cifs: ensure we revalidate the inode after readdir if cifsacl is enabled

Otherwise, "ls -l" will simply show the ownership of the files as
the default mnt_uid/gid. This may make "ls -l" performance on large
directories super-suck in some cases, but that's the cost of cifsacl.

One possibility to make it suck less would be to somehow proactively
dispatch the ACL requests asynchronously from readdir codepath, but
that's non-trivial to implement.

Reviewed-by: Shirish Pargaonkar <shirishpargaonkar@gmail.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>

diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index f9b5d3d6cf33461081f4945d61a5d604fe2ef468..96fe44b480af68c8d12762cb1b9af84b1a0b9b59 100644 (file)
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -134,6 +134,16 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb)
        if (fattr->cf_cifsattrs & ATTR_READONLY)
                fattr->cf_mode &= ~S_IWUGO;
 
+       /*
+        * We of course don't get ACL info in FIND_FIRST/NEXT results, so
+        * mark it for revalidation so that "ls -l" will look right. It might
+        * be super-slow, but if we don't do this then the ownership of files
+        * may look wrong since the inodes may not have timed out by the time
+        * "ls" does a stat() call on them.
+        */
+       if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL)
+               fattr->cf_flags |= CIFS_FATTR_NEED_REVAL;
+
        if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL &&
            fattr->cf_cifsattrs & ATTR_SYSTEM) {
                if (fattr->cf_eof == 0)  {