Commit 85bcfab4 authored by Dan Carpenter's avatar Dan Carpenter Committed by Greg Kroah-Hartman

staging: lustre: validate size in lustre_posix_acl_xattr_filter()

This is mostly to silence static checker complaints.

In theory the problem here is that if size is in the 1-3 range then we
don't allocate enough memory for the posix_acl_xattr_header struct and
it results in memory corruption.  But in reality kmalloc() return values
are aligned at sizeof(long) so corrupting this small amount of data is
not harmful.

The "size" variable should be type size_t.  The value of size is checked
in setxattr() so we know it is a number between "0 - XATTR_SIZE_MAX".
There is no need to check for negative sizes.
Signed-off-by: default avatarDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
parent 31e7213a
......@@ -74,7 +74,7 @@ typedef struct {
extern ext_acl_xattr_header *
lustre_posix_acl_xattr_2ext(posix_acl_xattr_header *header, int size);
extern int
lustre_posix_acl_xattr_filter(posix_acl_xattr_header *header, int size,
lustre_posix_acl_xattr_filter(posix_acl_xattr_header *header, size_t size,
posix_acl_xattr_header **out);
extern void
lustre_posix_acl_xattr_free(posix_acl_xattr_header *header, int size);
......
......@@ -171,17 +171,17 @@ EXPORT_SYMBOL(lustre_posix_acl_xattr_2ext);
/*
* Filter out the "nobody" entries in the posix ACL.
*/
int lustre_posix_acl_xattr_filter(posix_acl_xattr_header *header, int size,
int lustre_posix_acl_xattr_filter(posix_acl_xattr_header *header, size_t size,
posix_acl_xattr_header **out)
{
int count, i, j, rc = 0;
__u32 id;
posix_acl_xattr_header *new;
if (unlikely(size < 0))
return -EINVAL;
else if (!size)
if (!size)
return 0;
if (size < sizeof(*new))
return -EINVAL;
OBD_ALLOC(new, size);
if (unlikely(new == NULL))
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment