On Wed, Jan 28, 2015 at 06:09:52PM +0100, Christoph Hellwig wrote:
> On Mon, Jan 26, 2015 at 10:16:53PM -0800, Omar Sandoval wrote:
> > get_acl gets a reference which we must release in the error cases.
> >
> > Signed-off-by: Omar Sandoval <osandov@xxxxxxxxxxx>
>
> Looks good, but at this point goto-based unwinding might be in order.
Hi, Christoph,
There are already a couple of return paths in posix_acl_create, and
there are only these two error cases, so I think gotos might actually
make the code more confusing. In any case, here's an idea:
posix_acl: fix reference leaks in posix_acl_create
get_acl gets a reference which we must release in the error cases.
Signed-off-by: Omar Sandoval <osandov@xxxxxxxxxxx>
diff --git a/fs/posix_acl.c b/fs/posix_acl.c
index 0855f77..515d315 100644
--- a/fs/posix_acl.c
+++ b/fs/posix_acl.c
@@ -564,13 +564,11 @@ posix_acl_create(struct inode *dir, umode_t *mode,
*acl = posix_acl_clone(p, GFP_NOFS);
if (!*acl)
- return -ENOMEM;
+ goto no_mem;
ret = posix_acl_create_masq(*acl, mode);
- if (ret < 0) {
- posix_acl_release(*acl);
- return -ENOMEM;
- }
+ if (ret < 0)
+ goto no_mem_clone;
if (ret == 0) {
posix_acl_release(*acl);
@@ -591,6 +589,12 @@ no_acl:
*default_acl = NULL;
*acl = NULL;
return 0;
+
+no_mem_clone:
+ posix_acl_release(*acl);
+no_mem:
+ posix_acl_release(p);
+ return -ENOMEM;
}
EXPORT_SYMBOL_GPL(posix_acl_create);
--
Omar
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
