Skip to contentSkip to conversation

UBI: Fix potential NULL-pointer dereference

PATCH In review Alexander Stein 3 participants 4 messages Mar 28, 2014 – Mar 31, 2014

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 UBI: Fix potential NULL-pointer dereference

It's no use to debug print a string before checking on NULL.

---
 drivers/mtd/ubi/kapi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/mtd/ubi/kapi.c +2 -2
index 3aac1ac..e40864f 100644
@@ -242,11 +242,11 @@ struct ubi_volume_desc *ubi_open_volume_nm(int ubi_num, const char *name,
 	struct ubi_device *ubi;
 	struct ubi_volume_desc *ret;
 
-	dbg_gen("open device %d, volume %s, mode %d", ubi_num, name, mode);
-
 	if (!name)
 		return ERR_PTR(-EINVAL);
 
+	dbg_gen("open device %d, volume %s, mode %d", ubi_num, name, mode);
+
 	len = strnlen(name, UBI_VOL_NAME_MAX + 1);
 	if (len > UBI_VOL_NAME_MAX)
 		return ERR_PTR(-EINVAL);
Signature · 1 line
1.9.1
#1 lore raw
Quoted 3 lines · Alexander Stein

It's no use to debug print a string before checking on NULL.

If the string is NULL, that could be very useful to know when debugging and the print function should handle that gracefully, i.e. by not printing anything for %s.

BR // Mats

#2 lore raw
Quoted 8 lines · Mats Kärrman
Quoted 3 lines · Alexander Stein

It's no use to debug print a string before checking on NULL.

If the string is NULL, that could be very useful to know when debugging and the print function should handle that gracefully, i.e. by not printing anything for %s.

%s will just result in (null), I do not see any issue with this.

Signature · 2 lines
Best Regards,
Artem Bityutskiy
#3 lore raw

On Monday 31 March 2014, 12:46:40 wrote Artem Bityutskiy:

Quoted 11 lines · Mats Kärrman
Quoted 8 lines · Alexander Stein
Quoted 3 lines

It's no use to debug print a string before checking on NULL.

If the string is NULL, that could be very useful to know when debugging and the print function should handle that gracefully, i.e. by not printing anything for %s.

%s will just result in (null), I do not see any issue with this.

Oh, it does? I was not aware of that (kernel-specific) behavior. Sorry for ne noise.

Best regards,
Alexander