On 18.12.2017 12:03, Nikolay Borisov wrote:
> Currently if a mounted-btrfs instance is mounted for the 2nd time
> without first unmounting the first instance then we hit a memory leak
> in btrfs_mount_root due to the fs_info of the acquired superblock is
> different than the newly allocated fs info. Fix this by specifically
> checking if the fs_info instance of the newly acquired superblock is
> the same as ours and free it if not.
> 
> Reproducer:
> 
> mount /dev/vdc /media/scratch
> mount /dev/vdc /media/scratch2 <- memory leak hit
> 
> Signed-off-by: Nikolay Borisov <[email protected]>

This one fixes:

4da82a9e60a1 btrfs: add btrfs_mount_root() and new file_system_type

Which seems to be in for-next only.

> ---
>  fs/btrfs/super.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
> index e57eb6e70278..ea3bca85be44 100644
> --- a/fs/btrfs/super.c
> +++ b/fs/btrfs/super.c
> @@ -1612,7 +1612,11 @@ static struct dentry *btrfs_mount_root(struct 
> file_system_type *fs_type,
>               goto error_sec_opts;
>       }
>  
> -     fs_info = btrfs_sb(s);
> +     if (btrfs_sb(s) != fs_info) {
> +             free_fs_info(fs_info);
> +             fs_info = btrfs_sb(s);
> +     }
> +
>       error = setup_security_options(fs_info, s, &new_sec_opts);
>       if (error) {
>               deactivate_locked_super(s);
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to