Commit 9564e138 authored by Adam Litke's avatar Adam Litke Committed by Rusty Russell

virtio: Add memory statistics reporting to the balloon driver (V4)

Changes since V3:
 - Do not do endian conversions as they will be done in the host
 - Report stats that reference a quantity of memory in bytes
 - Minor coding style updates

Changes since V2:
 - Increase stat field size to 64 bits
 - Report all sizes in kb (not pages)
 - Drop anon_pages stat and fix endianness conversion

Changes since V1:
 - Use a virtqueue instead of the device config space

When using ballooning to manage overcommitted memory on a host, a system for
guests to communicate their memory usage to the host can provide information
that will minimize the impact of ballooning on the guests.  The current method
employs a daemon running in each guest that communicates memory statistics to a
host daemon at a specified time interval.  The host daemon aggregates this
information and inflates and/or deflates balloons according to the level of
host memory pressure.  This approach is effective but overly complex since a
daemon must be installed inside each guest and coordinated to communicate with
the host.  A simpler approach is to collect memory statistics in the virtio
balloon driver and communicate them directly to the hypervisor.

This patch enables the guest-side support by adding stats collection and
reporting to the virtio balloon driver.
Signed-off-by: default avatarAdam Litke <>
Cc: Anthony Liguori <>
Signed-off-by: Rusty Russell <> (minor fixes)
parent 1f08b833
......@@ -28,7 +28,7 @@
struct virtio_balloon
struct virtio_device *vdev;
struct virtqueue *inflate_vq, *deflate_vq;
struct virtqueue *inflate_vq, *deflate_vq, *stats_vq;
/* Where the ballooning thread waits for config to change. */
wait_queue_head_t config_change;
......@@ -49,6 +49,9 @@ struct virtio_balloon
/* The array of pfns we tell the Host about. */
unsigned int num_pfns;
u32 pfns[256];
/* Memory statistics */
struct virtio_balloon_stat stats[VIRTIO_BALLOON_S_NR];
static struct virtio_device_id id_table[] = {
......@@ -154,6 +157,62 @@ static void leak_balloon(struct virtio_balloon *vb, size_t num)
static inline void update_stat(struct virtio_balloon *vb, int idx,
u16 tag, u64 val)
vb->stats[idx].tag = tag;
vb->stats[idx].val = val;
#define pages_to_bytes(x) ((u64)(x) << PAGE_SHIFT)
static void update_balloon_stats(struct virtio_balloon *vb)
unsigned long events[NR_VM_EVENT_ITEMS];
struct sysinfo i;
int idx = 0;
update_stat(vb, idx++, VIRTIO_BALLOON_S_SWAP_IN,
update_stat(vb, idx++, VIRTIO_BALLOON_S_SWAP_OUT,
update_stat(vb, idx++, VIRTIO_BALLOON_S_MAJFLT, events[PGMAJFAULT]);
update_stat(vb, idx++, VIRTIO_BALLOON_S_MINFLT, events[PGFAULT]);
update_stat(vb, idx++, VIRTIO_BALLOON_S_MEMFREE,
update_stat(vb, idx++, VIRTIO_BALLOON_S_MEMTOT,
* While most virtqueues communicate guest-initiated requests to the hypervisor,
* the stats queue operates in reverse. The driver initializes the virtqueue
* with a single buffer. From that point forward, all conversations consist of
* a hypervisor request (a call to this function) which directs us to refill
* the virtqueue with a fresh stats buffer.
static void stats_ack(struct virtqueue *vq)
struct virtio_balloon *vb;
unsigned int len;
struct scatterlist sg;
vb = vq->vq_ops->get_buf(vq, &len);
if (!vb)
sg_init_one(&sg, vb->stats, sizeof(vb->stats));
if (vq->vq_ops->add_buf(vq, &sg, 1, 0, vb) < 0)
static void virtballoon_changed(struct virtio_device *vdev)
struct virtio_balloon *vb = vdev->priv;
......@@ -204,10 +263,10 @@ static int balloon(void *_vballoon)
static int virtballoon_probe(struct virtio_device *vdev)
struct virtio_balloon *vb;
struct virtqueue *vqs[2];
vq_callback_t *callbacks[] = { balloon_ack, balloon_ack };
const char *names[] = { "inflate", "deflate" };
int err;
struct virtqueue *vqs[3];
vq_callback_t *callbacks[] = { balloon_ack, balloon_ack, stats_ack };
const char *names[] = { "inflate", "deflate", "stats" };
int err, nvqs;
vdev->priv = vb = kmalloc(sizeof(*vb), GFP_KERNEL);
if (!vb) {
......@@ -220,13 +279,29 @@ static int virtballoon_probe(struct virtio_device *vdev)
vb->vdev = vdev;
/* We expect two virtqueues. */
err = vdev->config->find_vqs(vdev, 2, vqs, callbacks, names);
/* We expect two virtqueues: inflate and deflate,
* and optionally stat. */
nvqs = virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_STATS_VQ) ? 3 : 2;
err = vdev->config->find_vqs(vdev, nvqs, vqs, callbacks, names);
if (err)
goto out_free_vb;
vb->inflate_vq = vqs[0];
vb->deflate_vq = vqs[1];
if (virtio_has_feature(vb->vdev, VIRTIO_BALLOON_F_STATS_VQ)) {
struct scatterlist sg;
vb->stats_vq = vqs[2];
* Prime this virtqueue with one buffer so the hypervisor can
* use it to signal us later.
sg_init_one(&sg, vb->stats, sizeof vb->stats);
if (vb->stats_vq->vq_ops->add_buf(vb->stats_vq,
&sg, 1, 0, vb) < 0)
vb->thread = kthread_run(balloon, vb, "vballoon");
if (IS_ERR(vb->thread)) {
......@@ -264,7 +339,10 @@ static void __devexit virtballoon_remove(struct virtio_device *vdev)
static unsigned int features[] = { VIRTIO_BALLOON_F_MUST_TELL_HOST };
static unsigned int features[] = {
static struct virtio_driver virtio_balloon_driver = {
.feature_table = features,
......@@ -7,6 +7,7 @@
/* The feature bitmap for virtio balloon */
#define VIRTIO_BALLOON_F_MUST_TELL_HOST 0 /* Tell before reclaiming pages */
#define VIRTIO_BALLOON_F_STATS_VQ 1 /* Memory Stats virtqueue */
/* Size of a PFN in the balloon interface. */
......@@ -18,4 +19,18 @@ struct virtio_balloon_config
/* Number of pages we've actually got in balloon. */
__le32 actual;
#define VIRTIO_BALLOON_S_SWAP_IN 0 /* Amount of memory swapped in */
#define VIRTIO_BALLOON_S_SWAP_OUT 1 /* Amount of memory swapped out */
#define VIRTIO_BALLOON_S_MAJFLT 2 /* Number of major faults */
#define VIRTIO_BALLOON_S_MINFLT 3 /* Number of minor faults */
#define VIRTIO_BALLOON_S_MEMFREE 4 /* Total amount of free memory */
#define VIRTIO_BALLOON_S_MEMTOT 5 /* Total amount of memory */
struct virtio_balloon_stat {
u16 tag;
u64 val;
} __attribute__((packed));
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment