From: Mikulas Patocka <mpatocka@redhat.com>
Date: Fri, 8 May 2009 18:38:41 -0400
Subject: [md] s390: I/O stall when performing random CHPID off/on
Message-id: Pine.LNX.4.64.0905081835350.1316@hs20-bc2-1.build.redhat.com
O-Subject: Re: FW: Re: [RHEL5 U3 PATCH 1/1] s390 - I/O stall when performing CHPID off/on random times on LPAR
Bugzilla: 500729
RH-Acked-by: Alasdair G Kergon <agk@redhat.com>

I'd commit this. It is the upstream patch + flush_scheduled_work() to
avoid a crash. If it doesn't work and deadlocks again, please post the
backtrace of this deadlock.

Mikulas

diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c
index d466fd6..6cac8a2 100644
--- a/drivers/md/dm-mpath.c
+++ b/drivers/md/dm-mpath.c
@@ -858,6 +858,7 @@ static void multipath_dtr(struct dm_target *ti)
 
 	flush_workqueue(kmpath_handlerd);
 	flush_workqueue(kmultipathd);
+	flush_scheduled_work();
 	free_multipath(m);
 }
 
@@ -915,7 +916,7 @@ static int fail_path(struct pgpath *pgpath)
 	dm_path_uevent(DM_UEVENT_PATH_FAILED, m->ti,
 		      pgpath->path.dev->name, m->nr_valid_paths);
 
-	queue_work(kmultipathd, &m->trigger_event);
+	schedule_work(&m->trigger_event);
 
 out:
 	spin_unlock_irqrestore(&m->lock, flags);
@@ -957,7 +958,7 @@ static int reinstate_path(struct pgpath *pgpath)
 	dm_path_uevent(DM_UEVENT_PATH_REINSTATED, m->ti,
 		      pgpath->path.dev->name, m->nr_valid_paths);
 
-	queue_work(kmultipathd, &m->trigger_event);
+	schedule_work(&m->trigger_event);
 
 out:
 	spin_unlock_irqrestore(&m->lock, flags);
@@ -1001,7 +1002,7 @@ static void bypass_pg(struct multipath *m, struct priority_group *pg,
 
 	spin_unlock_irqrestore(&m->lock, flags);
 
-	queue_work(kmultipathd, &m->trigger_event);
+	schedule_work(&m->trigger_event);
 }
 
 /*
@@ -1031,7 +1032,7 @@ static int switch_pg_num(struct multipath *m, const char *pgstr)
 	}
 	spin_unlock_irqrestore(&m->lock, flags);
 
-	queue_work(kmultipathd, &m->trigger_event);
+	schedule_work(&m->trigger_event);
 	return 0;
 }