[CIFS] remove bkl usage from umount begin

The lock_kernel call moved into the fs for umount_begin
is not needed.  This adds a check to make sure we don't
call umount_begin twice on the same fs.

umount_begin for cifs is probably not needed and
may eventually be able to be removed, but in
the meantime this smaller patch is safe and
gets rid of the bkl from this path which provides
some benefit.

Acked-by: Jeff Layton <redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>

diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index b5e9f398c2e58026b99fb3c33b283edd59dd1bde..9f669f982c4d4fe74b4640b221f04d9bd183b597 100644 (file)
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -537,9 +537,14 @@ static void cifs_umount_begin(struct super_block *sb)
        if (tcon == NULL)
                return;
 
-       lock_kernel();
        read_lock(&cifs_tcp_ses_lock);
-       if (tcon->tc_count == 1)
+       if ((tcon->tc_count > 1) || (tcon->tidStatus == CifsExiting)) {
+               /* we have other mounts to same share or we have
+                  already tried to force umount this and woken up
+                  all waiting network requests, nothing to do */
+               read_unlock(&cifs_tcp_ses_lock);
+               return;
+       } else if (tcon->tc_count == 1)
                tcon->tidStatus = CifsExiting;
        read_unlock(&cifs_tcp_ses_lock);
 
@@ -554,9 +559,7 @@ static void cifs_umount_begin(struct super_block *sb)
                wake_up_all(&tcon->ses->server->response_q);
                msleep(1);
        }
-/* BB FIXME - finish add checks for tidStatus BB */
 
-       unlock_kernel();
        return;
 }