[BK PATCH] Driver Core fixes for 2.6.10-rc2

[BK PATCH] Driver Core fixes for 2.6.10-rc2

[Greg KH]

Hi,

Here are two bugfixes for the driver core / sysfs against 2.6.10-rc2.

Please pull from:
	bk://kernel.bkbits.net/gregkh/linux/driver-2.6

thanks,

greg k-h

p.s. I'll send these as patches in response to this email to lkml for
those who want to see them.

 fs/sysfs/dir.c |    3 ++-
 lib/kobject.c  |    6 ++++--
 2 files changed, 6 insertions(+), 3 deletions(-)
-----


Gerd Knorr:
  o fix kobject varargs bug

Maneesh Soni:
  o fix oops in sysfs_remove_dir()

[PATCH] Driver Core fixes for 2.6.10-rc2

[Greg KH]

ChangeSet 1.2164, 2004/11/19 08:47:13-08:00, akpm@osdl.org

[PATCH] fix kobject varargs bug

From: Gerd Knorr <kraxel@bytesex.org>

It uses the varargs list twice in a illegal way.  That doesn't harm on i386
by pure luck, but blows things up on amd64 machines.

Using var args list twice without calling va_start twice is illegal.

Signed-off-by: Gerd Knorr <kraxel@bytesex.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>


 lib/kobject.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)


diff -Nru a/lib/kobject.c b/lib/kobject.c
--- a/lib/kobject.c	2004-11-19 11:36:44 -08:00
+++ b/lib/kobject.c	2004-11-19 11:36:44 -08:00
@@ -232,11 +232,12 @@
 	va_list args;
 	char * name;
 
-	va_start(args,fmt);
 	/* 
 	 * First, try the static array 
 	 */
+	va_start(args,fmt);
 	need = vsnprintf(kobj->name,limit,fmt,args);
+	va_end(args);
 	if (need < limit) 
 		name = kobj->name;
 	else {
@@ -249,7 +250,9 @@
 			error = -ENOMEM;
 			goto Done;
 		}
+		va_start(args,fmt);
 		need = vsnprintf(name,limit,fmt,args);
+		va_end(args);
 
 		/* Still? Give up. */
 		if (need >= limit) {
@@ -266,7 +269,6 @@
 	/* Now, set the new name */
 	kobj->k_name = name;
  Done:
-	va_end(args);
 	return error;
 }
 

Re: [PATCH] Driver Core fixes for 2.6.10-rc2

[Greg KH]

ChangeSet 1.2165, 2004/11/19 08:47:38-08:00, maneesh@in.ibm.com

[PATCH] fix oops in sysfs_remove_dir()

The following patch should avoid the sysfs_remove_dir() oops you are
seeing while device removal. It anyway fixes the obvious error and is
needed. But it will not make any change to the first error you are
seeing while connecting the device.


o Following patch avoids the sysfs_remove_dir() oops when it is passed
  a kobject with NULL dentry.

Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>


 fs/sysfs/dir.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletion(-)


diff -Nru a/fs/sysfs/dir.c b/fs/sysfs/dir.c
--- a/fs/sysfs/dir.c	2004-11-19 11:36:39 -08:00
+++ b/fs/sysfs/dir.c	2004-11-19 11:36:39 -08:00
@@ -268,7 +268,7 @@
 void sysfs_remove_dir(struct kobject * kobj)
 {
 	struct dentry * dentry = dget(kobj->dentry);
-	struct sysfs_dirent * parent_sd = dentry->d_fsdata;
+	struct sysfs_dirent * parent_sd;
 	struct sysfs_dirent * sd, * tmp;
 
 	if (!dentry)
@@ -276,6 +276,7 @@
 
 	pr_debug("sysfs %s: removing dir\n",dentry->d_name.name);
 	down(&dentry->d_inode->i_sem);
+	parent_sd = dentry->d_fsdata;
 	list_for_each_entry_safe(sd, tmp, &parent_sd->s_children, s_sibling) {
 		if (!sd->s_element || !(sd->s_type & SYSFS_NOT_PINNED))
 			continue;