From: Glauber de Oliveira Costa <gcosta@redhat.com>
Subject: [RHEL-5 PATCH] Avoid touching the softlockup watchdog when gone for too long (#216467)
Date: Wed, 29 Nov 2006 12:31:33 -0200
Bugzilla: 216467
Message-Id: <20061129143132.GG23434@redhat.com>
Changelog: Xen: Avoid touching the watchdog when gone for too long


This patch, pulled directly from upstream avoids the softlockup watchdog
getting called when a domain is paused for a long timeframe, and then
unpaused. 

Being without a call to the softlockup watchdog in this situation is
normal, and all we have to do is call it when we're back. We can find
out that we were back by looking into the stolen variable.

It fixes #216467, but it won't work properly without the patch for
#209534, sent some time ago, but not yet included (I resent it to
dzickus today)

please ACK for inclusion

-- 
Glauber de Oliveira Costa
Red Hat Inc.
"Free as in Freedom"

# HG changeset patch
# User kfraser@localhost.localdomain
# Date 1164802619 0
# Node ID 53795f0a41b116d0fb428bd3ee4a099b42fca97f
# Parent  f32abd365ae006e4dfda7ec206a9955b3cd57da2
[LINUX] Avoid triggering the softlockup BUG when offline for too long.

After being offline for a long time, the softlockup  watchdog triggers
a BUG() on our faces. This is expected, as in fact, we spent more than
a fixed 10*HZ amount of time without touching the watchdog.

However, by inspecting the contents of stolen inside timer irq handler,
we can gain awareness of the fact, and do better than that.

Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com>

Index: latest/arch/i386/kernel/time-xen.c
===================================================================
--- latest.orig/arch/i386/kernel/time-xen.c
+++ latest/arch/i386/kernel/time-xen.c
@@ -709,6 +709,10 @@ irqreturn_t timer_interrupt(int irq, voi
 					    (cputime_t)delta_cpu);
 	}
 
+	/* Offlined for more than a few seconds? Avoid lockup warnings. */
+	if (stolen > 5*HZ)
+		touch_softlockup_watchdog();
+
 	/* Local timer processing (see update_process_times()). */
 	run_local_timers();
 	if (rcu_pending(cpu))