qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v3] balloon: Fix failure of updating guest memor


From: Michael S. Tsirkin
Subject: Re: [Qemu-devel] [PATCH v3] balloon: Fix failure of updating guest memory status
Date: Tue, 16 Aug 2016 02:08:28 +0300

On Fri, Aug 05, 2016 at 01:36:32PM +0200, Ladi Prosek wrote:
> The stats_vq_elem field keeps track of the state of the balloon stats
> virtqueue protocol but it wasn't preserved across migrations, resulting
> in losing guest memory status updates on the receiving VM.
> 
> This commit adds a new VM state change handler which resets stats_vq_elem
> to NULL when the VM is stopped, eliminating the need for the field to be
> migrated. When the VM starts running again, the field is reinitialized by
> re-popping the element from the virtqueue.
> 
> Signed-off-by: Ladi Prosek <address@hidden>
> Suggested-by: Michael S. Tsirkin <address@hidden>

I think that for the 2.7 machine type, we should
also migrate the guest-stats-polling-interval property.

Otherwise the value that was set on source
is not forwarded to the destination
properly.

And at that point, one wanders whether we should
check the interval is set before receive calling
stats.


> ---
>  hw/virtio/virtio-balloon.c         | 23 +++++++++++++++++++++++
>  include/hw/virtio/virtio-balloon.h |  1 +
>  2 files changed, 24 insertions(+)
> 
> diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
> index 5af429a..65457e9 100644
> --- a/hw/virtio/virtio-balloon.c
> +++ b/hw/virtio/virtio-balloon.c
> @@ -423,6 +423,26 @@ static int virtio_balloon_load_device(VirtIODevice 
> *vdev, QEMUFile *f,
>      return 0;
>  }
>  
> +static void virtio_balloon_vmstate_cb(void *opaque, int running,
> +                                      RunState state)
> +{
> +    VirtIOBalloon *s = opaque;
> +
> +    if (!running) {
> +        /* put the stats element back if the VM is not running */
> +        if (s->stats_vq_elem != NULL) {
> +            virtqueue_discard(s->svq, s->stats_vq_elem, s->stats_vq_offset);
> +            g_free(s->stats_vq_elem);
> +            s->stats_vq_elem = NULL;
> +        }
> +
> +    } else {
> +        /* poll stats queue for the element we may have discarded
> +         * when the VM was stopped */
> +        virtio_balloon_receive_stats(VIRTIO_DEVICE(s), s->svq);
> +    }
> +}
> +
>  static void virtio_balloon_device_realize(DeviceState *dev, Error **errp)
>  {
>      VirtIODevice *vdev = VIRTIO_DEVICE(dev);
> @@ -446,6 +466,8 @@ static void virtio_balloon_device_realize(DeviceState 
> *dev, Error **errp)
>      s->svq = virtio_add_queue(vdev, 128, virtio_balloon_receive_stats);
>  
>      reset_stats(s);
> +
> +    s->change = qemu_add_vm_change_state_handler(virtio_balloon_vmstate_cb, 
> s);
>  }
>  
>  static void virtio_balloon_device_unrealize(DeviceState *dev, Error **errp)
> @@ -453,6 +475,7 @@ static void virtio_balloon_device_unrealize(DeviceState 
> *dev, Error **errp)
>      VirtIODevice *vdev = VIRTIO_DEVICE(dev);
>      VirtIOBalloon *s = VIRTIO_BALLOON(dev);
>  
> +    qemu_del_vm_change_state_handler(s->change);
>      balloon_stats_destroy_timer(s);
>      qemu_remove_balloon_handler(s);
>      virtio_cleanup(vdev);
> diff --git a/include/hw/virtio/virtio-balloon.h 
> b/include/hw/virtio/virtio-balloon.h
> index 1ea13bd..d72ff7f 100644
> --- a/include/hw/virtio/virtio-balloon.h
> +++ b/include/hw/virtio/virtio-balloon.h
> @@ -43,6 +43,7 @@ typedef struct VirtIOBalloon {
>      int64_t stats_last_update;
>      int64_t stats_poll_interval;
>      uint32_t host_features;
> +    VMChangeStateEntry *change;
>  } VirtIOBalloon;
>  
>  #endif
> -- 
> 2.5.5



reply via email to

[Prev in Thread] Current Thread [Next in Thread]