* [patch] cxgb4: memory corruption in debugfs
@ 2015-08-18 9:31 Dan Carpenter
2015-08-18 10:28 ` Tetsuo Handa
2015-08-19 2:07 ` David Miller
0 siblings, 2 replies; 4+ messages in thread
From: Dan Carpenter @ 2015-08-18 9:31 UTC (permalink / raw)
To: Hariprasad S; +Cc: netdev, kernel-janitors
You can't use kstrtoul() with an int or it causes memory corruption.
Also j should be unsigned or we have underflow bugs.
I considered changing "j" to unsigned long but everything fits in a u32.
Fixes: 8e3d04fd7d70 ('cxgb4: Add MPS tracing support')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
index 1732e29..0a87a32 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
@@ -1289,13 +1289,14 @@ static unsigned int xdigit2int(unsigned char c)
static ssize_t mps_trc_write(struct file *file, const char __user *buf,
size_t count, loff_t *pos)
{
- int i, j, enable, ret;
+ int i, enable, ret;
u32 *data, *mask;
struct trace_params tp;
const struct inode *ino;
unsigned int trcidx;
char *s, *p, *word, *end;
struct adapter *adap;
+ u32 j;
ino = file_inode(file);
trcidx = (uintptr_t)ino->i_private & 3;
@@ -1340,7 +1341,7 @@ static ssize_t mps_trc_write(struct file *file, const char __user *buf,
if (!strncmp(word, "qid=", 4)) {
end = (char *)word + 4;
- ret = kstrtoul(end, 10, (unsigned long *)&j);
+ ret = kstrtouint(end, 10, &j);
if (ret)
goto out;
if (!adap->trace_rss) {
@@ -1369,7 +1370,7 @@ static ssize_t mps_trc_write(struct file *file, const char __user *buf,
}
if (!strncmp(word, "snaplen=", 8)) {
end = (char *)word + 8;
- ret = kstrtoul(end, 10, (unsigned long *)&j);
+ ret = kstrtouint(end, 10, &j);
if (ret || j > 9600) {
inval: count = -EINVAL;
goto out;
@@ -1379,7 +1380,7 @@ inval: count = -EINVAL;
}
if (!strncmp(word, "minlen=", 7)) {
end = (char *)word + 7;
- ret = kstrtoul(end, 10, (unsigned long *)&j);
+ ret = kstrtouint(end, 10, &j);
if (ret || j > TFMINPKTSIZE_M)
goto inval;
tp.min_len = j;
@@ -1453,7 +1454,7 @@ inval: count = -EINVAL;
}
if (*word == '@') {
end = (char *)word + 1;
- ret = kstrtoul(end, 10, (unsigned long *)&j);
+ ret = kstrtouint(end, 10, &j);
if (*end && *end != '\n')
goto inval;
if (j & 7) /* doesn't start at multiple of 8 */
^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [patch] cxgb4: memory corruption in debugfs
2015-08-18 9:31 [patch] cxgb4: memory corruption in debugfs Dan Carpenter
@ 2015-08-18 10:28 ` Tetsuo Handa
2015-08-18 10:38 ` Dan Carpenter
2015-08-19 2:07 ` David Miller
1 sibling, 1 reply; 4+ messages in thread
From: Tetsuo Handa @ 2015-08-18 10:28 UTC (permalink / raw)
To: dan.carpenter, hariprasad; +Cc: netdev, kernel-janitors
Dan Carpenter wrote:
> You can't use kstrtoul() with an int or it causes memory corruption.
> Also j should be unsigned or we have underflow bugs.
>
> I considered changing "j" to unsigned long but everything fits in a u32.
Excuse me, but kstrtouint()'s last argument is not "u32 *" but "unsigned int *".
Aren't there architectures where sizeof(unsigned int) > sizeof(u32) ?
>
> Fixes: 8e3d04fd7d70 ('cxgb4: Add MPS tracing support')
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
>
> diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
> index 1732e29..0a87a32 100644
> --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
> +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c
> @@ -1289,13 +1289,14 @@ static unsigned int xdigit2int(unsigned char c)
> static ssize_t mps_trc_write(struct file *file, const char __user *buf,
> size_t count, loff_t *pos)
> {
> - int i, j, enable, ret;
> + int i, enable, ret;
> u32 *data, *mask;
> struct trace_params tp;
> const struct inode *ino;
> unsigned int trcidx;
> char *s, *p, *word, *end;
> struct adapter *adap;
> + u32 j;
>
> ino = file_inode(file);
> trcidx = (uintptr_t)ino->i_private & 3;
> @@ -1340,7 +1341,7 @@ static ssize_t mps_trc_write(struct file *file, const char __user *buf,
>
> if (!strncmp(word, "qid=", 4)) {
> end = (char *)word + 4;
> - ret = kstrtoul(end, 10, (unsigned long *)&j);
> + ret = kstrtouint(end, 10, &j);
> if (ret)
> goto out;
> if (!adap->trace_rss) {
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [patch] cxgb4: memory corruption in debugfs
2015-08-18 10:28 ` Tetsuo Handa
@ 2015-08-18 10:38 ` Dan Carpenter
0 siblings, 0 replies; 4+ messages in thread
From: Dan Carpenter @ 2015-08-18 10:38 UTC (permalink / raw)
To: Tetsuo Handa; +Cc: hariprasad, netdev, kernel-janitors
On Tue, Aug 18, 2015 at 07:28:53PM +0900, Tetsuo Handa wrote:
> Dan Carpenter wrote:
> > You can't use kstrtoul() with an int or it causes memory corruption.
> > Also j should be unsigned or we have underflow bugs.
> >
> > I considered changing "j" to unsigned long but everything fits in a u32.
>
> Excuse me, but kstrtouint()'s last argument is not "u32 *" but "unsigned int *".
> Aren't there architectures where sizeof(unsigned int) > sizeof(u32) ?
No.
regards,
dan carpenter
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [patch] cxgb4: memory corruption in debugfs
2015-08-18 9:31 [patch] cxgb4: memory corruption in debugfs Dan Carpenter
2015-08-18 10:28 ` Tetsuo Handa
@ 2015-08-19 2:07 ` David Miller
1 sibling, 0 replies; 4+ messages in thread
From: David Miller @ 2015-08-19 2:07 UTC (permalink / raw)
To: dan.carpenter; +Cc: hariprasad, netdev, kernel-janitors
From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Tue, 18 Aug 2015 12:31:44 +0300
> You can't use kstrtoul() with an int or it causes memory corruption.
> Also j should be unsigned or we have underflow bugs.
>
> I considered changing "j" to unsigned long but everything fits in a u32.
>
> Fixes: 8e3d04fd7d70 ('cxgb4: Add MPS tracing support')
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Applied, thanks Dan.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2015-08-19 2:07 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-08-18 9:31 [patch] cxgb4: memory corruption in debugfs Dan Carpenter
2015-08-18 10:28 ` Tetsuo Handa
2015-08-18 10:38 ` Dan Carpenter
2015-08-19 2:07 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).