[Qemu-stable] [PATCH] rbd: add an asynchronous flush

[Josh Durgin]

The existing bdrv_co_flush_to_disk implementation uses rbd_flush(),
which is sychronous and causes the main qemu thread to block until it
is complete. This results in unresponsiveness and extra latency for
the guest.

Fix this by using an asynchronous version of flush.  This was added to
librbd with a special #define to indicate its presence, since it will
be backported to stable versions. Thus, there is no need to check the
version of librbd.

Implement this as bdrv_aio_flush, since it matches other aio functions
in the rbd block driver, and leave out bdrv_co_flush_to_disk when the
asynchronous version is available.

Reported-by: Oliver Francke <address@hidden>
Signed-off-by: Josh Durgin <address@hidden>
Signed-off-by: Stefan Hajnoczi <address@hidden>
(cherry picked from commit dc7588c1eb3008bda53dde1d6b890cd299758155)
---

Please add to the 1.4.2 stable series. It's a significant guest
performance increase, especially for interactivity
(see http://tracker.ceph.com/issues/3737).

 block/rbd.c |   37 +++++++++++++++++++++++++++++++++----
 1 files changed, 33 insertions(+), 4 deletions(-)

diff --git a/block/rbd.c b/block/rbd.c
index 8cd10a7..6296223 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -63,7 +63,8 @@
 typedef enum {
     RBD_AIO_READ,
     RBD_AIO_WRITE,
-    RBD_AIO_DISCARD
+    RBD_AIO_DISCARD,
+    RBD_AIO_FLUSH
 } RBDAIOCmd;
 
 typedef struct RBDAIOCB {
@@ -379,8 +380,7 @@ static void qemu_rbd_complete_aio(RADOSCB *rcb)
 
     r = rcb->ret;
 
-    if (acb->cmd == RBD_AIO_WRITE ||
-        acb->cmd == RBD_AIO_DISCARD) {
+    if (acb->cmd != RBD_AIO_READ) {
         if (r < 0) {
             acb->ret = r;
             acb->error = 1;
@@ -658,6 +658,16 @@ static int rbd_aio_discard_wrapper(rbd_image_t image,
 #endif
 }
 
+static int rbd_aio_flush_wrapper(rbd_image_t image,
+                                 rbd_completion_t comp)
+{
+#ifdef LIBRBD_SUPPORTS_AIO_FLUSH
+    return rbd_aio_flush(image, comp);
+#else
+    return -ENOTSUP;
+#endif
+}
+
 static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
                                        int64_t sector_num,
                                        QEMUIOVector *qiov,
@@ -678,7 +688,7 @@ static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
     acb = qemu_aio_get(&rbd_aiocb_info, bs, cb, opaque);
     acb->cmd = cmd;
     acb->qiov = qiov;
-    if (cmd == RBD_AIO_DISCARD) {
+    if (cmd == RBD_AIO_DISCARD || cmd == RBD_AIO_FLUSH) {
         acb->bounce = NULL;
     } else {
         acb->bounce = qemu_blockalign(bs, qiov->size);
@@ -722,6 +732,9 @@ static BlockDriverAIOCB *rbd_start_aio(BlockDriverState *bs,
     case RBD_AIO_DISCARD:
         r = rbd_aio_discard_wrapper(s->image, off, size, c);
         break;
+    case RBD_AIO_FLUSH:
+        r = rbd_aio_flush_wrapper(s->image, c);
+        break;
     default:
         r = -EINVAL;
     }
@@ -761,6 +774,16 @@ static BlockDriverAIOCB 
*qemu_rbd_aio_writev(BlockDriverState *bs,
                          RBD_AIO_WRITE);
 }
 
+#ifdef LIBRBD_SUPPORTS_AIO_FLUSH
+static BlockDriverAIOCB *qemu_rbd_aio_flush(BlockDriverState *bs,
+                                            BlockDriverCompletionFunc *cb,
+                                            void *opaque)
+{
+    return rbd_start_aio(bs, 0, NULL, 0, cb, opaque, RBD_AIO_FLUSH);
+}
+
+#else
+
 static int qemu_rbd_co_flush(BlockDriverState *bs)
 {
 #if LIBRBD_VERSION_CODE >= LIBRBD_VERSION(0, 1, 1)
@@ -771,6 +794,7 @@ static int qemu_rbd_co_flush(BlockDriverState *bs)
     return 0;
 #endif
 }
+#endif
 
 static int qemu_rbd_getinfo(BlockDriverState *bs, BlockDriverInfo *bdi)
 {
@@ -948,7 +972,12 @@ static BlockDriver bdrv_rbd = {
 
     .bdrv_aio_readv         = qemu_rbd_aio_readv,
     .bdrv_aio_writev        = qemu_rbd_aio_writev,
+
+#ifdef LIBRBD_SUPPORTS_AIO_FLUSH
+    .bdrv_aio_flush         = qemu_rbd_aio_flush,
+#else
     .bdrv_co_flush_to_disk  = qemu_rbd_co_flush,
+#endif
 
 #ifdef LIBRBD_SUPPORTS_DISCARD
     .bdrv_aio_discard       = qemu_rbd_aio_discard,
-- 
1.7.2.5