From: Longpeng <longpe...@huawei.com>
We find an issue when repeat reboot in guest during migration, it cause the
migration thread never be waken up again.
<main loop> |<migration_thread>
|
LOCK BQL |
... |
main_loop_should_exit |
pause_all_vcpus |
1. set all cpus ->stop=true |
and then kick |
2. return if all cpus is paused |
(by '->stopped == true'), else|
3. qemu_cond_wait [BQL UNLOCK] |
|LOCK BQL
|...
|do_vm_stop
| pause_all_vcpus
| (A)set all cpus ->stop=true
| and then kick
| (B)return if all cpus is paused
| (by '->stopped == true'), else
| (C)qemu_cond_wait [BQL UNLOCK]
4. be waken up and LOCK BQL | (D)be waken up BUT wait for BQL
5. goto 2. |
(BQL is still LOCKed) |
resume_all_vcpus |
1. set all cpus ->stop=false |
and ->stopped=false |
... |
BQL UNLOCK | (E)LOCK BQL
| (F)goto B. [but stopped is false now!]
|Finally, sleep at step 3 forever.
Note: This patch is just for discuss this issue, I'm looking forward to
your suggestions, thanks!
Cc: Peter Maydell <peter.mayd...@linaro.org>
Cc: Dr. David Alan Gilbert <dgilb...@redhat.com>
Cc: qemu-devel@nongnu.org <qemu-devel@nongnu.org>
Signed-off-by: Longpeng <longpe...@huawei.com>
---
cpus.c | 41 ++++++++++++++++++++++++++++++++++++-----
1 file changed, 36 insertions(+), 5 deletions(-)
diff --git a/cpus.c b/cpus.c
index b4f8b84..15e8b21 100644
--- a/cpus.c
+++ b/cpus.c
@@ -1857,10 +1857,30 @@ static bool all_vcpus_paused(void)
return true;
}
+static bool all_vcpus_resumed(void)
+{
+ CPUState *cpu;
+
+ CPU_FOREACH(cpu) {
+ if (cpu->stopped) {
+ return false;
+ }
+ }
+
+ return true;
+}
+
void pause_all_vcpus(void)
{
CPUState *cpu;
+ /* We need to drop the replay_lock so any vCPU threads woken up
+ * can finish their replay tasks
+ */
+retry_unlock:
+ replay_mutex_unlock();
+
+retry_pause:
qemu_clock_enable(QEMU_CLOCK_VIRTUAL, false);
CPU_FOREACH(cpu) {
if (qemu_cpu_is_self(cpu)) {
@@ -1871,13 +1891,17 @@ void pause_all_vcpus(void)
}
}
- /* We need to drop the replay_lock so any vCPU threads woken up
- * can finish their replay tasks
- */
- replay_mutex_unlock();
-
while (!all_vcpus_paused()) {
qemu_cond_wait(&qemu_pause_cond, &qemu_global_mutex);
+ /*
+ * All of the vcpus maybe resumed due to the race with other
+ * threads that doing pause && resume, and we'll stuck as a
+ * result. So we need to request again if the race occurs.
+ */
+ if (all_vcpus_resumed()) {
+ goto retry_pause;
+ }
+
CPU_FOREACH(cpu) {
qemu_cpu_kick(cpu);
}
@@ -1886,6 +1910,13 @@ void pause_all_vcpus(void)
qemu_mutex_unlock_iothread();
replay_mutex_lock();
qemu_mutex_lock_iothread();
+ /*
+ * The vcpus maybe resumed during the mutex is unlocking, we must
+ * make sure all of the vcpus are paused before return.
+ */
+ if (!all_vcpus_paused()) {
+ goto retry_unlock;
+ }
}
void cpu_resume(CPUState *cpu)
--
1.8.3.1
