for_each_cpu() unintuitively reports CPU0 as set independent of the actual
cpumask content on UP kernels. This causes an unexpected PIT interrupt
storm on a UP kernel running in an SMP virtual machine on Hyper-V, and as
a result, the virtual machine can suffer from a strange random delay of 1~20
minutes during boot-up, and sometimes it can hang forever.
Link: https://lkml.org/lkml/2018/5/9/63
Link: https://lkml.org/lkml/2018/5/15/747
Signed-off-by: Dexuan Cui <[email protected]>
Cc: [email protected]
---
Some part of the changelog and comments are copied from
Thomas Gleixner's 115ef3b7e61a ("watchdog/hardlockup/perf: Cure UP damage") :-)
kernel/time/tick-broadcast.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index b398c2e..2fc59ad 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -612,6 +612,14 @@ static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
now = ktime_get();
/* Find all expired events */
for_each_cpu(cpu, tick_broadcast_oneshot_mask) {
+#ifndef CONFIG_SMP
+ /*
+ * Required because for_each_cpu() reports unconditionally
+ * CPU0 as set on UP kernels.
+ */
+ if (cpumask_empty(tick_broadcast_oneshot_mask))
+ break;
+#endif
td = &per_cpu(tick_cpu_device, cpu);
if (td->evtdev->next_event <= now) {
cpumask_set_cpu(cpu, tmpmask);
--
2.7.4
Commit-ID: 5596fe34495cf0f645f417eb928ef224df3e3cb4
Gitweb: https://git.kernel.org/tip/5596fe34495cf0f645f417eb928ef224df3e3cb4
Author: Dexuan Cui <[email protected]>
AuthorDate: Tue, 15 May 2018 19:52:50 +0000
Committer: Thomas Gleixner <[email protected]>
CommitDate: Tue, 15 May 2018 22:45:54 +0200
tick/broadcast: Use for_each_cpu() specially on UP kernels
for_each_cpu() unintuitively reports CPU0 as set independent of the actual
cpumask content on UP kernels. This causes an unexpected PIT interrupt
storm on a UP kernel running in an SMP virtual machine on Hyper-V, and as
a result, the virtual machine can suffer from a strange random delay of 1~20
minutes during boot-up, and sometimes it can hang forever.
Protect if by checking whether the cpumask is empty before entering the
for_each_cpu() loop.
[ tglx: Use !IS_ENABLED(CONFIG_SMP) instead of #ifdeffery ]
Signed-off-by: Dexuan Cui <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Cc: Josh Poulson <[email protected]>
Cc: "Michael Kelley (EOSG)" <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Cc: [email protected]
Cc: Rakib Mullick <[email protected]>
Cc: Jork Loeser <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: KY Srinivasan <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Alexey Dobriyan <[email protected]>
Cc: Dmitry Vyukov <[email protected]>
Link: https://lkml.kernel.org/r/KL1P15301MB000678289FE55BA365B3279ABF990@KL1P15301MB0006.APCP153.PROD.OUTLOOK.COM
Link: https://lkml.kernel.org/r/KL1P15301MB0006FA63BC22BEB64902EAA0BF930@KL1P15301MB0006.APCP153.PROD.OUTLOOK.COM
---
kernel/time/tick-broadcast.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index b398c2ea69b2..aa2094d5dd27 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -612,6 +612,14 @@ static void tick_handle_oneshot_broadcast(struct clock_event_device *dev)
now = ktime_get();
/* Find all expired events */
for_each_cpu(cpu, tick_broadcast_oneshot_mask) {
+ /*
+ * Required for !SMP because for_each_cpu() reports
+ * unconditionally CPU0 as set on UP kernels.
+ */
+ if (!IS_ENABLED(CONFIG_SMP) &&
+ cpumask_empty(tick_broadcast_oneshot_mask))
+ break;
+
td = &per_cpu(tick_cpu_device, cpu);
if (td->evtdev->next_event <= now) {
cpumask_set_cpu(cpu, tmpmask);