If the root dentry implements d_dname instead of ignoring it in __dentry_path call d_dname and return the result. The motivating example are files under /proc/<pid>/ns/ are bind mounted into other locations. This change allow the name of the dentry that is bind mounted to show up in /proc/<pid>/mountinfo. However in general this is correct behavior any time dentry_path is called on a dentry that implements d_dname. Signed-off-by: "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx> --- fs/dcache.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/fs/dcache.c b/fs/dcache.c index c5c7847ff84b..a7a925fb3ce7 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -3162,6 +3162,8 @@ restart: done_seqretry(&rename_lock, seq); if (error) goto Elong; + if (dentry->d_op && dentry->d_op->d_dname) + retval = dentry->d_op->d_dname(dentry, buf, len); return retval; Elong: return ERR_PTR(-ENAMETOOLONG); -- 1.7.5.4 -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
