Re: [Qemu-devel] [PATCH] Disable virtio-balloon memory stats interface

[Adam Litke <agl@us.ibm.com>]

Ok, I can see how this will work better for the migration case.

Acked-by: Adam Litke <agl@us.ibm.com>

On Tue, 2010-09-14 at 11:09 -0300, Eduardo Habkost wrote:
> This field is guest-visible, won't this cause problems on migration?
> 
> Isn't it better to disable it on the "info balloon" side, so the guest
> knows that the host may start requesting stat info eventually? I suggest
> doing this:
> 
> ---
> diff --git a/hw/virtio-balloon.c b/hw/virtio-balloon.c
> index 1e4dfdd..92e9057 100644
> --- a/hw/virtio-balloon.c
> +++ b/hw/virtio-balloon.c
> @@ -30,6 +30,10 @@
>  #include <sys/mman.h>
>  #endif
> 
> +/* Disable guest-provided stats by now (bz#623903, bz#626544) */
> +#define ENABLE_GUEST_STATS   0
> +
> +
>  typedef struct VirtIOBalloon
>  {
>      VirtIODevice vdev;
> @@ -84,12 +88,14 @@ static QObject *get_stats_qobject(VirtIOBalloon *dev)
>                                    VIRTIO_BALLOON_PFN_SHIFT);
> 
>      stat_put(dict, "actual", actual);
> +#if ENABLE_GUEST_STATS
>      stat_put(dict, "mem_swapped_in", dev->stats[VIRTIO_BALLOON_S_SWAP_IN]);
>      stat_put(dict, "mem_swapped_out", dev->stats[VIRTIO_BALLOON_S_SWAP_OUT]);
>      stat_put(dict, "major_page_faults", dev->stats[VIRTIO_BALLOON_S_MAJFLT]);
>      stat_put(dict, "minor_page_faults", dev->stats[VIRTIO_BALLOON_S_MINFLT]);
>      stat_put(dict, "free_mem", dev->stats[VIRTIO_BALLOON_S_MEMFREE]);
>      stat_put(dict, "total_mem", dev->stats[VIRTIO_BALLOON_S_MEMTOT]);
> +#endif
> 
>      return QOBJECT(dict);
>  }
> @@ -215,7 +221,7 @@ static void virtio_balloon_to_target(void *opaque, ram_addr_t target,
>          }
>          dev->stats_callback = cb;
>          dev->stats_opaque_callback_data = cb_data; 
> -        if (dev->vdev.guest_features & (1 << VIRTIO_BALLOON_F_STATS_VQ)) {
> +        if (ENABLE_GUEST_STATS && (dev->vdev.guest_features & (1 << VIRTIO_BALLOON_F_STATS_VQ))) {
>              virtqueue_push(dev->svq, &dev->stats_vq_elem, dev->stats_vq_offset);
>              virtio_notify(&dev->vdev, dev->svq);
>          } else {
> 

-- 
Thanks,
Adam