[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PATCH 11/11] rbd: add an asynchronous flush
From: |
Stefan Hajnoczi |
Subject: |
[Qemu-devel] [PATCH 11/11] rbd: add an asynchronous flush |
Date: |
Mon, 15 Apr 2013 10:22:44 +0200 |
From: Josh Durgin <address@hidden>
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>
---
block/rbd.c | 37 +++++++++++++++++++++++++++++++++----
1 file changed, 33 insertions(+), 4 deletions(-)
diff --git a/block/rbd.c b/block/rbd.c
index 1a8ea6d..141b488 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;
@@ -659,6 +659,16 @@ 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,
@@ -679,7 +689,7 @@ 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);
@@ -723,6 +733,9 @@ 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;
}
@@ -762,6 +775,16 @@ 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)
@@ -772,6 +795,7 @@ static int qemu_rbd_co_flush(BlockDriverState *bs)
return 0;
#endif
}
+#endif
static int qemu_rbd_getinfo(BlockDriverState *bs, BlockDriverInfo *bdi)
{
@@ -949,7 +973,12 @@ 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,
--
1.8.1.4
- [Qemu-devel] [PATCH 01/11] block: Introduce bdrv_writev_vmstate, (continued)
- [Qemu-devel] [PATCH 01/11] block: Introduce bdrv_writev_vmstate, Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 02/11] savevm: Implement block_writev_buffer(), Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 03/11] block: Introduce bdrv_pwritev() for qcow2_save_vmstate, Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 04/11] qemu-iotests: A few more bdrv_pread/pwrite tests, Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 05/11] qemu-iotests: Add test for -drive options, Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 07/11] ide: refuse WIN_READ_NATIVE_MAX on empty device, Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 06/11] qemu-iotests: filter QEMU_PROG in 051.out, Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 10/11] iotests: Add 'check -ssh' option to test Secure Shell block device., Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 09/11] block: ssh: Use libssh2_sftp_fsync (if supported by libssh2) to flush to disk., Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 08/11] block: Add support for Secure Shell (ssh) block device., Stefan Hajnoczi, 2013/04/15
- [Qemu-devel] [PATCH 11/11] rbd: add an asynchronous flush,
Stefan Hajnoczi <=
- [Qemu-devel] [PULL 00/11] Block patches, Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 01/11] block: Introduce bdrv_writev_vmstate, Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 01/11] block/ssh: Require libssh2 >= 1.2.8., Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 02/11] savevm: Implement block_writev_buffer(), Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 03/11] block: Introduce bdrv_pwritev() for qcow2_save_vmstate, Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 02/11] sheepdog: add discard/trim support for sheepdog, Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 03/11] sheepdog: use BDRV_SECTOR_SIZE, Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 04/11] qemu-iotests: A few more bdrv_pread/pwrite tests, Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 04/11] sheepdog: implement .bdrv_co_is_allocated(), Stefan Hajnoczi, 2013/04/26
- [Qemu-devel] [PATCH 05/11] block: Disable driver-specific options for 1.5, Stefan Hajnoczi, 2013/04/26