2022-07-15 08:19:40

by williamsukatube

Subject: [PATCH] fuse: Fix a potential memory leak for kstrdup()

From: William Dean <[email protected]>

kfree() is missing on an error path to free the memory allocated by
kstrdup():

sb->s_subtype = kstrdup(parent_sb->s_subtype, GFP_KERNEL);

So it is better to free it via kfree(sb->s_subtype).

Fixes: 1866d779d5d2a ("fuse: Allow fuse_fill_super_common() for submounts")
Reported-by: Hacash Robot <[email protected]>
Signed-off-by: William Dean <[email protected]>
---
fs/fuse/inode.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 8c0665c5dff8..2d10afad07f8 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -1449,8 +1449,10 @@ static int fuse_fill_super_submount(struct super_block *sb,
get_fuse_inode(root)->nlookup--;
sb->s_d_op = &fuse_dentry_operations;
sb->s_root = d_make_root(root);
- if (!sb->s_root)
+ if (!sb->s_root) {
+ kfree(sb->s_subtype);
return -ENOMEM;
+ }

return 0;
}
--
2.25.1

2022-07-15 14:59:23

by Miklos Szeredi

Subject: Re: [PATCH] fuse: Fix a potential memory leak for kstrdup()

On Fri, 15 Jul 2022 at 09:54, <[email protected]> wrote:
>
> From: William Dean <[email protected]>
>
> kfree() is missing on an error path to free the memory allocated by
> kstrdup():
>
> sb->s_subtype = kstrdup(parent_sb->s_subtype, GFP_KERNEL);
>
> So it is better to free it via kfree(sb->s_subtype).

Not needed. s_subtype will be freed in __put_super().

Thanks,
Miklos