diff mbox series

[v2] Fixed a QEMU hang when guest poweroff in COLO mode

Message ID 1636596693-8477-1-git-send-email-lei.rao@intel.com
State New
Headers show
Series [v2] Fixed a QEMU hang when guest poweroff in COLO mode | expand

Commit Message

Lei Rao Nov. 11, 2021, 2:11 a.m. UTC
From: "Rao, Lei" <lei.rao@intel.com>

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>
---
 include/migration/colo.h |  1 +
 migration/colo.c         | 20 ++++++++++++++++++++
 migration/migration.c    |  6 ++++++
 3 files changed, 27 insertions(+)

Comments

Zhang, Chen Nov. 11, 2021, 2:25 a.m. UTC | #1
> -----Original Message-----
> From: Rao, Lei <lei.rao@intel.com>
> Sent: Thursday, November 11, 2021 10:12 AM
> To: Zhang, Chen <chen.zhang@intel.com>;
> zhang.zhanghailiang@huawei.com; quintela@redhat.com;
> dgilbert@redhat.com
> Cc: qemu-devel@nongnu.org; Rao, Lei <lei.rao@intel.com>
> Subject: [PATCH v2] Fixed a QEMU hang when guest poweroff in COLO mode
> 
> From: "Rao, Lei" <lei.rao@intel.com>
> 
> 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>


Looks good for me.

Reviewed-by: Zhang Chen <chen.zhang@intel.com>

Thanks
Chen

> ---
>  include/migration/colo.h |  1 +
>  migration/colo.c         | 20 ++++++++++++++++++++
>  migration/migration.c    |  6 ++++++
>  3 files changed, 27 insertions(+)
> 
> diff --git a/include/migration/colo.h b/include/migration/colo.h index
> 768e1f0..5fbe1a6 100644
> --- 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 2415325..0d3d98f
> 100644
> --- a/migration/colo.c
> +++ b/migration/colo.c
> @@ -820,6 +820,26 @@ 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 abaf6f9..c0ab86e 100644
> --- a/migration/migration.c
> +++ b/migration/migration.c
> @@ -226,6 +226,12 @@ 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
>       */
> --
> 1.8.3.1
Juan Quintela Nov. 16, 2021, 4:36 p.m. UTC | #2
"Rao, Lei" <lei.rao@intel.com> wrote:
> From: "Rao, Lei" <lei.rao@intel.com>
>
> 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: Juan Quintela <quintela@redhat.com>
diff mbox series

Patch

diff --git a/include/migration/colo.h b/include/migration/colo.h
index 768e1f0..5fbe1a6 100644
--- 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 2415325..0d3d98f 100644
--- a/migration/colo.c
+++ b/migration/colo.c
@@ -820,6 +820,26 @@  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 abaf6f9..c0ab86e 100644
--- a/migration/migration.c
+++ b/migration/migration.c
@@ -226,6 +226,12 @@  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
      */