rpms/qemu-kvm
7
0
Fork 0
Code Issues Pull Requests Releases Activity

* Mon Jun 16 2025 Jon Maloy <jmaloy@redhat.com> - 9.1.0-24

Browse Source 
- kvm-s390x-pci-add-support-for-guests-that-request-direct.patch [RHEL-11430]
- kvm-s390x-pci-indicate-QEMU-supports-relaxed-translation.patch [RHEL-11430]
- kvm-block-Expand-block-status-mode-from-bool-to-flags.patch [RHEL-82906 RHEL-83015]
- kvm-file-posix-gluster-Handle-zero-block-status-hint-bet.patch [RHEL-82906 RHEL-83015]
- kvm-block-Let-bdrv_co_is_zero_fast-consolidate-adjacent-.patch [RHEL-82906 RHEL-83015]
- kvm-block-Add-new-bdrv_co_is_all_zeroes-function.patch [RHEL-82906 RHEL-83015]
- kvm-iotests-Improve-iotest-194-to-mirror-data.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Minor-refactoring.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Pass-full-sync-mode-rather-than-bool-to-inter.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Allow-QMP-override-to-declare-target-already-.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Drop-redundant-zero_target-parameter.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Skip-pre-zeroing-destination-if-it-is-already.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Skip-writing-zeroes-when-target-is-already-ze.patch [RHEL-82906 RHEL-83015]
- kvm-iotests-common.rc-add-disk_usage-function.patch [RHEL-82906 RHEL-83015]
- kvm-tests-Add-iotest-mirror-sparse-for-recent-patches.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Reduce-I-O-when-destination-is-detect-zeroes-.patch [RHEL-82906 RHEL-83015]
- Resolves: RHEL-11430
  ([IBM 9.7 FEAT] KVM: Performance Enhanced Refresh PCI Translation - qemu part)
- Resolves: RHEL-82906
  (--migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7])
- Resolves: RHEL-83015
  (Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7])
This commit is contained in:
Jon Maloy 2025-06-16 17:59:51 -04:00
parent fe09579d2f
commit 0dc0c63dbf
17 changed files with 3402 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

145
kvm-block-Add-new-bdrv_co_is_all_zeroes-function.patch Normal file
View File

@ -0,0 +1,145 @@
From f2cd96a040dd7863484d22a3995a2904605dadde Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:21 -0500
Subject: [PATCH 06/16] block: Add new bdrv_co_is_all_zeroes() function
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [4/14] aabcba8323df698a72842f299e9242a5eee3aea6 (ebblake/centos-qemu-kvm)
There are some optimizations that require knowing if an image starts
out as reading all zeroes, such as making blockdev-mirror faster by
skipping the copying of source zeroes to the destination. The
existing bdrv_co_is_zero_fast() is a good building block for answering
this question, but it tends to give an answer of 0 for a file we just
created via QMP 'blockdev-create' or similar (such as 'qemu-img create
-f raw'). Why? Because file-posix.c insists on allocating a tiny
header to any file rather than leaving it 100% sparse, due to some
filesystems that are unable to answer alignment probes on a hole. But
teaching file-posix.c to read the tiny header doesn't scale - the
problem of a small header is also visible when libvirt sets up an NBD
client to a just-created file on a migration destination host.
So, we need a wrapper function that handles a bit more complexity in a
common manner for all block devices - when the BDS is mostly a hole,
but has a small non-hole header, it is still worth the time to read
that header and check if it reads as all zeroes before giving up and
returning a pessimistic answer.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-19-eblake@redhat.com>
(cherry picked from commit 52726096707c5c8b90597c445de897fa64d56e73)
Conflicts:
block/io.c - context with header names
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/io.c | 62 ++++++++++++++++++++++++++++++++++++++++
include/block/block-io.h | 2 ++
2 files changed, 64 insertions(+)
diff --git a/block/io.c b/block/io.c
index 293c5dd393..1f01337599 100644
--- a/block/io.c
+++ b/block/io.c
@@ -38,10 +38,14 @@
#include "qemu/error-report.h"
#include "qemu/main-loop.h"
#include "sysemu/replay.h"
+#include "qemu/units.h"
/* Maximum bounce buffer for copy-on-read and write zeroes, in bytes */
#define MAX_BOUNCE_BUFFER (32768 << BDRV_SECTOR_BITS)
+/* Maximum read size for checking if data reads as zero, in bytes */
+#define MAX_ZERO_CHECK_BUFFER (128 * KiB)
+
static void coroutine_fn GRAPH_RDLOCK
bdrv_parent_cb_resize(BlockDriverState *bs);
@@ -2774,6 +2778,64 @@ int coroutine_fn bdrv_co_is_zero_fast(BlockDriverState *bs, int64_t offset,
return 1;
}
+/*
+ * Check @bs (and its backing chain) to see if the entire image is known
+ * to read as zeroes.
+ * Return 1 if that is the case, 0 otherwise and -errno on error.
+ * This test is meant to be fast rather than accurate so returning 0
+ * does not guarantee non-zero data; however, a return of 1 is reliable,
+ * and this function can report 1 in more cases than bdrv_co_is_zero_fast.
+ */
+int coroutine_fn bdrv_co_is_all_zeroes(BlockDriverState *bs)
+{
+ int ret;
+ int64_t pnum, bytes;
+ char *buf;
+ QEMUIOVector local_qiov;
+ IO_CODE();
+
+ bytes = bdrv_co_getlength(bs);
+ if (bytes < 0) {
+ return bytes;
+ }
+
+ /* First probe - see if the entire image reads as zero */
+ ret = bdrv_co_common_block_status_above(bs, NULL, false, BDRV_WANT_ZERO,
+ 0, bytes, &pnum, NULL, NULL,
+ NULL);
+ if (ret < 0) {
+ return ret;
+ }
+ if (ret & BDRV_BLOCK_ZERO) {
+ return bdrv_co_is_zero_fast(bs, pnum, bytes - pnum);
+ }
+
+ /*
+ * Because of the way 'blockdev-create' works, raw files tend to
+ * be created with a non-sparse region at the front to make
+ * alignment probing easier. If the block starts with only a
+ * small allocated region, it is still worth the effort to see if
+ * the rest of the image is still sparse, coupled with manually
+ * reading the first region to see if it reads zero after all.
+ */
+ if (pnum > MAX_ZERO_CHECK_BUFFER) {
+ return 0;
+ }
+ ret = bdrv_co_is_zero_fast(bs, pnum, bytes - pnum);
+ if (ret <= 0) {
+ return ret;
+ }
+ /* Only the head of the image is unknown, and it's small. Read it. */
+ buf = qemu_blockalign(bs, pnum);
+ qemu_iovec_init_buf(&local_qiov, buf, pnum);
+ ret = bdrv_driver_preadv(bs, 0, pnum, &local_qiov, 0, 0);
+ if (ret >= 0) {
+ ret = buffer_is_zero(buf, pnum);
+ }
+ qemu_vfree(buf);
+ return ret;
+}
+
int coroutine_fn bdrv_co_is_allocated(BlockDriverState *bs, int64_t offset,
int64_t bytes, int64_t *pnum)
{
diff --git a/include/block/block-io.h b/include/block/block-io.h
index b49e0537dd..b99cc98d26 100644
--- a/include/block/block-io.h
+++ b/include/block/block-io.h
@@ -161,6 +161,8 @@ bdrv_is_allocated_above(BlockDriverState *bs, BlockDriverState *base,
int coroutine_fn GRAPH_RDLOCK
bdrv_co_is_zero_fast(BlockDriverState *bs, int64_t offset, int64_t bytes);
+int coroutine_fn GRAPH_RDLOCK
+bdrv_co_is_all_zeroes(BlockDriverState *bs);
int GRAPH_RDLOCK
bdrv_apply_auto_read_only(BlockDriverState *bs, const char *errmsg,
--
2.48.1

689
kvm-block-Expand-block-status-mode-from-bool-to-flags.patch Normal file
View File

@ -0,0 +1,689 @@
From 26f5d221dd16137bed3527ee120cdf085e2c7e23 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:18 -0500
Subject: [PATCH 03/16] block: Expand block status mode from bool to flags
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [1/14] 9de5245def80e9815ed306e4abce9caec56cef6f (ebblake/centos-qemu-kvm)
This patch is purely mechanical, changing bool want_zero into an
unsigned int for bitwise-or of flags. As of this patch, all
implementations are unchanged (the old want_zero==true is now
mode==BDRV_WANT_PRECISE which is a superset of BDRV_WANT_ZERO); but
the callers in io.c that used to pass want_zero==false are now
prepared for future driver changes that can now distinguish bewteen
BDRV_WANT_ZERO vs. BDRV_WANT_ALLOCATED. The next patch will actually
change the file-posix driver along those lines, now that we have
more-specific hints.
As for the background why this patch is useful: right now, the
file-posix driver recognizes that if allocation is being queried, the
entire image can be reported as allocated (there is no backing file to
refer to) - but this throws away information on whether the entire
image reads as zero (trivially true if lseek(SEEK_HOLE) at offset 0
returns -ENXIO, a bit more complicated to prove if the raw file was
created with 'qemu-img create' since we intentionally allocate a small
chunk of all-zero data to help with alignment probing). Later patches
will add a generic algorithm for seeing if an entire file reads as
zeroes.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-16-eblake@redhat.com>
(cherry picked from commit c33159dec79069514f78faecfe268439226b0f5b)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/blkdebug.c | 6 ++--
block/copy-before-write.c | 4 +--
block/coroutines.h | 4 +--
block/file-posix.c | 4 +--
block/gluster.c | 4 +--
block/io.c | 51 ++++++++++++++++----------------
block/iscsi.c | 6 ++--
block/nbd.c | 4 +--
block/null.c | 6 ++--
block/parallels.c | 6 ++--
block/qcow.c | 2 +-
block/qcow2.c | 6 ++--
block/qed.c | 6 ++--
block/quorum.c | 4 +--
block/raw-format.c | 4 +--
block/rbd.c | 6 ++--
block/snapshot-access.c | 4 +--
block/vdi.c | 4 +--
block/vmdk.c | 2 +-
block/vpc.c | 2 +-
block/vvfat.c | 6 ++--
include/block/block-common.h | 11 +++++++
include/block/block_int-common.h | 27 +++++++++--------
include/block/block_int-io.h | 4 +--
tests/unit/test-block-iothread.c | 2 +-
25 files changed, 99 insertions(+), 86 deletions(-)
diff --git a/block/blkdebug.c b/block/blkdebug.c
index c95c818c38..736ae2b56b 100644
--- a/block/blkdebug.c
+++ b/block/blkdebug.c
@@ -751,9 +751,9 @@ blkdebug_co_pdiscard(BlockDriverState *bs, int64_t offset, int64_t bytes)
}
static int coroutine_fn GRAPH_RDLOCK
-blkdebug_co_block_status(BlockDriverState *bs, bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum, int64_t *map,
- BlockDriverState **file)
+blkdebug_co_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes, int64_t *pnum,
+ int64_t *map, BlockDriverState **file)
{
int err;
diff --git a/block/copy-before-write.c b/block/copy-before-write.c
index 853e01a1eb..36488cdeca 100644
--- a/block/copy-before-write.c
+++ b/block/copy-before-write.c
@@ -290,8 +290,8 @@ cbw_co_preadv_snapshot(BlockDriverState *bs, int64_t offset, int64_t bytes,
}
static int coroutine_fn GRAPH_RDLOCK
-cbw_co_snapshot_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset, int64_t bytes,
+cbw_co_snapshot_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes,
int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
diff --git a/block/coroutines.h b/block/coroutines.h
index f3226682d6..811ef12e43 100644
--- a/block/coroutines.h
+++ b/block/coroutines.h
@@ -47,7 +47,7 @@ int coroutine_fn GRAPH_RDLOCK
bdrv_co_common_block_status_above(BlockDriverState *bs,
BlockDriverState *base,
bool include_base,
- bool want_zero,
+ unsigned int mode,
int64_t offset,
int64_t bytes,
int64_t *pnum,
@@ -78,7 +78,7 @@ int co_wrapper_mixed_bdrv_rdlock
bdrv_common_block_status_above(BlockDriverState *bs,
BlockDriverState *base,
bool include_base,
- bool want_zero,
+ unsigned int mode,
int64_t offset,
int64_t bytes,
int64_t *pnum,
diff --git a/block/file-posix.c b/block/file-posix.c
index f17a3f4d10..9ca55620ca 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -3277,7 +3277,7 @@ static int find_allocation(BlockDriverState *bs, off_t start,
* well exceed it.
*/
static int coroutine_fn raw_co_block_status(BlockDriverState *bs,
- bool want_zero,
+ unsigned int mode,
int64_t offset,
int64_t bytes, int64_t *pnum,
int64_t *map,
@@ -3293,7 +3293,7 @@ static int coroutine_fn raw_co_block_status(BlockDriverState *bs,
return ret;
}
- if (!want_zero) {
+ if (mode != BDRV_WANT_PRECISE) {
*pnum = bytes;
*map = offset;
*file = bs;
diff --git a/block/gluster.c b/block/gluster.c
index f8b415f381..ae5c45666b 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -1466,7 +1466,7 @@ exit:
* (Based on raw_co_block_status() from file-posix.c.)
*/
static int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs,
- bool want_zero,
+ unsigned int mode,
int64_t offset,
int64_t bytes,
int64_t *pnum,
@@ -1483,7 +1483,7 @@ static int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs,
return ret;
}
- if (!want_zero) {
+ if (mode != BDRV_WANT_PRECISE) {
*pnum = bytes;
*map = offset;
*file = bs;
diff --git a/block/io.c b/block/io.c
index 3e189837a1..daaafe00d7 100644
--- a/block/io.c
+++ b/block/io.c
@@ -2360,10 +2360,8 @@ int bdrv_flush_all(void)
* Drivers not implementing the functionality are assumed to not support
* backing files, hence all their sectors are reported as allocated.
*
- * If 'want_zero' is true, the caller is querying for mapping
- * purposes, with a focus on valid BDRV_BLOCK_OFFSET_VALID, _DATA, and
- * _ZERO where possible; otherwise, the result favors larger 'pnum',
- * with a focus on accurate BDRV_BLOCK_ALLOCATED.
+ * 'mode' serves as a hint as to which results are favored; see the
+ * BDRV_WANT_* macros for details.
*
* If 'offset' is beyond the end of the disk image the return value is
* BDRV_BLOCK_EOF and 'pnum' is set to 0.
@@ -2383,7 +2381,7 @@ int bdrv_flush_all(void)
* set to the host mapping and BDS corresponding to the guest offset.
*/
static int coroutine_fn GRAPH_RDLOCK
-bdrv_co_do_block_status(BlockDriverState *bs, bool want_zero,
+bdrv_co_do_block_status(BlockDriverState *bs, unsigned int mode,
int64_t offset, int64_t bytes,
int64_t *pnum, int64_t *map, BlockDriverState **file)
{
@@ -2472,7 +2470,7 @@ bdrv_co_do_block_status(BlockDriverState *bs, bool want_zero,
local_file = bs;
local_map = aligned_offset;
} else {
- ret = bs->drv->bdrv_co_block_status(bs, want_zero, aligned_offset,
+ ret = bs->drv->bdrv_co_block_status(bs, mode, aligned_offset,
aligned_bytes, pnum, &local_map,
&local_file);
@@ -2484,10 +2482,10 @@ bdrv_co_do_block_status(BlockDriverState *bs, bool want_zero,
* the cache requires an RCU update, so double check here to avoid
* such an update if possible.
*
- * Check want_zero, because we only want to update the cache when we
+ * Check mode, because we only want to update the cache when we
* have accurate information about what is zero and what is data.
*/
- if (want_zero &&
+ if (mode == BDRV_WANT_PRECISE &&
ret == (BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID) &&
QLIST_EMPTY(&bs->children))
{
@@ -2544,7 +2542,7 @@ bdrv_co_do_block_status(BlockDriverState *bs, bool want_zero,
if (ret & BDRV_BLOCK_RAW) {
assert(ret & BDRV_BLOCK_OFFSET_VALID && local_file);
- ret = bdrv_co_do_block_status(local_file, want_zero, local_map,
+ ret = bdrv_co_do_block_status(local_file, mode, local_map,
*pnum, pnum, &local_map, &local_file);
goto out;
}
@@ -2556,7 +2554,7 @@ bdrv_co_do_block_status(BlockDriverState *bs, bool want_zero,
if (!cow_bs) {
ret |= BDRV_BLOCK_ZERO;
- } else if (want_zero) {
+ } else if (mode == BDRV_WANT_PRECISE) {
int64_t size2 = bdrv_co_getlength(cow_bs);
if (size2 >= 0 && offset >= size2) {
@@ -2565,14 +2563,14 @@ bdrv_co_do_block_status(BlockDriverState *bs, bool want_zero,
}
}
- if (want_zero && ret & BDRV_BLOCK_RECURSE &&
+ if (mode == BDRV_WANT_PRECISE && ret & BDRV_BLOCK_RECURSE &&
local_file && local_file != bs &&
(ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO) &&
(ret & BDRV_BLOCK_OFFSET_VALID)) {
int64_t file_pnum;
int ret2;
- ret2 = bdrv_co_do_block_status(local_file, want_zero, local_map,
+ ret2 = bdrv_co_do_block_status(local_file, mode, local_map,
*pnum, &file_pnum, NULL, NULL);
if (ret2 >= 0) {
/* Ignore errors. This is just providing extra information, it
@@ -2623,7 +2621,7 @@ int coroutine_fn
bdrv_co_common_block_status_above(BlockDriverState *bs,
BlockDriverState *base,
bool include_base,
- bool want_zero,
+ unsigned int mode,
int64_t offset,
int64_t bytes,
int64_t *pnum,
@@ -2650,7 +2648,7 @@ bdrv_co_common_block_status_above(BlockDriverState *bs,
return 0;
}
- ret = bdrv_co_do_block_status(bs, want_zero, offset, bytes, pnum,
+ ret = bdrv_co_do_block_status(bs, mode, offset, bytes, pnum,
map, file);
++*depth;
if (ret < 0 || *pnum == 0 || ret & BDRV_BLOCK_ALLOCATED || bs == base) {
@@ -2667,7 +2665,7 @@ bdrv_co_common_block_status_above(BlockDriverState *bs,
for (p = bdrv_filter_or_cow_bs(bs); include_base || p != base;
p = bdrv_filter_or_cow_bs(p))
{
- ret = bdrv_co_do_block_status(p, want_zero, offset, bytes, pnum,
+ ret = bdrv_co_do_block_status(p, mode, offset, bytes, pnum,
map, file);
++*depth;
if (ret < 0) {
@@ -2730,7 +2728,8 @@ int coroutine_fn bdrv_co_block_status_above(BlockDriverState *bs,
BlockDriverState **file)
{
IO_CODE();
- return bdrv_co_common_block_status_above(bs, base, false, true, offset,
+ return bdrv_co_common_block_status_above(bs, base, false,
+ BDRV_WANT_PRECISE, offset,
bytes, pnum, map, file, NULL);
}
@@ -2761,8 +2760,9 @@ int coroutine_fn bdrv_co_is_zero_fast(BlockDriverState *bs, int64_t offset,
return 1;
}
- ret = bdrv_co_common_block_status_above(bs, NULL, false, false, offset,
- bytes, &pnum, NULL, NULL, NULL);
+ ret = bdrv_co_common_block_status_above(bs, NULL, false, BDRV_WANT_ZERO,
+ offset, bytes, &pnum, NULL, NULL,
+ NULL);
if (ret < 0) {
return ret;
@@ -2778,9 +2778,9 @@ int coroutine_fn bdrv_co_is_allocated(BlockDriverState *bs, int64_t offset,
int64_t dummy;
IO_CODE();
- ret = bdrv_co_common_block_status_above(bs, bs, true, false, offset,
- bytes, pnum ? pnum : &dummy, NULL,
- NULL, NULL);
+ ret = bdrv_co_common_block_status_above(bs, bs, true, BDRV_WANT_ALLOCATED,
+ offset, bytes, pnum ? pnum : &dummy,
+ NULL, NULL, NULL);
if (ret < 0) {
return ret;
}
@@ -2813,7 +2813,8 @@ int coroutine_fn bdrv_co_is_allocated_above(BlockDriverState *bs,
int ret;
IO_CODE();
- ret = bdrv_co_common_block_status_above(bs, base, include_base, false,
+ ret = bdrv_co_common_block_status_above(bs, base, include_base,
+ BDRV_WANT_ALLOCATED,
offset, bytes, pnum, NULL, NULL,
&depth);
if (ret < 0) {
@@ -3710,8 +3711,8 @@ bdrv_co_preadv_snapshot(BdrvChild *child, int64_t offset, int64_t bytes,
}
int coroutine_fn
-bdrv_co_snapshot_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset, int64_t bytes,
+bdrv_co_snapshot_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes,
int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
@@ -3729,7 +3730,7 @@ bdrv_co_snapshot_block_status(BlockDriverState *bs,
}
bdrv_inc_in_flight(bs);
- ret = drv->bdrv_co_snapshot_block_status(bs, want_zero, offset, bytes,
+ ret = drv->bdrv_co_snapshot_block_status(bs, mode, offset, bytes,
pnum, map, file);
bdrv_dec_in_flight(bs);
diff --git a/block/iscsi.c b/block/iscsi.c
index 979bf90cb7..d7caa4b363 100644
--- a/block/iscsi.c
+++ b/block/iscsi.c
@@ -694,9 +694,9 @@ out_unlock:
static int coroutine_fn iscsi_co_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum,
- int64_t *map,
+ unsigned int mode,
+ int64_t offset, int64_t bytes,
+ int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
IscsiLun *iscsilun = bs->opaque;
diff --git a/block/nbd.c b/block/nbd.c
index d464315766..a359aa236e 100644
--- a/block/nbd.c
+++ b/block/nbd.c
@@ -1397,8 +1397,8 @@ nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int64_t bytes)
}
static int coroutine_fn GRAPH_RDLOCK nbd_client_co_block_status(
- BlockDriverState *bs, bool want_zero, int64_t offset, int64_t bytes,
- int64_t *pnum, int64_t *map, BlockDriverState **file)
+ BlockDriverState *bs, unsigned int mode, int64_t offset,
+ int64_t bytes, int64_t *pnum, int64_t *map, BlockDriverState **file)
{
int ret, request_ret;
NBDExtent64 extent = { 0 };
diff --git a/block/null.c b/block/null.c
index 4730acc1eb..95021230c8 100644
--- a/block/null.c
+++ b/block/null.c
@@ -227,9 +227,9 @@ static int null_reopen_prepare(BDRVReopenState *reopen_state,
}
static int coroutine_fn null_co_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum,
- int64_t *map,
+ unsigned int mode,
+ int64_t offset, int64_t bytes,
+ int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
BDRVNullState *s = bs->opaque;
diff --git a/block/parallels.c b/block/parallels.c
index 9205a0864f..22ea7834fd 100644
--- a/block/parallels.c
+++ b/block/parallels.c
@@ -416,9 +416,9 @@ parallels_co_flush_to_os(BlockDriverState *bs)
}
static int coroutine_fn GRAPH_RDLOCK
-parallels_co_block_status(BlockDriverState *bs, bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum, int64_t *map,
- BlockDriverState **file)
+parallels_co_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes, int64_t *pnum,
+ int64_t *map, BlockDriverState **file)
{
BDRVParallelsState *s = bs->opaque;
int count;
diff --git a/block/qcow.c b/block/qcow.c
index c2f89db055..2e18c42d8f 100644
--- a/block/qcow.c
+++ b/block/qcow.c
@@ -530,7 +530,7 @@ get_cluster_offset(BlockDriverState *bs, uint64_t offset, int allocate,
}
static int coroutine_fn GRAPH_RDLOCK
-qcow_co_block_status(BlockDriverState *bs, bool want_zero,
+qcow_co_block_status(BlockDriverState *bs, unsigned int mode,
int64_t offset, int64_t bytes, int64_t *pnum,
int64_t *map, BlockDriverState **file)
{
diff --git a/block/qcow2.c b/block/qcow2.c
index a4cffb628c..788da07fee 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2147,9 +2147,9 @@ static void qcow2_join_options(QDict *options, QDict *old_options)
}
static int coroutine_fn GRAPH_RDLOCK
-qcow2_co_block_status(BlockDriverState *bs, bool want_zero, int64_t offset,
- int64_t count, int64_t *pnum, int64_t *map,
- BlockDriverState **file)
+qcow2_co_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t count, int64_t *pnum,
+ int64_t *map, BlockDriverState **file)
{
BDRVQcow2State *s = bs->opaque;
uint64_t host_offset;
diff --git a/block/qed.c b/block/qed.c
index fa5bc11085..b135e981e5 100644
--- a/block/qed.c
+++ b/block/qed.c
@@ -832,9 +832,9 @@ fail:
}
static int coroutine_fn GRAPH_RDLOCK
-bdrv_qed_co_block_status(BlockDriverState *bs, bool want_zero, int64_t pos,
- int64_t bytes, int64_t *pnum, int64_t *map,
- BlockDriverState **file)
+bdrv_qed_co_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t pos, int64_t bytes, int64_t *pnum,
+ int64_t *map, BlockDriverState **file)
{
BDRVQEDState *s = bs->opaque;
size_t len = MIN(bytes, SIZE_MAX);
diff --git a/block/quorum.c b/block/quorum.c
index db8fe891c4..bb4ed9483e 100644
--- a/block/quorum.c
+++ b/block/quorum.c
@@ -1226,7 +1226,7 @@ static void quorum_child_perm(BlockDriverState *bs, BdrvChild *c,
* region contains zeroes, and BDRV_BLOCK_DATA otherwise.
*/
static int coroutine_fn GRAPH_RDLOCK
-quorum_co_block_status(BlockDriverState *bs, bool want_zero,
+quorum_co_block_status(BlockDriverState *bs, unsigned int mode,
int64_t offset, int64_t count,
int64_t *pnum, int64_t *map, BlockDriverState **file)
{
@@ -1238,7 +1238,7 @@ quorum_co_block_status(BlockDriverState *bs, bool want_zero,
for (i = 0; i < s->num_children; i++) {
int64_t bytes;
ret = bdrv_co_common_block_status_above(s->children[i]->bs, NULL, false,
- want_zero, offset, count,
+ mode, offset, count,
&bytes, NULL, NULL, NULL);
if (ret < 0) {
quorum_report_bad(QUORUM_OP_TYPE_READ, offset, count,
diff --git a/block/raw-format.c b/block/raw-format.c
index ac7e8495f6..623bca87a6 100644
--- a/block/raw-format.c
+++ b/block/raw-format.c
@@ -283,8 +283,8 @@ fail:
}
static int coroutine_fn GRAPH_RDLOCK
-raw_co_block_status(BlockDriverState *bs, bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum, int64_t *map,
+raw_co_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes, int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
BDRVRawState *s = bs->opaque;
diff --git a/block/rbd.c b/block/rbd.c
index 9c0fd0cb3f..627f8eb05a 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1504,9 +1504,9 @@ static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
}
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,
+ unsigned int mode,
+ int64_t offset, int64_t bytes,
+ int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
BDRVRBDState *s = bs->opaque;
diff --git a/block/snapshot-access.c b/block/snapshot-access.c
index 84d0d13f86..972b8f2e68 100644
--- a/block/snapshot-access.c
+++ b/block/snapshot-access.c
@@ -41,11 +41,11 @@ snapshot_access_co_preadv_part(BlockDriverState *bs,
static int coroutine_fn GRAPH_RDLOCK
snapshot_access_co_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset,
+ unsigned int mode, int64_t offset,
int64_t bytes, int64_t *pnum,
int64_t *map, BlockDriverState **file)
{
- return bdrv_co_snapshot_block_status(bs->file->bs, want_zero, offset,
+ return bdrv_co_snapshot_block_status(bs->file->bs, mode, offset,
bytes, pnum, map, file);
}
diff --git a/block/vdi.c b/block/vdi.c
index 6363da08ce..028fe68488 100644
--- a/block/vdi.c
+++ b/block/vdi.c
@@ -521,8 +521,8 @@ static int vdi_reopen_prepare(BDRVReopenState *state,
}
static int coroutine_fn GRAPH_RDLOCK
-vdi_co_block_status(BlockDriverState *bs, bool want_zero, int64_t offset,
- int64_t bytes, int64_t *pnum, int64_t *map,
+vdi_co_block_status(BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes, int64_t *pnum, int64_t *map,
BlockDriverState **file)
{
BDRVVdiState *s = (BDRVVdiState *)bs->opaque;
diff --git a/block/vmdk.c b/block/vmdk.c
index 78f6433607..6f1af82078 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@ -1777,7 +1777,7 @@ static inline uint64_t vmdk_find_offset_in_cluster(VmdkExtent *extent,
}
static int coroutine_fn GRAPH_RDLOCK
-vmdk_co_block_status(BlockDriverState *bs, bool want_zero,
+vmdk_co_block_status(BlockDriverState *bs, unsigned int mode,
int64_t offset, int64_t bytes, int64_t *pnum,
int64_t *map, BlockDriverState **file)
{
diff --git a/block/vpc.c b/block/vpc.c
index d95a204612..0dd641b614 100644
--- a/block/vpc.c
+++ b/block/vpc.c
@@ -721,7 +721,7 @@ fail:
}
static int coroutine_fn GRAPH_RDLOCK
-vpc_co_block_status(BlockDriverState *bs, bool want_zero,
+vpc_co_block_status(BlockDriverState *bs, unsigned int mode,
int64_t offset, int64_t bytes,
int64_t *pnum, int64_t *map,
BlockDriverState **file)
diff --git a/block/vvfat.c b/block/vvfat.c
index 8ffe8b3b9b..d59231357e 100644
--- a/block/vvfat.c
+++ b/block/vvfat.c
@@ -3135,9 +3135,9 @@ vvfat_co_pwritev(BlockDriverState *bs, int64_t offset, int64_t bytes,
}
static int coroutine_fn vvfat_co_block_status(BlockDriverState *bs,
- bool want_zero, int64_t offset,
- int64_t bytes, int64_t *n,
- int64_t *map,
+ unsigned int mode,
+ int64_t offset, int64_t bytes,
+ int64_t *n, int64_t *map,
BlockDriverState **file)
{
*n = bytes;
diff --git a/include/block/block-common.h b/include/block/block-common.h
index 7030669f04..5beee6402b 100644
--- a/include/block/block-common.h
+++ b/include/block/block-common.h
@@ -333,6 +333,17 @@ typedef enum {
#define BDRV_BLOCK_RECURSE 0x40
#define BDRV_BLOCK_COMPRESSED 0x80
+/*
+ * Block status hints: the bitwise-or of these flags emphasize what
+ * the caller hopes to learn, and some drivers may be able to give
+ * faster answers by doing less work when the hint permits.
+ */
+#define BDRV_WANT_ZERO BDRV_BLOCK_ZERO
+#define BDRV_WANT_OFFSET_VALID BDRV_BLOCK_OFFSET_VALID
+#define BDRV_WANT_ALLOCATED BDRV_BLOCK_ALLOCATED
+#define BDRV_WANT_PRECISE (BDRV_WANT_ZERO | BDRV_WANT_OFFSET_VALID | \
+ BDRV_WANT_OFFSET_VALID)
+
typedef QTAILQ_HEAD(BlockReopenQueue, BlockReopenQueueEntry) BlockReopenQueue;
typedef struct BDRVReopenState {
diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h
index ebb4e56a50..a9c0daa2a4 100644
--- a/include/block/block_int-common.h
+++ b/include/block/block_int-common.h
@@ -608,15 +608,16 @@ struct BlockDriver {
* according to the current layer, and should only need to set
* BDRV_BLOCK_DATA, BDRV_BLOCK_ZERO, BDRV_BLOCK_OFFSET_VALID,
* and/or BDRV_BLOCK_RAW; if the current layer defers to a backing
- * layer, the result should be 0 (and not BDRV_BLOCK_ZERO). See
- * block.h for the overall meaning of the bits. As a hint, the
- * flag want_zero is true if the caller cares more about precise
- * mappings (favor accurate _OFFSET_VALID/_ZERO) or false for
- * overall allocation (favor larger *pnum, perhaps by reporting
- * _DATA instead of _ZERO). The block layer guarantees input
- * clamped to bdrv_getlength() and aligned to request_alignment,
- * as well as non-NULL pnum, map, and file; in turn, the driver
- * must return an error or set pnum to an aligned non-zero value.
+ * layer, the result should be 0 (and not BDRV_BLOCK_ZERO). The
+ * caller will synthesize BDRV_BLOCK_ALLOCATED based on the
+ * non-zero results. See block.h for the overall meaning of the
+ * bits. As a hint, the flags in @mode may include a bitwise-or
+ * of BDRV_WANT_ALLOCATED, BDRV_WANT_OFFSET_VALID, or
+ * BDRV_WANT_ZERO based on what the caller is looking for in the
+ * results. The block layer guarantees input clamped to
+ * bdrv_getlength() and aligned to request_alignment, as well as
+ * non-NULL pnum, map, and file; in turn, the driver must return
+ * an error or set pnum to an aligned non-zero value.
*
* Note that @bytes is just a hint on how big of a region the
* caller wants to inspect. It is not a limit on *pnum.
@@ -628,8 +629,8 @@ struct BlockDriver {
* to clamping *pnum for return to its caller.
*/
int coroutine_fn GRAPH_RDLOCK_PTR (*bdrv_co_block_status)(
- BlockDriverState *bs,
- bool want_zero, int64_t offset, int64_t bytes, int64_t *pnum,
+ BlockDriverState *bs, unsigned int mode,
+ int64_t offset, int64_t bytes, int64_t *pnum,
int64_t *map, BlockDriverState **file);
/*
@@ -653,8 +654,8 @@ struct BlockDriver {
QEMUIOVector *qiov, size_t qiov_offset);
int coroutine_fn GRAPH_RDLOCK_PTR (*bdrv_co_snapshot_block_status)(
- BlockDriverState *bs, bool want_zero, int64_t offset, int64_t bytes,
- int64_t *pnum, int64_t *map, BlockDriverState **file);
+ BlockDriverState *bs, unsigned int mode, int64_t offset,
+ int64_t bytes, int64_t *pnum, int64_t *map, BlockDriverState **file);
int coroutine_fn GRAPH_RDLOCK_PTR (*bdrv_co_pdiscard_snapshot)(
BlockDriverState *bs, int64_t offset, int64_t bytes);
diff --git a/include/block/block_int-io.h b/include/block/block_int-io.h
index 4a7cf2b4fd..4f94eb3c5a 100644
--- a/include/block/block_int-io.h
+++ b/include/block/block_int-io.h
@@ -38,8 +38,8 @@
int coroutine_fn GRAPH_RDLOCK bdrv_co_preadv_snapshot(BdrvChild *child,
int64_t offset, int64_t bytes, QEMUIOVector *qiov, size_t qiov_offset);
int coroutine_fn GRAPH_RDLOCK bdrv_co_snapshot_block_status(
- BlockDriverState *bs, bool want_zero, int64_t offset, int64_t bytes,
- int64_t *pnum, int64_t *map, BlockDriverState **file);
+ BlockDriverState *bs, unsigned int mode, int64_t offset,
+ int64_t bytes, int64_t *pnum, int64_t *map, BlockDriverState **file);
int coroutine_fn GRAPH_RDLOCK bdrv_co_pdiscard_snapshot(BlockDriverState *bs,
int64_t offset, int64_t bytes);
diff --git a/tests/unit/test-block-iothread.c b/tests/unit/test-block-iothread.c
index 3766d5de6b..373b72fdd8 100644
--- a/tests/unit/test-block-iothread.c
+++ b/tests/unit/test-block-iothread.c
@@ -63,7 +63,7 @@ bdrv_test_co_truncate(BlockDriverState *bs, int64_t offset, bool exact,
}
static int coroutine_fn bdrv_test_co_block_status(BlockDriverState *bs,
- bool want_zero,
+ unsigned int mode,
int64_t offset, int64_t count,
int64_t *pnum, int64_t *map,
BlockDriverState **file)
--
2.48.1

90
kvm-block-Let-bdrv_co_is_zero_fast-consolidate-adjacent-.patch Normal file
View File

@ -0,0 +1,90 @@
From 9f8158e56beae4221e91feb5a98cb4db9076cac4 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:20 -0500
Subject: [PATCH 05/16] block: Let bdrv_co_is_zero_fast consolidate adjacent
extents
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [3/14] 98bf9ff773d9a36f8a8e294e38629e3f20c41334 (ebblake/centos-qemu-kvm)
Some BDS drivers have a cap on how much block status they can supply
in one query (for example, NBD talking to an older server cannot
inspect more than 4G per query; and qcow2 tends to cap its answers
rather than cross a cluster boundary of an L1 table). Although the
existing callers of bdrv_co_is_zero_fast are not passing in that large
of a 'bytes' parameter, an upcoming caller wants to query the entire
image at once, and will thus benefit from being able to treat adjacent
zero regions in a coalesced manner, rather than claiming the region is
non-zero merely because pnum was truncated and didn't match the
incoming bytes.
While refactoring this into a loop, note that there is no need to
assign pnum prior to calling bdrv_co_common_block_status_above() (it
is guaranteed to be assigned deeper in the callstack).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-18-eblake@redhat.com>
(cherry picked from commit 31bf15d97dd1d205a3b264675f9a1b3bd1939068)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/io.c | 27 +++++++++++++++------------
1 file changed, 15 insertions(+), 12 deletions(-)
diff --git a/block/io.c b/block/io.c
index daaafe00d7..293c5dd393 100644
--- a/block/io.c
+++ b/block/io.c
@@ -2747,28 +2747,31 @@ int coroutine_fn bdrv_co_block_status(BlockDriverState *bs, int64_t offset,
* by @offset and @bytes is known to read as zeroes.
* Return 1 if that is the case, 0 otherwise and -errno on error.
* This test is meant to be fast rather than accurate so returning 0
- * does not guarantee non-zero data.
+ * does not guarantee non-zero data; but a return of 1 is reliable.
*/
int coroutine_fn bdrv_co_is_zero_fast(BlockDriverState *bs, int64_t offset,
int64_t bytes)
{
int ret;
- int64_t pnum = bytes;
+ int64_t pnum;
IO_CODE();
- if (!bytes) {
- return 1;
- }
-
- ret = bdrv_co_common_block_status_above(bs, NULL, false, BDRV_WANT_ZERO,
- offset, bytes, &pnum, NULL, NULL,
- NULL);
+ while (bytes) {
+ ret = bdrv_co_common_block_status_above(bs, NULL, false,
+ BDRV_WANT_ZERO, offset, bytes,
+ &pnum, NULL, NULL, NULL);
- if (ret < 0) {
- return ret;
+ if (ret < 0) {
+ return ret;
+ }
+ if (!(ret & BDRV_BLOCK_ZERO)) {
+ return 0;
+ }
+ offset += pnum;
+ bytes -= pnum;
}
- return (pnum == bytes) && (ret & BDRV_BLOCK_ZERO);
+ return 1;
}
int coroutine_fn bdrv_co_is_allocated(BlockDriverState *bs, int64_t offset,
--
2.48.1

64
kvm-file-posix-gluster-Handle-zero-block-status-hint-bet.patch Normal file
View File

@ -0,0 +1,64 @@
From 39e0c370357a414abacd64fb6a172e7b25eb4d82 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:19 -0500
Subject: [PATCH 04/16] file-posix, gluster: Handle zero block status hint
better
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [2/14] 1f7b47ce5f5fb321aee41a16accf5bce3d1bfe95 (ebblake/centos-qemu-kvm)
Although the previous patch to change 'bool want_zero' into a bitmask
made no semantic change, it is now time to differentiate. When the
caller specifically wants to know what parts of the file read as zero,
we need to use lseek and actually reporting holes, rather than
short-circuiting and advertising full allocation.
This change will be utilized in later patches to let mirroring
optimize for the case when the destination already reads as zeroes.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-17-eblake@redhat.com>
(cherry picked from commit a6a0a7fb0e327d17594c971b4a39de14e025b415)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/file-posix.c | 3 ++-
block/gluster.c | 2 +-
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
index 9ca55620ca..ce5da2b4c2 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -3293,7 +3293,8 @@ static int coroutine_fn raw_co_block_status(BlockDriverState *bs,
return ret;
}
- if (mode != BDRV_WANT_PRECISE) {
+ if (!(mode & BDRV_WANT_ZERO)) {
+ /* There is no backing file - all bytes are allocated in this file. */
*pnum = bytes;
*map = offset;
*file = bs;
diff --git a/block/gluster.c b/block/gluster.c
index ae5c45666b..175c70164c 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -1483,7 +1483,7 @@ static int coroutine_fn qemu_gluster_co_block_status(BlockDriverState *bs,
return ret;
}
- if (mode != BDRV_WANT_PRECISE) {
+ if (!(mode & BDRV_WANT_ZERO)) {
*pnum = bytes;
*map = offset;
*file = bs;
--
2.48.1

42
kvm-iotests-Improve-iotest-194-to-mirror-data.patch Normal file
View File

@ -0,0 +1,42 @@
From 8832268a98104ba3065a57dedcd3db43231512ba Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:22 -0500
Subject: [PATCH 07/16] iotests: Improve iotest 194 to mirror data
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [5/14] bfbe8eab1035480cef9d69d1974ba66b755b1b60 (ebblake/centos-qemu-kvm)
Mirroring a completely sparse image to a sparse destination should be
practically instantaneous. It isn't yet, but the test will be more
realistic if it has some non-zero to mirror as well as the holes.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-20-eblake@redhat.com>
(cherry picked from commit eb89627899bb84148d272394e885725eff456ae9)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
tests/qemu-iotests/194 | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/qemu-iotests/194 b/tests/qemu-iotests/194
index c0ce82dd25..d0b9c084f5 100755
--- a/tests/qemu-iotests/194
+++ b/tests/qemu-iotests/194
@@ -34,6 +34,7 @@ with iotests.FilePath('source.img') as source_img_path, \
img_size = '1G'
iotests.qemu_img_create('-f', iotests.imgfmt, source_img_path, img_size)
+ iotests.qemu_io('-f', iotests.imgfmt, '-c', 'write 512M 1M', source_img_path)
iotests.qemu_img_create('-f', iotests.imgfmt, dest_img_path, img_size)
iotests.log('Launching VMs...')
--
2.48.1

68
kvm-iotests-common.rc-add-disk_usage-function.patch Normal file
View File

@ -0,0 +1,68 @@
From 644f39de9e2466a9570833b1070acf47a53863ea Mon Sep 17 00:00:00 2001
From: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>
Date: Fri, 9 May 2025 15:40:29 -0500
Subject: [PATCH 14/16] iotests/common.rc: add disk_usage function
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [12/14] 0e5d4217f97fe6e952de23eedbc2b8d9c7600665 (ebblake/centos-qemu-kvm)
Move the definition from iotests/250 to common.rc. This is used to
detect real disk usage of sparse files. In particular, we want to use
it for checking subclusters-based discards.
Signed-off-by: Andrey Drobyshev <andrey.drobyshev@virtuozzo.com>
Reviewed-by: Alexander Ivanov <alexander.ivanov@virtuozzo.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Message-ID: <20240913163942.423050-6-andrey.drobyshev@virtuozzo.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-27-eblake@redhat.com>
(cherry picked from commit be9bac072ede6e6aa27079f59efcf17b56bd7b26)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
tests/qemu-iotests/250 | 5 -----
tests/qemu-iotests/common.rc | 6 ++++++
2 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/tests/qemu-iotests/250 b/tests/qemu-iotests/250
index af48f83aba..c0a0dbc0ff 100755
--- a/tests/qemu-iotests/250
+++ b/tests/qemu-iotests/250
@@ -52,11 +52,6 @@ _unsupported_imgopts data_file
# bdrv_co_truncate(bs->file) call in qcow2_co_truncate(), which might succeed
# anyway.
-disk_usage()
-{
- du --block-size=1 $1 | awk '{print $1}'
-}
-
size=2100M
_make_test_img -o "cluster_size=1M,preallocation=metadata" $size
diff --git a/tests/qemu-iotests/common.rc b/tests/qemu-iotests/common.rc
index 95c12577dd..237f746af8 100644
--- a/tests/qemu-iotests/common.rc
+++ b/tests/qemu-iotests/common.rc
@@ -140,6 +140,12 @@ _optstr_add()
fi
}
+# report real disk usage for sparse files
+disk_usage()
+{
+ du --block-size=1 "$1" | awk '{print $1}'
+}
+
# Set the variables to the empty string to turn Valgrind off
# for specific processes, e.g.
# $ VALGRIND_QEMU_IO= ./check -qcow2 -valgrind 015
--
2.48.1

295
kvm-mirror-Allow-QMP-override-to-declare-target-already-.patch Normal file
View File

@ -0,0 +1,295 @@
From a5f6042a0c80daf3672fa071b724cb05e6f6e928 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:25 -0500
Subject: [PATCH 10/16] mirror: Allow QMP override to declare target already
zero
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [8/14] fb054864175d83e9d232464295b170808bee0e6c (ebblake/centos-qemu-kvm)
QEMU has an optimization for a just-created drive-mirror destination
that is not possible for blockdev-mirror (which can't create the
destination) - any time we know the destination starts life as all
zeroes, we can skip a pre-zeroing pass on the destination. Recent
patches have added an improved heuristic for detecting if a file
contains all zeroes, and we plan to use that heuristic in upcoming
patches. But since a heuristic cannot quickly detect all scenarios,
and there may be cases where the caller is aware of information that
QEMU cannot learn quickly, it makes sense to have a way to tell QEMU
to assume facts about the destination that can make the mirror
operation faster. Given our existing example of "qemu-img convert
--target-is-zero", it is time to expose this override in QMP for
blockdev-mirror as well.
This patch results in some slight redundancy between the older
s->zero_target (set any time mode==FULL and the destination image was
not just created - ie. clear if drive-mirror is asking to skip the
pre-zero pass) and the newly-introduced s->target_is_zero (in addition
to the QMP override, it is set when drive-mirror creates the
destination image); this will be cleaned up in the next patch.
There is also a subtlety that we must consider. When drive-mirror is
passing target_is_zero on behalf of a just-created image, we know the
image is sparse (skipping the pre-zeroing keeps it that way), so it
doesn't matter whether the destination also has "discard":"unmap" and
"detect-zeroes":"unmap". But now that we are letting the user set the
knob for target-is-zero, if the user passes a pre-existing file that
is fully allocated, it is fine to leave the file fully allocated under
"detect-zeroes":"on", but if the file is open with
"detect-zeroes":"unmap", we should really be trying harder to punch
holes in the destination for every region of zeroes copied from the
source. The easiest way to do this is to still run the pre-zeroing
pass (turning the entire destination file sparse before populating
just the allocated portions of the source), even though that currently
results in double I/O to the portions of the file that are allocated.
A later patch will add further optimizations to reduce redundant
zeroing I/O during the mirror operation.
Since "target-is-zero":true is designed for optimizations, it is okay
to silently ignore the parameter rather than erroring if the user ever
sets the parameter in a scenario where the mirror job can't exploit it
(for example, when doing "sync":"top" instead of "sync":"full", we
can't pre-zero, so setting the parameter won't make a speed
difference).
Signed-off-by: Eric Blake <eblake@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-ID: <20250509204341.3553601-23-eblake@redhat.com>
Reviewed-by: Sunny Zhu <sunnyzhyy@qq.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit d17a34bfb94bda3a89d7320ae67255ded1d8c939)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 27 ++++++++++++++++++++++----
blockdev.c | 18 ++++++++++-------
include/block/block_int-global-state.h | 3 ++-
qapi/block-core.json | 8 +++++++-
tests/unit/test-block-iothread.c | 2 +-
5 files changed, 44 insertions(+), 14 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index c8bbaa0b35..bba3e3b05c 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -55,6 +55,8 @@ typedef struct MirrorBlockJob {
BlockMirrorBackingMode backing_mode;
/* Whether the target image requires explicit zero-initialization */
bool zero_target;
+ /* Whether the target should be assumed to be already zero initialized */
+ bool target_is_zero;
/*
* To be accesssed with atomics. Written only under the BQL (required by the
* current implementation of mirror_change()).
@@ -844,12 +846,26 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
BlockDriverState *target_bs = blk_bs(s->target);
int ret = -EIO;
int64_t count;
+ bool punch_holes =
+ target_bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP &&
+ bdrv_can_write_zeroes_with_unmap(target_bs);
bdrv_graph_co_rdlock();
bs = s->mirror_top_bs->backing->bs;
bdrv_graph_co_rdunlock();
- if (s->zero_target) {
+ if (s->zero_target && (!s->target_is_zero || punch_holes)) {
+ /*
+ * Here, we are in FULL mode; our goal is to avoid writing
+ * zeroes if the destination already reads as zero, except
+ * when we are trying to punch holes. This is possible if
+ * zeroing happened externally (s->target_is_zero) or if we
+ * have a fast way to pre-zero the image (the dirty bitmap
+ * will be populated later by the non-zero portions, the same
+ * as for TOP mode). If pre-zeroing is not fast, or we need
+ * to punch holes, then our only recourse is to write the
+ * entire image.
+ */
if (!bdrv_can_write_zeroes_with_unmap(target_bs)) {
bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, s->bdev_length);
return 0;
@@ -1714,7 +1730,7 @@ static BlockJob *mirror_start_job(
uint32_t granularity, int64_t buf_size,
MirrorSyncMode sync_mode,
BlockMirrorBackingMode backing_mode,
- bool zero_target,
+ bool zero_target, bool target_is_zero,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap,
@@ -1883,6 +1899,7 @@ static BlockJob *mirror_start_job(
s->sync_mode = sync_mode;
s->backing_mode = backing_mode;
s->zero_target = zero_target;
+ s->target_is_zero = target_is_zero;
qatomic_set(&s->copy_mode, copy_mode);
s->base = base;
s->base_overlay = bdrv_find_overlay(bs, base);
@@ -2011,7 +2028,7 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
int creation_flags, int64_t speed,
uint32_t granularity, int64_t buf_size,
MirrorSyncMode mode, BlockMirrorBackingMode backing_mode,
- bool zero_target,
+ bool zero_target, bool target_is_zero,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap, const char *filter_node_name,
@@ -2034,7 +2051,8 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
mirror_start_job(job_id, bs, creation_flags, target, replaces,
speed, granularity, buf_size, mode, backing_mode,
- zero_target, on_source_error, on_target_error, unmap,
+ zero_target,
+ target_is_zero, on_source_error, on_target_error, unmap,
NULL, NULL, &mirror_job_driver, base, false,
filter_node_name, true, copy_mode, false, errp);
}
@@ -2062,6 +2080,7 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
job = mirror_start_job(
job_id, bs, creation_flags, base, NULL, speed, 0, 0,
MIRROR_SYNC_MODE_TOP, MIRROR_LEAVE_BACKING_CHAIN, false,
+ false,
on_error, on_error, true, cb, opaque,
&commit_active_job_driver, base, auto_complete,
filter_node_name, false, MIRROR_COPY_MODE_BACKGROUND,
diff --git a/blockdev.c b/blockdev.c
index 70046b6690..db11a99312 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2795,7 +2795,7 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
const char *replaces,
enum MirrorSyncMode sync,
BlockMirrorBackingMode backing_mode,
- bool zero_target,
+ bool zero_target, bool target_is_zero,
bool has_speed, int64_t speed,
bool has_granularity, uint32_t granularity,
bool has_buf_size, int64_t buf_size,
@@ -2906,11 +2906,10 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
/* pass the node name to replace to mirror start since it's loose coupling
* and will allow to check whether the node still exist at mirror completion
*/
- mirror_start(job_id, bs, target,
- replaces, job_flags,
+ mirror_start(job_id, bs, target, replaces, job_flags,
speed, granularity, buf_size, sync, backing_mode, zero_target,
- on_source_error, on_target_error, unmap, filter_node_name,
- copy_mode, errp);
+ target_is_zero, on_source_error, on_target_error, unmap,
+ filter_node_name, copy_mode, errp);
}
void qmp_drive_mirror(DriveMirror *arg, Error **errp)
@@ -2925,6 +2924,7 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
int64_t size;
const char *format = arg->format;
bool zero_target;
+ bool target_is_zero;
int ret;
bs = qmp_get_root_bs(arg->device, errp);
@@ -3041,6 +3041,8 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
zero_target = (arg->sync == MIRROR_SYNC_MODE_FULL &&
(arg->mode == NEW_IMAGE_MODE_EXISTING ||
!bdrv_has_zero_init(target_bs)));
+ target_is_zero = (arg->mode != NEW_IMAGE_MODE_EXISTING &&
+ bdrv_has_zero_init(target_bs));
bdrv_graph_rdunlock_main_loop();
@@ -3052,7 +3054,7 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
blockdev_mirror_common(arg->job_id, bs, target_bs,
arg->replaces, arg->sync,
- backing_mode, zero_target,
+ backing_mode, zero_target, target_is_zero,
arg->has_speed, arg->speed,
arg->has_granularity, arg->granularity,
arg->has_buf_size, arg->buf_size,
@@ -3082,6 +3084,7 @@ void qmp_blockdev_mirror(const char *job_id,
bool has_copy_mode, MirrorCopyMode copy_mode,
bool has_auto_finalize, bool auto_finalize,
bool has_auto_dismiss, bool auto_dismiss,
+ bool has_target_is_zero, bool target_is_zero,
Error **errp)
{
BlockDriverState *bs;
@@ -3112,7 +3115,8 @@ void qmp_blockdev_mirror(const char *job_id,
blockdev_mirror_common(job_id, bs, target_bs,
replaces, sync, backing_mode,
- zero_target, has_speed, speed,
+ zero_target, has_target_is_zero && target_is_zero,
+ has_speed, speed,
has_granularity, granularity,
has_buf_size, buf_size,
has_on_source_error, on_source_error,
diff --git a/include/block/block_int-global-state.h b/include/block/block_int-global-state.h
index eb2d92a226..8cf0003ce7 100644
--- a/include/block/block_int-global-state.h
+++ b/include/block/block_int-global-state.h
@@ -140,6 +140,7 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
* @mode: Whether to collapse all images in the chain to the target.
* @backing_mode: How to establish the target's backing chain after completion.
* @zero_target: Whether the target should be explicitly zero-initialized
+ * @target_is_zero: Whether the target already is zero-initialized.
* @on_source_error: The action to take upon error reading from the source.
* @on_target_error: The action to take upon error writing to the target.
* @unmap: Whether to unmap target where source sectors only contain zeroes.
@@ -159,7 +160,7 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
int creation_flags, int64_t speed,
uint32_t granularity, int64_t buf_size,
MirrorSyncMode mode, BlockMirrorBackingMode backing_mode,
- bool zero_target,
+ bool zero_target, bool target_is_zero,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap, const char *filter_node_name,
diff --git a/qapi/block-core.json b/qapi/block-core.json
index c1af3d1f7d..3969c60b93 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
@@ -2535,6 +2535,11 @@
# disappear from the query list without user intervention.
# Defaults to true. (Since 3.1)
#
+# @target-is-zero: Assume the destination reads as all zeroes before
+# the mirror started. Setting this to true can speed up the
+# mirror. Setting this to true when the destination is not
+# actually all zero can corrupt the destination. (Since 10.1)
+#
# Since: 2.6
#
# .. qmp-example::
@@ -2554,7 +2559,8 @@
'*on-target-error': 'BlockdevOnError',
'*filter-node-name': 'str',
'*copy-mode': 'MirrorCopyMode',
- '*auto-finalize': 'bool', '*auto-dismiss': 'bool' },
+ '*auto-finalize': 'bool', '*auto-dismiss': 'bool',
+ '*target-is-zero': 'bool'},
'allow-preconfig': true }
##
diff --git a/tests/unit/test-block-iothread.c b/tests/unit/test-block-iothread.c
index 373b72fdd8..033711d8d7 100644
--- a/tests/unit/test-block-iothread.c
+++ b/tests/unit/test-block-iothread.c
@@ -755,7 +755,7 @@ static void test_propagate_mirror(void)
/* Start a mirror job */
mirror_start("job0", src, target, NULL, JOB_DEFAULT, 0, 0, 0,
- MIRROR_SYNC_MODE_NONE, MIRROR_OPEN_BACKING_CHAIN, false,
+ MIRROR_SYNC_MODE_NONE, MIRROR_OPEN_BACKING_CHAIN, false, false,
BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
false, "filter_node", MIRROR_COPY_MODE_BACKGROUND,
&error_abort);
--
2.48.1

241
kvm-mirror-Drop-redundant-zero_target-parameter.patch Normal file
View File

@ -0,0 +1,241 @@
From 5040f835f07f3355ae80b3da2ae83ce35de022e0 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:26 -0500
Subject: [PATCH 11/16] mirror: Drop redundant zero_target parameter
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [9/14] b84a938c69e3761211b9fee4c59b465d55f61855 (ebblake/centos-qemu-kvm)
The two callers to a mirror job (drive-mirror and blockdev-mirror) set
zero_target precisely when sync mode == FULL, with the one exception
that drive-mirror skips zeroing the target if it was newly created and
reads as zero. But given the previous patch, that exception is
equally captured by target_is_zero.
Meanwhile, there is another slight wrinkle, fortunately caught by
iotest 185: if the caller uses "sync":"top" but the source has no
backing file, the code in blockdev.c was changing sync to be FULL, but
only after it had set zero_target=false. In mirror.c, prior to recent
patches, this didn't matter: the only places that inspected sync were
setting is_none_mode (both TOP and FULL had set that to false), and
mirror_start() setting base = mode == MIRROR_SYNC_MODE_TOP ?
bdrv_backing_chain_next(bs) : NULL. But now that we are passing sync
around, the slammed sync mode would result in a new pre-zeroing pass
even when the user had passed "sync":"top" in an effort to skip
pre-zeroing. Fortunately, the assignment of base when bs has no
backing chain still works out to NULL if we don't slam things. So
with the forced change of sync ripped out of blockdev.c, the sync mode
is passed through the full callstack unmolested, and we can now
reliably reconstruct the same settings as what used to be passed in by
zero_target=false, without the redundant parameter.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20250509204341.3553601-24-eblake@redhat.com>
Reviewed-by: Sunny Zhu <sunnyzhyy@qq.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
[eblake: Fix regression in iotest 185]
Signed-off-by: Eric Blake <eblake@redhat.com>
(cherry picked from commit 253b43a29077de9266351e120c600a73b82e9c49)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 13 +++++--------
blockdev.c | 19 ++++---------------
include/block/block_int-global-state.h | 3 +--
tests/unit/test-block-iothread.c | 2 +-
4 files changed, 11 insertions(+), 26 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index bba3e3b05c..b35d12adaa 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -53,8 +53,6 @@ typedef struct MirrorBlockJob {
Error *replace_blocker;
MirrorSyncMode sync_mode;
BlockMirrorBackingMode backing_mode;
- /* Whether the target image requires explicit zero-initialization */
- bool zero_target;
/* Whether the target should be assumed to be already zero initialized */
bool target_is_zero;
/*
@@ -854,7 +852,9 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
bs = s->mirror_top_bs->backing->bs;
bdrv_graph_co_rdunlock();
- if (s->zero_target && (!s->target_is_zero || punch_holes)) {
+ if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
+ /* In TOP mode, there is no benefit to a pre-zeroing pass. */
+ } else if (!s->target_is_zero || punch_holes) {
/*
* Here, we are in FULL mode; our goal is to avoid writing
* zeroes if the destination already reads as zero, except
@@ -1730,7 +1730,7 @@ static BlockJob *mirror_start_job(
uint32_t granularity, int64_t buf_size,
MirrorSyncMode sync_mode,
BlockMirrorBackingMode backing_mode,
- bool zero_target, bool target_is_zero,
+ bool target_is_zero,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap,
@@ -1898,7 +1898,6 @@ static BlockJob *mirror_start_job(
s->on_target_error = on_target_error;
s->sync_mode = sync_mode;
s->backing_mode = backing_mode;
- s->zero_target = zero_target;
s->target_is_zero = target_is_zero;
qatomic_set(&s->copy_mode, copy_mode);
s->base = base;
@@ -2028,7 +2027,7 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
int creation_flags, int64_t speed,
uint32_t granularity, int64_t buf_size,
MirrorSyncMode mode, BlockMirrorBackingMode backing_mode,
- bool zero_target, bool target_is_zero,
+ bool target_is_zero,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap, const char *filter_node_name,
@@ -2051,7 +2050,6 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
mirror_start_job(job_id, bs, creation_flags, target, replaces,
speed, granularity, buf_size, mode, backing_mode,
- zero_target,
target_is_zero, on_source_error, on_target_error, unmap,
NULL, NULL, &mirror_job_driver, base, false,
filter_node_name, true, copy_mode, false, errp);
@@ -2080,7 +2078,6 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
job = mirror_start_job(
job_id, bs, creation_flags, base, NULL, speed, 0, 0,
MIRROR_SYNC_MODE_TOP, MIRROR_LEAVE_BACKING_CHAIN, false,
- false,
on_error, on_error, true, cb, opaque,
&commit_active_job_driver, base, auto_complete,
filter_node_name, false, MIRROR_COPY_MODE_BACKGROUND,
diff --git a/blockdev.c b/blockdev.c
index db11a99312..04fa759e30 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2795,7 +2795,7 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
const char *replaces,
enum MirrorSyncMode sync,
BlockMirrorBackingMode backing_mode,
- bool zero_target, bool target_is_zero,
+ bool target_is_zero,
bool has_speed, int64_t speed,
bool has_granularity, uint32_t granularity,
bool has_buf_size, int64_t buf_size,
@@ -2862,10 +2862,6 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
return;
}
- if (!bdrv_backing_chain_next(bs) && sync == MIRROR_SYNC_MODE_TOP) {
- sync = MIRROR_SYNC_MODE_FULL;
- }
-
if (!replaces) {
/* We want to mirror from @bs, but keep implicit filters on top */
unfiltered_bs = bdrv_skip_implicit_filters(bs);
@@ -2907,7 +2903,7 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
* and will allow to check whether the node still exist at mirror completion
*/
mirror_start(job_id, bs, target, replaces, job_flags,
- speed, granularity, buf_size, sync, backing_mode, zero_target,
+ speed, granularity, buf_size, sync, backing_mode,
target_is_zero, on_source_error, on_target_error, unmap,
filter_node_name, copy_mode, errp);
}
@@ -2923,7 +2919,6 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
int flags;
int64_t size;
const char *format = arg->format;
- bool zero_target;
bool target_is_zero;
int ret;
@@ -3038,9 +3033,6 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
}
bdrv_graph_rdlock_main_loop();
- zero_target = (arg->sync == MIRROR_SYNC_MODE_FULL &&
- (arg->mode == NEW_IMAGE_MODE_EXISTING ||
- !bdrv_has_zero_init(target_bs)));
target_is_zero = (arg->mode != NEW_IMAGE_MODE_EXISTING &&
bdrv_has_zero_init(target_bs));
bdrv_graph_rdunlock_main_loop();
@@ -3054,7 +3046,7 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
blockdev_mirror_common(arg->job_id, bs, target_bs,
arg->replaces, arg->sync,
- backing_mode, zero_target, target_is_zero,
+ backing_mode, target_is_zero,
arg->has_speed, arg->speed,
arg->has_granularity, arg->granularity,
arg->has_buf_size, arg->buf_size,
@@ -3091,7 +3083,6 @@ void qmp_blockdev_mirror(const char *job_id,
BlockDriverState *target_bs;
AioContext *aio_context;
BlockMirrorBackingMode backing_mode = MIRROR_LEAVE_BACKING_CHAIN;
- bool zero_target;
int ret;
bs = qmp_get_root_bs(device, errp);
@@ -3104,8 +3095,6 @@ void qmp_blockdev_mirror(const char *job_id,
return;
}
- zero_target = (sync == MIRROR_SYNC_MODE_FULL);
-
aio_context = bdrv_get_aio_context(bs);
ret = bdrv_try_change_aio_context(target_bs, aio_context, NULL, errp);
@@ -3115,7 +3104,7 @@ void qmp_blockdev_mirror(const char *job_id,
blockdev_mirror_common(job_id, bs, target_bs,
replaces, sync, backing_mode,
- zero_target, has_target_is_zero && target_is_zero,
+ has_target_is_zero && target_is_zero,
has_speed, speed,
has_granularity, granularity,
has_buf_size, buf_size,
diff --git a/include/block/block_int-global-state.h b/include/block/block_int-global-state.h
index 8cf0003ce7..d21bd7fd2f 100644
--- a/include/block/block_int-global-state.h
+++ b/include/block/block_int-global-state.h
@@ -139,7 +139,6 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
* @buf_size: The amount of data that can be in flight at one time.
* @mode: Whether to collapse all images in the chain to the target.
* @backing_mode: How to establish the target's backing chain after completion.
- * @zero_target: Whether the target should be explicitly zero-initialized
* @target_is_zero: Whether the target already is zero-initialized.
* @on_source_error: The action to take upon error reading from the source.
* @on_target_error: The action to take upon error writing to the target.
@@ -160,7 +159,7 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
int creation_flags, int64_t speed,
uint32_t granularity, int64_t buf_size,
MirrorSyncMode mode, BlockMirrorBackingMode backing_mode,
- bool zero_target, bool target_is_zero,
+ bool target_is_zero,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
bool unmap, const char *filter_node_name,
diff --git a/tests/unit/test-block-iothread.c b/tests/unit/test-block-iothread.c
index 033711d8d7..373b72fdd8 100644
--- a/tests/unit/test-block-iothread.c
+++ b/tests/unit/test-block-iothread.c
@@ -755,7 +755,7 @@ static void test_propagate_mirror(void)
/* Start a mirror job */
mirror_start("job0", src, target, NULL, JOB_DEFAULT, 0, 0, 0,
- MIRROR_SYNC_MODE_NONE, MIRROR_OPEN_BACKING_CHAIN, false, false,
+ MIRROR_SYNC_MODE_NONE, MIRROR_OPEN_BACKING_CHAIN, false,
BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
false, "filter_node", MIRROR_COPY_MODE_BACKGROUND,
&error_abort);
--
2.48.1

92
kvm-mirror-Minor-refactoring.patch Normal file
View File

@ -0,0 +1,92 @@
From 0102da22fe5aefde9d398d539fc290ab062346f1 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:23 -0500
Subject: [PATCH 08/16] mirror: Minor refactoring
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [6/14] 886fa2e3249f48f89d3e04ba619d370031851d89 (ebblake/centos-qemu-kvm)
Commit 5791ba52 (v9.2) pre-initialized ret in mirror_dirty_init to
silence a false positive compiler warning, even though in all code
paths where ret is used, it was guaranteed to be reassigned
beforehand. But since the function returns -errno, and -1 is not
always the right errno, it's better to initialize to -EIO.
An upcoming patch wants to track two bitmaps in
do_sync_target_write(); this will be easier if the current variables
related to the dirty bitmap are renamed.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20250509204341.3553601-21-eblake@redhat.com>
(cherry picked from commit 870f8963cf1a84f8ec929b05a6d68906974a76c5)
Conflicts:
block/mirror.c - commit 5791ba52 not present
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 61f0a717b7..22f8bd98c4 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -841,7 +841,7 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
int64_t offset;
BlockDriverState *bs;
BlockDriverState *target_bs = blk_bs(s->target);
- int ret;
+ int ret = -EIO;
int64_t count;
bdrv_graph_co_rdlock();
@@ -1341,7 +1341,7 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
{
int ret;
size_t qiov_offset = 0;
- int64_t bitmap_offset, bitmap_end;
+ int64_t dirty_bitmap_offset, dirty_bitmap_end;
if (!QEMU_IS_ALIGNED(offset, job->granularity) &&
bdrv_dirty_bitmap_get(job->dirty_bitmap, offset))
@@ -1388,11 +1388,11 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
* Tails are either clean or shrunk, so for bitmap resetting
* we safely align the range down.
*/
- bitmap_offset = QEMU_ALIGN_UP(offset, job->granularity);
- bitmap_end = QEMU_ALIGN_DOWN(offset + bytes, job->granularity);
- if (bitmap_offset < bitmap_end) {
- bdrv_reset_dirty_bitmap(job->dirty_bitmap, bitmap_offset,
- bitmap_end - bitmap_offset);
+ dirty_bitmap_offset = QEMU_ALIGN_UP(offset, job->granularity);
+ dirty_bitmap_end = QEMU_ALIGN_DOWN(offset + bytes, job->granularity);
+ if (dirty_bitmap_offset < dirty_bitmap_end) {
+ bdrv_reset_dirty_bitmap(job->dirty_bitmap, dirty_bitmap_offset,
+ dirty_bitmap_end - dirty_bitmap_offset);
}
job_progress_increase_remaining(&job->common.job, bytes);
@@ -1430,10 +1430,10 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
* at function start, and they must be still dirty, as we've locked
* the region for in-flight op.
*/
- bitmap_offset = QEMU_ALIGN_DOWN(offset, job->granularity);
- bitmap_end = QEMU_ALIGN_UP(offset + bytes, job->granularity);
- bdrv_set_dirty_bitmap(job->dirty_bitmap, bitmap_offset,
- bitmap_end - bitmap_offset);
+ dirty_bitmap_offset = QEMU_ALIGN_DOWN(offset, job->granularity);
+ dirty_bitmap_end = QEMU_ALIGN_UP(offset + bytes, job->granularity);
+ bdrv_set_dirty_bitmap(job->dirty_bitmap, dirty_bitmap_offset,
+ dirty_bitmap_end - dirty_bitmap_offset);
qatomic_set(&job->actively_synced, false);
action = mirror_error_action(job, false, -ret);
--
2.48.1

139
kvm-mirror-Pass-full-sync-mode-rather-than-bool-to-inter.patch Normal file
View File

@ -0,0 +1,139 @@
From 482db3e637a16d5877e523e87c53ddb2579b4b66 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:24 -0500
Subject: [PATCH 09/16] mirror: Pass full sync mode rather than bool to
internals
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [7/14] f45a83a14b0eea07517176d44ab0c49db8233ea0 (ebblake/centos-qemu-kvm)
Out of the five possible values for MirrorSyncMode, INCREMENTAL and
BITMAP are already rejected up front in mirror_start, leaving NONE,
TOP, and FULL as the remaining values that the code was collapsing
into a single bool is_none_mode. Furthermore, mirror_dirty_init() is
only reachable for modes TOP and FULL, as further guided by
s->zero_target. However, upcoming patches want to further optimize
the pre-zeroing pass of a sync=full mirror in mirror_dirty_init(),
while avoiding that pass on a sync=top action. Instead of throwing
away context by collapsing these two values into
s->is_none_mode=false, it is better to pass s->sync_mode throughout
the entire operation. For active commit, the desired semantics match
sync mode TOP.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20250509204341.3553601-22-eblake@redhat.com>
Reviewed-by: Sunny Zhu <sunnyzhyy@qq.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit 9474d97bd7421b4fe7c806ab0949697514d11e88)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 24 ++++++++++++------------
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 22f8bd98c4..c8bbaa0b35 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -51,7 +51,7 @@ typedef struct MirrorBlockJob {
BlockDriverState *to_replace;
/* Used to block operations on the drive-mirror-replace target */
Error *replace_blocker;
- bool is_none_mode;
+ MirrorSyncMode sync_mode;
BlockMirrorBackingMode backing_mode;
/* Whether the target image requires explicit zero-initialization */
bool zero_target;
@@ -723,9 +723,10 @@ static int mirror_exit_common(Job *job)
&error_abort);
if (!abort && s->backing_mode == MIRROR_SOURCE_BACKING_CHAIN) {
- BlockDriverState *backing = s->is_none_mode ? src : s->base;
+ BlockDriverState *backing;
BlockDriverState *unfiltered_target = bdrv_skip_filters(target_bs);
+ backing = s->sync_mode == MIRROR_SYNC_MODE_NONE ? src : s->base;
if (bdrv_cow_bs(unfiltered_target) != backing) {
bdrv_set_backing_hd(unfiltered_target, backing, &local_err);
if (local_err) {
@@ -1020,7 +1021,7 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
mirror_free_init(s);
s->last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
- if (!s->is_none_mode) {
+ if (s->sync_mode != MIRROR_SYNC_MODE_NONE) {
ret = mirror_dirty_init(s);
if (ret < 0 || job_is_cancelled(&s->common.job)) {
goto immediate_exit;
@@ -1711,6 +1712,7 @@ static BlockJob *mirror_start_job(
int creation_flags, BlockDriverState *target,
const char *replaces, int64_t speed,
uint32_t granularity, int64_t buf_size,
+ MirrorSyncMode sync_mode,
BlockMirrorBackingMode backing_mode,
bool zero_target,
BlockdevOnError on_source_error,
@@ -1719,7 +1721,7 @@ static BlockJob *mirror_start_job(
BlockCompletionFunc *cb,
void *opaque,
const BlockJobDriver *driver,
- bool is_none_mode, BlockDriverState *base,
+ BlockDriverState *base,
bool auto_complete, const char *filter_node_name,
bool is_mirror, MirrorCopyMode copy_mode,
bool base_ro,
@@ -1878,7 +1880,7 @@ static BlockJob *mirror_start_job(
s->replaces = g_strdup(replaces);
s->on_source_error = on_source_error;
s->on_target_error = on_target_error;
- s->is_none_mode = is_none_mode;
+ s->sync_mode = sync_mode;
s->backing_mode = backing_mode;
s->zero_target = zero_target;
qatomic_set(&s->copy_mode, copy_mode);
@@ -2015,7 +2017,6 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
bool unmap, const char *filter_node_name,
MirrorCopyMode copy_mode, Error **errp)
{
- bool is_none_mode;
BlockDriverState *base;
GLOBAL_STATE_CODE();
@@ -2028,14 +2029,13 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
}
bdrv_graph_rdlock_main_loop();
- is_none_mode = mode == MIRROR_SYNC_MODE_NONE;
base = mode == MIRROR_SYNC_MODE_TOP ? bdrv_backing_chain_next(bs) : NULL;
bdrv_graph_rdunlock_main_loop();
mirror_start_job(job_id, bs, creation_flags, target, replaces,
- speed, granularity, buf_size, backing_mode, zero_target,
- on_source_error, on_target_error, unmap, NULL, NULL,
- &mirror_job_driver, is_none_mode, base, false,
+ speed, granularity, buf_size, mode, backing_mode,
+ zero_target, on_source_error, on_target_error, unmap,
+ NULL, NULL, &mirror_job_driver, base, false,
filter_node_name, true, copy_mode, false, errp);
}
@@ -2061,9 +2061,9 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
job = mirror_start_job(
job_id, bs, creation_flags, base, NULL, speed, 0, 0,
- MIRROR_LEAVE_BACKING_CHAIN, false,
+ MIRROR_SYNC_MODE_TOP, MIRROR_LEAVE_BACKING_CHAIN, false,
on_error, on_error, true, cb, opaque,
- &commit_active_job_driver, false, base, auto_complete,
+ &commit_active_job_driver, base, auto_complete,
filter_node_name, false, MIRROR_COPY_MODE_BACKGROUND,
base_read_only, errp);
if (!job) {
--
2.48.1

58
kvm-mirror-Reduce-I-O-when-destination-is-detect-zeroes-.patch Normal file
View File

@ -0,0 +1,58 @@
From be6ce2c91fe949d1c264de974ab4f6c4efc6976e Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Tue, 13 May 2025 17:00:45 -0500
Subject: [PATCH 16/16] mirror: Reduce I/O when destination is
detect-zeroes:unmap
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [14/14] 66f3de2ba9f977c9bc1c54f67d76b366df132e62 (ebblake/centos-qemu-kvm)
If we are going to punch holes in the mirror destination even for the
portions where the source image is unallocated, it is nicer to treat
the entire image as dirty and punch as we go, rather than pre-zeroing
the entire image just to re-do I/O to the allocated portions of the
image.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20250513220142.535200-2-eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit 9abfc81246c9cc1845080eec5920779961187c07)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 7f3b5477ce..87c19ddf0d 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -920,11 +920,16 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
* zeroing happened externally (ret > 0) or if we have a fast
* way to pre-zero the image (the dirty bitmap will be
* populated later by the non-zero portions, the same as for
- * TOP mode). If pre-zeroing is not fast, then our only
- * recourse is to mark the entire image dirty. The act of
- * pre-zeroing will populate the zero bitmap.
+ * TOP mode). If pre-zeroing is not fast, or we need to visit
+ * the entire image in order to punch holes even in the
+ * non-allocated regions of the source, then just mark the
+ * entire image dirty and leave the zero bitmap clear at this
+ * point in time. Otherwise, it can be faster to pre-zero the
+ * image now, even if we re-write the allocated portions of
+ * the disk later, and the pre-zero pass will populate the
+ * zero bitmap.
*/
- if (!bdrv_can_write_zeroes_with_unmap(target_bs)) {
+ if (!bdrv_can_write_zeroes_with_unmap(target_bs) || punch_holes) {
bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, s->bdev_length);
return 0;
}
--
2.48.1

180
kvm-mirror-Skip-pre-zeroing-destination-if-it-is-already.patch Normal file
View File

@ -0,0 +1,180 @@
From 423ce7727eecae647330287e1264ac0d938fa7f9 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:27 -0500
Subject: [PATCH 12/16] mirror: Skip pre-zeroing destination if it is already
zero
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [10/14] e754ae559123099f4aed322f6a4287cf3323f54d (ebblake/centos-qemu-kvm)
When doing a sync=full mirroring, we can skip pre-zeroing the
destination if it already reads as zeroes and we are not also trying
to punch holes due to detect-zeroes. With this patch, there are fewer
scenarios that have to pass in an explicit target-is-zero, while still
resulting in a sparse destination remaining sparse.
A later patch will then further improve things to skip writing to the
destination for parts of the image where the source is zero; but even
with just this patch, it is possible to see a difference for any
source that does not report itself as fully allocated, coupled with a
destination BDS that can quickly report that it already reads as zero.
(For a source that reports as fully allocated, such as a file, the
rest of mirror_dirty_init() still sets the entire dirty bitmap to
true, so even though we avoided the pre-zeroing, we are not yet
avoiding all redundant I/O).
Iotest 194 detects the difference made by this patch: for a file
source (where block status reports the entire image as allocated, and
therefore we end up writing zeroes everywhere in the destination
anyways), the job length remains the same. But for a qcow2 source and
a destination that reads as all zeroes, the dirty bitmap changes to
just tracking the allocated portions of the source, which results in
faster completion and smaller job statistics. For the test to pass
with both ./check -file and -qcow2, a new python filter is needed to
mask out the now-varying job amounts (this matches the shell filters
_filter_block_job_{offset,len} in common.filter). A later test will
also be added which further validates expected sparseness, so it does
not matter that 194 is no longer explicitly looking at how many bytes
were copied.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20250509204341.3553601-25-eblake@redhat.com>
Reviewed-by: Sunny Zhu <sunnyzhyy@qq.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit 181a63667adf16c35b57e446def3e41c70f1fea6)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 24 ++++++++++++++++--------
tests/qemu-iotests/194 | 6 ++++--
tests/qemu-iotests/194.out | 4 ++--
tests/qemu-iotests/iotests.py | 12 +++++++++++-
4 files changed, 33 insertions(+), 13 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index b35d12adaa..29cac1777c 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -848,23 +848,31 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
target_bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP &&
bdrv_can_write_zeroes_with_unmap(target_bs);
+ /* Determine if the image is already zero, regardless of sync mode. */
bdrv_graph_co_rdlock();
bs = s->mirror_top_bs->backing->bs;
+ if (s->target_is_zero) {
+ ret = 1;
+ } else {
+ ret = bdrv_co_is_all_zeroes(target_bs);
+ }
bdrv_graph_co_rdunlock();
- if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
+ /* Determine if a pre-zeroing pass is necessary. */
+ if (ret < 0) {
+ return ret;
+ } else if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
/* In TOP mode, there is no benefit to a pre-zeroing pass. */
- } else if (!s->target_is_zero || punch_holes) {
+ } else if (ret == 0 || punch_holes) {
/*
* Here, we are in FULL mode; our goal is to avoid writing
* zeroes if the destination already reads as zero, except
* when we are trying to punch holes. This is possible if
- * zeroing happened externally (s->target_is_zero) or if we
- * have a fast way to pre-zero the image (the dirty bitmap
- * will be populated later by the non-zero portions, the same
- * as for TOP mode). If pre-zeroing is not fast, or we need
- * to punch holes, then our only recourse is to write the
- * entire image.
+ * zeroing happened externally (ret > 0) or if we have a fast
+ * way to pre-zero the image (the dirty bitmap will be
+ * populated later by the non-zero portions, the same as for
+ * TOP mode). If pre-zeroing is not fast, or we need to punch
+ * holes, then our only recourse is to write the entire image.
*/
if (!bdrv_can_write_zeroes_with_unmap(target_bs)) {
bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, s->bdev_length);
diff --git a/tests/qemu-iotests/194 b/tests/qemu-iotests/194
index d0b9c084f5..e114c0b269 100755
--- a/tests/qemu-iotests/194
+++ b/tests/qemu-iotests/194
@@ -62,7 +62,8 @@ with iotests.FilePath('source.img') as source_img_path, \
iotests.log('Waiting for `drive-mirror` to complete...')
iotests.log(source_vm.event_wait('BLOCK_JOB_READY'),
- filters=[iotests.filter_qmp_event])
+ filters=[iotests.filter_qmp_event,
+ iotests.filter_block_job])
iotests.log('Starting migration...')
capabilities = [{'capability': 'events', 'state': True},
@@ -88,7 +89,8 @@ with iotests.FilePath('source.img') as source_img_path, \
while True:
event2 = source_vm.event_wait('BLOCK_JOB_COMPLETED')
- iotests.log(event2, filters=[iotests.filter_qmp_event])
+ iotests.log(event2, filters=[iotests.filter_qmp_event,
+ iotests.filter_block_job])
if event2['event'] == 'BLOCK_JOB_COMPLETED':
iotests.log('Stopping the NBD server on destination...')
iotests.log(dest_vm.qmp('nbd-server-stop'))
diff --git a/tests/qemu-iotests/194.out b/tests/qemu-iotests/194.out
index 376ed1d2e6..84e0fc34be 100644
--- a/tests/qemu-iotests/194.out
+++ b/tests/qemu-iotests/194.out
@@ -7,7 +7,7 @@ Launching NBD server on destination...
Starting `drive-mirror` on source...
{"return": {}}
Waiting for `drive-mirror` to complete...
-{"data": {"device": "mirror-job0", "len": 1073741824, "offset": 1073741824, "speed": 0, "type": "mirror"}, "event": "BLOCK_JOB_READY", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
+{"data": {"device": "mirror-job0", "len": "LEN", "offset": "OFFSET", "speed": 0, "type": "mirror"}, "event": "BLOCK_JOB_READY", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
Starting migration...
{"return": {}}
{"execute": "migrate-start-postcopy", "arguments": {}}
@@ -17,7 +17,7 @@ Starting migration...
{"data": {"status": "completed"}, "event": "MIGRATION", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
Gracefully ending the `drive-mirror` job on source...
{"return": {}}
-{"data": {"device": "mirror-job0", "len": 1073741824, "offset": 1073741824, "speed": 0, "type": "mirror"}, "event": "BLOCK_JOB_COMPLETED", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
+{"data": {"device": "mirror-job0", "len": "LEN", "offset": "OFFSET", "speed": 0, "type": "mirror"}, "event": "BLOCK_JOB_COMPLETED", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
Stopping the NBD server on destination...
{"return": {}}
Wait for migration completion on target...
diff --git a/tests/qemu-iotests/iotests.py b/tests/qemu-iotests/iotests.py
index c8cb028c2d..978bef1499 100644
--- a/tests/qemu-iotests/iotests.py
+++ b/tests/qemu-iotests/iotests.py
@@ -601,13 +601,23 @@ def filter_chown(msg):
return chown_re.sub("chown UID:GID", msg)
def filter_qmp_event(event):
- '''Filter a QMP event dict'''
+ '''Filter the timestamp of a QMP event dict'''
event = dict(event)
if 'timestamp' in event:
event['timestamp']['seconds'] = 'SECS'
event['timestamp']['microseconds'] = 'USECS'
return event
+def filter_block_job(event):
+ '''Filter the offset and length of a QMP block job event dict'''
+ event = dict(event)
+ if 'data' in event:
+ if 'offset' in event['data']:
+ event['data']['offset'] = 'OFFSET'
+ if 'len' in event['data']:
+ event['data']['len'] = 'LEN'
+ return event
+
def filter_qmp(qmsg, filter_fn):
'''Given a string filter, filter a QMP object's values.
filter_fn takes a (key, value) pair.'''
--
2.48.1

355
kvm-mirror-Skip-writing-zeroes-when-target-is-already-ze.patch Normal file
View File

@ -0,0 +1,355 @@
From 8a2e660ff3ec7f7506fbd4197d4dc8f53db7859a Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:28 -0500
Subject: [PATCH 13/16] mirror: Skip writing zeroes when target is already zero
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [11/14] f6bb5e0cecee07af0389aa18c3bddb47d6c5cf54 (ebblake/centos-qemu-kvm)
When mirroring, the goal is to ensure that the destination reads the
same as the source; this goal is met whether the destination is sparse
or fully-allocated (except when explicitly punching holes, then merely
reading zero is not enough to know if it is sparse, so we still want
to punch the hole). Avoiding a redundant write to zero (whether in
the background because the zero cluster was marked in the dirty
bitmap, or in the foreground because the guest is writing zeroes) when
the destination already reads as zero makes mirroring faster, and
avoids allocating the destination merely because the source reports as
allocated.
The effect is especially pronounced when the source is a raw file.
That's because when the source is a qcow2 file, the dirty bitmap only
visits the portions of the source that are allocated, which tend to be
non-zero. But when the source is a raw file,
bdrv_co_is_allocated_above() reports the entire file as allocated so
mirror_dirty_init sets the entire dirty bitmap, and it is only later
during mirror_iteration that we change to consulting the more precise
bdrv_co_block_status_above() to learn where the source reads as zero.
Remember that since a mirror operation can write a cluster more than
once (every time the guest changes the source, the destination is also
changed to keep up), and the guest can change whether a given cluster
reads as zero, is discarded, or has non-zero data over the course of
the mirror operation, we can't take the shortcut of relying on
s->target_is_zero (which is static for the life of the job) in
mirror_co_zero() to see if the destination is already zero, because
that information may be stale. Any solution we use must be dynamic in
the face of the guest writing or discarding a cluster while the mirror
has been ongoing.
We could just teach mirror_co_zero() to do a block_status() probe of
the destination, and skip the zeroes if the destination already reads
as zero, but we know from past experience that extra block_status()
calls are not always cheap (tmpfs, anyone?), especially when they are
random access rather than linear. Use of block_status() of the source
by the background task in a linear fashion is not our bottleneck (it's
a background task, after all); but since mirroring can be done while
the source is actively being changed, we don't want a slow
block_status() of the destination to occur on the hot path of the
guest trying to do random-access writes to the source.
So this patch takes a slightly different approach: any time we have to
track dirty clusters, we can also track which clusters are known to
read as zero. For sync=TOP or when we are punching holes from
"detect-zeroes":"unmap", the zero bitmap starts out empty, but
prevents a second write zero to a cluster that was already zero by an
earlier pass; for sync=FULL when we are not punching holes, the zero
bitmap starts out full if the destination reads as zero during
initialization. Either way, I/O to the destination can now avoid
redundant write zero to a cluster that already reads as zero, all
without having to do a block_status() per write on the destination.
With this patch, if I create a raw sparse destination file, connect it
with QMP 'blockdev-add' while leaving it at the default "discard":
"ignore", then run QMP 'blockdev-mirror' with "sync": "full", the
destination remains sparse rather than fully allocated. Meanwhile, a
destination image that is already fully allocated remains so unless it
was opened with "detect-zeroes": "unmap". And any time writing zeroes
is skipped, the job counters are not incremented.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20250509204341.3553601-26-eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit 7e277545b90874171128804e256a538fb0e8dd7e)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
block/mirror.c | 107 ++++++++++++++++++++++++++++++++++++++++++-------
1 file changed, 93 insertions(+), 14 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
index 29cac1777c..7f3b5477ce 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -73,6 +73,7 @@ typedef struct MirrorBlockJob {
size_t buf_size;
int64_t bdev_length;
unsigned long *cow_bitmap;
+ unsigned long *zero_bitmap;
BdrvDirtyBitmap *dirty_bitmap;
BdrvDirtyBitmapIter *dbi;
uint8_t *buf;
@@ -108,9 +109,12 @@ struct MirrorOp {
int64_t offset;
uint64_t bytes;
- /* The pointee is set by mirror_co_read(), mirror_co_zero(), and
- * mirror_co_discard() before yielding for the first time */
+ /*
+ * These pointers are set by mirror_co_read(), mirror_co_zero(), and
+ * mirror_co_discard() before yielding for the first time
+ */
int64_t *bytes_handled;
+ bool *io_skipped;
bool is_pseudo_op;
bool is_active_write;
@@ -408,15 +412,34 @@ static void coroutine_fn mirror_co_read(void *opaque)
static void coroutine_fn mirror_co_zero(void *opaque)
{
MirrorOp *op = opaque;
- int ret;
+ bool write_needed = true;
+ int ret = 0;
op->s->in_flight++;
op->s->bytes_in_flight += op->bytes;
*op->bytes_handled = op->bytes;
op->is_in_flight = true;
- ret = blk_co_pwrite_zeroes(op->s->target, op->offset, op->bytes,
- op->s->unmap ? BDRV_REQ_MAY_UNMAP : 0);
+ if (op->s->zero_bitmap) {
+ unsigned long end = DIV_ROUND_UP(op->offset + op->bytes,
+ op->s->granularity);
+ assert(QEMU_IS_ALIGNED(op->offset, op->s->granularity));
+ assert(QEMU_IS_ALIGNED(op->bytes, op->s->granularity) ||
+ op->offset + op->bytes == op->s->bdev_length);
+ if (find_next_zero_bit(op->s->zero_bitmap, end,
+ op->offset / op->s->granularity) == end) {
+ write_needed = false;
+ *op->io_skipped = true;
+ }
+ }
+ if (write_needed) {
+ ret = blk_co_pwrite_zeroes(op->s->target, op->offset, op->bytes,
+ op->s->unmap ? BDRV_REQ_MAY_UNMAP : 0);
+ }
+ if (ret >= 0 && op->s->zero_bitmap) {
+ bitmap_set(op->s->zero_bitmap, op->offset / op->s->granularity,
+ DIV_ROUND_UP(op->bytes, op->s->granularity));
+ }
mirror_write_complete(op, ret);
}
@@ -435,29 +458,43 @@ static void coroutine_fn mirror_co_discard(void *opaque)
}
static unsigned mirror_perform(MirrorBlockJob *s, int64_t offset,
- unsigned bytes, MirrorMethod mirror_method)
+ unsigned bytes, MirrorMethod mirror_method,
+ bool *io_skipped)
{
MirrorOp *op;
Coroutine *co;
int64_t bytes_handled = -1;
+ assert(QEMU_IS_ALIGNED(offset, s->granularity));
+ assert(QEMU_IS_ALIGNED(bytes, s->granularity) ||
+ offset + bytes == s->bdev_length);
op = g_new(MirrorOp, 1);
*op = (MirrorOp){
.s = s,
.offset = offset,
.bytes = bytes,
.bytes_handled = &bytes_handled,
+ .io_skipped = io_skipped,
};
qemu_co_queue_init(&op->waiting_requests);
switch (mirror_method) {
case MIRROR_METHOD_COPY:
+ if (s->zero_bitmap) {
+ bitmap_clear(s->zero_bitmap, offset / s->granularity,
+ DIV_ROUND_UP(bytes, s->granularity));
+ }
co = qemu_coroutine_create(mirror_co_read, op);
break;
case MIRROR_METHOD_ZERO:
+ /* s->zero_bitmap handled in mirror_co_zero */
co = qemu_coroutine_create(mirror_co_zero, op);
break;
case MIRROR_METHOD_DISCARD:
+ if (s->zero_bitmap) {
+ bitmap_clear(s->zero_bitmap, offset / s->granularity,
+ DIV_ROUND_UP(bytes, s->granularity));
+ }
co = qemu_coroutine_create(mirror_co_discard, op);
break;
default:
@@ -568,6 +605,7 @@ static void coroutine_fn GRAPH_UNLOCKED mirror_iteration(MirrorBlockJob *s)
int ret;
int64_t io_bytes;
int64_t io_bytes_acct;
+ bool io_skipped = false;
MirrorMethod mirror_method = MIRROR_METHOD_COPY;
assert(!(offset % s->granularity));
@@ -611,8 +649,10 @@ static void coroutine_fn GRAPH_UNLOCKED mirror_iteration(MirrorBlockJob *s)
}
io_bytes = mirror_clip_bytes(s, offset, io_bytes);
- io_bytes = mirror_perform(s, offset, io_bytes, mirror_method);
- if (mirror_method != MIRROR_METHOD_COPY && write_zeroes_ok) {
+ io_bytes = mirror_perform(s, offset, io_bytes, mirror_method,
+ &io_skipped);
+ if (io_skipped ||
+ (mirror_method != MIRROR_METHOD_COPY && write_zeroes_ok)) {
io_bytes_acct = 0;
} else {
io_bytes_acct = io_bytes;
@@ -847,8 +887,10 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
bool punch_holes =
target_bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP &&
bdrv_can_write_zeroes_with_unmap(target_bs);
+ int64_t bitmap_length = DIV_ROUND_UP(s->bdev_length, s->granularity);
/* Determine if the image is already zero, regardless of sync mode. */
+ s->zero_bitmap = bitmap_new(bitmap_length);
bdrv_graph_co_rdlock();
bs = s->mirror_top_bs->backing->bs;
if (s->target_is_zero) {
@@ -862,7 +904,14 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
if (ret < 0) {
return ret;
} else if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
- /* In TOP mode, there is no benefit to a pre-zeroing pass. */
+ /*
+ * In TOP mode, there is no benefit to a pre-zeroing pass, but
+ * the zero bitmap can be set if the destination already reads
+ * as zero and we are not punching holes.
+ */
+ if (ret > 0 && !punch_holes) {
+ bitmap_set(s->zero_bitmap, 0, bitmap_length);
+ }
} else if (ret == 0 || punch_holes) {
/*
* Here, we are in FULL mode; our goal is to avoid writing
@@ -871,8 +920,9 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
* zeroing happened externally (ret > 0) or if we have a fast
* way to pre-zero the image (the dirty bitmap will be
* populated later by the non-zero portions, the same as for
- * TOP mode). If pre-zeroing is not fast, or we need to punch
- * holes, then our only recourse is to write the entire image.
+ * TOP mode). If pre-zeroing is not fast, then our only
+ * recourse is to mark the entire image dirty. The act of
+ * pre-zeroing will populate the zero bitmap.
*/
if (!bdrv_can_write_zeroes_with_unmap(target_bs)) {
bdrv_set_dirty_bitmap(s->dirty_bitmap, 0, s->bdev_length);
@@ -883,6 +933,7 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
for (offset = 0; offset < s->bdev_length; ) {
int bytes = MIN(s->bdev_length - offset,
QEMU_ALIGN_DOWN(INT_MAX, s->granularity));
+ bool ignored;
mirror_throttle(s);
@@ -898,12 +949,15 @@ static int coroutine_fn GRAPH_UNLOCKED mirror_dirty_init(MirrorBlockJob *s)
continue;
}
- mirror_perform(s, offset, bytes, MIRROR_METHOD_ZERO);
+ mirror_perform(s, offset, bytes, MIRROR_METHOD_ZERO, &ignored);
offset += bytes;
}
mirror_wait_for_all_io(s);
s->initial_zeroing_ongoing = false;
+ } else {
+ /* In FULL mode, and image already reads as zero. */
+ bitmap_set(s->zero_bitmap, 0, bitmap_length);
}
/* First part, loop on the sectors and initialize the dirty bitmap. */
@@ -1188,6 +1242,7 @@ immediate_exit:
assert(s->in_flight == 0);
qemu_vfree(s->buf);
g_free(s->cow_bitmap);
+ g_free(s->zero_bitmap);
g_free(s->in_flight_bitmap);
bdrv_dirty_iter_free(s->dbi);
@@ -1367,6 +1422,7 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
int ret;
size_t qiov_offset = 0;
int64_t dirty_bitmap_offset, dirty_bitmap_end;
+ int64_t zero_bitmap_offset, zero_bitmap_end;
if (!QEMU_IS_ALIGNED(offset, job->granularity) &&
bdrv_dirty_bitmap_get(job->dirty_bitmap, offset))
@@ -1410,8 +1466,9 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
}
/*
- * Tails are either clean or shrunk, so for bitmap resetting
- * we safely align the range down.
+ * Tails are either clean or shrunk, so for dirty bitmap resetting
+ * we safely align the range narrower. But for zero bitmap, round
+ * range wider for checking or clearing, and narrower for setting.
*/
dirty_bitmap_offset = QEMU_ALIGN_UP(offset, job->granularity);
dirty_bitmap_end = QEMU_ALIGN_DOWN(offset + bytes, job->granularity);
@@ -1419,22 +1476,44 @@ do_sync_target_write(MirrorBlockJob *job, MirrorMethod method,
bdrv_reset_dirty_bitmap(job->dirty_bitmap, dirty_bitmap_offset,
dirty_bitmap_end - dirty_bitmap_offset);
}
+ zero_bitmap_offset = offset / job->granularity;
+ zero_bitmap_end = DIV_ROUND_UP(offset + bytes, job->granularity);
job_progress_increase_remaining(&job->common.job, bytes);
job->active_write_bytes_in_flight += bytes;
switch (method) {
case MIRROR_METHOD_COPY:
+ if (job->zero_bitmap) {
+ bitmap_clear(job->zero_bitmap, zero_bitmap_offset,
+ zero_bitmap_end - zero_bitmap_offset);
+ }
ret = blk_co_pwritev_part(job->target, offset, bytes,
qiov, qiov_offset, flags);
break;
case MIRROR_METHOD_ZERO:
+ if (job->zero_bitmap) {
+ if (find_next_zero_bit(job->zero_bitmap, zero_bitmap_end,
+ zero_bitmap_offset) == zero_bitmap_end) {
+ ret = 0;
+ break;
+ }
+ }
assert(!qiov);
ret = blk_co_pwrite_zeroes(job->target, offset, bytes, flags);
+ if (job->zero_bitmap && ret >= 0) {
+ bitmap_set(job->zero_bitmap, dirty_bitmap_offset / job->granularity,
+ (dirty_bitmap_end - dirty_bitmap_offset) /
+ job->granularity);
+ }
break;
case MIRROR_METHOD_DISCARD:
+ if (job->zero_bitmap) {
+ bitmap_clear(job->zero_bitmap, zero_bitmap_offset,
+ zero_bitmap_end - zero_bitmap_offset);
+ }
assert(!qiov);
ret = blk_co_pdiscard(job->target, offset, bytes);
break;
--
2.48.1

256
kvm-s390x-pci-add-support-for-guests-that-request-direct.patch Normal file
View File

@ -0,0 +1,256 @@
From c60d0770ff3f9124e6e9d7beb03e1ef8067e8e26 Mon Sep 17 00:00:00 2001
From: Christoph Schlameuss <cschlame@redhat.com>
Date: Thu, 12 Jun 2025 13:25:32 +0200
Subject: [PATCH 01/16] s390x/pci: add support for guests that request direct
mapping
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Christoph Schlameuss <None>
RH-MergeRequest: 376: Draft: KVM: Performance Enhanced Refresh PCI Translation
RH-Jira: RHEL-11430
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cédric Le Goater <clg@redhat.com>
RH-Commit: [1/2] 11d1dd9a5add55ae43d5d922588a33945ecbfe27 (cschlame/qemu-kvm)
JIRA: https://issues.redhat.com/browse/RHEL-11430
Conflicts: hw/s390x/s390-pci-bus.c old s390_pci_device_properties[] still has DEFINE_PROP_END_OF_LIST()
hw/s390x/s390-pci-inst.c hw_accel.h is still in sysemu
hw/s390x/s390-virtio-ccw.c changes from ccw_machine_9_2_class_options() moved to ccw_rhel_machine_9_6_0_class_options()
commit dfcee1ea4c52ac60e0a06221eafb7b6253eb10c3
Author: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Wed Feb 26 16:00:12 2025 -0500
s390x/pci: add support for guests that request direct mapping
When receiving a guest mpcifc(4) or mpcifc(6) instruction without the T
bit set, treat this as a request to perform direct mapping instead of
address translation. In order to facilitate this, pin the entirety of
guest memory into the host iommu.
Pinning for the direct mapping case is handled via vfio and its memory
listener. Additionally, ram discard settings are inherited from vfio:
coordinated discards (e.g. virtio-mem) are allowed while uncoordinated
discards (e.g. virtio-balloon) are disabled.
Subsequent guest DMA operations are all expected to be of the format
guest_phys+sdma, allowing them to be used as lookup into the host
iommu table.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-ID: <20250226210013.238349-2-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Christoph Schlameuss <cschlame@redhat.com>
---
hw/s390x/s390-pci-bus.c | 39 +++++++++++++++++++++++++++++++--
hw/s390x/s390-pci-inst.c | 13 +++++++++--
hw/s390x/s390-pci-vfio.c | 23 +++++++++++++++----
hw/s390x/s390-virtio-ccw.c | 5 +++++
include/hw/s390x/s390-pci-bus.h | 3 +++
5 files changed, 75 insertions(+), 8 deletions(-)
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index 3e57d5faca..13bc02d837 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -18,6 +18,8 @@
#include "hw/s390x/s390-pci-inst.h"
#include "hw/s390x/s390-pci-kvm.h"
#include "hw/s390x/s390-pci-vfio.h"
+#include "hw/s390x/s390-virtio-ccw.h"
+#include "hw/boards.h"
#include "hw/pci/pci_bus.h"
#include "hw/qdev-properties.h"
#include "hw/pci/pci_bridge.h"
@@ -724,12 +726,42 @@ void s390_pci_iommu_enable(S390PCIIOMMU *iommu)
g_free(name);
}
+void s390_pci_iommu_direct_map_enable(S390PCIIOMMU *iommu)
+{
+ MachineState *ms = MACHINE(qdev_get_machine());
+ S390CcwMachineState *s390ms = S390_CCW_MACHINE(ms);
+
+ /*
+ * For direct-mapping we must map the entire guest address space. Rather
+ * than using an iommu, create a memory region alias that maps GPA X to
+ * IOVA X + SDMA. VFIO will handle pinning via its memory listener.
+ */
+ g_autofree char *name = g_strdup_printf("iommu-dm-s390-%04x",
+ iommu->pbdev->uid);
+
+ iommu->dm_mr = g_malloc0(sizeof(*iommu->dm_mr));
+ memory_region_init_alias(iommu->dm_mr, OBJECT(&iommu->mr), name,
+ get_system_memory(), 0,
+ s390_get_memory_limit(s390ms));
+ iommu->enabled = true;
+ memory_region_add_subregion(&iommu->mr, iommu->pbdev->zpci_fn.sdma,
+ iommu->dm_mr);
+}
+
void s390_pci_iommu_disable(S390PCIIOMMU *iommu)
{
iommu->enabled = false;
g_hash_table_remove_all(iommu->iotlb);
- memory_region_del_subregion(&iommu->mr, MEMORY_REGION(&iommu->iommu_mr));
- object_unparent(OBJECT(&iommu->iommu_mr));
+ if (iommu->dm_mr) {
+ memory_region_del_subregion(&iommu->mr, iommu->dm_mr);
+ object_unparent(OBJECT(iommu->dm_mr));
+ g_free(iommu->dm_mr);
+ iommu->dm_mr = NULL;
+ } else {
+ memory_region_del_subregion(&iommu->mr,
+ MEMORY_REGION(&iommu->iommu_mr));
+ object_unparent(OBJECT(&iommu->iommu_mr));
+ }
}
static void s390_pci_iommu_free(S390pciState *s, PCIBus *bus, int32_t devfn)
@@ -1130,6 +1162,7 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
/* Always intercept emulated devices */
pbdev->interp = false;
pbdev->forwarding_assist = false;
+ pbdev->rtr_avail = false;
}
if (s390_pci_msix_init(pbdev) && !pbdev->interp) {
@@ -1488,6 +1521,8 @@ static Property s390_pci_device_properties[] = {
DEFINE_PROP_BOOL("interpret", S390PCIBusDevice, interp, true),
DEFINE_PROP_BOOL("forwarding-assist", S390PCIBusDevice, forwarding_assist,
true),
+ DEFINE_PROP_BOOL("relaxed-translation", S390PCIBusDevice, rtr_avail,
+ true),
DEFINE_PROP_END_OF_LIST(),
};
diff --git a/hw/s390x/s390-pci-inst.c b/hw/s390x/s390-pci-inst.c
index 30149546c0..803ebcd9b3 100644
--- a/hw/s390x/s390-pci-inst.c
+++ b/hw/s390x/s390-pci-inst.c
@@ -16,6 +16,7 @@
#include "exec/memory.h"
#include "qemu/error-report.h"
#include "sysemu/hw_accel.h"
+#include "hw/boards.h"
#include "hw/pci/pci_device.h"
#include "hw/s390x/s390-pci-inst.h"
#include "hw/s390x/s390-pci-bus.h"
@@ -1008,17 +1009,25 @@ static int reg_ioat(CPUS390XState *env, S390PCIBusDevice *pbdev, ZpciFib fib,
}
/* currently we only support designation type 1 with translation */
- if (!(dt == ZPCI_IOTA_RTTO && t)) {
+ if (t && dt != ZPCI_IOTA_RTTO) {
error_report("unsupported ioat dt %d t %d", dt, t);
s390_program_interrupt(env, PGM_OPERAND, ra);
return -EINVAL;
+ } else if (!t && !pbdev->rtr_avail) {
+ error_report("relaxed translation not allowed");
+ s390_program_interrupt(env, PGM_OPERAND, ra);
+ return -EINVAL;
}
iommu->pba = pba;
iommu->pal = pal;
iommu->g_iota = g_iota;
- s390_pci_iommu_enable(iommu);
+ if (t) {
+ s390_pci_iommu_enable(iommu);
+ } else {
+ s390_pci_iommu_direct_map_enable(iommu);
+ }
return 0;
}
diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
index 7dbbc76823..443e222912 100644
--- a/hw/s390x/s390-pci-vfio.c
+++ b/hw/s390x/s390-pci-vfio.c
@@ -131,13 +131,28 @@ static void s390_pci_read_base(S390PCIBusDevice *pbdev,
/* Store function type separately for type-specific behavior */
pbdev->pft = cap->pft;
+ /*
+ * If the device is a passthrough ISM device, disallow relaxed
+ * translation.
+ */
+ if (pbdev->pft == ZPCI_PFT_ISM) {
+ pbdev->rtr_avail = false;
+ }
+
/*
* If appropriate, reduce the size of the supported DMA aperture reported
- * to the guest based upon the vfio DMA limit.
+ * to the guest based upon the vfio DMA limit. This is applicable for
+ * devices that are guaranteed to not use relaxed translation. If the
+ * device is capable of relaxed translation then we must advertise the
+ * full aperture. In this case, if translation is used then we will
+ * rely on the vfio DMA limit counting and use RPCIT CC1 / status 16
+ * to request that the guest free DMA mappings as necessary.
*/
- vfio_size = pbdev->iommu->max_dma_limit << TARGET_PAGE_BITS;
- if (vfio_size > 0 && vfio_size < cap->end_dma - cap->start_dma + 1) {
- pbdev->zpci_fn.edma = cap->start_dma + vfio_size - 1;
+ if (!pbdev->rtr_avail) {
+ vfio_size = pbdev->iommu->max_dma_limit << TARGET_PAGE_BITS;
+ if (vfio_size > 0 && vfio_size < cap->end_dma - cap->start_dma + 1) {
+ pbdev->zpci_fn.edma = cap->start_dma + vfio_size - 1;
+ }
}
}
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index 312e8f18aa..77a1bde71e 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -1348,8 +1348,13 @@ static void ccw_rhel_machine_9_6_0_instance_options(MachineState *machine)
static void ccw_rhel_machine_9_6_0_class_options(MachineClass *mc)
{
+ static GlobalProperty compat[] = {
+ { TYPE_S390_PCI_DEVICE, "relaxed-translation", "off", },
+ };
+
/* NB: remember to move this line to the *latest* RHEL 9 machine */
compat_props_add(mc->compat_props, hw_compat_rhel_9, hw_compat_rhel_9_len);
+ compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
}
DEFINE_CCW_MACHINE_AS_LATEST(9, 6, 0);
diff --git a/include/hw/s390x/s390-pci-bus.h b/include/hw/s390x/s390-pci-bus.h
index 2c43ea123f..04944d4fed 100644
--- a/include/hw/s390x/s390-pci-bus.h
+++ b/include/hw/s390x/s390-pci-bus.h
@@ -277,6 +277,7 @@ struct S390PCIIOMMU {
AddressSpace as;
MemoryRegion mr;
IOMMUMemoryRegion iommu_mr;
+ MemoryRegion *dm_mr;
bool enabled;
uint64_t g_iota;
uint64_t pba;
@@ -362,6 +363,7 @@ struct S390PCIBusDevice {
bool interp;
bool forwarding_assist;
bool aif;
+ bool rtr_avail;
QTAILQ_ENTRY(S390PCIBusDevice) link;
};
@@ -389,6 +391,7 @@ int pci_chsc_sei_nt2_have_event(void);
void s390_pci_sclp_configure(SCCB *sccb);
void s390_pci_sclp_deconfigure(SCCB *sccb);
void s390_pci_iommu_enable(S390PCIIOMMU *iommu);
+void s390_pci_iommu_direct_map_enable(S390PCIIOMMU *iommu);
void s390_pci_iommu_disable(S390PCIIOMMU *iommu);
void s390_pci_generate_error_event(uint16_t pec, uint32_t fh, uint32_t fid,
uint64_t faddr, uint32_t e);
--
2.48.1

72
kvm-s390x-pci-indicate-QEMU-supports-relaxed-translation.patch Normal file
View File

@ -0,0 +1,72 @@
From 13e8ddbd282da692c8199a6cb9ca847334089e29 Mon Sep 17 00:00:00 2001
From: Christoph Schlameuss <cschlame@redhat.com>
Date: Thu, 12 Jun 2025 11:48:41 +0200
Subject: [PATCH 02/16] s390x/pci: indicate QEMU supports relaxed translation
for passthrough
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Christoph Schlameuss <None>
RH-MergeRequest: 376: Draft: KVM: Performance Enhanced Refresh PCI Translation
RH-Jira: RHEL-11430
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cédric Le Goater <clg@redhat.com>
RH-Commit: [2/2] afd514268347d0b434a60d7c6c09d20b84e5d902 (cschlame/qemu-kvm)
JIRA: https://issues.redhat.com/browse/RHEL-11430
commit d9b5dfc7122559e5b5959ecf534788b90c3dd102
Author: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Wed Feb 26 16:00:13 2025 -0500
s390x/pci: indicate QEMU supports relaxed translation for passthrough
Specifying this bit in the guest CLP response indicates that the guest
can optionally choose to skip translation and instead use
identity-mapped operations.
Tested-by: Niklas Schnelle <schnelle@linux.ibm.com>
Reviewed-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Message-ID: <20250226210013.238349-3-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Christoph Schlameuss <cschlame@redhat.com>
---
hw/s390x/s390-pci-vfio.c | 5 ++++-
include/hw/s390x/s390-pci-clp.h | 1 +
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
index 443e222912..6236ac7f1e 100644
--- a/hw/s390x/s390-pci-vfio.c
+++ b/hw/s390x/s390-pci-vfio.c
@@ -238,8 +238,11 @@ static void s390_pci_read_group(S390PCIBusDevice *pbdev,
pbdev->pci_group = s390_group_create(pbdev->zpci_fn.pfgid, start_gid);
resgrp = &pbdev->pci_group->zpci_group;
+ if (pbdev->rtr_avail) {
+ resgrp->fr |= CLP_RSP_QPCIG_MASK_RTR;
+ }
if (cap->flags & VFIO_DEVICE_INFO_ZPCI_FLAG_REFRESH) {
- resgrp->fr = 1;
+ resgrp->fr |= CLP_RSP_QPCIG_MASK_REFRESH;
}
resgrp->dasm = cap->dasm;
resgrp->msia = cap->msi_addr;
diff --git a/include/hw/s390x/s390-pci-clp.h b/include/hw/s390x/s390-pci-clp.h
index 03b7f9ba5f..6a635d693b 100644
--- a/include/hw/s390x/s390-pci-clp.h
+++ b/include/hw/s390x/s390-pci-clp.h
@@ -158,6 +158,7 @@ typedef struct ClpRspQueryPciGrp {
#define CLP_RSP_QPCIG_MASK_NOI 0xfff
uint16_t i;
uint8_t version;
+#define CLP_RSP_QPCIG_MASK_RTR 0x20
#define CLP_RSP_QPCIG_MASK_FRAME 0x2
#define CLP_RSP_QPCIG_MASK_REFRESH 0x1
uint8_t fr;
--
2.48.1

545
kvm-tests-Add-iotest-mirror-sparse-for-recent-patches.patch Normal file
View File

@ -0,0 +1,545 @@
From e72aaba2efda48e083d92e6dacfe58667bdfa958 Mon Sep 17 00:00:00 2001
From: Eric Blake <eblake@redhat.com>
Date: Fri, 9 May 2025 15:40:30 -0500
Subject: [PATCH 15/16] tests: Add iotest mirror-sparse for recent patches
RH-Author: Eric Blake <eblake@redhat.com>
RH-MergeRequest: 365: blockdev-mirror: More efficient handling of sparse mirrors
RH-Jira: RHEL-82906 RHEL-83015
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [13/14] 6b7792e85b81e45d11c2664349db75d905e72adf (ebblake/centos-qemu-kvm)
Prove that blockdev-mirror can now result in sparse raw destination
files, regardless of whether the source is raw or qcow2. By making
this a separate test, it was possible to test effects of individual
patches for the various pieces that all have to work together for a
sparse mirror to be successful.
Note that ./check -file produces different job lengths than ./check
-qcow2 (the test uses a filter to normalize); that's because when
deciding how much of the image to be mirrored, the code looks at how
much of the source image was allocated (for qcow2, this is only the
written clusters; for raw, it is the entire file). But the important
part is that the destination file ends up smaller than 3M, rather than
the 20M it used to be before this patch series.
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-ID: <20250509204341.3553601-28-eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit c0ddcb2cbc146e64f666eaae4edc7b5db7e5814d)
Jira: https://issues.redhat.com/browse/RHEL-82906
Jira: https://issues.redhat.com/browse/RHEL-83015
Signed-off-by: Eric Blake <eblake@redhat.com>
---
tests/qemu-iotests/tests/mirror-sparse | 125 +++++++
tests/qemu-iotests/tests/mirror-sparse.out | 365 +++++++++++++++++++++
2 files changed, 490 insertions(+)
create mode 100755 tests/qemu-iotests/tests/mirror-sparse
create mode 100644 tests/qemu-iotests/tests/mirror-sparse.out
diff --git a/tests/qemu-iotests/tests/mirror-sparse b/tests/qemu-iotests/tests/mirror-sparse
new file mode 100755
index 0000000000..8c52a4e244
--- /dev/null
+++ b/tests/qemu-iotests/tests/mirror-sparse
@@ -0,0 +1,125 @@
+#!/usr/bin/env bash
+# group: rw auto quick
+#
+# Test blockdev-mirror with raw sparse destination
+#
+# Copyright (C) 2025 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+#
+
+seq="$(basename $0)"
+echo "QA output created by $seq"
+
+status=1 # failure is the default!
+
+_cleanup()
+{
+ _cleanup_test_img
+ _cleanup_qemu
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+cd ..
+. ./common.rc
+. ./common.filter
+. ./common.qemu
+
+_supported_fmt qcow2 raw # Format of the source. dst is always raw file
+_supported_proto file
+_supported_os Linux
+
+echo
+echo "=== Initial image setup ==="
+echo
+
+TEST_IMG="$TEST_IMG.base" _make_test_img 20M
+$QEMU_IO -c 'w 8M 2M' -f $IMGFMT "$TEST_IMG.base" | _filter_qemu_io
+
+_launch_qemu \
+ -blockdev '{"driver":"file", "cache":{"direct":true, "no-flush":false},
+ "filename":"'"$TEST_IMG.base"'", "node-name":"src-file"}' \
+ -blockdev '{"driver":"'$IMGFMT'", "node-name":"src", "file":"src-file"}'
+h1=$QEMU_HANDLE
+_send_qemu_cmd $h1 '{"execute": "qmp_capabilities"}' 'return'
+
+# Check several combinations; most should result in a sparse destination;
+# the destination should only be fully allocated if pre-allocated
+# and not punching holes due to detect-zeroes
+# do_test creation discard zeroes result
+do_test() {
+ creation=$1
+ discard=$2
+ zeroes=$3
+ expected=$4
+
+echo
+echo "=== Testing creation=$creation discard=$discard zeroes=$zeroes ==="
+echo
+
+rm -f $TEST_IMG
+if test $creation = external; then
+ truncate --size=20M $TEST_IMG
+else
+ _send_qemu_cmd $h1 '{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"'$TEST_IMG'",
+ "size":'$((20*1024*1024))', "preallocation":"'$creation'"},
+ "job-id":"job1"}}' 'concluded'
+ _send_qemu_cmd $h1 '{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}' 'return'
+fi
+_send_qemu_cmd $h1 '{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"'$TEST_IMG'", "aio":"threads",
+ "auto-read-only":true, "discard":"'$discard'",
+ "detect-zeroes":"'$zeroes'"}}' 'return'
+_send_qemu_cmd $h1 '{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}' 'return'
+_timed_wait_for $h1 '"ready"'
+_send_qemu_cmd $h1 '{"execute": "job-complete", "arguments":
+ {"id":"job2"}}' 'return' \
+ | _filter_block_job_offset | _filter_block_job_len
+_send_qemu_cmd $h1 '{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}' 'return' \
+ | _filter_block_job_offset | _filter_block_job_len
+$QEMU_IMG compare -U -f $IMGFMT -F raw $TEST_IMG.base $TEST_IMG
+result=$(disk_usage $TEST_IMG)
+if test $result -lt $((3*1024*1024)); then
+ actual=sparse
+elif test $result = $((20*1024*1024)); then
+ actual=full
+else
+ actual=unknown
+fi
+echo "Destination is $actual; expected $expected"
+}
+
+do_test external ignore off sparse
+do_test external unmap off sparse
+do_test external unmap unmap sparse
+do_test off ignore off sparse
+do_test off unmap off sparse
+do_test off unmap unmap sparse
+do_test full ignore off full
+do_test full unmap off sparse
+do_test full unmap unmap sparse
+
+_send_qemu_cmd $h1 '{"execute":"quit"}' ''
+
+# success, all done
+echo '*** done'
+rm -f $seq.full
+status=0
diff --git a/tests/qemu-iotests/tests/mirror-sparse.out b/tests/qemu-iotests/tests/mirror-sparse.out
new file mode 100644
index 0000000000..2103b891c3
--- /dev/null
+++ b/tests/qemu-iotests/tests/mirror-sparse.out
@@ -0,0 +1,365 @@
+QA output created by mirror-sparse
+
+=== Initial image setup ===
+
+Formatting 'TEST_DIR/t.IMGFMT.base', fmt=IMGFMT size=20971520
+wrote 2097152/2097152 bytes at offset 8388608
+2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+{"execute": "qmp_capabilities"}
+{"return": {}}
+
+=== Testing creation=external discard=ignore zeroes=off ===
+
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"ignore",
+ "detect-zeroes":"off"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=external discard=unmap zeroes=off ===
+
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"unmap",
+ "detect-zeroes":"off"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=external discard=unmap zeroes=unmap ===
+
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"unmap",
+ "detect-zeroes":"unmap"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=off discard=ignore zeroes=off ===
+
+{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"TEST_DIR/t.IMGFMT",
+ "size":20971520, "preallocation":"off"},
+ "job-id":"job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job1"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job1"}}
+{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job1"}}
+{"return": {}}
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"ignore",
+ "detect-zeroes":"off"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=off discard=unmap zeroes=off ===
+
+{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"TEST_DIR/t.IMGFMT",
+ "size":20971520, "preallocation":"off"},
+ "job-id":"job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job1"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job1"}}
+{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job1"}}
+{"return": {}}
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"unmap",
+ "detect-zeroes":"off"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=off discard=unmap zeroes=unmap ===
+
+{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"TEST_DIR/t.IMGFMT",
+ "size":20971520, "preallocation":"off"},
+ "job-id":"job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job1"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job1"}}
+{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job1"}}
+{"return": {}}
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"unmap",
+ "detect-zeroes":"unmap"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=full discard=ignore zeroes=off ===
+
+{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"TEST_DIR/t.IMGFMT",
+ "size":20971520, "preallocation":"full"},
+ "job-id":"job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job1"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job1"}}
+{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job1"}}
+{"return": {}}
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"ignore",
+ "detect-zeroes":"off"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is full; expected full
+
+=== Testing creation=full discard=unmap zeroes=off ===
+
+{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"TEST_DIR/t.IMGFMT",
+ "size":20971520, "preallocation":"full"},
+ "job-id":"job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job1"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job1"}}
+{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job1"}}
+{"return": {}}
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"unmap",
+ "detect-zeroes":"off"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+
+=== Testing creation=full discard=unmap zeroes=unmap ===
+
+{"execute": "blockdev-create", "arguments":
+ {"options": {"driver":"file", "filename":"TEST_DIR/t.IMGFMT",
+ "size":20971520, "preallocation":"full"},
+ "job-id":"job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job1"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job1"}}
+{"execute": "job-dismiss", "arguments":
+ {"id": "job1"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job1"}}
+{"return": {}}
+{"execute": "blockdev-add", "arguments":
+ {"node-name": "dst", "driver":"file",
+ "filename":"TEST_DIR/t.IMGFMT", "aio":"threads",
+ "auto-read-only":true, "discard":"unmap",
+ "detect-zeroes":"unmap"}}
+{"return": {}}
+{"execute":"blockdev-mirror", "arguments":
+ {"sync":"full", "device":"src", "target":"dst",
+ "job-id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "created", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "running", "id": "job2"}}
+{"return": {}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "ready", "id": "job2"}}
+{"execute": "job-complete", "arguments":
+ {"id":"job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_READY", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"return": {}}
+{"execute": "blockdev-del", "arguments":
+ {"node-name": "dst"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "waiting", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "pending", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "BLOCK_JOB_COMPLETED", "data": {"device": "job2", "len": LEN, "offset": OFFSET, "speed": 0, "type": "mirror"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "concluded", "id": "job2"}}
+{"timestamp": {"seconds": TIMESTAMP, "microseconds": TIMESTAMP}, "event": "JOB_STATUS_CHANGE", "data": {"status": "null", "id": "job2"}}
+{"return": {}}
+Images are identical.
+Destination is sparse; expected sparse
+{"execute":"quit"}
+*** done
--
2.48.1

72
qemu-kvm.spec
View File

@ -149,7 +149,7 @@ Obsoletes: %{name}-block-ssh <= %{epoch}:%{version} \
Summary: QEMU is a machine emulator and virtualizer Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm Name: qemu-kvm
Version: 9.1.0 Version: 9.1.0
Release: 23%{?rcrel}%{?dist}%{?cc_suffix} Release: 24%{?rcrel}%{?dist}%{?cc_suffix}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped # Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
# Epoch 15 used for RHEL 8 # Epoch 15 used for RHEL 8
# Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5) # Epoch 17 used for RHEL 9 (due to release versioning offset in RHEL 8.5)
@ -575,6 +575,52 @@ Patch201: kvm-net-vhost-user-add-QAPI-events-to-report-connection-.patch
Patch202: kvm-file-posix-Define-DM_MPATH_PROBE_PATHS.patch Patch202: kvm-file-posix-Define-DM_MPATH_PROBE_PATHS.patch
# For RHEL-95408 - Support multipath failover with scsi-block [rhel-9] # For RHEL-95408 - Support multipath failover with scsi-block [rhel-9]
Patch203: kvm-file-posix-Probe-paths-and-retry-SG_IO-on-potential-.patch Patch203: kvm-file-posix-Probe-paths-and-retry-SG_IO-on-potential-.patch
# For RHEL-11430 - [IBM 9.7 FEAT] KVM: Performance Enhanced Refresh PCI Translation - qemu part
Patch204: kvm-s390x-pci-add-support-for-guests-that-request-direct.patch
# For RHEL-11430 - [IBM 9.7 FEAT] KVM: Performance Enhanced Refresh PCI Translation - qemu part
Patch205: kvm-s390x-pci-indicate-QEMU-supports-relaxed-translation.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch206: kvm-block-Expand-block-status-mode-from-bool-to-flags.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch207: kvm-file-posix-gluster-Handle-zero-block-status-hint-bet.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch208: kvm-block-Let-bdrv_co_is_zero_fast-consolidate-adjacent-.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch209: kvm-block-Add-new-bdrv_co_is_all_zeroes-function.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch210: kvm-iotests-Improve-iotest-194-to-mirror-data.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch211: kvm-mirror-Minor-refactoring.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch212: kvm-mirror-Pass-full-sync-mode-rather-than-bool-to-inter.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch213: kvm-mirror-Allow-QMP-override-to-declare-target-already-.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch214: kvm-mirror-Drop-redundant-zero_target-parameter.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch215: kvm-mirror-Skip-pre-zeroing-destination-if-it-is-already.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch216: kvm-mirror-Skip-writing-zeroes-when-target-is-already-ze.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch217: kvm-iotests-common.rc-add-disk_usage-function.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch218: kvm-tests-Add-iotest-mirror-sparse-for-recent-patches.patch
# For RHEL-82906 - --migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7]
# For RHEL-83015 - Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7]
Patch219: kvm-mirror-Reduce-I-O-when-destination-is-detect-zeroes-.patch
%if %{have_clang} %if %{have_clang}
BuildRequires: clang BuildRequires: clang
@ -1650,6 +1696,30 @@ useradd -r -u 107 -g qemu -G kvm -d / -s /sbin/nologin \
%endif %endif
%changelog %changelog
* Mon Jun 16 2025 Jon Maloy <jmaloy@redhat.com> - 9.1.0-24
- kvm-s390x-pci-add-support-for-guests-that-request-direct.patch [RHEL-11430]
- kvm-s390x-pci-indicate-QEMU-supports-relaxed-translation.patch [RHEL-11430]
- kvm-block-Expand-block-status-mode-from-bool-to-flags.patch [RHEL-82906 RHEL-83015]
- kvm-file-posix-gluster-Handle-zero-block-status-hint-bet.patch [RHEL-82906 RHEL-83015]
- kvm-block-Let-bdrv_co_is_zero_fast-consolidate-adjacent-.patch [RHEL-82906 RHEL-83015]
- kvm-block-Add-new-bdrv_co_is_all_zeroes-function.patch [RHEL-82906 RHEL-83015]
- kvm-iotests-Improve-iotest-194-to-mirror-data.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Minor-refactoring.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Pass-full-sync-mode-rather-than-bool-to-inter.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Allow-QMP-override-to-declare-target-already-.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Drop-redundant-zero_target-parameter.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Skip-pre-zeroing-destination-if-it-is-already.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Skip-writing-zeroes-when-target-is-already-ze.patch [RHEL-82906 RHEL-83015]
- kvm-iotests-common.rc-add-disk_usage-function.patch [RHEL-82906 RHEL-83015]
- kvm-tests-Add-iotest-mirror-sparse-for-recent-patches.patch [RHEL-82906 RHEL-83015]
- kvm-mirror-Reduce-I-O-when-destination-is-detect-zeroes-.patch [RHEL-82906 RHEL-83015]
- Resolves: RHEL-11430
([IBM 9.7 FEAT] KVM: Performance Enhanced Refresh PCI Translation - qemu part)
- Resolves: RHEL-82906
(--migrate-disks-detect-zeroes doesn't take effect for disk migration [rhel-9.7])
- Resolves: RHEL-83015
(Disk size of target raw image is full allocated when doing mirror with default discard value [rhel-9.7])
* Mon Jun 09 2025 Jon Maloy <jmaloy@redhat.com> - 9.1.0-23 * Mon Jun 09 2025 Jon Maloy <jmaloy@redhat.com> - 9.1.0-23
- kvm-net-vhost-user-add-QAPI-events-to-report-connection-.patch [RHEL-95120] - kvm-net-vhost-user-add-QAPI-events-to-report-connection-.patch [RHEL-95120]
- kvm-file-posix-Define-DM_MPATH_PROBE_PATHS.patch [RHEL-95408] - kvm-file-posix-Define-DM_MPATH_PROBE_PATHS.patch [RHEL-95408]