This patch fixes a possible null pointer dereference in
check_quota_exceeded, detected by the static checker smatch, with the
following warning:
fs/ceph/quota.c:240 check_quota_exceeded()
error: we previously assumed 'realm' could be null (see line 188)
Reported-by: Dan Carpenter <[email protected]>
Link: https://lkml.kernel.org/n/[email protected]
Fixes: b7a2921765cf ("ceph: quota: support for ceph.quota.max_files")
Signed-off-by: Luis Henriques <[email protected]>
---
fs/ceph/quota.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c
index 32d4f13784ba..03f4d24db8fe 100644
--- a/fs/ceph/quota.c
+++ b/fs/ceph/quota.c
@@ -237,7 +237,8 @@ static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op,
ceph_put_snap_realm(mdsc, realm);
realm = next;
}
- ceph_put_snap_realm(mdsc, realm);
+ if (realm)
+ ceph_put_snap_realm(mdsc, realm);
up_read(&mdsc->snap_rwsem);
return exceeded;
> On Nov 5, 2018, at 19:00, Luis Henriques <[email protected]> wrote:
>
> This patch fixes a possible null pointer dereference in
> check_quota_exceeded, detected by the static checker smatch, with the
> following warning:
>
> fs/ceph/quota.c:240 check_quota_exceeded()
> error: we previously assumed 'realm' could be null (see line 188)
>
> Reported-by: Dan Carpenter <[email protected]>
> Link: https://lkml.kernel.org/n/[email protected]
> Fixes: b7a2921765cf ("ceph: quota: support for ceph.quota.max_files")
> Signed-off-by: Luis Henriques <[email protected]>
> ---
> fs/ceph/quota.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c
> index 32d4f13784ba..03f4d24db8fe 100644
> --- a/fs/ceph/quota.c
> +++ b/fs/ceph/quota.c
> @@ -237,7 +237,8 @@ static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op,
> ceph_put_snap_realm(mdsc, realm);
> realm = next;
> }
> - ceph_put_snap_realm(mdsc, realm);
> + if (realm)
> + ceph_put_snap_realm(mdsc, realm);
> up_read(&mdsc->snap_rwsem);
>
> return exceeded;
Applied, thanks.
Yan, Zheng