[PATCH 1/1] powerpc: Increase pseries_cpu_die delay

[PATCH 1/1] powerpc: Increase pseries_cpu_die delay

[Brian King]

In testing SMT disable, we have been regularly seeing the following message:

Querying DEAD? cpu %i (%i) shows %i

This indicates the current delay in pseries_cpu_die where we wait
for the specified CPU to die, is insufficient. Usually, this does
not cause a problem, but we've seen this result in BUG_ON's going
off in the timer code when we try to migrate the timers off the
dead cpu while a timer is still running. Increasing this delay,
as is done in this patch, seems to resolve this issue.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
---

 arch/powerpc/platforms/pseries/hotplug-cpu.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff -puN arch/powerpc/platforms/pseries/hotplug-cpu.c~powerpc_cpu_die_delay arch/powerpc/platforms/pseries/hotplug-cpu.c
--- linux-2.6/arch/powerpc/platforms/pseries/hotplug-cpu.c~powerpc_cpu_die_delay	2010-07-20 09:24:01.000000000 -0500
+++ linux-2.6-bjking1/arch/powerpc/platforms/pseries/hotplug-cpu.c	2010-07-20 09:24:20.000000000 -0500
@@ -193,12 +193,12 @@ static void pseries_cpu_die(unsigned int
 
 	if (get_preferred_offline_state(cpu) == CPU_STATE_INACTIVE) {
 		cpu_status = 1;
-		for (tries = 0; tries < 1000; tries++) {
+		for (tries = 0; tries < 5000; tries++) {
 			if (get_cpu_current_state(cpu) == CPU_STATE_INACTIVE) {
 				cpu_status = 0;
 				break;
 			}
-			cpu_relax();
+			msleep(1);
 		}
 	} else if (get_preferred_offline_state(cpu) == CPU_STATE_OFFLINE) {
 
_