[PATCH] mac80211 failure to check kmalloc return value in key_key_readcould lead to unpleasant surprises

From: Jesper Juhl
Date: Fri Oct 29 2010 - 10:28:09 EST


I noticed two small issues in mac80211/debugfs_key.c::key_key_read while
reading through the code. Patch below.

The key_key_read() function returns ssize_t and the value that's actually
returned is the return value of simple_read_from_buffer() which also
returns ssize_t, so let's hold the return value in a ssize_t local
variable rather than a int one.

Also, memory is allocated dynamically with kmalloc() which can fail, but
the return value of kmalloc() is not checked, so we may end up operating
on a null pointer further on. So check for a NULL return and bail out with
-ENOMEM in that case.

Please keep me on CC on replies since I'm not subscribed to
linux-wireless.


Signed-off-by: Jesper Juhl <jj@xxxxxxxxxxxxx>
---
debugfs_key.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/net/mac80211/debugfs_key.c b/net/mac80211/debugfs_key.c
index 4aa47d0..1243d1d 100644
--- a/net/mac80211/debugfs_key.c
+++ b/net/mac80211/debugfs_key.c
@@ -203,9 +203,13 @@ static ssize_t key_key_read(struct file *file, char __user *userbuf,
size_t count, loff_t *ppos)
{
struct ieee80211_key *key = file->private_data;
- int i, res, bufsize = 2 * key->conf.keylen + 2;
+ int i, bufsize = 2 * key->conf.keylen + 2;
char *buf = kmalloc(bufsize, GFP_KERNEL);
char *p = buf;
+ ssize_t res;
+
+ if (!buf)
+ return -ENOMEM;

for (i = 0; i < key->conf.keylen; i++)
p += scnprintf(p, bufsize + buf - p, "%02x", key->conf.key[i]);


--
Jesper Juhl <jj@xxxxxxxxxxxxx> http://www.chaosbits.net/
Plain text mails only, please http://www.expita.com/nomime.html
Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/