Re: [PATCH v3 08/11] cifs: Fix potential deadlock when updating vol in cifs_reconnect()

["Aurélien Aptel" <aaptel@suse.com>]

"Paulo Alcantara (SUSE)" <pc@cjr.nz> writes:
> We can't acquire vol_lock while refreshing the DFS cache because
> cifs_reconnect() may call dfs_cache_update_vol() while we are walking
> through the volume list.
>
> To prevent that, make vol_info refcounted, create a temp list with all
> volumes eligible for refreshing, and then use it without any locks
> held.
>
> Also turn vol_lock into a spinlock rather than a mutex.
>
> Signed-off-by: Paulo Alcantara (SUSE) <pc@cjr.nz>
> ---
>  fs/cifs/dfs_cache.c | 104 +++++++++++++++++++++++++++++---------------
>  1 file changed, 69 insertions(+), 35 deletions(-)
>
> diff --git a/fs/cifs/dfs_cache.c b/fs/cifs/dfs_cache.c
> index 1d1f7c03931b..629190926197 100644
> --- a/fs/cifs/dfs_cache.c
> +++ b/fs/cifs/dfs_cache.c
> @@ -52,12 +52,14 @@ struct vol_info {
>  	struct smb_vol smb_vol;
>  	char *mntdata;
>  	struct list_head list;
> +	struct list_head rlist;
> +	int vol_count;
>  };

If vol_count is going to be accessed/modified from multiple threads
without locks there might be races. I think you should use kref

https://www.kernel.org/doc/Documentation/kref.txt

-- 
Aurélien Aptel / SUSE Labs Samba Team
GPG: 1839 CB5F 9F5B FB9B AA97  8C99 03C8 A49B 521B D5D3
SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg, DE
GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 247165 (AG München)