> 
> -- 
> I speak only for myself.
> Rafael J. Wysocki, Intel Open Source Technology Center.

Re: [PATCH] acpi: fan.c: printk replacement

[Joe Perches]

On Tue, 2014-08-26 at 23:02 +0200, Rafael J. Wysocki wrote:
> On Tuesday, August 26, 2014 09:00:39 PM Sudip Mukherjee wrote:
> > On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> > > On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > > > printk replaced with corresponding dev_err and dev_info
> > > > fixed one broken user-visible string
> > > > multiine comment edited for correct commenting style
> > > > asm/uaccess.h replaced with linux/uaccess.h
> > > > 
> > > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > > ---
> > > >  drivers/acpi/fan.c | 18 +++++++++---------
> > > >  1 file changed, 9 insertions(+), 9 deletions(-)
> > > > 
> > > > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > > > index 8acf53e..7900d55 100644
> > > > --- a/drivers/acpi/fan.c
> > > > +++ b/drivers/acpi/fan.c
> > > > @@ -27,7 +27,7 @@
> > > >  #include <linux/module.h>
> > > >  #include <linux/init.h>
> > > >  #include <linux/types.h>
> > > > -#include <asm/uaccess.h>
> > > > +#include <linux/uaccess.h>
> > > >  #include <linux/thermal.h>
> > > >  #include <linux/acpi.h>
> > > >  
> > > > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> > > >  };
> > > >  
> > > >  /* --------------------------------------------------------------------------
> > > > -                                 Driver Interface
> > > > -   -------------------------------------------------------------------------- */
> > > > + *                               Driver Interface
> > > > + * --------------------------------------------------------------------------
> > > > +*/
> > > >  
> > > >  static int acpi_fan_add(struct acpi_device *device)
> > > >  {
> > > > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> > > >  
> > > >  	result = acpi_bus_update_power(device->handle, NULL);
> > > >  	if (result) {
> > > > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > > > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> > > 
> > > While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> > > too.
> > > 
> > > Otherwise I don't see a compelling reason to apply this.
> > > 
> > 
> > Hi,
> > Since in the patch I am not using any pr_* , so I am unable to understand why
> > you are asking for a proper pr_fmt().
> 
> Never mind, I was confused somehow, not exactly sure why.  Sorry about that.
> 
> > I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
> > My understanding was dev_* is more preffered than pr_*.
> > waiting for your suggestion on this.
> 
> Well, that really depends on the particular case.  It really is better to use
> dev_err() here, but then PREFIX with it is not really useful, so please just
> drop PREFIX from the new messages.

PREFIX is "ACPI: " so I think the idea is
to be able to grep for that.

Re: [PATCH] acpi: fan.c: printk replacement

[Rafael J. Wysocki]

On Tuesday, August 26, 2014 09:00:39 PM Sudip Mukherjee wrote:
> On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> > On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > > printk replaced with corresponding dev_err and dev_info
> > > fixed one broken user-visible string
> > > multiine comment edited for correct commenting style
> > > asm/uaccess.h replaced with linux/uaccess.h
> > > 
> > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > ---
> > >  drivers/acpi/fan.c | 18 +++++++++---------
> > >  1 file changed, 9 insertions(+), 9 deletions(-)
> > > 
> > > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > > index 8acf53e..7900d55 100644
> > > --- a/drivers/acpi/fan.c
> > > +++ b/drivers/acpi/fan.c
> > > @@ -27,7 +27,7 @@
> > >  #include <linux/module.h>
> > >  #include <linux/init.h>
> > >  #include <linux/types.h>
> > > -#include <asm/uaccess.h>
> > > +#include <linux/uaccess.h>
> > >  #include <linux/thermal.h>
> > >  #include <linux/acpi.h>
> > >  
> > > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> > >  };
> > >  
> > >  /* --------------------------------------------------------------------------
> > > -                                 Driver Interface
> > > -   -------------------------------------------------------------------------- */
> > > + *                               Driver Interface
> > > + * --------------------------------------------------------------------------
> > > +*/
> > >  
> > >  static int acpi_fan_add(struct acpi_device *device)
> > >  {
> > > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> > >  
> > >  	result = acpi_bus_update_power(device->handle, NULL);
> > >  	if (result) {
> > > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> > 
> > While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> > too.
> > 
> > Otherwise I don't see a compelling reason to apply this.
> > 
> 
> Hi,
> Since in the patch I am not using any pr_* , so I am unable to understand why
> you are asking for a proper pr_fmt().

Never mind, I was confused somehow, not exactly sure why.  Sorry about that.

> I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
> My understanding was dev_* is more preffered than pr_*.
> waiting for your suggestion on this.

Well, that really depends on the particular case.  It really is better to use
dev_err() here, but then PREFIX with it is not really useful, so please just
drop PREFIX from the new messages.

Rafael

Re: [PATCH] acpi: fan.c: printk replacement

[Rafael J. Wysocki]

On Tuesday, August 26, 2014 01:59:02 PM Joe Perches wrote:
> On Tue, 2014-08-26 at 23:02 +0200, Rafael J. Wysocki wrote:
> > On Tuesday, August 26, 2014 09:00:39 PM Sudip Mukherjee wrote:
> > > On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> > > > On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > > > > printk replaced with corresponding dev_err and dev_info
> > > > > fixed one broken user-visible string
> > > > > multiine comment edited for correct commenting style
> > > > > asm/uaccess.h replaced with linux/uaccess.h
> > > > > 
> > > > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > > > ---
> > > > >  drivers/acpi/fan.c | 18 +++++++++---------
> > > > >  1 file changed, 9 insertions(+), 9 deletions(-)
> > > > > 
> > > > > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > > > > index 8acf53e..7900d55 100644
> > > > > --- a/drivers/acpi/fan.c
> > > > > +++ b/drivers/acpi/fan.c
> > > > > @@ -27,7 +27,7 @@
> > > > >  #include <linux/module.h>
> > > > >  #include <linux/init.h>
> > > > >  #include <linux/types.h>
> > > > > -#include <asm/uaccess.h>
> > > > > +#include <linux/uaccess.h>
> > > > >  #include <linux/thermal.h>
> > > > >  #include <linux/acpi.h>
> > > > >  
> > > > > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> > > > >  };
> > > > >  
> > > > >  /* --------------------------------------------------------------------------
> > > > > -                                 Driver Interface
> > > > > -   -------------------------------------------------------------------------- */
> > > > > + *                               Driver Interface
> > > > > + * --------------------------------------------------------------------------
> > > > > +*/
> > > > >  
> > > > >  static int acpi_fan_add(struct acpi_device *device)
> > > > >  {
> > > > > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> > > > >  
> > > > >  	result = acpi_bus_update_power(device->handle, NULL);
> > > > >  	if (result) {
> > > > > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > > > > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> > > > 
> > > > While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> > > > too.
> > > > 
> > > > Otherwise I don't see a compelling reason to apply this.
> > > > 
> > > 
> > > Hi,
> > > Since in the patch I am not using any pr_* , so I am unable to understand why
> > > you are asking for a proper pr_fmt().
> > 
> > Never mind, I was confused somehow, not exactly sure why.  Sorry about that.
> > 
> > > I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
> > > My understanding was dev_* is more preffered than pr_*.
> > > waiting for your suggestion on this.
> > 
> > Well, that really depends on the particular case.  It really is better to use
> > dev_err() here, but then PREFIX with it is not really useful, so please just
> > drop PREFIX from the new messages.
> 
> PREFIX is "ACPI: " so I think the idea is
> to be able to grep for that.

I'm not sure how useful that is in this particular case.  You can grep for "power
state" istead just fine ...

Rafael

Re: [PATCH] acpi: fan.c: printk replacement

[Sudip Mukherjee]

On Tue, Aug 26, 2014 at 11:22:12PM +0200, Rafael J. Wysocki wrote:
> On Tuesday, August 26, 2014 01:59:02 PM Joe Perches wrote:
> > On Tue, 2014-08-26 at 23:02 +0200, Rafael J. Wysocki wrote:
> > > On Tuesday, August 26, 2014 09:00:39 PM Sudip Mukherjee wrote:
> > > > On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> > > > > On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > > > > > printk replaced with corresponding dev_err and dev_info
> > > > > > fixed one broken user-visible string
> > > > > > multiine comment edited for correct commenting style
> > > > > > asm/uaccess.h replaced with linux/uaccess.h
> > > > > > 
> > > > > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > > > > ---
> > > > > >  drivers/acpi/fan.c | 18 +++++++++---------
> > > > > >  1 file changed, 9 insertions(+), 9 deletions(-)
> > > > > > 
> > > > > > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > > > > > index 8acf53e..7900d55 100644
> > > > > > --- a/drivers/acpi/fan.c
> > > > > > +++ b/drivers/acpi/fan.c
> > > > > > @@ -27,7 +27,7 @@
> > > > > >  #include <linux/module.h>
> > > > > >  #include <linux/init.h>
> > > > > >  #include <linux/types.h>
> > > > > > -#include <asm/uaccess.h>
> > > > > > +#include <linux/uaccess.h>
> > > > > >  #include <linux/thermal.h>
> > > > > >  #include <linux/acpi.h>
> > > > > >  
> > > > > > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> > > > > >  };
> > > > > >  
> > > > > >  /* --------------------------------------------------------------------------
> > > > > > -                                 Driver Interface
> > > > > > -   -------------------------------------------------------------------------- */
> > > > > > + *                               Driver Interface
> > > > > > + * --------------------------------------------------------------------------
> > > > > > +*/
> > > > > >  
> > > > > >  static int acpi_fan_add(struct acpi_device *device)
> > > > > >  {
> > > > > > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> > > > > >  
> > > > > >  	result = acpi_bus_update_power(device->handle, NULL);
> > > > > >  	if (result) {
> > > > > > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > > > > > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> > > > > 
> > > > > While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> > > > > too.
> > > > > 
> > > > > Otherwise I don't see a compelling reason to apply this.
> > > > > 
> > > > 
> > > > Hi,
> > > > Since in the patch I am not using any pr_* , so I am unable to understand why
> > > > you are asking for a proper pr_fmt().
> > > 
> > > Never mind, I was confused somehow, not exactly sure why.  Sorry about that.
> > > 
> > > > I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
> > > > My understanding was dev_* is more preffered than pr_*.
> > > > waiting for your suggestion on this.
> > > 
> > > Well, that really depends on the particular case.  It really is better to use
> > > dev_err() here, but then PREFIX with it is not really useful, so please just
> > > drop PREFIX from the new messages.
> > 
> > PREFIX is "ACPI: " so I think the idea is
> > to be able to grep for that.
> 
> I'm not sure how useful that is in this particular case.  You can grep for "power
> state" istead just fine ...
> 
> Rafael
> 

Then there is one more printk which prints fan state is on or off :

>dev_info(&device->dev, PREFIX "%s [%s] (%s)\n",
>	  acpi_device_name(device), acpi_device_bid(device),
>              !device->power.state ? "on" : "off");

So if we drop the PREFIX and if some one wants to grep for this fan on/off , then how to do that ..
after removing PREFIX , in dmesg I am getting it as :
[    2.056204] fan PNP0C0B:00: Fan [FAN0] (off)
[    2.056225] fan PNP0C0B:01: Fan [FAN1] (off)
[    2.056245] fan PNP0C0B:02: Fan [FAN2] (off)
[    2.056263] fan PNP0C0B:03: Fan [FAN3] (off)
[    2.056283] fan PNP0C0B:04: Fan [FAN4] (off)


thanks
sudip

Re: [PATCH] acpi: fan.c: printk replacement

[Rafael J. Wysocki]

On Wednesday, August 27, 2014 10:27:47 AM Sudip Mukherjee wrote:
> On Tue, Aug 26, 2014 at 11:22:12PM +0200, Rafael J. Wysocki wrote:
> > On Tuesday, August 26, 2014 01:59:02 PM Joe Perches wrote:
> > > On Tue, 2014-08-26 at 23:02 +0200, Rafael J. Wysocki wrote:
> > > > On Tuesday, August 26, 2014 09:00:39 PM Sudip Mukherjee wrote:
> > > > > On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> > > > > > On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > > > > > > printk replaced with corresponding dev_err and dev_info
> > > > > > > fixed one broken user-visible string
> > > > > > > multiine comment edited for correct commenting style
> > > > > > > asm/uaccess.h replaced with linux/uaccess.h
> > > > > > > 
> > > > > > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > > > > > ---
> > > > > > >  drivers/acpi/fan.c | 18 +++++++++---------
> > > > > > >  1 file changed, 9 insertions(+), 9 deletions(-)
> > > > > > > 
> > > > > > > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > > > > > > index 8acf53e..7900d55 100644
> > > > > > > --- a/drivers/acpi/fan.c
> > > > > > > +++ b/drivers/acpi/fan.c
> > > > > > > @@ -27,7 +27,7 @@
> > > > > > >  #include <linux/module.h>
> > > > > > >  #include <linux/init.h>
> > > > > > >  #include <linux/types.h>
> > > > > > > -#include <asm/uaccess.h>
> > > > > > > +#include <linux/uaccess.h>
> > > > > > >  #include <linux/thermal.h>
> > > > > > >  #include <linux/acpi.h>
> > > > > > >  
> > > > > > > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> > > > > > >  };
> > > > > > >  
> > > > > > >  /* --------------------------------------------------------------------------
> > > > > > > -                                 Driver Interface
> > > > > > > -   -------------------------------------------------------------------------- */
> > > > > > > + *                               Driver Interface
> > > > > > > + * --------------------------------------------------------------------------
> > > > > > > +*/
> > > > > > >  
> > > > > > >  static int acpi_fan_add(struct acpi_device *device)
> > > > > > >  {
> > > > > > > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> > > > > > >  
> > > > > > >  	result = acpi_bus_update_power(device->handle, NULL);
> > > > > > >  	if (result) {
> > > > > > > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > > > > > > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> > > > > > 
> > > > > > While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> > > > > > too.
> > > > > > 
> > > > > > Otherwise I don't see a compelling reason to apply this.
> > > > > > 
> > > > > 
> > > > > Hi,
> > > > > Since in the patch I am not using any pr_* , so I am unable to understand why
> > > > > you are asking for a proper pr_fmt().
> > > > 
> > > > Never mind, I was confused somehow, not exactly sure why.  Sorry about that.
> > > > 
> > > > > I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
> > > > > My understanding was dev_* is more preffered than pr_*.
> > > > > waiting for your suggestion on this.
> > > > 
> > > > Well, that really depends on the particular case.  It really is better to use
> > > > dev_err() here, but then PREFIX with it is not really useful, so please just
> > > > drop PREFIX from the new messages.
> > > 
> > > PREFIX is "ACPI: " so I think the idea is
> > > to be able to grep for that.
> > 
> > I'm not sure how useful that is in this particular case.  You can grep for "power
> > state" istead just fine ...
> > 
> > Rafael
> > 
> 
> Then there is one more printk which prints fan state is on or off :
> 
> >dev_info(&device->dev, PREFIX "%s [%s] (%s)\n",
> >	  acpi_device_name(device), acpi_device_bid(device),
> >              !device->power.state ? "on" : "off");
> 
> So if we drop the PREFIX and if some one wants to grep for this fan on/off , then how to do that ..
> after removing PREFIX , in dmesg I am getting it as :
> [    2.056204] fan PNP0C0B:00: Fan [FAN0] (off)
> [    2.056225] fan PNP0C0B:01: Fan [FAN1] (off)
> [    2.056245] fan PNP0C0B:02: Fan [FAN2] (off)
> [    2.056263] fan PNP0C0B:03: Fan [FAN3] (off)
> [    2.056283] fan PNP0C0B:04: Fan [FAN4] (off)

$ dmesg | grep fan

?

You can simply say "ACPI fan [%s] (%s)\n", acpi_device_bid(device) " in that
dev_info too.

Rafael

Re: [PATCH] acpi: fan.c: printk replacement

[Sudip Mukherjee]

On Thu, Aug 28, 2014 at 01:26:58AM +0200, Rafael J. Wysocki wrote:
> On Wednesday, August 27, 2014 10:27:47 AM Sudip Mukherjee wrote:
> > On Tue, Aug 26, 2014 at 11:22:12PM +0200, Rafael J. Wysocki wrote:
> > > On Tuesday, August 26, 2014 01:59:02 PM Joe Perches wrote:
> > > > On Tue, 2014-08-26 at 23:02 +0200, Rafael J. Wysocki wrote:
> > > > > On Tuesday, August 26, 2014 09:00:39 PM Sudip Mukherjee wrote:
> > > > > > On Tue, Aug 26, 2014 at 12:45:20AM +0200, Rafael J. Wysocki wrote:
> > > > > > > On Friday, August 22, 2014 05:33:21 PM Sudip Mukherjee wrote:
> > > > > > > > printk replaced with corresponding dev_err and dev_info
> > > > > > > > fixed one broken user-visible string
> > > > > > > > multiine comment edited for correct commenting style
> > > > > > > > asm/uaccess.h replaced with linux/uaccess.h
> > > > > > > > 
> > > > > > > > Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
> > > > > > > > ---
> > > > > > > >  drivers/acpi/fan.c | 18 +++++++++---------
> > > > > > > >  1 file changed, 9 insertions(+), 9 deletions(-)
> > > > > > > > 
> > > > > > > > diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c
> > > > > > > > index 8acf53e..7900d55 100644
> > > > > > > > --- a/drivers/acpi/fan.c
> > > > > > > > +++ b/drivers/acpi/fan.c
> > > > > > > > @@ -27,7 +27,7 @@
> > > > > > > >  #include <linux/module.h>
> > > > > > > >  #include <linux/init.h>
> > > > > > > >  #include <linux/types.h>
> > > > > > > > -#include <asm/uaccess.h>
> > > > > > > > +#include <linux/uaccess.h>
> > > > > > > >  #include <linux/thermal.h>
> > > > > > > >  #include <linux/acpi.h>
> > > > > > > >  
> > > > > > > > @@ -127,8 +127,9 @@ static const struct thermal_cooling_device_ops fan_cooling_ops = {
> > > > > > > >  };
> > > > > > > >  
> > > > > > > >  /* --------------------------------------------------------------------------
> > > > > > > > -                                 Driver Interface
> > > > > > > > -   -------------------------------------------------------------------------- */
> > > > > > > > + *                               Driver Interface
> > > > > > > > + * --------------------------------------------------------------------------
> > > > > > > > +*/
> > > > > > > >  
> > > > > > > >  static int acpi_fan_add(struct acpi_device *device)
> > > > > > > >  {
> > > > > > > > @@ -143,7 +144,7 @@ static int acpi_fan_add(struct acpi_device *device)
> > > > > > > >  
> > > > > > > >  	result = acpi_bus_update_power(device->handle, NULL);
> > > > > > > >  	if (result) {
> > > > > > > > -		printk(KERN_ERR PREFIX "Setting initial power state\n");
> > > > > > > > +		dev_err(&device->dev, PREFIX "Setting initial power state\n");
> > > > > > > 
> > > > > > > While at it, please define a proper pr_fmt() for this file and get rid of PREFIX
> > > > > > > too.
> > > > > > > 
> > > > > > > Otherwise I don't see a compelling reason to apply this.
> > > > > > > 
> > > > > > 
> > > > > > Hi,
> > > > > > Since in the patch I am not using any pr_* , so I am unable to understand why
> > > > > > you are asking for a proper pr_fmt().
> > > > > 
> > > > > Never mind, I was confused somehow, not exactly sure why.  Sorry about that.
> > > > > 
> > > > > > I can get rid of the PREFIX . Then should I use pr_* in the patch instead of dev_* ? 
> > > > > > My understanding was dev_* is more preffered than pr_*.
> > > > > > waiting for your suggestion on this.
> > > > > 
> > > > > Well, that really depends on the particular case.  It really is better to use
> > > > > dev_err() here, but then PREFIX with it is not really useful, so please just
> > > > > drop PREFIX from the new messages.
> > > > 
> > > > PREFIX is "ACPI: " so I think the idea is
> > > > to be able to grep for that.
> > > 
> > > I'm not sure how useful that is in this particular case.  You can grep for "power
> > > state" istead just fine ...
> > > 
> > > Rafael
> > > 
> > 
> > Then there is one more printk which prints fan state is on or off :
> > 
> > >dev_info(&device->dev, PREFIX "%s [%s] (%s)\n",
> > >	  acpi_device_name(device), acpi_device_bid(device),
> > >              !device->power.state ? "on" : "off");
> > 
> > So if we drop the PREFIX and if some one wants to grep for this fan on/off , then how to do that ..
> > after removing PREFIX , in dmesg I am getting it as :
> > [    2.056204] fan PNP0C0B:00: Fan [FAN0] (off)
> > [    2.056225] fan PNP0C0B:01: Fan [FAN1] (off)
> > [    2.056245] fan PNP0C0B:02: Fan [FAN2] (off)
> > [    2.056263] fan PNP0C0B:03: Fan [FAN3] (off)
> > [    2.056283] fan PNP0C0B:04: Fan [FAN4] (off)
> 
> $ dmesg | grep fan
> 
> ?
> 
> You can simply say "ACPI fan [%s] (%s)\n", acpi_device_bid(device) " in that
> dev_info too.
> 
> Rafael
> 
I was under the impression that grep means using git grep and searching in the kernel code.
Anyways, I have sent a modified patch , can you please review it ..

thanks
sudip