2024-02-23 09:37:51

by Su Hui

Subject: [PATCH next] orangefs: re-fix sb refcount leak when allocate sb info failed.

This problem was previously fixed by commit ac2c63757f4f ("orangefs: Fix sb
refcount leak when allocate sb info failed.").
Add a judgement to fix NULL deference problem and also avoid refcount
leak problem.

Fixes: 9bf93dcfc453 ("Julia Lawall reported this null pointer dereference, this should fix it.")
Signed-off-by: Su Hui <[email protected]>
---
ps: Thanks for Dan's suggestion[1].
[1]: https://lore.kernel.org/all/[email protected]/
fs/orangefs/super.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/orangefs/super.c b/fs/orangefs/super.c
index 34849b4a3243..e189e1aae00b 100644
--- a/fs/orangefs/super.c
+++ b/fs/orangefs/super.c
@@ -527,7 +527,7 @@ struct dentry *orangefs_mount(struct file_system_type *fst,
sb->s_fs_info = kzalloc(sizeof(struct orangefs_sb_info_s), GFP_KERNEL);
if (!ORANGEFS_SB(sb)) {
d = ERR_PTR(-ENOMEM);
- goto free_op;
+ goto free_sb_and_op;
}

ret = orangefs_fill_sb(sb,
@@ -581,7 +581,8 @@ struct dentry *orangefs_mount(struct file_system_type *fst,

free_sb_and_op:
/* Will call orangefs_kill_sb with sb not in list. */
- ORANGEFS_SB(sb)->no_list = 1;
+ if (ORANGEFS_SB(sb))
+ ORANGEFS_SB(sb)->no_list = 1;
/* ORANGEFS_VFS_OP_FS_UMOUNT is done by orangefs_kill_sb. */
deactivate_locked_super(sb);
free_op:
--
2.30.2

2024-02-23 10:14:23

by Dan Carpenter

Subject: Re: [PATCH next] orangefs: re-fix sb refcount leak when allocate sb info failed.

On Fri, Feb 23, 2024 at 05:36:41PM +0800, Su Hui wrote:
> This problem was previously fixed by commit ac2c63757f4f ("orangefs: Fix sb
> refcount leak when allocate sb info failed.").
> Add a judgement to fix NULL deference problem and also avoid refcount
> leak problem.
>
> Fixes: 9bf93dcfc453 ("Julia Lawall reported this null pointer dereference, this should fix it.")
> Signed-off-by: Su Hui <[email protected]>
> ---
> ps: Thanks for Dan's suggestion[1].
> [1]: https://lore.kernel.org/all/[email protected]/

Thanks!

Reviewed-by: Dan Carpenter <[email protected]>

regards,
dan carpenter