[PATCH] mtd: orion_nand: convert printk to dev_*

[PATCH] mtd: orion_nand: convert printk to dev_*

[Andy Shevchenko]

It's better to use actual device name as a prefix in error messages.

Signed-off-by: Andy Shevchenko <andy.shevchenko@gmail.com>
---
 drivers/mtd/nand/orion_nand.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/nand/orion_nand.c b/drivers/mtd/nand/orion_nand.c
index 8fbd002..ba38853 100644
--- a/drivers/mtd/nand/orion_nand.c
+++ b/drivers/mtd/nand/orion_nand.c
@@ -87,7 +87,7 @@ static int __init orion_nand_probe(struct platform_device *pdev)
 
 	nc = kzalloc(sizeof(struct nand_chip) + sizeof(struct mtd_info), GFP_KERNEL);
 	if (!nc) {
-		printk(KERN_ERR "orion_nand: failed to allocate device structure.\n");
+		dev_err(&pdev->dev, "failed to allocate device structure.\n");
 		ret = -ENOMEM;
 		goto no_res;
 	}
@@ -101,7 +101,7 @@ static int __init orion_nand_probe(struct platform_device *pdev)
 
 	io_base = ioremap(res->start, resource_size(res));
 	if (!io_base) {
-		printk(KERN_ERR "orion_nand: ioremap failed\n");
+		dev_err(&pdev->dev, "ioremap failed\n");
 		ret = -EIO;
 		goto no_res;
 	}
@@ -110,7 +110,8 @@ static int __init orion_nand_probe(struct platform_device *pdev)
 		board = devm_kzalloc(&pdev->dev, sizeof(struct orion_nand_data),
 					GFP_KERNEL);
 		if (!board) {
-			printk(KERN_ERR "orion_nand: failed to allocate board structure.\n");
+			dev_err(&pdev->dev,
+				"failed to allocate board structure.\n");
 			ret = -ENOMEM;
 			goto no_res;
 		}
-- 
1.8.3.101.g727a46b

Re: [PATCH] mtd: orion_nand: convert printk to dev_*

[Dan Carpenter]

On Mon, Jun 10, 2013 at 12:16:54AM +0300, Andy Shevchenko wrote:
> It's better to use actual device name as a prefix in error messages.
> 

Gar...

This doesn't apply on top of Emil's patch.  Emil's patch was first
so it should go in first.  Also Emil's patch was a bugfix so it is
more important than this printk cleanup patch.

Really, the kzalloc error messages should just be deleted.  They are
cargo cult error messages as in "we need an error message because
we are copy and pasting from other code without thinking."  Kzalloc
already has far better messages built in.

Ah well...

regards,
dan carpenter

Re: [PATCH] mtd: orion_nand: convert printk to dev_*

[Andy Shevchenko]

On Mon, Jun 10, 2013 at 1:18 AM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> On Mon, Jun 10, 2013 at 12:16:54AM +0300, Andy Shevchenko wrote:
>> It's better to use actual device name as a prefix in error messages.
>>
>
> Gar...
>
> This doesn't apply on top of Emil's patch.  Emil's patch was first
> so it should go in first.  Also Emil's patch was a bugfix so it is
> more important than this printk cleanup patch.

Which tree has that patch incorporated? I'm okay to rebase it against that tree.
Current version was done against recent linux-next.

--
With Best Regards,
Andy Shevchenko

Re: [PATCH] mtd: orion_nand: convert printk to dev_*

[Emil Goode]

Hello Dan,

Thank you for pointing this out, I was going to send a second version
and fix that typo anyway so I will send a patch that applies on top
of Andy's patch and also remove that error message for kzalloc.

Best regards,

Emil

On Mon, Jun 10, 2013 at 01:18:08AM +0300, Dan Carpenter wrote:
> On Mon, Jun 10, 2013 at 12:16:54AM +0300, Andy Shevchenko wrote:
> > It's better to use actual device name as a prefix in error messages.
> > 
> 
> Gar...
> 
> This doesn't apply on top of Emil's patch.  Emil's patch was first
> so it should go in first.  Also Emil's patch was a bugfix so it is
> more important than this printk cleanup patch.
> 
> Really, the kzalloc error messages should just be deleted.  They are
> cargo cult error messages as in "we need an error message because
> we are copy and pasting from other code without thinking."  Kzalloc
> already has far better messages built in.
> 
> Ah well...
> 
> regards,
> dan carpenter

Re: [PATCH] mtd: orion_nand: convert printk to dev_*

[Emil Goode]

Hello Andy,

Thanks for the hint about converting to dev*. But I agree with Dan that error
messages for kzalloc calls are a bit pointless and after removing them all the
dev* calls are gone. So I will send a second version that fixes the typo and
removes the error messages for kzalloc. If you have any objections let me know. 

Best regards,

Emil

On Mon, Jun 10, 2013 at 01:31:27AM +0300, Andy Shevchenko wrote:
> On Mon, Jun 10, 2013 at 1:18 AM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> > On Mon, Jun 10, 2013 at 12:16:54AM +0300, Andy Shevchenko wrote:
> >> It's better to use actual device name as a prefix in error messages.
> >>
> >
> > Gar...
> >
> > This doesn't apply on top of Emil's patch.  Emil's patch was first
> > so it should go in first.  Also Emil's patch was a bugfix so it is
> > more important than this printk cleanup patch.
> 
> Which tree has that patch incorporated? I'm okay to rebase it against that tree.
> Current version was done against recent linux-next.
> 
> --
> With Best Regards,
> Andy Shevchenko