pve-qemu-qoup/debian/patches/pve/0053-Revert-block-rbd-implement-bdrv_co_block_status.patch
Fabian Ebner dc265df350 add revert to work around performance regression when backing up large RBD disk
resulting in QMP timeouts and very slow backups. The plan is to figure
out (ideally together with upstream) a way to make the implementation
of bdrv_co_block_status for RBD more efficient. But for now, revert
the problematic change as a stop-gap measure.

Upstream bug report:
https://gitlab.com/qemu-project/qemu/-/issues/1026

Forum threads:
https://forum.proxmox.com/threads/109272/
https://forum.proxmox.com/threads/109448/
https://forum.proxmox.com/threads/101334/ (partially)

Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
2022-05-19 09:23:38 +02:00

162 lines
5.3 KiB
Diff

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Fabian Ebner <f.ebner@proxmox.com>
Date: Tue, 17 May 2022 09:46:02 +0200
Subject: [PATCH] Revert "block/rbd: implement bdrv_co_block_status"
During backup, bdrv_co_block_status is called for each block copy
chunk. When RBD is used, the current implementation with
rbd_diff_iterate2() using whole_object=true takes about linearly more
time, depending on the image size. Since there are linearly more
chunks, the slowdown is quadratic, becoming unacceptable for large
images (starting somewhere between 500-1000 GiB in my testing).
This reverts commit 0347a8fd4c3faaedf119be04c197804be40a384b as a
stop-gap measure, until it's clear how to make the implemenation
more efficient.
Upstream bug report:
https://gitlab.com/qemu-project/qemu/-/issues/1026
Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
---
block/rbd.c | 112 ----------------------------------------------------
1 file changed, 112 deletions(-)
diff --git a/block/rbd.c b/block/rbd.c
index a4b8fb482c..3393b06a4e 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -97,12 +97,6 @@ typedef struct RBDTask {
int64_t ret;
} RBDTask;
-typedef struct RBDDiffIterateReq {
- uint64_t offs;
- uint64_t bytes;
- bool exists;
-} RBDDiffIterateReq;
-
static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
BlockdevOptionsRbd *opts, bool cache,
const char *keypairs, const char *secretid,
@@ -1267,111 +1261,6 @@ static ImageInfoSpecific *qemu_rbd_get_specific_info(BlockDriverState *bs,
return spec_info;
}
-/*
- * rbd_diff_iterate2 allows to interrupt the exection by returning a negative
- * value in the callback routine. Choose a value that does not conflict with
- * an existing exitcode and return it if we want to prematurely stop the
- * execution because we detected a change in the allocation status.
- */
-#define QEMU_RBD_EXIT_DIFF_ITERATE2 -9000
-
-static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
- int exists, void *opaque)
-{
- RBDDiffIterateReq *req = opaque;
-
- assert(req->offs + req->bytes <= offs);
- /*
- * we do not diff against a snapshot so we should never receive a callback
- * for a hole.
- */
- assert(exists);
-
- if (!req->exists && offs > req->offs) {
- /*
- * we started in an unallocated area and hit the first allocated
- * block. req->bytes must be set to the length of the unallocated area
- * before the allocated area. stop further processing.
- */
- req->bytes = offs - req->offs;
- return QEMU_RBD_EXIT_DIFF_ITERATE2;
- }
-
- if (req->exists && offs > req->offs + req->bytes) {
- /*
- * we started in an allocated area and jumped over an unallocated area,
- * req->bytes contains the length of the allocated area before the
- * unallocated area. stop further processing.
- */
- return QEMU_RBD_EXIT_DIFF_ITERATE2;
- }
-
- req->bytes += len;
- req->exists = true;
-
- return 0;
-}
-
-static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum,
- int64_t *map,
- BlockDriverState **file)
-{
- BDRVRBDState *s = bs->opaque;
- int status, r;
- RBDDiffIterateReq req = { .offs = offset };
- uint64_t features, flags;
-
- assert(offset + bytes <= s->image_size);
-
- /* default to all sectors allocated */
- status = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID;
- *map = offset;
- *file = bs;
- *pnum = bytes;
-
- /* check if RBD image supports fast-diff */
- r = rbd_get_features(s->image, &features);
- if (r < 0) {
- return status;
- }
- if (!(features & RBD_FEATURE_FAST_DIFF)) {
- return status;
- }
-
- /* check if RBD fast-diff result is valid */
- r = rbd_get_flags(s->image, &flags);
- if (r < 0) {
- return status;
- }
- if (flags & RBD_FLAG_FAST_DIFF_INVALID) {
- return status;
- }
-
- r = rbd_diff_iterate2(s->image, NULL, offset, bytes, true, true,
- qemu_rbd_diff_iterate_cb, &req);
- if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
- return status;
- }
- assert(req.bytes <= bytes);
- if (!req.exists) {
- if (r == 0) {
- /*
- * rbd_diff_iterate2 does not invoke callbacks for unallocated
- * areas. This here catches the case where no callback was
- * invoked at all (req.bytes == 0).
- */
- assert(req.bytes == 0);
- req.bytes = bytes;
- }
- status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
- }
-
- *pnum = req.bytes;
- return status;
-}
-
static int64_t qemu_rbd_getlength(BlockDriverState *bs)
{
BDRVRBDState *s = bs->opaque;
@@ -1607,7 +1496,6 @@ static BlockDriver bdrv_rbd = {
#ifdef LIBRBD_SUPPORTS_WRITE_ZEROES
.bdrv_co_pwrite_zeroes = qemu_rbd_co_pwrite_zeroes,
#endif
- .bdrv_co_block_status = qemu_rbd_co_block_status,
.bdrv_snapshot_create = qemu_rbd_snap_create,
.bdrv_snapshot_delete = qemu_rbd_snap_remove,