Mailing List Archive

# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1299061403 0
# Node ID 666006a3a4ad3e4015fa0610aae6ebda3d271523
# Parent bde351b18fd0fce23e6d4ae24627ed15a9b7c768
x86: Disable IRQs and do WBINVD first in play_dead().

When we advertise we are dead via cpu_exit_clear(), it is no longer
safe to handle interrupts as our per-cpu vars can go away.

In future, we may want to think about NMI handling in this scenario
too.

Signed-off-by: Keir Fraser <keir@xen.org>
---


diff -r bde351b18fd0 -r 666006a3a4ad xen/arch/x86/domain.c
--- a/xen/arch/x86/domain.c Wed Mar 02 10:01:55 2011 +0000
+++ b/xen/arch/x86/domain.c Wed Mar 02 10:23:23 2011 +0000
@@ -99,19 +99,20 @@

static void play_dead(void)
{
+ local_irq_disable();
+ wbinvd();
+
/*
* NOTE: After cpu_exit_clear, per-cpu variables are no longer accessible,
* as they may be freed at any time. In this case, heap corruption or
* #PF can occur (when heap debugging is enabled). For example, even
* printk() can involve tasklet scheduling, which touches per-cpu vars.
*
- * Consider very carefully when adding code to this path. Most hypervisor
+ * Consider very carefully when adding code to *dead_idle. Most hypervisor
* subsystems are unsafe to call.
*/
cpu_exit_clear(smp_processor_id());
- mb();
- local_irq_disable();
- wbinvd();
+
(*dead_idle)();
}


_______________________________________________
Xen-changelog mailing list
Xen-changelog@lists.xensource.com
http://lists.xensource.com/xen-changelog