Fixed a QEMU hang when guest poweroff in COLO mode

When the PVM guest poweroff, the COLO thread may wait a semaphore
in colo_process_checkpoint().So, we should wake up the COLO thread
before migration shutdown.

Signed-off-by: Lei Rao <lei.rao@intel.com>
Reviewed-by: Zhang Chen <chen.zhang@intel.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>

diff --git a/include/migration/colo.h b/include/migration/colo.h
index 768e1f04c3a60fc9ee98928333be46b6a10f5baa..5fbe1a6d5d30418419b2eef8cab5134c60f11c50 100644 (file)
--- a/include/migration/colo.h
+++ b/include/migration/colo.h
@@ -37,4 +37,5 @@ COLOMode get_colo_mode(void);
 void colo_do_failover(void);
 
 void colo_checkpoint_notify(void *opaque);
+void colo_shutdown(void);
 #endif

diff --git a/migration/colo.c b/migration/colo.c
index c8fadae95672fbbe7bc30f9e969cecbdf455ad63..2a855049666118646966d7a21891f098da4dfcda 100644 (file)
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -819,6 +819,26 @@ static void colo_wait_handle_message(MigrationIncomingState *mis,
     }
 }
 
+void colo_shutdown(void)
+{
+    MigrationIncomingState *mis = NULL;
+    MigrationState *s = NULL;
+
+    switch (get_colo_mode()) {
+    case COLO_MODE_PRIMARY:
+        s = migrate_get_current();
+        qemu_event_set(&s->colo_checkpoint_event);
+        qemu_sem_post(&s->colo_exit_sem);
+        break;
+    case COLO_MODE_SECONDARY:
+        mis = migration_incoming_get_current();
+        qemu_sem_post(&mis->colo_incoming_sem);
+        break;
+    default:
+        break;
+    }
+}
+
 void *colo_process_incoming_thread(void *opaque)
 {
     MigrationIncomingState *mis = opaque;

diff --git a/migration/migration.c b/migration/migration.c
index abaf6f9e3d790fa2d0d9528ca0fcce070f0429cf..c0ab86e9a5b0b4be01c3f3579ba184ee2451751f 100644 (file)
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -225,6 +225,12 @@ void migration_cancel(const Error *error)
 
 void migration_shutdown(void)
 {
+    /*
+     * When the QEMU main thread exit, the COLO thread
+     * may wait a semaphore. So, we should wakeup the
+     * COLO thread before migration shutdown.
+     */
+    colo_shutdown();
     /*
      * Cancel the current migration - that will (eventually)
      * stop the migration using this structure