aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorThomas Gleixner <tglx@linutronix.de>2007-03-23 17:14:37 -0700
committerLinus Torvalds <torvalds@woody.linux-foundation.org>2007-03-23 20:21:45 -0700
commitf33bc55c472295966e520c9347822fdd8b1082cd (patch)
tree9532d70dbbead840164fbfd4ad429791476d9f21
parent6ea65ff79ceb36a1195761be857da1fdf9878450 (diff)
[PATCH] x86_64: avoid sending LOCAL_TIMER_VECTOR IPI to itself
Ray Lee reported, that on an UP kernel with "noapic" command line option set, the box locks hard during boot. Adding some debug printks revealed, that the last action on the box before stalling was "Send IPI" - a debug printk which was put into smp_send_timer_broadcast_ipi(). It seems that send_IPI_mask(mask, LOCAL_TIMER_VECTOR) fails when "noapic" is set on the command line on an UP kernel. Aside of that it does not make much sense to trigger an interrupt instead of calling the function directly on the CPU which gets the PIT/HPET interrupt in case of broadcasting. Reported-by: Ray Lee <ray-lk@madrabbit.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: Ray Lee <ray-lk@madrabbit.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-rw-r--r--arch/x86_64/kernel/apic.c8
1 files changed, 8 insertions, 0 deletions
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 46acf4f2f1e..bd3e45d47c3 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -934,9 +934,17 @@ EXPORT_SYMBOL(switch_APIC_timer_to_ipi);
void smp_send_timer_broadcast_ipi(void)
{
+ int cpu = smp_processor_id();
cpumask_t mask;
cpus_and(mask, cpu_online_map, timer_interrupt_broadcast_ipi_mask);
+
+ if (cpu_isset(cpu, mask)) {
+ cpu_clear(cpu, mask);
+ add_pda(apic_timer_irqs, 1);
+ smp_local_timer_interrupt();
+ }
+
if (!cpus_empty(mask)) {
send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
}