X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=fs%2Focfs2%2Fdlm%2Fdlmrecovery.c;h=1e232000f3f7e1826dfaecab947afb937c222bd6;hb=987b0145d94eecf292d8b301228356f44611ab7c;hp=805cbabac051d5007c492ac5de04f4f81c84e832;hpb=9464c7cf61b9433057924c36e6e02f303a00e768;p=linux-2.6.git

diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 805cbabac..1e232000f 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -58,7 +58,7 @@ static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node);
 static int dlm_recovery_thread(void *data);
 void dlm_complete_recovery_thread(struct dlm_ctxt *dlm);
 int dlm_launch_recovery_thread(struct dlm_ctxt *dlm);
-void dlm_kick_recovery_thread(struct dlm_ctxt *dlm);
+static void dlm_kick_recovery_thread(struct dlm_ctxt *dlm);
 static int dlm_do_recovery(struct dlm_ctxt *dlm);
 
 static int dlm_pick_recovery_master(struct dlm_ctxt *dlm);
@@ -78,9 +78,15 @@ static int dlm_send_mig_lockres_msg(struct dlm_ctxt *dlm,
 				    u8 send_to,
 				    struct dlm_lock_resource *res,
 				    int total_locks);
+static int dlm_lockres_master_requery(struct dlm_ctxt *dlm,
+				      struct dlm_lock_resource *res,
+				      u8 *real_master);
 static int dlm_process_recovery_data(struct dlm_ctxt *dlm,
 				     struct dlm_lock_resource *res,
 				     struct dlm_migratable_lockres *mres);
+static int dlm_do_master_requery(struct dlm_ctxt *dlm,
+				 struct dlm_lock_resource *res,
+				 u8 nodenum, u8 *real_master);
 static int dlm_send_finalize_reco_message(struct dlm_ctxt *dlm);
 static int dlm_send_all_done_msg(struct dlm_ctxt *dlm,
 				 u8 dead_node, u8 send_to);
@@ -159,7 +165,7 @@ void dlm_dispatch_work(void *data)
  * RECOVERY THREAD
  */
 
-void dlm_kick_recovery_thread(struct dlm_ctxt *dlm)
+static void dlm_kick_recovery_thread(struct dlm_ctxt *dlm)
 {
 	/* wake the recovery thread
 	 * this will wake the reco thread in one of three places
@@ -744,12 +750,10 @@ static void dlm_request_all_locks_worker(struct dlm_work_item *item, void *data)
 		     dlm->name, dlm->reco.dead_node, dlm->reco.new_master,
 		     dead_node, reco_master);
 		mlog(ML_ERROR, "%s: name=%.*s master=%u locks=%u/%u flags=%u "
-		     "entry[0]={c=%u:%llu,l=%u,f=%u,t=%d,ct=%d,hb=%d,n=%u}\n",
+		     "entry[0]={c=%"MLFu64",l=%u,f=%u,t=%d,ct=%d,hb=%d,n=%u}\n",
 		     dlm->name, mres->lockname_len, mres->lockname, mres->master,
 		     mres->num_locks, mres->total_locks, mres->flags,
-		     dlm_get_lock_cookie_node(mres->ml[0].cookie),
-		     dlm_get_lock_cookie_seq(mres->ml[0].cookie),
-		     mres->ml[0].list, mres->ml[0].flags,
+		     mres->ml[0].cookie, mres->ml[0].list, mres->ml[0].flags,
 		     mres->ml[0].type, mres->ml[0].convert_type,
 		     mres->ml[0].highest_blocked, mres->ml[0].node);
 		BUG();
@@ -1312,8 +1316,9 @@ leave:
 
 
 
-int dlm_lockres_master_requery(struct dlm_ctxt *dlm,
-			       struct dlm_lock_resource *res, u8 *real_master)
+static int dlm_lockres_master_requery(struct dlm_ctxt *dlm,
+				      struct dlm_lock_resource *res,
+				      u8 *real_master)
 {
 	struct dlm_node_iter iter;
 	int nodenum;
@@ -1355,10 +1360,8 @@ int dlm_lockres_master_requery(struct dlm_ctxt *dlm,
 		ret = dlm_do_master_requery(dlm, res, nodenum, real_master);
 		if (ret < 0) {
 			mlog_errno(ret);
-			if (!dlm_is_host_down(ret))
-				BUG();
-			/* host is down, so answer for that node would be
-			 * DLM_LOCK_RES_OWNER_UNKNOWN.  continue. */
+			BUG();
+			/* TODO: need to figure a way to restart this */
 		}
 		if (*real_master != DLM_LOCK_RES_OWNER_UNKNOWN) {
 			mlog(0, "lock master is %u\n", *real_master);
@@ -1369,8 +1372,9 @@ int dlm_lockres_master_requery(struct dlm_ctxt *dlm,
 }
 
 
-int dlm_do_master_requery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res,
-			  u8 nodenum, u8 *real_master)
+static int dlm_do_master_requery(struct dlm_ctxt *dlm,
+				 struct dlm_lock_resource *res,
+				 u8 nodenum, u8 *real_master)
 {
 	int ret = -EINVAL;
 	struct dlm_master_requery req;
@@ -1515,11 +1519,9 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm,
 			/* lock is always created locally first, and
 			 * destroyed locally last.  it must be on the list */
 			if (!lock) {
-				u64 c = ml->cookie;
 				mlog(ML_ERROR, "could not find local lock "
-					       "with cookie %u:%llu!\n",
-					       dlm_get_lock_cookie_node(c),
-					       dlm_get_lock_cookie_seq(c));
+					       "with cookie %"MLFu64"!\n",
+				     ml->cookie);
 				BUG();
 			}
 			BUG_ON(lock->ml.node != ml->node);
@@ -1737,13 +1739,6 @@ static void dlm_finish_local_lockres_recovery(struct dlm_ctxt *dlm,
 				} else
 					continue;
 
-				if (!list_empty(&res->recovering)) {
-					mlog(0, "%s:%.*s: lockres was "
-					     "marked RECOVERING, owner=%u\n",
-					     dlm->name, res->lockname.len,
-					     res->lockname.name, res->owner);
-					list_del_init(&res->recovering);
-				}
 				spin_lock(&res->spinlock);
 				dlm_change_lockres_owner(dlm, res, new_master);
 				res->state &= ~DLM_LOCK_RES_RECOVERING;
@@ -2263,10 +2258,7 @@ int dlm_begin_reco_handler(struct o2net_msg *msg, u32 len, void *data)
 			mlog(0, "%u not in domain/live_nodes map "
 			     "so setting it in reco map manually\n",
 			     br->dead_node);
-		/* force the recovery cleanup in __dlm_hb_node_down
-		 * both of these will be cleared in a moment */
-		set_bit(br->dead_node, dlm->domain_map);
-		set_bit(br->dead_node, dlm->live_nodes_map);
+		set_bit(br->dead_node, dlm->recovery_map);
 		__dlm_hb_node_down(dlm, br->dead_node);
 	}
 	spin_unlock(&dlm->spinlock);