[PATCH] wireless: clean up sysfs code using %pM

[PATCH] wireless: clean up sysfs code using %pM

[Johannes Berg]

Remove converting the MAC address to a string by a direct byte
conversion and use %pM instead, since the code is now boilerplate
use a macro to define the show functions, and also use the shorter
__ATTR_RO macro to define the attributes.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
---
John, just FYI, it needs to go through net-next-2.6 since it depends on
%pM. Won't cause conflicts either :)

 net/wireless/sysfs.c |   25 ++++++++++---------------
 1 file changed, 10 insertions(+), 15 deletions(-)

--- test-tree.orig/net/wireless/sysfs.c	2008-10-28 13:45:27.000000000 +0100
+++ test-tree/net/wireless/sysfs.c	2008-10-28 13:56:50.000000000 +0100
@@ -23,25 +23,20 @@ static inline struct cfg80211_registered
 	return container_of(dev, struct cfg80211_registered_device, wiphy.dev);
 }
 
-static ssize_t _show_index(struct device *dev, struct device_attribute *attr,
-			   char *buf)
-{
-	return sprintf(buf, "%d\n", dev_to_rdev(dev)->idx);
+#define SHOW_FMT(name, fmt, member)					\
+static ssize_t name ## _show(struct device *dev,			\
+			      struct device_attribute *attr,		\
+			      char *buf)				\
+{									\
+	return sprintf(buf, fmt "\n", dev_to_rdev(dev)->member);	\
 }
 
-static ssize_t _show_permaddr(struct device *dev,
-			      struct device_attribute *attr,
-			      char *buf)
-{
-	unsigned char *addr = dev_to_rdev(dev)->wiphy.perm_addr;
-
-	return sprintf(buf, "%.2x:%.2x:%.2x:%.2x:%.2x:%.2x\n",
-		       addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]);
-}
+SHOW_FMT(index, "%d", idx);
+SHOW_FMT(macaddress, "%pM", wiphy.perm_addr);
 
 static struct device_attribute ieee80211_dev_attrs[] = {
-	__ATTR(index, S_IRUGO, _show_index, NULL),
-	__ATTR(macaddress, S_IRUGO, _show_permaddr, NULL),
+	__ATTR_RO(index),
+	__ATTR_RO(macaddress),
 	{}
 };
 

Re: [PATCH] wireless: clean up sysfs code using %pM

[David Miller]

From: Johannes Berg <johannes@sipsolutions.net>
Date: Tue, 28 Oct 2008 14:00:54 +0100

> Remove converting the MAC address to a string by a direct byte
> conversion and use %pM instead, since the code is now boilerplate
> use a macro to define the show functions, and also use the shorter
> __ATTR_RO macro to define the attributes.
> 
> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
> ---
> John, just FYI, it needs to go through net-next-2.6 since it depends on
> %pM. Won't cause conflicts either :)

I'll let John get this one.

Re: [PATCH] wireless: clean up sysfs code using %pM

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 848 bytes --]

On Tue, 2008-10-28 at 22:07 -0700, David Miller wrote:
> From: Johannes Berg <johannes@sipsolutions.net>
> Date: Tue, 28 Oct 2008 14:00:54 +0100
> 
> > Remove converting the MAC address to a string by a direct byte
> > conversion and use %pM instead, since the code is now boilerplate
> > use a macro to define the show functions, and also use the shorter
> > __ATTR_RO macro to define the attributes.
> > 
> > Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
> > ---
> > John, just FYI, it needs to go through net-next-2.6 since it depends on
> > %pM. Won't cause conflicts either :)
> 
> I'll let John get this one.

Hmm. That means it won't get into 2.6.29 because John doesn't branch
from your next tree but rather from Linus's tree, and thus we can't use
%pM yet. I can resend then, doesn't matter much.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] wireless: clean up sysfs code using %pM

[John W. Linville]

On Wed, Oct 29, 2008 at 12:54:51PM +0100, Johannes Berg wrote:
> On Tue, 2008-10-28 at 22:07 -0700, David Miller wrote:
> > From: Johannes Berg <johannes@sipsolutions.net>
> > Date: Tue, 28 Oct 2008 14:00:54 +0100
> > 
> > > Remove converting the MAC address to a string by a direct byte
> > > conversion and use %pM instead, since the code is now boilerplate
> > > use a macro to define the show functions, and also use the shorter
> > > __ATTR_RO macro to define the attributes.
> > > 
> > > Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
> > > ---
> > > John, just FYI, it needs to go through net-next-2.6 since it depends on
> > > %pM. Won't cause conflicts either :)
> > 
> > I'll let John get this one.
> 
> Hmm. That means it won't get into 2.6.29 because John doesn't branch
> from your next tree but rather from Linus's tree, and thus we can't use
> %pM yet. I can resend then, doesn't matter much.

Not quite true -- wireless-testing doesn't branch from net-next-2.6,
but wireless-next-2.6 does.  Often that is a benefit for "just
wireless" development, but sometimes not...

I can push this through wireless-next-2.6 only (i.e. not in
wireless-testing), Dave could take it directly, or it could just wait.
Any preferences?

John
-- 
John W. Linville		Linux should be at the core
linville@tuxdriver.com			of your literate lifestyle.

Re: [PATCH] wireless: clean up sysfs code using %pM

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 910 bytes --]

On Wed, 2008-10-29 at 08:52 -0400, John W. Linville wrote:

> > > I'll let John get this one.
> > 
> > Hmm. That means it won't get into 2.6.29 because John doesn't branch
> > from your next tree but rather from Linus's tree, and thus we can't use
> > %pM yet. I can resend then, doesn't matter much.
> 
> Not quite true -- wireless-testing doesn't branch from net-next-2.6,
> but wireless-next-2.6 does.  Often that is a benefit for "just
> wireless" development, but sometimes not...

Ah, ok.

> I can push this through wireless-next-2.6 only (i.e. not in
> wireless-testing), Dave could take it directly, or it could just wait.
> Any preferences?

If you'd take it to wireless-next-2.6 only then... we haven't touched
this code in a long time and probably won't again soon so it won't
matter much if it's not in wireless-testing. That way I don't have to
keep track of it.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]