ocfs2/dlm: Use ast_lock to protect ast_list
authorSunil Mushran <sunil.mushran@oracle.com>
Tue, 3 Feb 2009 20:37:15 +0000 (12:37 -0800)
committerMark Fasheh <mfasheh@suse.com>
Thu, 26 Feb 2009 19:51:09 +0000 (11:51 -0800)
The code was using dlm->spinlock instead of dlm->ast_lock to protect the
ast_list. This patch fixes the issue.

Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Acked-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
fs/ocfs2/dlm/dlmunlock.c

index 86ca085ef3246b8066d7f9f976b0866e7e63e168..fcf879ed69308e9518d0504428d3afbe13ee9f8f 100644 (file)
@@ -117,11 +117,11 @@ static enum dlm_status dlmunlock_common(struct dlm_ctxt *dlm,
        else
                BUG_ON(res->owner == dlm->node_num);
 
-       spin_lock(&dlm->spinlock);
+       spin_lock(&dlm->ast_lock);
        /* We want to be sure that we're not freeing a lock
         * that still has AST's pending... */
        in_use = !list_empty(&lock->ast_list);
-       spin_unlock(&dlm->spinlock);
+       spin_unlock(&dlm->ast_lock);
        if (in_use) {
               mlog(ML_ERROR, "lockres %.*s: Someone is calling dlmunlock "
                    "while waiting for an ast!", res->lockname.len,