From mboxrd@z Thu Jan  1 00:00:00 1970
From: George Dunlap <george.dunlap@eu.citrix.com>
Subject: [PATCH 1 of 3] xen: Fix schedule's grabbing of the
	schedule lock
Date: Thu, 5 Apr 2012 12:54:52 +0100
Message-ID: <fd0d2fd2824b63d724ea.1333626892@kodo2>
References: <patchbomb.1333626891@kodo2>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
In-Reply-To: <patchbomb.1333626891@kodo2>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: xen-devel@lists.xensource.com
Cc: george.dunlap@eu.citrix.com
List-Id: xen-devel@lists.xenproject.org

Because the location of the lock can change between the time you read
it and the time you grab it, the per-cpu schedule locks need to check
after lock acquisition that the lock location hasn't changed, and
release and re-try if so.  This change was effected throughout the
source code, but one very important place was apparently missed: in
schedule() itself.

Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>

diff -r 197e09bf0322 -r fd0d2fd2824b xen/common/schedule.c
--- a/xen/common/schedule.c	Wed Apr 04 11:29:43 2012 +0100
+++ b/xen/common/schedule.c	Thu Apr 05 12:20:04 2012 +0100
@@ -1075,6 +1075,7 @@ static void schedule(void)
     bool_t                tasklet_work_scheduled = 0;
     struct schedule_data *sd;
     struct task_slice     next_slice;
+    int cpu = smp_processor_id();
 
     ASSERT(!in_atomic());
 
@@ -1099,7 +1100,7 @@ static void schedule(void)
         BUG();
     }
 
-    spin_lock_irq(sd->schedule_lock);
+    pcpu_schedule_lock_irq(cpu);
 
     stop_timer(&sd->s_timer);
     
@@ -1116,7 +1117,7 @@ static void schedule(void)
 
     if ( unlikely(prev == next) )
     {
-        spin_unlock_irq(sd->schedule_lock);
+        pcpu_schedule_unlock_irq(cpu);
         trace_continue_running(next);
         return continue_running(prev);
     }
@@ -1154,7 +1155,7 @@ static void schedule(void)
     ASSERT(!next->is_running);
     next->is_running = 1;
 
-    spin_unlock_irq(sd->schedule_lock);
+    pcpu_schedule_unlock_irq(cpu);
 
     perfc_incr(sched_ctx);