5 files changed, 145 insertions(+), 29 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
-index d8ecb9efa2..d95a7d7940 100644
+index 3c4ab1159d..f2eca983f1 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -51,7 +51,7 @@ typedef struct MirrorBlockJob {
errp);
if (!job) {
diff --git a/blockdev.c b/blockdev.c
-index e46e831212..fa601838a3 100644
+index 9230888e34..9a1a3118ed 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -2951,6 +2951,10 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
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 0f21b0570b..e7f901d048 100644
+index b49f4eb35b..9d744db618 100644
--- a/include/block/block_int-global-state.h
+++ b/include/block/block_int-global-state.h
-@@ -148,7 +148,9 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
+@@ -149,7 +149,9 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
BlockDriverState *target, const char *replaces,
int creation_flags, int64_t speed,
uint32_t granularity, int64_t buf_size,
BlockdevOnError on_source_error,
BlockdevOnError on_target_error,
diff --git a/qapi/block-core.json b/qapi/block-core.json
-index beeb91952a..fb25c2b245 100644
+index 2173e7734a..e1857e7094 100644
--- a/qapi/block-core.json
+++ b/qapi/block-core.json
-@@ -1993,10 +1993,19 @@
+@@ -2000,10 +2000,19 @@
# (all the disk, only the sectors allocated in the topmost image, or
# only new I/O).
#
#
# @buf-size: maximum amount of data in flight from source to
# target (since 1.4).
-@@ -2034,7 +2043,9 @@
+@@ -2043,7 +2052,9 @@
{ 'struct': 'DriveMirror',
'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
'*format': 'str', '*node-name': 'str', '*replaces': 'str',
'*speed': 'int', '*granularity': 'uint32',
'*buf-size': 'int', '*on-source-error': 'BlockdevOnError',
'*on-target-error': 'BlockdevOnError',
-@@ -2306,10 +2317,19 @@
+@@ -2322,10 +2333,19 @@
# (all the disk, only the sectors allocated in the topmost image, or
# only new I/O).
#
#
# @buf-size: maximum amount of data in flight from source to
# target
-@@ -2358,7 +2378,8 @@
+@@ -2375,7 +2395,8 @@
{ 'command': 'blockdev-mirror',
'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
'*replaces': 'str',
'*buf-size': 'int', '*on-source-error': 'BlockdevOnError',
'*on-target-error': 'BlockdevOnError',
diff --git a/tests/unit/test-block-iothread.c b/tests/unit/test-block-iothread.c
-index 94718c9319..7977ac14f4 100644
+index 8b55eccc89..f4650be8e5 100644
--- a/tests/unit/test-block-iothread.c
+++ b/tests/unit/test-block-iothread.c
-@@ -626,8 +626,8 @@ static void test_propagate_mirror(void)
+@@ -753,8 +753,8 @@ static void test_propagate_mirror(void)
/* Start a mirror job */
mirror_start("job0", src, target, NULL, JOB_DEFAULT, 0, 0, 0,
1 file changed, 18 insertions(+), 6 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
-index d95a7d7940..2c79ee41af 100644
+index f2eca983f1..b6475d50ad 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -673,8 +673,6 @@ static int mirror_exit_common(Job *job)
1 file changed, 3 insertions(+)
diff --git a/blockdev.c b/blockdev.c
-index fa601838a3..93fd0a61a4 100644
+index 9a1a3118ed..a57b0af2e7 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -3050,6 +3050,9 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
-index 2c79ee41af..e04a3ea6f4 100644
+index b6475d50ad..8b3342f9ec 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -779,8 +779,8 @@ static int mirror_exit_common(Job *job)
3 files changed, 70 insertions(+), 59 deletions(-)
diff --git a/block/mirror.c b/block/mirror.c
-index e04a3ea6f4..4feec2a002 100644
+index 8b3342f9ec..1d4ff0efad 100644
--- a/block/mirror.c
+++ b/block/mirror.c
@@ -1634,31 +1634,13 @@ static BlockJob *mirror_start_job(
if (bitmap_mode != BITMAP_SYNC_MODE_NEVER) {
diff --git a/blockdev.c b/blockdev.c
-index 93fd0a61a4..1af5a1fcb2 100644
+index a57b0af2e7..ce62a9b439 100644
--- a/blockdev.c
+++ b/blockdev.c
@@ -3029,7 +3029,36 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
6 files changed, 59 insertions(+), 5 deletions(-)
diff --git a/include/monitor/monitor.h b/include/monitor/monitor.h
-index cc4cc6c6ad..82f649d9ab 100644
+index a4b40e8391..d64ae8f34e 100644
--- a/include/monitor/monitor.h
+++ b/include/monitor/monitor.h
@@ -16,6 +16,7 @@ extern QemuOptsList qemu_mon_opts;
/**
diff --git a/monitor/monitor.c b/monitor/monitor.c
-index 21c7a68758..ad9813567a 100644
+index 86949024f6..c306cadcf4 100644
--- a/monitor/monitor.c
+++ b/monitor/monitor.c
@@ -135,6 +135,21 @@ bool monitor_cur_is_qmp(void)
aio_bh_schedule_oneshot(qemu_get_aio_context(), do_qmp_dispatch_bh,
&data);
diff --git a/stubs/monitor-core.c b/stubs/monitor-core.c
-index d058a2a00d..3290b58120 100644
+index afa477aae6..d3ff124bf3 100644
--- a/stubs/monitor-core.c
+++ b/stubs/monitor-core.c
-@@ -13,6 +13,11 @@ Monitor *monitor_set_cur(Coroutine *co, Monitor *mon)
+@@ -12,6 +12,11 @@ Monitor *monitor_set_cur(Coroutine *co, Monitor *mon)
return NULL;
}
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Fri, 20 May 2022 09:59:22 +0200
-Subject: [PATCH] block/gluster: correctly set max_pdiscard
-
-On 64-bit platforms, assigning SIZE_MAX to the int64_t max_pdiscard
-results in a negative value, and the following assertion would trigger
-down the line (it's not the same max_pdiscard, but computed from the
-other one):
-qemu-system-x86_64: ../block/io.c:3166: bdrv_co_pdiscard: Assertion
-`max_pdiscard >= bs->bl.request_alignment' failed.
-
-On 32-bit platforms, it's fine to keep using SIZE_MAX.
-
-The assertion in qemu_gluster_co_pdiscard() is checking that the value
-of 'bytes' can safely be passed to glfs_discard_async(), which takes a
-size_t for the argument in question, so it is kept as is. And since
-max_pdiscard is still <= SIZE_MAX, relying on max_pdiscard is still
-fine.
-
-Fixes: 0c8022876f ("block: use int64_t instead of int in driver discard handlers")
-Cc: qemu-stable@nongnu.org
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Message-Id: <20220520075922.43972-1-f.ebner@proxmox.com>
-Reviewed-by: Eric Blake <eblake@redhat.com>
-Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@yandex-team.ru>
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-(cherry-picked from commit 9b38fc56c054c7de65fa3bf7cdd82b32654f6b7d)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/gluster.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/block/gluster.c b/block/gluster.c
-index 80b75cb96c..1079b6186b 100644
---- a/block/gluster.c
-+++ b/block/gluster.c
-@@ -901,7 +901,7 @@ out:
- static void qemu_gluster_refresh_limits(BlockDriverState *bs, Error **errp)
- {
- bs->bl.max_transfer = GLUSTER_MAX_TRANSFER;
-- bs->bl.max_pdiscard = SIZE_MAX;
-+ bs->bl.max_pdiscard = MIN(SIZE_MAX, INT64_MAX);
- }
-
- static int qemu_gluster_reopen_prepare(BDRVReopenState *state,
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Hanna Reitz <hreitz@redhat.com>
-Date: Mon, 14 Mar 2022 17:27:18 +0100
-Subject: [PATCH] block/vmdk: Fix reopening bs->file
-
-VMDK disk data is stored in extents, which may or may not be separate
-from bs->file. VmdkExtent.file points to where they are stored. Each
-that is stored in bs->file will simply reuse the exact pointer value of
-bs->file.
-
-(That is why vmdk_free_extents() will unref VmdkExtent.file (e->file)
-only if e->file != bs->file.)
-
-Reopen operations can change bs->file (they will replace the whole
-BdrvChild object, not just the BDS stored in that BdrvChild), and then
-we will need to change all .file pointers of all such VmdkExtents to
-point to the new BdrvChild.
-
-In vmdk_reopen_prepare(), we have to check which VmdkExtents are
-affected, and in vmdk_reopen_commit(), we can modify them. We have to
-split this because:
-- The new BdrvChild is created only after prepare, so we can change
- VmdkExtent.file only in commit
-- In commit, there no longer is any (valid) reference to the old
- BdrvChild object, so there would be nothing to compare VmdkExtent.file
- against to see whether it was equal to bs->file before reopening
- (There is BDRVReopenState.old_file_bs, but the old bs->file
- BdrvChild's .bs pointer will be NULL-ed when the new BdrvChild is
- created, and so we cannot compare VmdkExtent.file->bs against
- BDRVReopenState.old_file_bs)
-
-Signed-off-by: Hanna Reitz <hreitz@redhat.com>
-Message-Id: <20220314162719.65384-2-hreitz@redhat.com>
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-(cherry-picked from commit 6d17e2879854d7d0e623c06a9286085e97bf3545)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/vmdk.c | 56 +++++++++++++++++++++++++++++++++++++++++++++++++++-
- 1 file changed, 55 insertions(+), 1 deletion(-)
-
-diff --git a/block/vmdk.c b/block/vmdk.c
-index 37c0946066..38e5ab3806 100644
---- a/block/vmdk.c
-+++ b/block/vmdk.c
-@@ -178,6 +178,10 @@ typedef struct BDRVVmdkState {
- char *create_type;
- } BDRVVmdkState;
-
-+typedef struct BDRVVmdkReopenState {
-+ bool *extents_using_bs_file;
-+} BDRVVmdkReopenState;
-+
- typedef struct VmdkMetaData {
- unsigned int l1_index;
- unsigned int l2_index;
-@@ -400,15 +404,63 @@ static int vmdk_is_cid_valid(BlockDriverState *bs)
- return 1;
- }
-
--/* We have nothing to do for VMDK reopen, stubs just return success */
- static int vmdk_reopen_prepare(BDRVReopenState *state,
- BlockReopenQueue *queue, Error **errp)
- {
-+ BDRVVmdkState *s;
-+ BDRVVmdkReopenState *rs;
-+ int i;
-+
- assert(state != NULL);
- assert(state->bs != NULL);
-+ assert(state->opaque == NULL);
-+
-+ s = state->bs->opaque;
-+
-+ rs = g_new0(BDRVVmdkReopenState, 1);
-+ state->opaque = rs;
-+
-+ /*
-+ * Check whether there are any extents stored in bs->file; if bs->file
-+ * changes, we will need to update their .file pointers to follow suit
-+ */
-+ rs->extents_using_bs_file = g_new(bool, s->num_extents);
-+ for (i = 0; i < s->num_extents; i++) {
-+ rs->extents_using_bs_file[i] = s->extents[i].file == state->bs->file;
-+ }
-+
- return 0;
- }
-
-+static void vmdk_reopen_clean(BDRVReopenState *state)
-+{
-+ BDRVVmdkReopenState *rs = state->opaque;
-+
-+ g_free(rs->extents_using_bs_file);
-+ g_free(rs);
-+ state->opaque = NULL;
-+}
-+
-+static void vmdk_reopen_commit(BDRVReopenState *state)
-+{
-+ BDRVVmdkState *s = state->bs->opaque;
-+ BDRVVmdkReopenState *rs = state->opaque;
-+ int i;
-+
-+ for (i = 0; i < s->num_extents; i++) {
-+ if (rs->extents_using_bs_file[i]) {
-+ s->extents[i].file = state->bs->file;
-+ }
-+ }
-+
-+ vmdk_reopen_clean(state);
-+}
-+
-+static void vmdk_reopen_abort(BDRVReopenState *state)
-+{
-+ vmdk_reopen_clean(state);
-+}
-+
- static int vmdk_parent_open(BlockDriverState *bs)
- {
- char *p_name;
-@@ -3072,6 +3124,8 @@ static BlockDriver bdrv_vmdk = {
- .bdrv_open = vmdk_open,
- .bdrv_co_check = vmdk_co_check,
- .bdrv_reopen_prepare = vmdk_reopen_prepare,
-+ .bdrv_reopen_commit = vmdk_reopen_commit,
-+ .bdrv_reopen_abort = vmdk_reopen_abort,
- .bdrv_child_perm = bdrv_default_perms,
- .bdrv_co_preadv = vmdk_co_preadv,
- .bdrv_co_pwritev = vmdk_co_pwritev,
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Hajnoczi <stefanha@redhat.com>
-Date: Thu, 9 Jun 2022 17:47:11 +0100
-Subject: [PATCH] linux-aio: fix unbalanced plugged counter in laio_io_unplug()
-
-Every laio_io_plug() call has a matching laio_io_unplug() call. There is
-a plugged counter that tracks the number of levels of plugging and
-allows for nesting.
-
-The plugged counter must reflect the balance between laio_io_plug() and
-laio_io_unplug() calls accurately. Otherwise I/O stalls occur since
-io_submit(2) calls are skipped while plugged.
-
-Reported-by: Nikolay Tenev <nt@storpool.com>
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
-Message-id: 20220609164712.1539045-2-stefanha@redhat.com
-Cc: Stefano Garzarella <sgarzare@redhat.com>
-Fixes: 68d7946648 ("linux-aio: add `dev_max_batch` parameter to laio_io_unplug()")
-[Stefano Garzarella suggested adding a Fixes tag.
---Stefan]
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-(cherry-picked from commit f387cac5af030a58ac5a0dacf64cab5e5a4fe5c7)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/linux-aio.c | 4 +++-
- 1 file changed, 3 insertions(+), 1 deletion(-)
-
-diff --git a/block/linux-aio.c b/block/linux-aio.c
-index 4c423fcccf..6078da7e42 100644
---- a/block/linux-aio.c
-+++ b/block/linux-aio.c
-@@ -363,8 +363,10 @@ void laio_io_unplug(BlockDriverState *bs, LinuxAioState *s,
- uint64_t dev_max_batch)
- {
- assert(s->io_q.plugged);
-+ s->io_q.plugged--;
-+
- if (s->io_q.in_queue >= laio_max_batch(s, dev_max_batch) ||
-- (--s->io_q.plugged == 0 &&
-+ (!s->io_q.plugged &&
- !s->io_q.blocked && !QSIMPLEQ_EMPTY(&s->io_q.pending))) {
- ioq_submit(s);
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Claudio Fontana <cfontana@suse.de>
-Date: Tue, 31 May 2022 13:47:07 +0200
-Subject: [PATCH] pci: fix overflow in snprintf string formatting
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-the code in pcibus_get_fw_dev_path contained the potential for a
-stack buffer overflow of 1 byte, potentially writing to the stack an
-extra NUL byte.
-
-This overflow could happen if the PCI slot is >= 0x10000000,
-and the PCI function is >= 0x10000000, due to the size parameter
-of snprintf being incorrectly calculated in the call:
-
- if (PCI_FUNC(d->devfn))
- snprintf(path + off, sizeof(path) + off, ",%x", PCI_FUNC(d->devfn));
-
-since the off obtained from a previous call to snprintf is added
-instead of subtracted from the total available size of the buffer.
-
-Without the accurate size guard from snprintf, we end up writing in the
-worst case:
-
-name (32) + "@" (1) + SLOT (8) + "," (1) + FUNC (8) + term NUL (1) = 51 bytes
-
-In order to provide something more robust, replace all of the code in
-pcibus_get_fw_dev_path with a single call to g_strdup_printf,
-so there is no need to rely on manual calculations.
-
-Found by compiling QEMU with FORTIFY_SOURCE=3 as the error:
-
-*** buffer overflow detected ***: terminated
-
-Thread 1 "qemu-system-x86" received signal SIGABRT, Aborted.
-[Switching to Thread 0x7ffff642c380 (LWP 121307)]
-0x00007ffff71ff55c in __pthread_kill_implementation () from /lib64/libc.so.6
-(gdb) bt
- #0 0x00007ffff71ff55c in __pthread_kill_implementation () at /lib64/libc.so.6
- #1 0x00007ffff71ac6f6 in raise () at /lib64/libc.so.6
- #2 0x00007ffff7195814 in abort () at /lib64/libc.so.6
- #3 0x00007ffff71f279e in __libc_message () at /lib64/libc.so.6
- #4 0x00007ffff729767a in __fortify_fail () at /lib64/libc.so.6
- #5 0x00007ffff7295c36 in () at /lib64/libc.so.6
- #6 0x00007ffff72957f5 in __snprintf_chk () at /lib64/libc.so.6
- #7 0x0000555555b1c1fd in pcibus_get_fw_dev_path ()
- #8 0x0000555555f2bde4 in qdev_get_fw_dev_path_helper.constprop ()
- #9 0x0000555555f2bd86 in qdev_get_fw_dev_path_helper.constprop ()
- #10 0x00005555559a6e5d in get_boot_device_path ()
- #11 0x00005555559a712c in get_boot_devices_list ()
- #12 0x0000555555b1a3d0 in fw_cfg_machine_reset ()
- #13 0x0000555555bf4c2d in pc_machine_reset ()
- #14 0x0000555555c66988 in qemu_system_reset ()
- #15 0x0000555555a6dff6 in qdev_machine_creation_done ()
- #16 0x0000555555c79186 in qmp_x_exit_preconfig.part ()
- #17 0x0000555555c7b459 in qemu_init ()
- #18 0x0000555555960a29 in main ()
-
-Found-by: Dario Faggioli <Dario Faggioli <dfaggioli@suse.com>
-Found-by: Martin Liška <martin.liska@suse.com>
-Cc: qemu-stable@nongnu.org
-Signed-off-by: Claudio Fontana <cfontana@suse.de>
-Message-Id: <20220531114707.18830-1-cfontana@suse.de>
-Reviewed-by: Ani Sinha <ani@anisinha.ca>
-(cherry-picked from commit 36f18c6989a3d1ff1d7a0e50b0868ef3958299b4)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/pci/pci.c | 18 +++++++++---------
- 1 file changed, 9 insertions(+), 9 deletions(-)
-
-diff --git a/hw/pci/pci.c b/hw/pci/pci.c
-index dae9119bfe..c69b412434 100644
---- a/hw/pci/pci.c
-+++ b/hw/pci/pci.c
-@@ -2625,15 +2625,15 @@ static char *pci_dev_fw_name(DeviceState *dev, char *buf, int len)
- static char *pcibus_get_fw_dev_path(DeviceState *dev)
- {
- PCIDevice *d = (PCIDevice *)dev;
-- char path[50], name[33];
-- int off;
--
-- off = snprintf(path, sizeof(path), "%s@%x",
-- pci_dev_fw_name(dev, name, sizeof name),
-- PCI_SLOT(d->devfn));
-- if (PCI_FUNC(d->devfn))
-- snprintf(path + off, sizeof(path) + off, ",%x", PCI_FUNC(d->devfn));
-- return g_strdup(path);
-+ char name[33];
-+ int has_func = !!PCI_FUNC(d->devfn);
-+
-+ return g_strdup_printf("%s@%x%s%.*x",
-+ pci_dev_fw_name(dev, name, sizeof(name)),
-+ PCI_SLOT(d->devfn),
-+ has_func ? "," : "",
-+ has_func,
-+ PCI_FUNC(d->devfn));
- }
-
- static char *pcibus_get_dev_path(DeviceState *dev)
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: "Maciej S. Szmigiero" <maciej.szmigiero@oracle.com>
-Date: Mon, 23 May 2022 18:26:58 +0200
-Subject: [PATCH] target/i386/kvm: Fix disabling MPX on "-cpu host" with
- MPX-capable host
-
-Since KVM commit 5f76f6f5ff96 ("KVM: nVMX: Do not expose MPX VMX controls when guest MPX disabled")
-it is not possible to disable MPX on a "-cpu host" just by adding "-mpx"
-there if the host CPU does indeed support MPX.
-QEMU will fail to set MSR_IA32_VMX_TRUE_{EXIT,ENTRY}_CTLS MSRs in this case
-and so trigger an assertion failure.
-
-Instead, besides "-mpx" one has to explicitly add also
-"-vmx-exit-clear-bndcfgs" and "-vmx-entry-load-bndcfgs" to QEMU command
-line to make it work, which is a bit convoluted.
-
-Make the MPX-related bits in FEAT_VMX_{EXIT,ENTRY}_CTLS dependent on MPX
-being actually enabled so such workarounds are no longer necessary.
-
-Signed-off-by: Maciej S. Szmigiero <maciej.szmigiero@oracle.com>
-Message-Id: <51aa2125c76363204cc23c27165e778097c33f0b.1653323077.git.maciej.szmigiero@oracle.com>
-Cc: qemu-stable@nongnu.org
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 267b5e7e378afd260004cb37a66a6fcd641e3b53)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- target/i386/cpu.c | 8 ++++++++
- 1 file changed, 8 insertions(+)
-
-diff --git a/target/i386/cpu.c b/target/i386/cpu.c
-index cb6b5467d0..6e6945139b 100644
---- a/target/i386/cpu.c
-+++ b/target/i386/cpu.c
-@@ -1327,6 +1327,14 @@ static FeatureDep feature_dependencies[] = {
- .from = { FEAT_7_0_EBX, CPUID_7_0_EBX_INVPCID },
- .to = { FEAT_VMX_SECONDARY_CTLS, VMX_SECONDARY_EXEC_ENABLE_INVPCID },
- },
-+ {
-+ .from = { FEAT_7_0_EBX, CPUID_7_0_EBX_MPX },
-+ .to = { FEAT_VMX_EXIT_CTLS, VMX_VM_EXIT_CLEAR_BNDCFGS },
-+ },
-+ {
-+ .from = { FEAT_7_0_EBX, CPUID_7_0_EBX_MPX },
-+ .to = { FEAT_VMX_ENTRY_CTLS, VMX_VM_ENTRY_LOAD_BNDCFGS },
-+ },
- {
- .from = { FEAT_7_0_EBX, CPUID_7_0_EBX_RDSEED },
- .to = { FEAT_VMX_SECONDARY_CTLS, VMX_SECONDARY_EXEC_RDSEED_EXITING },
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Hajnoczi <stefanha@redhat.com>
-Date: Mon, 7 Mar 2022 15:38:51 +0000
-Subject: [PATCH] coroutine-ucontext: use QEMU_DEFINE_STATIC_CO_TLS()
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Thread-Local Storage variables cannot be used directly from coroutine
-code because the compiler may optimize TLS variable accesses across
-qemu_coroutine_yield() calls. When the coroutine is re-entered from
-another thread the TLS variables from the old thread must no longer be
-used.
-
-Use QEMU_DEFINE_STATIC_CO_TLS() for the current and leader variables.
-
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-Message-Id: <20220307153853.602859-2-stefanha@redhat.com>
-Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-(cherry-picked from commit 34145a307d849d0b6734d0222a7aa0bb9eef7407)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- util/coroutine-ucontext.c | 38 ++++++++++++++++++++++++--------------
- 1 file changed, 24 insertions(+), 14 deletions(-)
-
-diff --git a/util/coroutine-ucontext.c b/util/coroutine-ucontext.c
-index 904b375192..127d5a13c8 100644
---- a/util/coroutine-ucontext.c
-+++ b/util/coroutine-ucontext.c
-@@ -25,6 +25,7 @@
- #include "qemu/osdep.h"
- #include <ucontext.h>
- #include "qemu/coroutine_int.h"
-+#include "qemu/coroutine-tls.h"
-
- #ifdef CONFIG_VALGRIND_H
- #include <valgrind/valgrind.h>
-@@ -66,8 +67,8 @@ typedef struct {
- /**
- * Per-thread coroutine bookkeeping
- */
--static __thread CoroutineUContext leader;
--static __thread Coroutine *current;
-+QEMU_DEFINE_STATIC_CO_TLS(Coroutine *, current);
-+QEMU_DEFINE_STATIC_CO_TLS(CoroutineUContext, leader);
-
- /*
- * va_args to makecontext() must be type 'int', so passing
-@@ -97,14 +98,15 @@ static inline __attribute__((always_inline))
- void finish_switch_fiber(void *fake_stack_save)
- {
- #ifdef CONFIG_ASAN
-+ CoroutineUContext *leaderp = get_ptr_leader();
- const void *bottom_old;
- size_t size_old;
-
- __sanitizer_finish_switch_fiber(fake_stack_save, &bottom_old, &size_old);
-
-- if (!leader.stack) {
-- leader.stack = (void *)bottom_old;
-- leader.stack_size = size_old;
-+ if (!leaderp->stack) {
-+ leaderp->stack = (void *)bottom_old;
-+ leaderp->stack_size = size_old;
- }
- #endif
- #ifdef CONFIG_TSAN
-@@ -161,8 +163,10 @@ static void coroutine_trampoline(int i0, int i1)
-
- /* Initialize longjmp environment and switch back the caller */
- if (!sigsetjmp(self->env, 0)) {
-- start_switch_fiber_asan(COROUTINE_YIELD, &fake_stack_save, leader.stack,
-- leader.stack_size);
-+ CoroutineUContext *leaderp = get_ptr_leader();
-+
-+ start_switch_fiber_asan(COROUTINE_YIELD, &fake_stack_save,
-+ leaderp->stack, leaderp->stack_size);
- start_switch_fiber_tsan(&fake_stack_save, self, true); /* true=caller */
- siglongjmp(*(sigjmp_buf *)co->entry_arg, 1);
- }
-@@ -297,7 +301,7 @@ qemu_coroutine_switch(Coroutine *from_, Coroutine *to_,
- int ret;
- void *fake_stack_save = NULL;
-
-- current = to_;
-+ set_current(to_);
-
- ret = sigsetjmp(from->env, 0);
- if (ret == 0) {
-@@ -315,18 +319,24 @@ qemu_coroutine_switch(Coroutine *from_, Coroutine *to_,
-
- Coroutine *qemu_coroutine_self(void)
- {
-- if (!current) {
-- current = &leader.base;
-+ Coroutine *self = get_current();
-+ CoroutineUContext *leaderp = get_ptr_leader();
-+
-+ if (!self) {
-+ self = &leaderp->base;
-+ set_current(self);
- }
- #ifdef CONFIG_TSAN
-- if (!leader.tsan_co_fiber) {
-- leader.tsan_co_fiber = __tsan_get_current_fiber();
-+ if (!leaderp->tsan_co_fiber) {
-+ leaderp->tsan_co_fiber = __tsan_get_current_fiber();
- }
- #endif
-- return current;
-+ return self;
- }
-
- bool qemu_in_coroutine(void)
- {
-- return current && current->caller;
-+ Coroutine *self = get_current();
-+
-+ return self && self->caller;
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Hajnoczi <stefanha@redhat.com>
-Date: Mon, 7 Mar 2022 15:38:52 +0000
-Subject: [PATCH] coroutine: use QEMU_DEFINE_STATIC_CO_TLS()
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Thread-Local Storage variables cannot be used directly from coroutine
-code because the compiler may optimize TLS variable accesses across
-qemu_coroutine_yield() calls. When the coroutine is re-entered from
-another thread the TLS variables from the old thread must no longer be
-used.
-
-Use QEMU_DEFINE_STATIC_CO_TLS() for the current and leader variables.
-The alloc_pool QSLIST needs a typedef so the return value of
-get_ptr_alloc_pool() can be stored in a local variable.
-
-One example of why this code is necessary: a coroutine that yields
-before calling qemu_coroutine_create() to create another coroutine is
-affected by the TLS issue.
-
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-Message-Id: <20220307153853.602859-3-stefanha@redhat.com>
-Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-(cherry-picked from commit ac387a08a9c9f6b36757da912f0339c25f421f90)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- util/qemu-coroutine.c | 41 ++++++++++++++++++++++++-----------------
- 1 file changed, 24 insertions(+), 17 deletions(-)
-
-diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
-index c03b2422ff..f3e8300c8d 100644
---- a/util/qemu-coroutine.c
-+++ b/util/qemu-coroutine.c
-@@ -18,6 +18,7 @@
- #include "qemu/atomic.h"
- #include "qemu/coroutine.h"
- #include "qemu/coroutine_int.h"
-+#include "qemu/coroutine-tls.h"
- #include "block/aio.h"
-
- /** Initial batch size is 64, and is increased on demand */
-@@ -29,17 +30,20 @@ enum {
- static QSLIST_HEAD(, Coroutine) release_pool = QSLIST_HEAD_INITIALIZER(pool);
- static unsigned int pool_batch_size = POOL_INITIAL_BATCH_SIZE;
- static unsigned int release_pool_size;
--static __thread QSLIST_HEAD(, Coroutine) alloc_pool = QSLIST_HEAD_INITIALIZER(pool);
--static __thread unsigned int alloc_pool_size;
--static __thread Notifier coroutine_pool_cleanup_notifier;
-+
-+typedef QSLIST_HEAD(, Coroutine) CoroutineQSList;
-+QEMU_DEFINE_STATIC_CO_TLS(CoroutineQSList, alloc_pool);
-+QEMU_DEFINE_STATIC_CO_TLS(unsigned int, alloc_pool_size);
-+QEMU_DEFINE_STATIC_CO_TLS(Notifier, coroutine_pool_cleanup_notifier);
-
- static void coroutine_pool_cleanup(Notifier *n, void *value)
- {
- Coroutine *co;
- Coroutine *tmp;
-+ CoroutineQSList *alloc_pool = get_ptr_alloc_pool();
-
-- QSLIST_FOREACH_SAFE(co, &alloc_pool, pool_next, tmp) {
-- QSLIST_REMOVE_HEAD(&alloc_pool, pool_next);
-+ QSLIST_FOREACH_SAFE(co, alloc_pool, pool_next, tmp) {
-+ QSLIST_REMOVE_HEAD(alloc_pool, pool_next);
- qemu_coroutine_delete(co);
- }
- }
-@@ -49,27 +53,30 @@ Coroutine *qemu_coroutine_create(CoroutineEntry *entry, void *opaque)
- Coroutine *co = NULL;
-
- if (CONFIG_COROUTINE_POOL) {
-- co = QSLIST_FIRST(&alloc_pool);
-+ CoroutineQSList *alloc_pool = get_ptr_alloc_pool();
-+
-+ co = QSLIST_FIRST(alloc_pool);
- if (!co) {
- if (release_pool_size > qatomic_read(&pool_batch_size)) {
- /* Slow path; a good place to register the destructor, too. */
-- if (!coroutine_pool_cleanup_notifier.notify) {
-- coroutine_pool_cleanup_notifier.notify = coroutine_pool_cleanup;
-- qemu_thread_atexit_add(&coroutine_pool_cleanup_notifier);
-+ Notifier *notifier = get_ptr_coroutine_pool_cleanup_notifier();
-+ if (!notifier->notify) {
-+ notifier->notify = coroutine_pool_cleanup;
-+ qemu_thread_atexit_add(notifier);
- }
-
- /* This is not exact; there could be a little skew between
- * release_pool_size and the actual size of release_pool. But
- * it is just a heuristic, it does not need to be perfect.
- */
-- alloc_pool_size = qatomic_xchg(&release_pool_size, 0);
-- QSLIST_MOVE_ATOMIC(&alloc_pool, &release_pool);
-- co = QSLIST_FIRST(&alloc_pool);
-+ set_alloc_pool_size(qatomic_xchg(&release_pool_size, 0));
-+ QSLIST_MOVE_ATOMIC(alloc_pool, &release_pool);
-+ co = QSLIST_FIRST(alloc_pool);
- }
- }
- if (co) {
-- QSLIST_REMOVE_HEAD(&alloc_pool, pool_next);
-- alloc_pool_size--;
-+ QSLIST_REMOVE_HEAD(alloc_pool, pool_next);
-+ set_alloc_pool_size(get_alloc_pool_size() - 1);
- }
- }
-
-@@ -93,9 +100,9 @@ static void coroutine_delete(Coroutine *co)
- qatomic_inc(&release_pool_size);
- return;
- }
-- if (alloc_pool_size < qatomic_read(&pool_batch_size)) {
-- QSLIST_INSERT_HEAD(&alloc_pool, co, pool_next);
-- alloc_pool_size++;
-+ if (get_alloc_pool_size() < qatomic_read(&pool_batch_size)) {
-+ QSLIST_INSERT_HEAD(get_ptr_alloc_pool(), co, pool_next);
-+ set_alloc_pool_size(get_alloc_pool_size() + 1);
- return;
- }
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Kevin Wolf <kwolf@redhat.com>
-Date: Tue, 10 May 2022 17:10:19 +0200
-Subject: [PATCH] coroutine: Rename qemu_coroutine_inc/dec_pool_size()
-
-It's true that these functions currently affect the batch size in which
-coroutines are reused (i.e. moved from the global release pool to the
-allocation pool of a specific thread), but this is a bug and will be
-fixed in a separate patch.
-
-In fact, the comment in the header file already just promises that it
-influences the pool size, so reflect this in the name of the functions.
-As a nice side effect, the shorter function name makes some line
-wrapping unnecessary.
-
-Cc: qemu-stable@nongnu.org
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-Message-Id: <20220510151020.105528-2-kwolf@redhat.com>
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-(cherry-picked from commit 98e3ab35054b946f7c2aba5408822532b0920b53)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/block/virtio-blk.c | 6 ++----
- include/qemu/coroutine.h | 6 +++---
- util/qemu-coroutine.c | 4 ++--
- 3 files changed, 7 insertions(+), 9 deletions(-)
-
-diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
-index 540c38f829..6a1cc41877 100644
---- a/hw/block/virtio-blk.c
-+++ b/hw/block/virtio-blk.c
-@@ -1215,8 +1215,7 @@ static void virtio_blk_device_realize(DeviceState *dev, Error **errp)
- for (i = 0; i < conf->num_queues; i++) {
- virtio_add_queue(vdev, conf->queue_size, virtio_blk_handle_output);
- }
-- qemu_coroutine_increase_pool_batch_size(conf->num_queues * conf->queue_size
-- / 2);
-+ qemu_coroutine_inc_pool_size(conf->num_queues * conf->queue_size / 2);
- virtio_blk_data_plane_create(vdev, conf, &s->dataplane, &err);
- if (err != NULL) {
- error_propagate(errp, err);
-@@ -1253,8 +1252,7 @@ static void virtio_blk_device_unrealize(DeviceState *dev)
- for (i = 0; i < conf->num_queues; i++) {
- virtio_del_queue(vdev, i);
- }
-- qemu_coroutine_decrease_pool_batch_size(conf->num_queues * conf->queue_size
-- / 2);
-+ qemu_coroutine_dec_pool_size(conf->num_queues * conf->queue_size / 2);
- qemu_del_vm_change_state_handler(s->change);
- blockdev_mark_auto_del(s->blk);
- virtio_cleanup(vdev);
-diff --git a/include/qemu/coroutine.h b/include/qemu/coroutine.h
-index c828a95ee0..5b621d1295 100644
---- a/include/qemu/coroutine.h
-+++ b/include/qemu/coroutine.h
-@@ -334,12 +334,12 @@ void coroutine_fn yield_until_fd_readable(int fd);
- /**
- * Increase coroutine pool size
- */
--void qemu_coroutine_increase_pool_batch_size(unsigned int additional_pool_size);
-+void qemu_coroutine_inc_pool_size(unsigned int additional_pool_size);
-
- /**
-- * Devcrease coroutine pool size
-+ * Decrease coroutine pool size
- */
--void qemu_coroutine_decrease_pool_batch_size(unsigned int additional_pool_size);
-+void qemu_coroutine_dec_pool_size(unsigned int additional_pool_size);
-
- #include "qemu/lockable.h"
-
-diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
-index f3e8300c8d..ea23929a74 100644
---- a/util/qemu-coroutine.c
-+++ b/util/qemu-coroutine.c
-@@ -212,12 +212,12 @@ AioContext *coroutine_fn qemu_coroutine_get_aio_context(Coroutine *co)
- return co->ctx;
- }
-
--void qemu_coroutine_increase_pool_batch_size(unsigned int additional_pool_size)
-+void qemu_coroutine_inc_pool_size(unsigned int additional_pool_size)
- {
- qatomic_add(&pool_batch_size, additional_pool_size);
- }
-
--void qemu_coroutine_decrease_pool_batch_size(unsigned int removing_pool_size)
-+void qemu_coroutine_dec_pool_size(unsigned int removing_pool_size)
- {
- qatomic_sub(&pool_batch_size, removing_pool_size);
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Kevin Wolf <kwolf@redhat.com>
-Date: Tue, 10 May 2022 17:10:20 +0200
-Subject: [PATCH] coroutine: Revert to constant batch size
-
-Commit 4c41c69e changed the way the coroutine pool is sized because for
-virtio-blk devices with a large queue size and heavy I/O, it was just
-too small and caused coroutines to be deleted and reallocated soon
-afterwards. The change made the size dynamic based on the number of
-queues and the queue size of virtio-blk devices.
-
-There are two important numbers here: Slightly simplified, when a
-coroutine terminates, it is generally stored in the global release pool
-up to a certain pool size, and if the pool is full, it is freed.
-Conversely, when allocating a new coroutine, the coroutines in the
-release pool are reused if the pool already has reached a certain
-minimum size (the batch size), otherwise we allocate new coroutines.
-
-The problem after commit 4c41c69e is that it not only increases the
-maximum pool size (which is the intended effect), but also the batch
-size for reusing coroutines (which is a bug). It means that in cases
-with many devices and/or a large queue size (which defaults to the
-number of vcpus for virtio-blk-pci), many thousand coroutines could be
-sitting in the release pool without being reused.
-
-This is not only a waste of memory and allocations, but it actually
-makes the QEMU process likely to hit the vm.max_map_count limit on Linux
-because each coroutine requires two mappings (its stack and the guard
-page for the stack), causing it to abort() in qemu_alloc_stack() because
-when the limit is hit, mprotect() starts to fail with ENOMEM.
-
-In order to fix the problem, change the batch size back to 64 to avoid
-uselessly accumulating coroutines in the release pool, but keep the
-dynamic maximum pool size so that coroutines aren't freed too early
-in heavy I/O scenarios.
-
-Note that this fix doesn't strictly make it impossible to hit the limit,
-but this would only happen if most of the coroutines are actually in use
-at the same time, not just sitting in a pool. This is the same behaviour
-as we already had before commit 4c41c69e. Fully preventing this would
-require allowing qemu_coroutine_create() to return an error, but it
-doesn't seem to be a scenario that people hit in practice.
-
-Cc: qemu-stable@nongnu.org
-Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2079938
-Fixes: 4c41c69e05fe28c0f95f8abd2ebf407e95a4f04b
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-Message-Id: <20220510151020.105528-3-kwolf@redhat.com>
-Tested-by: Hiroki Narukawa <hnarukaw@yahoo-corp.jp>
-Signed-off-by: Kevin Wolf <kwolf@redhat.com>
-(cherry-picked from commit 9ec7a59b5aad4b736871c378d30f5ef5ec51cb52)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- util/qemu-coroutine.c | 22 ++++++++++++++--------
- 1 file changed, 14 insertions(+), 8 deletions(-)
-
-diff --git a/util/qemu-coroutine.c b/util/qemu-coroutine.c
-index ea23929a74..4a8bd63ef0 100644
---- a/util/qemu-coroutine.c
-+++ b/util/qemu-coroutine.c
-@@ -21,14 +21,20 @@
- #include "qemu/coroutine-tls.h"
- #include "block/aio.h"
-
--/** Initial batch size is 64, and is increased on demand */
-+/**
-+ * The minimal batch size is always 64, coroutines from the release_pool are
-+ * reused as soon as there are 64 coroutines in it. The maximum pool size starts
-+ * with 64 and is increased on demand so that coroutines are not deleted even if
-+ * they are not immediately reused.
-+ */
- enum {
-- POOL_INITIAL_BATCH_SIZE = 64,
-+ POOL_MIN_BATCH_SIZE = 64,
-+ POOL_INITIAL_MAX_SIZE = 64,
- };
-
- /** Free list to speed up creation */
- static QSLIST_HEAD(, Coroutine) release_pool = QSLIST_HEAD_INITIALIZER(pool);
--static unsigned int pool_batch_size = POOL_INITIAL_BATCH_SIZE;
-+static unsigned int pool_max_size = POOL_INITIAL_MAX_SIZE;
- static unsigned int release_pool_size;
-
- typedef QSLIST_HEAD(, Coroutine) CoroutineQSList;
-@@ -57,7 +63,7 @@ Coroutine *qemu_coroutine_create(CoroutineEntry *entry, void *opaque)
-
- co = QSLIST_FIRST(alloc_pool);
- if (!co) {
-- if (release_pool_size > qatomic_read(&pool_batch_size)) {
-+ if (release_pool_size > POOL_MIN_BATCH_SIZE) {
- /* Slow path; a good place to register the destructor, too. */
- Notifier *notifier = get_ptr_coroutine_pool_cleanup_notifier();
- if (!notifier->notify) {
-@@ -95,12 +101,12 @@ static void coroutine_delete(Coroutine *co)
- co->caller = NULL;
-
- if (CONFIG_COROUTINE_POOL) {
-- if (release_pool_size < qatomic_read(&pool_batch_size) * 2) {
-+ if (release_pool_size < qatomic_read(&pool_max_size) * 2) {
- QSLIST_INSERT_HEAD_ATOMIC(&release_pool, co, pool_next);
- qatomic_inc(&release_pool_size);
- return;
- }
-- if (get_alloc_pool_size() < qatomic_read(&pool_batch_size)) {
-+ if (get_alloc_pool_size() < qatomic_read(&pool_max_size)) {
- QSLIST_INSERT_HEAD(get_ptr_alloc_pool(), co, pool_next);
- set_alloc_pool_size(get_alloc_pool_size() + 1);
- return;
-@@ -214,10 +220,10 @@ AioContext *coroutine_fn qemu_coroutine_get_aio_context(Coroutine *co)
-
- void qemu_coroutine_inc_pool_size(unsigned int additional_pool_size)
- {
-- qatomic_add(&pool_batch_size, additional_pool_size);
-+ qatomic_add(&pool_max_size, additional_pool_size);
- }
-
- void qemu_coroutine_dec_pool_size(unsigned int removing_pool_size)
- {
-- qatomic_sub(&pool_batch_size, removing_pool_size);
-+ qatomic_sub(&pool_max_size, removing_pool_size);
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Paolo Bonzini <pbonzini@redhat.com>
-Date: Fri, 29 Apr 2022 21:16:28 +0200
-Subject: [PATCH] target/i386: do not consult nonexistent host leaves
-
-When cache_info_passthrough is requested, QEMU passes the host values
-of the cache information CPUID leaves down to the guest. However,
-it blindly assumes that the CPUID leaf exists on the host, and this
-cannot be guaranteed: for example, KVM has recently started to
-synthesize AMD leaves up to 0x80000021 in order to provide accurate
-CPU bug information to guests.
-
-Querying a nonexistent host leaf fills the output arguments of
-host_cpuid with data that (albeit deterministic) is nonsensical
-as cache information, namely the data in the highest Intel CPUID
-leaf. If said highest leaf is not ECX-dependent, this can even
-cause an infinite loop when kvm_arch_init_vcpu prepares the input
-to KVM_SET_CPUID2. The infinite loop is only terminated by an
-abort() when the array gets full.
-
-Reported-by: Maxim Levitsky <mlevitsk@redhat.com>
-Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
-Cc: qemu-stable@nongnu.org
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 798d8ec0dacd4cc0034298d94f430c14f23e2919)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- target/i386/cpu.c | 41 ++++++++++++++++++++++++++++++++++++-----
- 1 file changed, 36 insertions(+), 5 deletions(-)
-
-diff --git a/target/i386/cpu.c b/target/i386/cpu.c
-index 6e6945139b..c79e151887 100644
---- a/target/i386/cpu.c
-+++ b/target/i386/cpu.c
-@@ -5030,6 +5030,37 @@ uint64_t x86_cpu_get_supported_feature_word(FeatureWord w,
- return r;
- }
-
-+static void x86_cpu_get_cache_cpuid(uint32_t func, uint32_t index,
-+ uint32_t *eax, uint32_t *ebx,
-+ uint32_t *ecx, uint32_t *edx)
-+{
-+ uint32_t level, unused;
-+
-+ /* Only return valid host leaves. */
-+ switch (func) {
-+ case 2:
-+ case 4:
-+ host_cpuid(0, 0, &level, &unused, &unused, &unused);
-+ break;
-+ case 0x80000005:
-+ case 0x80000006:
-+ case 0x8000001d:
-+ host_cpuid(0x80000000, 0, &level, &unused, &unused, &unused);
-+ break;
-+ default:
-+ return;
-+ }
-+
-+ if (func > level) {
-+ *eax = 0;
-+ *ebx = 0;
-+ *ecx = 0;
-+ *edx = 0;
-+ } else {
-+ host_cpuid(func, index, eax, ebx, ecx, edx);
-+ }
-+}
-+
- /*
- * Only for builtin_x86_defs models initialized with x86_register_cpudef_types.
- */
-@@ -5288,7 +5319,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
- case 2:
- /* cache info: needed for Pentium Pro compatibility */
- if (cpu->cache_info_passthrough) {
-- host_cpuid(index, 0, eax, ebx, ecx, edx);
-+ x86_cpu_get_cache_cpuid(index, 0, eax, ebx, ecx, edx);
- break;
- } else if (cpu->vendor_cpuid_only && IS_AMD_CPU(env)) {
- *eax = *ebx = *ecx = *edx = 0;
-@@ -5308,7 +5339,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
- case 4:
- /* cache info: needed for Core compatibility */
- if (cpu->cache_info_passthrough) {
-- host_cpuid(index, count, eax, ebx, ecx, edx);
-+ x86_cpu_get_cache_cpuid(index, count, eax, ebx, ecx, edx);
- /* QEMU gives out its own APIC IDs, never pass down bits 31..26. */
- *eax &= ~0xFC000000;
- if ((*eax & 31) && cs->nr_cores > 1) {
-@@ -5710,7 +5741,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
- case 0x80000005:
- /* cache info (L1 cache) */
- if (cpu->cache_info_passthrough) {
-- host_cpuid(index, 0, eax, ebx, ecx, edx);
-+ x86_cpu_get_cache_cpuid(index, 0, eax, ebx, ecx, edx);
- break;
- }
- *eax = (L1_DTLB_2M_ASSOC << 24) | (L1_DTLB_2M_ENTRIES << 16) |
-@@ -5723,7 +5754,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
- case 0x80000006:
- /* cache info (L2 cache) */
- if (cpu->cache_info_passthrough) {
-- host_cpuid(index, 0, eax, ebx, ecx, edx);
-+ x86_cpu_get_cache_cpuid(index, 0, eax, ebx, ecx, edx);
- break;
- }
- *eax = (AMD_ENC_ASSOC(L2_DTLB_2M_ASSOC) << 28) |
-@@ -5783,7 +5814,7 @@ void cpu_x86_cpuid(CPUX86State *env, uint32_t index, uint32_t count,
- case 0x8000001D:
- *eax = 0;
- if (cpu->cache_info_passthrough) {
-- host_cpuid(index, count, eax, ebx, ecx, edx);
-+ x86_cpu_get_cache_cpuid(index, count, eax, ebx, ecx, edx);
- break;
- }
- switch (count) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Hajnoczi <stefanha@redhat.com>
-Date: Wed, 27 Apr 2022 15:35:36 +0100
-Subject: [PATCH] virtio-scsi: fix ctrl and event handler functions in
- dataplane mode
-
-Commit f34e8d8b8d48d73f36a67b6d5e492ef9784b5012 ("virtio-scsi: prepare
-virtio_scsi_handle_cmd for dataplane") prepared the virtio-scsi cmd
-virtqueue handler function to be used in both the dataplane and
-non-datpalane code paths.
-
-It failed to convert the ctrl and event virtqueue handler functions,
-which are not designed to be called from the dataplane code path but
-will be since the ioeventfd is set up for those virtqueues when
-dataplane starts.
-
-Convert the ctrl and event virtqueue handler functions now so they
-operate correctly when called from the dataplane code path. Avoid code
-duplication by extracting this code into a helper function.
-
-Fixes: f34e8d8b8d48d73f36a67b6d5e492ef9784b5012 ("virtio-scsi: prepare virtio_scsi_handle_cmd for dataplane")
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
-Message-id: 20220427143541.119567-2-stefanha@redhat.com
-[Fixed s/by used/be used/ typo pointed out by Michael Tokarev
-<mjt@tls.msk.ru>.
---Stefan]
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-(cherry-picked from commit 2f743ef6366c2df4ef51ef3ae318138cdc0125ab)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/scsi/virtio-scsi.c | 42 +++++++++++++++++++++++++++---------------
- 1 file changed, 27 insertions(+), 15 deletions(-)
-
-diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
-index 34a968ecfb..417fbc71d6 100644
---- a/hw/scsi/virtio-scsi.c
-+++ b/hw/scsi/virtio-scsi.c
-@@ -472,16 +472,32 @@ bool virtio_scsi_handle_ctrl_vq(VirtIOSCSI *s, VirtQueue *vq)
- return progress;
- }
-
-+/*
-+ * If dataplane is configured but not yet started, do so now and return true on
-+ * success.
-+ *
-+ * Dataplane is started by the core virtio code but virtqueue handler functions
-+ * can also be invoked when a guest kicks before DRIVER_OK, so this helper
-+ * function helps us deal with manually starting ioeventfd in that case.
-+ */
-+static bool virtio_scsi_defer_to_dataplane(VirtIOSCSI *s)
-+{
-+ if (!s->ctx || s->dataplane_started) {
-+ return false;
-+ }
-+
-+ virtio_device_start_ioeventfd(&s->parent_obj.parent_obj);
-+ return !s->dataplane_fenced;
-+}
-+
- static void virtio_scsi_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
- {
- VirtIOSCSI *s = (VirtIOSCSI *)vdev;
-
-- if (s->ctx) {
-- virtio_device_start_ioeventfd(vdev);
-- if (!s->dataplane_fenced) {
-- return;
-- }
-+ if (virtio_scsi_defer_to_dataplane(s)) {
-+ return;
- }
-+
- virtio_scsi_acquire(s);
- virtio_scsi_handle_ctrl_vq(s, vq);
- virtio_scsi_release(s);
-@@ -720,12 +736,10 @@ static void virtio_scsi_handle_cmd(VirtIODevice *vdev, VirtQueue *vq)
- /* use non-QOM casts in the data path */
- VirtIOSCSI *s = (VirtIOSCSI *)vdev;
-
-- if (s->ctx && !s->dataplane_started) {
-- virtio_device_start_ioeventfd(vdev);
-- if (!s->dataplane_fenced) {
-- return;
-- }
-+ if (virtio_scsi_defer_to_dataplane(s)) {
-+ return;
- }
-+
- virtio_scsi_acquire(s);
- virtio_scsi_handle_cmd_vq(s, vq);
- virtio_scsi_release(s);
-@@ -855,12 +869,10 @@ static void virtio_scsi_handle_event(VirtIODevice *vdev, VirtQueue *vq)
- {
- VirtIOSCSI *s = VIRTIO_SCSI(vdev);
-
-- if (s->ctx) {
-- virtio_device_start_ioeventfd(vdev);
-- if (!s->dataplane_fenced) {
-- return;
-- }
-+ if (virtio_scsi_defer_to_dataplane(s)) {
-+ return;
- }
-+
- virtio_scsi_acquire(s);
- virtio_scsi_handle_event_vq(s, vq);
- virtio_scsi_release(s);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Hajnoczi <stefanha@redhat.com>
-Date: Wed, 27 Apr 2022 15:35:37 +0100
-Subject: [PATCH] virtio-scsi: don't waste CPU polling the event virtqueue
-
-The virtio-scsi event virtqueue is not emptied by its handler function.
-This is typical for rx virtqueues where the device uses buffers when
-some event occurs (e.g. a packet is received, an error condition
-happens, etc).
-
-Polling non-empty virtqueues wastes CPU cycles. We are not waiting for
-new buffers to become available, we are waiting for an event to occur,
-so it's a misuse of CPU resources to poll for buffers.
-
-Introduce the new virtio_queue_aio_attach_host_notifier_no_poll() API,
-which is identical to virtio_queue_aio_attach_host_notifier() except
-that it does not poll the virtqueue.
-
-Before this patch the following command-line consumed 100% CPU in the
-IOThread polling and calling virtio_scsi_handle_event():
-
- $ qemu-system-x86_64 -M accel=kvm -m 1G -cpu host \
- --object iothread,id=iothread0 \
- --device virtio-scsi-pci,iothread=iothread0 \
- --blockdev file,filename=test.img,aio=native,cache.direct=on,node-name=drive0 \
- --device scsi-hd,drive=drive0
-
-After this patch CPU is no longer wasted.
-
-Reported-by: Nir Soffer <nsoffer@redhat.com>
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-Tested-by: Nir Soffer <nsoffer@redhat.com>
-Message-id: 20220427143541.119567-3-stefanha@redhat.com
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-(cherry-picked from commit 38738f7dbbda90fbc161757b7f4be35b52205552)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/scsi/virtio-scsi-dataplane.c | 2 +-
- hw/virtio/virtio.c | 13 +++++++++++++
- include/hw/virtio/virtio.h | 1 +
- 3 files changed, 15 insertions(+), 1 deletion(-)
-
-diff --git a/hw/scsi/virtio-scsi-dataplane.c b/hw/scsi/virtio-scsi-dataplane.c
-index 29575cbaf6..8bb6e6acfc 100644
---- a/hw/scsi/virtio-scsi-dataplane.c
-+++ b/hw/scsi/virtio-scsi-dataplane.c
-@@ -138,7 +138,7 @@ int virtio_scsi_dataplane_start(VirtIODevice *vdev)
-
- aio_context_acquire(s->ctx);
- virtio_queue_aio_attach_host_notifier(vs->ctrl_vq, s->ctx);
-- virtio_queue_aio_attach_host_notifier(vs->event_vq, s->ctx);
-+ virtio_queue_aio_attach_host_notifier_no_poll(vs->event_vq, s->ctx);
-
- for (i = 0; i < vs->conf.num_queues; i++) {
- virtio_queue_aio_attach_host_notifier(vs->cmd_vqs[i], s->ctx);
-diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
-index 9d637e043e..67a873f54a 100644
---- a/hw/virtio/virtio.c
-+++ b/hw/virtio/virtio.c
-@@ -3534,6 +3534,19 @@ void virtio_queue_aio_attach_host_notifier(VirtQueue *vq, AioContext *ctx)
- virtio_queue_host_notifier_aio_poll_end);
- }
-
-+/*
-+ * Same as virtio_queue_aio_attach_host_notifier() but without polling. Use
-+ * this for rx virtqueues and similar cases where the virtqueue handler
-+ * function does not pop all elements. When the virtqueue is left non-empty
-+ * polling consumes CPU cycles and should not be used.
-+ */
-+void virtio_queue_aio_attach_host_notifier_no_poll(VirtQueue *vq, AioContext *ctx)
-+{
-+ aio_set_event_notifier(ctx, &vq->host_notifier, true,
-+ virtio_queue_host_notifier_read,
-+ NULL, NULL);
-+}
-+
- void virtio_queue_aio_detach_host_notifier(VirtQueue *vq, AioContext *ctx)
- {
- aio_set_event_notifier(ctx, &vq->host_notifier, true, NULL, NULL, NULL);
-diff --git a/include/hw/virtio/virtio.h b/include/hw/virtio/virtio.h
-index b31c4507f5..b62a35fdca 100644
---- a/include/hw/virtio/virtio.h
-+++ b/include/hw/virtio/virtio.h
-@@ -317,6 +317,7 @@ EventNotifier *virtio_queue_get_host_notifier(VirtQueue *vq);
- void virtio_queue_set_host_notifier_enabled(VirtQueue *vq, bool enabled);
- void virtio_queue_host_notifier_read(EventNotifier *n);
- void virtio_queue_aio_attach_host_notifier(VirtQueue *vq, AioContext *ctx);
-+void virtio_queue_aio_attach_host_notifier_no_poll(VirtQueue *vq, AioContext *ctx);
- void virtio_queue_aio_detach_host_notifier(VirtQueue *vq, AioContext *ctx);
- VirtQueue *virtio_vector_first_queue(VirtIODevice *vdev, uint16_t vector);
- VirtQueue *virtio_vector_next_queue(VirtQueue *vq);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
-Date: Thu, 12 May 2022 19:57:42 +0200
-Subject: [PATCH] vhost: Track descriptor chain in private at SVQ
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-The device could have access to modify them, and it definitely have
-access when we implement packed vq. Harden SVQ maintaining a private
-copy of the descriptor chain. Other fields like buffer addresses are
-already maintained sepparatedly.
-
-Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
-Message-Id: <20220512175747.142058-2-eperezma@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 495fe3a78749c39c0e772c4e1a55d6cb8a7e5292)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-shadow-virtqueue.c | 12 +++++++-----
- hw/virtio/vhost-shadow-virtqueue.h | 6 ++++++
- 2 files changed, 13 insertions(+), 5 deletions(-)
-
-diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-virtqueue.c
-index b232803d1b..3155801f50 100644
---- a/hw/virtio/vhost-shadow-virtqueue.c
-+++ b/hw/virtio/vhost-shadow-virtqueue.c
-@@ -138,6 +138,7 @@ static void vhost_vring_write_descs(VhostShadowVirtqueue *svq, hwaddr *sg,
- for (n = 0; n < num; n++) {
- if (more_descs || (n + 1 < num)) {
- descs[i].flags = flags | cpu_to_le16(VRING_DESC_F_NEXT);
-+ descs[i].next = cpu_to_le16(svq->desc_next[i]);
- } else {
- descs[i].flags = flags;
- }
-@@ -145,10 +146,10 @@ static void vhost_vring_write_descs(VhostShadowVirtqueue *svq, hwaddr *sg,
- descs[i].len = cpu_to_le32(iovec[n].iov_len);
-
- last = i;
-- i = cpu_to_le16(descs[i].next);
-+ i = cpu_to_le16(svq->desc_next[i]);
- }
-
-- svq->free_head = le16_to_cpu(descs[last].next);
-+ svq->free_head = le16_to_cpu(svq->desc_next[last]);
- }
-
- static bool vhost_svq_add_split(VhostShadowVirtqueue *svq,
-@@ -336,7 +337,6 @@ static void vhost_svq_disable_notification(VhostShadowVirtqueue *svq)
- static VirtQueueElement *vhost_svq_get_buf(VhostShadowVirtqueue *svq,
- uint32_t *len)
- {
-- vring_desc_t *descs = svq->vring.desc;
- const vring_used_t *used = svq->vring.used;
- vring_used_elem_t used_elem;
- uint16_t last_used;
-@@ -365,7 +365,7 @@ static VirtQueueElement *vhost_svq_get_buf(VhostShadowVirtqueue *svq,
- return NULL;
- }
-
-- descs[used_elem.id].next = svq->free_head;
-+ svq->desc_next[used_elem.id] = svq->free_head;
- svq->free_head = used_elem.id;
-
- *len = used_elem.len;
-@@ -540,8 +540,9 @@ void vhost_svq_start(VhostShadowVirtqueue *svq, VirtIODevice *vdev,
- svq->vring.used = qemu_memalign(qemu_real_host_page_size, device_size);
- memset(svq->vring.used, 0, device_size);
- svq->ring_id_maps = g_new0(VirtQueueElement *, svq->vring.num);
-+ svq->desc_next = g_new0(uint16_t, svq->vring.num);
- for (unsigned i = 0; i < svq->vring.num - 1; i++) {
-- svq->vring.desc[i].next = cpu_to_le16(i + 1);
-+ svq->desc_next[i] = cpu_to_le16(i + 1);
- }
- }
-
-@@ -574,6 +575,7 @@ void vhost_svq_stop(VhostShadowVirtqueue *svq)
- virtqueue_detach_element(svq->vq, next_avail_elem, 0);
- }
- svq->vq = NULL;
-+ g_free(svq->desc_next);
- g_free(svq->ring_id_maps);
- qemu_vfree(svq->vring.desc);
- qemu_vfree(svq->vring.used);
-diff --git a/hw/virtio/vhost-shadow-virtqueue.h b/hw/virtio/vhost-shadow-virtqueue.h
-index e5e24c536d..c132c994e9 100644
---- a/hw/virtio/vhost-shadow-virtqueue.h
-+++ b/hw/virtio/vhost-shadow-virtqueue.h
-@@ -53,6 +53,12 @@ typedef struct VhostShadowVirtqueue {
- /* Next VirtQueue element that guest made available */
- VirtQueueElement *next_guest_avail_elem;
-
-+ /*
-+ * Backup next field for each descriptor so we can recover securely, not
-+ * needing to trust the device access.
-+ */
-+ uint16_t *desc_next;
-+
- /* Next head to expose to the device */
- uint16_t shadow_avail_idx;
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
-Date: Thu, 12 May 2022 19:57:43 +0200
-Subject: [PATCH] vhost: Fix device's used descriptor dequeue
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Only the first one of them were properly enqueued back.
-
-Fixes: 100890f7ca ("vhost: Shadow virtqueue buffers forwarding")
-
-Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
-Message-Id: <20220512175747.142058-3-eperezma@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 81abfa5724c9a6502d7a1d3a67c55f2a303a1170)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-shadow-virtqueue.c | 17 +++++++++++++++--
- 1 file changed, 15 insertions(+), 2 deletions(-)
-
-diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-virtqueue.c
-index 3155801f50..31fc50907d 100644
---- a/hw/virtio/vhost-shadow-virtqueue.c
-+++ b/hw/virtio/vhost-shadow-virtqueue.c
-@@ -334,12 +334,22 @@ static void vhost_svq_disable_notification(VhostShadowVirtqueue *svq)
- svq->vring.avail->flags |= cpu_to_le16(VRING_AVAIL_F_NO_INTERRUPT);
- }
-
-+static uint16_t vhost_svq_last_desc_of_chain(const VhostShadowVirtqueue *svq,
-+ uint16_t num, uint16_t i)
-+{
-+ for (uint16_t j = 0; j < (num - 1); ++j) {
-+ i = le16_to_cpu(svq->desc_next[i]);
-+ }
-+
-+ return i;
-+}
-+
- static VirtQueueElement *vhost_svq_get_buf(VhostShadowVirtqueue *svq,
- uint32_t *len)
- {
- const vring_used_t *used = svq->vring.used;
- vring_used_elem_t used_elem;
-- uint16_t last_used;
-+ uint16_t last_used, last_used_chain, num;
-
- if (!vhost_svq_more_used(svq)) {
- return NULL;
-@@ -365,7 +375,10 @@ static VirtQueueElement *vhost_svq_get_buf(VhostShadowVirtqueue *svq,
- return NULL;
- }
-
-- svq->desc_next[used_elem.id] = svq->free_head;
-+ num = svq->ring_id_maps[used_elem.id]->in_num +
-+ svq->ring_id_maps[used_elem.id]->out_num;
-+ last_used_chain = vhost_svq_last_desc_of_chain(svq, num, used_elem.id);
-+ svq->desc_next[last_used_chain] = svq->free_head;
- svq->free_head = used_elem.id;
-
- *len = used_elem.len;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
-Date: Thu, 12 May 2022 19:57:44 +0200
-Subject: [PATCH] vdpa: Fix bad index calculus at vhost_vdpa_get_vring_base
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Fixes: 6d0b222666 ("vdpa: Adapt vhost_vdpa_get_vring_base to SVQ")
-
-Acked-by: Jason Wang <jasowang@redhat.com>
-Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
-Message-Id: <20220512175747.142058-4-eperezma@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 639036477ef890958415967e753ca2cbb348c16c)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-vdpa.c | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
-index 8adf7c0b92..8555a84f87 100644
---- a/hw/virtio/vhost-vdpa.c
-+++ b/hw/virtio/vhost-vdpa.c
-@@ -1170,11 +1170,11 @@ static int vhost_vdpa_get_vring_base(struct vhost_dev *dev,
- struct vhost_vring_state *ring)
- {
- struct vhost_vdpa *v = dev->opaque;
-+ int vdpa_idx = ring->index - dev->vq_index;
- int ret;
-
- if (v->shadow_vqs_enabled) {
-- VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs,
-- ring->index);
-+ VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, vdpa_idx);
-
- /*
- * Setting base as last used idx, so destination will see as available
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
-Date: Thu, 12 May 2022 19:57:45 +0200
-Subject: [PATCH] vdpa: Fix index calculus at vhost_vdpa_svqs_start
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-With the introduction of MQ the index of the vq needs to be calculated
-with the device model vq_index.
-
-Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
-Acked-by: Jason Wang <jasowang@redhat.com>
-Message-Id: <20220512175747.142058-5-eperezma@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 1c82fdfef8a227518ffecae9d419bcada995c202)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-vdpa.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
-index 8555a84f87..aa43cfa7d9 100644
---- a/hw/virtio/vhost-vdpa.c
-+++ b/hw/virtio/vhost-vdpa.c
-@@ -1016,7 +1016,7 @@ static bool vhost_vdpa_svqs_start(struct vhost_dev *dev)
- VirtQueue *vq = virtio_get_queue(dev->vdev, dev->vq_index + i);
- VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
- struct vhost_vring_addr addr = {
-- .index = i,
-+ .index = dev->vq_index + i,
- };
- int r;
- bool ok = vhost_vdpa_svq_setup(dev, svq, i, &err);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Philippe=20Mathieu-Daud=C3=A9?= <philmd@redhat.com>
-Date: Thu, 12 May 2022 19:57:46 +0200
-Subject: [PATCH] hw/virtio: Replace g_memdup() by g_memdup2()
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Per https://discourse.gnome.org/t/port-your-module-from-g-memdup-to-g-memdup2-now/5538
-
- The old API took the size of the memory to duplicate as a guint,
- whereas most memory functions take memory sizes as a gsize. This
- made it easy to accidentally pass a gsize to g_memdup(). For large
- values, that would lead to a silent truncation of the size from 64
- to 32 bits, and result in a heap area being returned which is
- significantly smaller than what the caller expects. This can likely
- be exploited in various modules to cause a heap buffer overflow.
-
-Replace g_memdup() by the safer g_memdup2() wrapper.
-
-Acked-by: Jason Wang <jasowang@redhat.com>
-Acked-by: Eugenio Pérez <eperezma@redhat.com>
-Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
-Message-Id: <20220512175747.142058-6-eperezma@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit d792199de55ca5cb5334016884039c740290b5c7)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/net/virtio-net.c | 3 ++-
- hw/virtio/virtio-crypto.c | 6 +++---
- 2 files changed, 5 insertions(+), 4 deletions(-)
-
-diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
-index 1067e72b39..e4748a7e6c 100644
---- a/hw/net/virtio-net.c
-+++ b/hw/net/virtio-net.c
-@@ -1443,7 +1443,8 @@ static void virtio_net_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
- }
-
- iov_cnt = elem->out_num;
-- iov2 = iov = g_memdup(elem->out_sg, sizeof(struct iovec) * elem->out_num);
-+ iov2 = iov = g_memdup2(elem->out_sg,
-+ sizeof(struct iovec) * elem->out_num);
- s = iov_to_buf(iov, iov_cnt, 0, &ctrl, sizeof(ctrl));
- iov_discard_front(&iov, &iov_cnt, sizeof(ctrl));
- if (s != sizeof(ctrl)) {
-diff --git a/hw/virtio/virtio-crypto.c b/hw/virtio/virtio-crypto.c
-index dcd80b904d..0e31e3cc04 100644
---- a/hw/virtio/virtio-crypto.c
-+++ b/hw/virtio/virtio-crypto.c
-@@ -242,7 +242,7 @@ static void virtio_crypto_handle_ctrl(VirtIODevice *vdev, VirtQueue *vq)
- }
-
- out_num = elem->out_num;
-- out_iov_copy = g_memdup(elem->out_sg, sizeof(out_iov[0]) * out_num);
-+ out_iov_copy = g_memdup2(elem->out_sg, sizeof(out_iov[0]) * out_num);
- out_iov = out_iov_copy;
-
- in_num = elem->in_num;
-@@ -605,11 +605,11 @@ virtio_crypto_handle_request(VirtIOCryptoReq *request)
- }
-
- out_num = elem->out_num;
-- out_iov_copy = g_memdup(elem->out_sg, sizeof(out_iov[0]) * out_num);
-+ out_iov_copy = g_memdup2(elem->out_sg, sizeof(out_iov[0]) * out_num);
- out_iov = out_iov_copy;
-
- in_num = elem->in_num;
-- in_iov_copy = g_memdup(elem->in_sg, sizeof(in_iov[0]) * in_num);
-+ in_iov_copy = g_memdup2(elem->in_sg, sizeof(in_iov[0]) * in_num);
- in_iov = in_iov_copy;
-
- if (unlikely(iov_to_buf(out_iov, out_num, 0, &req, sizeof(req))
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
-Date: Thu, 12 May 2022 19:57:47 +0200
-Subject: [PATCH] vhost: Fix element in vhost_svq_add failure
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Coverity rightly reports that is not free in that case.
-
-Fixes: Coverity CID 1487559
-Fixes: 100890f7ca ("vhost: Shadow virtqueue buffers forwarding")
-
-Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
-Message-Id: <20220512175747.142058-7-eperezma@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 5181db132b587754dda3a520eec923b87a65bbb7)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-shadow-virtqueue.c | 8 ++++++++
- 1 file changed, 8 insertions(+)
-
-diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-virtqueue.c
-index 31fc50907d..06d0bb39d9 100644
---- a/hw/virtio/vhost-shadow-virtqueue.c
-+++ b/hw/virtio/vhost-shadow-virtqueue.c
-@@ -199,11 +199,19 @@ static bool vhost_svq_add_split(VhostShadowVirtqueue *svq,
- return true;
- }
-
-+/**
-+ * Add an element to a SVQ.
-+ *
-+ * The caller must check that there is enough slots for the new element. It
-+ * takes ownership of the element: In case of failure, it is free and the SVQ
-+ * is considered broken.
-+ */
- static bool vhost_svq_add(VhostShadowVirtqueue *svq, VirtQueueElement *elem)
- {
- unsigned qemu_head;
- bool ok = vhost_svq_add_split(svq, elem, &qemu_head);
- if (unlikely(!ok)) {
-+ g_free(elem);
- return false;
- }
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dominique Martinet <dominique.martinet@atmark-techno.com>
-Date: Thu, 30 Jun 2022 10:01:37 +0900
-Subject: [PATCH] io_uring: fix short read slow path
-
-sqeq.off here is the offset to read within the disk image, so obviously
-not 'nread' (the amount we just read), but as the author meant to write
-its current value incremented by the amount we just read.
-
-Normally recent versions of linux will not issue short reads,
-but it can happen so we should fix this.
-
-This lead to weird image corruptions when short read happened
-
-Fixes: 6663a0a33764 ("block/io_uring: implements interfaces for io_uring")
-Link: https://lkml.kernel.org/r/YrrFGO4A1jS0GI0G@atmark-techno.com
-Signed-off-by: Dominique Martinet <dominique.martinet@atmark-techno.com>
-Message-Id: <20220630010137.2518851-1-dominique.martinet@atmark-techno.com>
-Reviewed-by: Hanna Reitz <hreitz@redhat.com>
-Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
-(cherry-picked from commit c06fc7ce147e57ab493bad9263f1601b8298484b)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/io_uring.c | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/block/io_uring.c b/block/io_uring.c
-index 782afdb433..e451e55de0 100644
---- a/block/io_uring.c
-+++ b/block/io_uring.c
-@@ -89,7 +89,7 @@ static void luring_resubmit_short_read(LuringState *s, LuringAIOCB *luringcb,
- trace_luring_resubmit_short_read(s, luringcb, nread);
-
- /* Update read position */
-- luringcb->total_read = nread;
-+ luringcb->total_read += nread;
- remaining = luringcb->qiov->size - luringcb->total_read;
-
- /* Shorten qiov */
-@@ -103,7 +103,7 @@ static void luring_resubmit_short_read(LuringState *s, LuringAIOCB *luringcb,
- remaining);
-
- /* Update sqe */
-- luringcb->sqeq.off = nread;
-+ luringcb->sqeq.off += nread;
- luringcb->sqeq.addr = (__u64)(uintptr_t)luringcb->resubmit_qiov.iov;
- luringcb->sqeq.len = luringcb->resubmit_qiov.niov;
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Ding Hui <dinghui@sangfor.com.cn>
-Date: Wed, 29 Jun 2022 17:40:26 +0800
-Subject: [PATCH] e1000: set RX descriptor status in a separate operation
-
-The code of setting RX descriptor status field maybe work fine in
-previously, however with the update of glibc version, it shows two
-issues when guest using dpdk receive packets:
-
- 1. The dpdk has a certain probability getting wrong buffer_addr
-
- this impact may be not obvious, such as lost a packet once in
- a while
-
- 2. The dpdk may consume a packet twice when scan the RX desc queue
- over again
-
- this impact will lead a infinite wait in Qemu, since the RDT
- (tail pointer) be inscreased to equal to RDH by unexpected,
- which regard as the RX desc queue is full
-
-Write a whole of RX desc with DD flag on is not quite correct, because
-when the underlying implementation of memcpy using XMM registers to
-copy e1000_rx_desc (when AVX or something else CPU feature is usable),
-the bytes order of desc writing to memory is indeterminacy
-
-We can use full-scale test case to reproduce the issue-2 by
-https://github.com/BASM/qemu_dpdk_e1000_test (thanks to Leonid Myravjev)
-
-I also write a POC test case at https://github.com/cdkey/e1000_poc
-which can reproduce both of them, and easy to verify the patch effect.
-
-The hw watchpoint also shows that, when Qemu using XMM related instructions
-writing 16 bytes e1000_rx_desc, concurrent with DPDK using movb
-writing 1 byte status, the final result of writing to memory will be one
-of them, if it made by Qemu which DD flag is on, DPDK will consume it
-again.
-
-Setting DD status in a separate operation, can prevent the impact of
-disorder memory writing by memcpy, also avoid unexpected data when
-concurrent writing status by qemu and guest dpdk.
-
-Links: https://lore.kernel.org/qemu-devel/20200102110504.GG121208@stefanha-x1.localdomain/T/
-
-Reported-by: Leonid Myravjev <asm@asm.pp.ru>
-Cc: Stefan Hajnoczi <stefanha@gmail.com>
-Cc: Paolo Bonzini <pbonzini@redhat.com>
-Cc: Michael S. Tsirkin <mst@redhat.com>
-Cc: qemu-stable@nongnu.org
-Tested-by: Jing Zhang <zhangjing@sangfor.com.cn>
-Reviewed-by: Frank Lee <lifan38153@sangfor.com.cn>
-Signed-off-by: Ding Hui <dinghui@sangfor.com.cn>
-Signed-off-by: Jason Wang <jasowang@redhat.com>
-(cherry-picked from commit 034d00d4858161e1d4cff82d8d230bce874a04d3)
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- hw/net/e1000.c | 5 ++++-
- 1 file changed, 4 insertions(+), 1 deletion(-)
-
-diff --git a/hw/net/e1000.c b/hw/net/e1000.c
-index f5bc81296d..e26e0a64c1 100644
---- a/hw/net/e1000.c
-+++ b/hw/net/e1000.c
-@@ -979,7 +979,7 @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
- base = rx_desc_base(s) + sizeof(desc) * s->mac_reg[RDH];
- pci_dma_read(d, base, &desc, sizeof(desc));
- desc.special = vlan_special;
-- desc.status |= (vlan_status | E1000_RXD_STAT_DD);
-+ desc.status &= ~E1000_RXD_STAT_DD;
- if (desc.buffer_addr) {
- if (desc_offset < size) {
- size_t iov_copy;
-@@ -1013,6 +1013,9 @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
- DBGOUT(RX, "Null RX descriptor!!\n");
- }
- pci_dma_write(d, base, &desc, sizeof(desc));
-+ desc.status |= (vlan_status | E1000_RXD_STAT_DD);
-+ pci_dma_write(d, base + offsetof(struct e1000_rx_desc, status),
-+ &desc.status, sizeof(desc.status));
-
- if (++s->mac_reg[RDH] * sizeof(desc) >= s->mac_reg[RDLEN])
- s->mac_reg[RDH] = 0;
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
-index 39a3d6dbe6..e5bf5d59bf 100644
+index 48cd096624..3d60b80286 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
-@@ -554,7 +554,7 @@ static QemuOptsList raw_runtime_opts = {
+@@ -553,7 +553,7 @@ static QemuOptsList raw_runtime_opts = {
{
.name = "locking",
.type = QEMU_OPT_STRING,
},
{
.name = "pr-manager",
-@@ -654,7 +654,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
+@@ -653,7 +653,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
s->use_lock = false;
break;
case ON_OFF_AUTO_AUTO:
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
-index 982c532353..5395e7e471 100644
+index 82004b65b9..4868db8f94 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
-@@ -2087,9 +2087,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
+@@ -2133,9 +2133,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
#define CPU_RESOLVING_TYPE TYPE_X86_CPU
#ifdef TARGET_X86_64
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/block/gluster.c b/block/gluster.c
-index 592e71b22a..aebfece6eb 100644
+index b60213ab80..93da76bc31 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -42,7 +42,7 @@
1 file changed, 2 insertions(+)
diff --git a/block/rbd.c b/block/rbd.c
-index 6caf35cbba..0cec24c86d 100644
+index f826410f40..64a8d7d48b 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -820,6 +820,8 @@ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
3 files changed, 43 insertions(+)
diff --git a/net/net.c b/net/net.c
-index f0d14dbfc1..6d476c47ef 100644
+index 2db160e063..8329347891 100644
--- a/net/net.c
+++ b/net/net.c
-@@ -1334,6 +1334,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
+@@ -1343,6 +1343,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
}
}
{
NetClientState *nc;
diff --git a/qapi/net.json b/qapi/net.json
-index b92f3f5fb4..52c7e1f82b 100644
+index 75ba2cb989..a3c93ab88f 100644
--- a/qapi/net.json
+++ b/qapi/net.json
@@ -35,6 +35,21 @@
# @netdev_add:
#
diff --git a/qapi/pragma.json b/qapi/pragma.json
-index e6a021c19c..41139d8645 100644
+index 7f810b0e97..a2358e303a 100644
--- a/qapi/pragma.json
+++ b/qapi/pragma.json
-@@ -25,6 +25,7 @@
- 'system_reset',
+@@ -26,6 +26,7 @@
'system_wakeup' ],
+ # Commands allowed to return a non-dictionary
'command-returns-exceptions': [
+ 'get_link_status',
'human-monitor-command',
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/block/gluster.c b/block/gluster.c
-index aebfece6eb..3b7ee2f649 100644
+index 93da76bc31..1079b6186b 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -57,6 +57,7 @@ typedef struct GlusterAIOCB {
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/qemu-img.c b/qemu-img.c
-index 1caddfb23a..2782e181c0 100644
+index 7d4b33b3da..bb36f42dd2 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -3005,7 +3005,8 @@ static int img_info(int argc, char **argv)
+@@ -3010,7 +3010,8 @@ static int img_info(int argc, char **argv)
list = collect_image_info_list(image_opts, filename, fmt, chain,
force_share);
if (!list) {
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
---
qemu-img-cmds.hx | 4 +-
- qemu-img.c | 187 +++++++++++++++++++++++++++++------------------
- 2 files changed, 119 insertions(+), 72 deletions(-)
+ qemu-img.c | 202 ++++++++++++++++++++++++++++++-----------------
+ 2 files changed, 133 insertions(+), 73 deletions(-)
diff --git a/qemu-img-cmds.hx b/qemu-img-cmds.hx
index 1b1dab5b17..d1616c045a 100644
DEF("info", img_info,
diff --git a/qemu-img.c b/qemu-img.c
-index 2782e181c0..8cd43b3601 100644
+index bb36f42dd2..74afcb79ef 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -4821,10 +4821,12 @@ static int img_bitmap(int argc, char **argv)
+@@ -4826,10 +4826,12 @@ static int img_bitmap(int argc, char **argv)
#define C_IF 04
#define C_OF 010
#define C_SKIP 020
};
struct DdIo {
-@@ -4900,6 +4902,19 @@ static int img_dd_skip(const char *arg,
+@@ -4905,6 +4907,19 @@ static int img_dd_skip(const char *arg,
return 0;
}
static int img_dd(int argc, char **argv)
{
int ret = 0;
-@@ -4940,6 +4955,7 @@ static int img_dd(int argc, char **argv)
+@@ -4945,6 +4960,7 @@ static int img_dd(int argc, char **argv)
{ "if", img_dd_if, C_IF },
{ "of", img_dd_of, C_OF },
{ "skip", img_dd_skip, C_SKIP },
{ NULL, NULL, 0 }
};
const struct option long_options[] = {
-@@ -5015,91 +5031,112 @@ static int img_dd(int argc, char **argv)
+@@ -5020,91 +5036,112 @@ static int img_dd(int argc, char **argv)
arg = NULL;
}
}
- create_opts = qemu_opts_append(create_opts, drv->create_opts);
- create_opts = qemu_opts_append(create_opts, proto_drv->create_opts);
-
-- opts = qemu_opts_create(create_opts, NULL, 0, &error_abort);
-
+- opts = qemu_opts_create(create_opts, NULL, 0, &error_abort);
+
- size = blk_getlength(blk1);
- if (size < 0) {
- error_report("Failed to get size for '%s'", in.filename);
}
if (dd.flags & C_SKIP && (in.offset > INT64_MAX / in.bsz ||
-@@ -5117,11 +5154,17 @@ static int img_dd(int argc, char **argv)
+@@ -5121,20 +5158,43 @@ static int img_dd(int argc, char **argv)
+ in.buf = g_new(uint8_t, in.bsz);
for (out_pos = 0; in_pos < size; block_count++) {
- int in_ret, out_ret;
++ int in_ret, out_ret;
+ int bytes = (in_pos + in.bsz > size) ? size - in_pos : in.bsz;
-
-- if (in_pos + in.bsz > size) {
-- in_ret = blk_pread(blk1, in_pos, in.buf, size - in_pos);
-+ size_t in_bsz = in_pos + in.bsz > size ? size - in_pos : in.bsz;
+- ret = blk_pread(blk1, in_pos, bytes, in.buf, 0);
+- if (ret < 0) {
+ if (blk1) {
-+ in_ret = blk_pread(blk1, in_pos, in.buf, in_bsz);
- } else {
-- in_ret = blk_pread(blk1, in_pos, in.buf, in.bsz);
-+ in_ret = read(STDIN_FILENO, in.buf, in_bsz);
++ in_ret = blk_pread(blk1, in_pos, bytes, in.buf, 0);
++ if (in_ret == 0) {
++ in_ret = bytes;
++ }
++ } else {
++ in_ret = read(STDIN_FILENO, in.buf, bytes);
+ if (in_ret == 0) {
+ /* early EOF is considered an error */
+ error_report("Input ended unexpectedly");
+ ret = -1;
+ goto out;
+ }
- }
- if (in_ret < 0) {
++ }
++ if (in_ret < 0) {
error_report("error while reading from input image file: %s",
-@@ -5131,9 +5174,13 @@ static int img_dd(int argc, char **argv)
+- strerror(-ret));
++ strerror(-in_ret));
++ ret = -1;
+ goto out;
}
- in_pos += in_ret;
+ in_pos += bytes;
-- out_ret = blk_pwrite(blk2, out_pos, in.buf, in_ret, 0);
+- ret = blk_pwrite(blk2, out_pos, bytes, in.buf, 0);
+- if (ret < 0) {
+ if (blk2) {
-+ out_ret = blk_pwrite(blk2, out_pos, in.buf, in_ret, 0);
++ out_ret = blk_pwrite(blk2, out_pos, in_ret, in.buf, 0);
++ if (out_ret == 0) {
++ out_ret = in_ret;
++ }
+ } else {
+ out_ret = write(STDOUT_FILENO, in.buf, in_ret);
+ }
-
-- if (out_ret < 0) {
++
+ if (out_ret != in_ret) {
error_report("error while writing to output image file: %s",
- strerror(-out_ret));
- ret = -1;
+- strerror(-ret));
++ strerror(-out_ret));
++ ret = -1;
+ goto out;
+ }
+ out_pos += bytes;
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
---
qemu-img.c | 28 +++++++++++++++++++++++++---
1 file changed, 25 insertions(+), 3 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
-index 8cd43b3601..67033b2d2c 100644
+index 74afcb79ef..14594d44b6 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -4822,11 +4822,13 @@ static int img_bitmap(int argc, char **argv)
+@@ -4827,11 +4827,13 @@ static int img_bitmap(int argc, char **argv)
#define C_OF 010
#define C_SKIP 020
#define C_OSIZE 040
};
struct DdIo {
-@@ -4915,6 +4917,19 @@ static int img_dd_osize(const char *arg,
+@@ -4920,6 +4922,19 @@ static int img_dd_osize(const char *arg,
return 0;
}
static int img_dd(int argc, char **argv)
{
int ret = 0;
-@@ -4929,12 +4944,14 @@ static int img_dd(int argc, char **argv)
+@@ -4934,12 +4949,14 @@ static int img_dd(int argc, char **argv)
int c, i;
const char *out_fmt = "raw";
const char *fmt = NULL;
};
struct DdIo in = {
.bsz = 512, /* Block size is by default 512 bytes */
-@@ -4956,6 +4973,7 @@ static int img_dd(int argc, char **argv)
+@@ -4961,6 +4978,7 @@ static int img_dd(int argc, char **argv)
{ "of", img_dd_of, C_OF },
{ "skip", img_dd_skip, C_SKIP },
{ "osize", img_dd_osize, C_OSIZE },
{ NULL, NULL, 0 }
};
const struct option long_options[] = {
-@@ -5152,14 +5170,18 @@ static int img_dd(int argc, char **argv)
+@@ -5157,9 +5175,10 @@ static int img_dd(int argc, char **argv)
in.buf = g_new(uint8_t, in.bsz);
+ readsize = (dd.isize > 0) ? dd.isize : size;
+ for (out_pos = 0; in_pos < readsize; block_count++) {
int in_ret, out_ret;
-- size_t in_bsz = in_pos + in.bsz > size ? size - in_pos : in.bsz;
-+ size_t in_bsz = in_pos + in.bsz > readsize ? readsize - in_pos : in.bsz;
+- int bytes = (in_pos + in.bsz > size) ? size - in_pos : in.bsz;
++ int bytes = (in_pos + in.bsz > readsize) ? readsize - in_pos : in.bsz;
if (blk1) {
- in_ret = blk_pread(blk1, in_pos, in.buf, in_bsz);
+ in_ret = blk_pread(blk1, in_pos, bytes, in.buf, 0);
+ if (in_ret == 0) {
+@@ -5168,6 +5187,9 @@ static int img_dd(int argc, char **argv)
} else {
- in_ret = read(STDIN_FILENO, in.buf, in_bsz);
+ in_ret = read(STDIN_FILENO, in.buf, bytes);
if (in_ret == 0) {
+ if (dd.isize == 0) {
+ goto out;
3 files changed, 26 insertions(+), 12 deletions(-)
diff --git a/docs/tools/qemu-img.rst b/docs/tools/qemu-img.rst
-index 8885ea11cf..33979b7430 100644
+index 85a6e05b35..699229eef6 100644
--- a/docs/tools/qemu-img.rst
+++ b/docs/tools/qemu-img.rst
@@ -208,6 +208,10 @@ Parameters to convert subcommand:
DEF("info", img_info,
diff --git a/qemu-img.c b/qemu-img.c
-index 67033b2d2c..35c2bdc95c 100644
+index 14594d44b6..c6b4a5567d 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -4946,7 +4946,7 @@ static int img_dd(int argc, char **argv)
+@@ -4951,7 +4951,7 @@ static int img_dd(int argc, char **argv)
const char *fmt = NULL;
int64_t size = 0, readsize = 0;
int64_t block_count = 0, out_pos, in_pos;
struct DdInfo dd = {
.flags = 0,
.count = 0,
-@@ -4984,7 +4984,7 @@ static int img_dd(int argc, char **argv)
+@@ -4989,7 +4989,7 @@ static int img_dd(int argc, char **argv)
{ 0, 0, 0, 0 }
};
if (c == EOF) {
break;
}
-@@ -5004,6 +5004,9 @@ static int img_dd(int argc, char **argv)
+@@ -5009,6 +5009,9 @@ static int img_dd(int argc, char **argv)
case 'h':
help();
break;
case 'U':
force_share = true;
break;
-@@ -5134,13 +5137,15 @@ static int img_dd(int argc, char **argv)
+@@ -5139,13 +5142,15 @@ static int img_dd(int argc, char **argv)
size - in.bsz * in.offset, &error_abort);
}
3 files changed, 81 insertions(+), 4 deletions(-)
diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
-index 163d244eb4..389907f1f8 100644
+index 73ac5eb675..bbfe7eca62 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
-@@ -813,8 +813,37 @@ static uint64_t virtio_balloon_get_features(VirtIODevice *vdev, uint64_t f,
+@@ -806,8 +806,37 @@ static uint64_t virtio_balloon_get_features(VirtIODevice *vdev, uint64_t f,
static void virtio_balloon_stat(void *opaque, BalloonInfo *info)
{
VirtIOBalloon *dev = opaque;
static void virtio_balloon_to_target(void *opaque, ram_addr_t target)
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 634968498b..5482dd0569 100644
+index c6cd6f91dd..15572befb1 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
-@@ -708,7 +708,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
+@@ -715,7 +715,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
return;
}
qapi_free_BalloonInfo(info);
}
diff --git a/qapi/machine.json b/qapi/machine.json
-index d25a481ce4..3627172aed 100644
+index 6afd1936b0..8b4be9b718 100644
--- a/qapi/machine.json
+++ b/qapi/machine.json
-@@ -1018,10 +1018,30 @@
+@@ -1054,9 +1054,29 @@
# @actual: the logical size of the VM in bytes
# Formula used: logical_vm_size = vm_ram_size - balloon_size
#
+# @max_mem: amount of memory (in bytes) assigned to the guest
+#
# Since: 0.14
- #
##
-{ 'struct': 'BalloonInfo', 'data': {'actual': 'int' } }
+{ 'struct': 'BalloonInfo',
info->default_cpu_type = g_strdup(mc->default_cpu_type);
info->has_default_cpu_type = true;
diff --git a/qapi/machine.json b/qapi/machine.json
-index 3627172aed..ca133e68ce 100644
+index 8b4be9b718..555458f785 100644
--- a/qapi/machine.json
+++ b/qapi/machine.json
-@@ -139,6 +139,8 @@
+@@ -138,6 +138,8 @@
#
# @is-default: whether the machine is default
#
# @cpu-max: maximum number of CPUs supported by the machine type
# (since 1.5)
#
-@@ -160,7 +162,7 @@
+@@ -159,7 +161,7 @@
##
{ 'struct': 'MachineInfo',
'data': { 'name': 'str', '*alias': 'str',
2 files changed, 8 insertions(+)
diff --git a/qapi/ui.json b/qapi/ui.json
-index 13a8bb82aa..4c34436ba7 100644
+index cf58ab4283..0be2388941 100644
--- a/qapi/ui.json
+++ b/qapi/ui.json
-@@ -300,11 +300,14 @@
+@@ -310,11 +310,14 @@
#
# @channels: a list of @SpiceChannel for each active spice channel
#
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Thu, 13 Oct 2022 11:33:50 +0200
+Subject: [PATCH] PVE: add IOChannel implementation for savevm-async
+
+based on migration/channel-block.c and the implementation that was
+present in migration/savevm-async.c before QEMU 7.1.
+
+Passes along read/write requests to the given BlockBackend, while
+ensuring that a read request going beyond the end results in a
+graceful short read.
+
+Additionally, allows tracking the current position from the outside
+(intended to be used for progress tracking).
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ migration/channel-savevm-async.c | 181 +++++++++++++++++++++++++++++++
+ migration/channel-savevm-async.h | 51 +++++++++
+ migration/meson.build | 1 +
+ 3 files changed, 233 insertions(+)
+ create mode 100644 migration/channel-savevm-async.c
+ create mode 100644 migration/channel-savevm-async.h
+
+diff --git a/migration/channel-savevm-async.c b/migration/channel-savevm-async.c
+new file mode 100644
+index 0000000000..efea3842cf
+--- /dev/null
++++ b/migration/channel-savevm-async.c
+@@ -0,0 +1,181 @@
++/*
++ * QIO Channel implementation to be used by savevm-async QMP calls
++ */
++#include "qemu/osdep.h"
++#include "migration/channel-savevm-async.h"
++#include "qapi/error.h"
++#include "sysemu/block-backend.h"
++#include "trace.h"
++
++QIOChannelSavevmAsync *
++qio_channel_savevm_async_new(BlockBackend *be, size_t *bs_pos)
++{
++ QIOChannelSavevmAsync *ioc;
++
++ ioc = QIO_CHANNEL_SAVEVM_ASYNC(object_new(TYPE_QIO_CHANNEL_SAVEVM_ASYNC));
++
++ bdrv_ref(blk_bs(be));
++ ioc->be = be;
++ ioc->bs_pos = bs_pos;
++
++ return ioc;
++}
++
++
++static void
++qio_channel_savevm_async_finalize(Object *obj)
++{
++ QIOChannelSavevmAsync *ioc = QIO_CHANNEL_SAVEVM_ASYNC(obj);
++
++ if (ioc->be) {
++ bdrv_unref(blk_bs(ioc->be));
++ ioc->be = NULL;
++ }
++ ioc->bs_pos = NULL;
++}
++
++
++static ssize_t
++qio_channel_savevm_async_readv(QIOChannel *ioc,
++ const struct iovec *iov,
++ size_t niov,
++ int **fds,
++ size_t *nfds,
++ Error **errp)
++{
++ QIOChannelSavevmAsync *saioc = QIO_CHANNEL_SAVEVM_ASYNC(ioc);
++ BlockBackend *be = saioc->be;
++ int64_t maxlen = blk_getlength(be);
++ QEMUIOVector qiov;
++ size_t size;
++ int ret;
++
++ qemu_iovec_init_external(&qiov, (struct iovec *)iov, niov);
++
++ if (*saioc->bs_pos >= maxlen) {
++ error_setg(errp, "cannot read beyond maxlen");
++ return -1;
++ }
++
++ if (maxlen - *saioc->bs_pos < qiov.size) {
++ size = maxlen - *saioc->bs_pos;
++ } else {
++ size = qiov.size;
++ }
++
++ // returns 0 on success
++ ret = blk_preadv(be, *saioc->bs_pos, size, &qiov, 0);
++ if (ret < 0) {
++ error_setg(errp, "blk_preadv returned %d", ret);
++ return -1;
++ }
++
++ *saioc->bs_pos += size;
++ return size;
++}
++
++
++static ssize_t
++qio_channel_savevm_async_writev(QIOChannel *ioc,
++ const struct iovec *iov,
++ size_t niov,
++ int *fds,
++ size_t nfds,
++ int flags,
++ Error **errp)
++{
++ QIOChannelSavevmAsync *saioc = QIO_CHANNEL_SAVEVM_ASYNC(ioc);
++ BlockBackend *be = saioc->be;
++ QEMUIOVector qiov;
++ int ret;
++
++ qemu_iovec_init_external(&qiov, (struct iovec *)iov, niov);
++
++ if (qemu_in_coroutine()) {
++ ret = blk_co_pwritev(be, *saioc->bs_pos, qiov.size, &qiov, 0);
++ aio_wait_kick();
++ } else {
++ ret = blk_pwritev(be, *saioc->bs_pos, qiov.size, &qiov, 0);
++ }
++
++ if (ret < 0) {
++ return ret;
++ }
++
++ *saioc->bs_pos += qiov.size;
++ return qiov.size;
++}
++
++
++static int
++qio_channel_savevm_async_set_blocking(QIOChannel *ioc,
++ bool enabled,
++ Error **errp)
++{
++ if (!enabled) {
++ error_setg(errp, "Non-blocking mode not supported for savevm-async");
++ return -1;
++ }
++ return 0;
++}
++
++
++static int
++qio_channel_savevm_async_close(QIOChannel *ioc,
++ Error **errp)
++{
++ QIOChannelSavevmAsync *saioc = QIO_CHANNEL_SAVEVM_ASYNC(ioc);
++ int rv = bdrv_flush(blk_bs(saioc->be));
++
++ if (rv < 0) {
++ error_setg_errno(errp, -rv, "Unable to flush VMState");
++ return -1;
++ }
++
++ bdrv_unref(blk_bs(saioc->be));
++ saioc->be = NULL;
++ saioc->bs_pos = NULL;
++
++ return 0;
++}
++
++
++static void
++qio_channel_savevm_async_set_aio_fd_handler(QIOChannel *ioc,
++ AioContext *ctx,
++ IOHandler *io_read,
++ IOHandler *io_write,
++ void *opaque)
++{
++ // if channel-block starts doing something, check if this needs adaptation
++}
++
++
++static void
++qio_channel_savevm_async_class_init(ObjectClass *klass,
++ void *class_data G_GNUC_UNUSED)
++{
++ QIOChannelClass *ioc_klass = QIO_CHANNEL_CLASS(klass);
++
++ ioc_klass->io_writev = qio_channel_savevm_async_writev;
++ ioc_klass->io_readv = qio_channel_savevm_async_readv;
++ ioc_klass->io_set_blocking = qio_channel_savevm_async_set_blocking;
++ ioc_klass->io_close = qio_channel_savevm_async_close;
++ ioc_klass->io_set_aio_fd_handler = qio_channel_savevm_async_set_aio_fd_handler;
++}
++
++static const TypeInfo qio_channel_savevm_async_info = {
++ .parent = TYPE_QIO_CHANNEL,
++ .name = TYPE_QIO_CHANNEL_SAVEVM_ASYNC,
++ .instance_size = sizeof(QIOChannelSavevmAsync),
++ .instance_finalize = qio_channel_savevm_async_finalize,
++ .class_init = qio_channel_savevm_async_class_init,
++};
++
++static void
++qio_channel_savevm_async_register_types(void)
++{
++ type_register_static(&qio_channel_savevm_async_info);
++}
++
++type_init(qio_channel_savevm_async_register_types);
+diff --git a/migration/channel-savevm-async.h b/migration/channel-savevm-async.h
+new file mode 100644
+index 0000000000..17ae2cb261
+--- /dev/null
++++ b/migration/channel-savevm-async.h
+@@ -0,0 +1,51 @@
++/*
++ * QEMU I/O channels driver for savevm-async.c
++ *
++ * Copyright (c) 2022 Proxmox Server Solutions
++ *
++ * Authors:
++ * Fiona Ebner (f.ebner@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ */
++
++#ifndef QIO_CHANNEL_SAVEVM_ASYNC_H
++#define QIO_CHANNEL_SAVEVM_ASYNC_H
++
++#include "io/channel.h"
++#include "qom/object.h"
++
++#define TYPE_QIO_CHANNEL_SAVEVM_ASYNC "qio-channel-savevm-async"
++OBJECT_DECLARE_SIMPLE_TYPE(QIOChannelSavevmAsync, QIO_CHANNEL_SAVEVM_ASYNC)
++
++
++/**
++ * QIOChannelSavevmAsync:
++ *
++ * The QIOChannelBlock object provides a channel implementation that is able to
++ * perform I/O on any BlockBackend whose BlockDriverState directly contains a
++ * VMState (as opposed to indirectly, like qcow2). It allows tracking the
++ * current position from the outside.
++ */
++struct QIOChannelSavevmAsync {
++ QIOChannel parent;
++ BlockBackend *be;
++ size_t *bs_pos;
++};
++
++
++/**
++ * qio_channel_savevm_async_new:
++ * @be: the block backend
++ * @bs_pos: used to keep track of the IOChannels current position
++ *
++ * Create a new IO channel object that can perform I/O on a BlockBackend object
++ * whose BlockDriverState directly contains a VMState.
++ *
++ * Returns: the new channel object
++ */
++QIOChannelSavevmAsync *
++qio_channel_savevm_async_new(BlockBackend *be, size_t *bs_pos);
++
++#endif /* QIO_CHANNEL_SAVEVM_ASYNC_H */
+diff --git a/migration/meson.build b/migration/meson.build
+index 690487cf1a..8cac83c06c 100644
+--- a/migration/meson.build
++++ b/migration/meson.build
+@@ -13,6 +13,7 @@ softmmu_ss.add(files(
+ 'block-dirty-bitmap.c',
+ 'channel.c',
+ 'channel-block.c',
++ 'channel-savevm-async.c',
+ 'colo-failover.c',
+ 'colo.c',
+ 'exec.c',
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:46 +0200
-Subject: [PATCH] PVE: add savevm-async for background state snapshots
-
-Put qemu_savevm_state_{header,setup} into the main loop and the rest
-of the iteration into a coroutine. The former need to lock the
-iothread (and we can't unlock it in the coroutine), and the latter
-can't deal with being in a separate thread, so a coroutine it must
-be.
-
-Truncate output file at 1024 boundary.
-
-Do not block the VM and save the state on aborting a snapshot, as the
-snapshot will be invalid anyway.
-
-Also, when aborting, wait for the target file to be closed, otherwise a
-client might run into race-conditions when trying to remove the file
-still opened by QEMU.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-[improve aborting]
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-[FE: further improve aborting]
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hmp-commands-info.hx | 13 +
- hmp-commands.hx | 33 ++
- include/migration/snapshot.h | 2 +
- include/monitor/hmp.h | 5 +
- migration/meson.build | 1 +
- migration/savevm-async.c | 604 +++++++++++++++++++++++++++++++++++
- monitor/hmp-cmds.c | 57 ++++
- qapi/migration.json | 34 ++
- qapi/misc.json | 32 ++
- qemu-options.hx | 12 +
- softmmu/vl.c | 10 +
- 11 files changed, 803 insertions(+)
- create mode 100644 migration/savevm-async.c
-
-diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
-index adfa085a9b..925e680e5a 100644
---- a/hmp-commands-info.hx
-+++ b/hmp-commands-info.hx
-@@ -538,6 +538,19 @@ SRST
- Show current migration parameters.
- ERST
-
-+ {
-+ .name = "savevm",
-+ .args_type = "",
-+ .params = "",
-+ .help = "show savevm status",
-+ .cmd = hmp_info_savevm,
-+ },
-+
-+SRST
-+ ``info savevm``
-+ Show savevm status.
-+ERST
-+
- {
- .name = "balloon",
- .args_type = "",
-diff --git a/hmp-commands.hx b/hmp-commands.hx
-index 8476277aa9..7f0ac498c4 100644
---- a/hmp-commands.hx
-+++ b/hmp-commands.hx
-@@ -1746,3 +1746,36 @@ ERST
- "\n\t\t\t -b to specify dirty bitmap as method of calculation)",
- .cmd = hmp_calc_dirty_rate,
- },
-+
-+ {
-+ .name = "savevm-start",
-+ .args_type = "statefile:s?",
-+ .params = "[statefile]",
-+ .help = "Prepare for snapshot and halt VM. Save VM state to statefile.",
-+ .cmd = hmp_savevm_start,
-+ },
-+
-+ {
-+ .name = "snapshot-drive",
-+ .args_type = "device:s,name:s",
-+ .params = "device name",
-+ .help = "Create internal snapshot.",
-+ .cmd = hmp_snapshot_drive,
-+ },
-+
-+ {
-+ .name = "delete-drive-snapshot",
-+ .args_type = "device:s,name:s",
-+ .params = "device name",
-+ .help = "Delete internal snapshot.",
-+ .cmd = hmp_delete_drive_snapshot,
-+ },
-+
-+ {
-+ .name = "savevm-end",
-+ .args_type = "",
-+ .params = "",
-+ .help = "Resume VM after snaphot.",
-+ .cmd = hmp_savevm_end,
-+ .coroutine = true,
-+ },
-diff --git a/include/migration/snapshot.h b/include/migration/snapshot.h
-index e72083b117..c846d37806 100644
---- a/include/migration/snapshot.h
-+++ b/include/migration/snapshot.h
-@@ -61,4 +61,6 @@ bool delete_snapshot(const char *name,
- bool has_devices, strList *devices,
- Error **errp);
-
-+int load_snapshot_from_blockdev(const char *filename, Error **errp);
-+
- #endif
-diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
-index 96d014826a..3a39ba41b5 100644
---- a/include/monitor/hmp.h
-+++ b/include/monitor/hmp.h
-@@ -26,6 +26,7 @@ void hmp_info_status(Monitor *mon, const QDict *qdict);
- void hmp_info_uuid(Monitor *mon, const QDict *qdict);
- void hmp_info_chardev(Monitor *mon, const QDict *qdict);
- void hmp_info_mice(Monitor *mon, const QDict *qdict);
-+void hmp_info_savevm(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_capabilities(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict);
-@@ -80,6 +81,10 @@ void hmp_netdev_add(Monitor *mon, const QDict *qdict);
- void hmp_netdev_del(Monitor *mon, const QDict *qdict);
- void hmp_getfd(Monitor *mon, const QDict *qdict);
- void hmp_closefd(Monitor *mon, const QDict *qdict);
-+void hmp_savevm_start(Monitor *mon, const QDict *qdict);
-+void hmp_snapshot_drive(Monitor *mon, const QDict *qdict);
-+void hmp_delete_drive_snapshot(Monitor *mon, const QDict *qdict);
-+void hmp_savevm_end(Monitor *mon, const QDict *qdict);
- void hmp_sendkey(Monitor *mon, const QDict *qdict);
- void hmp_screendump(Monitor *mon, const QDict *qdict);
- void hmp_chardev_add(Monitor *mon, const QDict *qdict);
-diff --git a/migration/meson.build b/migration/meson.build
-index 8b5ca5c047..1e2aec8486 100644
---- a/migration/meson.build
-+++ b/migration/meson.build
-@@ -23,6 +23,7 @@ softmmu_ss.add(files(
- 'multifd-zlib.c',
- 'postcopy-ram.c',
- 'savevm.c',
-+ 'savevm-async.c',
- 'socket.c',
- 'tls.c',
- ), gnutls)
-diff --git a/migration/savevm-async.c b/migration/savevm-async.c
-new file mode 100644
-index 0000000000..b9a43c56bc
---- /dev/null
-+++ b/migration/savevm-async.c
-@@ -0,0 +1,604 @@
-+#include "qemu/osdep.h"
-+#include "migration/migration.h"
-+#include "migration/savevm.h"
-+#include "migration/snapshot.h"
-+#include "migration/global_state.h"
-+#include "migration/ram.h"
-+#include "migration/qemu-file.h"
-+#include "sysemu/sysemu.h"
-+#include "sysemu/runstate.h"
-+#include "block/block.h"
-+#include "sysemu/block-backend.h"
-+#include "qapi/error.h"
-+#include "qapi/qmp/qerror.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qapi-commands-migration.h"
-+#include "qapi/qapi-commands-misc.h"
-+#include "qapi/qapi-commands-block.h"
-+#include "qemu/cutils.h"
-+#include "qemu/timer.h"
-+#include "qemu/main-loop.h"
-+#include "qemu/rcu.h"
-+
-+/* #define DEBUG_SAVEVM_STATE */
-+
-+/* used while emulated sync operation in progress */
-+#define NOT_DONE -EINPROGRESS
-+
-+#ifdef DEBUG_SAVEVM_STATE
-+#define DPRINTF(fmt, ...) \
-+ do { printf("savevm-async: " fmt, ## __VA_ARGS__); } while (0)
-+#else
-+#define DPRINTF(fmt, ...) \
-+ do { } while (0)
-+#endif
-+
-+enum {
-+ SAVE_STATE_DONE,
-+ SAVE_STATE_ERROR,
-+ SAVE_STATE_ACTIVE,
-+ SAVE_STATE_COMPLETED,
-+ SAVE_STATE_CANCELLED
-+};
-+
-+
-+static struct SnapshotState {
-+ BlockBackend *target;
-+ size_t bs_pos;
-+ int state;
-+ Error *error;
-+ Error *blocker;
-+ int saved_vm_running;
-+ QEMUFile *file;
-+ int64_t total_time;
-+ QEMUBH *finalize_bh;
-+ Coroutine *co;
-+ QemuCoSleep target_close_wait;
-+} snap_state;
-+
-+static bool savevm_aborted(void)
-+{
-+ return snap_state.state == SAVE_STATE_CANCELLED ||
-+ snap_state.state == SAVE_STATE_ERROR;
-+}
-+
-+SaveVMInfo *qmp_query_savevm(Error **errp)
-+{
-+ SaveVMInfo *info = g_malloc0(sizeof(*info));
-+ struct SnapshotState *s = &snap_state;
-+
-+ if (s->state != SAVE_STATE_DONE) {
-+ info->has_bytes = true;
-+ info->bytes = s->bs_pos;
-+ switch (s->state) {
-+ case SAVE_STATE_ERROR:
-+ info->has_status = true;
-+ info->status = g_strdup("failed");
-+ info->has_total_time = true;
-+ info->total_time = s->total_time;
-+ if (s->error) {
-+ info->has_error = true;
-+ info->error = g_strdup(error_get_pretty(s->error));
-+ }
-+ break;
-+ case SAVE_STATE_ACTIVE:
-+ info->has_status = true;
-+ info->status = g_strdup("active");
-+ info->has_total_time = true;
-+ info->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME)
-+ - s->total_time;
-+ break;
-+ case SAVE_STATE_COMPLETED:
-+ info->has_status = true;
-+ info->status = g_strdup("completed");
-+ info->has_total_time = true;
-+ info->total_time = s->total_time;
-+ break;
-+ }
-+ }
-+
-+ return info;
-+}
-+
-+static int save_snapshot_cleanup(void)
-+{
-+ int ret = 0;
-+
-+ DPRINTF("save_snapshot_cleanup\n");
-+
-+ snap_state.total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) -
-+ snap_state.total_time;
-+
-+ if (snap_state.file) {
-+ ret = qemu_fclose(snap_state.file);
-+ }
-+
-+ if (snap_state.target) {
-+ if (!savevm_aborted()) {
-+ /* try to truncate, but ignore errors (will fail on block devices).
-+ * note1: bdrv_read() need whole blocks, so we need to round up
-+ * note2: PVE requires 1024 (BDRV_SECTOR_SIZE*2) alignment
-+ */
-+ size_t size = QEMU_ALIGN_UP(snap_state.bs_pos, BDRV_SECTOR_SIZE*2);
-+ blk_truncate(snap_state.target, size, false, PREALLOC_MODE_OFF, 0, NULL);
-+ }
-+ blk_op_unblock_all(snap_state.target, snap_state.blocker);
-+ error_free(snap_state.blocker);
-+ snap_state.blocker = NULL;
-+ blk_unref(snap_state.target);
-+ snap_state.target = NULL;
-+
-+ qemu_co_sleep_wake(&snap_state.target_close_wait);
-+ }
-+
-+ return ret;
-+}
-+
-+static void save_snapshot_error(const char *fmt, ...)
-+{
-+ va_list ap;
-+ char *msg;
-+
-+ va_start(ap, fmt);
-+ msg = g_strdup_vprintf(fmt, ap);
-+ va_end(ap);
-+
-+ DPRINTF("save_snapshot_error: %s\n", msg);
-+
-+ if (!snap_state.error) {
-+ error_set(&snap_state.error, ERROR_CLASS_GENERIC_ERROR, "%s", msg);
-+ }
-+
-+ g_free (msg);
-+
-+ snap_state.state = SAVE_STATE_ERROR;
-+}
-+
-+static int block_state_close(void *opaque, Error **errp)
-+{
-+ snap_state.file = NULL;
-+ return blk_flush(snap_state.target);
-+}
-+
-+typedef struct BlkRwCo {
-+ int64_t offset;
-+ QEMUIOVector *qiov;
-+ ssize_t ret;
-+} BlkRwCo;
-+
-+static void coroutine_fn block_state_write_entry(void *opaque) {
-+ BlkRwCo *rwco = opaque;
-+ rwco->ret = blk_co_pwritev(snap_state.target, rwco->offset, rwco->qiov->size,
-+ rwco->qiov, 0);
-+ aio_wait_kick();
-+}
-+
-+static ssize_t block_state_writev_buffer(void *opaque, struct iovec *iov,
-+ int iovcnt, int64_t pos, Error **errp)
-+{
-+ QEMUIOVector qiov;
-+ BlkRwCo rwco;
-+
-+ assert(pos == snap_state.bs_pos);
-+ rwco = (BlkRwCo) {
-+ .offset = pos,
-+ .qiov = &qiov,
-+ .ret = NOT_DONE,
-+ };
-+
-+ qemu_iovec_init_external(&qiov, iov, iovcnt);
-+
-+ if (qemu_in_coroutine()) {
-+ block_state_write_entry(&rwco);
-+ } else {
-+ Coroutine *co = qemu_coroutine_create(&block_state_write_entry, &rwco);
-+ bdrv_coroutine_enter(blk_bs(snap_state.target), co);
-+ BDRV_POLL_WHILE(blk_bs(snap_state.target), rwco.ret == NOT_DONE);
-+ }
-+ if (rwco.ret < 0) {
-+ return rwco.ret;
-+ }
-+
-+ snap_state.bs_pos += qiov.size;
-+ return qiov.size;
-+}
-+
-+static const QEMUFileOps block_file_ops = {
-+ .writev_buffer = block_state_writev_buffer,
-+ .close = block_state_close,
-+};
-+
-+static void process_savevm_finalize(void *opaque)
-+{
-+ int ret;
-+ AioContext *iohandler_ctx = iohandler_get_aio_context();
-+ MigrationState *ms = migrate_get_current();
-+
-+ bool aborted = savevm_aborted();
-+
-+#ifdef DEBUG_SAVEVM_STATE
-+ int64_t start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-+#endif
-+
-+ qemu_bh_delete(snap_state.finalize_bh);
-+ snap_state.finalize_bh = NULL;
-+ snap_state.co = NULL;
-+
-+ /* We need to own the target bdrv's context for the following functions,
-+ * so move it back. It can stay in the main context and live out its live
-+ * there, since we're done with it after this method ends anyway.
-+ */
-+ aio_context_acquire(iohandler_ctx);
-+ blk_set_aio_context(snap_state.target, qemu_get_aio_context(), NULL);
-+ aio_context_release(iohandler_ctx);
-+
-+ ret = vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
-+ if (ret < 0) {
-+ save_snapshot_error("vm_stop_force_state error %d", ret);
-+ }
-+
-+ if (!aborted) {
-+ /* skip state saving if we aborted, snapshot will be invalid anyway */
-+ (void)qemu_savevm_state_complete_precopy(snap_state.file, false, false);
-+ ret = qemu_file_get_error(snap_state.file);
-+ if (ret < 0) {
-+ save_snapshot_error("qemu_savevm_state_iterate error %d", ret);
-+ }
-+ }
-+
-+ DPRINTF("state saving complete\n");
-+ DPRINTF("timing: process_savevm_finalize (state saving) took %ld ms\n",
-+ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time);
-+
-+ /* clear migration state */
-+ migrate_set_state(&ms->state, MIGRATION_STATUS_SETUP,
-+ ret || aborted ? MIGRATION_STATUS_FAILED : MIGRATION_STATUS_COMPLETED);
-+ ms->to_dst_file = NULL;
-+
-+ qemu_savevm_state_cleanup();
-+
-+ ret = save_snapshot_cleanup();
-+ if (ret < 0) {
-+ save_snapshot_error("save_snapshot_cleanup error %d", ret);
-+ } else if (snap_state.state == SAVE_STATE_ACTIVE) {
-+ snap_state.state = SAVE_STATE_COMPLETED;
-+ } else if (aborted) {
-+ /*
-+ * If there was an error, there's no need to set a new one here.
-+ * If the snapshot was canceled, leave setting the state to
-+ * qmp_savevm_end(), which is waked by save_snapshot_cleanup().
-+ */
-+ } else {
-+ save_snapshot_error("process_savevm_cleanup: invalid state: %d",
-+ snap_state.state);
-+ }
-+ if (snap_state.saved_vm_running) {
-+ vm_start();
-+ snap_state.saved_vm_running = false;
-+ }
-+
-+ DPRINTF("timing: process_savevm_finalize (full) took %ld ms\n",
-+ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time);
-+}
-+
-+static void coroutine_fn process_savevm_co(void *opaque)
-+{
-+ int ret;
-+ int64_t maxlen;
-+ BdrvNextIterator it;
-+ BlockDriverState *bs = NULL;
-+
-+#ifdef DEBUG_SAVEVM_STATE
-+ int64_t start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-+#endif
-+
-+ ret = qemu_file_get_error(snap_state.file);
-+ if (ret < 0) {
-+ save_snapshot_error("qemu_savevm_state_setup failed");
-+ return;
-+ }
-+
-+ while (snap_state.state == SAVE_STATE_ACTIVE) {
-+ uint64_t pending_size, pend_precopy, pend_compatible, pend_postcopy;
-+
-+ /* pending is expected to be called without iothread lock */
-+ qemu_mutex_unlock_iothread();
-+ qemu_savevm_state_pending(snap_state.file, 0, &pend_precopy, &pend_compatible, &pend_postcopy);
-+ qemu_mutex_lock_iothread();
-+
-+ pending_size = pend_precopy + pend_compatible + pend_postcopy;
-+
-+ maxlen = blk_getlength(snap_state.target) - 30*1024*1024;
-+
-+ if (pending_size > 400000 && snap_state.bs_pos + pending_size < maxlen) {
-+ ret = qemu_savevm_state_iterate(snap_state.file, false);
-+ if (ret < 0) {
-+ save_snapshot_error("qemu_savevm_state_iterate error %d", ret);
-+ break;
-+ }
-+ DPRINTF("savevm iterate pending size %lu ret %d\n", pending_size, ret);
-+ } else {
-+ qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
-+ ret = global_state_store();
-+ if (ret) {
-+ save_snapshot_error("global_state_store error %d", ret);
-+ break;
-+ }
-+
-+ DPRINTF("savevm iterate complete\n");
-+ break;
-+ }
-+ }
-+
-+ DPRINTF("timing: process_savevm_co took %ld ms\n",
-+ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time);
-+
-+#ifdef DEBUG_SAVEVM_STATE
-+ int64_t start_time_flush = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-+#endif
-+ /* If a drive runs in an IOThread we can flush it async, and only
-+ * need to sync-flush whatever IO happens between now and
-+ * vm_stop_force_state. bdrv_next can only be called from main AioContext,
-+ * so move there now and after every flush.
-+ */
-+ aio_co_reschedule_self(qemu_get_aio_context());
-+ for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
-+ /* target has BDRV_O_NO_FLUSH, no sense calling bdrv_flush on it */
-+ if (bs == blk_bs(snap_state.target)) {
-+ continue;
-+ }
-+
-+ AioContext *bs_ctx = bdrv_get_aio_context(bs);
-+ if (bs_ctx != qemu_get_aio_context()) {
-+ DPRINTF("savevm: async flushing drive %s\n", bs->filename);
-+ aio_co_reschedule_self(bs_ctx);
-+ bdrv_flush(bs);
-+ aio_co_reschedule_self(qemu_get_aio_context());
-+ }
-+ }
-+
-+ DPRINTF("timing: async flushing took %ld ms\n",
-+ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time_flush);
-+
-+ qemu_bh_schedule(snap_state.finalize_bh);
-+}
-+
-+void qmp_savevm_start(bool has_statefile, const char *statefile, Error **errp)
-+{
-+ Error *local_err = NULL;
-+ MigrationState *ms = migrate_get_current();
-+ AioContext *iohandler_ctx = iohandler_get_aio_context();
-+
-+ int bdrv_oflags = BDRV_O_RDWR | BDRV_O_RESIZE | BDRV_O_NO_FLUSH;
-+
-+ if (snap_state.state != SAVE_STATE_DONE) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ "VM snapshot already started\n");
-+ return;
-+ }
-+
-+ if (migration_is_running(ms->state)) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, QERR_MIGRATION_ACTIVE);
-+ return;
-+ }
-+
-+ if (migrate_use_block()) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ "Block migration and snapshots are incompatible");
-+ return;
-+ }
-+
-+ /* initialize snapshot info */
-+ snap_state.saved_vm_running = runstate_is_running();
-+ snap_state.bs_pos = 0;
-+ snap_state.total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
-+ snap_state.blocker = NULL;
-+ snap_state.target_close_wait.to_wake = NULL;
-+
-+ if (snap_state.error) {
-+ error_free(snap_state.error);
-+ snap_state.error = NULL;
-+ }
-+
-+ if (!has_statefile) {
-+ vm_stop(RUN_STATE_SAVE_VM);
-+ snap_state.state = SAVE_STATE_COMPLETED;
-+ return;
-+ }
-+
-+ if (qemu_savevm_state_blocked(errp)) {
-+ return;
-+ }
-+
-+ /* Open the image */
-+ QDict *options = NULL;
-+ options = qdict_new();
-+ qdict_put_str(options, "driver", "raw");
-+ snap_state.target = blk_new_open(statefile, NULL, options, bdrv_oflags, &local_err);
-+ if (!snap_state.target) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
-+ goto restart;
-+ }
-+
-+ snap_state.file = qemu_fopen_ops(&snap_state, &block_file_ops);
-+
-+ if (!snap_state.file) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
-+ goto restart;
-+ }
-+
-+ /*
-+ * qemu_savevm_* paths use migration code and expect a migration state.
-+ * State is cleared in process_savevm_co, but has to be initialized
-+ * here (blocking main thread, from QMP) to avoid race conditions.
-+ */
-+ migrate_init(ms);
-+ memset(&ram_counters, 0, sizeof(ram_counters));
-+ ms->to_dst_file = snap_state.file;
-+
-+ error_setg(&snap_state.blocker, "block device is in use by savevm");
-+ blk_op_block_all(snap_state.target, snap_state.blocker);
-+
-+ snap_state.state = SAVE_STATE_ACTIVE;
-+ snap_state.finalize_bh = qemu_bh_new(process_savevm_finalize, &snap_state);
-+ snap_state.co = qemu_coroutine_create(&process_savevm_co, NULL);
-+ qemu_mutex_unlock_iothread();
-+ qemu_savevm_state_header(snap_state.file);
-+ qemu_savevm_state_setup(snap_state.file);
-+ qemu_mutex_lock_iothread();
-+
-+ /* Async processing from here on out happens in iohandler context, so let
-+ * the target bdrv have its home there.
-+ */
-+ blk_set_aio_context(snap_state.target, iohandler_ctx, &local_err);
-+
-+ aio_co_schedule(iohandler_ctx, snap_state.co);
-+
-+ return;
-+
-+restart:
-+
-+ save_snapshot_error("setup failed");
-+
-+ if (snap_state.saved_vm_running) {
-+ vm_start();
-+ snap_state.saved_vm_running = false;
-+ }
-+}
-+
-+void coroutine_fn qmp_savevm_end(Error **errp)
-+{
-+ int64_t timeout;
-+
-+ if (snap_state.state == SAVE_STATE_DONE) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ "VM snapshot not started\n");
-+ return;
-+ }
-+
-+ if (snap_state.state == SAVE_STATE_ACTIVE) {
-+ snap_state.state = SAVE_STATE_CANCELLED;
-+ goto wait_for_close;
-+ }
-+
-+ if (snap_state.saved_vm_running) {
-+ vm_start();
-+ snap_state.saved_vm_running = false;
-+ }
-+
-+ snap_state.state = SAVE_STATE_DONE;
-+
-+wait_for_close:
-+ if (!snap_state.target) {
-+ DPRINTF("savevm-end: no target file open\n");
-+ return;
-+ }
-+
-+ /* wait until cleanup is done before returning, this ensures that after this
-+ * call exits the statefile will be closed and can be removed immediately */
-+ DPRINTF("savevm-end: waiting for cleanup\n");
-+ timeout = 30L * 1000 * 1000 * 1000;
-+ qemu_co_sleep_ns_wakeable(&snap_state.target_close_wait,
-+ QEMU_CLOCK_REALTIME, timeout);
-+ if (snap_state.target) {
-+ save_snapshot_error("timeout waiting for target file close in "
-+ "qmp_savevm_end");
-+ /* we cannot assume the snapshot finished in this case, so leave the
-+ * state alone - caller has to figure something out */
-+ return;
-+ }
-+
-+ // File closed and no other error, so ensure next snapshot can be started.
-+ if (snap_state.state != SAVE_STATE_ERROR) {
-+ snap_state.state = SAVE_STATE_DONE;
-+ }
-+
-+ DPRINTF("savevm-end: cleanup done\n");
-+}
-+
-+// FIXME: Deprecated
-+void qmp_snapshot_drive(const char *device, const char *name, Error **errp)
-+{
-+ // Compatibility to older qemu-server.
-+ qmp_blockdev_snapshot_internal_sync(device, name, errp);
-+}
-+
-+// FIXME: Deprecated
-+void qmp_delete_drive_snapshot(const char *device, const char *name,
-+ Error **errp)
-+{
-+ // Compatibility to older qemu-server.
-+ (void)qmp_blockdev_snapshot_delete_internal_sync(device, false, NULL,
-+ true, name, errp);
-+}
-+
-+static ssize_t loadstate_get_buffer(void *opaque, uint8_t *buf, int64_t pos,
-+ size_t size, Error **errp)
-+{
-+ BlockBackend *be = opaque;
-+ int64_t maxlen = blk_getlength(be);
-+ if (pos > maxlen) {
-+ return -EIO;
-+ }
-+ if ((pos + size) > maxlen) {
-+ size = maxlen - pos - 1;
-+ }
-+ if (size == 0) {
-+ return 0;
-+ }
-+ return blk_pread(be, pos, buf, size);
-+}
-+
-+static const QEMUFileOps loadstate_file_ops = {
-+ .get_buffer = loadstate_get_buffer,
-+};
-+
-+int load_snapshot_from_blockdev(const char *filename, Error **errp)
-+{
-+ BlockBackend *be;
-+ Error *local_err = NULL;
-+ Error *blocker = NULL;
-+
-+ QEMUFile *f;
-+ int ret = -EINVAL;
-+
-+ be = blk_new_open(filename, NULL, NULL, 0, &local_err);
-+
-+ if (!be) {
-+ error_setg(errp, "Could not open VM state file");
-+ goto the_end;
-+ }
-+
-+ error_setg(&blocker, "block device is in use by load state");
-+ blk_op_block_all(be, blocker);
-+
-+ /* restore the VM state */
-+ f = qemu_fopen_ops(be, &loadstate_file_ops);
-+ if (!f) {
-+ error_setg(errp, "Could not open VM state file");
-+ goto the_end;
-+ }
-+
-+ qemu_system_reset(SHUTDOWN_CAUSE_NONE);
-+ ret = qemu_loadvm_state(f);
-+
-+ /* dirty bitmap migration has a special case we need to trigger manually */
-+ dirty_bitmap_mig_before_vm_start();
-+
-+ qemu_fclose(f);
-+ migration_incoming_state_destroy();
-+ if (ret < 0) {
-+ error_setg_errno(errp, -ret, "Error while loading VM state");
-+ goto the_end;
-+ }
-+
-+ ret = 0;
-+
-+ the_end:
-+ if (be) {
-+ blk_op_unblock_all(be, blocker);
-+ error_free(blocker);
-+ blk_unref(be);
-+ }
-+ return ret;
-+}
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 5482dd0569..c64b959738 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -1906,6 +1906,63 @@ void hmp_info_memory_devices(Monitor *mon, const QDict *qdict)
- hmp_handle_error(mon, err);
- }
-
-+void hmp_savevm_start(Monitor *mon, const QDict *qdict)
-+{
-+ Error *errp = NULL;
-+ const char *statefile = qdict_get_try_str(qdict, "statefile");
-+
-+ qmp_savevm_start(statefile != NULL, statefile, &errp);
-+ hmp_handle_error(mon, errp);
-+}
-+
-+void hmp_snapshot_drive(Monitor *mon, const QDict *qdict)
-+{
-+ Error *errp = NULL;
-+ const char *name = qdict_get_str(qdict, "name");
-+ const char *device = qdict_get_str(qdict, "device");
-+
-+ qmp_snapshot_drive(device, name, &errp);
-+ hmp_handle_error(mon, errp);
-+}
-+
-+void hmp_delete_drive_snapshot(Monitor *mon, const QDict *qdict)
-+{
-+ Error *errp = NULL;
-+ const char *name = qdict_get_str(qdict, "name");
-+ const char *device = qdict_get_str(qdict, "device");
-+
-+ qmp_delete_drive_snapshot(device, name, &errp);
-+ hmp_handle_error(mon, errp);
-+}
-+
-+void coroutine_fn hmp_savevm_end(Monitor *mon, const QDict *qdict)
-+{
-+ Error *errp = NULL;
-+
-+ qmp_savevm_end(&errp);
-+ hmp_handle_error(mon, errp);
-+}
-+
-+void hmp_info_savevm(Monitor *mon, const QDict *qdict)
-+{
-+ SaveVMInfo *info;
-+ info = qmp_query_savevm(NULL);
-+
-+ if (info->has_status) {
-+ monitor_printf(mon, "savevm status: %s\n", info->status);
-+ monitor_printf(mon, "total time: %" PRIu64 " milliseconds\n",
-+ info->total_time);
-+ } else {
-+ monitor_printf(mon, "savevm status: not running\n");
-+ }
-+ if (info->has_bytes) {
-+ monitor_printf(mon, "Bytes saved: %"PRIu64"\n", info->bytes);
-+ }
-+ if (info->has_error) {
-+ monitor_printf(mon, "Error: %s\n", info->error);
-+ }
-+}
-+
- void hmp_info_iothreads(Monitor *mon, const QDict *qdict)
- {
- IOThreadInfoList *info_list = qmp_query_iothreads(NULL);
-diff --git a/qapi/migration.json b/qapi/migration.json
-index 27d7b28158..31d6cc902e 100644
---- a/qapi/migration.json
-+++ b/qapi/migration.json
-@@ -258,6 +258,40 @@
- '*compression': 'CompressionStats',
- '*socket-address': ['SocketAddress'] } }
-
-+##
-+# @SaveVMInfo:
-+#
-+# Information about current migration process.
-+#
-+# @status: string describing the current savevm status.
-+# This can be 'active', 'completed', 'failed'.
-+# If this field is not returned, no savevm process
-+# has been initiated
-+#
-+# @error: string containing error message is status is failed.
-+#
-+# @total-time: total amount of milliseconds since savevm started.
-+# If savevm has ended, it returns the total save time
-+#
-+# @bytes: total amount of data transfered
-+#
-+# Since: 1.3
-+##
-+{ 'struct': 'SaveVMInfo',
-+ 'data': {'*status': 'str', '*error': 'str',
-+ '*total-time': 'int', '*bytes': 'int'} }
-+
-+##
-+# @query-savevm:
-+#
-+# Returns information about current savevm process.
-+#
-+# Returns: @SaveVMInfo
-+#
-+# Since: 1.3
-+##
-+{ 'command': 'query-savevm', 'returns': 'SaveVMInfo' }
-+
- ##
- # @query-migrate:
- #
-diff --git a/qapi/misc.json b/qapi/misc.json
-index b83cc39029..1e5dd7db29 100644
---- a/qapi/misc.json
-+++ b/qapi/misc.json
-@@ -435,6 +435,38 @@
- ##
- { 'command': 'query-fdsets', 'returns': ['FdsetInfo'] }
-
-+##
-+# @savevm-start:
-+#
-+# Prepare for snapshot and halt VM. Save VM state to statefile.
-+#
-+##
-+{ 'command': 'savevm-start', 'data': { '*statefile': 'str' } }
-+
-+##
-+# @snapshot-drive:
-+#
-+# Create an internal drive snapshot.
-+#
-+##
-+{ 'command': 'snapshot-drive', 'data': { 'device': 'str', 'name': 'str' } }
-+
-+##
-+# @delete-drive-snapshot:
-+#
-+# Delete a drive snapshot.
-+#
-+##
-+{ 'command': 'delete-drive-snapshot', 'data': { 'device': 'str', 'name': 'str' } }
-+
-+##
-+# @savevm-end:
-+#
-+# Resume VM after a snapshot.
-+#
-+##
-+{ 'command': 'savevm-end', 'coroutine': true }
-+
- ##
- # @CommandLineParameterType:
- #
-diff --git a/qemu-options.hx b/qemu-options.hx
-index 34e9b32a5c..aeade4ef80 100644
---- a/qemu-options.hx
-+++ b/qemu-options.hx
-@@ -4254,6 +4254,18 @@ SRST
- Start right away with a saved state (``loadvm`` in monitor)
- ERST
-
-+DEF("loadstate", HAS_ARG, QEMU_OPTION_loadstate, \
-+ "-loadstate file\n" \
-+ " start right away with a saved state\n",
-+ QEMU_ARCH_ALL)
-+SRST
-+``-loadstate file``
-+ Start right away with a saved state. This option does not rollback
-+ disk state like @code{loadvm}, so user must make sure that disk
-+ have correct state. @var{file} can be any valid device URL. See the section
-+ for "Device URL Syntax" for more information.
-+ERST
-+
- #ifndef _WIN32
- DEF("daemonize", 0, QEMU_OPTION_daemonize, \
- "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
-diff --git a/softmmu/vl.c b/softmmu/vl.c
-index 6f646531a0..a3f2a3818c 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -157,6 +157,7 @@ static const char *incoming;
- static const char *loadvm;
- static const char *accelerators;
- static QDict *machine_opts_dict;
-+static const char *loadstate;
- static QTAILQ_HEAD(, ObjectOption) object_opts = QTAILQ_HEAD_INITIALIZER(object_opts);
- static QTAILQ_HEAD(, DeviceOption) device_opts = QTAILQ_HEAD_INITIALIZER(device_opts);
- static ram_addr_t maxram_size;
-@@ -2749,6 +2750,12 @@ void qmp_x_exit_preconfig(Error **errp)
-
- if (loadvm) {
- load_snapshot(loadvm, NULL, false, NULL, &error_fatal);
-+ } else if (loadstate) {
-+ Error *local_err = NULL;
-+ if (load_snapshot_from_blockdev(loadstate, &local_err) < 0) {
-+ error_report_err(local_err);
-+ autostart = 0;
-+ }
- }
- if (replay_mode != REPLAY_MODE_NONE) {
- replay_vmstate_init();
-@@ -3289,6 +3296,9 @@ void qemu_init(int argc, char **argv, char **envp)
- case QEMU_OPTION_loadvm:
- loadvm = optarg;
- break;
-+ case QEMU_OPTION_loadstate:
-+ loadstate = optarg;
-+ break;
- case QEMU_OPTION_full_screen:
- dpy.has_full_screen = true;
- dpy.full_screen = true;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 4 May 2020 11:05:08 +0200
-Subject: [PATCH] PVE: add optional buffer size to QEMUFile
-
-So we can use a 4M buffer for savevm-async which should
-increase performance storing the state onto ceph.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-[increase max IOV count in QEMUFile to actually write more data]
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- migration/qemu-file.c | 38 +++++++++++++++++++++++++-------------
- migration/qemu-file.h | 1 +
- migration/savevm-async.c | 4 ++--
- 3 files changed, 28 insertions(+), 15 deletions(-)
-
-diff --git a/migration/qemu-file.c b/migration/qemu-file.c
-index 1479cddad9..21a3355ae2 100644
---- a/migration/qemu-file.c
-+++ b/migration/qemu-file.c
-@@ -31,8 +31,8 @@
- #include "trace.h"
- #include "qapi/error.h"
-
--#define IO_BUF_SIZE 32768
--#define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 64)
-+#define DEFAULT_IO_BUF_SIZE 32768
-+#define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 256)
-
- struct QEMUFile {
- const QEMUFileOps *ops;
-@@ -46,7 +46,8 @@ struct QEMUFile {
- when reading */
- int buf_index;
- int buf_size; /* 0 when writing */
-- uint8_t buf[IO_BUF_SIZE];
-+ size_t buf_allocated_size;
-+ uint8_t *buf;
-
- DECLARE_BITMAP(may_free, MAX_IOV_SIZE);
- struct iovec iov[MAX_IOV_SIZE];
-@@ -104,7 +105,7 @@ bool qemu_file_mode_is_not_valid(const char *mode)
- return false;
- }
-
--QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops, bool has_ioc)
-+QEMUFile *qemu_fopen_ops_sized(void *opaque, const QEMUFileOps *ops, bool has_ioc, size_t buffer_size)
- {
- QEMUFile *f;
-
-@@ -113,9 +114,17 @@ QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops, bool has_ioc)
- f->opaque = opaque;
- f->ops = ops;
- f->has_ioc = has_ioc;
-+ f->buf_allocated_size = buffer_size;
-+ f->buf = malloc(buffer_size);
-+
- return f;
- }
-
-+QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops, bool has_ioc)
-+{
-+ return qemu_fopen_ops_sized(opaque, ops, has_ioc, DEFAULT_IO_BUF_SIZE);
-+}
-+
-
- void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks)
- {
-@@ -350,7 +359,7 @@ static ssize_t qemu_fill_buffer(QEMUFile *f)
- }
-
- len = f->ops->get_buffer(f->opaque, f->buf + pending, f->pos,
-- IO_BUF_SIZE - pending, &local_error);
-+ f->buf_allocated_size - pending, &local_error);
- if (len > 0) {
- f->buf_size += len;
- f->pos += len;
-@@ -390,6 +399,9 @@ int qemu_fclose(QEMUFile *f)
- ret = ret2;
- }
- }
-+
-+ free(f->buf);
-+
- /* If any error was spotted before closing, we should report it
- * instead of the close() return value.
- */
-@@ -444,7 +456,7 @@ static void add_buf_to_iovec(QEMUFile *f, size_t len)
- {
- if (!add_to_iovec(f, f->buf + f->buf_index, len, false)) {
- f->buf_index += len;
-- if (f->buf_index == IO_BUF_SIZE) {
-+ if (f->buf_index == f->buf_allocated_size) {
- qemu_fflush(f);
- }
- }
-@@ -470,7 +482,7 @@ void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, size_t size)
- }
-
- while (size > 0) {
-- l = IO_BUF_SIZE - f->buf_index;
-+ l = f->buf_allocated_size - f->buf_index;
- if (l > size) {
- l = size;
- }
-@@ -517,8 +529,8 @@ size_t qemu_peek_buffer(QEMUFile *f, uint8_t **buf, size_t size, size_t offset)
- size_t index;
-
- assert(!qemu_file_is_writable(f));
-- assert(offset < IO_BUF_SIZE);
-- assert(size <= IO_BUF_SIZE - offset);
-+ assert(offset < f->buf_allocated_size);
-+ assert(size <= f->buf_allocated_size - offset);
-
- /* The 1st byte to read from */
- index = f->buf_index + offset;
-@@ -568,7 +580,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
- size_t res;
- uint8_t *src;
-
-- res = qemu_peek_buffer(f, &src, MIN(pending, IO_BUF_SIZE), 0);
-+ res = qemu_peek_buffer(f, &src, MIN(pending, f->buf_allocated_size), 0);
- if (res == 0) {
- return done;
- }
-@@ -602,7 +614,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
- */
- size_t qemu_get_buffer_in_place(QEMUFile *f, uint8_t **buf, size_t size)
- {
-- if (size < IO_BUF_SIZE) {
-+ if (size < f->buf_allocated_size) {
- size_t res;
- uint8_t *src = NULL;
-
-@@ -627,7 +639,7 @@ int qemu_peek_byte(QEMUFile *f, int offset)
- int index = f->buf_index + offset;
-
- assert(!qemu_file_is_writable(f));
-- assert(offset < IO_BUF_SIZE);
-+ assert(offset < f->buf_allocated_size);
-
- if (index >= f->buf_size) {
- qemu_fill_buffer(f);
-@@ -779,7 +791,7 @@ static int qemu_compress_data(z_stream *stream, uint8_t *dest, size_t dest_len,
- ssize_t qemu_put_compression_data(QEMUFile *f, z_stream *stream,
- const uint8_t *p, size_t size)
- {
-- ssize_t blen = IO_BUF_SIZE - f->buf_index - sizeof(int32_t);
-+ ssize_t blen = f->buf_allocated_size - f->buf_index - sizeof(int32_t);
-
- if (blen < compressBound(size)) {
- return -1;
-diff --git a/migration/qemu-file.h b/migration/qemu-file.h
-index 3f36d4dc8c..67501fd9cf 100644
---- a/migration/qemu-file.h
-+++ b/migration/qemu-file.h
-@@ -121,6 +121,7 @@ typedef struct QEMUFileHooks {
- } QEMUFileHooks;
-
- QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops, bool has_ioc);
-+QEMUFile *qemu_fopen_ops_sized(void *opaque, const QEMUFileOps *ops, bool has_ioc, size_t buffer_size);
- void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks);
- int qemu_get_fd(QEMUFile *f);
- int qemu_fclose(QEMUFile *f);
-diff --git a/migration/savevm-async.c b/migration/savevm-async.c
-index b9a43c56bc..0bc5799cf8 100644
---- a/migration/savevm-async.c
-+++ b/migration/savevm-async.c
-@@ -420,7 +420,7 @@ void qmp_savevm_start(bool has_statefile, const char *statefile, Error **errp)
- goto restart;
- }
-
-- snap_state.file = qemu_fopen_ops(&snap_state, &block_file_ops);
-+ snap_state.file = qemu_fopen_ops_sized(&snap_state, &block_file_ops, false, 4 * 1024 * 1024);
-
- if (!snap_state.file) {
- error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
-@@ -573,7 +573,7 @@ int load_snapshot_from_blockdev(const char *filename, Error **errp)
- blk_op_block_all(be, blocker);
-
- /* restore the VM state */
-- f = qemu_fopen_ops(be, &loadstate_file_ops);
-+ f = qemu_fopen_ops_sized(be, &loadstate_file_ops, false, 4 * 1024 * 1024);
- if (!f) {
- error_setg(errp, "Could not open VM state file");
- goto the_end;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:46 +0200
+Subject: [PATCH] PVE: add savevm-async for background state snapshots
+
+Put qemu_savevm_state_{header,setup} into the main loop and the rest
+of the iteration into a coroutine. The former need to lock the
+iothread (and we can't unlock it in the coroutine), and the latter
+can't deal with being in a separate thread, so a coroutine it must
+be.
+
+Truncate output file at 1024 boundary.
+
+Do not block the VM and save the state on aborting a snapshot, as the
+snapshot will be invalid anyway.
+
+Also, when aborting, wait for the target file to be closed, otherwise a
+client might run into race-conditions when trying to remove the file
+still opened by QEMU.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+[improve aborting]
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+[FE: further improve aborting
+ adapt to removal of QEMUFileOps]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hmp-commands-info.hx | 13 +
+ hmp-commands.hx | 33 +++
+ include/migration/snapshot.h | 2 +
+ include/monitor/hmp.h | 5 +
+ migration/meson.build | 1 +
+ migration/savevm-async.c | 531 +++++++++++++++++++++++++++++++++++
+ monitor/hmp-cmds.c | 57 ++++
+ qapi/migration.json | 34 +++
+ qapi/misc.json | 32 +++
+ qemu-options.hx | 12 +
+ softmmu/vl.c | 10 +
+ 11 files changed, 730 insertions(+)
+ create mode 100644 migration/savevm-async.c
+
+diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
+index 188d9ece3b..97b88eaaad 100644
+--- a/hmp-commands-info.hx
++++ b/hmp-commands-info.hx
+@@ -538,6 +538,19 @@ SRST
+ Show current migration parameters.
+ ERST
+
++ {
++ .name = "savevm",
++ .args_type = "",
++ .params = "",
++ .help = "show savevm status",
++ .cmd = hmp_info_savevm,
++ },
++
++SRST
++ ``info savevm``
++ Show savevm status.
++ERST
++
+ {
+ .name = "balloon",
+ .args_type = "",
+diff --git a/hmp-commands.hx b/hmp-commands.hx
+index 182e639d14..bbcc73e942 100644
+--- a/hmp-commands.hx
++++ b/hmp-commands.hx
+@@ -1800,3 +1800,36 @@ ERST
+ "\n\t\t\t\t\t limit on a specified virtual cpu",
+ .cmd = hmp_cancel_vcpu_dirty_limit,
+ },
++
++ {
++ .name = "savevm-start",
++ .args_type = "statefile:s?",
++ .params = "[statefile]",
++ .help = "Prepare for snapshot and halt VM. Save VM state to statefile.",
++ .cmd = hmp_savevm_start,
++ },
++
++ {
++ .name = "snapshot-drive",
++ .args_type = "device:s,name:s",
++ .params = "device name",
++ .help = "Create internal snapshot.",
++ .cmd = hmp_snapshot_drive,
++ },
++
++ {
++ .name = "delete-drive-snapshot",
++ .args_type = "device:s,name:s",
++ .params = "device name",
++ .help = "Delete internal snapshot.",
++ .cmd = hmp_delete_drive_snapshot,
++ },
++
++ {
++ .name = "savevm-end",
++ .args_type = "",
++ .params = "",
++ .help = "Resume VM after snaphot.",
++ .cmd = hmp_savevm_end,
++ .coroutine = true,
++ },
+diff --git a/include/migration/snapshot.h b/include/migration/snapshot.h
+index e72083b117..c846d37806 100644
+--- a/include/migration/snapshot.h
++++ b/include/migration/snapshot.h
+@@ -61,4 +61,6 @@ bool delete_snapshot(const char *name,
+ bool has_devices, strList *devices,
+ Error **errp);
+
++int load_snapshot_from_blockdev(const char *filename, Error **errp);
++
+ #endif
+diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
+index a618eb1e4e..55067beff1 100644
+--- a/include/monitor/hmp.h
++++ b/include/monitor/hmp.h
+@@ -26,6 +26,7 @@ void hmp_info_status(Monitor *mon, const QDict *qdict);
+ void hmp_info_uuid(Monitor *mon, const QDict *qdict);
+ void hmp_info_chardev(Monitor *mon, const QDict *qdict);
+ void hmp_info_mice(Monitor *mon, const QDict *qdict);
++void hmp_info_savevm(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_capabilities(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict);
+@@ -80,6 +81,10 @@ void hmp_netdev_add(Monitor *mon, const QDict *qdict);
+ void hmp_netdev_del(Monitor *mon, const QDict *qdict);
+ void hmp_getfd(Monitor *mon, const QDict *qdict);
+ void hmp_closefd(Monitor *mon, const QDict *qdict);
++void hmp_savevm_start(Monitor *mon, const QDict *qdict);
++void hmp_snapshot_drive(Monitor *mon, const QDict *qdict);
++void hmp_delete_drive_snapshot(Monitor *mon, const QDict *qdict);
++void hmp_savevm_end(Monitor *mon, const QDict *qdict);
+ void hmp_sendkey(Monitor *mon, const QDict *qdict);
+ void hmp_screendump(Monitor *mon, const QDict *qdict);
+ void hmp_chardev_add(Monitor *mon, const QDict *qdict);
+diff --git a/migration/meson.build b/migration/meson.build
+index 8cac83c06c..0842d00cd2 100644
+--- a/migration/meson.build
++++ b/migration/meson.build
+@@ -24,6 +24,7 @@ softmmu_ss.add(files(
+ 'multifd-zlib.c',
+ 'postcopy-ram.c',
+ 'savevm.c',
++ 'savevm-async.c',
+ 'socket.c',
+ 'tls.c',
+ ), gnutls)
+diff --git a/migration/savevm-async.c b/migration/savevm-async.c
+new file mode 100644
+index 0000000000..b3692739a0
+--- /dev/null
++++ b/migration/savevm-async.c
+@@ -0,0 +1,531 @@
++#include "qemu/osdep.h"
++#include "migration/channel-savevm-async.h"
++#include "migration/migration.h"
++#include "migration/savevm.h"
++#include "migration/snapshot.h"
++#include "migration/global_state.h"
++#include "migration/ram.h"
++#include "migration/qemu-file.h"
++#include "sysemu/sysemu.h"
++#include "sysemu/runstate.h"
++#include "block/block.h"
++#include "sysemu/block-backend.h"
++#include "qapi/error.h"
++#include "qapi/qmp/qerror.h"
++#include "qapi/qmp/qdict.h"
++#include "qapi/qapi-commands-migration.h"
++#include "qapi/qapi-commands-misc.h"
++#include "qapi/qapi-commands-block.h"
++#include "qemu/cutils.h"
++#include "qemu/timer.h"
++#include "qemu/main-loop.h"
++#include "qemu/rcu.h"
++
++/* #define DEBUG_SAVEVM_STATE */
++
++#ifdef DEBUG_SAVEVM_STATE
++#define DPRINTF(fmt, ...) \
++ do { printf("savevm-async: " fmt, ## __VA_ARGS__); } while (0)
++#else
++#define DPRINTF(fmt, ...) \
++ do { } while (0)
++#endif
++
++enum {
++ SAVE_STATE_DONE,
++ SAVE_STATE_ERROR,
++ SAVE_STATE_ACTIVE,
++ SAVE_STATE_COMPLETED,
++ SAVE_STATE_CANCELLED
++};
++
++
++static struct SnapshotState {
++ BlockBackend *target;
++ size_t bs_pos;
++ int state;
++ Error *error;
++ Error *blocker;
++ int saved_vm_running;
++ QEMUFile *file;
++ int64_t total_time;
++ QEMUBH *finalize_bh;
++ Coroutine *co;
++ QemuCoSleep target_close_wait;
++} snap_state;
++
++static bool savevm_aborted(void)
++{
++ return snap_state.state == SAVE_STATE_CANCELLED ||
++ snap_state.state == SAVE_STATE_ERROR;
++}
++
++SaveVMInfo *qmp_query_savevm(Error **errp)
++{
++ SaveVMInfo *info = g_malloc0(sizeof(*info));
++ struct SnapshotState *s = &snap_state;
++
++ if (s->state != SAVE_STATE_DONE) {
++ info->has_bytes = true;
++ info->bytes = s->bs_pos;
++ switch (s->state) {
++ case SAVE_STATE_ERROR:
++ info->has_status = true;
++ info->status = g_strdup("failed");
++ info->has_total_time = true;
++ info->total_time = s->total_time;
++ if (s->error) {
++ info->has_error = true;
++ info->error = g_strdup(error_get_pretty(s->error));
++ }
++ break;
++ case SAVE_STATE_ACTIVE:
++ info->has_status = true;
++ info->status = g_strdup("active");
++ info->has_total_time = true;
++ info->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME)
++ - s->total_time;
++ break;
++ case SAVE_STATE_COMPLETED:
++ info->has_status = true;
++ info->status = g_strdup("completed");
++ info->has_total_time = true;
++ info->total_time = s->total_time;
++ break;
++ }
++ }
++
++ return info;
++}
++
++static int save_snapshot_cleanup(void)
++{
++ int ret = 0;
++
++ DPRINTF("save_snapshot_cleanup\n");
++
++ snap_state.total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME) -
++ snap_state.total_time;
++
++ if (snap_state.file) {
++ ret = qemu_fclose(snap_state.file);
++ snap_state.file = NULL;
++ }
++
++ if (snap_state.target) {
++ if (!savevm_aborted()) {
++ /* try to truncate, but ignore errors (will fail on block devices).
++ * note1: bdrv_read() need whole blocks, so we need to round up
++ * note2: PVE requires 1024 (BDRV_SECTOR_SIZE*2) alignment
++ */
++ size_t size = QEMU_ALIGN_UP(snap_state.bs_pos, BDRV_SECTOR_SIZE*2);
++ blk_truncate(snap_state.target, size, false, PREALLOC_MODE_OFF, 0, NULL);
++ }
++ blk_op_unblock_all(snap_state.target, snap_state.blocker);
++ error_free(snap_state.blocker);
++ snap_state.blocker = NULL;
++ blk_unref(snap_state.target);
++ snap_state.target = NULL;
++
++ qemu_co_sleep_wake(&snap_state.target_close_wait);
++ }
++
++ return ret;
++}
++
++static void save_snapshot_error(const char *fmt, ...)
++{
++ va_list ap;
++ char *msg;
++
++ va_start(ap, fmt);
++ msg = g_strdup_vprintf(fmt, ap);
++ va_end(ap);
++
++ DPRINTF("save_snapshot_error: %s\n", msg);
++
++ if (!snap_state.error) {
++ error_set(&snap_state.error, ERROR_CLASS_GENERIC_ERROR, "%s", msg);
++ }
++
++ g_free (msg);
++
++ snap_state.state = SAVE_STATE_ERROR;
++}
++
++static void process_savevm_finalize(void *opaque)
++{
++ int ret;
++ AioContext *iohandler_ctx = iohandler_get_aio_context();
++ MigrationState *ms = migrate_get_current();
++
++ bool aborted = savevm_aborted();
++
++#ifdef DEBUG_SAVEVM_STATE
++ int64_t start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
++#endif
++
++ qemu_bh_delete(snap_state.finalize_bh);
++ snap_state.finalize_bh = NULL;
++ snap_state.co = NULL;
++
++ /* We need to own the target bdrv's context for the following functions,
++ * so move it back. It can stay in the main context and live out its live
++ * there, since we're done with it after this method ends anyway.
++ */
++ aio_context_acquire(iohandler_ctx);
++ blk_set_aio_context(snap_state.target, qemu_get_aio_context(), NULL);
++ aio_context_release(iohandler_ctx);
++
++ ret = vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
++ if (ret < 0) {
++ save_snapshot_error("vm_stop_force_state error %d", ret);
++ }
++
++ if (!aborted) {
++ /* skip state saving if we aborted, snapshot will be invalid anyway */
++ (void)qemu_savevm_state_complete_precopy(snap_state.file, false, false);
++ ret = qemu_file_get_error(snap_state.file);
++ if (ret < 0) {
++ save_snapshot_error("qemu_savevm_state_iterate error %d", ret);
++ }
++ }
++
++ DPRINTF("state saving complete\n");
++ DPRINTF("timing: process_savevm_finalize (state saving) took %ld ms\n",
++ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time);
++
++ /* clear migration state */
++ migrate_set_state(&ms->state, MIGRATION_STATUS_SETUP,
++ ret || aborted ? MIGRATION_STATUS_FAILED : MIGRATION_STATUS_COMPLETED);
++ ms->to_dst_file = NULL;
++
++ qemu_savevm_state_cleanup();
++
++ ret = save_snapshot_cleanup();
++ if (ret < 0) {
++ save_snapshot_error("save_snapshot_cleanup error %d", ret);
++ } else if (snap_state.state == SAVE_STATE_ACTIVE) {
++ snap_state.state = SAVE_STATE_COMPLETED;
++ } else if (aborted) {
++ /*
++ * If there was an error, there's no need to set a new one here.
++ * If the snapshot was canceled, leave setting the state to
++ * qmp_savevm_end(), which is waked by save_snapshot_cleanup().
++ */
++ } else {
++ save_snapshot_error("process_savevm_cleanup: invalid state: %d",
++ snap_state.state);
++ }
++ if (snap_state.saved_vm_running) {
++ vm_start();
++ snap_state.saved_vm_running = false;
++ }
++
++ DPRINTF("timing: process_savevm_finalize (full) took %ld ms\n",
++ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time);
++}
++
++static void coroutine_fn process_savevm_co(void *opaque)
++{
++ int ret;
++ int64_t maxlen;
++ BdrvNextIterator it;
++ BlockDriverState *bs = NULL;
++
++#ifdef DEBUG_SAVEVM_STATE
++ int64_t start_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
++#endif
++
++ ret = qemu_file_get_error(snap_state.file);
++ if (ret < 0) {
++ save_snapshot_error("qemu_savevm_state_setup failed");
++ return;
++ }
++
++ while (snap_state.state == SAVE_STATE_ACTIVE) {
++ uint64_t pending_size, pend_precopy, pend_compatible, pend_postcopy;
++
++ /* pending is expected to be called without iothread lock */
++ qemu_mutex_unlock_iothread();
++ qemu_savevm_state_pending(snap_state.file, 0, &pend_precopy, &pend_compatible, &pend_postcopy);
++ qemu_mutex_lock_iothread();
++
++ pending_size = pend_precopy + pend_compatible + pend_postcopy;
++
++ maxlen = blk_getlength(snap_state.target) - 30*1024*1024;
++
++ if (pending_size > 400000 && snap_state.bs_pos + pending_size < maxlen) {
++ ret = qemu_savevm_state_iterate(snap_state.file, false);
++ if (ret < 0) {
++ save_snapshot_error("qemu_savevm_state_iterate error %d", ret);
++ break;
++ }
++ DPRINTF("savevm iterate pending size %lu ret %d\n", pending_size, ret);
++ } else {
++ qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
++ ret = global_state_store();
++ if (ret) {
++ save_snapshot_error("global_state_store error %d", ret);
++ break;
++ }
++
++ DPRINTF("savevm iterate complete\n");
++ break;
++ }
++ }
++
++ DPRINTF("timing: process_savevm_co took %ld ms\n",
++ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time);
++
++#ifdef DEBUG_SAVEVM_STATE
++ int64_t start_time_flush = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
++#endif
++ /* If a drive runs in an IOThread we can flush it async, and only
++ * need to sync-flush whatever IO happens between now and
++ * vm_stop_force_state. bdrv_next can only be called from main AioContext,
++ * so move there now and after every flush.
++ */
++ aio_co_reschedule_self(qemu_get_aio_context());
++ for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
++ /* target has BDRV_O_NO_FLUSH, no sense calling bdrv_flush on it */
++ if (bs == blk_bs(snap_state.target)) {
++ continue;
++ }
++
++ AioContext *bs_ctx = bdrv_get_aio_context(bs);
++ if (bs_ctx != qemu_get_aio_context()) {
++ DPRINTF("savevm: async flushing drive %s\n", bs->filename);
++ aio_co_reschedule_self(bs_ctx);
++ bdrv_flush(bs);
++ aio_co_reschedule_self(qemu_get_aio_context());
++ }
++ }
++
++ DPRINTF("timing: async flushing took %ld ms\n",
++ qemu_clock_get_ms(QEMU_CLOCK_REALTIME) - start_time_flush);
++
++ qemu_bh_schedule(snap_state.finalize_bh);
++}
++
++void qmp_savevm_start(bool has_statefile, const char *statefile, Error **errp)
++{
++ Error *local_err = NULL;
++ MigrationState *ms = migrate_get_current();
++ AioContext *iohandler_ctx = iohandler_get_aio_context();
++
++ int bdrv_oflags = BDRV_O_RDWR | BDRV_O_RESIZE | BDRV_O_NO_FLUSH;
++
++ if (snap_state.state != SAVE_STATE_DONE) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
++ "VM snapshot already started\n");
++ return;
++ }
++
++ if (migration_is_running(ms->state)) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, QERR_MIGRATION_ACTIVE);
++ return;
++ }
++
++ if (migrate_use_block()) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
++ "Block migration and snapshots are incompatible");
++ return;
++ }
++
++ /* initialize snapshot info */
++ snap_state.saved_vm_running = runstate_is_running();
++ snap_state.bs_pos = 0;
++ snap_state.total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
++ snap_state.blocker = NULL;
++ snap_state.target_close_wait.to_wake = NULL;
++
++ if (snap_state.error) {
++ error_free(snap_state.error);
++ snap_state.error = NULL;
++ }
++
++ if (!has_statefile) {
++ vm_stop(RUN_STATE_SAVE_VM);
++ snap_state.state = SAVE_STATE_COMPLETED;
++ return;
++ }
++
++ if (qemu_savevm_state_blocked(errp)) {
++ return;
++ }
++
++ /* Open the image */
++ QDict *options = NULL;
++ options = qdict_new();
++ qdict_put_str(options, "driver", "raw");
++ snap_state.target = blk_new_open(statefile, NULL, options, bdrv_oflags, &local_err);
++ if (!snap_state.target) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
++ goto restart;
++ }
++
++ QIOChannel *ioc = QIO_CHANNEL(qio_channel_savevm_async_new(snap_state.target,
++ &snap_state.bs_pos));
++ snap_state.file = qemu_file_new_output(ioc);
++
++ if (!snap_state.file) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
++ goto restart;
++ }
++
++ /*
++ * qemu_savevm_* paths use migration code and expect a migration state.
++ * State is cleared in process_savevm_co, but has to be initialized
++ * here (blocking main thread, from QMP) to avoid race conditions.
++ */
++ migrate_init(ms);
++ memset(&ram_counters, 0, sizeof(ram_counters));
++ ms->to_dst_file = snap_state.file;
++
++ error_setg(&snap_state.blocker, "block device is in use by savevm");
++ blk_op_block_all(snap_state.target, snap_state.blocker);
++
++ snap_state.state = SAVE_STATE_ACTIVE;
++ snap_state.finalize_bh = qemu_bh_new(process_savevm_finalize, &snap_state);
++ snap_state.co = qemu_coroutine_create(&process_savevm_co, NULL);
++ qemu_mutex_unlock_iothread();
++ qemu_savevm_state_header(snap_state.file);
++ qemu_savevm_state_setup(snap_state.file);
++ qemu_mutex_lock_iothread();
++
++ /* Async processing from here on out happens in iohandler context, so let
++ * the target bdrv have its home there.
++ */
++ blk_set_aio_context(snap_state.target, iohandler_ctx, &local_err);
++
++ aio_co_schedule(iohandler_ctx, snap_state.co);
++
++ return;
++
++restart:
++
++ save_snapshot_error("setup failed");
++
++ if (snap_state.saved_vm_running) {
++ vm_start();
++ snap_state.saved_vm_running = false;
++ }
++}
++
++void coroutine_fn qmp_savevm_end(Error **errp)
++{
++ int64_t timeout;
++
++ if (snap_state.state == SAVE_STATE_DONE) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
++ "VM snapshot not started\n");
++ return;
++ }
++
++ if (snap_state.state == SAVE_STATE_ACTIVE) {
++ snap_state.state = SAVE_STATE_CANCELLED;
++ goto wait_for_close;
++ }
++
++ if (snap_state.saved_vm_running) {
++ vm_start();
++ snap_state.saved_vm_running = false;
++ }
++
++ snap_state.state = SAVE_STATE_DONE;
++
++wait_for_close:
++ if (!snap_state.target) {
++ DPRINTF("savevm-end: no target file open\n");
++ return;
++ }
++
++ /* wait until cleanup is done before returning, this ensures that after this
++ * call exits the statefile will be closed and can be removed immediately */
++ DPRINTF("savevm-end: waiting for cleanup\n");
++ timeout = 30L * 1000 * 1000 * 1000;
++ qemu_co_sleep_ns_wakeable(&snap_state.target_close_wait,
++ QEMU_CLOCK_REALTIME, timeout);
++ if (snap_state.target) {
++ save_snapshot_error("timeout waiting for target file close in "
++ "qmp_savevm_end");
++ /* we cannot assume the snapshot finished in this case, so leave the
++ * state alone - caller has to figure something out */
++ return;
++ }
++
++ // File closed and no other error, so ensure next snapshot can be started.
++ if (snap_state.state != SAVE_STATE_ERROR) {
++ snap_state.state = SAVE_STATE_DONE;
++ }
++
++ DPRINTF("savevm-end: cleanup done\n");
++}
++
++// FIXME: Deprecated
++void qmp_snapshot_drive(const char *device, const char *name, Error **errp)
++{
++ // Compatibility to older qemu-server.
++ qmp_blockdev_snapshot_internal_sync(device, name, errp);
++}
++
++// FIXME: Deprecated
++void qmp_delete_drive_snapshot(const char *device, const char *name,
++ Error **errp)
++{
++ // Compatibility to older qemu-server.
++ (void)qmp_blockdev_snapshot_delete_internal_sync(device, false, NULL,
++ true, name, errp);
++}
++
++int load_snapshot_from_blockdev(const char *filename, Error **errp)
++{
++ BlockBackend *be;
++ Error *local_err = NULL;
++ Error *blocker = NULL;
++
++ QEMUFile *f;
++ size_t bs_pos = 0;
++ int ret = -EINVAL;
++
++ be = blk_new_open(filename, NULL, NULL, 0, &local_err);
++
++ if (!be) {
++ error_setg(errp, "Could not open VM state file");
++ goto the_end;
++ }
++
++ error_setg(&blocker, "block device is in use by load state");
++ blk_op_block_all(be, blocker);
++
++ /* restore the VM state */
++ f = qemu_file_new_input(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)));
++ if (!f) {
++ error_setg(errp, "Could not open VM state file");
++ goto the_end;
++ }
++
++ qemu_system_reset(SHUTDOWN_CAUSE_NONE);
++ ret = qemu_loadvm_state(f);
++
++ /* dirty bitmap migration has a special case we need to trigger manually */
++ dirty_bitmap_mig_before_vm_start();
++
++ qemu_fclose(f);
++ migration_incoming_state_destroy();
++ if (ret < 0) {
++ error_setg_errno(errp, -ret, "Error while loading VM state");
++ goto the_end;
++ }
++
++ ret = 0;
++
++ the_end:
++ if (be) {
++ blk_op_unblock_all(be, blocker);
++ error_free(blocker);
++ blk_unref(be);
++ }
++ return ret;
++}
+diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
+index 15572befb1..1507180990 100644
+--- a/monitor/hmp-cmds.c
++++ b/monitor/hmp-cmds.c
+@@ -1925,6 +1925,63 @@ void hmp_info_memory_devices(Monitor *mon, const QDict *qdict)
+ hmp_handle_error(mon, err);
+ }
+
++void hmp_savevm_start(Monitor *mon, const QDict *qdict)
++{
++ Error *errp = NULL;
++ const char *statefile = qdict_get_try_str(qdict, "statefile");
++
++ qmp_savevm_start(statefile != NULL, statefile, &errp);
++ hmp_handle_error(mon, errp);
++}
++
++void hmp_snapshot_drive(Monitor *mon, const QDict *qdict)
++{
++ Error *errp = NULL;
++ const char *name = qdict_get_str(qdict, "name");
++ const char *device = qdict_get_str(qdict, "device");
++
++ qmp_snapshot_drive(device, name, &errp);
++ hmp_handle_error(mon, errp);
++}
++
++void hmp_delete_drive_snapshot(Monitor *mon, const QDict *qdict)
++{
++ Error *errp = NULL;
++ const char *name = qdict_get_str(qdict, "name");
++ const char *device = qdict_get_str(qdict, "device");
++
++ qmp_delete_drive_snapshot(device, name, &errp);
++ hmp_handle_error(mon, errp);
++}
++
++void coroutine_fn hmp_savevm_end(Monitor *mon, const QDict *qdict)
++{
++ Error *errp = NULL;
++
++ qmp_savevm_end(&errp);
++ hmp_handle_error(mon, errp);
++}
++
++void hmp_info_savevm(Monitor *mon, const QDict *qdict)
++{
++ SaveVMInfo *info;
++ info = qmp_query_savevm(NULL);
++
++ if (info->has_status) {
++ monitor_printf(mon, "savevm status: %s\n", info->status);
++ monitor_printf(mon, "total time: %" PRIu64 " milliseconds\n",
++ info->total_time);
++ } else {
++ monitor_printf(mon, "savevm status: not running\n");
++ }
++ if (info->has_bytes) {
++ monitor_printf(mon, "Bytes saved: %"PRIu64"\n", info->bytes);
++ }
++ if (info->has_error) {
++ monitor_printf(mon, "Error: %s\n", info->error);
++ }
++}
++
+ void hmp_info_iothreads(Monitor *mon, const QDict *qdict)
+ {
+ IOThreadInfoList *info_list = qmp_query_iothreads(NULL);
+diff --git a/qapi/migration.json b/qapi/migration.json
+index 81185d4311..3129f71fa8 100644
+--- a/qapi/migration.json
++++ b/qapi/migration.json
+@@ -261,6 +261,40 @@
+ '*compression': 'CompressionStats',
+ '*socket-address': ['SocketAddress'] } }
+
++##
++# @SaveVMInfo:
++#
++# Information about current migration process.
++#
++# @status: string describing the current savevm status.
++# This can be 'active', 'completed', 'failed'.
++# If this field is not returned, no savevm process
++# has been initiated
++#
++# @error: string containing error message is status is failed.
++#
++# @total-time: total amount of milliseconds since savevm started.
++# If savevm has ended, it returns the total save time
++#
++# @bytes: total amount of data transfered
++#
++# Since: 1.3
++##
++{ 'struct': 'SaveVMInfo',
++ 'data': {'*status': 'str', '*error': 'str',
++ '*total-time': 'int', '*bytes': 'int'} }
++
++##
++# @query-savevm:
++#
++# Returns information about current savevm process.
++#
++# Returns: @SaveVMInfo
++#
++# Since: 1.3
++##
++{ 'command': 'query-savevm', 'returns': 'SaveVMInfo' }
++
+ ##
+ # @query-migrate:
+ #
+diff --git a/qapi/misc.json b/qapi/misc.json
+index 27ef5a2b20..b3ce75dcae 100644
+--- a/qapi/misc.json
++++ b/qapi/misc.json
+@@ -435,6 +435,38 @@
+ ##
+ { 'command': 'query-fdsets', 'returns': ['FdsetInfo'] }
+
++##
++# @savevm-start:
++#
++# Prepare for snapshot and halt VM. Save VM state to statefile.
++#
++##
++{ 'command': 'savevm-start', 'data': { '*statefile': 'str' } }
++
++##
++# @snapshot-drive:
++#
++# Create an internal drive snapshot.
++#
++##
++{ 'command': 'snapshot-drive', 'data': { 'device': 'str', 'name': 'str' } }
++
++##
++# @delete-drive-snapshot:
++#
++# Delete a drive snapshot.
++#
++##
++{ 'command': 'delete-drive-snapshot', 'data': { 'device': 'str', 'name': 'str' } }
++
++##
++# @savevm-end:
++#
++# Resume VM after a snapshot.
++#
++##
++{ 'command': 'savevm-end', 'coroutine': true }
++
+ ##
+ # @CommandLineParameterType:
+ #
+diff --git a/qemu-options.hx b/qemu-options.hx
+index 31c04f7eea..c2ca6e91b5 100644
+--- a/qemu-options.hx
++++ b/qemu-options.hx
+@@ -4341,6 +4341,18 @@ SRST
+ Start right away with a saved state (``loadvm`` in monitor)
+ ERST
+
++DEF("loadstate", HAS_ARG, QEMU_OPTION_loadstate, \
++ "-loadstate file\n" \
++ " start right away with a saved state\n",
++ QEMU_ARCH_ALL)
++SRST
++``-loadstate file``
++ Start right away with a saved state. This option does not rollback
++ disk state like @code{loadvm}, so user must make sure that disk
++ have correct state. @var{file} can be any valid device URL. See the section
++ for "Device URL Syntax" for more information.
++ERST
++
+ #ifndef _WIN32
+ DEF("daemonize", 0, QEMU_OPTION_daemonize, \
+ "-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
+diff --git a/softmmu/vl.c b/softmmu/vl.c
+index 706bd7cff7..b8637c4262 100644
+--- a/softmmu/vl.c
++++ b/softmmu/vl.c
+@@ -165,6 +165,7 @@ static const char *accelerators;
+ static bool have_custom_ram_size;
+ static const char *ram_memdev_id;
+ static QDict *machine_opts_dict;
++static const char *loadstate;
+ static QTAILQ_HEAD(, ObjectOption) object_opts = QTAILQ_HEAD_INITIALIZER(object_opts);
+ static QTAILQ_HEAD(, DeviceOption) device_opts = QTAILQ_HEAD_INITIALIZER(device_opts);
+ static int display_remote;
+@@ -2584,6 +2585,12 @@ void qmp_x_exit_preconfig(Error **errp)
+
+ if (loadvm) {
+ load_snapshot(loadvm, NULL, false, NULL, &error_fatal);
++ } else if (loadstate) {
++ Error *local_err = NULL;
++ if (load_snapshot_from_blockdev(loadstate, &local_err) < 0) {
++ error_report_err(local_err);
++ autostart = 0;
++ }
+ }
+ if (replay_mode != REPLAY_MODE_NONE) {
+ replay_vmstate_init();
+@@ -3133,6 +3140,9 @@ void qemu_init(int argc, char **argv, char **envp)
+ case QEMU_OPTION_loadvm:
+ loadvm = optarg;
+ break;
++ case QEMU_OPTION_loadstate:
++ loadstate = optarg;
++ break;
+ case QEMU_OPTION_full_screen:
+ dpy.has_full_screen = true;
+ dpy.full_screen = true;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Mon, 4 May 2020 11:05:08 +0200
+Subject: [PATCH] PVE: add optional buffer size to QEMUFile
+
+So we can use a 4M buffer for savevm-async which should
+increase performance storing the state onto ceph.
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+[increase max IOV count in QEMUFile to actually write more data]
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: adapt to removal of QEMUFileOps]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ migration/qemu-file.c | 49 +++++++++++++++++++++++++++-------------
+ migration/qemu-file.h | 2 ++
+ migration/savevm-async.c | 5 ++--
+ 3 files changed, 38 insertions(+), 18 deletions(-)
+
+diff --git a/migration/qemu-file.c b/migration/qemu-file.c
+index 4f400c2e52..21e8998867 100644
+--- a/migration/qemu-file.c
++++ b/migration/qemu-file.c
+@@ -31,8 +31,8 @@
+ #include "trace.h"
+ #include "qapi/error.h"
+
+-#define IO_BUF_SIZE 32768
+-#define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 64)
++#define DEFAULT_IO_BUF_SIZE 32768
++#define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 256)
+
+ struct QEMUFile {
+ const QEMUFileHooks *hooks;
+@@ -55,7 +55,8 @@ struct QEMUFile {
+
+ int buf_index;
+ int buf_size; /* 0 when writing */
+- uint8_t buf[IO_BUF_SIZE];
++ size_t buf_allocated_size;
++ uint8_t *buf;
+
+ DECLARE_BITMAP(may_free, MAX_IOV_SIZE);
+ struct iovec iov[MAX_IOV_SIZE];
+@@ -106,7 +107,9 @@ bool qemu_file_mode_is_not_valid(const char *mode)
+ return false;
+ }
+
+-static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
++static QEMUFile *qemu_file_new_impl(QIOChannel *ioc,
++ bool is_writable,
++ size_t buffer_size)
+ {
+ QEMUFile *f;
+
+@@ -115,6 +118,8 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
+ object_ref(ioc);
+ f->ioc = ioc;
+ f->is_writable = is_writable;
++ f->buf_allocated_size = buffer_size;
++ f->buf = malloc(buffer_size);
+
+ return f;
+ }
+@@ -125,17 +130,27 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
+ */
+ QEMUFile *qemu_file_get_return_path(QEMUFile *f)
+ {
+- return qemu_file_new_impl(f->ioc, !f->is_writable);
++ return qemu_file_new_impl(f->ioc, !f->is_writable, DEFAULT_IO_BUF_SIZE);
+ }
+
+ QEMUFile *qemu_file_new_output(QIOChannel *ioc)
+ {
+- return qemu_file_new_impl(ioc, true);
++ return qemu_file_new_impl(ioc, true, DEFAULT_IO_BUF_SIZE);
++}
++
++QEMUFile *qemu_file_new_output_sized(QIOChannel *ioc, size_t buffer_size)
++{
++ return qemu_file_new_impl(ioc, true, buffer_size);
+ }
+
+ QEMUFile *qemu_file_new_input(QIOChannel *ioc)
+ {
+- return qemu_file_new_impl(ioc, false);
++ return qemu_file_new_impl(ioc, false, DEFAULT_IO_BUF_SIZE);
++}
++
++QEMUFile *qemu_file_new_input_sized(QIOChannel *ioc, size_t buffer_size)
++{
++ return qemu_file_new_impl(ioc, false, buffer_size);
+ }
+
+ void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks)
+@@ -393,7 +408,7 @@ static ssize_t qemu_fill_buffer(QEMUFile *f)
+ do {
+ len = qio_channel_read(f->ioc,
+ (char *)f->buf + pending,
+- IO_BUF_SIZE - pending,
++ f->buf_allocated_size - pending,
+ &local_error);
+ if (len == QIO_CHANNEL_ERR_BLOCK) {
+ if (qemu_in_coroutine()) {
+@@ -443,6 +458,8 @@ int qemu_fclose(QEMUFile *f)
+ }
+ g_clear_pointer(&f->ioc, object_unref);
+
++ free(f->buf);
++
+ /* If any error was spotted before closing, we should report it
+ * instead of the close() return value.
+ */
+@@ -497,7 +514,7 @@ static void add_buf_to_iovec(QEMUFile *f, size_t len)
+ {
+ if (!add_to_iovec(f, f->buf + f->buf_index, len, false)) {
+ f->buf_index += len;
+- if (f->buf_index == IO_BUF_SIZE) {
++ if (f->buf_index == f->buf_allocated_size) {
+ qemu_fflush(f);
+ }
+ }
+@@ -523,7 +540,7 @@ void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, size_t size)
+ }
+
+ while (size > 0) {
+- l = IO_BUF_SIZE - f->buf_index;
++ l = f->buf_allocated_size - f->buf_index;
+ if (l > size) {
+ l = size;
+ }
+@@ -570,8 +587,8 @@ size_t qemu_peek_buffer(QEMUFile *f, uint8_t **buf, size_t size, size_t offset)
+ size_t index;
+
+ assert(!qemu_file_is_writable(f));
+- assert(offset < IO_BUF_SIZE);
+- assert(size <= IO_BUF_SIZE - offset);
++ assert(offset < f->buf_allocated_size);
++ assert(size <= f->buf_allocated_size - offset);
+
+ /* The 1st byte to read from */
+ index = f->buf_index + offset;
+@@ -621,7 +638,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
+ size_t res;
+ uint8_t *src;
+
+- res = qemu_peek_buffer(f, &src, MIN(pending, IO_BUF_SIZE), 0);
++ res = qemu_peek_buffer(f, &src, MIN(pending, f->buf_allocated_size), 0);
+ if (res == 0) {
+ return done;
+ }
+@@ -655,7 +672,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
+ */
+ size_t qemu_get_buffer_in_place(QEMUFile *f, uint8_t **buf, size_t size)
+ {
+- if (size < IO_BUF_SIZE) {
++ if (size < f->buf_allocated_size) {
+ size_t res;
+ uint8_t *src = NULL;
+
+@@ -680,7 +697,7 @@ int qemu_peek_byte(QEMUFile *f, int offset)
+ int index = f->buf_index + offset;
+
+ assert(!qemu_file_is_writable(f));
+- assert(offset < IO_BUF_SIZE);
++ assert(offset < f->buf_allocated_size);
+
+ if (index >= f->buf_size) {
+ qemu_fill_buffer(f);
+@@ -832,7 +849,7 @@ static int qemu_compress_data(z_stream *stream, uint8_t *dest, size_t dest_len,
+ ssize_t qemu_put_compression_data(QEMUFile *f, z_stream *stream,
+ const uint8_t *p, size_t size)
+ {
+- ssize_t blen = IO_BUF_SIZE - f->buf_index - sizeof(int32_t);
++ ssize_t blen = f->buf_allocated_size - f->buf_index - sizeof(int32_t);
+
+ if (blen < compressBound(size)) {
+ return -1;
+diff --git a/migration/qemu-file.h b/migration/qemu-file.h
+index fa13d04d78..914f1a63a8 100644
+--- a/migration/qemu-file.h
++++ b/migration/qemu-file.h
+@@ -63,7 +63,9 @@ typedef struct QEMUFileHooks {
+ } QEMUFileHooks;
+
+ QEMUFile *qemu_file_new_input(QIOChannel *ioc);
++QEMUFile *qemu_file_new_input_sized(QIOChannel *ioc, size_t buffer_size);
+ QEMUFile *qemu_file_new_output(QIOChannel *ioc);
++QEMUFile *qemu_file_new_output_sized(QIOChannel *ioc, size_t buffer_size);
+ void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks);
+ int qemu_fclose(QEMUFile *f);
+
+diff --git a/migration/savevm-async.c b/migration/savevm-async.c
+index b3692739a0..e65a5e3482 100644
+--- a/migration/savevm-async.c
++++ b/migration/savevm-async.c
+@@ -367,7 +367,7 @@ void qmp_savevm_start(bool has_statefile, const char *statefile, Error **errp)
+
+ QIOChannel *ioc = QIO_CHANNEL(qio_channel_savevm_async_new(snap_state.target,
+ &snap_state.bs_pos));
+- snap_state.file = qemu_file_new_output(ioc);
++ snap_state.file = qemu_file_new_output_sized(ioc, 4 * 1024 * 1024);
+
+ if (!snap_state.file) {
+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
+@@ -500,7 +500,8 @@ int load_snapshot_from_blockdev(const char *filename, Error **errp)
+ blk_op_block_all(be, blocker);
+
+ /* restore the VM state */
+- f = qemu_file_new_input(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)));
++ f = qemu_file_new_input_sized(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)),
++ 4 * 1024 * 1024);
+ if (!f) {
+ error_setg(errp, "Could not open VM state file");
+ goto the_end;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:47 +0200
-Subject: [PATCH] PVE: block: add the zeroinit block driver filter
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[adapt to changed function signatures]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/meson.build | 1 +
- block/zeroinit.c | 196 ++++++++++++++++++++++++++++++++++++++++++++++
- 2 files changed, 197 insertions(+)
- create mode 100644 block/zeroinit.c
-
-diff --git a/block/meson.build b/block/meson.build
-index 0b2a60c99b..9451fc3ccd 100644
---- a/block/meson.build
-+++ b/block/meson.build
-@@ -43,6 +43,7 @@ block_ss.add(files(
- 'vmdk.c',
- 'vpc.c',
- 'write-threshold.c',
-+ 'zeroinit.c',
- ), zstd, zlib, gnutls)
-
- softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
-diff --git a/block/zeroinit.c b/block/zeroinit.c
-new file mode 100644
-index 0000000000..20ee611f22
---- /dev/null
-+++ b/block/zeroinit.c
-@@ -0,0 +1,196 @@
-+/*
-+ * Filter to fake a zero-initialized block device.
-+ *
-+ * Copyright (c) 2016 Wolfgang Bumiller <w.bumiller@proxmox.com>
-+ * Copyright (c) 2016 Proxmox Server Solutions GmbH
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ */
-+
-+#include "qemu/osdep.h"
-+#include "qapi/error.h"
-+#include "block/block_int.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qstring.h"
-+#include "qemu/cutils.h"
-+#include "qemu/option.h"
-+#include "qemu/module.h"
-+
-+typedef struct {
-+ bool has_zero_init;
-+ int64_t extents;
-+} BDRVZeroinitState;
-+
-+/* Valid blkverify filenames look like blkverify:path/to/raw_image:path/to/image */
-+static void zeroinit_parse_filename(const char *filename, QDict *options,
-+ Error **errp)
-+{
-+ QString *raw_path;
-+
-+ /* Parse the blkverify: prefix */
-+ if (!strstart(filename, "zeroinit:", &filename)) {
-+ /* There was no prefix; therefore, all options have to be already
-+ present in the QDict (except for the filename) */
-+ return;
-+ }
-+
-+ raw_path = qstring_from_str(filename);
-+ qdict_put(options, "x-next", raw_path);
-+}
-+
-+static QemuOptsList runtime_opts = {
-+ .name = "zeroinit",
-+ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
-+ .desc = {
-+ {
-+ .name = "x-next",
-+ .type = QEMU_OPT_STRING,
-+ .help = "[internal use only, will be removed]",
-+ },
-+ {
-+ .name = "x-zeroinit",
-+ .type = QEMU_OPT_BOOL,
-+ .help = "set has_initialized_zero flag",
-+ },
-+ { /* end of list */ }
-+ },
-+};
-+
-+static int zeroinit_open(BlockDriverState *bs, QDict *options, int flags,
-+ Error **errp)
-+{
-+ BDRVZeroinitState *s = bs->opaque;
-+ QemuOpts *opts;
-+ Error *local_err = NULL;
-+ int ret;
-+
-+ s->extents = 0;
-+
-+ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
-+ qemu_opts_absorb_qdict(opts, options, &local_err);
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ ret = -EINVAL;
-+ goto fail;
-+ }
-+
-+ /* Open the raw file */
-+ bs->file = bdrv_open_child(qemu_opt_get(opts, "x-next"), options, "next",
-+ bs, &child_of_bds, BDRV_CHILD_FILTERED, false, &local_err);
-+ if (local_err) {
-+ ret = -EINVAL;
-+ error_propagate(errp, local_err);
-+ goto fail;
-+ }
-+
-+ /* set the options */
-+ s->has_zero_init = qemu_opt_get_bool(opts, "x-zeroinit", true);
-+
-+ ret = 0;
-+fail:
-+ if (ret < 0) {
-+ bdrv_unref_child(bs, bs->file);
-+ }
-+ qemu_opts_del(opts);
-+ return ret;
-+}
-+
-+static void zeroinit_close(BlockDriverState *bs)
-+{
-+ BDRVZeroinitState *s = bs->opaque;
-+ (void)s;
-+}
-+
-+static int64_t zeroinit_getlength(BlockDriverState *bs)
-+{
-+ return bdrv_getlength(bs->file->bs);
-+}
-+
-+static int coroutine_fn zeroinit_co_preadv(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
-+{
-+ return bdrv_co_preadv(bs->file, offset, bytes, qiov, flags);
-+}
-+
-+static int coroutine_fn zeroinit_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
-+ int64_t bytes, BdrvRequestFlags flags)
-+{
-+ BDRVZeroinitState *s = bs->opaque;
-+ if (offset >= s->extents)
-+ return 0;
-+ return bdrv_pwrite_zeroes(bs->file, offset, bytes, flags);
-+}
-+
-+static int coroutine_fn zeroinit_co_pwritev(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
-+{
-+ BDRVZeroinitState *s = bs->opaque;
-+ int64_t extents = offset + bytes;
-+ if (extents > s->extents)
-+ s->extents = extents;
-+ return bdrv_co_pwritev(bs->file, offset, bytes, qiov, flags);
-+}
-+
-+static coroutine_fn int zeroinit_co_flush(BlockDriverState *bs)
-+{
-+ return bdrv_co_flush(bs->file->bs);
-+}
-+
-+static int zeroinit_has_zero_init(BlockDriverState *bs)
-+{
-+ BDRVZeroinitState *s = bs->opaque;
-+ return s->has_zero_init;
-+}
-+
-+static int coroutine_fn zeroinit_co_pdiscard(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes)
-+{
-+ return bdrv_co_pdiscard(bs->file, offset, bytes);
-+}
-+
-+static int zeroinit_co_truncate(BlockDriverState *bs, int64_t offset,
-+ _Bool exact, PreallocMode prealloc,
-+ BdrvRequestFlags req_flags, Error **errp)
-+{
-+ return bdrv_co_truncate(bs->file, offset, exact, prealloc, req_flags, errp);
-+}
-+
-+static int zeroinit_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
-+{
-+ return bdrv_get_info(bs->file->bs, bdi);
-+}
-+
-+static BlockDriver bdrv_zeroinit = {
-+ .format_name = "zeroinit",
-+ .protocol_name = "zeroinit",
-+ .instance_size = sizeof(BDRVZeroinitState),
-+
-+ .bdrv_parse_filename = zeroinit_parse_filename,
-+ .bdrv_file_open = zeroinit_open,
-+ .bdrv_close = zeroinit_close,
-+ .bdrv_getlength = zeroinit_getlength,
-+ .bdrv_child_perm = bdrv_default_perms,
-+ .bdrv_co_flush_to_disk = zeroinit_co_flush,
-+
-+ .bdrv_co_pwrite_zeroes = zeroinit_co_pwrite_zeroes,
-+ .bdrv_co_pwritev = zeroinit_co_pwritev,
-+ .bdrv_co_preadv = zeroinit_co_preadv,
-+ .bdrv_co_flush = zeroinit_co_flush,
-+
-+ .is_filter = true,
-+
-+ .bdrv_has_zero_init = zeroinit_has_zero_init,
-+
-+ .bdrv_co_pdiscard = zeroinit_co_pdiscard,
-+
-+ .bdrv_co_truncate = zeroinit_co_truncate,
-+ .bdrv_get_info = zeroinit_get_info,
-+};
-+
-+static void bdrv_zeroinit_init(void)
-+{
-+ bdrv_register(&bdrv_zeroinit);
-+}
-+
-+block_init(bdrv_zeroinit_init);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:48 +0200
-Subject: [PATCH] PVE: Add dummy -id command line parameter
-
-This used to be part of the qemu-side PVE authentication for
-VNC. Now this does nothing.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- qemu-options.hx | 3 +++
- softmmu/vl.c | 8 ++++++++
- 2 files changed, 11 insertions(+)
-
-diff --git a/qemu-options.hx b/qemu-options.hx
-index aeade4ef80..a26f0b0400 100644
---- a/qemu-options.hx
-+++ b/qemu-options.hx
-@@ -1075,6 +1075,9 @@ DEFHEADING()
-
- DEFHEADING(Block device options:)
-
-+DEF("id", HAS_ARG, QEMU_OPTION_id,
-+ "-id n set the VMID", QEMU_ARCH_ALL)
-+
- DEF("fda", HAS_ARG, QEMU_OPTION_fda,
- "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
- DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
-diff --git a/softmmu/vl.c b/softmmu/vl.c
-index a3f2a3818c..4208142685 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -2785,6 +2785,7 @@ void qemu_init(int argc, char **argv, char **envp)
- MachineClass *machine_class;
- bool userconfig = true;
- FILE *vmstate_dump_file = NULL;
-+ long vm_id;
-
- qemu_add_opts(&qemu_drive_opts);
- qemu_add_drive_opts(&qemu_legacy_drive_opts);
-@@ -3420,6 +3421,13 @@ void qemu_init(int argc, char **argv, char **envp)
- machine_parse_property_opt(qemu_find_opts("smp-opts"),
- "smp", optarg);
- break;
-+ case QEMU_OPTION_id:
-+ vm_id = strtol(optarg, (char **)&optarg, 10);
-+ if (*optarg != 0 || vm_id < 100 || vm_id > INT_MAX) {
-+ error_report("invalid -id argument %s", optarg);
-+ exit(1);
-+ }
-+ break;
- case QEMU_OPTION_vnc:
- vnc_parse(optarg);
- break;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:47 +0200
+Subject: [PATCH] PVE: block: add the zeroinit block driver filter
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[adapt to changed function signatures]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/meson.build | 1 +
+ block/zeroinit.c | 196 ++++++++++++++++++++++++++++++++++++++++++++++
+ 2 files changed, 197 insertions(+)
+ create mode 100644 block/zeroinit.c
+
+diff --git a/block/meson.build b/block/meson.build
+index 60bc305597..ad40c10b6a 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -43,6 +43,7 @@ block_ss.add(files(
+ 'vmdk.c',
+ 'vpc.c',
+ 'write-threshold.c',
++ 'zeroinit.c',
+ ), zstd, zlib, gnutls)
+
+ softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+diff --git a/block/zeroinit.c b/block/zeroinit.c
+new file mode 100644
+index 0000000000..20ee611f22
+--- /dev/null
++++ b/block/zeroinit.c
+@@ -0,0 +1,196 @@
++/*
++ * Filter to fake a zero-initialized block device.
++ *
++ * Copyright (c) 2016 Wolfgang Bumiller <w.bumiller@proxmox.com>
++ * Copyright (c) 2016 Proxmox Server Solutions GmbH
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ */
++
++#include "qemu/osdep.h"
++#include "qapi/error.h"
++#include "block/block_int.h"
++#include "qapi/qmp/qdict.h"
++#include "qapi/qmp/qstring.h"
++#include "qemu/cutils.h"
++#include "qemu/option.h"
++#include "qemu/module.h"
++
++typedef struct {
++ bool has_zero_init;
++ int64_t extents;
++} BDRVZeroinitState;
++
++/* Valid blkverify filenames look like blkverify:path/to/raw_image:path/to/image */
++static void zeroinit_parse_filename(const char *filename, QDict *options,
++ Error **errp)
++{
++ QString *raw_path;
++
++ /* Parse the blkverify: prefix */
++ if (!strstart(filename, "zeroinit:", &filename)) {
++ /* There was no prefix; therefore, all options have to be already
++ present in the QDict (except for the filename) */
++ return;
++ }
++
++ raw_path = qstring_from_str(filename);
++ qdict_put(options, "x-next", raw_path);
++}
++
++static QemuOptsList runtime_opts = {
++ .name = "zeroinit",
++ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
++ .desc = {
++ {
++ .name = "x-next",
++ .type = QEMU_OPT_STRING,
++ .help = "[internal use only, will be removed]",
++ },
++ {
++ .name = "x-zeroinit",
++ .type = QEMU_OPT_BOOL,
++ .help = "set has_initialized_zero flag",
++ },
++ { /* end of list */ }
++ },
++};
++
++static int zeroinit_open(BlockDriverState *bs, QDict *options, int flags,
++ Error **errp)
++{
++ BDRVZeroinitState *s = bs->opaque;
++ QemuOpts *opts;
++ Error *local_err = NULL;
++ int ret;
++
++ s->extents = 0;
++
++ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
++ qemu_opts_absorb_qdict(opts, options, &local_err);
++ if (local_err) {
++ error_propagate(errp, local_err);
++ ret = -EINVAL;
++ goto fail;
++ }
++
++ /* Open the raw file */
++ bs->file = bdrv_open_child(qemu_opt_get(opts, "x-next"), options, "next",
++ bs, &child_of_bds, BDRV_CHILD_FILTERED, false, &local_err);
++ if (local_err) {
++ ret = -EINVAL;
++ error_propagate(errp, local_err);
++ goto fail;
++ }
++
++ /* set the options */
++ s->has_zero_init = qemu_opt_get_bool(opts, "x-zeroinit", true);
++
++ ret = 0;
++fail:
++ if (ret < 0) {
++ bdrv_unref_child(bs, bs->file);
++ }
++ qemu_opts_del(opts);
++ return ret;
++}
++
++static void zeroinit_close(BlockDriverState *bs)
++{
++ BDRVZeroinitState *s = bs->opaque;
++ (void)s;
++}
++
++static int64_t zeroinit_getlength(BlockDriverState *bs)
++{
++ return bdrv_getlength(bs->file->bs);
++}
++
++static int coroutine_fn zeroinit_co_preadv(BlockDriverState *bs,
++ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++{
++ return bdrv_co_preadv(bs->file, offset, bytes, qiov, flags);
++}
++
++static int coroutine_fn zeroinit_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
++ int64_t bytes, BdrvRequestFlags flags)
++{
++ BDRVZeroinitState *s = bs->opaque;
++ if (offset >= s->extents)
++ return 0;
++ return bdrv_pwrite_zeroes(bs->file, offset, bytes, flags);
++}
++
++static int coroutine_fn zeroinit_co_pwritev(BlockDriverState *bs,
++ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++{
++ BDRVZeroinitState *s = bs->opaque;
++ int64_t extents = offset + bytes;
++ if (extents > s->extents)
++ s->extents = extents;
++ return bdrv_co_pwritev(bs->file, offset, bytes, qiov, flags);
++}
++
++static coroutine_fn int zeroinit_co_flush(BlockDriverState *bs)
++{
++ return bdrv_co_flush(bs->file->bs);
++}
++
++static int zeroinit_has_zero_init(BlockDriverState *bs)
++{
++ BDRVZeroinitState *s = bs->opaque;
++ return s->has_zero_init;
++}
++
++static int coroutine_fn zeroinit_co_pdiscard(BlockDriverState *bs,
++ int64_t offset, int64_t bytes)
++{
++ return bdrv_co_pdiscard(bs->file, offset, bytes);
++}
++
++static int zeroinit_co_truncate(BlockDriverState *bs, int64_t offset,
++ _Bool exact, PreallocMode prealloc,
++ BdrvRequestFlags req_flags, Error **errp)
++{
++ return bdrv_co_truncate(bs->file, offset, exact, prealloc, req_flags, errp);
++}
++
++static int zeroinit_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
++{
++ return bdrv_get_info(bs->file->bs, bdi);
++}
++
++static BlockDriver bdrv_zeroinit = {
++ .format_name = "zeroinit",
++ .protocol_name = "zeroinit",
++ .instance_size = sizeof(BDRVZeroinitState),
++
++ .bdrv_parse_filename = zeroinit_parse_filename,
++ .bdrv_file_open = zeroinit_open,
++ .bdrv_close = zeroinit_close,
++ .bdrv_getlength = zeroinit_getlength,
++ .bdrv_child_perm = bdrv_default_perms,
++ .bdrv_co_flush_to_disk = zeroinit_co_flush,
++
++ .bdrv_co_pwrite_zeroes = zeroinit_co_pwrite_zeroes,
++ .bdrv_co_pwritev = zeroinit_co_pwritev,
++ .bdrv_co_preadv = zeroinit_co_preadv,
++ .bdrv_co_flush = zeroinit_co_flush,
++
++ .is_filter = true,
++
++ .bdrv_has_zero_init = zeroinit_has_zero_init,
++
++ .bdrv_co_pdiscard = zeroinit_co_pdiscard,
++
++ .bdrv_co_truncate = zeroinit_co_truncate,
++ .bdrv_get_info = zeroinit_get_info,
++};
++
++static void bdrv_zeroinit_init(void)
++{
++ bdrv_register(&bdrv_zeroinit);
++}
++
++block_init(bdrv_zeroinit_init);
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:48 +0200
+Subject: [PATCH] PVE: Add dummy -id command line parameter
+
+This used to be part of the qemu-side PVE authentication for
+VNC. Now this does nothing.
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ qemu-options.hx | 3 +++
+ softmmu/vl.c | 8 ++++++++
+ 2 files changed, 11 insertions(+)
+
+diff --git a/qemu-options.hx b/qemu-options.hx
+index c2ca6e91b5..ab4734ef32 100644
+--- a/qemu-options.hx
++++ b/qemu-options.hx
+@@ -1118,6 +1118,9 @@ backend describes how QEMU handles the data.
+
+ ERST
+
++DEF("id", HAS_ARG, QEMU_OPTION_id,
++ "-id n set the VMID", QEMU_ARCH_ALL)
++
+ DEF("fda", HAS_ARG, QEMU_OPTION_fda,
+ "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
+ DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
+diff --git a/softmmu/vl.c b/softmmu/vl.c
+index b8637c4262..39f149924e 100644
+--- a/softmmu/vl.c
++++ b/softmmu/vl.c
+@@ -2620,6 +2620,7 @@ void qemu_init(int argc, char **argv, char **envp)
+ MachineClass *machine_class;
+ bool userconfig = true;
+ FILE *vmstate_dump_file = NULL;
++ long vm_id;
+
+ qemu_add_opts(&qemu_drive_opts);
+ qemu_add_drive_opts(&qemu_legacy_drive_opts);
+@@ -3245,6 +3246,13 @@ void qemu_init(int argc, char **argv, char **envp)
+ machine_parse_property_opt(qemu_find_opts("smp-opts"),
+ "smp", optarg);
+ break;
++ case QEMU_OPTION_id:
++ vm_id = strtol(optarg, (char **)&optarg, 10);
++ if (*optarg != 0 || vm_id < 100 || vm_id > INT_MAX) {
++ error_report("invalid -id argument %s", optarg);
++ exit(1);
++ }
++ break;
+ case QEMU_OPTION_vnc:
+ vnc_parse(optarg);
+ break;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:49 +0200
-Subject: [PATCH] PVE: [Config] Revert "target-i386: disable LINT0 after reset"
-
-This reverts commit b8eb5512fd8a115f164edbbe897cdf8884920ccb.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/intc/apic_common.c | 9 +++++++++
- 1 file changed, 9 insertions(+)
-
-diff --git a/hw/intc/apic_common.c b/hw/intc/apic_common.c
-index 2a20982066..7968ad5a93 100644
---- a/hw/intc/apic_common.c
-+++ b/hw/intc/apic_common.c
-@@ -278,6 +278,15 @@ static void apic_reset_common(DeviceState *dev)
- info->vapic_base_update(s);
-
- apic_init_reset(dev);
-+
-+ if (bsp) {
-+ /*
-+ * LINT0 delivery mode on CPU #0 is set to ExtInt at initialization
-+ * time typically by BIOS, so PIC interrupt can be delivered to the
-+ * processor when local APIC is enabled.
-+ */
-+ s->lvt[APIC_LVT_LINT0] = 0x700;
-+ }
- }
-
- static const VMStateDescription vmstate_apic_common;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:49 +0200
+Subject: [PATCH] PVE: [Config] Revert "target-i386: disable LINT0 after reset"
+
+This reverts commit b8eb5512fd8a115f164edbbe897cdf8884920ccb.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ hw/intc/apic_common.c | 9 +++++++++
+ 1 file changed, 9 insertions(+)
+
+diff --git a/hw/intc/apic_common.c b/hw/intc/apic_common.c
+index 2a20982066..7968ad5a93 100644
+--- a/hw/intc/apic_common.c
++++ b/hw/intc/apic_common.c
+@@ -278,6 +278,15 @@ static void apic_reset_common(DeviceState *dev)
+ info->vapic_base_update(s);
+
+ apic_init_reset(dev);
++
++ if (bsp) {
++ /*
++ * LINT0 delivery mode on CPU #0 is set to ExtInt at initialization
++ * time typically by BIOS, so PIC interrupt can be delivered to the
++ * processor when local APIC is enabled.
++ */
++ s->lvt[APIC_LVT_LINT0] = 0x700;
++ }
+ }
+
+ static const VMStateDescription vmstate_apic_common;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:50 +0200
-Subject: [PATCH] PVE: [Up+Config] file-posix: make locking optiono on create
-
-Otherwise creating images on nfs/cifs can be problematic.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/file-posix.c | 59 ++++++++++++++++++++++++++++++--------------
- qapi/block-core.json | 3 ++-
- 2 files changed, 42 insertions(+), 20 deletions(-)
-
-diff --git a/block/file-posix.c b/block/file-posix.c
-index e5bf5d59bf..b013668dce 100644
---- a/block/file-posix.c
-+++ b/block/file-posix.c
-@@ -2461,6 +2461,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
- int fd;
- uint64_t perm, shared;
- int result = 0;
-+ bool locked = false;
-
- /* Validate options and set default values */
- assert(options->driver == BLOCKDEV_DRIVER_FILE);
-@@ -2501,19 +2502,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
- perm = BLK_PERM_WRITE | BLK_PERM_RESIZE;
- shared = BLK_PERM_ALL & ~BLK_PERM_RESIZE;
-
-- /* Step one: Take locks */
-- result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
-- if (result < 0) {
-- goto out_close;
-- }
-+ if (file_opts->locking != ON_OFF_AUTO_OFF) {
-+ /* Step one: Take locks */
-+ result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
-+ if (result < 0) {
-+ goto out_close;
-+ }
-+ locked = true;
-
-- /* Step two: Check that nobody else has taken conflicting locks */
-- result = raw_check_lock_bytes(fd, perm, shared, errp);
-- if (result < 0) {
-- error_append_hint(errp,
-- "Is another process using the image [%s]?\n",
-- file_opts->filename);
-- goto out_unlock;
-+ /* Step two: Check that nobody else has taken conflicting locks */
-+ result = raw_check_lock_bytes(fd, perm, shared, errp);
-+ if (result < 0) {
-+ error_append_hint(errp,
-+ "Is another process using the image [%s]?\n",
-+ file_opts->filename);
-+ goto out_unlock;
-+ }
- }
-
- /* Clear the file by truncating it to 0 */
-@@ -2567,13 +2571,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
- }
-
- out_unlock:
-- raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
-- if (local_err) {
-- /* The above call should not fail, and if it does, that does
-- * not mean the whole creation operation has failed. So
-- * report it the user for their convenience, but do not report
-- * it to the caller. */
-- warn_report_err(local_err);
-+ if (locked) {
-+ raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
-+ if (local_err) {
-+ /* The above call should not fail, and if it does, that does
-+ * not mean the whole creation operation has failed. So
-+ * report it the user for their convenience, but do not report
-+ * it to the caller. */
-+ warn_report_err(local_err);
-+ }
- }
-
- out_close:
-@@ -2598,6 +2604,7 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
- PreallocMode prealloc;
- char *buf = NULL;
- Error *local_err = NULL;
-+ OnOffAuto locking;
-
- /* Skip file: protocol prefix */
- strstart(filename, "file:", &filename);
-@@ -2620,6 +2627,18 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
- return -EINVAL;
- }
-
-+ locking = qapi_enum_parse(&OnOffAuto_lookup,
-+ qemu_opt_get(opts, "locking"),
-+ ON_OFF_AUTO_AUTO, &local_err);
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ return -EINVAL;
-+ }
-+
-+ if (locking == ON_OFF_AUTO_AUTO) {
-+ locking = ON_OFF_AUTO_OFF;
-+ }
-+
- options = (BlockdevCreateOptions) {
- .driver = BLOCKDEV_DRIVER_FILE,
- .u.file = {
-@@ -2631,6 +2650,8 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
- .nocow = nocow,
- .has_extent_size_hint = has_extent_size_hint,
- .extent_size_hint = extent_size_hint,
-+ .has_locking = true,
-+ .locking = locking,
- },
- };
- return raw_co_create(&options, errp);
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index fb25c2b245..564b6b43f7 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -4484,7 +4484,8 @@
- 'size': 'size',
- '*preallocation': 'PreallocMode',
- '*nocow': 'bool',
-- '*extent-size-hint': 'size'} }
-+ '*extent-size-hint': 'size',
-+ '*locking': 'OnOffAuto' } }
-
- ##
- # @BlockdevCreateOptionsGluster:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:50 +0200
+Subject: [PATCH] PVE: [Up+Config] file-posix: make locking optiono on create
+
+Otherwise creating images on nfs/cifs can be problematic.
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/file-posix.c | 59 ++++++++++++++++++++++++++++++--------------
+ qapi/block-core.json | 3 ++-
+ 2 files changed, 42 insertions(+), 20 deletions(-)
+
+diff --git a/block/file-posix.c b/block/file-posix.c
+index 3d60b80286..49ee1db5f9 100644
+--- a/block/file-posix.c
++++ b/block/file-posix.c
+@@ -2475,6 +2475,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+ int fd;
+ uint64_t perm, shared;
+ int result = 0;
++ bool locked = false;
+
+ /* Validate options and set default values */
+ assert(options->driver == BLOCKDEV_DRIVER_FILE);
+@@ -2515,19 +2516,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+ perm = BLK_PERM_WRITE | BLK_PERM_RESIZE;
+ shared = BLK_PERM_ALL & ~BLK_PERM_RESIZE;
+
+- /* Step one: Take locks */
+- result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
+- if (result < 0) {
+- goto out_close;
+- }
++ if (file_opts->locking != ON_OFF_AUTO_OFF) {
++ /* Step one: Take locks */
++ result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
++ if (result < 0) {
++ goto out_close;
++ }
++ locked = true;
+
+- /* Step two: Check that nobody else has taken conflicting locks */
+- result = raw_check_lock_bytes(fd, perm, shared, errp);
+- if (result < 0) {
+- error_append_hint(errp,
+- "Is another process using the image [%s]?\n",
+- file_opts->filename);
+- goto out_unlock;
++ /* Step two: Check that nobody else has taken conflicting locks */
++ result = raw_check_lock_bytes(fd, perm, shared, errp);
++ if (result < 0) {
++ error_append_hint(errp,
++ "Is another process using the image [%s]?\n",
++ file_opts->filename);
++ goto out_unlock;
++ }
+ }
+
+ /* Clear the file by truncating it to 0 */
+@@ -2581,13 +2585,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+ }
+
+ out_unlock:
+- raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
+- if (local_err) {
+- /* The above call should not fail, and if it does, that does
+- * not mean the whole creation operation has failed. So
+- * report it the user for their convenience, but do not report
+- * it to the caller. */
+- warn_report_err(local_err);
++ if (locked) {
++ raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
++ if (local_err) {
++ /* The above call should not fail, and if it does, that does
++ * not mean the whole creation operation has failed. So
++ * report it the user for their convenience, but do not report
++ * it to the caller. */
++ warn_report_err(local_err);
++ }
+ }
+
+ out_close:
+@@ -2612,6 +2618,7 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+ PreallocMode prealloc;
+ char *buf = NULL;
+ Error *local_err = NULL;
++ OnOffAuto locking;
+
+ /* Skip file: protocol prefix */
+ strstart(filename, "file:", &filename);
+@@ -2634,6 +2641,18 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+ return -EINVAL;
+ }
+
++ locking = qapi_enum_parse(&OnOffAuto_lookup,
++ qemu_opt_get(opts, "locking"),
++ ON_OFF_AUTO_AUTO, &local_err);
++ if (local_err) {
++ error_propagate(errp, local_err);
++ return -EINVAL;
++ }
++
++ if (locking == ON_OFF_AUTO_AUTO) {
++ locking = ON_OFF_AUTO_OFF;
++ }
++
+ options = (BlockdevCreateOptions) {
+ .driver = BLOCKDEV_DRIVER_FILE,
+ .u.file = {
+@@ -2645,6 +2664,8 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+ .nocow = nocow,
+ .has_extent_size_hint = has_extent_size_hint,
+ .extent_size_hint = extent_size_hint,
++ .has_locking = true,
++ .locking = locking,
+ },
+ };
+ return raw_co_create(&options, errp);
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index e1857e7094..ddac91e8f6 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -4537,7 +4537,8 @@
+ 'size': 'size',
+ '*preallocation': 'PreallocMode',
+ '*nocow': 'bool',
+- '*extent-size-hint': 'size'} }
++ '*extent-size-hint': 'size',
++ '*locking': 'OnOffAuto' } }
+
+ ##
+ # @BlockdevCreateOptionsGluster:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:53 +0200
-Subject: [PATCH] PVE: monitor: disable oob capability
-
-A bisect revealed that commit 8258292e18c3
-("monitor: Remove "x-oob", offer capability "oob" unconditionally")
-causes unexpected hangs when restoring live snapshots from some
-types of block devices (particularly RBD).
-We need to figure out what's happnening there. For now, since we
-had this disabled before and probably don't need it now either,
-disable oob, so we can get a functioning qemu out...
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- monitor/qmp.c | 3 +--
- 1 file changed, 1 insertion(+), 2 deletions(-)
-
-diff --git a/monitor/qmp.c b/monitor/qmp.c
-index 6b8cfcf6d8..3ec67e32d3 100644
---- a/monitor/qmp.c
-+++ b/monitor/qmp.c
-@@ -519,8 +519,7 @@ void monitor_init_qmp(Chardev *chr, bool pretty, Error **errp)
- qemu_chr_fe_set_echo(&mon->common.chr, true);
-
- /* Note: we run QMP monitor in I/O thread when @chr supports that */
-- monitor_data_init(&mon->common, true, false,
-- qemu_chr_has_feature(chr, QEMU_CHAR_FEATURE_GCONTEXT));
-+ monitor_data_init(&mon->common, true, false, false);
-
- mon->pretty = pretty;
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:54 +0200
-Subject: [PATCH] PVE: [Compat]: 4.0 used balloon qemu-4-0-config-size false
- here
-
-The underlying issue why this change from upstream to us arised in
-the first place is that QEMU 4.0 was already released at the point we
-run into this migration issue, so we did the then obvious fallback to
-false for virtio-balloon-device qemu-4-0-config-size.
-
-QEMU made that switch back in 4.1, where it now uses a backward
-compatible mechanism to detect if the bigger CFG sizes should be
-used, i.e., checking the VIRTIO_BALLOON_F_PAGE_POISON or
-VIRTIO_BALLOON_F_FREE_PAGE_HINT balloon feature flags.
-As for them, upstream released version 4.0 had this to true they keep
-it to true in their compatibility record for the 4.0 machine, to
-allow live migrations from 4.0 to 4.1.
-As for us, downstream released version 4.0 (first public release of
-this QEMU) had this to false, we change it back to false again, for
-the same reason.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/core/machine.c | 3 ++-
- 1 file changed, 2 insertions(+), 1 deletion(-)
-
-diff --git a/hw/core/machine.c b/hw/core/machine.c
-index 1e23fdc14b..8a50445ee8 100644
---- a/hw/core/machine.c
-+++ b/hw/core/machine.c
-@@ -118,7 +118,8 @@ GlobalProperty hw_compat_4_0[] = {
- { "virtio-vga", "edid", "false" },
- { "virtio-gpu-device", "edid", "false" },
- { "virtio-device", "use-started", "false" },
-- { "virtio-balloon-device", "qemu-4-0-config-size", "true" },
-+ // PVE differed from upstream for 4.0 balloon cfg size
-+ { "virtio-balloon-device", "qemu-4-0-config-size", "false" },
- { "pl031", "migrate-tick-offset", "false" },
- };
- const size_t hw_compat_4_0_len = G_N_ELEMENTS(hw_compat_4_0);
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:53 +0200
+Subject: [PATCH] PVE: monitor: disable oob capability
+
+A bisect revealed that commit 8258292e18c3
+("monitor: Remove "x-oob", offer capability "oob" unconditionally")
+causes unexpected hangs when restoring live snapshots from some
+types of block devices (particularly RBD).
+We need to figure out what's happnening there. For now, since we
+had this disabled before and probably don't need it now either,
+disable oob, so we can get a functioning qemu out...
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ monitor/qmp.c | 3 +--
+ 1 file changed, 1 insertion(+), 2 deletions(-)
+
+diff --git a/monitor/qmp.c b/monitor/qmp.c
+index 6b8cfcf6d8..3ec67e32d3 100644
+--- a/monitor/qmp.c
++++ b/monitor/qmp.c
+@@ -519,8 +519,7 @@ void monitor_init_qmp(Chardev *chr, bool pretty, Error **errp)
+ qemu_chr_fe_set_echo(&mon->common.chr, true);
+
+ /* Note: we run QMP monitor in I/O thread when @chr supports that */
+- monitor_data_init(&mon->common, true, false,
+- qemu_chr_has_feature(chr, QEMU_CHAR_FEATURE_GCONTEXT));
++ monitor_data_init(&mon->common, true, false, false);
+
+ mon->pretty = pretty;
+
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:55 +0200
-Subject: [PATCH] PVE: Allow version code in machine type
-
-E.g. pc-i440fx-4.0+pve3 would print 'pve3' as version code while
-selecting pc-i440fx-4.0 as machine type.
-
-Version is made available as 'pve-version' in query-machines (same as,
-and only if 'is-current').
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/core/machine-qmp-cmds.c | 6 ++++++
- include/hw/boards.h | 2 ++
- qapi/machine.json | 4 +++-
- softmmu/vl.c | 25 +++++++++++++++++++++++++
- 4 files changed, 36 insertions(+), 1 deletion(-)
-
-diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
-index 76fff60a6b..ec9201fb9a 100644
---- a/hw/core/machine-qmp-cmds.c
-+++ b/hw/core/machine-qmp-cmds.c
-@@ -103,6 +103,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
- if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
- info->has_is_current = true;
- info->is_current = true;
-+
-+ // PVE version string only exists for current machine
-+ if (mc->pve_version) {
-+ info->has_pve_version = true;
-+ info->pve_version = g_strdup(mc->pve_version);
-+ }
- }
-
- if (mc->default_cpu_type) {
-diff --git a/include/hw/boards.h b/include/hw/boards.h
-index c92ac8815c..572d911ed4 100644
---- a/include/hw/boards.h
-+++ b/include/hw/boards.h
-@@ -230,6 +230,8 @@ struct MachineClass {
- const char *desc;
- const char *deprecation_reason;
-
-+ const char *pve_version;
-+
- void (*init)(MachineState *state);
- void (*reset)(MachineState *state);
- void (*wakeup)(MachineState *state);
-diff --git a/qapi/machine.json b/qapi/machine.json
-index ca133e68ce..39ba204b52 100644
---- a/qapi/machine.json
-+++ b/qapi/machine.json
-@@ -158,6 +158,8 @@
- #
- # @default-ram-id: the default ID of initial RAM memory backend (since 5.2)
- #
-+# @pve-version: custom PVE version suffix specified as 'machine+pveN'
-+#
- # Since: 1.2
- ##
- { 'struct': 'MachineInfo',
-@@ -165,7 +167,7 @@
- '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
- 'hotpluggable-cpus': 'bool', 'numa-mem-supported': 'bool',
- 'deprecated': 'bool', '*default-cpu-type': 'str',
-- '*default-ram-id': 'str' } }
-+ '*default-ram-id': 'str', '*pve-version': 'str' } }
-
- ##
- # @query-machines:
-diff --git a/softmmu/vl.c b/softmmu/vl.c
-index 4208142685..37d85f1b96 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -1659,6 +1659,7 @@ static const QEMUOption *lookup_opt(int argc, char **argv,
- static MachineClass *select_machine(QDict *qdict, Error **errp)
- {
- const char *optarg = qdict_get_try_str(qdict, "type");
-+ const char *pvever = qdict_get_try_str(qdict, "pvever");
- GSList *machines = object_class_get_list(TYPE_MACHINE, false);
- MachineClass *machine_class;
- Error *local_err = NULL;
-@@ -1676,6 +1677,11 @@ static MachineClass *select_machine(QDict *qdict, Error **errp)
- }
- }
-
-+ if (machine_class) {
-+ machine_class->pve_version = g_strdup(pvever);
-+ qdict_del(qdict, "pvever");
-+ }
-+
- g_slist_free(machines);
- if (local_err) {
- error_append_hint(&local_err, "Use -machine help to list supported machines\n");
-@@ -3362,12 +3368,31 @@ void qemu_init(int argc, char **argv, char **envp)
- case QEMU_OPTION_machine:
- {
- bool help;
-+ size_t pvever_index, name_len;
-+ const gchar *name;
-+ gchar *name_clean, *pvever;
-
- keyval_parse_into(machine_opts_dict, optarg, "type", &help, &error_fatal);
- if (help) {
- machine_help_func(machine_opts_dict);
- exit(EXIT_SUCCESS);
- }
-+
-+ // PVE version is specified with '+' as seperator, e.g. pc-i440fx+pvever
-+ name = qdict_get_try_str(machine_opts_dict, "type");
-+ if (name != NULL) {
-+ name_len = strlen(name);
-+ pvever_index = strcspn(name, "+");
-+ if (pvever_index < name_len) {
-+ name_clean = g_strndup(name, pvever_index);
-+ pvever = g_strndup(name + pvever_index + 1, name_len - pvever_index - 1);
-+ qdict_put_str(machine_opts_dict, "pvever", pvever);
-+ qdict_put_str(machine_opts_dict, "type", name_clean);
-+ g_free(name_clean);
-+ g_free(pvever);
-+ }
-+ }
-+
- break;
- }
- case QEMU_OPTION_accel:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:54 +0200
+Subject: [PATCH] PVE: [Compat]: 4.0 used balloon qemu-4-0-config-size false
+ here
+
+The underlying issue why this change from upstream to us arised in
+the first place is that QEMU 4.0 was already released at the point we
+run into this migration issue, so we did the then obvious fallback to
+false for virtio-balloon-device qemu-4-0-config-size.
+
+QEMU made that switch back in 4.1, where it now uses a backward
+compatible mechanism to detect if the bigger CFG sizes should be
+used, i.e., checking the VIRTIO_BALLOON_F_PAGE_POISON or
+VIRTIO_BALLOON_F_FREE_PAGE_HINT balloon feature flags.
+As for them, upstream released version 4.0 had this to true they keep
+it to true in their compatibility record for the 4.0 machine, to
+allow live migrations from 4.0 to 4.1.
+As for us, downstream released version 4.0 (first public release of
+this QEMU) had this to false, we change it back to false again, for
+the same reason.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ hw/core/machine.c | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/hw/core/machine.c b/hw/core/machine.c
+index a673302cce..fa424440bd 100644
+--- a/hw/core/machine.c
++++ b/hw/core/machine.c
+@@ -127,7 +127,8 @@ GlobalProperty hw_compat_4_0[] = {
+ { "virtio-vga", "edid", "false" },
+ { "virtio-gpu-device", "edid", "false" },
+ { "virtio-device", "use-started", "false" },
+- { "virtio-balloon-device", "qemu-4-0-config-size", "true" },
++ // PVE differed from upstream for 4.0 balloon cfg size
++ { "virtio-balloon-device", "qemu-4-0-config-size", "false" },
+ { "pl031", "migrate-tick-offset", "false" },
+ };
+ const size_t hw_compat_4_0_len = G_N_ELEMENTS(hw_compat_4_0);
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:55 +0200
+Subject: [PATCH] PVE: Allow version code in machine type
+
+E.g. pc-i440fx-4.0+pve3 would print 'pve3' as version code while
+selecting pc-i440fx-4.0 as machine type.
+
+Version is made available as 'pve-version' in query-machines (same as,
+and only if 'is-current').
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ hw/core/machine-qmp-cmds.c | 6 ++++++
+ include/hw/boards.h | 2 ++
+ qapi/machine.json | 4 +++-
+ softmmu/vl.c | 25 +++++++++++++++++++++++++
+ 4 files changed, 36 insertions(+), 1 deletion(-)
+
+diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
+index 76fff60a6b..ec9201fb9a 100644
+--- a/hw/core/machine-qmp-cmds.c
++++ b/hw/core/machine-qmp-cmds.c
+@@ -103,6 +103,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
+ if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
+ info->has_is_current = true;
+ info->is_current = true;
++
++ // PVE version string only exists for current machine
++ if (mc->pve_version) {
++ info->has_pve_version = true;
++ info->pve_version = g_strdup(mc->pve_version);
++ }
+ }
+
+ if (mc->default_cpu_type) {
+diff --git a/include/hw/boards.h b/include/hw/boards.h
+index 7b416c9787..8ae15c51aa 100644
+--- a/include/hw/boards.h
++++ b/include/hw/boards.h
+@@ -230,6 +230,8 @@ struct MachineClass {
+ const char *desc;
+ const char *deprecation_reason;
+
++ const char *pve_version;
++
+ void (*init)(MachineState *state);
+ void (*reset)(MachineState *state);
+ void (*wakeup)(MachineState *state);
+diff --git a/qapi/machine.json b/qapi/machine.json
+index 555458f785..d868e4d31d 100644
+--- a/qapi/machine.json
++++ b/qapi/machine.json
+@@ -157,6 +157,8 @@
+ #
+ # @default-ram-id: the default ID of initial RAM memory backend (since 5.2)
+ #
++# @pve-version: custom PVE version suffix specified as 'machine+pveN'
++#
+ # Since: 1.2
+ ##
+ { 'struct': 'MachineInfo',
+@@ -164,7 +166,7 @@
+ '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
+ 'hotpluggable-cpus': 'bool', 'numa-mem-supported': 'bool',
+ 'deprecated': 'bool', '*default-cpu-type': 'str',
+- '*default-ram-id': 'str' } }
++ '*default-ram-id': 'str', '*pve-version': 'str' } }
+
+ ##
+ # @query-machines:
+diff --git a/softmmu/vl.c b/softmmu/vl.c
+index 39f149924e..0d233d55f3 100644
+--- a/softmmu/vl.c
++++ b/softmmu/vl.c
+@@ -1580,6 +1580,7 @@ static const QEMUOption *lookup_opt(int argc, char **argv,
+ static MachineClass *select_machine(QDict *qdict, Error **errp)
+ {
+ const char *optarg = qdict_get_try_str(qdict, "type");
++ const char *pvever = qdict_get_try_str(qdict, "pvever");
+ GSList *machines = object_class_get_list(TYPE_MACHINE, false);
+ MachineClass *machine_class;
+ Error *local_err = NULL;
+@@ -1597,6 +1598,11 @@ static MachineClass *select_machine(QDict *qdict, Error **errp)
+ }
+ }
+
++ if (machine_class) {
++ machine_class->pve_version = g_strdup(pvever);
++ qdict_del(qdict, "pvever");
++ }
++
+ g_slist_free(machines);
+ if (local_err) {
+ error_append_hint(&local_err, "Use -machine help to list supported machines\n");
+@@ -3187,12 +3193,31 @@ void qemu_init(int argc, char **argv, char **envp)
+ case QEMU_OPTION_machine:
+ {
+ bool help;
++ size_t pvever_index, name_len;
++ const gchar *name;
++ gchar *name_clean, *pvever;
+
+ keyval_parse_into(machine_opts_dict, optarg, "type", &help, &error_fatal);
+ if (help) {
+ machine_help_func(machine_opts_dict);
+ exit(EXIT_SUCCESS);
+ }
++
++ // PVE version is specified with '+' as seperator, e.g. pc-i440fx+pvever
++ name = qdict_get_try_str(machine_opts_dict, "type");
++ if (name != NULL) {
++ name_len = strlen(name);
++ pvever_index = strcspn(name, "+");
++ if (pvever_index < name_len) {
++ name_clean = g_strndup(name, pvever_index);
++ pvever = g_strndup(name + pvever_index + 1, name_len - pvever_index - 1);
++ qdict_put_str(machine_opts_dict, "pvever", pvever);
++ qdict_put_str(machine_opts_dict, "type", name_clean);
++ g_free(name_clean);
++ g_free(pvever);
++ }
++ }
++
+ break;
+ }
+ case QEMU_OPTION_accel:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 2 Mar 2022 08:35:05 +0100
-Subject: [PATCH] block/backup: move bcs bitmap initialization to job creation
-
-For backing up the state of multiple disks from the same time, a job
-for each disk has to be created. It's convenient if the jobs don't
-have to be started at the same time and if operation of the VM can be
-resumed after job creation. This would lead to a window between job
-creation and running the job, where writes can happen. But no writes
-should happen between setting up the copy-before-write filter and
-setting up the block copy state bitmap, because then new writes would
-just pass through.
-
-Commit 06e0a9c16405c0a4c1eca33cf286cc04c42066a2 moved initalization of
-the bitmap to setting up the copy-before-write filter when sync_mode
-is not MIRROR_SYNC_MODE_BITMAP. Ensure that the bitmap is initialized
-upon job creation for the remaining case too, by moving the
-backup_init_bcs_bitmap call to backup_job_create.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/backup.c | 8 ++++----
- 1 file changed, 4 insertions(+), 4 deletions(-)
-
-diff --git a/block/backup.c b/block/backup.c
-index 5cfd0b999c..07b899035c 100644
---- a/block/backup.c
-+++ b/block/backup.c
-@@ -239,8 +239,8 @@ static void backup_init_bcs_bitmap(BackupBlockJob *job)
- assert(ret);
- } else if (job->sync_mode == MIRROR_SYNC_MODE_TOP) {
- /*
-- * We can't hog the coroutine to initialize this thoroughly.
-- * Set a flag and resume work when we are able to yield safely.
-+ * Initialization is costly here. Simply set a flag and let the
-+ * backup_run coroutine resume work once it can yield safely.
- */
- block_copy_set_skip_unallocated(job->bcs, true);
- }
-@@ -254,8 +254,6 @@ static int coroutine_fn backup_run(Job *job, Error **errp)
- BackupBlockJob *s = container_of(job, BackupBlockJob, common.job);
- int ret;
-
-- backup_init_bcs_bitmap(s);
--
- if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
- int64_t offset = 0;
- int64_t count;
-@@ -494,6 +492,8 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- block_job_add_bdrv(&job->common, "target", target, 0, BLK_PERM_ALL,
- &error_abort);
-
-+ backup_init_bcs_bitmap(job);
-+
- return &job->common;
-
- error:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:57 +0200
-Subject: [PATCH] PVE-Backup: add vma backup format code
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[FE: create: register all streams before entering coroutines]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/meson.build | 2 +
- meson.build | 5 +
- vma-reader.c | 860 ++++++++++++++++++++++++++++++++++++++++++++++
- vma-writer.c | 791 ++++++++++++++++++++++++++++++++++++++++++
- vma.c | 850 +++++++++++++++++++++++++++++++++++++++++++++
- vma.h | 150 ++++++++
- 6 files changed, 2658 insertions(+)
- create mode 100644 vma-reader.c
- create mode 100644 vma-writer.c
- create mode 100644 vma.c
- create mode 100644 vma.h
-
-diff --git a/block/meson.build b/block/meson.build
-index 9451fc3ccd..819eb80951 100644
---- a/block/meson.build
-+++ b/block/meson.build
-@@ -46,6 +46,8 @@ block_ss.add(files(
- 'zeroinit.c',
- ), zstd, zlib, gnutls)
-
-+block_ss.add(files('../vma-writer.c'), libuuid)
-+
- softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
-
- if get_option('qcow1').allowed()
-diff --git a/meson.build b/meson.build
-index 861de93c4f..96a91b95e4 100644
---- a/meson.build
-+++ b/meson.build
-@@ -1267,6 +1267,8 @@ keyutils = dependency('libkeyutils', required: false,
-
- has_gettid = cc.has_function('gettid')
-
-+libuuid = cc.find_library('uuid', required: true)
-+
- # libselinux
- selinux = dependency('libselinux',
- required: get_option('selinux'),
-@@ -3359,6 +3361,9 @@ if have_tools
- dependencies: [blockdev, qemuutil, gnutls, selinux],
- install: true)
-
-+ vma = executable('vma', files('vma.c', 'vma-reader.c') + genh,
-+ dependencies: [authz, block, crypto, io, qom], install: true)
-+
- subdir('storage-daemon')
- subdir('contrib/rdmacm-mux')
- subdir('contrib/elf2dmp')
-diff --git a/vma-reader.c b/vma-reader.c
-new file mode 100644
-index 0000000000..4f4ee2b47b
---- /dev/null
-+++ b/vma-reader.c
-@@ -0,0 +1,860 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) 2012 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+#include <uuid/uuid.h>
-+
-+#include "qemu-common.h"
-+#include "qemu/timer.h"
-+#include "qemu/ratelimit.h"
-+#include "vma.h"
-+#include "block/block.h"
-+#include "sysemu/block-backend.h"
-+
-+static unsigned char zero_vma_block[VMA_BLOCK_SIZE];
-+
-+typedef struct VmaRestoreState {
-+ BlockBackend *target;
-+ bool write_zeroes;
-+ unsigned long *bitmap;
-+ int bitmap_size;
-+} VmaRestoreState;
-+
-+struct VmaReader {
-+ int fd;
-+ GChecksum *md5csum;
-+ GHashTable *blob_hash;
-+ unsigned char *head_data;
-+ VmaDeviceInfo devinfo[256];
-+ VmaRestoreState rstate[256];
-+ GList *cdata_list;
-+ guint8 vmstate_stream;
-+ uint32_t vmstate_clusters;
-+ /* to show restore percentage if run with -v */
-+ time_t start_time;
-+ int64_t cluster_count;
-+ int64_t clusters_read;
-+ int64_t zero_cluster_data;
-+ int64_t partial_zero_cluster_data;
-+ int clusters_read_per;
-+};
-+
-+static guint
-+g_int32_hash(gconstpointer v)
-+{
-+ return *(const uint32_t *)v;
-+}
-+
-+static gboolean
-+g_int32_equal(gconstpointer v1, gconstpointer v2)
-+{
-+ return *((const uint32_t *)v1) == *((const uint32_t *)v2);
-+}
-+
-+static int vma_reader_get_bitmap(VmaRestoreState *rstate, int64_t cluster_num)
-+{
-+ assert(rstate);
-+ assert(rstate->bitmap);
-+
-+ unsigned long val, idx, bit;
-+
-+ idx = cluster_num / BITS_PER_LONG;
-+
-+ assert(rstate->bitmap_size > idx);
-+
-+ bit = cluster_num % BITS_PER_LONG;
-+ val = rstate->bitmap[idx];
-+
-+ return !!(val & (1UL << bit));
-+}
-+
-+static void vma_reader_set_bitmap(VmaRestoreState *rstate, int64_t cluster_num,
-+ int dirty)
-+{
-+ assert(rstate);
-+ assert(rstate->bitmap);
-+
-+ unsigned long val, idx, bit;
-+
-+ idx = cluster_num / BITS_PER_LONG;
-+
-+ assert(rstate->bitmap_size > idx);
-+
-+ bit = cluster_num % BITS_PER_LONG;
-+ val = rstate->bitmap[idx];
-+ if (dirty) {
-+ if (!(val & (1UL << bit))) {
-+ val |= 1UL << bit;
-+ }
-+ } else {
-+ if (val & (1UL << bit)) {
-+ val &= ~(1UL << bit);
-+ }
-+ }
-+ rstate->bitmap[idx] = val;
-+}
-+
-+typedef struct VmaBlob {
-+ uint32_t start;
-+ uint32_t len;
-+ void *data;
-+} VmaBlob;
-+
-+static const VmaBlob *get_header_blob(VmaReader *vmar, uint32_t pos)
-+{
-+ assert(vmar);
-+ assert(vmar->blob_hash);
-+
-+ return g_hash_table_lookup(vmar->blob_hash, &pos);
-+}
-+
-+static const char *get_header_str(VmaReader *vmar, uint32_t pos)
-+{
-+ const VmaBlob *blob = get_header_blob(vmar, pos);
-+ if (!blob) {
-+ return NULL;
-+ }
-+ const char *res = (char *)blob->data;
-+ if (res[blob->len-1] != '\0') {
-+ return NULL;
-+ }
-+ return res;
-+}
-+
-+static ssize_t
-+safe_read(int fd, unsigned char *buf, size_t count)
-+{
-+ ssize_t n;
-+
-+ do {
-+ n = read(fd, buf, count);
-+ } while (n < 0 && errno == EINTR);
-+
-+ return n;
-+}
-+
-+static ssize_t
-+full_read(int fd, unsigned char *buf, size_t len)
-+{
-+ ssize_t n;
-+ size_t total;
-+
-+ total = 0;
-+
-+ while (len > 0) {
-+ n = safe_read(fd, buf, len);
-+
-+ if (n == 0) {
-+ return total;
-+ }
-+
-+ if (n <= 0) {
-+ break;
-+ }
-+
-+ buf += n;
-+ total += n;
-+ len -= n;
-+ }
-+
-+ if (len) {
-+ return -1;
-+ }
-+
-+ return total;
-+}
-+
-+void vma_reader_destroy(VmaReader *vmar)
-+{
-+ assert(vmar);
-+
-+ if (vmar->fd >= 0) {
-+ close(vmar->fd);
-+ }
-+
-+ if (vmar->cdata_list) {
-+ g_list_free(vmar->cdata_list);
-+ }
-+
-+ int i;
-+ for (i = 1; i < 256; i++) {
-+ if (vmar->rstate[i].bitmap) {
-+ g_free(vmar->rstate[i].bitmap);
-+ }
-+ if (vmar->rstate[i].target) {
-+ blk_unref(vmar->rstate[i].target);
-+ }
-+ }
-+
-+ if (vmar->md5csum) {
-+ g_checksum_free(vmar->md5csum);
-+ }
-+
-+ if (vmar->blob_hash) {
-+ g_hash_table_destroy(vmar->blob_hash);
-+ }
-+
-+ if (vmar->head_data) {
-+ g_free(vmar->head_data);
-+ }
-+
-+ g_free(vmar);
-+
-+};
-+
-+static int vma_reader_read_head(VmaReader *vmar, Error **errp)
-+{
-+ assert(vmar);
-+ assert(errp);
-+ assert(*errp == NULL);
-+
-+ unsigned char md5sum[16];
-+ int i;
-+ int ret = 0;
-+
-+ vmar->head_data = g_malloc(sizeof(VmaHeader));
-+
-+ if (full_read(vmar->fd, vmar->head_data, sizeof(VmaHeader)) !=
-+ sizeof(VmaHeader)) {
-+ error_setg(errp, "can't read vma header - %s",
-+ errno ? g_strerror(errno) : "got EOF");
-+ return -1;
-+ }
-+
-+ VmaHeader *h = (VmaHeader *)vmar->head_data;
-+
-+ if (h->magic != VMA_MAGIC) {
-+ error_setg(errp, "not a vma file - wrong magic number");
-+ return -1;
-+ }
-+
-+ uint32_t header_size = GUINT32_FROM_BE(h->header_size);
-+ int need = header_size - sizeof(VmaHeader);
-+ if (need <= 0) {
-+ error_setg(errp, "wrong vma header size %d", header_size);
-+ return -1;
-+ }
-+
-+ vmar->head_data = g_realloc(vmar->head_data, header_size);
-+ h = (VmaHeader *)vmar->head_data;
-+
-+ if (full_read(vmar->fd, vmar->head_data + sizeof(VmaHeader), need) !=
-+ need) {
-+ error_setg(errp, "can't read vma header data - %s",
-+ errno ? g_strerror(errno) : "got EOF");
-+ return -1;
-+ }
-+
-+ memcpy(md5sum, h->md5sum, 16);
-+ memset(h->md5sum, 0, 16);
-+
-+ g_checksum_reset(vmar->md5csum);
-+ g_checksum_update(vmar->md5csum, vmar->head_data, header_size);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmar->md5csum, (guint8 *)(h->md5sum), &csize);
-+
-+ if (memcmp(md5sum, h->md5sum, 16) != 0) {
-+ error_setg(errp, "wrong vma header chechsum");
-+ return -1;
-+ }
-+
-+ /* we can modify header data after checksum verify */
-+ h->header_size = header_size;
-+
-+ h->version = GUINT32_FROM_BE(h->version);
-+ if (h->version != 1) {
-+ error_setg(errp, "wrong vma version %d", h->version);
-+ return -1;
-+ }
-+
-+ h->ctime = GUINT64_FROM_BE(h->ctime);
-+ h->blob_buffer_offset = GUINT32_FROM_BE(h->blob_buffer_offset);
-+ h->blob_buffer_size = GUINT32_FROM_BE(h->blob_buffer_size);
-+
-+ uint32_t bstart = h->blob_buffer_offset + 1;
-+ uint32_t bend = h->blob_buffer_offset + h->blob_buffer_size;
-+
-+ if (bstart <= sizeof(VmaHeader)) {
-+ error_setg(errp, "wrong vma blob buffer offset %d",
-+ h->blob_buffer_offset);
-+ return -1;
-+ }
-+
-+ if (bend > header_size) {
-+ error_setg(errp, "wrong vma blob buffer size %d/%d",
-+ h->blob_buffer_offset, h->blob_buffer_size);
-+ return -1;
-+ }
-+
-+ while ((bstart + 2) <= bend) {
-+ uint32_t size = vmar->head_data[bstart] +
-+ (vmar->head_data[bstart+1] << 8);
-+ if ((bstart + size + 2) <= bend) {
-+ VmaBlob *blob = g_new0(VmaBlob, 1);
-+ blob->start = bstart - h->blob_buffer_offset;
-+ blob->len = size;
-+ blob->data = vmar->head_data + bstart + 2;
-+ g_hash_table_insert(vmar->blob_hash, &blob->start, blob);
-+ }
-+ bstart += size + 2;
-+ }
-+
-+
-+ int count = 0;
-+ for (i = 1; i < 256; i++) {
-+ VmaDeviceInfoHeader *dih = &h->dev_info[i];
-+ uint32_t devname_ptr = GUINT32_FROM_BE(dih->devname_ptr);
-+ uint64_t size = GUINT64_FROM_BE(dih->size);
-+ const char *devname = get_header_str(vmar, devname_ptr);
-+
-+ if (size && devname) {
-+ count++;
-+ vmar->devinfo[i].size = size;
-+ vmar->devinfo[i].devname = devname;
-+
-+ if (strcmp(devname, "vmstate") == 0) {
-+ vmar->vmstate_stream = i;
-+ }
-+ }
-+ }
-+
-+ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
-+ uint32_t name_ptr = GUINT32_FROM_BE(h->config_names[i]);
-+ uint32_t data_ptr = GUINT32_FROM_BE(h->config_data[i]);
-+
-+ if (!(name_ptr && data_ptr)) {
-+ continue;
-+ }
-+ const char *name = get_header_str(vmar, name_ptr);
-+ const VmaBlob *blob = get_header_blob(vmar, data_ptr);
-+
-+ if (!(name && blob)) {
-+ error_setg(errp, "vma contains invalid data pointers");
-+ return -1;
-+ }
-+
-+ VmaConfigData *cdata = g_new0(VmaConfigData, 1);
-+ cdata->name = name;
-+ cdata->data = blob->data;
-+ cdata->len = blob->len;
-+
-+ vmar->cdata_list = g_list_append(vmar->cdata_list, cdata);
-+ }
-+
-+ return ret;
-+};
-+
-+VmaReader *vma_reader_create(const char *filename, Error **errp)
-+{
-+ assert(filename);
-+ assert(errp);
-+
-+ VmaReader *vmar = g_new0(VmaReader, 1);
-+
-+ if (strcmp(filename, "-") == 0) {
-+ vmar->fd = dup(0);
-+ } else {
-+ vmar->fd = open(filename, O_RDONLY);
-+ }
-+
-+ if (vmar->fd < 0) {
-+ error_setg(errp, "can't open file %s - %s\n", filename,
-+ g_strerror(errno));
-+ goto err;
-+ }
-+
-+ vmar->md5csum = g_checksum_new(G_CHECKSUM_MD5);
-+ if (!vmar->md5csum) {
-+ error_setg(errp, "can't allocate cmsum\n");
-+ goto err;
-+ }
-+
-+ vmar->blob_hash = g_hash_table_new_full(g_int32_hash, g_int32_equal,
-+ NULL, g_free);
-+
-+ if (vma_reader_read_head(vmar, errp) < 0) {
-+ goto err;
-+ }
-+
-+ return vmar;
-+
-+err:
-+ if (vmar) {
-+ vma_reader_destroy(vmar);
-+ }
-+
-+ return NULL;
-+}
-+
-+VmaHeader *vma_reader_get_header(VmaReader *vmar)
-+{
-+ assert(vmar);
-+ assert(vmar->head_data);
-+
-+ return (VmaHeader *)(vmar->head_data);
-+}
-+
-+GList *vma_reader_get_config_data(VmaReader *vmar)
-+{
-+ assert(vmar);
-+ assert(vmar->head_data);
-+
-+ return vmar->cdata_list;
-+}
-+
-+VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id)
-+{
-+ assert(vmar);
-+ assert(dev_id);
-+
-+ if (vmar->devinfo[dev_id].size && vmar->devinfo[dev_id].devname) {
-+ return &vmar->devinfo[dev_id];
-+ }
-+
-+ return NULL;
-+}
-+
-+static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
-+ BlockBackend *target, bool write_zeroes)
-+{
-+ assert(vmar);
-+ assert(dev_id);
-+
-+ vmar->rstate[dev_id].target = target;
-+ vmar->rstate[dev_id].write_zeroes = write_zeroes;
-+
-+ int64_t size = vmar->devinfo[dev_id].size;
-+
-+ int64_t bitmap_size = (size/BDRV_SECTOR_SIZE) +
-+ (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG - 1;
-+ bitmap_size /= (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG;
-+
-+ vmar->rstate[dev_id].bitmap_size = bitmap_size;
-+ vmar->rstate[dev_id].bitmap = g_new0(unsigned long, bitmap_size);
-+
-+ vmar->cluster_count += size/VMA_CLUSTER_SIZE;
-+}
-+
-+int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id, BlockBackend *target,
-+ bool write_zeroes, Error **errp)
-+{
-+ assert(vmar);
-+ assert(target != NULL);
-+ assert(dev_id);
-+ assert(vmar->rstate[dev_id].target == NULL);
-+
-+ int64_t size = blk_getlength(target);
-+ int64_t size_diff = size - vmar->devinfo[dev_id].size;
-+
-+ /* storage types can have different size restrictions, so it
-+ * is not always possible to create an image with exact size.
-+ * So we tolerate a size difference up to 4MB.
-+ */
-+ if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
-+ error_setg(errp, "vma_reader_register_bs for stream %s failed - "
-+ "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
-+ size, vmar->devinfo[dev_id].size);
-+ return -1;
-+ }
-+
-+ allocate_rstate(vmar, dev_id, target, write_zeroes);
-+
-+ return 0;
-+}
-+
-+static ssize_t safe_write(int fd, void *buf, size_t count)
-+{
-+ ssize_t n;
-+
-+ do {
-+ n = write(fd, buf, count);
-+ } while (n < 0 && errno == EINTR);
-+
-+ return n;
-+}
-+
-+static size_t full_write(int fd, void *buf, size_t len)
-+{
-+ ssize_t n;
-+ size_t total;
-+
-+ total = 0;
-+
-+ while (len > 0) {
-+ n = safe_write(fd, buf, len);
-+ if (n < 0) {
-+ return n;
-+ }
-+ buf += n;
-+ total += n;
-+ len -= n;
-+ }
-+
-+ if (len) {
-+ /* incomplete write ? */
-+ return -1;
-+ }
-+
-+ return total;
-+}
-+
-+static int restore_write_data(VmaReader *vmar, guint8 dev_id,
-+ BlockBackend *target, int vmstate_fd,
-+ unsigned char *buf, int64_t sector_num,
-+ int nb_sectors, Error **errp)
-+{
-+ assert(vmar);
-+
-+ if (dev_id == vmar->vmstate_stream) {
-+ if (vmstate_fd >= 0) {
-+ int len = nb_sectors * BDRV_SECTOR_SIZE;
-+ int res = full_write(vmstate_fd, buf, len);
-+ if (res < 0) {
-+ error_setg(errp, "write vmstate failed %d", res);
-+ return -1;
-+ }
-+ }
-+ } else {
-+ int res = blk_pwrite(target, sector_num * BDRV_SECTOR_SIZE, buf, nb_sectors * BDRV_SECTOR_SIZE, 0);
-+ if (res < 0) {
-+ error_setg(errp, "blk_pwrite to %s failed (%d)",
-+ bdrv_get_device_name(blk_bs(target)), res);
-+ return -1;
-+ }
-+ }
-+ return 0;
-+}
-+
-+static int restore_extent(VmaReader *vmar, unsigned char *buf,
-+ int extent_size, int vmstate_fd,
-+ bool verbose, bool verify, Error **errp)
-+{
-+ assert(vmar);
-+ assert(buf);
-+
-+ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
-+ int start = VMA_EXTENT_HEADER_SIZE;
-+ int i;
-+
-+ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
-+ uint64_t block_info = GUINT64_FROM_BE(ehead->blockinfo[i]);
-+ uint64_t cluster_num = block_info & 0xffffffff;
-+ uint8_t dev_id = (block_info >> 32) & 0xff;
-+ uint16_t mask = block_info >> (32+16);
-+ int64_t max_sector;
-+
-+ if (!dev_id) {
-+ continue;
-+ }
-+
-+ VmaRestoreState *rstate = &vmar->rstate[dev_id];
-+ BlockBackend *target = NULL;
-+
-+ if (dev_id != vmar->vmstate_stream) {
-+ target = rstate->target;
-+ if (!verify && !target) {
-+ error_setg(errp, "got wrong dev id %d", dev_id);
-+ return -1;
-+ }
-+
-+ if (vma_reader_get_bitmap(rstate, cluster_num)) {
-+ error_setg(errp, "found duplicated cluster %zd for stream %s",
-+ cluster_num, vmar->devinfo[dev_id].devname);
-+ return -1;
-+ }
-+ vma_reader_set_bitmap(rstate, cluster_num, 1);
-+
-+ max_sector = vmar->devinfo[dev_id].size/BDRV_SECTOR_SIZE;
-+ } else {
-+ max_sector = G_MAXINT64;
-+ if (cluster_num != vmar->vmstate_clusters) {
-+ error_setg(errp, "found out of order vmstate data");
-+ return -1;
-+ }
-+ vmar->vmstate_clusters++;
-+ }
-+
-+ vmar->clusters_read++;
-+
-+ if (verbose) {
-+ time_t duration = time(NULL) - vmar->start_time;
-+ int percent = (vmar->clusters_read*100)/vmar->cluster_count;
-+ if (percent != vmar->clusters_read_per) {
-+ printf("progress %d%% (read %zd bytes, duration %zd sec)\n",
-+ percent, vmar->clusters_read*VMA_CLUSTER_SIZE,
-+ duration);
-+ fflush(stdout);
-+ vmar->clusters_read_per = percent;
-+ }
-+ }
-+
-+ /* try to write whole clusters to speedup restore */
-+ if (mask == 0xffff) {
-+ if ((start + VMA_CLUSTER_SIZE) > extent_size) {
-+ error_setg(errp, "short vma extent - too many blocks");
-+ return -1;
-+ }
-+ int64_t sector_num = (cluster_num * VMA_CLUSTER_SIZE) /
-+ BDRV_SECTOR_SIZE;
-+ int64_t end_sector = sector_num +
-+ VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE;
-+
-+ if (end_sector > max_sector) {
-+ end_sector = max_sector;
-+ }
-+
-+ if (end_sector <= sector_num) {
-+ error_setg(errp, "got wrong block address - write beyond end");
-+ return -1;
-+ }
-+
-+ if (!verify) {
-+ int nb_sectors = end_sector - sector_num;
-+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
-+ buf + start, sector_num, nb_sectors,
-+ errp) < 0) {
-+ return -1;
-+ }
-+ }
-+
-+ start += VMA_CLUSTER_SIZE;
-+ } else {
-+ int j;
-+ int bit = 1;
-+
-+ for (j = 0; j < 16; j++) {
-+ int64_t sector_num = (cluster_num*VMA_CLUSTER_SIZE +
-+ j*VMA_BLOCK_SIZE)/BDRV_SECTOR_SIZE;
-+
-+ int64_t end_sector = sector_num +
-+ VMA_BLOCK_SIZE/BDRV_SECTOR_SIZE;
-+ if (end_sector > max_sector) {
-+ end_sector = max_sector;
-+ }
-+
-+ if (mask & bit) {
-+ if ((start + VMA_BLOCK_SIZE) > extent_size) {
-+ error_setg(errp, "short vma extent - too many blocks");
-+ return -1;
-+ }
-+
-+ if (end_sector <= sector_num) {
-+ error_setg(errp, "got wrong block address - "
-+ "write beyond end");
-+ return -1;
-+ }
-+
-+ if (!verify) {
-+ int nb_sectors = end_sector - sector_num;
-+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
-+ buf + start, sector_num,
-+ nb_sectors, errp) < 0) {
-+ return -1;
-+ }
-+ }
-+
-+ start += VMA_BLOCK_SIZE;
-+
-+ } else {
-+
-+
-+ if (end_sector > sector_num) {
-+ /* Todo: use bdrv_co_write_zeroes (but that need to
-+ * be run inside coroutine?)
-+ */
-+ int nb_sectors = end_sector - sector_num;
-+ int zero_size = BDRV_SECTOR_SIZE*nb_sectors;
-+ vmar->zero_cluster_data += zero_size;
-+ if (mask != 0) {
-+ vmar->partial_zero_cluster_data += zero_size;
-+ }
-+
-+ if (rstate->write_zeroes && !verify) {
-+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
-+ zero_vma_block, sector_num,
-+ nb_sectors, errp) < 0) {
-+ return -1;
-+ }
-+ }
-+ }
-+ }
-+
-+ bit = bit << 1;
-+ }
-+ }
-+ }
-+
-+ if (start != extent_size) {
-+ error_setg(errp, "vma extent error - missing blocks");
-+ return -1;
-+ }
-+
-+ return 0;
-+}
-+
-+static int vma_reader_restore_full(VmaReader *vmar, int vmstate_fd,
-+ bool verbose, bool verify,
-+ Error **errp)
-+{
-+ assert(vmar);
-+ assert(vmar->head_data);
-+
-+ int ret = 0;
-+ unsigned char buf[VMA_MAX_EXTENT_SIZE];
-+ int buf_pos = 0;
-+ unsigned char md5sum[16];
-+ VmaHeader *h = (VmaHeader *)vmar->head_data;
-+
-+ vmar->start_time = time(NULL);
-+
-+ while (1) {
-+ int bytes = full_read(vmar->fd, buf + buf_pos, sizeof(buf) - buf_pos);
-+ if (bytes < 0) {
-+ error_setg(errp, "read failed - %s", g_strerror(errno));
-+ return -1;
-+ }
-+
-+ buf_pos += bytes;
-+
-+ if (!buf_pos) {
-+ break; /* EOF */
-+ }
-+
-+ if (buf_pos < VMA_EXTENT_HEADER_SIZE) {
-+ error_setg(errp, "read short extent (%d bytes)", buf_pos);
-+ return -1;
-+ }
-+
-+ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
-+
-+ /* extract md5sum */
-+ memcpy(md5sum, ehead->md5sum, sizeof(ehead->md5sum));
-+ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
-+
-+ g_checksum_reset(vmar->md5csum);
-+ g_checksum_update(vmar->md5csum, buf, VMA_EXTENT_HEADER_SIZE);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmar->md5csum, ehead->md5sum, &csize);
-+
-+ if (memcmp(md5sum, ehead->md5sum, 16) != 0) {
-+ error_setg(errp, "wrong vma extent header chechsum");
-+ return -1;
-+ }
-+
-+ if (memcmp(h->uuid, ehead->uuid, sizeof(ehead->uuid)) != 0) {
-+ error_setg(errp, "wrong vma extent uuid");
-+ return -1;
-+ }
-+
-+ if (ehead->magic != VMA_EXTENT_MAGIC || ehead->reserved1 != 0) {
-+ error_setg(errp, "wrong vma extent header magic");
-+ return -1;
-+ }
-+
-+ int block_count = GUINT16_FROM_BE(ehead->block_count);
-+ int extent_size = VMA_EXTENT_HEADER_SIZE + block_count*VMA_BLOCK_SIZE;
-+
-+ if (buf_pos < extent_size) {
-+ error_setg(errp, "short vma extent (%d < %d)", buf_pos,
-+ extent_size);
-+ return -1;
-+ }
-+
-+ if (restore_extent(vmar, buf, extent_size, vmstate_fd, verbose,
-+ verify, errp) < 0) {
-+ return -1;
-+ }
-+
-+ if (buf_pos > extent_size) {
-+ memmove(buf, buf + extent_size, buf_pos - extent_size);
-+ buf_pos = buf_pos - extent_size;
-+ } else {
-+ buf_pos = 0;
-+ }
-+ }
-+
-+ bdrv_drain_all();
-+
-+ int i;
-+ for (i = 1; i < 256; i++) {
-+ VmaRestoreState *rstate = &vmar->rstate[i];
-+ if (!rstate->target) {
-+ continue;
-+ }
-+
-+ if (blk_flush(rstate->target) < 0) {
-+ error_setg(errp, "vma blk_flush %s failed",
-+ vmar->devinfo[i].devname);
-+ return -1;
-+ }
-+
-+ if (vmar->devinfo[i].size &&
-+ (strcmp(vmar->devinfo[i].devname, "vmstate") != 0)) {
-+ assert(rstate->bitmap);
-+
-+ int64_t cluster_num, end;
-+
-+ end = (vmar->devinfo[i].size + VMA_CLUSTER_SIZE - 1) /
-+ VMA_CLUSTER_SIZE;
-+
-+ for (cluster_num = 0; cluster_num < end; cluster_num++) {
-+ if (!vma_reader_get_bitmap(rstate, cluster_num)) {
-+ error_setg(errp, "detected missing cluster %zd "
-+ "for stream %s", cluster_num,
-+ vmar->devinfo[i].devname);
-+ return -1;
-+ }
-+ }
-+ }
-+ }
-+
-+ if (verbose) {
-+ if (vmar->clusters_read) {
-+ printf("total bytes read %zd, sparse bytes %zd (%.3g%%)\n",
-+ vmar->clusters_read*VMA_CLUSTER_SIZE,
-+ vmar->zero_cluster_data,
-+ (double)(100.0*vmar->zero_cluster_data)/
-+ (vmar->clusters_read*VMA_CLUSTER_SIZE));
-+
-+ int64_t datasize = vmar->clusters_read*VMA_CLUSTER_SIZE-vmar->zero_cluster_data;
-+ if (datasize) { // this does not make sense for empty files
-+ printf("space reduction due to 4K zero blocks %.3g%%\n",
-+ (double)(100.0*vmar->partial_zero_cluster_data) / datasize);
-+ }
-+ } else {
-+ printf("vma archive contains no image data\n");
-+ }
-+ }
-+ return ret;
-+}
-+
-+int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
-+ Error **errp)
-+{
-+ return vma_reader_restore_full(vmar, vmstate_fd, verbose, false, errp);
-+}
-+
-+int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp)
-+{
-+ guint8 dev_id;
-+
-+ for (dev_id = 1; dev_id < 255; dev_id++) {
-+ if (vma_reader_get_device_info(vmar, dev_id)) {
-+ allocate_rstate(vmar, dev_id, NULL, false);
-+ }
-+ }
-+
-+ return vma_reader_restore_full(vmar, -1, verbose, true, errp);
-+}
-+
-diff --git a/vma-writer.c b/vma-writer.c
-new file mode 100644
-index 0000000000..df4b20793d
---- /dev/null
-+++ b/vma-writer.c
-@@ -0,0 +1,791 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) 2012 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+#include <uuid/uuid.h>
-+
-+#include "vma.h"
-+#include "block/block.h"
-+#include "monitor/monitor.h"
-+#include "qemu/main-loop.h"
-+#include "qemu/coroutine.h"
-+#include "qemu/cutils.h"
-+#include "qemu/memalign.h"
-+
-+#define DEBUG_VMA 0
-+
-+#define DPRINTF(fmt, ...)\
-+ do { if (DEBUG_VMA) { printf("vma: " fmt, ## __VA_ARGS__); } } while (0)
-+
-+#define WRITE_BUFFERS 5
-+#define HEADER_CLUSTERS 8
-+#define HEADERBUF_SIZE (VMA_CLUSTER_SIZE*HEADER_CLUSTERS)
-+
-+struct VmaWriter {
-+ int fd;
-+ FILE *cmd;
-+ int status;
-+ char errmsg[8192];
-+ uuid_t uuid;
-+ bool header_written;
-+ bool closed;
-+
-+ /* we always write extents */
-+ unsigned char *outbuf;
-+ int outbuf_pos; /* in bytes */
-+ int outbuf_count; /* in VMA_BLOCKS */
-+ uint64_t outbuf_block_info[VMA_BLOCKS_PER_EXTENT];
-+
-+ unsigned char *headerbuf;
-+
-+ GChecksum *md5csum;
-+ CoMutex flush_lock;
-+ Coroutine *co_writer;
-+
-+ /* drive informations */
-+ VmaStreamInfo stream_info[256];
-+ guint stream_count;
-+
-+ guint8 vmstate_stream;
-+ uint32_t vmstate_clusters;
-+
-+ /* header blob table */
-+ char *header_blob_table;
-+ uint32_t header_blob_table_size;
-+ uint32_t header_blob_table_pos;
-+
-+ /* store for config blobs */
-+ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+ uint32_t config_count;
-+};
-+
-+void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...)
-+{
-+ va_list ap;
-+
-+ if (vmaw->status < 0) {
-+ return;
-+ }
-+
-+ vmaw->status = -1;
-+
-+ va_start(ap, fmt);
-+ g_vsnprintf(vmaw->errmsg, sizeof(vmaw->errmsg), fmt, ap);
-+ va_end(ap);
-+
-+ DPRINTF("vma_writer_set_error: %s\n", vmaw->errmsg);
-+}
-+
-+static uint32_t allocate_header_blob(VmaWriter *vmaw, const char *data,
-+ size_t len)
-+{
-+ if (len > 65535) {
-+ return 0;
-+ }
-+
-+ if (!vmaw->header_blob_table ||
-+ (vmaw->header_blob_table_size <
-+ (vmaw->header_blob_table_pos + len + 2))) {
-+ int newsize = vmaw->header_blob_table_size + ((len + 2 + 511)/512)*512;
-+
-+ vmaw->header_blob_table = g_realloc(vmaw->header_blob_table, newsize);
-+ memset(vmaw->header_blob_table + vmaw->header_blob_table_size,
-+ 0, newsize - vmaw->header_blob_table_size);
-+ vmaw->header_blob_table_size = newsize;
-+ }
-+
-+ uint32_t cpos = vmaw->header_blob_table_pos;
-+ vmaw->header_blob_table[cpos] = len & 255;
-+ vmaw->header_blob_table[cpos+1] = (len >> 8) & 255;
-+ memcpy(vmaw->header_blob_table + cpos + 2, data, len);
-+ vmaw->header_blob_table_pos += len + 2;
-+ return cpos;
-+}
-+
-+static uint32_t allocate_header_string(VmaWriter *vmaw, const char *str)
-+{
-+ assert(vmaw);
-+
-+ size_t len = strlen(str) + 1;
-+
-+ return allocate_header_blob(vmaw, str, len);
-+}
-+
-+int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
-+ gsize len)
-+{
-+ assert(vmaw);
-+ assert(!vmaw->header_written);
-+ assert(vmaw->config_count < VMA_MAX_CONFIGS);
-+ assert(name);
-+ assert(data);
-+
-+ gchar *basename = g_path_get_basename(name);
-+ uint32_t name_ptr = allocate_header_string(vmaw, basename);
-+ g_free(basename);
-+
-+ if (!name_ptr) {
-+ return -1;
-+ }
-+
-+ uint32_t data_ptr = allocate_header_blob(vmaw, data, len);
-+ if (!data_ptr) {
-+ return -1;
-+ }
-+
-+ vmaw->config_names[vmaw->config_count] = name_ptr;
-+ vmaw->config_data[vmaw->config_count] = data_ptr;
-+
-+ vmaw->config_count++;
-+
-+ return 0;
-+}
-+
-+int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
-+ size_t size)
-+{
-+ assert(vmaw);
-+ assert(devname);
-+ assert(!vmaw->status);
-+
-+ if (vmaw->header_written) {
-+ vma_writer_set_error(vmaw, "vma_writer_register_stream: header "
-+ "already written");
-+ return -1;
-+ }
-+
-+ guint n = vmaw->stream_count + 1;
-+
-+ /* we can have dev_ids form 1 to 255 (0 reserved)
-+ * 255(-1) reseverd for safety
-+ */
-+ if (n > 254) {
-+ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
-+ "too many drives");
-+ return -1;
-+ }
-+
-+ if (size <= 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
-+ "got strange size %zd", size);
-+ return -1;
-+ }
-+
-+ DPRINTF("vma_writer_register_stream %s %zu %d\n", devname, size, n);
-+
-+ vmaw->stream_info[n].devname = g_strdup(devname);
-+ vmaw->stream_info[n].size = size;
-+
-+ vmaw->stream_info[n].cluster_count = (size + VMA_CLUSTER_SIZE - 1) /
-+ VMA_CLUSTER_SIZE;
-+
-+ vmaw->stream_count = n;
-+
-+ if (strcmp(devname, "vmstate") == 0) {
-+ vmaw->vmstate_stream = n;
-+ }
-+
-+ return n;
-+}
-+
-+static void coroutine_fn yield_until_fd_writable(int fd)
-+{
-+ assert(qemu_in_coroutine());
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ aio_set_fd_handler(ctx, fd, false, NULL, (IOHandler *)qemu_coroutine_enter,
-+ NULL, NULL, qemu_coroutine_self());
-+ qemu_coroutine_yield();
-+ aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL, NULL);
-+}
-+
-+static ssize_t coroutine_fn
-+vma_queue_write(VmaWriter *vmaw, const void *buf, size_t bytes)
-+{
-+ DPRINTF("vma_queue_write enter %zd\n", bytes);
-+
-+ assert(vmaw);
-+ assert(buf);
-+ assert(bytes <= VMA_MAX_EXTENT_SIZE);
-+
-+ size_t done = 0;
-+ ssize_t ret;
-+
-+ assert(vmaw->co_writer == NULL);
-+
-+ vmaw->co_writer = qemu_coroutine_self();
-+
-+ while (done < bytes) {
-+ if (vmaw->status < 0) {
-+ DPRINTF("vma_queue_write detected canceled backup\n");
-+ done = -1;
-+ break;
-+ }
-+ yield_until_fd_writable(vmaw->fd);
-+ ret = write(vmaw->fd, buf + done, bytes - done);
-+ if (ret > 0) {
-+ done += ret;
-+ DPRINTF("vma_queue_write written %zd %zd\n", done, ret);
-+ } else if (ret < 0) {
-+ if (errno == EAGAIN || errno == EWOULDBLOCK) {
-+ /* try again */
-+ } else {
-+ vma_writer_set_error(vmaw, "vma_queue_write: write error - %s",
-+ g_strerror(errno));
-+ done = -1; /* always return failure for partial writes */
-+ break;
-+ }
-+ } else if (ret == 0) {
-+ /* should not happen - simply try again */
-+ }
-+ }
-+
-+ vmaw->co_writer = NULL;
-+
-+ return (done == bytes) ? bytes : -1;
-+}
-+
-+VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
-+{
-+ const char *p;
-+
-+ assert(sizeof(VmaHeader) == (4096 + 8192));
-+ assert(G_STRUCT_OFFSET(VmaHeader, config_names) == 2044);
-+ assert(G_STRUCT_OFFSET(VmaHeader, config_data) == 3068);
-+ assert(G_STRUCT_OFFSET(VmaHeader, dev_info) == 4096);
-+ assert(sizeof(VmaExtentHeader) == 512);
-+
-+ VmaWriter *vmaw = g_new0(VmaWriter, 1);
-+ vmaw->fd = -1;
-+
-+ vmaw->md5csum = g_checksum_new(G_CHECKSUM_MD5);
-+ if (!vmaw->md5csum) {
-+ error_setg(errp, "can't allocate cmsum\n");
-+ goto err;
-+ }
-+
-+ if (strstart(filename, "exec:", &p)) {
-+ vmaw->cmd = popen(p, "w");
-+ if (vmaw->cmd == NULL) {
-+ error_setg(errp, "can't popen command '%s' - %s\n", p,
-+ g_strerror(errno));
-+ goto err;
-+ }
-+ vmaw->fd = fileno(vmaw->cmd);
-+
-+ /* try to use O_NONBLOCK */
-+ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
-+
-+ } else {
-+ struct stat st;
-+ int oflags;
-+ const char *tmp_id_str;
-+
-+ if ((stat(filename, &st) == 0) && S_ISFIFO(st.st_mode)) {
-+ oflags = O_NONBLOCK|O_WRONLY;
-+ vmaw->fd = qemu_open(filename, oflags, errp);
-+ } else if (strstart(filename, "/dev/fdset/", &tmp_id_str)) {
-+ oflags = O_NONBLOCK|O_WRONLY;
-+ vmaw->fd = qemu_open(filename, oflags, errp);
-+ } else if (strstart(filename, "/dev/fdname/", &tmp_id_str)) {
-+ vmaw->fd = monitor_get_fd(monitor_cur(), tmp_id_str, errp);
-+ if (vmaw->fd < 0) {
-+ goto err;
-+ }
-+ /* try to use O_NONBLOCK */
-+ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
-+ } else {
-+ oflags = O_NONBLOCK|O_DIRECT|O_WRONLY|O_EXCL;
-+ vmaw->fd = qemu_create(filename, oflags, 0644, errp);
-+ }
-+
-+ if (vmaw->fd < 0) {
-+ error_setg(errp, "can't open file %s - %s\n", filename,
-+ g_strerror(errno));
-+ goto err;
-+ }
-+ }
-+
-+ /* we use O_DIRECT, so we need to align IO buffers */
-+
-+ vmaw->outbuf = qemu_memalign(512, VMA_MAX_EXTENT_SIZE);
-+ vmaw->headerbuf = qemu_memalign(512, HEADERBUF_SIZE);
-+
-+ vmaw->outbuf_count = 0;
-+ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
-+
-+ vmaw->header_blob_table_pos = 1; /* start at pos 1 */
-+
-+ qemu_co_mutex_init(&vmaw->flush_lock);
-+
-+ uuid_copy(vmaw->uuid, uuid);
-+
-+ return vmaw;
-+
-+err:
-+ if (vmaw) {
-+ if (vmaw->cmd) {
-+ pclose(vmaw->cmd);
-+ } else if (vmaw->fd >= 0) {
-+ close(vmaw->fd);
-+ }
-+
-+ if (vmaw->md5csum) {
-+ g_checksum_free(vmaw->md5csum);
-+ }
-+
-+ g_free(vmaw);
-+ }
-+
-+ return NULL;
-+}
-+
-+static int coroutine_fn vma_write_header(VmaWriter *vmaw)
-+{
-+ assert(vmaw);
-+ unsigned char *buf = vmaw->headerbuf;
-+ VmaHeader *head = (VmaHeader *)buf;
-+
-+ int i;
-+
-+ DPRINTF("VMA WRITE HEADER\n");
-+
-+ if (vmaw->status < 0) {
-+ return vmaw->status;
-+ }
-+
-+ memset(buf, 0, HEADERBUF_SIZE);
-+
-+ head->magic = VMA_MAGIC;
-+ head->version = GUINT32_TO_BE(1); /* v1 */
-+ memcpy(head->uuid, vmaw->uuid, 16);
-+
-+ time_t ctime = time(NULL);
-+ head->ctime = GUINT64_TO_BE(ctime);
-+
-+ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
-+ head->config_names[i] = GUINT32_TO_BE(vmaw->config_names[i]);
-+ head->config_data[i] = GUINT32_TO_BE(vmaw->config_data[i]);
-+ }
-+
-+ /* 32 bytes per device (12 used currently) = 8192 bytes max */
-+ for (i = 1; i <= 254; i++) {
-+ VmaStreamInfo *si = &vmaw->stream_info[i];
-+ if (si->size) {
-+ assert(si->devname);
-+ uint32_t devname_ptr = allocate_header_string(vmaw, si->devname);
-+ if (!devname_ptr) {
-+ return -1;
-+ }
-+ head->dev_info[i].devname_ptr = GUINT32_TO_BE(devname_ptr);
-+ head->dev_info[i].size = GUINT64_TO_BE(si->size);
-+ }
-+ }
-+
-+ uint32_t header_size = sizeof(VmaHeader) + vmaw->header_blob_table_size;
-+ head->header_size = GUINT32_TO_BE(header_size);
-+
-+ if (header_size > HEADERBUF_SIZE) {
-+ return -1; /* just to be sure */
-+ }
-+
-+ uint32_t blob_buffer_offset = sizeof(VmaHeader);
-+ memcpy(buf + blob_buffer_offset, vmaw->header_blob_table,
-+ vmaw->header_blob_table_size);
-+ head->blob_buffer_offset = GUINT32_TO_BE(blob_buffer_offset);
-+ head->blob_buffer_size = GUINT32_TO_BE(vmaw->header_blob_table_pos);
-+
-+ g_checksum_reset(vmaw->md5csum);
-+ g_checksum_update(vmaw->md5csum, (const guchar *)buf, header_size);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmaw->md5csum, (guint8 *)(head->md5sum), &csize);
-+
-+ return vma_queue_write(vmaw, buf, header_size);
-+}
-+
-+static int coroutine_fn vma_writer_flush(VmaWriter *vmaw)
-+{
-+ assert(vmaw);
-+
-+ int ret;
-+ int i;
-+
-+ if (vmaw->status < 0) {
-+ return vmaw->status;
-+ }
-+
-+ if (!vmaw->header_written) {
-+ vmaw->header_written = true;
-+ ret = vma_write_header(vmaw);
-+ if (ret < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_flush: write header failed");
-+ return ret;
-+ }
-+ }
-+
-+ DPRINTF("VMA WRITE FLUSH %d %d\n", vmaw->outbuf_count, vmaw->outbuf_pos);
-+
-+
-+ VmaExtentHeader *ehead = (VmaExtentHeader *)vmaw->outbuf;
-+
-+ ehead->magic = VMA_EXTENT_MAGIC;
-+ ehead->reserved1 = 0;
-+
-+ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
-+ ehead->blockinfo[i] = GUINT64_TO_BE(vmaw->outbuf_block_info[i]);
-+ }
-+
-+ guint16 block_count = (vmaw->outbuf_pos - VMA_EXTENT_HEADER_SIZE) /
-+ VMA_BLOCK_SIZE;
-+
-+ ehead->block_count = GUINT16_TO_BE(block_count);
-+
-+ memcpy(ehead->uuid, vmaw->uuid, sizeof(ehead->uuid));
-+ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
-+
-+ g_checksum_reset(vmaw->md5csum);
-+ g_checksum_update(vmaw->md5csum, vmaw->outbuf, VMA_EXTENT_HEADER_SIZE);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmaw->md5csum, ehead->md5sum, &csize);
-+
-+ int bytes = vmaw->outbuf_pos;
-+ ret = vma_queue_write(vmaw, vmaw->outbuf, bytes);
-+ if (ret != bytes) {
-+ vma_writer_set_error(vmaw, "vma_writer_flush: failed write");
-+ }
-+
-+ vmaw->outbuf_count = 0;
-+ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
-+
-+ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
-+ vmaw->outbuf_block_info[i] = 0;
-+ }
-+
-+ return vmaw->status;
-+}
-+
-+static int vma_count_open_streams(VmaWriter *vmaw)
-+{
-+ g_assert(vmaw != NULL);
-+
-+ int i;
-+ int open_drives = 0;
-+ for (i = 0; i <= 255; i++) {
-+ if (vmaw->stream_info[i].size && !vmaw->stream_info[i].finished) {
-+ open_drives++;
-+ }
-+ }
-+
-+ return open_drives;
-+}
-+
-+
-+/**
-+ * You need to call this if the vma archive does not contain
-+ * any data stream.
-+ */
-+int coroutine_fn
-+vma_writer_flush_output(VmaWriter *vmaw)
-+{
-+ qemu_co_mutex_lock(&vmaw->flush_lock);
-+ int ret = vma_writer_flush(vmaw);
-+ qemu_co_mutex_unlock(&vmaw->flush_lock);
-+ if (ret < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_flush_header failed");
-+ }
-+ return ret;
-+}
-+
-+/**
-+ * all jobs should call this when there is no more data
-+ * Returns: number of remaining stream (0 ==> finished)
-+ */
-+int coroutine_fn
-+vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id)
-+{
-+ g_assert(vmaw != NULL);
-+
-+ DPRINTF("vma_writer_set_status %d\n", dev_id);
-+ if (!vmaw->stream_info[dev_id].size) {
-+ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
-+ "no such stream %d", dev_id);
-+ return -1;
-+ }
-+ if (vmaw->stream_info[dev_id].finished) {
-+ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
-+ "stream already closed %d", dev_id);
-+ return -1;
-+ }
-+
-+ vmaw->stream_info[dev_id].finished = true;
-+
-+ int open_drives = vma_count_open_streams(vmaw);
-+
-+ if (open_drives <= 0) {
-+ DPRINTF("vma_writer_set_status all drives completed\n");
-+ vma_writer_flush_output(vmaw);
-+ }
-+
-+ return open_drives;
-+}
-+
-+int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status)
-+{
-+ int i;
-+
-+ g_assert(vmaw != NULL);
-+
-+ if (status) {
-+ status->status = vmaw->status;
-+ g_strlcpy(status->errmsg, vmaw->errmsg, sizeof(status->errmsg));
-+ for (i = 0; i <= 255; i++) {
-+ status->stream_info[i] = vmaw->stream_info[i];
-+ }
-+
-+ uuid_unparse_lower(vmaw->uuid, status->uuid_str);
-+ }
-+
-+ status->closed = vmaw->closed;
-+
-+ return vmaw->status;
-+}
-+
-+static int vma_writer_get_buffer(VmaWriter *vmaw)
-+{
-+ int ret = 0;
-+
-+ qemu_co_mutex_lock(&vmaw->flush_lock);
-+
-+ /* wait until buffer is available */
-+ while (vmaw->outbuf_count >= (VMA_BLOCKS_PER_EXTENT - 1)) {
-+ ret = vma_writer_flush(vmaw);
-+ if (ret < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_get_buffer: flush failed");
-+ break;
-+ }
-+ }
-+
-+ qemu_co_mutex_unlock(&vmaw->flush_lock);
-+
-+ return ret;
-+}
-+
-+
-+int64_t coroutine_fn
-+vma_writer_write(VmaWriter *vmaw, uint8_t dev_id, int64_t cluster_num,
-+ const unsigned char *buf, size_t *zero_bytes)
-+{
-+ g_assert(vmaw != NULL);
-+ g_assert(zero_bytes != NULL);
-+
-+ *zero_bytes = 0;
-+
-+ if (vmaw->status < 0) {
-+ return vmaw->status;
-+ }
-+
-+ if (!dev_id || !vmaw->stream_info[dev_id].size) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "no such stream %d", dev_id);
-+ return -1;
-+ }
-+
-+ if (vmaw->stream_info[dev_id].finished) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "stream already closed %d", dev_id);
-+ return -1;
-+ }
-+
-+
-+ if (cluster_num >= (((uint64_t)1)<<32)) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "cluster number out of range");
-+ return -1;
-+ }
-+
-+ if (dev_id == vmaw->vmstate_stream) {
-+ if (cluster_num != vmaw->vmstate_clusters) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "non sequential vmstate write");
-+ }
-+ vmaw->vmstate_clusters++;
-+ } else if (cluster_num >= vmaw->stream_info[dev_id].cluster_count) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: cluster number too big");
-+ return -1;
-+ }
-+
-+ /* wait until buffer is available */
-+ if (vma_writer_get_buffer(vmaw) < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "vma_writer_get_buffer failed");
-+ return -1;
-+ }
-+
-+ DPRINTF("VMA WRITE %d %zd\n", dev_id, cluster_num);
-+
-+ uint64_t dev_size = vmaw->stream_info[dev_id].size;
-+ uint16_t mask = 0;
-+
-+ if (buf) {
-+ int i;
-+ int bit = 1;
-+ uint64_t byte_offset = cluster_num * VMA_CLUSTER_SIZE;
-+ for (i = 0; i < 16; i++) {
-+ const unsigned char *vmablock = buf + (i*VMA_BLOCK_SIZE);
-+
-+ // Note: If the source is not 64k-aligned, we might reach 4k blocks
-+ // after the end of the device. Always mark these as zero in the
-+ // mask, so the restore handles them correctly.
-+ if (byte_offset < dev_size &&
-+ !buffer_is_zero(vmablock, VMA_BLOCK_SIZE))
-+ {
-+ mask |= bit;
-+ memcpy(vmaw->outbuf + vmaw->outbuf_pos, vmablock,
-+ VMA_BLOCK_SIZE);
-+
-+ // prevent memory leakage on unaligned last block
-+ if (byte_offset + VMA_BLOCK_SIZE > dev_size) {
-+ uint64_t real_data_in_block = dev_size - byte_offset;
-+ memset(vmaw->outbuf + vmaw->outbuf_pos + real_data_in_block,
-+ 0, VMA_BLOCK_SIZE - real_data_in_block);
-+ }
-+
-+ vmaw->outbuf_pos += VMA_BLOCK_SIZE;
-+ } else {
-+ DPRINTF("VMA WRITE %zd ZERO BLOCK %d\n", cluster_num, i);
-+ vmaw->stream_info[dev_id].zero_bytes += VMA_BLOCK_SIZE;
-+ *zero_bytes += VMA_BLOCK_SIZE;
-+ }
-+
-+ byte_offset += VMA_BLOCK_SIZE;
-+ bit = bit << 1;
-+ }
-+ } else {
-+ DPRINTF("VMA WRITE %zd ZERO CLUSTER\n", cluster_num);
-+ vmaw->stream_info[dev_id].zero_bytes += VMA_CLUSTER_SIZE;
-+ *zero_bytes += VMA_CLUSTER_SIZE;
-+ }
-+
-+ uint64_t block_info = ((uint64_t)mask) << (32+16);
-+ block_info |= ((uint64_t)dev_id) << 32;
-+ block_info |= (cluster_num & 0xffffffff);
-+ vmaw->outbuf_block_info[vmaw->outbuf_count] = block_info;
-+
-+ DPRINTF("VMA WRITE MASK %zd %zx\n", cluster_num, block_info);
-+
-+ vmaw->outbuf_count++;
-+
-+ /** NOTE: We allways write whole clusters, but we correctly set
-+ * transferred bytes. So transferred == size when when everything
-+ * went OK.
-+ */
-+ size_t transferred = VMA_CLUSTER_SIZE;
-+
-+ if (dev_id != vmaw->vmstate_stream) {
-+ uint64_t last = (cluster_num + 1) * VMA_CLUSTER_SIZE;
-+ if (last > dev_size) {
-+ uint64_t diff = last - dev_size;
-+ if (diff >= VMA_CLUSTER_SIZE) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "read after last cluster");
-+ return -1;
-+ }
-+ transferred -= diff;
-+ }
-+ }
-+
-+ vmaw->stream_info[dev_id].transferred += transferred;
-+
-+ return transferred;
-+}
-+
-+void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp)
-+{
-+ if (vmaw->status < 0 && *errp == NULL) {
-+ error_setg(errp, "%s", vmaw->errmsg);
-+ }
-+}
-+
-+int vma_writer_close(VmaWriter *vmaw, Error **errp)
-+{
-+ g_assert(vmaw != NULL);
-+
-+ int i;
-+
-+ qemu_co_mutex_lock(&vmaw->flush_lock); // wait for pending writes
-+
-+ assert(vmaw->co_writer == NULL);
-+
-+ if (vmaw->cmd) {
-+ if (pclose(vmaw->cmd) < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "pclose failed - %s", g_strerror(errno));
-+ }
-+ } else {
-+ if (close(vmaw->fd) < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "close failed - %s", g_strerror(errno));
-+ }
-+ }
-+
-+ for (i = 0; i <= 255; i++) {
-+ VmaStreamInfo *si = &vmaw->stream_info[i];
-+ if (si->size) {
-+ if (!si->finished) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "detected open stream '%s'", si->devname);
-+ } else if ((si->transferred != si->size) &&
-+ (i != vmaw->vmstate_stream)) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "incomplete stream '%s' (%zd != %zd)",
-+ si->devname, si->transferred, si->size);
-+ }
-+ }
-+ }
-+
-+ for (i = 0; i <= 255; i++) {
-+ vmaw->stream_info[i].finished = 1; /* mark as closed */
-+ }
-+
-+ vmaw->closed = 1;
-+
-+ if (vmaw->status < 0 && *errp == NULL) {
-+ error_setg(errp, "%s", vmaw->errmsg);
-+ }
-+
-+ qemu_co_mutex_unlock(&vmaw->flush_lock);
-+
-+ return vmaw->status;
-+}
-+
-+void vma_writer_destroy(VmaWriter *vmaw)
-+{
-+ assert(vmaw);
-+
-+ int i;
-+
-+ for (i = 0; i <= 255; i++) {
-+ if (vmaw->stream_info[i].devname) {
-+ g_free(vmaw->stream_info[i].devname);
-+ }
-+ }
-+
-+ if (vmaw->md5csum) {
-+ g_checksum_free(vmaw->md5csum);
-+ }
-+
-+ qemu_vfree(vmaw->headerbuf);
-+ qemu_vfree(vmaw->outbuf);
-+ g_free(vmaw);
-+}
-diff --git a/vma.c b/vma.c
-new file mode 100644
-index 0000000000..91612d50a2
---- /dev/null
-+++ b/vma.c
-@@ -0,0 +1,850 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) 2012-2013 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+
-+#include "vma.h"
-+#include "qemu-common.h"
-+#include "qemu/module.h"
-+#include "qemu/error-report.h"
-+#include "qemu/main-loop.h"
-+#include "qemu/cutils.h"
-+#include "qemu/memalign.h"
-+#include "qapi/qmp/qdict.h"
-+#include "sysemu/block-backend.h"
-+
-+static void help(void)
-+{
-+ const char *help_msg =
-+ "usage: vma command [command options]\n"
-+ "\n"
-+ "vma list <filename>\n"
-+ "vma config <filename> [-c config]\n"
-+ "vma create <filename> [-c config] pathname ...\n"
-+ "vma extract <filename> [-r <fifo>] <targetdir>\n"
-+ "vma verify <filename> [-v]\n"
-+ ;
-+
-+ printf("%s", help_msg);
-+ exit(1);
-+}
-+
-+static const char *extract_devname(const char *path, char **devname, int index)
-+{
-+ assert(path);
-+
-+ const char *sep = strchr(path, '=');
-+
-+ if (sep) {
-+ *devname = g_strndup(path, sep - path);
-+ path = sep + 1;
-+ } else {
-+ if (index >= 0) {
-+ *devname = g_strdup_printf("disk%d", index);
-+ } else {
-+ *devname = NULL;
-+ }
-+ }
-+
-+ return path;
-+}
-+
-+static void print_content(VmaReader *vmar)
-+{
-+ assert(vmar);
-+
-+ VmaHeader *head = vma_reader_get_header(vmar);
-+
-+ GList *l = vma_reader_get_config_data(vmar);
-+ while (l && l->data) {
-+ VmaConfigData *cdata = (VmaConfigData *)l->data;
-+ l = g_list_next(l);
-+ printf("CFG: size: %d name: %s\n", cdata->len, cdata->name);
-+ }
-+
-+ int i;
-+ VmaDeviceInfo *di;
-+ for (i = 1; i < 255; i++) {
-+ di = vma_reader_get_device_info(vmar, i);
-+ if (di) {
-+ if (strcmp(di->devname, "vmstate") == 0) {
-+ printf("VMSTATE: dev_id=%d memory: %zd\n", i, di->size);
-+ } else {
-+ printf("DEV: dev_id=%d size: %zd devname: %s\n",
-+ i, di->size, di->devname);
-+ }
-+ }
-+ }
-+ /* ctime is the last entry we print */
-+ printf("CTIME: %s", ctime(&head->ctime));
-+ fflush(stdout);
-+}
-+
-+static int list_content(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ const char *filename;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "h");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ default:
-+ g_assert_not_reached();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 1) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ print_content(vmar);
-+
-+ vma_reader_destroy(vmar);
-+
-+ return ret;
-+}
-+
-+typedef struct RestoreMap {
-+ char *devname;
-+ char *path;
-+ char *format;
-+ uint64_t throttling_bps;
-+ char *throttling_group;
-+ char *cache;
-+ bool write_zero;
-+} RestoreMap;
-+
-+static bool try_parse_option(char **line, const char *optname, char **out, const char *inbuf) {
-+ size_t optlen = strlen(optname);
-+ if (strncmp(*line, optname, optlen) != 0 || (*line)[optlen] != '=') {
-+ return false;
-+ }
-+ if (*out) {
-+ g_error("read map failed - duplicate value for option '%s'", optname);
-+ }
-+ char *value = (*line) + optlen + 1; /* including a '=' */
-+ char *colon = strchr(value, ':');
-+ if (!colon) {
-+ g_error("read map failed - option '%s' not terminated ('%s')",
-+ optname, inbuf);
-+ }
-+ *line = colon+1;
-+ *out = g_strndup(value, colon - value);
-+ return true;
-+}
-+
-+static uint64_t verify_u64(const char *text) {
-+ uint64_t value;
-+ const char *endptr = NULL;
-+ if (qemu_strtou64(text, &endptr, 0, &value) != 0 || !endptr || *endptr) {
-+ g_error("read map failed - not a number: %s", text);
-+ }
-+ return value;
-+}
-+
-+static int extract_content(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ int verbose = 0;
-+ const char *filename;
-+ const char *dirname;
-+ const char *readmap = NULL;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hvr:");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'r':
-+ readmap = optarg;
-+ break;
-+ case 'v':
-+ verbose = 1;
-+ break;
-+ default:
-+ help();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 2) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+ dirname = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ if (mkdir(dirname, 0777) < 0) {
-+ g_error("unable to create target directory %s - %s",
-+ dirname, g_strerror(errno));
-+ }
-+
-+ GList *l = vma_reader_get_config_data(vmar);
-+ while (l && l->data) {
-+ VmaConfigData *cdata = (VmaConfigData *)l->data;
-+ l = g_list_next(l);
-+ char *cfgfn = g_strdup_printf("%s/%s", dirname, cdata->name);
-+ GError *err = NULL;
-+ if (!g_file_set_contents(cfgfn, (gchar *)cdata->data, cdata->len,
-+ &err)) {
-+ g_error("unable to write file: %s", err->message);
-+ }
-+ }
-+
-+ GHashTable *devmap = g_hash_table_new(g_str_hash, g_str_equal);
-+
-+ if (readmap) {
-+ print_content(vmar);
-+
-+ FILE *map = fopen(readmap, "r");
-+ if (!map) {
-+ g_error("unable to open fifo %s - %s", readmap, g_strerror(errno));
-+ }
-+
-+ while (1) {
-+ char inbuf[8192];
-+ char *line = fgets(inbuf, sizeof(inbuf), map);
-+ char *format = NULL;
-+ char *bps = NULL;
-+ char *group = NULL;
-+ char *cache = NULL;
-+ if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
-+ break;
-+ }
-+ int len = strlen(line);
-+ if (line[len - 1] == '\n') {
-+ line[len - 1] = '\0';
-+ if (len == 1) {
-+ break;
-+ }
-+ }
-+
-+ while (1) {
-+ if (!try_parse_option(&line, "format", &format, inbuf) &&
-+ !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
-+ !try_parse_option(&line, "throttling.group", &group, inbuf) &&
-+ !try_parse_option(&line, "cache", &cache, inbuf))
-+ {
-+ break;
-+ }
-+ }
-+
-+ uint64_t bps_value = 0;
-+ if (bps) {
-+ bps_value = verify_u64(bps);
-+ g_free(bps);
-+ }
-+
-+ const char *path;
-+ bool write_zero;
-+ if (line[0] == '0' && line[1] == ':') {
-+ path = line + 2;
-+ write_zero = false;
-+ } else if (line[0] == '1' && line[1] == ':') {
-+ path = line + 2;
-+ write_zero = true;
-+ } else {
-+ g_error("read map failed - parse error ('%s')", inbuf);
-+ }
-+
-+ char *devname = NULL;
-+ path = extract_devname(path, &devname, -1);
-+ if (!devname) {
-+ g_error("read map failed - no dev name specified ('%s')",
-+ inbuf);
-+ }
-+
-+ RestoreMap *map = g_new0(RestoreMap, 1);
-+ map->devname = g_strdup(devname);
-+ map->path = g_strdup(path);
-+ map->format = format;
-+ map->throttling_bps = bps_value;
-+ map->throttling_group = group;
-+ map->cache = cache;
-+ map->write_zero = write_zero;
-+
-+ g_hash_table_insert(devmap, map->devname, map);
-+
-+ };
-+ }
-+
-+ int i;
-+ int vmstate_fd = -1;
-+ guint8 vmstate_stream = 0;
-+
-+ for (i = 1; i < 255; i++) {
-+ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
-+ if (di && (strcmp(di->devname, "vmstate") == 0)) {
-+ vmstate_stream = i;
-+ char *statefn = g_strdup_printf("%s/vmstate.bin", dirname);
-+ vmstate_fd = open(statefn, O_WRONLY|O_CREAT|O_EXCL, 0644);
-+ if (vmstate_fd < 0) {
-+ g_error("create vmstate file '%s' failed - %s", statefn,
-+ g_strerror(errno));
-+ }
-+ g_free(statefn);
-+ } else if (di) {
-+ char *devfn = NULL;
-+ const char *format = NULL;
-+ uint64_t throttling_bps = 0;
-+ const char *throttling_group = NULL;
-+ const char *cache = NULL;
-+ int flags = BDRV_O_RDWR;
-+ bool write_zero = true;
-+
-+ BlockBackend *blk = NULL;
-+
-+ if (readmap) {
-+ RestoreMap *map;
-+ map = (RestoreMap *)g_hash_table_lookup(devmap, di->devname);
-+ if (map == NULL) {
-+ g_error("no device name mapping for %s", di->devname);
-+ }
-+ devfn = map->path;
-+ format = map->format;
-+ throttling_bps = map->throttling_bps;
-+ throttling_group = map->throttling_group;
-+ cache = map->cache;
-+ write_zero = map->write_zero;
-+ } else {
-+ devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
-+ dirname, di->devname);
-+ printf("DEVINFO %s %zd\n", devfn, di->size);
-+
-+ bdrv_img_create(devfn, "raw", NULL, NULL, NULL, di->size,
-+ flags, true, &errp);
-+ if (errp) {
-+ g_error("can't create file %s: %s", devfn,
-+ error_get_pretty(errp));
-+ }
-+
-+ /* Note: we created an empty file above, so there is no
-+ * need to write zeroes (so we generate a sparse file)
-+ */
-+ write_zero = false;
-+ }
-+
-+ size_t devlen = strlen(devfn);
-+ QDict *options = NULL;
-+ bool writethrough;
-+ if (format) {
-+ /* explicit format from commandline */
-+ options = qdict_new();
-+ qdict_put_str(options, "driver", format);
-+ } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
-+ strncmp(devfn, "/dev/", 5) == 0)
-+ {
-+ /* This part is now deprecated for PVE as well (just as qemu
-+ * deprecated not specifying an explicit raw format, too.
-+ */
-+ /* explicit raw format */
-+ options = qdict_new();
-+ qdict_put_str(options, "driver", "raw");
-+ }
-+ if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
-+ g_error("invalid cache option: %s\n", cache);
-+ }
-+
-+ if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
-+ g_error("can't open file %s - %s", devfn,
-+ error_get_pretty(errp));
-+ }
-+
-+ if (cache) {
-+ blk_set_enable_write_cache(blk, !writethrough);
-+ }
-+
-+ if (throttling_group) {
-+ blk_io_limits_enable(blk, throttling_group);
-+ }
-+
-+ if (throttling_bps) {
-+ if (!throttling_group) {
-+ blk_io_limits_enable(blk, devfn);
-+ }
-+
-+ ThrottleConfig cfg;
-+ throttle_config_init(&cfg);
-+ cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
-+ Error *err = NULL;
-+ if (!throttle_is_valid(&cfg, &err)) {
-+ error_report_err(err);
-+ g_error("failed to apply throttling");
-+ }
-+ blk_set_io_limits(blk, &cfg);
-+ }
-+
-+ if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ if (!readmap) {
-+ g_free(devfn);
-+ }
-+ }
-+ }
-+
-+ if (vma_reader_restore(vmar, vmstate_fd, verbose, &errp) < 0) {
-+ g_error("restore failed - %s", error_get_pretty(errp));
-+ }
-+
-+ if (!readmap) {
-+ for (i = 1; i < 255; i++) {
-+ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
-+ if (di && (i != vmstate_stream)) {
-+ char *tmpfn = g_strdup_printf("%s/tmp-disk-%s.raw",
-+ dirname, di->devname);
-+ char *fn = g_strdup_printf("%s/disk-%s.raw",
-+ dirname, di->devname);
-+ if (rename(tmpfn, fn) != 0) {
-+ g_error("rename %s to %s failed - %s",
-+ tmpfn, fn, g_strerror(errno));
-+ }
-+ }
-+ }
-+ }
-+
-+ vma_reader_destroy(vmar);
-+
-+ bdrv_close_all();
-+
-+ return ret;
-+}
-+
-+static int verify_content(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ int verbose = 0;
-+ const char *filename;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hv");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'v':
-+ verbose = 1;
-+ break;
-+ default:
-+ help();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 1) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ if (verbose) {
-+ print_content(vmar);
-+ }
-+
-+ if (vma_reader_verify(vmar, verbose, &errp) < 0) {
-+ g_error("verify failed - %s", error_get_pretty(errp));
-+ }
-+
-+ vma_reader_destroy(vmar);
-+
-+ bdrv_close_all();
-+
-+ return ret;
-+}
-+
-+typedef struct BackupJob {
-+ BlockBackend *target;
-+ int64_t len;
-+ VmaWriter *vmaw;
-+ uint8_t dev_id;
-+} BackupJob;
-+
-+#define BACKUP_SECTORS_PER_CLUSTER (VMA_CLUSTER_SIZE / BDRV_SECTOR_SIZE)
-+
-+static void coroutine_fn backup_run_empty(void *opaque)
-+{
-+ VmaWriter *vmaw = (VmaWriter *)opaque;
-+
-+ vma_writer_flush_output(vmaw);
-+
-+ Error *err = NULL;
-+ if (vma_writer_close(vmaw, &err) != 0) {
-+ g_warning("vma_writer_close failed %s", error_get_pretty(err));
-+ }
-+}
-+
-+static void coroutine_fn backup_run(void *opaque)
-+{
-+ BackupJob *job = (BackupJob *)opaque;
-+ struct iovec iov;
-+ QEMUIOVector qiov;
-+
-+ int64_t start, end;
-+ int ret = 0;
-+
-+ unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
-+
-+ start = 0;
-+ end = DIV_ROUND_UP(job->len / BDRV_SECTOR_SIZE,
-+ BACKUP_SECTORS_PER_CLUSTER);
-+
-+ for (; start < end; start++) {
-+ iov.iov_base = buf;
-+ iov.iov_len = VMA_CLUSTER_SIZE;
-+ qemu_iovec_init_external(&qiov, &iov, 1);
-+
-+ ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
-+ VMA_CLUSTER_SIZE, &qiov, 0);
-+ if (ret < 0) {
-+ vma_writer_set_error(job->vmaw, "read error", -1);
-+ goto out;
-+ }
-+
-+ size_t zb = 0;
-+ if (vma_writer_write(job->vmaw, job->dev_id, start, buf, &zb) < 0) {
-+ vma_writer_set_error(job->vmaw, "backup_dump_cb vma_writer_write failed", -1);
-+ goto out;
-+ }
-+ }
-+
-+
-+out:
-+ if (vma_writer_close_stream(job->vmaw, job->dev_id) <= 0) {
-+ Error *err = NULL;
-+ if (vma_writer_close(job->vmaw, &err) != 0) {
-+ g_warning("vma_writer_close failed %s", error_get_pretty(err));
-+ }
-+ }
-+ qemu_vfree(buf);
-+}
-+
-+static int create_archive(int argc, char **argv)
-+{
-+ int i, c;
-+ int verbose = 0;
-+ const char *archivename;
-+ GList *backup_coroutines = NULL;
-+ GList *config_files = NULL;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hvc:");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'c':
-+ config_files = g_list_append(config_files, optarg);
-+ break;
-+ case 'v':
-+ verbose = 1;
-+ break;
-+ default:
-+ g_assert_not_reached();
-+ }
-+ }
-+
-+
-+ /* make sure we an archive name */
-+ if ((optind + 1) > argc) {
-+ help();
-+ }
-+
-+ archivename = argv[optind++];
-+
-+ uuid_t uuid;
-+ uuid_generate(uuid);
-+
-+ Error *local_err = NULL;
-+ VmaWriter *vmaw = vma_writer_create(archivename, uuid, &local_err);
-+
-+ if (vmaw == NULL) {
-+ g_error("%s", error_get_pretty(local_err));
-+ }
-+
-+ GList *l = config_files;
-+ while (l && l->data) {
-+ char *name = l->data;
-+ char *cdata = NULL;
-+ gsize clen = 0;
-+ GError *err = NULL;
-+ if (!g_file_get_contents(name, &cdata, &clen, &err)) {
-+ unlink(archivename);
-+ g_error("Unable to read file: %s", err->message);
-+ }
-+
-+ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
-+ unlink(archivename);
-+ g_error("Unable to append config data %s (len = %zd)",
-+ name, clen);
-+ }
-+ l = g_list_next(l);
-+ }
-+
-+ int devcount = 0;
-+ while (optind < argc) {
-+ const char *path = argv[optind++];
-+ char *devname = NULL;
-+ path = extract_devname(path, &devname, devcount++);
-+
-+ Error *errp = NULL;
-+ BlockBackend *target;
-+
-+ target = blk_new_open(path, NULL, NULL, 0, &errp);
-+ if (!target) {
-+ unlink(archivename);
-+ g_error("bdrv_open '%s' failed - %s", path, error_get_pretty(errp));
-+ }
-+ int64_t size = blk_getlength(target);
-+ int dev_id = vma_writer_register_stream(vmaw, devname, size);
-+ if (dev_id <= 0) {
-+ unlink(archivename);
-+ g_error("vma_writer_register_stream '%s' failed", devname);
-+ }
-+
-+ BackupJob *job = g_new0(BackupJob, 1);
-+ job->len = size;
-+ job->target = target;
-+ job->vmaw = vmaw;
-+ job->dev_id = dev_id;
-+
-+ Coroutine *co = qemu_coroutine_create(backup_run, job);
-+ // Don't enter coroutine yet, because it might write the header before
-+ // all streams can be registered.
-+ backup_coroutines = g_list_append(backup_coroutines, co);
-+ }
-+
-+ VmaStatus vmastat;
-+ int percent = 0;
-+ int last_percent = -1;
-+
-+ if (devcount) {
-+ GList *entry = backup_coroutines;
-+ while (entry && entry->data) {
-+ Coroutine *co = entry->data;
-+ qemu_coroutine_enter(co);
-+ entry = g_list_next(entry);
-+ }
-+
-+ while (1) {
-+ main_loop_wait(false);
-+ vma_writer_get_status(vmaw, &vmastat);
-+
-+ if (verbose) {
-+
-+ uint64_t total = 0;
-+ uint64_t transferred = 0;
-+ uint64_t zero_bytes = 0;
-+
-+ int i;
-+ for (i = 0; i < 256; i++) {
-+ if (vmastat.stream_info[i].size) {
-+ total += vmastat.stream_info[i].size;
-+ transferred += vmastat.stream_info[i].transferred;
-+ zero_bytes += vmastat.stream_info[i].zero_bytes;
-+ }
-+ }
-+ percent = (transferred*100)/total;
-+ if (percent != last_percent) {
-+ fprintf(stderr, "progress %d%% %zd/%zd %zd\n", percent,
-+ transferred, total, zero_bytes);
-+ fflush(stderr);
-+
-+ last_percent = percent;
-+ }
-+ }
-+
-+ if (vmastat.closed) {
-+ break;
-+ }
-+ }
-+ } else {
-+ Coroutine *co = qemu_coroutine_create(backup_run_empty, vmaw);
-+ qemu_coroutine_enter(co);
-+ while (1) {
-+ main_loop_wait(false);
-+ vma_writer_get_status(vmaw, &vmastat);
-+ if (vmastat.closed) {
-+ break;
-+ }
-+ }
-+ }
-+
-+ bdrv_drain_all();
-+
-+ vma_writer_get_status(vmaw, &vmastat);
-+
-+ if (verbose) {
-+ for (i = 0; i < 256; i++) {
-+ VmaStreamInfo *si = &vmastat.stream_info[i];
-+ if (si->size) {
-+ fprintf(stderr, "image %s: size=%zd zeros=%zd saved=%zd\n",
-+ si->devname, si->size, si->zero_bytes,
-+ si->size - si->zero_bytes);
-+ }
-+ }
-+ }
-+
-+ if (vmastat.status < 0) {
-+ unlink(archivename);
-+ g_error("creating vma archive failed");
-+ }
-+
-+ g_list_free(backup_coroutines);
-+ g_list_free(config_files);
-+ vma_writer_destroy(vmaw);
-+ return 0;
-+}
-+
-+static int dump_config(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ const char *filename;
-+ const char *config_name = "qemu-server.conf";
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hc:");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'c':
-+ config_name = optarg;
-+ break;
-+ default:
-+ help();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 1) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ int found = 0;
-+ GList *l = vma_reader_get_config_data(vmar);
-+ while (l && l->data) {
-+ VmaConfigData *cdata = (VmaConfigData *)l->data;
-+ l = g_list_next(l);
-+ if (strcmp(cdata->name, config_name) == 0) {
-+ found = 1;
-+ fwrite(cdata->data, cdata->len, 1, stdout);
-+ break;
-+ }
-+ }
-+
-+ vma_reader_destroy(vmar);
-+
-+ bdrv_close_all();
-+
-+ if (!found) {
-+ fprintf(stderr, "unable to find configuration data '%s'\n", config_name);
-+ return -1;
-+ }
-+
-+ return ret;
-+}
-+
-+int main(int argc, char **argv)
-+{
-+ const char *cmdname;
-+ Error *main_loop_err = NULL;
-+
-+ error_init(argv[0]);
-+ module_call_init(MODULE_INIT_TRACE);
-+ qemu_init_exec_dir(argv[0]);
-+
-+ if (qemu_init_main_loop(&main_loop_err)) {
-+ g_error("%s", error_get_pretty(main_loop_err));
-+ }
-+
-+ bdrv_init();
-+ module_call_init(MODULE_INIT_QOM);
-+
-+ if (argc < 2) {
-+ help();
-+ }
-+
-+ cmdname = argv[1];
-+ argc--; argv++;
-+
-+
-+ if (!strcmp(cmdname, "list")) {
-+ return list_content(argc, argv);
-+ } else if (!strcmp(cmdname, "create")) {
-+ return create_archive(argc, argv);
-+ } else if (!strcmp(cmdname, "extract")) {
-+ return extract_content(argc, argv);
-+ } else if (!strcmp(cmdname, "verify")) {
-+ return verify_content(argc, argv);
-+ } else if (!strcmp(cmdname, "config")) {
-+ return dump_config(argc, argv);
-+ }
-+
-+ help();
-+ return 0;
-+}
-diff --git a/vma.h b/vma.h
-new file mode 100644
-index 0000000000..c895c97f6d
---- /dev/null
-+++ b/vma.h
-@@ -0,0 +1,150 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#ifndef BACKUP_VMA_H
-+#define BACKUP_VMA_H
-+
-+#include <uuid/uuid.h>
-+#include "qapi/error.h"
-+#include "block/block.h"
-+
-+#define VMA_BLOCK_BITS 12
-+#define VMA_BLOCK_SIZE (1<<VMA_BLOCK_BITS)
-+#define VMA_CLUSTER_BITS (VMA_BLOCK_BITS+4)
-+#define VMA_CLUSTER_SIZE (1<<VMA_CLUSTER_BITS)
-+
-+#if VMA_CLUSTER_SIZE != 65536
-+#error unexpected cluster size
-+#endif
-+
-+#define VMA_EXTENT_HEADER_SIZE 512
-+#define VMA_BLOCKS_PER_EXTENT 59
-+#define VMA_MAX_CONFIGS 256
-+
-+#define VMA_MAX_EXTENT_SIZE \
-+ (VMA_EXTENT_HEADER_SIZE+VMA_CLUSTER_SIZE*VMA_BLOCKS_PER_EXTENT)
-+#if VMA_MAX_EXTENT_SIZE != 3867136
-+#error unexpected VMA_EXTENT_SIZE
-+#endif
-+
-+/* File Format Definitions */
-+
-+#define VMA_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|0x00))
-+#define VMA_EXTENT_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|'E'))
-+
-+typedef struct VmaDeviceInfoHeader {
-+ uint32_t devname_ptr; /* offset into blob_buffer table */
-+ uint32_t reserved0;
-+ uint64_t size; /* device size in bytes */
-+ uint64_t reserved1;
-+ uint64_t reserved2;
-+} VmaDeviceInfoHeader;
-+
-+typedef struct VmaHeader {
-+ uint32_t magic;
-+ uint32_t version;
-+ unsigned char uuid[16];
-+ int64_t ctime;
-+ unsigned char md5sum[16];
-+
-+ uint32_t blob_buffer_offset;
-+ uint32_t blob_buffer_size;
-+ uint32_t header_size;
-+
-+ unsigned char reserved[1984];
-+
-+ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+
-+ uint32_t reserved1;
-+
-+ VmaDeviceInfoHeader dev_info[256];
-+} VmaHeader;
-+
-+typedef struct VmaExtentHeader {
-+ uint32_t magic;
-+ uint16_t reserved1;
-+ uint16_t block_count;
-+ unsigned char uuid[16];
-+ unsigned char md5sum[16];
-+ uint64_t blockinfo[VMA_BLOCKS_PER_EXTENT];
-+} VmaExtentHeader;
-+
-+/* functions/definitions to read/write vma files */
-+
-+typedef struct VmaReader VmaReader;
-+
-+typedef struct VmaWriter VmaWriter;
-+
-+typedef struct VmaConfigData {
-+ const char *name;
-+ const void *data;
-+ uint32_t len;
-+} VmaConfigData;
-+
-+typedef struct VmaStreamInfo {
-+ uint64_t size;
-+ uint64_t cluster_count;
-+ uint64_t transferred;
-+ uint64_t zero_bytes;
-+ int finished;
-+ char *devname;
-+} VmaStreamInfo;
-+
-+typedef struct VmaStatus {
-+ int status;
-+ bool closed;
-+ char errmsg[8192];
-+ char uuid_str[37];
-+ VmaStreamInfo stream_info[256];
-+} VmaStatus;
-+
-+typedef struct VmaDeviceInfo {
-+ uint64_t size; /* device size in bytes */
-+ const char *devname;
-+} VmaDeviceInfo;
-+
-+VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp);
-+int vma_writer_close(VmaWriter *vmaw, Error **errp);
-+void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp);
-+void vma_writer_destroy(VmaWriter *vmaw);
-+int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
-+ size_t len);
-+int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
-+ size_t size);
-+
-+int64_t coroutine_fn vma_writer_write(VmaWriter *vmaw, uint8_t dev_id,
-+ int64_t cluster_num,
-+ const unsigned char *buf,
-+ size_t *zero_bytes);
-+
-+int coroutine_fn vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id);
-+int coroutine_fn vma_writer_flush_output(VmaWriter *vmaw);
-+
-+int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status);
-+void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...);
-+
-+
-+VmaReader *vma_reader_create(const char *filename, Error **errp);
-+void vma_reader_destroy(VmaReader *vmar);
-+VmaHeader *vma_reader_get_header(VmaReader *vmar);
-+GList *vma_reader_get_config_data(VmaReader *vmar);
-+VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id);
-+int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id,
-+ BlockBackend *target, bool write_zeroes,
-+ Error **errp);
-+int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
-+ Error **errp);
-+int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp);
-+
-+#endif /* BACKUP_VMA_H */
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 2 Mar 2022 08:35:05 +0100
+Subject: [PATCH] block/backup: move bcs bitmap initialization to job creation
+
+For backing up the state of multiple disks from the same time, a job
+for each disk has to be created. It's convenient if the jobs don't
+have to be started at the same time and if operation of the VM can be
+resumed after job creation. This would lead to a window between job
+creation and running the job, where writes can happen. But no writes
+should happen between setting up the copy-before-write filter and
+setting up the block copy state bitmap, because then new writes would
+just pass through.
+
+Commit 06e0a9c16405c0a4c1eca33cf286cc04c42066a2 moved initalization of
+the bitmap to setting up the copy-before-write filter when sync_mode
+is not MIRROR_SYNC_MODE_BITMAP. Ensure that the bitmap is initialized
+upon job creation for the remaining case too, by moving the
+backup_init_bcs_bitmap call to backup_job_create.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/backup.c | 8 ++++----
+ 1 file changed, 4 insertions(+), 4 deletions(-)
+
+diff --git a/block/backup.c b/block/backup.c
+index b2b649e305..b6fa9e8a69 100644
+--- a/block/backup.c
++++ b/block/backup.c
+@@ -237,8 +237,8 @@ static void backup_init_bcs_bitmap(BackupBlockJob *job)
+ true);
+ } else if (job->sync_mode == MIRROR_SYNC_MODE_TOP) {
+ /*
+- * We can't hog the coroutine to initialize this thoroughly.
+- * Set a flag and resume work when we are able to yield safely.
++ * Initialization is costly here. Simply set a flag and let the
++ * backup_run coroutine resume work once it can yield safely.
+ */
+ block_copy_set_skip_unallocated(job->bcs, true);
+ }
+@@ -252,8 +252,6 @@ static int coroutine_fn backup_run(Job *job, Error **errp)
+ BackupBlockJob *s = container_of(job, BackupBlockJob, common.job);
+ int ret;
+
+- backup_init_bcs_bitmap(s);
+-
+ if (s->sync_mode == MIRROR_SYNC_MODE_TOP) {
+ int64_t offset = 0;
+ int64_t count;
+@@ -492,6 +490,8 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+ block_job_add_bdrv(&job->common, "target", target, 0, BLK_PERM_ALL,
+ &error_abort);
+
++ backup_init_bcs_bitmap(job);
++
+ return &job->common;
+
+ error:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:58 +0200
-Subject: [PATCH] PVE-Backup: add backup-dump block driver
-
-- add backup-dump block driver block/backup-dump.c
-- move BackupBlockJob declaration from block/backup.c to include/block/block_int.h
-- block/backup.c - backup-job-create: also consider source cluster size
-- job.c: make job_should_pause non-static
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/backup-dump.c | 168 +++++++++++++++++++++++++++++++
- block/backup.c | 30 ++----
- block/meson.build | 1 +
- include/block/block_int-common.h | 35 +++++++
- job.c | 3 +-
- 5 files changed, 214 insertions(+), 23 deletions(-)
- create mode 100644 block/backup-dump.c
-
-diff --git a/block/backup-dump.c b/block/backup-dump.c
-new file mode 100644
-index 0000000000..93d7f46950
---- /dev/null
-+++ b/block/backup-dump.c
-@@ -0,0 +1,168 @@
-+/*
-+ * BlockDriver to send backup data stream to a callback function
-+ *
-+ * Copyright (C) 2020 Proxmox Server Solutions GmbH
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include "qemu-common.h"
-+#include "qom/object_interfaces.h"
-+#include "block/block_int.h"
-+
-+typedef struct {
-+ int dump_cb_block_size;
-+ uint64_t byte_size;
-+ BackupDumpFunc *dump_cb;
-+ void *dump_cb_data;
-+} BDRVBackupDumpState;
-+
-+static int qemu_backup_dump_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
-+{
-+ BDRVBackupDumpState *s = bs->opaque;
-+
-+ bdi->cluster_size = s->dump_cb_block_size;
-+ return 0;
-+}
-+
-+static int qemu_backup_dump_check_perm(
-+ BlockDriverState *bs,
-+ uint64_t perm,
-+ uint64_t shared,
-+ Error **errp)
-+{
-+ /* Nothing to do. */
-+ return 0;
-+}
-+
-+static void qemu_backup_dump_set_perm(
-+ BlockDriverState *bs,
-+ uint64_t perm,
-+ uint64_t shared)
-+{
-+ /* Nothing to do. */
-+}
-+
-+static void qemu_backup_dump_abort_perm_update(BlockDriverState *bs)
-+{
-+ /* Nothing to do. */
-+}
-+
-+static void qemu_backup_dump_refresh_limits(BlockDriverState *bs, Error **errp)
-+{
-+ bs->bl.request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O */
-+}
-+
-+static void qemu_backup_dump_close(BlockDriverState *bs)
-+{
-+ /* Nothing to do. */
-+}
-+
-+static int64_t qemu_backup_dump_getlength(BlockDriverState *bs)
-+{
-+ BDRVBackupDumpState *s = bs->opaque;
-+
-+ return s->byte_size;
-+}
-+
-+static coroutine_fn int qemu_backup_dump_co_writev(
-+ BlockDriverState *bs,
-+ int64_t sector_num,
-+ int nb_sectors,
-+ QEMUIOVector *qiov,
-+ int flags)
-+{
-+ /* flags can be only values we set in supported_write_flags */
-+ assert(flags == 0);
-+
-+ BDRVBackupDumpState *s = bs->opaque;
-+ off_t offset = sector_num * BDRV_SECTOR_SIZE;
-+
-+ uint64_t written = 0;
-+
-+ for (int i = 0; i < qiov->niov; ++i) {
-+ const struct iovec *v = &qiov->iov[i];
-+
-+ int rc = s->dump_cb(s->dump_cb_data, offset, v->iov_len, v->iov_base);
-+ if (rc < 0) {
-+ return rc;
-+ }
-+
-+ if (rc != v->iov_len) {
-+ return -EIO;
-+ }
-+
-+ written += v->iov_len;
-+ offset += v->iov_len;
-+ }
-+
-+ return written;
-+}
-+
-+static void qemu_backup_dump_child_perm(
-+ BlockDriverState *bs,
-+ BdrvChild *c,
-+ BdrvChildRole role,
-+ BlockReopenQueue *reopen_queue,
-+ uint64_t perm, uint64_t shared,
-+ uint64_t *nperm, uint64_t *nshared)
-+{
-+ *nperm = BLK_PERM_ALL;
-+ *nshared = BLK_PERM_ALL;
-+}
-+
-+static BlockDriver bdrv_backup_dump_drive = {
-+ .format_name = "backup-dump-drive",
-+ .protocol_name = "backup-dump",
-+ .instance_size = sizeof(BDRVBackupDumpState),
-+
-+ .bdrv_close = qemu_backup_dump_close,
-+ .bdrv_has_zero_init = bdrv_has_zero_init_1,
-+ .bdrv_getlength = qemu_backup_dump_getlength,
-+ .bdrv_get_info = qemu_backup_dump_get_info,
-+
-+ .bdrv_co_writev = qemu_backup_dump_co_writev,
-+
-+ .bdrv_refresh_limits = qemu_backup_dump_refresh_limits,
-+ .bdrv_check_perm = qemu_backup_dump_check_perm,
-+ .bdrv_set_perm = qemu_backup_dump_set_perm,
-+ .bdrv_abort_perm_update = qemu_backup_dump_abort_perm_update,
-+ .bdrv_child_perm = qemu_backup_dump_child_perm,
-+};
-+
-+static void bdrv_backup_dump_init(void)
-+{
-+ bdrv_register(&bdrv_backup_dump_drive);
-+}
-+
-+block_init(bdrv_backup_dump_init);
-+
-+
-+BlockDriverState *bdrv_backup_dump_create(
-+ int dump_cb_block_size,
-+ uint64_t byte_size,
-+ BackupDumpFunc *dump_cb,
-+ void *dump_cb_data,
-+ Error **errp)
-+{
-+ BDRVBackupDumpState *state;
-+ BlockDriverState *bs = bdrv_new_open_driver(
-+ &bdrv_backup_dump_drive, NULL, BDRV_O_RDWR, errp);
-+
-+ if (!bs) {
-+ return NULL;
-+ }
-+
-+ bs->total_sectors = byte_size / BDRV_SECTOR_SIZE;
-+ bs->opaque = state = g_new0(BDRVBackupDumpState, 1);
-+
-+ state->dump_cb_block_size = dump_cb_block_size;
-+ state->byte_size = byte_size;
-+ state->dump_cb = dump_cb;
-+ state->dump_cb_data = dump_cb_data;
-+
-+ return bs;
-+}
-diff --git a/block/backup.c b/block/backup.c
-index 07b899035c..7b5d02f580 100644
---- a/block/backup.c
-+++ b/block/backup.c
-@@ -29,28 +29,6 @@
-
- #include "block/copy-before-write.h"
-
--typedef struct BackupBlockJob {
-- BlockJob common;
-- BlockDriverState *cbw;
-- BlockDriverState *source_bs;
-- BlockDriverState *target_bs;
--
-- BdrvDirtyBitmap *sync_bitmap;
--
-- MirrorSyncMode sync_mode;
-- BitmapSyncMode bitmap_mode;
-- BlockdevOnError on_source_error;
-- BlockdevOnError on_target_error;
-- uint64_t len;
-- int64_t cluster_size;
-- BackupPerf perf;
--
-- BlockCopyState *bcs;
--
-- bool wait;
-- BlockCopyCallState *bg_bcs_call;
--} BackupBlockJob;
--
- static const BlockJobDriver backup_job_driver;
-
- static void backup_cleanup_sync_bitmap(BackupBlockJob *job, int ret)
-@@ -456,6 +434,14 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- }
-
- cluster_size = block_copy_cluster_size(bcs);
-+ if (cluster_size < 0) {
-+ goto error;
-+ }
-+
-+ BlockDriverInfo bdi;
-+ if (bdrv_get_info(bs, &bdi) == 0) {
-+ cluster_size = MAX(cluster_size, bdi.cluster_size);
-+ }
-
- if (perf->max_chunk && perf->max_chunk < cluster_size) {
- error_setg(errp, "Required max-chunk (%" PRIi64 ") is less than backup "
-diff --git a/block/meson.build b/block/meson.build
-index 819eb80951..067708b7c0 100644
---- a/block/meson.build
-+++ b/block/meson.build
-@@ -4,6 +4,7 @@ block_ss.add(files(
- 'aio_task.c',
- 'amend.c',
- 'backup.c',
-+ 'backup-dump.c',
- 'copy-before-write.c',
- 'blkdebug.c',
- 'blklogwrites.c',
-diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h
-index 8947abab76..f272d0d8dc 100644
---- a/include/block/block_int-common.h
-+++ b/include/block/block_int-common.h
-@@ -26,6 +26,7 @@
-
- #include "block/accounting.h"
- #include "block/block.h"
-+#include "block/block-copy.h"
- #include "block/aio-wait.h"
- #include "qemu/queue.h"
- #include "qemu/coroutine.h"
-@@ -64,6 +65,40 @@
-
- #define BLOCK_PROBE_BUF_SIZE 512
-
-+typedef int BackupDumpFunc(void *opaque, uint64_t offset, uint64_t bytes, const void *buf);
-+
-+BlockDriverState *bdrv_backup_dump_create(
-+ int dump_cb_block_size,
-+ uint64_t byte_size,
-+ BackupDumpFunc *dump_cb,
-+ void *dump_cb_data,
-+ Error **errp);
-+
-+// Needs to be defined here, since it's used in blockdev.c to detect PVE backup
-+// jobs with source_bs
-+typedef struct BlockCopyState BlockCopyState;
-+typedef struct BackupBlockJob {
-+ BlockJob common;
-+ BlockDriverState *cbw;
-+ BlockDriverState *source_bs;
-+ BlockDriverState *target_bs;
-+
-+ BdrvDirtyBitmap *sync_bitmap;
-+
-+ MirrorSyncMode sync_mode;
-+ BitmapSyncMode bitmap_mode;
-+ BlockdevOnError on_source_error;
-+ BlockdevOnError on_target_error;
-+ uint64_t len;
-+ int64_t cluster_size;
-+ BackupPerf perf;
-+
-+ BlockCopyState *bcs;
-+
-+ bool wait;
-+ BlockCopyCallState *bg_bcs_call;
-+} BackupBlockJob;
-+
- enum BdrvTrackedRequestType {
- BDRV_TRACKED_READ,
- BDRV_TRACKED_WRITE,
-diff --git a/job.c b/job.c
-index 075c6f3a20..e5699ad200 100644
---- a/job.c
-+++ b/job.c
-@@ -276,7 +276,8 @@ static bool job_started(Job *job)
- return job->co;
- }
-
--static bool job_should_pause(Job *job)
-+bool job_should_pause(Job *job);
-+bool job_should_pause(Job *job)
- {
- return job->pause_count > 0;
- }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:57 +0200
+Subject: [PATCH] PVE-Backup: add vma backup format code
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: create: register all streams before entering coroutines]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/meson.build | 2 +
+ meson.build | 5 +
+ vma-reader.c | 859 ++++++++++++++++++++++++++++++++++++++++++++++
+ vma-writer.c | 791 ++++++++++++++++++++++++++++++++++++++++++
+ vma.c | 849 +++++++++++++++++++++++++++++++++++++++++++++
+ vma.h | 150 ++++++++
+ 6 files changed, 2656 insertions(+)
+ create mode 100644 vma-reader.c
+ create mode 100644 vma-writer.c
+ create mode 100644 vma.c
+ create mode 100644 vma.h
+
+diff --git a/block/meson.build b/block/meson.build
+index ad40c10b6a..3a0b84bc11 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -46,6 +46,8 @@ block_ss.add(files(
+ 'zeroinit.c',
+ ), zstd, zlib, gnutls)
+
++block_ss.add(files('../vma-writer.c'), libuuid)
++
+ softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+
+ if get_option('qcow1').allowed()
+diff --git a/meson.build b/meson.build
+index d5230eadd6..ffff66c0cc 100644
+--- a/meson.build
++++ b/meson.build
+@@ -1462,6 +1462,8 @@ keyutils = dependency('libkeyutils', required: false,
+
+ has_gettid = cc.has_function('gettid')
+
++libuuid = cc.find_library('uuid', required: true)
++
+ # libselinux
+ selinux = dependency('libselinux',
+ required: get_option('selinux'),
+@@ -3607,6 +3609,9 @@ if have_tools
+ dependencies: [blockdev, qemuutil, gnutls, selinux],
+ install: true)
+
++ vma = executable('vma', files('vma.c', 'vma-reader.c') + genh,
++ dependencies: [authz, block, crypto, io, qom], install: true)
++
+ subdir('storage-daemon')
+ subdir('contrib/rdmacm-mux')
+ subdir('contrib/elf2dmp')
+diff --git a/vma-reader.c b/vma-reader.c
+new file mode 100644
+index 0000000000..e65f1e8415
+--- /dev/null
++++ b/vma-reader.c
+@@ -0,0 +1,859 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) 2012 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++#include <uuid/uuid.h>
++
++#include "qemu/timer.h"
++#include "qemu/ratelimit.h"
++#include "vma.h"
++#include "block/block.h"
++#include "sysemu/block-backend.h"
++
++static unsigned char zero_vma_block[VMA_BLOCK_SIZE];
++
++typedef struct VmaRestoreState {
++ BlockBackend *target;
++ bool write_zeroes;
++ unsigned long *bitmap;
++ int bitmap_size;
++} VmaRestoreState;
++
++struct VmaReader {
++ int fd;
++ GChecksum *md5csum;
++ GHashTable *blob_hash;
++ unsigned char *head_data;
++ VmaDeviceInfo devinfo[256];
++ VmaRestoreState rstate[256];
++ GList *cdata_list;
++ guint8 vmstate_stream;
++ uint32_t vmstate_clusters;
++ /* to show restore percentage if run with -v */
++ time_t start_time;
++ int64_t cluster_count;
++ int64_t clusters_read;
++ int64_t zero_cluster_data;
++ int64_t partial_zero_cluster_data;
++ int clusters_read_per;
++};
++
++static guint
++g_int32_hash(gconstpointer v)
++{
++ return *(const uint32_t *)v;
++}
++
++static gboolean
++g_int32_equal(gconstpointer v1, gconstpointer v2)
++{
++ return *((const uint32_t *)v1) == *((const uint32_t *)v2);
++}
++
++static int vma_reader_get_bitmap(VmaRestoreState *rstate, int64_t cluster_num)
++{
++ assert(rstate);
++ assert(rstate->bitmap);
++
++ unsigned long val, idx, bit;
++
++ idx = cluster_num / BITS_PER_LONG;
++
++ assert(rstate->bitmap_size > idx);
++
++ bit = cluster_num % BITS_PER_LONG;
++ val = rstate->bitmap[idx];
++
++ return !!(val & (1UL << bit));
++}
++
++static void vma_reader_set_bitmap(VmaRestoreState *rstate, int64_t cluster_num,
++ int dirty)
++{
++ assert(rstate);
++ assert(rstate->bitmap);
++
++ unsigned long val, idx, bit;
++
++ idx = cluster_num / BITS_PER_LONG;
++
++ assert(rstate->bitmap_size > idx);
++
++ bit = cluster_num % BITS_PER_LONG;
++ val = rstate->bitmap[idx];
++ if (dirty) {
++ if (!(val & (1UL << bit))) {
++ val |= 1UL << bit;
++ }
++ } else {
++ if (val & (1UL << bit)) {
++ val &= ~(1UL << bit);
++ }
++ }
++ rstate->bitmap[idx] = val;
++}
++
++typedef struct VmaBlob {
++ uint32_t start;
++ uint32_t len;
++ void *data;
++} VmaBlob;
++
++static const VmaBlob *get_header_blob(VmaReader *vmar, uint32_t pos)
++{
++ assert(vmar);
++ assert(vmar->blob_hash);
++
++ return g_hash_table_lookup(vmar->blob_hash, &pos);
++}
++
++static const char *get_header_str(VmaReader *vmar, uint32_t pos)
++{
++ const VmaBlob *blob = get_header_blob(vmar, pos);
++ if (!blob) {
++ return NULL;
++ }
++ const char *res = (char *)blob->data;
++ if (res[blob->len-1] != '\0') {
++ return NULL;
++ }
++ return res;
++}
++
++static ssize_t
++safe_read(int fd, unsigned char *buf, size_t count)
++{
++ ssize_t n;
++
++ do {
++ n = read(fd, buf, count);
++ } while (n < 0 && errno == EINTR);
++
++ return n;
++}
++
++static ssize_t
++full_read(int fd, unsigned char *buf, size_t len)
++{
++ ssize_t n;
++ size_t total;
++
++ total = 0;
++
++ while (len > 0) {
++ n = safe_read(fd, buf, len);
++
++ if (n == 0) {
++ return total;
++ }
++
++ if (n <= 0) {
++ break;
++ }
++
++ buf += n;
++ total += n;
++ len -= n;
++ }
++
++ if (len) {
++ return -1;
++ }
++
++ return total;
++}
++
++void vma_reader_destroy(VmaReader *vmar)
++{
++ assert(vmar);
++
++ if (vmar->fd >= 0) {
++ close(vmar->fd);
++ }
++
++ if (vmar->cdata_list) {
++ g_list_free(vmar->cdata_list);
++ }
++
++ int i;
++ for (i = 1; i < 256; i++) {
++ if (vmar->rstate[i].bitmap) {
++ g_free(vmar->rstate[i].bitmap);
++ }
++ if (vmar->rstate[i].target) {
++ blk_unref(vmar->rstate[i].target);
++ }
++ }
++
++ if (vmar->md5csum) {
++ g_checksum_free(vmar->md5csum);
++ }
++
++ if (vmar->blob_hash) {
++ g_hash_table_destroy(vmar->blob_hash);
++ }
++
++ if (vmar->head_data) {
++ g_free(vmar->head_data);
++ }
++
++ g_free(vmar);
++
++};
++
++static int vma_reader_read_head(VmaReader *vmar, Error **errp)
++{
++ assert(vmar);
++ assert(errp);
++ assert(*errp == NULL);
++
++ unsigned char md5sum[16];
++ int i;
++ int ret = 0;
++
++ vmar->head_data = g_malloc(sizeof(VmaHeader));
++
++ if (full_read(vmar->fd, vmar->head_data, sizeof(VmaHeader)) !=
++ sizeof(VmaHeader)) {
++ error_setg(errp, "can't read vma header - %s",
++ errno ? g_strerror(errno) : "got EOF");
++ return -1;
++ }
++
++ VmaHeader *h = (VmaHeader *)vmar->head_data;
++
++ if (h->magic != VMA_MAGIC) {
++ error_setg(errp, "not a vma file - wrong magic number");
++ return -1;
++ }
++
++ uint32_t header_size = GUINT32_FROM_BE(h->header_size);
++ int need = header_size - sizeof(VmaHeader);
++ if (need <= 0) {
++ error_setg(errp, "wrong vma header size %d", header_size);
++ return -1;
++ }
++
++ vmar->head_data = g_realloc(vmar->head_data, header_size);
++ h = (VmaHeader *)vmar->head_data;
++
++ if (full_read(vmar->fd, vmar->head_data + sizeof(VmaHeader), need) !=
++ need) {
++ error_setg(errp, "can't read vma header data - %s",
++ errno ? g_strerror(errno) : "got EOF");
++ return -1;
++ }
++
++ memcpy(md5sum, h->md5sum, 16);
++ memset(h->md5sum, 0, 16);
++
++ g_checksum_reset(vmar->md5csum);
++ g_checksum_update(vmar->md5csum, vmar->head_data, header_size);
++ gsize csize = 16;
++ g_checksum_get_digest(vmar->md5csum, (guint8 *)(h->md5sum), &csize);
++
++ if (memcmp(md5sum, h->md5sum, 16) != 0) {
++ error_setg(errp, "wrong vma header chechsum");
++ return -1;
++ }
++
++ /* we can modify header data after checksum verify */
++ h->header_size = header_size;
++
++ h->version = GUINT32_FROM_BE(h->version);
++ if (h->version != 1) {
++ error_setg(errp, "wrong vma version %d", h->version);
++ return -1;
++ }
++
++ h->ctime = GUINT64_FROM_BE(h->ctime);
++ h->blob_buffer_offset = GUINT32_FROM_BE(h->blob_buffer_offset);
++ h->blob_buffer_size = GUINT32_FROM_BE(h->blob_buffer_size);
++
++ uint32_t bstart = h->blob_buffer_offset + 1;
++ uint32_t bend = h->blob_buffer_offset + h->blob_buffer_size;
++
++ if (bstart <= sizeof(VmaHeader)) {
++ error_setg(errp, "wrong vma blob buffer offset %d",
++ h->blob_buffer_offset);
++ return -1;
++ }
++
++ if (bend > header_size) {
++ error_setg(errp, "wrong vma blob buffer size %d/%d",
++ h->blob_buffer_offset, h->blob_buffer_size);
++ return -1;
++ }
++
++ while ((bstart + 2) <= bend) {
++ uint32_t size = vmar->head_data[bstart] +
++ (vmar->head_data[bstart+1] << 8);
++ if ((bstart + size + 2) <= bend) {
++ VmaBlob *blob = g_new0(VmaBlob, 1);
++ blob->start = bstart - h->blob_buffer_offset;
++ blob->len = size;
++ blob->data = vmar->head_data + bstart + 2;
++ g_hash_table_insert(vmar->blob_hash, &blob->start, blob);
++ }
++ bstart += size + 2;
++ }
++
++
++ int count = 0;
++ for (i = 1; i < 256; i++) {
++ VmaDeviceInfoHeader *dih = &h->dev_info[i];
++ uint32_t devname_ptr = GUINT32_FROM_BE(dih->devname_ptr);
++ uint64_t size = GUINT64_FROM_BE(dih->size);
++ const char *devname = get_header_str(vmar, devname_ptr);
++
++ if (size && devname) {
++ count++;
++ vmar->devinfo[i].size = size;
++ vmar->devinfo[i].devname = devname;
++
++ if (strcmp(devname, "vmstate") == 0) {
++ vmar->vmstate_stream = i;
++ }
++ }
++ }
++
++ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
++ uint32_t name_ptr = GUINT32_FROM_BE(h->config_names[i]);
++ uint32_t data_ptr = GUINT32_FROM_BE(h->config_data[i]);
++
++ if (!(name_ptr && data_ptr)) {
++ continue;
++ }
++ const char *name = get_header_str(vmar, name_ptr);
++ const VmaBlob *blob = get_header_blob(vmar, data_ptr);
++
++ if (!(name && blob)) {
++ error_setg(errp, "vma contains invalid data pointers");
++ return -1;
++ }
++
++ VmaConfigData *cdata = g_new0(VmaConfigData, 1);
++ cdata->name = name;
++ cdata->data = blob->data;
++ cdata->len = blob->len;
++
++ vmar->cdata_list = g_list_append(vmar->cdata_list, cdata);
++ }
++
++ return ret;
++};
++
++VmaReader *vma_reader_create(const char *filename, Error **errp)
++{
++ assert(filename);
++ assert(errp);
++
++ VmaReader *vmar = g_new0(VmaReader, 1);
++
++ if (strcmp(filename, "-") == 0) {
++ vmar->fd = dup(0);
++ } else {
++ vmar->fd = open(filename, O_RDONLY);
++ }
++
++ if (vmar->fd < 0) {
++ error_setg(errp, "can't open file %s - %s\n", filename,
++ g_strerror(errno));
++ goto err;
++ }
++
++ vmar->md5csum = g_checksum_new(G_CHECKSUM_MD5);
++ if (!vmar->md5csum) {
++ error_setg(errp, "can't allocate cmsum\n");
++ goto err;
++ }
++
++ vmar->blob_hash = g_hash_table_new_full(g_int32_hash, g_int32_equal,
++ NULL, g_free);
++
++ if (vma_reader_read_head(vmar, errp) < 0) {
++ goto err;
++ }
++
++ return vmar;
++
++err:
++ if (vmar) {
++ vma_reader_destroy(vmar);
++ }
++
++ return NULL;
++}
++
++VmaHeader *vma_reader_get_header(VmaReader *vmar)
++{
++ assert(vmar);
++ assert(vmar->head_data);
++
++ return (VmaHeader *)(vmar->head_data);
++}
++
++GList *vma_reader_get_config_data(VmaReader *vmar)
++{
++ assert(vmar);
++ assert(vmar->head_data);
++
++ return vmar->cdata_list;
++}
++
++VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id)
++{
++ assert(vmar);
++ assert(dev_id);
++
++ if (vmar->devinfo[dev_id].size && vmar->devinfo[dev_id].devname) {
++ return &vmar->devinfo[dev_id];
++ }
++
++ return NULL;
++}
++
++static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
++ BlockBackend *target, bool write_zeroes)
++{
++ assert(vmar);
++ assert(dev_id);
++
++ vmar->rstate[dev_id].target = target;
++ vmar->rstate[dev_id].write_zeroes = write_zeroes;
++
++ int64_t size = vmar->devinfo[dev_id].size;
++
++ int64_t bitmap_size = (size/BDRV_SECTOR_SIZE) +
++ (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG - 1;
++ bitmap_size /= (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG;
++
++ vmar->rstate[dev_id].bitmap_size = bitmap_size;
++ vmar->rstate[dev_id].bitmap = g_new0(unsigned long, bitmap_size);
++
++ vmar->cluster_count += size/VMA_CLUSTER_SIZE;
++}
++
++int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id, BlockBackend *target,
++ bool write_zeroes, Error **errp)
++{
++ assert(vmar);
++ assert(target != NULL);
++ assert(dev_id);
++ assert(vmar->rstate[dev_id].target == NULL);
++
++ int64_t size = blk_getlength(target);
++ int64_t size_diff = size - vmar->devinfo[dev_id].size;
++
++ /* storage types can have different size restrictions, so it
++ * is not always possible to create an image with exact size.
++ * So we tolerate a size difference up to 4MB.
++ */
++ if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
++ error_setg(errp, "vma_reader_register_bs for stream %s failed - "
++ "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
++ size, vmar->devinfo[dev_id].size);
++ return -1;
++ }
++
++ allocate_rstate(vmar, dev_id, target, write_zeroes);
++
++ return 0;
++}
++
++static ssize_t safe_write(int fd, void *buf, size_t count)
++{
++ ssize_t n;
++
++ do {
++ n = write(fd, buf, count);
++ } while (n < 0 && errno == EINTR);
++
++ return n;
++}
++
++static size_t full_write(int fd, void *buf, size_t len)
++{
++ ssize_t n;
++ size_t total;
++
++ total = 0;
++
++ while (len > 0) {
++ n = safe_write(fd, buf, len);
++ if (n < 0) {
++ return n;
++ }
++ buf += n;
++ total += n;
++ len -= n;
++ }
++
++ if (len) {
++ /* incomplete write ? */
++ return -1;
++ }
++
++ return total;
++}
++
++static int restore_write_data(VmaReader *vmar, guint8 dev_id,
++ BlockBackend *target, int vmstate_fd,
++ unsigned char *buf, int64_t sector_num,
++ int nb_sectors, Error **errp)
++{
++ assert(vmar);
++
++ if (dev_id == vmar->vmstate_stream) {
++ if (vmstate_fd >= 0) {
++ int len = nb_sectors * BDRV_SECTOR_SIZE;
++ int res = full_write(vmstate_fd, buf, len);
++ if (res < 0) {
++ error_setg(errp, "write vmstate failed %d", res);
++ return -1;
++ }
++ }
++ } else {
++ int res = blk_pwrite(target, sector_num * BDRV_SECTOR_SIZE, nb_sectors * BDRV_SECTOR_SIZE, buf, 0);
++ if (res < 0) {
++ error_setg(errp, "blk_pwrite to %s failed (%d)",
++ bdrv_get_device_name(blk_bs(target)), res);
++ return -1;
++ }
++ }
++ return 0;
++}
++
++static int restore_extent(VmaReader *vmar, unsigned char *buf,
++ int extent_size, int vmstate_fd,
++ bool verbose, bool verify, Error **errp)
++{
++ assert(vmar);
++ assert(buf);
++
++ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
++ int start = VMA_EXTENT_HEADER_SIZE;
++ int i;
++
++ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
++ uint64_t block_info = GUINT64_FROM_BE(ehead->blockinfo[i]);
++ uint64_t cluster_num = block_info & 0xffffffff;
++ uint8_t dev_id = (block_info >> 32) & 0xff;
++ uint16_t mask = block_info >> (32+16);
++ int64_t max_sector;
++
++ if (!dev_id) {
++ continue;
++ }
++
++ VmaRestoreState *rstate = &vmar->rstate[dev_id];
++ BlockBackend *target = NULL;
++
++ if (dev_id != vmar->vmstate_stream) {
++ target = rstate->target;
++ if (!verify && !target) {
++ error_setg(errp, "got wrong dev id %d", dev_id);
++ return -1;
++ }
++
++ if (vma_reader_get_bitmap(rstate, cluster_num)) {
++ error_setg(errp, "found duplicated cluster %zd for stream %s",
++ cluster_num, vmar->devinfo[dev_id].devname);
++ return -1;
++ }
++ vma_reader_set_bitmap(rstate, cluster_num, 1);
++
++ max_sector = vmar->devinfo[dev_id].size/BDRV_SECTOR_SIZE;
++ } else {
++ max_sector = G_MAXINT64;
++ if (cluster_num != vmar->vmstate_clusters) {
++ error_setg(errp, "found out of order vmstate data");
++ return -1;
++ }
++ vmar->vmstate_clusters++;
++ }
++
++ vmar->clusters_read++;
++
++ if (verbose) {
++ time_t duration = time(NULL) - vmar->start_time;
++ int percent = (vmar->clusters_read*100)/vmar->cluster_count;
++ if (percent != vmar->clusters_read_per) {
++ printf("progress %d%% (read %zd bytes, duration %zd sec)\n",
++ percent, vmar->clusters_read*VMA_CLUSTER_SIZE,
++ duration);
++ fflush(stdout);
++ vmar->clusters_read_per = percent;
++ }
++ }
++
++ /* try to write whole clusters to speedup restore */
++ if (mask == 0xffff) {
++ if ((start + VMA_CLUSTER_SIZE) > extent_size) {
++ error_setg(errp, "short vma extent - too many blocks");
++ return -1;
++ }
++ int64_t sector_num = (cluster_num * VMA_CLUSTER_SIZE) /
++ BDRV_SECTOR_SIZE;
++ int64_t end_sector = sector_num +
++ VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE;
++
++ if (end_sector > max_sector) {
++ end_sector = max_sector;
++ }
++
++ if (end_sector <= sector_num) {
++ error_setg(errp, "got wrong block address - write beyond end");
++ return -1;
++ }
++
++ if (!verify) {
++ int nb_sectors = end_sector - sector_num;
++ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
++ buf + start, sector_num, nb_sectors,
++ errp) < 0) {
++ return -1;
++ }
++ }
++
++ start += VMA_CLUSTER_SIZE;
++ } else {
++ int j;
++ int bit = 1;
++
++ for (j = 0; j < 16; j++) {
++ int64_t sector_num = (cluster_num*VMA_CLUSTER_SIZE +
++ j*VMA_BLOCK_SIZE)/BDRV_SECTOR_SIZE;
++
++ int64_t end_sector = sector_num +
++ VMA_BLOCK_SIZE/BDRV_SECTOR_SIZE;
++ if (end_sector > max_sector) {
++ end_sector = max_sector;
++ }
++
++ if (mask & bit) {
++ if ((start + VMA_BLOCK_SIZE) > extent_size) {
++ error_setg(errp, "short vma extent - too many blocks");
++ return -1;
++ }
++
++ if (end_sector <= sector_num) {
++ error_setg(errp, "got wrong block address - "
++ "write beyond end");
++ return -1;
++ }
++
++ if (!verify) {
++ int nb_sectors = end_sector - sector_num;
++ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
++ buf + start, sector_num,
++ nb_sectors, errp) < 0) {
++ return -1;
++ }
++ }
++
++ start += VMA_BLOCK_SIZE;
++
++ } else {
++
++
++ if (end_sector > sector_num) {
++ /* Todo: use bdrv_co_write_zeroes (but that need to
++ * be run inside coroutine?)
++ */
++ int nb_sectors = end_sector - sector_num;
++ int zero_size = BDRV_SECTOR_SIZE*nb_sectors;
++ vmar->zero_cluster_data += zero_size;
++ if (mask != 0) {
++ vmar->partial_zero_cluster_data += zero_size;
++ }
++
++ if (rstate->write_zeroes && !verify) {
++ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
++ zero_vma_block, sector_num,
++ nb_sectors, errp) < 0) {
++ return -1;
++ }
++ }
++ }
++ }
++
++ bit = bit << 1;
++ }
++ }
++ }
++
++ if (start != extent_size) {
++ error_setg(errp, "vma extent error - missing blocks");
++ return -1;
++ }
++
++ return 0;
++}
++
++static int vma_reader_restore_full(VmaReader *vmar, int vmstate_fd,
++ bool verbose, bool verify,
++ Error **errp)
++{
++ assert(vmar);
++ assert(vmar->head_data);
++
++ int ret = 0;
++ unsigned char buf[VMA_MAX_EXTENT_SIZE];
++ int buf_pos = 0;
++ unsigned char md5sum[16];
++ VmaHeader *h = (VmaHeader *)vmar->head_data;
++
++ vmar->start_time = time(NULL);
++
++ while (1) {
++ int bytes = full_read(vmar->fd, buf + buf_pos, sizeof(buf) - buf_pos);
++ if (bytes < 0) {
++ error_setg(errp, "read failed - %s", g_strerror(errno));
++ return -1;
++ }
++
++ buf_pos += bytes;
++
++ if (!buf_pos) {
++ break; /* EOF */
++ }
++
++ if (buf_pos < VMA_EXTENT_HEADER_SIZE) {
++ error_setg(errp, "read short extent (%d bytes)", buf_pos);
++ return -1;
++ }
++
++ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
++
++ /* extract md5sum */
++ memcpy(md5sum, ehead->md5sum, sizeof(ehead->md5sum));
++ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
++
++ g_checksum_reset(vmar->md5csum);
++ g_checksum_update(vmar->md5csum, buf, VMA_EXTENT_HEADER_SIZE);
++ gsize csize = 16;
++ g_checksum_get_digest(vmar->md5csum, ehead->md5sum, &csize);
++
++ if (memcmp(md5sum, ehead->md5sum, 16) != 0) {
++ error_setg(errp, "wrong vma extent header chechsum");
++ return -1;
++ }
++
++ if (memcmp(h->uuid, ehead->uuid, sizeof(ehead->uuid)) != 0) {
++ error_setg(errp, "wrong vma extent uuid");
++ return -1;
++ }
++
++ if (ehead->magic != VMA_EXTENT_MAGIC || ehead->reserved1 != 0) {
++ error_setg(errp, "wrong vma extent header magic");
++ return -1;
++ }
++
++ int block_count = GUINT16_FROM_BE(ehead->block_count);
++ int extent_size = VMA_EXTENT_HEADER_SIZE + block_count*VMA_BLOCK_SIZE;
++
++ if (buf_pos < extent_size) {
++ error_setg(errp, "short vma extent (%d < %d)", buf_pos,
++ extent_size);
++ return -1;
++ }
++
++ if (restore_extent(vmar, buf, extent_size, vmstate_fd, verbose,
++ verify, errp) < 0) {
++ return -1;
++ }
++
++ if (buf_pos > extent_size) {
++ memmove(buf, buf + extent_size, buf_pos - extent_size);
++ buf_pos = buf_pos - extent_size;
++ } else {
++ buf_pos = 0;
++ }
++ }
++
++ bdrv_drain_all();
++
++ int i;
++ for (i = 1; i < 256; i++) {
++ VmaRestoreState *rstate = &vmar->rstate[i];
++ if (!rstate->target) {
++ continue;
++ }
++
++ if (blk_flush(rstate->target) < 0) {
++ error_setg(errp, "vma blk_flush %s failed",
++ vmar->devinfo[i].devname);
++ return -1;
++ }
++
++ if (vmar->devinfo[i].size &&
++ (strcmp(vmar->devinfo[i].devname, "vmstate") != 0)) {
++ assert(rstate->bitmap);
++
++ int64_t cluster_num, end;
++
++ end = (vmar->devinfo[i].size + VMA_CLUSTER_SIZE - 1) /
++ VMA_CLUSTER_SIZE;
++
++ for (cluster_num = 0; cluster_num < end; cluster_num++) {
++ if (!vma_reader_get_bitmap(rstate, cluster_num)) {
++ error_setg(errp, "detected missing cluster %zd "
++ "for stream %s", cluster_num,
++ vmar->devinfo[i].devname);
++ return -1;
++ }
++ }
++ }
++ }
++
++ if (verbose) {
++ if (vmar->clusters_read) {
++ printf("total bytes read %zd, sparse bytes %zd (%.3g%%)\n",
++ vmar->clusters_read*VMA_CLUSTER_SIZE,
++ vmar->zero_cluster_data,
++ (double)(100.0*vmar->zero_cluster_data)/
++ (vmar->clusters_read*VMA_CLUSTER_SIZE));
++
++ int64_t datasize = vmar->clusters_read*VMA_CLUSTER_SIZE-vmar->zero_cluster_data;
++ if (datasize) { // this does not make sense for empty files
++ printf("space reduction due to 4K zero blocks %.3g%%\n",
++ (double)(100.0*vmar->partial_zero_cluster_data) / datasize);
++ }
++ } else {
++ printf("vma archive contains no image data\n");
++ }
++ }
++ return ret;
++}
++
++int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
++ Error **errp)
++{
++ return vma_reader_restore_full(vmar, vmstate_fd, verbose, false, errp);
++}
++
++int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp)
++{
++ guint8 dev_id;
++
++ for (dev_id = 1; dev_id < 255; dev_id++) {
++ if (vma_reader_get_device_info(vmar, dev_id)) {
++ allocate_rstate(vmar, dev_id, NULL, false);
++ }
++ }
++
++ return vma_reader_restore_full(vmar, -1, verbose, true, errp);
++}
++
+diff --git a/vma-writer.c b/vma-writer.c
+new file mode 100644
+index 0000000000..df4b20793d
+--- /dev/null
++++ b/vma-writer.c
+@@ -0,0 +1,791 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) 2012 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++#include <uuid/uuid.h>
++
++#include "vma.h"
++#include "block/block.h"
++#include "monitor/monitor.h"
++#include "qemu/main-loop.h"
++#include "qemu/coroutine.h"
++#include "qemu/cutils.h"
++#include "qemu/memalign.h"
++
++#define DEBUG_VMA 0
++
++#define DPRINTF(fmt, ...)\
++ do { if (DEBUG_VMA) { printf("vma: " fmt, ## __VA_ARGS__); } } while (0)
++
++#define WRITE_BUFFERS 5
++#define HEADER_CLUSTERS 8
++#define HEADERBUF_SIZE (VMA_CLUSTER_SIZE*HEADER_CLUSTERS)
++
++struct VmaWriter {
++ int fd;
++ FILE *cmd;
++ int status;
++ char errmsg[8192];
++ uuid_t uuid;
++ bool header_written;
++ bool closed;
++
++ /* we always write extents */
++ unsigned char *outbuf;
++ int outbuf_pos; /* in bytes */
++ int outbuf_count; /* in VMA_BLOCKS */
++ uint64_t outbuf_block_info[VMA_BLOCKS_PER_EXTENT];
++
++ unsigned char *headerbuf;
++
++ GChecksum *md5csum;
++ CoMutex flush_lock;
++ Coroutine *co_writer;
++
++ /* drive informations */
++ VmaStreamInfo stream_info[256];
++ guint stream_count;
++
++ guint8 vmstate_stream;
++ uint32_t vmstate_clusters;
++
++ /* header blob table */
++ char *header_blob_table;
++ uint32_t header_blob_table_size;
++ uint32_t header_blob_table_pos;
++
++ /* store for config blobs */
++ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++ uint32_t config_count;
++};
++
++void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...)
++{
++ va_list ap;
++
++ if (vmaw->status < 0) {
++ return;
++ }
++
++ vmaw->status = -1;
++
++ va_start(ap, fmt);
++ g_vsnprintf(vmaw->errmsg, sizeof(vmaw->errmsg), fmt, ap);
++ va_end(ap);
++
++ DPRINTF("vma_writer_set_error: %s\n", vmaw->errmsg);
++}
++
++static uint32_t allocate_header_blob(VmaWriter *vmaw, const char *data,
++ size_t len)
++{
++ if (len > 65535) {
++ return 0;
++ }
++
++ if (!vmaw->header_blob_table ||
++ (vmaw->header_blob_table_size <
++ (vmaw->header_blob_table_pos + len + 2))) {
++ int newsize = vmaw->header_blob_table_size + ((len + 2 + 511)/512)*512;
++
++ vmaw->header_blob_table = g_realloc(vmaw->header_blob_table, newsize);
++ memset(vmaw->header_blob_table + vmaw->header_blob_table_size,
++ 0, newsize - vmaw->header_blob_table_size);
++ vmaw->header_blob_table_size = newsize;
++ }
++
++ uint32_t cpos = vmaw->header_blob_table_pos;
++ vmaw->header_blob_table[cpos] = len & 255;
++ vmaw->header_blob_table[cpos+1] = (len >> 8) & 255;
++ memcpy(vmaw->header_blob_table + cpos + 2, data, len);
++ vmaw->header_blob_table_pos += len + 2;
++ return cpos;
++}
++
++static uint32_t allocate_header_string(VmaWriter *vmaw, const char *str)
++{
++ assert(vmaw);
++
++ size_t len = strlen(str) + 1;
++
++ return allocate_header_blob(vmaw, str, len);
++}
++
++int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
++ gsize len)
++{
++ assert(vmaw);
++ assert(!vmaw->header_written);
++ assert(vmaw->config_count < VMA_MAX_CONFIGS);
++ assert(name);
++ assert(data);
++
++ gchar *basename = g_path_get_basename(name);
++ uint32_t name_ptr = allocate_header_string(vmaw, basename);
++ g_free(basename);
++
++ if (!name_ptr) {
++ return -1;
++ }
++
++ uint32_t data_ptr = allocate_header_blob(vmaw, data, len);
++ if (!data_ptr) {
++ return -1;
++ }
++
++ vmaw->config_names[vmaw->config_count] = name_ptr;
++ vmaw->config_data[vmaw->config_count] = data_ptr;
++
++ vmaw->config_count++;
++
++ return 0;
++}
++
++int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
++ size_t size)
++{
++ assert(vmaw);
++ assert(devname);
++ assert(!vmaw->status);
++
++ if (vmaw->header_written) {
++ vma_writer_set_error(vmaw, "vma_writer_register_stream: header "
++ "already written");
++ return -1;
++ }
++
++ guint n = vmaw->stream_count + 1;
++
++ /* we can have dev_ids form 1 to 255 (0 reserved)
++ * 255(-1) reseverd for safety
++ */
++ if (n > 254) {
++ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
++ "too many drives");
++ return -1;
++ }
++
++ if (size <= 0) {
++ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
++ "got strange size %zd", size);
++ return -1;
++ }
++
++ DPRINTF("vma_writer_register_stream %s %zu %d\n", devname, size, n);
++
++ vmaw->stream_info[n].devname = g_strdup(devname);
++ vmaw->stream_info[n].size = size;
++
++ vmaw->stream_info[n].cluster_count = (size + VMA_CLUSTER_SIZE - 1) /
++ VMA_CLUSTER_SIZE;
++
++ vmaw->stream_count = n;
++
++ if (strcmp(devname, "vmstate") == 0) {
++ vmaw->vmstate_stream = n;
++ }
++
++ return n;
++}
++
++static void coroutine_fn yield_until_fd_writable(int fd)
++{
++ assert(qemu_in_coroutine());
++ AioContext *ctx = qemu_get_current_aio_context();
++ aio_set_fd_handler(ctx, fd, false, NULL, (IOHandler *)qemu_coroutine_enter,
++ NULL, NULL, qemu_coroutine_self());
++ qemu_coroutine_yield();
++ aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL, NULL);
++}
++
++static ssize_t coroutine_fn
++vma_queue_write(VmaWriter *vmaw, const void *buf, size_t bytes)
++{
++ DPRINTF("vma_queue_write enter %zd\n", bytes);
++
++ assert(vmaw);
++ assert(buf);
++ assert(bytes <= VMA_MAX_EXTENT_SIZE);
++
++ size_t done = 0;
++ ssize_t ret;
++
++ assert(vmaw->co_writer == NULL);
++
++ vmaw->co_writer = qemu_coroutine_self();
++
++ while (done < bytes) {
++ if (vmaw->status < 0) {
++ DPRINTF("vma_queue_write detected canceled backup\n");
++ done = -1;
++ break;
++ }
++ yield_until_fd_writable(vmaw->fd);
++ ret = write(vmaw->fd, buf + done, bytes - done);
++ if (ret > 0) {
++ done += ret;
++ DPRINTF("vma_queue_write written %zd %zd\n", done, ret);
++ } else if (ret < 0) {
++ if (errno == EAGAIN || errno == EWOULDBLOCK) {
++ /* try again */
++ } else {
++ vma_writer_set_error(vmaw, "vma_queue_write: write error - %s",
++ g_strerror(errno));
++ done = -1; /* always return failure for partial writes */
++ break;
++ }
++ } else if (ret == 0) {
++ /* should not happen - simply try again */
++ }
++ }
++
++ vmaw->co_writer = NULL;
++
++ return (done == bytes) ? bytes : -1;
++}
++
++VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
++{
++ const char *p;
++
++ assert(sizeof(VmaHeader) == (4096 + 8192));
++ assert(G_STRUCT_OFFSET(VmaHeader, config_names) == 2044);
++ assert(G_STRUCT_OFFSET(VmaHeader, config_data) == 3068);
++ assert(G_STRUCT_OFFSET(VmaHeader, dev_info) == 4096);
++ assert(sizeof(VmaExtentHeader) == 512);
++
++ VmaWriter *vmaw = g_new0(VmaWriter, 1);
++ vmaw->fd = -1;
++
++ vmaw->md5csum = g_checksum_new(G_CHECKSUM_MD5);
++ if (!vmaw->md5csum) {
++ error_setg(errp, "can't allocate cmsum\n");
++ goto err;
++ }
++
++ if (strstart(filename, "exec:", &p)) {
++ vmaw->cmd = popen(p, "w");
++ if (vmaw->cmd == NULL) {
++ error_setg(errp, "can't popen command '%s' - %s\n", p,
++ g_strerror(errno));
++ goto err;
++ }
++ vmaw->fd = fileno(vmaw->cmd);
++
++ /* try to use O_NONBLOCK */
++ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
++
++ } else {
++ struct stat st;
++ int oflags;
++ const char *tmp_id_str;
++
++ if ((stat(filename, &st) == 0) && S_ISFIFO(st.st_mode)) {
++ oflags = O_NONBLOCK|O_WRONLY;
++ vmaw->fd = qemu_open(filename, oflags, errp);
++ } else if (strstart(filename, "/dev/fdset/", &tmp_id_str)) {
++ oflags = O_NONBLOCK|O_WRONLY;
++ vmaw->fd = qemu_open(filename, oflags, errp);
++ } else if (strstart(filename, "/dev/fdname/", &tmp_id_str)) {
++ vmaw->fd = monitor_get_fd(monitor_cur(), tmp_id_str, errp);
++ if (vmaw->fd < 0) {
++ goto err;
++ }
++ /* try to use O_NONBLOCK */
++ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
++ } else {
++ oflags = O_NONBLOCK|O_DIRECT|O_WRONLY|O_EXCL;
++ vmaw->fd = qemu_create(filename, oflags, 0644, errp);
++ }
++
++ if (vmaw->fd < 0) {
++ error_setg(errp, "can't open file %s - %s\n", filename,
++ g_strerror(errno));
++ goto err;
++ }
++ }
++
++ /* we use O_DIRECT, so we need to align IO buffers */
++
++ vmaw->outbuf = qemu_memalign(512, VMA_MAX_EXTENT_SIZE);
++ vmaw->headerbuf = qemu_memalign(512, HEADERBUF_SIZE);
++
++ vmaw->outbuf_count = 0;
++ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
++
++ vmaw->header_blob_table_pos = 1; /* start at pos 1 */
++
++ qemu_co_mutex_init(&vmaw->flush_lock);
++
++ uuid_copy(vmaw->uuid, uuid);
++
++ return vmaw;
++
++err:
++ if (vmaw) {
++ if (vmaw->cmd) {
++ pclose(vmaw->cmd);
++ } else if (vmaw->fd >= 0) {
++ close(vmaw->fd);
++ }
++
++ if (vmaw->md5csum) {
++ g_checksum_free(vmaw->md5csum);
++ }
++
++ g_free(vmaw);
++ }
++
++ return NULL;
++}
++
++static int coroutine_fn vma_write_header(VmaWriter *vmaw)
++{
++ assert(vmaw);
++ unsigned char *buf = vmaw->headerbuf;
++ VmaHeader *head = (VmaHeader *)buf;
++
++ int i;
++
++ DPRINTF("VMA WRITE HEADER\n");
++
++ if (vmaw->status < 0) {
++ return vmaw->status;
++ }
++
++ memset(buf, 0, HEADERBUF_SIZE);
++
++ head->magic = VMA_MAGIC;
++ head->version = GUINT32_TO_BE(1); /* v1 */
++ memcpy(head->uuid, vmaw->uuid, 16);
++
++ time_t ctime = time(NULL);
++ head->ctime = GUINT64_TO_BE(ctime);
++
++ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
++ head->config_names[i] = GUINT32_TO_BE(vmaw->config_names[i]);
++ head->config_data[i] = GUINT32_TO_BE(vmaw->config_data[i]);
++ }
++
++ /* 32 bytes per device (12 used currently) = 8192 bytes max */
++ for (i = 1; i <= 254; i++) {
++ VmaStreamInfo *si = &vmaw->stream_info[i];
++ if (si->size) {
++ assert(si->devname);
++ uint32_t devname_ptr = allocate_header_string(vmaw, si->devname);
++ if (!devname_ptr) {
++ return -1;
++ }
++ head->dev_info[i].devname_ptr = GUINT32_TO_BE(devname_ptr);
++ head->dev_info[i].size = GUINT64_TO_BE(si->size);
++ }
++ }
++
++ uint32_t header_size = sizeof(VmaHeader) + vmaw->header_blob_table_size;
++ head->header_size = GUINT32_TO_BE(header_size);
++
++ if (header_size > HEADERBUF_SIZE) {
++ return -1; /* just to be sure */
++ }
++
++ uint32_t blob_buffer_offset = sizeof(VmaHeader);
++ memcpy(buf + blob_buffer_offset, vmaw->header_blob_table,
++ vmaw->header_blob_table_size);
++ head->blob_buffer_offset = GUINT32_TO_BE(blob_buffer_offset);
++ head->blob_buffer_size = GUINT32_TO_BE(vmaw->header_blob_table_pos);
++
++ g_checksum_reset(vmaw->md5csum);
++ g_checksum_update(vmaw->md5csum, (const guchar *)buf, header_size);
++ gsize csize = 16;
++ g_checksum_get_digest(vmaw->md5csum, (guint8 *)(head->md5sum), &csize);
++
++ return vma_queue_write(vmaw, buf, header_size);
++}
++
++static int coroutine_fn vma_writer_flush(VmaWriter *vmaw)
++{
++ assert(vmaw);
++
++ int ret;
++ int i;
++
++ if (vmaw->status < 0) {
++ return vmaw->status;
++ }
++
++ if (!vmaw->header_written) {
++ vmaw->header_written = true;
++ ret = vma_write_header(vmaw);
++ if (ret < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_flush: write header failed");
++ return ret;
++ }
++ }
++
++ DPRINTF("VMA WRITE FLUSH %d %d\n", vmaw->outbuf_count, vmaw->outbuf_pos);
++
++
++ VmaExtentHeader *ehead = (VmaExtentHeader *)vmaw->outbuf;
++
++ ehead->magic = VMA_EXTENT_MAGIC;
++ ehead->reserved1 = 0;
++
++ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
++ ehead->blockinfo[i] = GUINT64_TO_BE(vmaw->outbuf_block_info[i]);
++ }
++
++ guint16 block_count = (vmaw->outbuf_pos - VMA_EXTENT_HEADER_SIZE) /
++ VMA_BLOCK_SIZE;
++
++ ehead->block_count = GUINT16_TO_BE(block_count);
++
++ memcpy(ehead->uuid, vmaw->uuid, sizeof(ehead->uuid));
++ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
++
++ g_checksum_reset(vmaw->md5csum);
++ g_checksum_update(vmaw->md5csum, vmaw->outbuf, VMA_EXTENT_HEADER_SIZE);
++ gsize csize = 16;
++ g_checksum_get_digest(vmaw->md5csum, ehead->md5sum, &csize);
++
++ int bytes = vmaw->outbuf_pos;
++ ret = vma_queue_write(vmaw, vmaw->outbuf, bytes);
++ if (ret != bytes) {
++ vma_writer_set_error(vmaw, "vma_writer_flush: failed write");
++ }
++
++ vmaw->outbuf_count = 0;
++ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
++
++ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
++ vmaw->outbuf_block_info[i] = 0;
++ }
++
++ return vmaw->status;
++}
++
++static int vma_count_open_streams(VmaWriter *vmaw)
++{
++ g_assert(vmaw != NULL);
++
++ int i;
++ int open_drives = 0;
++ for (i = 0; i <= 255; i++) {
++ if (vmaw->stream_info[i].size && !vmaw->stream_info[i].finished) {
++ open_drives++;
++ }
++ }
++
++ return open_drives;
++}
++
++
++/**
++ * You need to call this if the vma archive does not contain
++ * any data stream.
++ */
++int coroutine_fn
++vma_writer_flush_output(VmaWriter *vmaw)
++{
++ qemu_co_mutex_lock(&vmaw->flush_lock);
++ int ret = vma_writer_flush(vmaw);
++ qemu_co_mutex_unlock(&vmaw->flush_lock);
++ if (ret < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_flush_header failed");
++ }
++ return ret;
++}
++
++/**
++ * all jobs should call this when there is no more data
++ * Returns: number of remaining stream (0 ==> finished)
++ */
++int coroutine_fn
++vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id)
++{
++ g_assert(vmaw != NULL);
++
++ DPRINTF("vma_writer_set_status %d\n", dev_id);
++ if (!vmaw->stream_info[dev_id].size) {
++ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
++ "no such stream %d", dev_id);
++ return -1;
++ }
++ if (vmaw->stream_info[dev_id].finished) {
++ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
++ "stream already closed %d", dev_id);
++ return -1;
++ }
++
++ vmaw->stream_info[dev_id].finished = true;
++
++ int open_drives = vma_count_open_streams(vmaw);
++
++ if (open_drives <= 0) {
++ DPRINTF("vma_writer_set_status all drives completed\n");
++ vma_writer_flush_output(vmaw);
++ }
++
++ return open_drives;
++}
++
++int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status)
++{
++ int i;
++
++ g_assert(vmaw != NULL);
++
++ if (status) {
++ status->status = vmaw->status;
++ g_strlcpy(status->errmsg, vmaw->errmsg, sizeof(status->errmsg));
++ for (i = 0; i <= 255; i++) {
++ status->stream_info[i] = vmaw->stream_info[i];
++ }
++
++ uuid_unparse_lower(vmaw->uuid, status->uuid_str);
++ }
++
++ status->closed = vmaw->closed;
++
++ return vmaw->status;
++}
++
++static int vma_writer_get_buffer(VmaWriter *vmaw)
++{
++ int ret = 0;
++
++ qemu_co_mutex_lock(&vmaw->flush_lock);
++
++ /* wait until buffer is available */
++ while (vmaw->outbuf_count >= (VMA_BLOCKS_PER_EXTENT - 1)) {
++ ret = vma_writer_flush(vmaw);
++ if (ret < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_get_buffer: flush failed");
++ break;
++ }
++ }
++
++ qemu_co_mutex_unlock(&vmaw->flush_lock);
++
++ return ret;
++}
++
++
++int64_t coroutine_fn
++vma_writer_write(VmaWriter *vmaw, uint8_t dev_id, int64_t cluster_num,
++ const unsigned char *buf, size_t *zero_bytes)
++{
++ g_assert(vmaw != NULL);
++ g_assert(zero_bytes != NULL);
++
++ *zero_bytes = 0;
++
++ if (vmaw->status < 0) {
++ return vmaw->status;
++ }
++
++ if (!dev_id || !vmaw->stream_info[dev_id].size) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "no such stream %d", dev_id);
++ return -1;
++ }
++
++ if (vmaw->stream_info[dev_id].finished) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "stream already closed %d", dev_id);
++ return -1;
++ }
++
++
++ if (cluster_num >= (((uint64_t)1)<<32)) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "cluster number out of range");
++ return -1;
++ }
++
++ if (dev_id == vmaw->vmstate_stream) {
++ if (cluster_num != vmaw->vmstate_clusters) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "non sequential vmstate write");
++ }
++ vmaw->vmstate_clusters++;
++ } else if (cluster_num >= vmaw->stream_info[dev_id].cluster_count) {
++ vma_writer_set_error(vmaw, "vma_writer_write: cluster number too big");
++ return -1;
++ }
++
++ /* wait until buffer is available */
++ if (vma_writer_get_buffer(vmaw) < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "vma_writer_get_buffer failed");
++ return -1;
++ }
++
++ DPRINTF("VMA WRITE %d %zd\n", dev_id, cluster_num);
++
++ uint64_t dev_size = vmaw->stream_info[dev_id].size;
++ uint16_t mask = 0;
++
++ if (buf) {
++ int i;
++ int bit = 1;
++ uint64_t byte_offset = cluster_num * VMA_CLUSTER_SIZE;
++ for (i = 0; i < 16; i++) {
++ const unsigned char *vmablock = buf + (i*VMA_BLOCK_SIZE);
++
++ // Note: If the source is not 64k-aligned, we might reach 4k blocks
++ // after the end of the device. Always mark these as zero in the
++ // mask, so the restore handles them correctly.
++ if (byte_offset < dev_size &&
++ !buffer_is_zero(vmablock, VMA_BLOCK_SIZE))
++ {
++ mask |= bit;
++ memcpy(vmaw->outbuf + vmaw->outbuf_pos, vmablock,
++ VMA_BLOCK_SIZE);
++
++ // prevent memory leakage on unaligned last block
++ if (byte_offset + VMA_BLOCK_SIZE > dev_size) {
++ uint64_t real_data_in_block = dev_size - byte_offset;
++ memset(vmaw->outbuf + vmaw->outbuf_pos + real_data_in_block,
++ 0, VMA_BLOCK_SIZE - real_data_in_block);
++ }
++
++ vmaw->outbuf_pos += VMA_BLOCK_SIZE;
++ } else {
++ DPRINTF("VMA WRITE %zd ZERO BLOCK %d\n", cluster_num, i);
++ vmaw->stream_info[dev_id].zero_bytes += VMA_BLOCK_SIZE;
++ *zero_bytes += VMA_BLOCK_SIZE;
++ }
++
++ byte_offset += VMA_BLOCK_SIZE;
++ bit = bit << 1;
++ }
++ } else {
++ DPRINTF("VMA WRITE %zd ZERO CLUSTER\n", cluster_num);
++ vmaw->stream_info[dev_id].zero_bytes += VMA_CLUSTER_SIZE;
++ *zero_bytes += VMA_CLUSTER_SIZE;
++ }
++
++ uint64_t block_info = ((uint64_t)mask) << (32+16);
++ block_info |= ((uint64_t)dev_id) << 32;
++ block_info |= (cluster_num & 0xffffffff);
++ vmaw->outbuf_block_info[vmaw->outbuf_count] = block_info;
++
++ DPRINTF("VMA WRITE MASK %zd %zx\n", cluster_num, block_info);
++
++ vmaw->outbuf_count++;
++
++ /** NOTE: We allways write whole clusters, but we correctly set
++ * transferred bytes. So transferred == size when when everything
++ * went OK.
++ */
++ size_t transferred = VMA_CLUSTER_SIZE;
++
++ if (dev_id != vmaw->vmstate_stream) {
++ uint64_t last = (cluster_num + 1) * VMA_CLUSTER_SIZE;
++ if (last > dev_size) {
++ uint64_t diff = last - dev_size;
++ if (diff >= VMA_CLUSTER_SIZE) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "read after last cluster");
++ return -1;
++ }
++ transferred -= diff;
++ }
++ }
++
++ vmaw->stream_info[dev_id].transferred += transferred;
++
++ return transferred;
++}
++
++void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp)
++{
++ if (vmaw->status < 0 && *errp == NULL) {
++ error_setg(errp, "%s", vmaw->errmsg);
++ }
++}
++
++int vma_writer_close(VmaWriter *vmaw, Error **errp)
++{
++ g_assert(vmaw != NULL);
++
++ int i;
++
++ qemu_co_mutex_lock(&vmaw->flush_lock); // wait for pending writes
++
++ assert(vmaw->co_writer == NULL);
++
++ if (vmaw->cmd) {
++ if (pclose(vmaw->cmd) < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "pclose failed - %s", g_strerror(errno));
++ }
++ } else {
++ if (close(vmaw->fd) < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "close failed - %s", g_strerror(errno));
++ }
++ }
++
++ for (i = 0; i <= 255; i++) {
++ VmaStreamInfo *si = &vmaw->stream_info[i];
++ if (si->size) {
++ if (!si->finished) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "detected open stream '%s'", si->devname);
++ } else if ((si->transferred != si->size) &&
++ (i != vmaw->vmstate_stream)) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "incomplete stream '%s' (%zd != %zd)",
++ si->devname, si->transferred, si->size);
++ }
++ }
++ }
++
++ for (i = 0; i <= 255; i++) {
++ vmaw->stream_info[i].finished = 1; /* mark as closed */
++ }
++
++ vmaw->closed = 1;
++
++ if (vmaw->status < 0 && *errp == NULL) {
++ error_setg(errp, "%s", vmaw->errmsg);
++ }
++
++ qemu_co_mutex_unlock(&vmaw->flush_lock);
++
++ return vmaw->status;
++}
++
++void vma_writer_destroy(VmaWriter *vmaw)
++{
++ assert(vmaw);
++
++ int i;
++
++ for (i = 0; i <= 255; i++) {
++ if (vmaw->stream_info[i].devname) {
++ g_free(vmaw->stream_info[i].devname);
++ }
++ }
++
++ if (vmaw->md5csum) {
++ g_checksum_free(vmaw->md5csum);
++ }
++
++ qemu_vfree(vmaw->headerbuf);
++ qemu_vfree(vmaw->outbuf);
++ g_free(vmaw);
++}
+diff --git a/vma.c b/vma.c
+new file mode 100644
+index 0000000000..e8dffb43e0
+--- /dev/null
++++ b/vma.c
+@@ -0,0 +1,849 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) 2012-2013 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++
++#include "vma.h"
++#include "qemu/module.h"
++#include "qemu/error-report.h"
++#include "qemu/main-loop.h"
++#include "qemu/cutils.h"
++#include "qemu/memalign.h"
++#include "qapi/qmp/qdict.h"
++#include "sysemu/block-backend.h"
++
++static void help(void)
++{
++ const char *help_msg =
++ "usage: vma command [command options]\n"
++ "\n"
++ "vma list <filename>\n"
++ "vma config <filename> [-c config]\n"
++ "vma create <filename> [-c config] pathname ...\n"
++ "vma extract <filename> [-r <fifo>] <targetdir>\n"
++ "vma verify <filename> [-v]\n"
++ ;
++
++ printf("%s", help_msg);
++ exit(1);
++}
++
++static const char *extract_devname(const char *path, char **devname, int index)
++{
++ assert(path);
++
++ const char *sep = strchr(path, '=');
++
++ if (sep) {
++ *devname = g_strndup(path, sep - path);
++ path = sep + 1;
++ } else {
++ if (index >= 0) {
++ *devname = g_strdup_printf("disk%d", index);
++ } else {
++ *devname = NULL;
++ }
++ }
++
++ return path;
++}
++
++static void print_content(VmaReader *vmar)
++{
++ assert(vmar);
++
++ VmaHeader *head = vma_reader_get_header(vmar);
++
++ GList *l = vma_reader_get_config_data(vmar);
++ while (l && l->data) {
++ VmaConfigData *cdata = (VmaConfigData *)l->data;
++ l = g_list_next(l);
++ printf("CFG: size: %d name: %s\n", cdata->len, cdata->name);
++ }
++
++ int i;
++ VmaDeviceInfo *di;
++ for (i = 1; i < 255; i++) {
++ di = vma_reader_get_device_info(vmar, i);
++ if (di) {
++ if (strcmp(di->devname, "vmstate") == 0) {
++ printf("VMSTATE: dev_id=%d memory: %zd\n", i, di->size);
++ } else {
++ printf("DEV: dev_id=%d size: %zd devname: %s\n",
++ i, di->size, di->devname);
++ }
++ }
++ }
++ /* ctime is the last entry we print */
++ printf("CTIME: %s", ctime(&head->ctime));
++ fflush(stdout);
++}
++
++static int list_content(int argc, char **argv)
++{
++ int c, ret = 0;
++ const char *filename;
++
++ for (;;) {
++ c = getopt(argc, argv, "h");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ default:
++ g_assert_not_reached();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 1) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ print_content(vmar);
++
++ vma_reader_destroy(vmar);
++
++ return ret;
++}
++
++typedef struct RestoreMap {
++ char *devname;
++ char *path;
++ char *format;
++ uint64_t throttling_bps;
++ char *throttling_group;
++ char *cache;
++ bool write_zero;
++} RestoreMap;
++
++static bool try_parse_option(char **line, const char *optname, char **out, const char *inbuf) {
++ size_t optlen = strlen(optname);
++ if (strncmp(*line, optname, optlen) != 0 || (*line)[optlen] != '=') {
++ return false;
++ }
++ if (*out) {
++ g_error("read map failed - duplicate value for option '%s'", optname);
++ }
++ char *value = (*line) + optlen + 1; /* including a '=' */
++ char *colon = strchr(value, ':');
++ if (!colon) {
++ g_error("read map failed - option '%s' not terminated ('%s')",
++ optname, inbuf);
++ }
++ *line = colon+1;
++ *out = g_strndup(value, colon - value);
++ return true;
++}
++
++static uint64_t verify_u64(const char *text) {
++ uint64_t value;
++ const char *endptr = NULL;
++ if (qemu_strtou64(text, &endptr, 0, &value) != 0 || !endptr || *endptr) {
++ g_error("read map failed - not a number: %s", text);
++ }
++ return value;
++}
++
++static int extract_content(int argc, char **argv)
++{
++ int c, ret = 0;
++ int verbose = 0;
++ const char *filename;
++ const char *dirname;
++ const char *readmap = NULL;
++
++ for (;;) {
++ c = getopt(argc, argv, "hvr:");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'r':
++ readmap = optarg;
++ break;
++ case 'v':
++ verbose = 1;
++ break;
++ default:
++ help();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 2) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++ dirname = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ if (mkdir(dirname, 0777) < 0) {
++ g_error("unable to create target directory %s - %s",
++ dirname, g_strerror(errno));
++ }
++
++ GList *l = vma_reader_get_config_data(vmar);
++ while (l && l->data) {
++ VmaConfigData *cdata = (VmaConfigData *)l->data;
++ l = g_list_next(l);
++ char *cfgfn = g_strdup_printf("%s/%s", dirname, cdata->name);
++ GError *err = NULL;
++ if (!g_file_set_contents(cfgfn, (gchar *)cdata->data, cdata->len,
++ &err)) {
++ g_error("unable to write file: %s", err->message);
++ }
++ }
++
++ GHashTable *devmap = g_hash_table_new(g_str_hash, g_str_equal);
++
++ if (readmap) {
++ print_content(vmar);
++
++ FILE *map = fopen(readmap, "r");
++ if (!map) {
++ g_error("unable to open fifo %s - %s", readmap, g_strerror(errno));
++ }
++
++ while (1) {
++ char inbuf[8192];
++ char *line = fgets(inbuf, sizeof(inbuf), map);
++ char *format = NULL;
++ char *bps = NULL;
++ char *group = NULL;
++ char *cache = NULL;
++ if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
++ break;
++ }
++ int len = strlen(line);
++ if (line[len - 1] == '\n') {
++ line[len - 1] = '\0';
++ if (len == 1) {
++ break;
++ }
++ }
++
++ while (1) {
++ if (!try_parse_option(&line, "format", &format, inbuf) &&
++ !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
++ !try_parse_option(&line, "throttling.group", &group, inbuf) &&
++ !try_parse_option(&line, "cache", &cache, inbuf))
++ {
++ break;
++ }
++ }
++
++ uint64_t bps_value = 0;
++ if (bps) {
++ bps_value = verify_u64(bps);
++ g_free(bps);
++ }
++
++ const char *path;
++ bool write_zero;
++ if (line[0] == '0' && line[1] == ':') {
++ path = line + 2;
++ write_zero = false;
++ } else if (line[0] == '1' && line[1] == ':') {
++ path = line + 2;
++ write_zero = true;
++ } else {
++ g_error("read map failed - parse error ('%s')", inbuf);
++ }
++
++ char *devname = NULL;
++ path = extract_devname(path, &devname, -1);
++ if (!devname) {
++ g_error("read map failed - no dev name specified ('%s')",
++ inbuf);
++ }
++
++ RestoreMap *map = g_new0(RestoreMap, 1);
++ map->devname = g_strdup(devname);
++ map->path = g_strdup(path);
++ map->format = format;
++ map->throttling_bps = bps_value;
++ map->throttling_group = group;
++ map->cache = cache;
++ map->write_zero = write_zero;
++
++ g_hash_table_insert(devmap, map->devname, map);
++
++ };
++ }
++
++ int i;
++ int vmstate_fd = -1;
++ guint8 vmstate_stream = 0;
++
++ for (i = 1; i < 255; i++) {
++ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
++ if (di && (strcmp(di->devname, "vmstate") == 0)) {
++ vmstate_stream = i;
++ char *statefn = g_strdup_printf("%s/vmstate.bin", dirname);
++ vmstate_fd = open(statefn, O_WRONLY|O_CREAT|O_EXCL, 0644);
++ if (vmstate_fd < 0) {
++ g_error("create vmstate file '%s' failed - %s", statefn,
++ g_strerror(errno));
++ }
++ g_free(statefn);
++ } else if (di) {
++ char *devfn = NULL;
++ const char *format = NULL;
++ uint64_t throttling_bps = 0;
++ const char *throttling_group = NULL;
++ const char *cache = NULL;
++ int flags = BDRV_O_RDWR;
++ bool write_zero = true;
++
++ BlockBackend *blk = NULL;
++
++ if (readmap) {
++ RestoreMap *map;
++ map = (RestoreMap *)g_hash_table_lookup(devmap, di->devname);
++ if (map == NULL) {
++ g_error("no device name mapping for %s", di->devname);
++ }
++ devfn = map->path;
++ format = map->format;
++ throttling_bps = map->throttling_bps;
++ throttling_group = map->throttling_group;
++ cache = map->cache;
++ write_zero = map->write_zero;
++ } else {
++ devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
++ dirname, di->devname);
++ printf("DEVINFO %s %zd\n", devfn, di->size);
++
++ bdrv_img_create(devfn, "raw", NULL, NULL, NULL, di->size,
++ flags, true, &errp);
++ if (errp) {
++ g_error("can't create file %s: %s", devfn,
++ error_get_pretty(errp));
++ }
++
++ /* Note: we created an empty file above, so there is no
++ * need to write zeroes (so we generate a sparse file)
++ */
++ write_zero = false;
++ }
++
++ size_t devlen = strlen(devfn);
++ QDict *options = NULL;
++ bool writethrough;
++ if (format) {
++ /* explicit format from commandline */
++ options = qdict_new();
++ qdict_put_str(options, "driver", format);
++ } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
++ strncmp(devfn, "/dev/", 5) == 0)
++ {
++ /* This part is now deprecated for PVE as well (just as qemu
++ * deprecated not specifying an explicit raw format, too.
++ */
++ /* explicit raw format */
++ options = qdict_new();
++ qdict_put_str(options, "driver", "raw");
++ }
++ if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
++ g_error("invalid cache option: %s\n", cache);
++ }
++
++ if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
++ g_error("can't open file %s - %s", devfn,
++ error_get_pretty(errp));
++ }
++
++ if (cache) {
++ blk_set_enable_write_cache(blk, !writethrough);
++ }
++
++ if (throttling_group) {
++ blk_io_limits_enable(blk, throttling_group);
++ }
++
++ if (throttling_bps) {
++ if (!throttling_group) {
++ blk_io_limits_enable(blk, devfn);
++ }
++
++ ThrottleConfig cfg;
++ throttle_config_init(&cfg);
++ cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
++ Error *err = NULL;
++ if (!throttle_is_valid(&cfg, &err)) {
++ error_report_err(err);
++ g_error("failed to apply throttling");
++ }
++ blk_set_io_limits(blk, &cfg);
++ }
++
++ if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ if (!readmap) {
++ g_free(devfn);
++ }
++ }
++ }
++
++ if (vma_reader_restore(vmar, vmstate_fd, verbose, &errp) < 0) {
++ g_error("restore failed - %s", error_get_pretty(errp));
++ }
++
++ if (!readmap) {
++ for (i = 1; i < 255; i++) {
++ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
++ if (di && (i != vmstate_stream)) {
++ char *tmpfn = g_strdup_printf("%s/tmp-disk-%s.raw",
++ dirname, di->devname);
++ char *fn = g_strdup_printf("%s/disk-%s.raw",
++ dirname, di->devname);
++ if (rename(tmpfn, fn) != 0) {
++ g_error("rename %s to %s failed - %s",
++ tmpfn, fn, g_strerror(errno));
++ }
++ }
++ }
++ }
++
++ vma_reader_destroy(vmar);
++
++ bdrv_close_all();
++
++ return ret;
++}
++
++static int verify_content(int argc, char **argv)
++{
++ int c, ret = 0;
++ int verbose = 0;
++ const char *filename;
++
++ for (;;) {
++ c = getopt(argc, argv, "hv");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'v':
++ verbose = 1;
++ break;
++ default:
++ help();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 1) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ if (verbose) {
++ print_content(vmar);
++ }
++
++ if (vma_reader_verify(vmar, verbose, &errp) < 0) {
++ g_error("verify failed - %s", error_get_pretty(errp));
++ }
++
++ vma_reader_destroy(vmar);
++
++ bdrv_close_all();
++
++ return ret;
++}
++
++typedef struct BackupJob {
++ BlockBackend *target;
++ int64_t len;
++ VmaWriter *vmaw;
++ uint8_t dev_id;
++} BackupJob;
++
++#define BACKUP_SECTORS_PER_CLUSTER (VMA_CLUSTER_SIZE / BDRV_SECTOR_SIZE)
++
++static void coroutine_fn backup_run_empty(void *opaque)
++{
++ VmaWriter *vmaw = (VmaWriter *)opaque;
++
++ vma_writer_flush_output(vmaw);
++
++ Error *err = NULL;
++ if (vma_writer_close(vmaw, &err) != 0) {
++ g_warning("vma_writer_close failed %s", error_get_pretty(err));
++ }
++}
++
++static void coroutine_fn backup_run(void *opaque)
++{
++ BackupJob *job = (BackupJob *)opaque;
++ struct iovec iov;
++ QEMUIOVector qiov;
++
++ int64_t start, end;
++ int ret = 0;
++
++ unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
++
++ start = 0;
++ end = DIV_ROUND_UP(job->len / BDRV_SECTOR_SIZE,
++ BACKUP_SECTORS_PER_CLUSTER);
++
++ for (; start < end; start++) {
++ iov.iov_base = buf;
++ iov.iov_len = VMA_CLUSTER_SIZE;
++ qemu_iovec_init_external(&qiov, &iov, 1);
++
++ ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
++ VMA_CLUSTER_SIZE, &qiov, 0);
++ if (ret < 0) {
++ vma_writer_set_error(job->vmaw, "read error", -1);
++ goto out;
++ }
++
++ size_t zb = 0;
++ if (vma_writer_write(job->vmaw, job->dev_id, start, buf, &zb) < 0) {
++ vma_writer_set_error(job->vmaw, "backup_dump_cb vma_writer_write failed", -1);
++ goto out;
++ }
++ }
++
++
++out:
++ if (vma_writer_close_stream(job->vmaw, job->dev_id) <= 0) {
++ Error *err = NULL;
++ if (vma_writer_close(job->vmaw, &err) != 0) {
++ g_warning("vma_writer_close failed %s", error_get_pretty(err));
++ }
++ }
++ qemu_vfree(buf);
++}
++
++static int create_archive(int argc, char **argv)
++{
++ int i, c;
++ int verbose = 0;
++ const char *archivename;
++ GList *backup_coroutines = NULL;
++ GList *config_files = NULL;
++
++ for (;;) {
++ c = getopt(argc, argv, "hvc:");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'c':
++ config_files = g_list_append(config_files, optarg);
++ break;
++ case 'v':
++ verbose = 1;
++ break;
++ default:
++ g_assert_not_reached();
++ }
++ }
++
++
++ /* make sure we an archive name */
++ if ((optind + 1) > argc) {
++ help();
++ }
++
++ archivename = argv[optind++];
++
++ uuid_t uuid;
++ uuid_generate(uuid);
++
++ Error *local_err = NULL;
++ VmaWriter *vmaw = vma_writer_create(archivename, uuid, &local_err);
++
++ if (vmaw == NULL) {
++ g_error("%s", error_get_pretty(local_err));
++ }
++
++ GList *l = config_files;
++ while (l && l->data) {
++ char *name = l->data;
++ char *cdata = NULL;
++ gsize clen = 0;
++ GError *err = NULL;
++ if (!g_file_get_contents(name, &cdata, &clen, &err)) {
++ unlink(archivename);
++ g_error("Unable to read file: %s", err->message);
++ }
++
++ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
++ unlink(archivename);
++ g_error("Unable to append config data %s (len = %zd)",
++ name, clen);
++ }
++ l = g_list_next(l);
++ }
++
++ int devcount = 0;
++ while (optind < argc) {
++ const char *path = argv[optind++];
++ char *devname = NULL;
++ path = extract_devname(path, &devname, devcount++);
++
++ Error *errp = NULL;
++ BlockBackend *target;
++
++ target = blk_new_open(path, NULL, NULL, 0, &errp);
++ if (!target) {
++ unlink(archivename);
++ g_error("bdrv_open '%s' failed - %s", path, error_get_pretty(errp));
++ }
++ int64_t size = blk_getlength(target);
++ int dev_id = vma_writer_register_stream(vmaw, devname, size);
++ if (dev_id <= 0) {
++ unlink(archivename);
++ g_error("vma_writer_register_stream '%s' failed", devname);
++ }
++
++ BackupJob *job = g_new0(BackupJob, 1);
++ job->len = size;
++ job->target = target;
++ job->vmaw = vmaw;
++ job->dev_id = dev_id;
++
++ Coroutine *co = qemu_coroutine_create(backup_run, job);
++ // Don't enter coroutine yet, because it might write the header before
++ // all streams can be registered.
++ backup_coroutines = g_list_append(backup_coroutines, co);
++ }
++
++ VmaStatus vmastat;
++ int percent = 0;
++ int last_percent = -1;
++
++ if (devcount) {
++ GList *entry = backup_coroutines;
++ while (entry && entry->data) {
++ Coroutine *co = entry->data;
++ qemu_coroutine_enter(co);
++ entry = g_list_next(entry);
++ }
++
++ while (1) {
++ main_loop_wait(false);
++ vma_writer_get_status(vmaw, &vmastat);
++
++ if (verbose) {
++
++ uint64_t total = 0;
++ uint64_t transferred = 0;
++ uint64_t zero_bytes = 0;
++
++ int i;
++ for (i = 0; i < 256; i++) {
++ if (vmastat.stream_info[i].size) {
++ total += vmastat.stream_info[i].size;
++ transferred += vmastat.stream_info[i].transferred;
++ zero_bytes += vmastat.stream_info[i].zero_bytes;
++ }
++ }
++ percent = (transferred*100)/total;
++ if (percent != last_percent) {
++ fprintf(stderr, "progress %d%% %zd/%zd %zd\n", percent,
++ transferred, total, zero_bytes);
++ fflush(stderr);
++
++ last_percent = percent;
++ }
++ }
++
++ if (vmastat.closed) {
++ break;
++ }
++ }
++ } else {
++ Coroutine *co = qemu_coroutine_create(backup_run_empty, vmaw);
++ qemu_coroutine_enter(co);
++ while (1) {
++ main_loop_wait(false);
++ vma_writer_get_status(vmaw, &vmastat);
++ if (vmastat.closed) {
++ break;
++ }
++ }
++ }
++
++ bdrv_drain_all();
++
++ vma_writer_get_status(vmaw, &vmastat);
++
++ if (verbose) {
++ for (i = 0; i < 256; i++) {
++ VmaStreamInfo *si = &vmastat.stream_info[i];
++ if (si->size) {
++ fprintf(stderr, "image %s: size=%zd zeros=%zd saved=%zd\n",
++ si->devname, si->size, si->zero_bytes,
++ si->size - si->zero_bytes);
++ }
++ }
++ }
++
++ if (vmastat.status < 0) {
++ unlink(archivename);
++ g_error("creating vma archive failed");
++ }
++
++ g_list_free(backup_coroutines);
++ g_list_free(config_files);
++ vma_writer_destroy(vmaw);
++ return 0;
++}
++
++static int dump_config(int argc, char **argv)
++{
++ int c, ret = 0;
++ const char *filename;
++ const char *config_name = "qemu-server.conf";
++
++ for (;;) {
++ c = getopt(argc, argv, "hc:");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'c':
++ config_name = optarg;
++ break;
++ default:
++ help();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 1) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ int found = 0;
++ GList *l = vma_reader_get_config_data(vmar);
++ while (l && l->data) {
++ VmaConfigData *cdata = (VmaConfigData *)l->data;
++ l = g_list_next(l);
++ if (strcmp(cdata->name, config_name) == 0) {
++ found = 1;
++ fwrite(cdata->data, cdata->len, 1, stdout);
++ break;
++ }
++ }
++
++ vma_reader_destroy(vmar);
++
++ bdrv_close_all();
++
++ if (!found) {
++ fprintf(stderr, "unable to find configuration data '%s'\n", config_name);
++ return -1;
++ }
++
++ return ret;
++}
++
++int main(int argc, char **argv)
++{
++ const char *cmdname;
++ Error *main_loop_err = NULL;
++
++ error_init(argv[0]);
++ module_call_init(MODULE_INIT_TRACE);
++ qemu_init_exec_dir(argv[0]);
++
++ if (qemu_init_main_loop(&main_loop_err)) {
++ g_error("%s", error_get_pretty(main_loop_err));
++ }
++
++ bdrv_init();
++ module_call_init(MODULE_INIT_QOM);
++
++ if (argc < 2) {
++ help();
++ }
++
++ cmdname = argv[1];
++ argc--; argv++;
++
++
++ if (!strcmp(cmdname, "list")) {
++ return list_content(argc, argv);
++ } else if (!strcmp(cmdname, "create")) {
++ return create_archive(argc, argv);
++ } else if (!strcmp(cmdname, "extract")) {
++ return extract_content(argc, argv);
++ } else if (!strcmp(cmdname, "verify")) {
++ return verify_content(argc, argv);
++ } else if (!strcmp(cmdname, "config")) {
++ return dump_config(argc, argv);
++ }
++
++ help();
++ return 0;
++}
+diff --git a/vma.h b/vma.h
+new file mode 100644
+index 0000000000..c895c97f6d
+--- /dev/null
++++ b/vma.h
+@@ -0,0 +1,150 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#ifndef BACKUP_VMA_H
++#define BACKUP_VMA_H
++
++#include <uuid/uuid.h>
++#include "qapi/error.h"
++#include "block/block.h"
++
++#define VMA_BLOCK_BITS 12
++#define VMA_BLOCK_SIZE (1<<VMA_BLOCK_BITS)
++#define VMA_CLUSTER_BITS (VMA_BLOCK_BITS+4)
++#define VMA_CLUSTER_SIZE (1<<VMA_CLUSTER_BITS)
++
++#if VMA_CLUSTER_SIZE != 65536
++#error unexpected cluster size
++#endif
++
++#define VMA_EXTENT_HEADER_SIZE 512
++#define VMA_BLOCKS_PER_EXTENT 59
++#define VMA_MAX_CONFIGS 256
++
++#define VMA_MAX_EXTENT_SIZE \
++ (VMA_EXTENT_HEADER_SIZE+VMA_CLUSTER_SIZE*VMA_BLOCKS_PER_EXTENT)
++#if VMA_MAX_EXTENT_SIZE != 3867136
++#error unexpected VMA_EXTENT_SIZE
++#endif
++
++/* File Format Definitions */
++
++#define VMA_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|0x00))
++#define VMA_EXTENT_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|'E'))
++
++typedef struct VmaDeviceInfoHeader {
++ uint32_t devname_ptr; /* offset into blob_buffer table */
++ uint32_t reserved0;
++ uint64_t size; /* device size in bytes */
++ uint64_t reserved1;
++ uint64_t reserved2;
++} VmaDeviceInfoHeader;
++
++typedef struct VmaHeader {
++ uint32_t magic;
++ uint32_t version;
++ unsigned char uuid[16];
++ int64_t ctime;
++ unsigned char md5sum[16];
++
++ uint32_t blob_buffer_offset;
++ uint32_t blob_buffer_size;
++ uint32_t header_size;
++
++ unsigned char reserved[1984];
++
++ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++
++ uint32_t reserved1;
++
++ VmaDeviceInfoHeader dev_info[256];
++} VmaHeader;
++
++typedef struct VmaExtentHeader {
++ uint32_t magic;
++ uint16_t reserved1;
++ uint16_t block_count;
++ unsigned char uuid[16];
++ unsigned char md5sum[16];
++ uint64_t blockinfo[VMA_BLOCKS_PER_EXTENT];
++} VmaExtentHeader;
++
++/* functions/definitions to read/write vma files */
++
++typedef struct VmaReader VmaReader;
++
++typedef struct VmaWriter VmaWriter;
++
++typedef struct VmaConfigData {
++ const char *name;
++ const void *data;
++ uint32_t len;
++} VmaConfigData;
++
++typedef struct VmaStreamInfo {
++ uint64_t size;
++ uint64_t cluster_count;
++ uint64_t transferred;
++ uint64_t zero_bytes;
++ int finished;
++ char *devname;
++} VmaStreamInfo;
++
++typedef struct VmaStatus {
++ int status;
++ bool closed;
++ char errmsg[8192];
++ char uuid_str[37];
++ VmaStreamInfo stream_info[256];
++} VmaStatus;
++
++typedef struct VmaDeviceInfo {
++ uint64_t size; /* device size in bytes */
++ const char *devname;
++} VmaDeviceInfo;
++
++VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp);
++int vma_writer_close(VmaWriter *vmaw, Error **errp);
++void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp);
++void vma_writer_destroy(VmaWriter *vmaw);
++int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
++ size_t len);
++int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
++ size_t size);
++
++int64_t coroutine_fn vma_writer_write(VmaWriter *vmaw, uint8_t dev_id,
++ int64_t cluster_num,
++ const unsigned char *buf,
++ size_t *zero_bytes);
++
++int coroutine_fn vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id);
++int coroutine_fn vma_writer_flush_output(VmaWriter *vmaw);
++
++int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status);
++void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...);
++
++
++VmaReader *vma_reader_create(const char *filename, Error **errp);
++void vma_reader_destroy(VmaReader *vmar);
++VmaHeader *vma_reader_get_header(VmaReader *vmar);
++GList *vma_reader_get_config_data(VmaReader *vmar);
++VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id);
++int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id,
++ BlockBackend *target, bool write_zeroes,
++ Error **errp);
++int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
++ Error **errp);
++int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp);
++
++#endif /* BACKUP_VMA_H */
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:58 +0200
+Subject: [PATCH] PVE-Backup: add backup-dump block driver
+
+- add backup-dump block driver block/backup-dump.c
+- move BackupBlockJob declaration from block/backup.c to include/block/block_int.h
+- block/backup.c - backup-job-create: also consider source cluster size
+- job.c: make job_should_pause non-static
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/backup-dump.c | 167 +++++++++++++++++++++++++++++++
+ block/backup.c | 30 ++----
+ block/meson.build | 1 +
+ include/block/block_int-common.h | 35 +++++++
+ job.c | 3 +-
+ 5 files changed, 213 insertions(+), 23 deletions(-)
+ create mode 100644 block/backup-dump.c
+
+diff --git a/block/backup-dump.c b/block/backup-dump.c
+new file mode 100644
+index 0000000000..04718a94e2
+--- /dev/null
++++ b/block/backup-dump.c
+@@ -0,0 +1,167 @@
++/*
++ * BlockDriver to send backup data stream to a callback function
++ *
++ * Copyright (C) 2020 Proxmox Server Solutions GmbH
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include "qom/object_interfaces.h"
++#include "block/block_int.h"
++
++typedef struct {
++ int dump_cb_block_size;
++ uint64_t byte_size;
++ BackupDumpFunc *dump_cb;
++ void *dump_cb_data;
++} BDRVBackupDumpState;
++
++static int qemu_backup_dump_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
++{
++ BDRVBackupDumpState *s = bs->opaque;
++
++ bdi->cluster_size = s->dump_cb_block_size;
++ return 0;
++}
++
++static int qemu_backup_dump_check_perm(
++ BlockDriverState *bs,
++ uint64_t perm,
++ uint64_t shared,
++ Error **errp)
++{
++ /* Nothing to do. */
++ return 0;
++}
++
++static void qemu_backup_dump_set_perm(
++ BlockDriverState *bs,
++ uint64_t perm,
++ uint64_t shared)
++{
++ /* Nothing to do. */
++}
++
++static void qemu_backup_dump_abort_perm_update(BlockDriverState *bs)
++{
++ /* Nothing to do. */
++}
++
++static void qemu_backup_dump_refresh_limits(BlockDriverState *bs, Error **errp)
++{
++ bs->bl.request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O */
++}
++
++static void qemu_backup_dump_close(BlockDriverState *bs)
++{
++ /* Nothing to do. */
++}
++
++static int64_t qemu_backup_dump_getlength(BlockDriverState *bs)
++{
++ BDRVBackupDumpState *s = bs->opaque;
++
++ return s->byte_size;
++}
++
++static coroutine_fn int qemu_backup_dump_co_writev(
++ BlockDriverState *bs,
++ int64_t sector_num,
++ int nb_sectors,
++ QEMUIOVector *qiov,
++ int flags)
++{
++ /* flags can be only values we set in supported_write_flags */
++ assert(flags == 0);
++
++ BDRVBackupDumpState *s = bs->opaque;
++ off_t offset = sector_num * BDRV_SECTOR_SIZE;
++
++ uint64_t written = 0;
++
++ for (int i = 0; i < qiov->niov; ++i) {
++ const struct iovec *v = &qiov->iov[i];
++
++ int rc = s->dump_cb(s->dump_cb_data, offset, v->iov_len, v->iov_base);
++ if (rc < 0) {
++ return rc;
++ }
++
++ if (rc != v->iov_len) {
++ return -EIO;
++ }
++
++ written += v->iov_len;
++ offset += v->iov_len;
++ }
++
++ return written;
++}
++
++static void qemu_backup_dump_child_perm(
++ BlockDriverState *bs,
++ BdrvChild *c,
++ BdrvChildRole role,
++ BlockReopenQueue *reopen_queue,
++ uint64_t perm, uint64_t shared,
++ uint64_t *nperm, uint64_t *nshared)
++{
++ *nperm = BLK_PERM_ALL;
++ *nshared = BLK_PERM_ALL;
++}
++
++static BlockDriver bdrv_backup_dump_drive = {
++ .format_name = "backup-dump-drive",
++ .protocol_name = "backup-dump",
++ .instance_size = sizeof(BDRVBackupDumpState),
++
++ .bdrv_close = qemu_backup_dump_close,
++ .bdrv_has_zero_init = bdrv_has_zero_init_1,
++ .bdrv_getlength = qemu_backup_dump_getlength,
++ .bdrv_get_info = qemu_backup_dump_get_info,
++
++ .bdrv_co_writev = qemu_backup_dump_co_writev,
++
++ .bdrv_refresh_limits = qemu_backup_dump_refresh_limits,
++ .bdrv_check_perm = qemu_backup_dump_check_perm,
++ .bdrv_set_perm = qemu_backup_dump_set_perm,
++ .bdrv_abort_perm_update = qemu_backup_dump_abort_perm_update,
++ .bdrv_child_perm = qemu_backup_dump_child_perm,
++};
++
++static void bdrv_backup_dump_init(void)
++{
++ bdrv_register(&bdrv_backup_dump_drive);
++}
++
++block_init(bdrv_backup_dump_init);
++
++
++BlockDriverState *bdrv_backup_dump_create(
++ int dump_cb_block_size,
++ uint64_t byte_size,
++ BackupDumpFunc *dump_cb,
++ void *dump_cb_data,
++ Error **errp)
++{
++ BDRVBackupDumpState *state;
++ BlockDriverState *bs = bdrv_new_open_driver(
++ &bdrv_backup_dump_drive, NULL, BDRV_O_RDWR, errp);
++
++ if (!bs) {
++ return NULL;
++ }
++
++ bs->total_sectors = byte_size / BDRV_SECTOR_SIZE;
++ bs->opaque = state = g_new0(BDRVBackupDumpState, 1);
++
++ state->dump_cb_block_size = dump_cb_block_size;
++ state->byte_size = byte_size;
++ state->dump_cb = dump_cb;
++ state->dump_cb_data = dump_cb_data;
++
++ return bs;
++}
+diff --git a/block/backup.c b/block/backup.c
+index b6fa9e8a69..789f8b7799 100644
+--- a/block/backup.c
++++ b/block/backup.c
+@@ -29,28 +29,6 @@
+
+ #include "block/copy-before-write.h"
+
+-typedef struct BackupBlockJob {
+- BlockJob common;
+- BlockDriverState *cbw;
+- BlockDriverState *source_bs;
+- BlockDriverState *target_bs;
+-
+- BdrvDirtyBitmap *sync_bitmap;
+-
+- MirrorSyncMode sync_mode;
+- BitmapSyncMode bitmap_mode;
+- BlockdevOnError on_source_error;
+- BlockdevOnError on_target_error;
+- uint64_t len;
+- int64_t cluster_size;
+- BackupPerf perf;
+-
+- BlockCopyState *bcs;
+-
+- bool wait;
+- BlockCopyCallState *bg_bcs_call;
+-} BackupBlockJob;
+-
+ static const BlockJobDriver backup_job_driver;
+
+ static void backup_cleanup_sync_bitmap(BackupBlockJob *job, int ret)
+@@ -454,6 +432,14 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+ }
+
+ cluster_size = block_copy_cluster_size(bcs);
++ if (cluster_size < 0) {
++ goto error;
++ }
++
++ BlockDriverInfo bdi;
++ if (bdrv_get_info(bs, &bdi) == 0) {
++ cluster_size = MAX(cluster_size, bdi.cluster_size);
++ }
+
+ if (perf->max_chunk && perf->max_chunk < cluster_size) {
+ error_setg(errp, "Required max-chunk (%" PRIi64 ") is less than backup "
+diff --git a/block/meson.build b/block/meson.build
+index 3a0b84bc11..7f22e7f177 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -4,6 +4,7 @@ block_ss.add(files(
+ 'aio_task.c',
+ 'amend.c',
+ 'backup.c',
++ 'backup-dump.c',
+ 'copy-before-write.c',
+ 'blkdebug.c',
+ 'blklogwrites.c',
+diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h
+index 8947abab76..f272d0d8dc 100644
+--- a/include/block/block_int-common.h
++++ b/include/block/block_int-common.h
+@@ -26,6 +26,7 @@
+
+ #include "block/accounting.h"
+ #include "block/block.h"
++#include "block/block-copy.h"
+ #include "block/aio-wait.h"
+ #include "qemu/queue.h"
+ #include "qemu/coroutine.h"
+@@ -64,6 +65,40 @@
+
+ #define BLOCK_PROBE_BUF_SIZE 512
+
++typedef int BackupDumpFunc(void *opaque, uint64_t offset, uint64_t bytes, const void *buf);
++
++BlockDriverState *bdrv_backup_dump_create(
++ int dump_cb_block_size,
++ uint64_t byte_size,
++ BackupDumpFunc *dump_cb,
++ void *dump_cb_data,
++ Error **errp);
++
++// Needs to be defined here, since it's used in blockdev.c to detect PVE backup
++// jobs with source_bs
++typedef struct BlockCopyState BlockCopyState;
++typedef struct BackupBlockJob {
++ BlockJob common;
++ BlockDriverState *cbw;
++ BlockDriverState *source_bs;
++ BlockDriverState *target_bs;
++
++ BdrvDirtyBitmap *sync_bitmap;
++
++ MirrorSyncMode sync_mode;
++ BitmapSyncMode bitmap_mode;
++ BlockdevOnError on_source_error;
++ BlockdevOnError on_target_error;
++ uint64_t len;
++ int64_t cluster_size;
++ BackupPerf perf;
++
++ BlockCopyState *bcs;
++
++ bool wait;
++ BlockCopyCallState *bg_bcs_call;
++} BackupBlockJob;
++
+ enum BdrvTrackedRequestType {
+ BDRV_TRACKED_READ,
+ BDRV_TRACKED_WRITE,
+diff --git a/job.c b/job.c
+index 075c6f3a20..e5699ad200 100644
+--- a/job.c
++++ b/job.c
+@@ -276,7 +276,8 @@ static bool job_started(Job *job)
+ return job->co;
+ }
+
+-static bool job_should_pause(Job *job)
++bool job_should_pause(Job *job);
++bool job_should_pause(Job *job)
+ {
+ return job->pause_count > 0;
+ }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:16:59 +0200
-Subject: [PATCH] PVE-Backup: proxmox backup patches for qemu
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
-[PVE-Backup: avoid coroutines to fix AIO freeze, cleanups]
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[add new force parameter to job_cancel_sync calls]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/meson.build | 5 +
- block/monitor/block-hmp-cmds.c | 33 ++
- blockdev.c | 1 +
- hmp-commands-info.hx | 14 +
- hmp-commands.hx | 29 +
- include/monitor/hmp.h | 3 +
- meson.build | 1 +
- monitor/hmp-cmds.c | 44 ++
- proxmox-backup-client.c | 176 ++++++
- proxmox-backup-client.h | 59 ++
- pve-backup.c | 959 +++++++++++++++++++++++++++++++++
- qapi/block-core.json | 109 ++++
- qapi/common.json | 13 +
- qapi/machine.json | 15 +-
- 14 files changed, 1448 insertions(+), 13 deletions(-)
- create mode 100644 proxmox-backup-client.c
- create mode 100644 proxmox-backup-client.h
- create mode 100644 pve-backup.c
-
-diff --git a/block/meson.build b/block/meson.build
-index 067708b7c0..42aac96dbb 100644
---- a/block/meson.build
-+++ b/block/meson.build
-@@ -48,6 +48,11 @@ block_ss.add(files(
- ), zstd, zlib, gnutls)
-
- block_ss.add(files('../vma-writer.c'), libuuid)
-+block_ss.add(files(
-+ '../proxmox-backup-client.c',
-+ '../pve-backup.c',
-+), libproxmox_backup_qemu)
-+
-
- softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index bfb3c043a0..89ca64444d 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1015,3 +1015,36 @@ void hmp_info_snapshots(Monitor *mon, const QDict *qdict)
- g_free(sn_tab);
- g_free(global_snapshots);
- }
-+
-+void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
-+{
-+ Error *error = NULL;
-+
-+ qmp_backup_cancel(&error);
-+
-+ hmp_handle_error(mon, error);
-+}
-+
-+void hmp_backup(Monitor *mon, const QDict *qdict)
-+{
-+ Error *error = NULL;
-+
-+ int dir = qdict_get_try_bool(qdict, "directory", 0);
-+ const char *backup_file = qdict_get_str(qdict, "backupfile");
-+ const char *devlist = qdict_get_try_str(qdict, "devlist");
-+ int64_t speed = qdict_get_try_int(qdict, "speed", 0);
-+
-+ qmp_backup(
-+ backup_file,
-+ false, NULL, // PBS password
-+ false, NULL, // PBS keyfile
-+ false, NULL, // PBS key_password
-+ false, NULL, // PBS fingerprint
-+ false, NULL, // PBS backup-id
-+ false, 0, // PBS backup-time
-+ true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
-+ false, NULL, false, NULL, !!devlist,
-+ devlist, qdict_haskey(qdict, "speed"), speed, &error);
-+
-+ hmp_handle_error(mon, error);
-+}
-diff --git a/blockdev.c b/blockdev.c
-index 1af5a1fcb2..9916a72334 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -36,6 +36,7 @@
- #include "hw/block/block.h"
- #include "block/blockjob.h"
- #include "block/qdict.h"
-+#include "block/blockjob_int.h"
- #include "block/throttle-groups.h"
- #include "monitor/monitor.h"
- #include "qemu/error-report.h"
-diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
-index 925e680e5a..d9b787d2c6 100644
---- a/hmp-commands-info.hx
-+++ b/hmp-commands-info.hx
-@@ -484,6 +484,20 @@ SRST
- Show the current VM UUID.
- ERST
-
-+
-+ {
-+ .name = "backup",
-+ .args_type = "",
-+ .params = "",
-+ .help = "show backup status",
-+ .cmd = hmp_info_backup,
-+ },
-+
-+SRST
-+ ``info backup``
-+ Show backup status.
-+ERST
-+
- #if defined(CONFIG_SLIRP)
- {
- .name = "usernet",
-diff --git a/hmp-commands.hx b/hmp-commands.hx
-index 7f0ac498c4..994445f301 100644
---- a/hmp-commands.hx
-+++ b/hmp-commands.hx
-@@ -99,6 +99,35 @@ ERST
- SRST
- ``block_stream``
- Copy data from a backing file into a block device.
-+ERST
-+
-+ {
-+ .name = "backup",
-+ .args_type = "directory:-d,backupfile:s,speed:o?,devlist:s?",
-+ .params = "[-d] backupfile [speed [devlist]]",
-+ .help = "create a VM Backup."
-+ "\n\t\t\t Use -d to dump data into a directory instead"
-+ "\n\t\t\t of using VMA format.",
-+ .cmd = hmp_backup,
-+ },
-+
-+SRST
-+``backup``
-+ Create a VM backup.
-+ERST
-+
-+ {
-+ .name = "backup_cancel",
-+ .args_type = "",
-+ .params = "",
-+ .help = "cancel the current VM backup",
-+ .cmd = hmp_backup_cancel,
-+ },
-+
-+SRST
-+``backup_cancel``
-+ Cancel the current VM backup.
-+
- ERST
-
- {
-diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
-index 3a39ba41b5..d269b4c99c 100644
---- a/include/monitor/hmp.h
-+++ b/include/monitor/hmp.h
-@@ -30,6 +30,7 @@ void hmp_info_savevm(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_capabilities(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict);
-+void hmp_info_backup(Monitor *mon, const QDict *qdict);
- void hmp_info_cpus(Monitor *mon, const QDict *qdict);
- void hmp_info_vnc(Monitor *mon, const QDict *qdict);
- void hmp_info_spice(Monitor *mon, const QDict *qdict);
-@@ -73,6 +74,8 @@ void hmp_x_colo_lost_heartbeat(Monitor *mon, const QDict *qdict);
- void hmp_set_password(Monitor *mon, const QDict *qdict);
- void hmp_expire_password(Monitor *mon, const QDict *qdict);
- void hmp_change(Monitor *mon, const QDict *qdict);
-+void hmp_backup(Monitor *mon, const QDict *qdict);
-+void hmp_backup_cancel(Monitor *mon, const QDict *qdict);
- void hmp_migrate(Monitor *mon, const QDict *qdict);
- void hmp_device_add(Monitor *mon, const QDict *qdict);
- void hmp_device_del(Monitor *mon, const QDict *qdict);
-diff --git a/meson.build b/meson.build
-index 96a91b95e4..76fcb23299 100644
---- a/meson.build
-+++ b/meson.build
-@@ -1268,6 +1268,7 @@ keyutils = dependency('libkeyutils', required: false,
- has_gettid = cc.has_function('gettid')
-
- libuuid = cc.find_library('uuid', required: true)
-+libproxmox_backup_qemu = cc.find_library('proxmox_backup_qemu', required: true)
-
- # libselinux
- selinux = dependency('libselinux',
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index c64b959738..ef654221c4 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -195,6 +195,50 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
- qapi_free_MouseInfoList(mice_list);
- }
-
-+void hmp_info_backup(Monitor *mon, const QDict *qdict)
-+{
-+ BackupStatus *info;
-+
-+ info = qmp_query_backup(NULL);
-+
-+ if (!info) {
-+ monitor_printf(mon, "Backup status: not initialized\n");
-+ return;
-+ }
-+
-+ if (info->has_status) {
-+ if (info->has_errmsg) {
-+ monitor_printf(mon, "Backup status: %s - %s\n",
-+ info->status, info->errmsg);
-+ } else {
-+ monitor_printf(mon, "Backup status: %s\n", info->status);
-+ }
-+ }
-+
-+ if (info->has_backup_file) {
-+ monitor_printf(mon, "Start time: %s", ctime(&info->start_time));
-+ if (info->end_time) {
-+ monitor_printf(mon, "End time: %s", ctime(&info->end_time));
-+ }
-+
-+ int per = (info->has_total && info->total &&
-+ info->has_transferred && info->transferred) ?
-+ (info->transferred * 100)/info->total : 0;
-+ int zero_per = (info->has_total && info->total &&
-+ info->has_zero_bytes && info->zero_bytes) ?
-+ (info->zero_bytes * 100)/info->total : 0;
-+ monitor_printf(mon, "Backup file: %s\n", info->backup_file);
-+ monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
-+ monitor_printf(mon, "Total size: %zd\n", info->total);
-+ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-+ info->transferred, per);
-+ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-+ info->zero_bytes, zero_per);
-+ }
-+
-+ qapi_free_BackupStatus(info);
-+}
-+
- static char *SocketAddress_to_str(SocketAddress *addr)
- {
- switch (addr->type) {
-diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
-new file mode 100644
-index 0000000000..a8f6653a81
---- /dev/null
-+++ b/proxmox-backup-client.c
-@@ -0,0 +1,176 @@
-+#include "proxmox-backup-client.h"
-+#include "qemu/main-loop.h"
-+#include "block/aio-wait.h"
-+#include "qapi/error.h"
-+
-+/* Proxmox Backup Server client bindings using coroutines */
-+
-+typedef struct BlockOnCoroutineWrapper {
-+ AioContext *ctx;
-+ CoroutineEntry *entry;
-+ void *entry_arg;
-+ bool finished;
-+} BlockOnCoroutineWrapper;
-+
-+static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
-+{
-+ BlockOnCoroutineWrapper *wrapper = opaque;
-+ wrapper->entry(wrapper->entry_arg);
-+ wrapper->finished = true;
-+ aio_wait_kick();
-+}
-+
-+void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
-+{
-+ assert(!qemu_in_coroutine());
-+
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ BlockOnCoroutineWrapper wrapper = {
-+ .finished = false,
-+ .entry = entry,
-+ .entry_arg = entry_arg,
-+ .ctx = ctx,
-+ };
-+ Coroutine *wrapper_co = qemu_coroutine_create(block_on_coroutine_wrapper, &wrapper);
-+ aio_co_enter(ctx, wrapper_co);
-+ AIO_WAIT_WHILE(ctx, !wrapper.finished);
-+}
-+
-+// This is called from another thread, so we use aio_co_schedule()
-+static void proxmox_backup_schedule_wake(void *data) {
-+ CoCtxData *waker = (CoCtxData *)data;
-+ aio_co_schedule(waker->ctx, waker->co);
-+}
-+
-+int coroutine_fn
-+proxmox_backup_co_connect(ProxmoxBackupHandle *pbs, Error **errp)
-+{
-+ Coroutine *co = qemu_coroutine_self();
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ CoCtxData waker = { .co = co, .ctx = ctx };
-+ char *pbs_err = NULL;
-+ int pbs_res = -1;
-+
-+ proxmox_backup_connect_async(pbs, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ qemu_coroutine_yield();
-+ if (pbs_res < 0) {
-+ if (errp) error_setg(errp, "backup connect failed: %s", pbs_err ? pbs_err : "unknown error");
-+ if (pbs_err) proxmox_backup_free_error(pbs_err);
-+ }
-+ return pbs_res;
-+}
-+
-+int coroutine_fn
-+proxmox_backup_co_add_config(
-+ ProxmoxBackupHandle *pbs,
-+ const char *name,
-+ const uint8_t *data,
-+ uint64_t size,
-+ Error **errp)
-+{
-+ Coroutine *co = qemu_coroutine_self();
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ CoCtxData waker = { .co = co, .ctx = ctx };
-+ char *pbs_err = NULL;
-+ int pbs_res = -1;
-+
-+ proxmox_backup_add_config_async(
-+ pbs, name, data, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ qemu_coroutine_yield();
-+ if (pbs_res < 0) {
-+ if (errp) error_setg(errp, "backup add_config %s failed: %s", name, pbs_err ? pbs_err : "unknown error");
-+ if (pbs_err) proxmox_backup_free_error(pbs_err);
-+ }
-+ return pbs_res;
-+}
-+
-+int coroutine_fn
-+proxmox_backup_co_register_image(
-+ ProxmoxBackupHandle *pbs,
-+ const char *device_name,
-+ uint64_t size,
-+ Error **errp)
-+{
-+ Coroutine *co = qemu_coroutine_self();
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ CoCtxData waker = { .co = co, .ctx = ctx };
-+ char *pbs_err = NULL;
-+ int pbs_res = -1;
-+
-+ proxmox_backup_register_image_async(
-+ pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ qemu_coroutine_yield();
-+ if (pbs_res < 0) {
-+ if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
-+ if (pbs_err) proxmox_backup_free_error(pbs_err);
-+ }
-+ return pbs_res;
-+}
-+
-+int coroutine_fn
-+proxmox_backup_co_finish(
-+ ProxmoxBackupHandle *pbs,
-+ Error **errp)
-+{
-+ Coroutine *co = qemu_coroutine_self();
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ CoCtxData waker = { .co = co, .ctx = ctx };
-+ char *pbs_err = NULL;
-+ int pbs_res = -1;
-+
-+ proxmox_backup_finish_async(
-+ pbs, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ qemu_coroutine_yield();
-+ if (pbs_res < 0) {
-+ if (errp) error_setg(errp, "backup finish failed: %s", pbs_err ? pbs_err : "unknown error");
-+ if (pbs_err) proxmox_backup_free_error(pbs_err);
-+ }
-+ return pbs_res;
-+}
-+
-+int coroutine_fn
-+proxmox_backup_co_close_image(
-+ ProxmoxBackupHandle *pbs,
-+ uint8_t dev_id,
-+ Error **errp)
-+{
-+ Coroutine *co = qemu_coroutine_self();
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ CoCtxData waker = { .co = co, .ctx = ctx };
-+ char *pbs_err = NULL;
-+ int pbs_res = -1;
-+
-+ proxmox_backup_close_image_async(
-+ pbs, dev_id, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ qemu_coroutine_yield();
-+ if (pbs_res < 0) {
-+ if (errp) error_setg(errp, "backup close image failed: %s", pbs_err ? pbs_err : "unknown error");
-+ if (pbs_err) proxmox_backup_free_error(pbs_err);
-+ }
-+ return pbs_res;
-+}
-+
-+int coroutine_fn
-+proxmox_backup_co_write_data(
-+ ProxmoxBackupHandle *pbs,
-+ uint8_t dev_id,
-+ const uint8_t *data,
-+ uint64_t offset,
-+ uint64_t size,
-+ Error **errp)
-+{
-+ Coroutine *co = qemu_coroutine_self();
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ CoCtxData waker = { .co = co, .ctx = ctx };
-+ char *pbs_err = NULL;
-+ int pbs_res = -1;
-+
-+ proxmox_backup_write_data_async(
-+ pbs, dev_id, data, offset, size, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ qemu_coroutine_yield();
-+ if (pbs_res < 0) {
-+ if (errp) error_setg(errp, "backup write data failed: %s", pbs_err ? pbs_err : "unknown error");
-+ if (pbs_err) proxmox_backup_free_error(pbs_err);
-+ }
-+ return pbs_res;
-+}
-diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
-new file mode 100644
-index 0000000000..1dda8b7d8f
---- /dev/null
-+++ b/proxmox-backup-client.h
-@@ -0,0 +1,59 @@
-+#ifndef PROXMOX_BACKUP_CLIENT_H
-+#define PROXMOX_BACKUP_CLIENT_H
-+
-+#include "qemu/osdep.h"
-+#include "qemu/coroutine.h"
-+#include "proxmox-backup-qemu.h"
-+
-+typedef struct CoCtxData {
-+ Coroutine *co;
-+ AioContext *ctx;
-+ void *data;
-+} CoCtxData;
-+
-+// FIXME: Remove once coroutines are supported for QMP
-+void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg);
-+
-+int coroutine_fn
-+proxmox_backup_co_connect(
-+ ProxmoxBackupHandle *pbs,
-+ Error **errp);
-+
-+int coroutine_fn
-+proxmox_backup_co_add_config(
-+ ProxmoxBackupHandle *pbs,
-+ const char *name,
-+ const uint8_t *data,
-+ uint64_t size,
-+ Error **errp);
-+
-+int coroutine_fn
-+proxmox_backup_co_register_image(
-+ ProxmoxBackupHandle *pbs,
-+ const char *device_name,
-+ uint64_t size,
-+ Error **errp);
-+
-+
-+int coroutine_fn
-+proxmox_backup_co_finish(
-+ ProxmoxBackupHandle *pbs,
-+ Error **errp);
-+
-+int coroutine_fn
-+proxmox_backup_co_close_image(
-+ ProxmoxBackupHandle *pbs,
-+ uint8_t dev_id,
-+ Error **errp);
-+
-+int coroutine_fn
-+proxmox_backup_co_write_data(
-+ ProxmoxBackupHandle *pbs,
-+ uint8_t dev_id,
-+ const uint8_t *data,
-+ uint64_t offset,
-+ uint64_t size,
-+ Error **errp);
-+
-+
-+#endif /* PROXMOX_BACKUP_CLIENT_H */
-diff --git a/pve-backup.c b/pve-backup.c
-new file mode 100644
-index 0000000000..88f5ee133f
---- /dev/null
-+++ b/pve-backup.c
-@@ -0,0 +1,959 @@
-+#include "proxmox-backup-client.h"
-+#include "vma.h"
-+
-+#include "qemu/osdep.h"
-+#include "qemu/module.h"
-+#include "sysemu/block-backend.h"
-+#include "sysemu/blockdev.h"
-+#include "block/blockjob.h"
-+#include "qapi/qapi-commands-block.h"
-+#include "qapi/qmp/qerror.h"
-+
-+/* PVE backup state and related function */
-+
-+/*
-+ * Note: A resume from a qemu_coroutine_yield can happen in a different thread,
-+ * so you may not use normal mutexes within coroutines:
-+ *
-+ * ---bad-example---
-+ * qemu_rec_mutex_lock(lock)
-+ * ...
-+ * qemu_coroutine_yield() // wait for something
-+ * // we are now inside a different thread
-+ * qemu_rec_mutex_unlock(lock) // Crash - wrong thread!!
-+ * ---end-bad-example--
-+ *
-+ * ==> Always use CoMutext inside coroutines.
-+ * ==> Never acquire/release AioContext withing coroutines (because that use QemuRecMutex)
-+ *
-+ */
-+
-+static struct PVEBackupState {
-+ struct {
-+ // Everithing accessed from qmp_backup_query command is protected using lock
-+ QemuMutex lock;
-+ Error *error;
-+ time_t start_time;
-+ time_t end_time;
-+ char *backup_file;
-+ uuid_t uuid;
-+ char uuid_str[37];
-+ size_t total;
-+ size_t transferred;
-+ size_t zero_bytes;
-+ } stat;
-+ int64_t speed;
-+ VmaWriter *vmaw;
-+ ProxmoxBackupHandle *pbs;
-+ GList *di_list;
-+ QemuMutex backup_mutex;
-+ CoMutex dump_callback_mutex;
-+} backup_state;
-+
-+static void pvebackup_init(void)
-+{
-+ qemu_mutex_init(&backup_state.stat.lock);
-+ qemu_mutex_init(&backup_state.backup_mutex);
-+ qemu_co_mutex_init(&backup_state.dump_callback_mutex);
-+}
-+
-+// initialize PVEBackupState at startup
-+opts_init(pvebackup_init);
-+
-+typedef struct PVEBackupDevInfo {
-+ BlockDriverState *bs;
-+ size_t size;
-+ uint8_t dev_id;
-+ bool completed;
-+ char targetfile[PATH_MAX];
-+ BlockDriverState *target;
-+} PVEBackupDevInfo;
-+
-+static void pvebackup_run_next_job(void);
-+
-+static BlockJob *
-+lookup_active_block_job(PVEBackupDevInfo *di)
-+{
-+ if (!di->completed && di->bs) {
-+ for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
-+ if (job->job.driver->job_type != JOB_TYPE_BACKUP) {
-+ continue;
-+ }
-+
-+ BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
-+ if (bjob && bjob->source_bs == di->bs) {
-+ return job;
-+ }
-+ }
-+ }
-+ return NULL;
-+}
-+
-+static void pvebackup_propagate_error(Error *err)
-+{
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ error_propagate(&backup_state.stat.error, err);
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+}
-+
-+static bool pvebackup_error_or_canceled(void)
-+{
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ bool error_or_canceled = !!backup_state.stat.error;
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+ return error_or_canceled;
-+}
-+
-+static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes)
-+{
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ backup_state.stat.zero_bytes += zero_bytes;
-+ backup_state.stat.transferred += transferred;
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+}
-+
-+// This may get called from multiple coroutines in multiple io-threads
-+// Note1: this may get called after job_cancel()
-+static int coroutine_fn
-+pvebackup_co_dump_pbs_cb(
-+ void *opaque,
-+ uint64_t start,
-+ uint64_t bytes,
-+ const void *pbuf)
-+{
-+ assert(qemu_in_coroutine());
-+
-+ const uint64_t size = bytes;
-+ const unsigned char *buf = pbuf;
-+ PVEBackupDevInfo *di = opaque;
-+
-+ assert(backup_state.pbs);
-+
-+ Error *local_err = NULL;
-+ int pbs_res = -1;
-+
-+ qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
-+
-+ // avoid deadlock if job is cancelled
-+ if (pvebackup_error_or_canceled()) {
-+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+ return -1;
-+ }
-+
-+ pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id, buf, start, size, &local_err);
-+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+
-+ if (pbs_res < 0) {
-+ pvebackup_propagate_error(local_err);
-+ return pbs_res;
-+ } else {
-+ pvebackup_add_transfered_bytes(size, !buf ? size : 0);
-+ }
-+
-+ return size;
-+}
-+
-+// This may get called from multiple coroutines in multiple io-threads
-+static int coroutine_fn
-+pvebackup_co_dump_vma_cb(
-+ void *opaque,
-+ uint64_t start,
-+ uint64_t bytes,
-+ const void *pbuf)
-+{
-+ assert(qemu_in_coroutine());
-+
-+ const uint64_t size = bytes;
-+ const unsigned char *buf = pbuf;
-+ PVEBackupDevInfo *di = opaque;
-+
-+ int ret = -1;
-+
-+ assert(backup_state.vmaw);
-+
-+ uint64_t remaining = size;
-+
-+ uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
-+ if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
-+ Error *local_err = NULL;
-+ error_setg(&local_err,
-+ "got unaligned write inside backup dump "
-+ "callback (sector %ld)", start);
-+ pvebackup_propagate_error(local_err);
-+ return -1; // not aligned to cluster size
-+ }
-+
-+ while (remaining > 0) {
-+ qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
-+ // avoid deadlock if job is cancelled
-+ if (pvebackup_error_or_canceled()) {
-+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+ return -1;
-+ }
-+
-+ size_t zero_bytes = 0;
-+ ret = vma_writer_write(backup_state.vmaw, di->dev_id, cluster_num, buf, &zero_bytes);
-+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+
-+ ++cluster_num;
-+ if (buf) {
-+ buf += VMA_CLUSTER_SIZE;
-+ }
-+ if (ret < 0) {
-+ Error *local_err = NULL;
-+ vma_writer_error_propagate(backup_state.vmaw, &local_err);
-+ pvebackup_propagate_error(local_err);
-+ return ret;
-+ } else {
-+ if (remaining >= VMA_CLUSTER_SIZE) {
-+ assert(ret == VMA_CLUSTER_SIZE);
-+ pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes);
-+ remaining -= VMA_CLUSTER_SIZE;
-+ } else {
-+ assert(ret == remaining);
-+ pvebackup_add_transfered_bytes(remaining, zero_bytes);
-+ remaining = 0;
-+ }
-+ }
-+ }
-+
-+ return size;
-+}
-+
-+// assumes the caller holds backup_mutex
-+static void coroutine_fn pvebackup_co_cleanup(void *unused)
-+{
-+ assert(qemu_in_coroutine());
-+
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ backup_state.stat.end_time = time(NULL);
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+ if (backup_state.vmaw) {
-+ Error *local_err = NULL;
-+ vma_writer_close(backup_state.vmaw, &local_err);
-+
-+ if (local_err != NULL) {
-+ pvebackup_propagate_error(local_err);
-+ }
-+
-+ backup_state.vmaw = NULL;
-+ }
-+
-+ if (backup_state.pbs) {
-+ if (!pvebackup_error_or_canceled()) {
-+ Error *local_err = NULL;
-+ proxmox_backup_co_finish(backup_state.pbs, &local_err);
-+ if (local_err != NULL) {
-+ pvebackup_propagate_error(local_err);
-+ }
-+ }
-+
-+ proxmox_backup_disconnect(backup_state.pbs);
-+ backup_state.pbs = NULL;
-+ }
-+
-+ g_list_free(backup_state.di_list);
-+ backup_state.di_list = NULL;
-+}
-+
-+// assumes the caller holds backup_mutex
-+static void coroutine_fn pvebackup_complete_stream(void *opaque)
-+{
-+ PVEBackupDevInfo *di = opaque;
-+
-+ bool error_or_canceled = pvebackup_error_or_canceled();
-+
-+ if (backup_state.vmaw) {
-+ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-+ }
-+
-+ if (backup_state.pbs && !error_or_canceled) {
-+ Error *local_err = NULL;
-+ proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
-+ if (local_err != NULL) {
-+ pvebackup_propagate_error(local_err);
-+ }
-+ }
-+}
-+
-+static void pvebackup_complete_cb(void *opaque, int ret)
-+{
-+ assert(!qemu_in_coroutine());
-+
-+ PVEBackupDevInfo *di = opaque;
-+
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+
-+ di->completed = true;
-+
-+ if (ret < 0) {
-+ Error *local_err = NULL;
-+ error_setg(&local_err, "job failed with err %d - %s", ret, strerror(-ret));
-+ pvebackup_propagate_error(local_err);
-+ }
-+
-+ di->bs = NULL;
-+
-+ assert(di->target == NULL);
-+
-+ block_on_coroutine_fn(pvebackup_complete_stream, di);
-+
-+ // remove self from job queue
-+ backup_state.di_list = g_list_remove(backup_state.di_list, di);
-+
-+ g_free(di);
-+
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ pvebackup_run_next_job();
-+}
-+
-+static void pvebackup_cancel(void)
-+{
-+ assert(!qemu_in_coroutine());
-+
-+ Error *cancel_err = NULL;
-+ error_setg(&cancel_err, "backup canceled");
-+ pvebackup_propagate_error(cancel_err);
-+
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+
-+ if (backup_state.vmaw) {
-+ /* make sure vma writer does not block anymore */
-+ vma_writer_set_error(backup_state.vmaw, "backup canceled");
-+ }
-+
-+ if (backup_state.pbs) {
-+ proxmox_backup_abort(backup_state.pbs, "backup canceled");
-+ }
-+
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ for(;;) {
-+
-+ BlockJob *next_job = NULL;
-+
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+
-+ GList *l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ BlockJob *job = lookup_active_block_job(di);
-+ if (job != NULL) {
-+ next_job = job;
-+ break;
-+ }
-+ }
-+
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ if (next_job) {
-+ AioContext *aio_context = next_job->job.aio_context;
-+ aio_context_acquire(aio_context);
-+ job_cancel_sync(&next_job->job, true);
-+ aio_context_release(aio_context);
-+ } else {
-+ break;
-+ }
-+ }
-+}
-+
-+void qmp_backup_cancel(Error **errp)
-+{
-+ pvebackup_cancel();
-+}
-+
-+// assumes the caller holds backup_mutex
-+static int coroutine_fn pvebackup_co_add_config(
-+ const char *file,
-+ const char *name,
-+ BackupFormat format,
-+ const char *backup_dir,
-+ VmaWriter *vmaw,
-+ ProxmoxBackupHandle *pbs,
-+ Error **errp)
-+{
-+ int res = 0;
-+
-+ char *cdata = NULL;
-+ gsize clen = 0;
-+ GError *err = NULL;
-+ if (!g_file_get_contents(file, &cdata, &clen, &err)) {
-+ error_setg(errp, "unable to read file '%s'", file);
-+ return 1;
-+ }
-+
-+ char *basename = g_path_get_basename(file);
-+ if (name == NULL) name = basename;
-+
-+ if (format == BACKUP_FORMAT_VMA) {
-+ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
-+ error_setg(errp, "unable to add %s config data to vma archive", file);
-+ goto err;
-+ }
-+ } else if (format == BACKUP_FORMAT_PBS) {
-+ if (proxmox_backup_co_add_config(pbs, name, (unsigned char *)cdata, clen, errp) < 0)
-+ goto err;
-+ } else if (format == BACKUP_FORMAT_DIR) {
-+ char config_path[PATH_MAX];
-+ snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, name);
-+ if (!g_file_set_contents(config_path, cdata, clen, &err)) {
-+ error_setg(errp, "unable to write config file '%s'", config_path);
-+ goto err;
-+ }
-+ }
-+
-+ out:
-+ g_free(basename);
-+ g_free(cdata);
-+ return res;
-+
-+ err:
-+ res = -1;
-+ goto out;
-+}
-+
-+bool job_should_pause(Job *job);
-+
-+static void pvebackup_run_next_job(void)
-+{
-+ assert(!qemu_in_coroutine());
-+
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+
-+ GList *l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ BlockJob *job = lookup_active_block_job(di);
-+
-+ if (job) {
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ AioContext *aio_context = job->job.aio_context;
-+ aio_context_acquire(aio_context);
-+
-+ if (job_should_pause(&job->job)) {
-+ bool error_or_canceled = pvebackup_error_or_canceled();
-+ if (error_or_canceled) {
-+ job_cancel_sync(&job->job, true);
-+ } else {
-+ job_resume(&job->job);
-+ }
-+ }
-+ aio_context_release(aio_context);
-+ return;
-+ }
-+ }
-+
-+ block_on_coroutine_fn(pvebackup_co_cleanup, NULL); // no more jobs, run cleanup
-+
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+}
-+
-+static bool create_backup_jobs(void) {
-+
-+ assert(!qemu_in_coroutine());
-+
-+ Error *local_err = NULL;
-+
-+ BackupPerf perf = { .max_workers = 16 };
-+
-+ /* create and start all jobs (paused state) */
-+ GList *l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ assert(di->target != NULL);
-+
-+ AioContext *aio_context = bdrv_get_aio_context(di->bs);
-+ aio_context_acquire(aio_context);
-+
-+ BlockJob *job = backup_job_create(
-+ NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
-+ BITMAP_SYNC_MODE_NEVER, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-+ JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
-+
-+ aio_context_release(aio_context);
-+
-+ if (!job || local_err != NULL) {
-+ Error *create_job_err = NULL;
-+ error_setg(&create_job_err, "backup_job_create failed: %s",
-+ local_err ? error_get_pretty(local_err) : "null");
-+
-+ pvebackup_propagate_error(create_job_err);
-+ break;
-+ }
-+ job_start(&job->job);
-+
-+ bdrv_unref(di->target);
-+ di->target = NULL;
-+ }
-+
-+ bool errors = pvebackup_error_or_canceled();
-+
-+ if (errors) {
-+ l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ if (di->target) {
-+ bdrv_unref(di->target);
-+ di->target = NULL;
-+ }
-+ }
-+ }
-+
-+ return errors;
-+}
-+
-+typedef struct QmpBackupTask {
-+ const char *backup_file;
-+ bool has_password;
-+ const char *password;
-+ bool has_keyfile;
-+ const char *keyfile;
-+ bool has_key_password;
-+ const char *key_password;
-+ bool has_backup_id;
-+ const char *backup_id;
-+ bool has_backup_time;
-+ const char *fingerprint;
-+ bool has_fingerprint;
-+ int64_t backup_time;
-+ bool has_format;
-+ BackupFormat format;
-+ bool has_config_file;
-+ const char *config_file;
-+ bool has_firewall_file;
-+ const char *firewall_file;
-+ bool has_devlist;
-+ const char *devlist;
-+ bool has_speed;
-+ int64_t speed;
-+ Error **errp;
-+ UuidInfo *result;
-+} QmpBackupTask;
-+
-+// assumes the caller holds backup_mutex
-+static void coroutine_fn pvebackup_co_prepare(void *opaque)
-+{
-+ assert(qemu_in_coroutine());
-+
-+ QmpBackupTask *task = opaque;
-+
-+ task->result = NULL; // just to be sure
-+
-+ BlockBackend *blk;
-+ BlockDriverState *bs = NULL;
-+ const char *backup_dir = NULL;
-+ Error *local_err = NULL;
-+ uuid_t uuid;
-+ VmaWriter *vmaw = NULL;
-+ ProxmoxBackupHandle *pbs = NULL;
-+ gchar **devs = NULL;
-+ GList *di_list = NULL;
-+ GList *l;
-+ UuidInfo *uuid_info;
-+
-+ const char *config_name = "qemu-server.conf";
-+ const char *firewall_name = "qemu-server.fw";
-+
-+ if (backup_state.di_list) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ "previous backup not finished");
-+ return;
-+ }
-+
-+ /* Todo: try to auto-detect format based on file name */
-+ BackupFormat format = task->has_format ? task->format : BACKUP_FORMAT_VMA;
-+
-+ if (task->has_devlist) {
-+ devs = g_strsplit_set(task->devlist, ",;:", -1);
-+
-+ gchar **d = devs;
-+ while (d && *d) {
-+ blk = blk_by_name(*d);
-+ if (blk) {
-+ bs = blk_bs(blk);
-+ if (!bdrv_is_inserted(bs)) {
-+ error_setg(task->errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
-+ goto err;
-+ }
-+ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
-+ di->bs = bs;
-+ di_list = g_list_append(di_list, di);
-+ } else {
-+ error_set(task->errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-+ "Device '%s' not found", *d);
-+ goto err;
-+ }
-+ d++;
-+ }
-+
-+ } else {
-+ BdrvNextIterator it;
-+
-+ bs = NULL;
-+ for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
-+ if (!bdrv_is_inserted(bs) || bdrv_is_read_only(bs)) {
-+ continue;
-+ }
-+
-+ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
-+ di->bs = bs;
-+ di_list = g_list_append(di_list, di);
-+ }
-+ }
-+
-+ if (!di_list) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
-+ goto err;
-+ }
-+
-+ size_t total = 0;
-+
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, task->errp)) {
-+ goto err;
-+ }
-+
-+ ssize_t size = bdrv_getlength(di->bs);
-+ if (size < 0) {
-+ error_setg_errno(task->errp, -di->size, "bdrv_getlength failed");
-+ goto err;
-+ }
-+ di->size = size;
-+ total += size;
-+ }
-+
-+ uuid_generate(uuid);
-+
-+ if (format == BACKUP_FORMAT_PBS) {
-+ if (!task->has_password) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
-+ goto err;
-+ }
-+ if (!task->has_backup_id) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
-+ goto err;
-+ }
-+ if (!task->has_backup_time) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
-+ goto err;
-+ }
-+
-+ int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
-+ firewall_name = "fw.conf";
-+
-+ char *pbs_err = NULL;
-+ pbs = proxmox_backup_new(
-+ task->backup_file,
-+ task->backup_id,
-+ task->backup_time,
-+ dump_cb_block_size,
-+ task->has_password ? task->password : NULL,
-+ task->has_keyfile ? task->keyfile : NULL,
-+ task->has_key_password ? task->key_password : NULL,
-+ task->has_fingerprint ? task->fingerprint : NULL,
-+ &pbs_err);
-+
-+ if (!pbs) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ "proxmox_backup_new failed: %s", pbs_err);
-+ proxmox_backup_free_error(pbs_err);
-+ goto err;
-+ }
-+
-+ if (proxmox_backup_co_connect(pbs, task->errp) < 0)
-+ goto err;
-+
-+ /* register all devices */
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ const char *devname = bdrv_get_device_name(di->bs);
-+
-+ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, task->errp);
-+ if (dev_id < 0)
-+ goto err;
-+
-+ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
-+ goto err;
-+ }
-+
-+ di->dev_id = dev_id;
-+ }
-+ } else if (format == BACKUP_FORMAT_VMA) {
-+ vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
-+ if (!vmaw) {
-+ if (local_err) {
-+ error_propagate(task->errp, local_err);
-+ }
-+ goto err;
-+ }
-+
-+ /* register all devices for vma writer */
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
-+ goto err;
-+ }
-+
-+ const char *devname = bdrv_get_device_name(di->bs);
-+ di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
-+ if (di->dev_id <= 0) {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ "register_stream failed");
-+ goto err;
-+ }
-+ }
-+ } else if (format == BACKUP_FORMAT_DIR) {
-+ if (mkdir(task->backup_file, 0640) != 0) {
-+ error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
-+ task->backup_file);
-+ goto err;
-+ }
-+ backup_dir = task->backup_file;
-+
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ const char *devname = bdrv_get_device_name(di->bs);
-+ snprintf(di->targetfile, PATH_MAX, "%s/%s.raw", backup_dir, devname);
-+
-+ int flags = BDRV_O_RDWR;
-+ bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
-+ di->size, flags, false, &local_err);
-+ if (local_err) {
-+ error_propagate(task->errp, local_err);
-+ goto err;
-+ }
-+
-+ di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
-+ if (!di->target) {
-+ error_propagate(task->errp, local_err);
-+ goto err;
-+ }
-+ }
-+ } else {
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
-+ goto err;
-+ }
-+
-+
-+ /* add configuration file to archive */
-+ if (task->has_config_file) {
-+ if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
-+ vmaw, pbs, task->errp) != 0) {
-+ goto err;
-+ }
-+ }
-+
-+ /* add firewall file to archive */
-+ if (task->has_firewall_file) {
-+ if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
-+ vmaw, pbs, task->errp) != 0) {
-+ goto err;
-+ }
-+ }
-+ /* initialize global backup_state now */
-+
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+
-+ if (backup_state.stat.error) {
-+ error_free(backup_state.stat.error);
-+ backup_state.stat.error = NULL;
-+ }
-+
-+ backup_state.stat.start_time = time(NULL);
-+ backup_state.stat.end_time = 0;
-+
-+ if (backup_state.stat.backup_file) {
-+ g_free(backup_state.stat.backup_file);
-+ }
-+ backup_state.stat.backup_file = g_strdup(task->backup_file);
-+
-+ uuid_copy(backup_state.stat.uuid, uuid);
-+ uuid_unparse_lower(uuid, backup_state.stat.uuid_str);
-+ char *uuid_str = g_strdup(backup_state.stat.uuid_str);
-+
-+ backup_state.stat.total = total;
-+ backup_state.stat.transferred = 0;
-+ backup_state.stat.zero_bytes = 0;
-+
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+ backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
-+
-+ backup_state.vmaw = vmaw;
-+ backup_state.pbs = pbs;
-+
-+ backup_state.di_list = di_list;
-+
-+ uuid_info = g_malloc0(sizeof(*uuid_info));
-+ uuid_info->UUID = uuid_str;
-+
-+ task->result = uuid_info;
-+ return;
-+
-+err:
-+
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ if (di->target) {
-+ bdrv_unref(di->target);
-+ }
-+
-+ if (di->targetfile[0]) {
-+ unlink(di->targetfile);
-+ }
-+ g_free(di);
-+ }
-+ g_list_free(di_list);
-+
-+ if (devs) {
-+ g_strfreev(devs);
-+ }
-+
-+ if (vmaw) {
-+ Error *err = NULL;
-+ vma_writer_close(vmaw, &err);
-+ unlink(task->backup_file);
-+ }
-+
-+ if (pbs) {
-+ proxmox_backup_disconnect(pbs);
-+ }
-+
-+ if (backup_dir) {
-+ rmdir(backup_dir);
-+ }
-+
-+ task->result = NULL;
-+ return;
-+}
-+
-+UuidInfo *qmp_backup(
-+ const char *backup_file,
-+ bool has_password, const char *password,
-+ bool has_keyfile, const char *keyfile,
-+ bool has_key_password, const char *key_password,
-+ bool has_fingerprint, const char *fingerprint,
-+ bool has_backup_id, const char *backup_id,
-+ bool has_backup_time, int64_t backup_time,
-+ bool has_format, BackupFormat format,
-+ bool has_config_file, const char *config_file,
-+ bool has_firewall_file, const char *firewall_file,
-+ bool has_devlist, const char *devlist,
-+ bool has_speed, int64_t speed, Error **errp)
-+{
-+ QmpBackupTask task = {
-+ .backup_file = backup_file,
-+ .has_password = has_password,
-+ .password = password,
-+ .has_key_password = has_key_password,
-+ .key_password = key_password,
-+ .has_fingerprint = has_fingerprint,
-+ .fingerprint = fingerprint,
-+ .has_backup_id = has_backup_id,
-+ .backup_id = backup_id,
-+ .has_backup_time = has_backup_time,
-+ .backup_time = backup_time,
-+ .has_format = has_format,
-+ .format = format,
-+ .has_config_file = has_config_file,
-+ .config_file = config_file,
-+ .has_firewall_file = has_firewall_file,
-+ .firewall_file = firewall_file,
-+ .has_devlist = has_devlist,
-+ .devlist = devlist,
-+ .has_speed = has_speed,
-+ .speed = speed,
-+ .errp = errp,
-+ };
-+
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+
-+ block_on_coroutine_fn(pvebackup_co_prepare, &task);
-+
-+ if (*errp == NULL) {
-+ create_backup_jobs();
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+ pvebackup_run_next_job();
-+ } else {
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+ }
-+
-+ return task.result;
-+}
-+
-+BackupStatus *qmp_query_backup(Error **errp)
-+{
-+ BackupStatus *info = g_malloc0(sizeof(*info));
-+
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+
-+ if (!backup_state.stat.start_time) {
-+ /* not started, return {} */
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+ return info;
-+ }
-+
-+ info->has_status = true;
-+ info->has_start_time = true;
-+ info->start_time = backup_state.stat.start_time;
-+
-+ if (backup_state.stat.backup_file) {
-+ info->has_backup_file = true;
-+ info->backup_file = g_strdup(backup_state.stat.backup_file);
-+ }
-+
-+ info->has_uuid = true;
-+ info->uuid = g_strdup(backup_state.stat.uuid_str);
-+
-+ if (backup_state.stat.end_time) {
-+ if (backup_state.stat.error) {
-+ info->status = g_strdup("error");
-+ info->has_errmsg = true;
-+ info->errmsg = g_strdup(error_get_pretty(backup_state.stat.error));
-+ } else {
-+ info->status = g_strdup("done");
-+ }
-+ info->has_end_time = true;
-+ info->end_time = backup_state.stat.end_time;
-+ } else {
-+ info->status = g_strdup("active");
-+ }
-+
-+ info->has_total = true;
-+ info->total = backup_state.stat.total;
-+ info->has_zero_bytes = true;
-+ info->zero_bytes = backup_state.stat.zero_bytes;
-+ info->has_transferred = true;
-+ info->transferred = backup_state.stat.transferred;
-+
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+ return info;
-+}
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 564b6b43f7..ffcc25367b 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -744,6 +744,115 @@
- { 'command': 'query-block', 'returns': ['BlockInfo'] }
-
-
-+##
-+# @BackupStatus:
-+#
-+# Detailed backup status.
-+#
-+# @status: string describing the current backup status.
-+# This can be 'active', 'done', 'error'. If this field is not
-+# returned, no backup process has been initiated
-+#
-+# @errmsg: error message (only returned if status is 'error')
-+#
-+# @total: total amount of bytes involved in the backup process
-+#
-+# @transferred: amount of bytes already backed up.
-+#
-+# @zero-bytes: amount of 'zero' bytes detected.
-+#
-+# @start-time: time (epoch) when backup job started.
-+#
-+# @end-time: time (epoch) when backup job finished.
-+#
-+# @backup-file: backup file name
-+#
-+# @uuid: uuid for this backup job
-+#
-+##
-+{ 'struct': 'BackupStatus',
-+ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
-+ '*transferred': 'int', '*zero-bytes': 'int',
-+ '*start-time': 'int', '*end-time': 'int',
-+ '*backup-file': 'str', '*uuid': 'str' } }
-+
-+##
-+# @BackupFormat:
-+#
-+# An enumeration of supported backup formats.
-+#
-+# @vma: Proxmox vma backup format
-+##
-+{ 'enum': 'BackupFormat',
-+ 'data': [ 'vma', 'dir', 'pbs' ] }
-+
-+##
-+# @backup:
-+#
-+# Starts a VM backup.
-+#
-+# @backup-file: the backup file name
-+#
-+# @format: format of the backup file
-+#
-+# @config-file: a configuration file to include into
-+# the backup archive.
-+#
-+# @speed: the maximum speed, in bytes per second
-+#
-+# @devlist: list of block device names (separated by ',', ';'
-+# or ':'). By default the backup includes all writable block devices.
-+#
-+# @password: backup server passsword (required for format 'pbs')
-+#
-+# @keyfile: keyfile used for encryption (optional for format 'pbs')
-+#
-+# @key-password: password for keyfile (optional for format 'pbs')
-+#
-+# @fingerprint: server cert fingerprint (optional for format 'pbs')
-+#
-+# @backup-id: backup ID (required for format 'pbs')
-+#
-+# @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
-+#
-+# Returns: the uuid of the backup job
-+#
-+##
-+{ 'command': 'backup', 'data': { 'backup-file': 'str',
-+ '*password': 'str',
-+ '*keyfile': 'str',
-+ '*key-password': 'str',
-+ '*fingerprint': 'str',
-+ '*backup-id': 'str',
-+ '*backup-time': 'int',
-+ '*format': 'BackupFormat',
-+ '*config-file': 'str',
-+ '*firewall-file': 'str',
-+ '*devlist': 'str', '*speed': 'int' },
-+ 'returns': 'UuidInfo' }
-+
-+##
-+# @query-backup:
-+#
-+# Returns information about current/last backup task.
-+#
-+# Returns: @BackupStatus
-+#
-+##
-+{ 'command': 'query-backup', 'returns': 'BackupStatus' }
-+
-+##
-+# @backup-cancel:
-+#
-+# Cancel the current executing backup process.
-+#
-+# Returns: nothing on success
-+#
-+# Notes: This command succeeds even if there is no backup process running.
-+#
-+##
-+{ 'command': 'backup-cancel' }
-+
- ##
- # @BlockDeviceTimedStats:
- #
-diff --git a/qapi/common.json b/qapi/common.json
-index 412cc4f5ae..3e7a77ea66 100644
---- a/qapi/common.json
-+++ b/qapi/common.json
-@@ -208,3 +208,16 @@
- ##
- { 'struct': 'HumanReadableText',
- 'data': { 'human-readable-text': 'str' } }
-+
-+##
-+# @UuidInfo:
-+#
-+# Guest UUID information (Universally Unique Identifier).
-+#
-+# @UUID: the UUID of the guest
-+#
-+# Since: 0.14.0
-+#
-+# Notes: If no UUID was specified for the guest, a null UUID is returned.
-+##
-+{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
-diff --git a/qapi/machine.json b/qapi/machine.json
-index 39ba204b52..091aafeff7 100644
---- a/qapi/machine.json
-+++ b/qapi/machine.json
-@@ -4,6 +4,8 @@
- # This work is licensed under the terms of the GNU GPL, version 2 or later.
- # See the COPYING file in the top-level directory.
-
-+{ 'include': 'common.json' }
-+
- ##
- # = Machines
- ##
-@@ -227,19 +229,6 @@
- ##
- { 'command': 'query-target', 'returns': 'TargetInfo' }
-
--##
--# @UuidInfo:
--#
--# Guest UUID information (Universally Unique Identifier).
--#
--# @UUID: the UUID of the guest
--#
--# Since: 0.14
--#
--# Notes: If no UUID was specified for the guest, a null UUID is returned.
--##
--{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
--
- ##
- # @query-uuid:
- #
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:17:01 +0200
-Subject: [PATCH] PVE-Backup: pbs-restore - new command to restore from proxmox
- backup server
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- meson.build | 4 +
- pbs-restore.c | 224 ++++++++++++++++++++++++++++++++++++++++++++++++++
- 2 files changed, 228 insertions(+)
- create mode 100644 pbs-restore.c
-
-diff --git a/meson.build b/meson.build
-index 76fcb23299..9476ccbe07 100644
---- a/meson.build
-+++ b/meson.build
-@@ -3365,6 +3365,10 @@ if have_tools
- vma = executable('vma', files('vma.c', 'vma-reader.c') + genh,
- dependencies: [authz, block, crypto, io, qom], install: true)
-
-+ pbs_restore = executable('pbs-restore', files('pbs-restore.c') + genh,
-+ dependencies: [authz, block, crypto, io, qom,
-+ libproxmox_backup_qemu], install: true)
-+
- subdir('storage-daemon')
- subdir('contrib/rdmacm-mux')
- subdir('contrib/elf2dmp')
-diff --git a/pbs-restore.c b/pbs-restore.c
-new file mode 100644
-index 0000000000..4d3f925a1b
---- /dev/null
-+++ b/pbs-restore.c
-@@ -0,0 +1,224 @@
-+/*
-+ * Qemu image restore helper for Proxmox Backup
-+ *
-+ * Copyright (C) 2019 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+#include <getopt.h>
-+#include <string.h>
-+
-+#include "qemu-common.h"
-+#include "qemu/module.h"
-+#include "qemu/error-report.h"
-+#include "qemu/main-loop.h"
-+#include "qemu/cutils.h"
-+#include "qapi/error.h"
-+#include "qapi/qmp/qdict.h"
-+#include "sysemu/block-backend.h"
-+
-+#include <proxmox-backup-qemu.h>
-+
-+static void help(void)
-+{
-+ const char *help_msg =
-+ "usage: pbs-restore [--repository <repo>] snapshot archive-name target [command options]\n"
-+ ;
-+
-+ printf("%s", help_msg);
-+ exit(1);
-+}
-+
-+typedef struct CallbackData {
-+ BlockBackend *target;
-+ uint64_t last_offset;
-+ bool skip_zero;
-+} CallbackData;
-+
-+static int write_callback(
-+ void *callback_data_ptr,
-+ uint64_t offset,
-+ const unsigned char *data,
-+ uint64_t data_len)
-+{
-+ int res = -1;
-+
-+ CallbackData *callback_data = (CallbackData *)callback_data_ptr;
-+
-+ uint64_t last_offset = callback_data->last_offset;
-+ if (offset > last_offset) callback_data->last_offset = offset;
-+
-+ if (data == NULL) {
-+ if (callback_data->skip_zero && offset > last_offset) {
-+ return 0;
-+ }
-+ res = blk_pwrite_zeroes(callback_data->target, offset, data_len, 0);
-+ } else {
-+ res = blk_pwrite(callback_data->target, offset, data, data_len, 0);
-+ }
-+
-+ if (res < 0) {
-+ fprintf(stderr, "blk_pwrite failed at offset %ld length %ld (%d) - %s\n", offset, data_len, res, strerror(-res));
-+ return res;
-+ }
-+
-+ return 0;
-+}
-+
-+int main(int argc, char **argv)
-+{
-+ Error *main_loop_err = NULL;
-+ const char *format = "raw";
-+ const char *repository = NULL;
-+ const char *keyfile = NULL;
-+ int verbose = false;
-+ bool skip_zero = false;
-+
-+ error_init(argv[0]);
-+
-+ for (;;) {
-+ static const struct option long_options[] = {
-+ {"help", no_argument, 0, 'h'},
-+ {"skip-zero", no_argument, 0, 'S'},
-+ {"verbose", no_argument, 0, 'v'},
-+ {"format", required_argument, 0, 'f'},
-+ {"repository", required_argument, 0, 'r'},
-+ {"keyfile", required_argument, 0, 'k'},
-+ {0, 0, 0, 0}
-+ };
-+ int c = getopt_long(argc, argv, "hvf:r:k:", long_options, NULL);
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case ':':
-+ fprintf(stderr, "missing argument for option '%s'\n", argv[optind - 1]);
-+ return -1;
-+ case '?':
-+ fprintf(stderr, "unrecognized option '%s'\n", argv[optind - 1]);
-+ return -1;
-+ case 'f':
-+ format = g_strdup(argv[optind - 1]);
-+ break;
-+ case 'r':
-+ repository = g_strdup(argv[optind - 1]);
-+ break;
-+ case 'k':
-+ keyfile = g_strdup(argv[optind - 1]);
-+ break;
-+ case 'v':
-+ verbose = true;
-+ break;
-+ case 'S':
-+ skip_zero = true;
-+ break;
-+ case 'h':
-+ help();
-+ return 0;
-+ }
-+ }
-+
-+ if (optind >= argc - 2) {
-+ fprintf(stderr, "missing arguments\n");
-+ help();
-+ return -1;
-+ }
-+
-+ if (repository == NULL) {
-+ repository = getenv("PBS_REPOSITORY");
-+ }
-+
-+ if (repository == NULL) {
-+ fprintf(stderr, "no repository specified\n");
-+ help();
-+ return -1;
-+ }
-+
-+ char *snapshot = argv[optind++];
-+ char *archive_name = argv[optind++];
-+ char *target = argv[optind++];
-+
-+ const char *password = getenv("PBS_PASSWORD");
-+ const char *fingerprint = getenv("PBS_FINGERPRINT");
-+ const char *key_password = getenv("PBS_ENCRYPTION_PASSWORD");
-+
-+ if (qemu_init_main_loop(&main_loop_err)) {
-+ g_error("%s", error_get_pretty(main_loop_err));
-+ }
-+
-+ bdrv_init();
-+ module_call_init(MODULE_INIT_QOM);
-+
-+ if (verbose) {
-+ fprintf(stderr, "connecting to repository '%s'\n", repository);
-+ }
-+ char *pbs_error = NULL;
-+ ProxmoxRestoreHandle *conn = proxmox_restore_new(
-+ repository, snapshot, password, keyfile, key_password, fingerprint, &pbs_error);
-+ if (conn == NULL) {
-+ fprintf(stderr, "restore failed: %s\n", pbs_error);
-+ return -1;
-+ }
-+
-+ int res = proxmox_restore_connect(conn, &pbs_error);
-+ if (res < 0 || pbs_error) {
-+ fprintf(stderr, "restore failed (connection error): %s\n", pbs_error);
-+ return -1;
-+ }
-+
-+ QDict *options = qdict_new();
-+
-+ if (format) {
-+ qdict_put_str(options, "driver", format);
-+ }
-+
-+
-+ if (verbose) {
-+ fprintf(stderr, "open block backend for target '%s'\n", target);
-+ }
-+ Error *local_err = NULL;
-+ int flags = BDRV_O_RDWR;
-+ BlockBackend *blk = blk_new_open(target, NULL, options, flags, &local_err);
-+ if (!blk) {
-+ fprintf(stderr, "%s\n", error_get_pretty(local_err));
-+ return -1;
-+ }
-+
-+ CallbackData *callback_data = calloc(sizeof(CallbackData), 1);
-+
-+ callback_data->target = blk;
-+ callback_data->skip_zero = skip_zero;
-+ callback_data->last_offset = 0;
-+
-+ // blk_set_enable_write_cache(blk, !writethrough);
-+
-+ if (verbose) {
-+ fprintf(stderr, "starting to restore snapshot '%s'\n", snapshot);
-+ fflush(stderr); // ensure we do not get printed after the progress log
-+ }
-+ res = proxmox_restore_image(
-+ conn,
-+ archive_name,
-+ write_callback,
-+ callback_data,
-+ &pbs_error,
-+ verbose);
-+
-+ proxmox_restore_disconnect(conn);
-+ blk_unref(blk);
-+
-+ if (res < 0) {
-+ fprintf(stderr, "restore failed: %s\n", pbs_error);
-+ return -1;
-+ }
-+
-+ return 0;
-+}
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Mon, 6 Apr 2020 12:16:59 +0200
+Subject: [PATCH] PVE-Backup: proxmox backup patches for qemu
+
+Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
+[PVE-Backup: avoid coroutines to fix AIO freeze, cleanups]
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[add new force parameter to job_cancel_sync calls]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/meson.build | 5 +
+ block/monitor/block-hmp-cmds.c | 33 ++
+ blockdev.c | 1 +
+ hmp-commands-info.hx | 14 +
+ hmp-commands.hx | 29 +
+ include/monitor/hmp.h | 3 +
+ meson.build | 1 +
+ monitor/hmp-cmds.c | 44 ++
+ proxmox-backup-client.c | 176 ++++++
+ proxmox-backup-client.h | 59 ++
+ pve-backup.c | 959 +++++++++++++++++++++++++++++++++
+ qapi/block-core.json | 109 ++++
+ qapi/common.json | 13 +
+ qapi/machine.json | 15 +-
+ 14 files changed, 1448 insertions(+), 13 deletions(-)
+ create mode 100644 proxmox-backup-client.c
+ create mode 100644 proxmox-backup-client.h
+ create mode 100644 pve-backup.c
+
+diff --git a/block/meson.build b/block/meson.build
+index 7f22e7f177..2783b77e9c 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -48,6 +48,11 @@ block_ss.add(files(
+ ), zstd, zlib, gnutls)
+
+ block_ss.add(files('../vma-writer.c'), libuuid)
++block_ss.add(files(
++ '../proxmox-backup-client.c',
++ '../pve-backup.c',
++), libproxmox_backup_qemu)
++
+
+ softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index bfb3c043a0..89ca64444d 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1015,3 +1015,36 @@ void hmp_info_snapshots(Monitor *mon, const QDict *qdict)
+ g_free(sn_tab);
+ g_free(global_snapshots);
+ }
++
++void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
++{
++ Error *error = NULL;
++
++ qmp_backup_cancel(&error);
++
++ hmp_handle_error(mon, error);
++}
++
++void hmp_backup(Monitor *mon, const QDict *qdict)
++{
++ Error *error = NULL;
++
++ int dir = qdict_get_try_bool(qdict, "directory", 0);
++ const char *backup_file = qdict_get_str(qdict, "backupfile");
++ const char *devlist = qdict_get_try_str(qdict, "devlist");
++ int64_t speed = qdict_get_try_int(qdict, "speed", 0);
++
++ qmp_backup(
++ backup_file,
++ false, NULL, // PBS password
++ false, NULL, // PBS keyfile
++ false, NULL, // PBS key_password
++ false, NULL, // PBS fingerprint
++ false, NULL, // PBS backup-id
++ false, 0, // PBS backup-time
++ true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
++ false, NULL, false, NULL, !!devlist,
++ devlist, qdict_haskey(qdict, "speed"), speed, &error);
++
++ hmp_handle_error(mon, error);
++}
+diff --git a/blockdev.c b/blockdev.c
+index ce62a9b439..1600b24eab 100644
+--- a/blockdev.c
++++ b/blockdev.c
+@@ -36,6 +36,7 @@
+ #include "hw/block/block.h"
+ #include "block/blockjob.h"
+ #include "block/qdict.h"
++#include "block/blockjob_int.h"
+ #include "block/throttle-groups.h"
+ #include "monitor/monitor.h"
+ #include "qemu/error-report.h"
+diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
+index 97b88eaaad..92a8867afb 100644
+--- a/hmp-commands-info.hx
++++ b/hmp-commands-info.hx
+@@ -484,6 +484,20 @@ SRST
+ Show the current VM UUID.
+ ERST
+
++
++ {
++ .name = "backup",
++ .args_type = "",
++ .params = "",
++ .help = "show backup status",
++ .cmd = hmp_info_backup,
++ },
++
++SRST
++ ``info backup``
++ Show backup status.
++ERST
++
+ #if defined(CONFIG_SLIRP)
+ {
+ .name = "usernet",
+diff --git a/hmp-commands.hx b/hmp-commands.hx
+index bbcc73e942..97f24942b3 100644
+--- a/hmp-commands.hx
++++ b/hmp-commands.hx
+@@ -101,6 +101,35 @@ ERST
+ SRST
+ ``block_stream``
+ Copy data from a backing file into a block device.
++ERST
++
++ {
++ .name = "backup",
++ .args_type = "directory:-d,backupfile:s,speed:o?,devlist:s?",
++ .params = "[-d] backupfile [speed [devlist]]",
++ .help = "create a VM Backup."
++ "\n\t\t\t Use -d to dump data into a directory instead"
++ "\n\t\t\t of using VMA format.",
++ .cmd = hmp_backup,
++ },
++
++SRST
++``backup``
++ Create a VM backup.
++ERST
++
++ {
++ .name = "backup_cancel",
++ .args_type = "",
++ .params = "",
++ .help = "cancel the current VM backup",
++ .cmd = hmp_backup_cancel,
++ },
++
++SRST
++``backup_cancel``
++ Cancel the current VM backup.
++
+ ERST
+
+ {
+diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
+index 55067beff1..5a98d2d927 100644
+--- a/include/monitor/hmp.h
++++ b/include/monitor/hmp.h
+@@ -30,6 +30,7 @@ void hmp_info_savevm(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_capabilities(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict);
++void hmp_info_backup(Monitor *mon, const QDict *qdict);
+ void hmp_info_cpus(Monitor *mon, const QDict *qdict);
+ void hmp_info_vnc(Monitor *mon, const QDict *qdict);
+ void hmp_info_spice(Monitor *mon, const QDict *qdict);
+@@ -73,6 +74,8 @@ void hmp_x_colo_lost_heartbeat(Monitor *mon, const QDict *qdict);
+ void hmp_set_password(Monitor *mon, const QDict *qdict);
+ void hmp_expire_password(Monitor *mon, const QDict *qdict);
+ void hmp_change(Monitor *mon, const QDict *qdict);
++void hmp_backup(Monitor *mon, const QDict *qdict);
++void hmp_backup_cancel(Monitor *mon, const QDict *qdict);
+ void hmp_migrate(Monitor *mon, const QDict *qdict);
+ void hmp_device_add(Monitor *mon, const QDict *qdict);
+ void hmp_device_del(Monitor *mon, const QDict *qdict);
+diff --git a/meson.build b/meson.build
+index ffff66c0cc..0bc2fb5b10 100644
+--- a/meson.build
++++ b/meson.build
+@@ -1463,6 +1463,7 @@ keyutils = dependency('libkeyutils', required: false,
+ has_gettid = cc.has_function('gettid')
+
+ libuuid = cc.find_library('uuid', required: true)
++libproxmox_backup_qemu = cc.find_library('proxmox_backup_qemu', required: true)
+
+ # libselinux
+ selinux = dependency('libselinux',
+diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
+index 1507180990..1168773da7 100644
+--- a/monitor/hmp-cmds.c
++++ b/monitor/hmp-cmds.c
+@@ -197,6 +197,50 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
+ qapi_free_MouseInfoList(mice_list);
+ }
+
++void hmp_info_backup(Monitor *mon, const QDict *qdict)
++{
++ BackupStatus *info;
++
++ info = qmp_query_backup(NULL);
++
++ if (!info) {
++ monitor_printf(mon, "Backup status: not initialized\n");
++ return;
++ }
++
++ if (info->has_status) {
++ if (info->has_errmsg) {
++ monitor_printf(mon, "Backup status: %s - %s\n",
++ info->status, info->errmsg);
++ } else {
++ monitor_printf(mon, "Backup status: %s\n", info->status);
++ }
++ }
++
++ if (info->has_backup_file) {
++ monitor_printf(mon, "Start time: %s", ctime(&info->start_time));
++ if (info->end_time) {
++ monitor_printf(mon, "End time: %s", ctime(&info->end_time));
++ }
++
++ int per = (info->has_total && info->total &&
++ info->has_transferred && info->transferred) ?
++ (info->transferred * 100)/info->total : 0;
++ int zero_per = (info->has_total && info->total &&
++ info->has_zero_bytes && info->zero_bytes) ?
++ (info->zero_bytes * 100)/info->total : 0;
++ monitor_printf(mon, "Backup file: %s\n", info->backup_file);
++ monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
++ monitor_printf(mon, "Total size: %zd\n", info->total);
++ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
++ info->transferred, per);
++ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
++ info->zero_bytes, zero_per);
++ }
++
++ qapi_free_BackupStatus(info);
++}
++
+ static char *SocketAddress_to_str(SocketAddress *addr)
+ {
+ switch (addr->type) {
+diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
+new file mode 100644
+index 0000000000..a8f6653a81
+--- /dev/null
++++ b/proxmox-backup-client.c
+@@ -0,0 +1,176 @@
++#include "proxmox-backup-client.h"
++#include "qemu/main-loop.h"
++#include "block/aio-wait.h"
++#include "qapi/error.h"
++
++/* Proxmox Backup Server client bindings using coroutines */
++
++typedef struct BlockOnCoroutineWrapper {
++ AioContext *ctx;
++ CoroutineEntry *entry;
++ void *entry_arg;
++ bool finished;
++} BlockOnCoroutineWrapper;
++
++static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
++{
++ BlockOnCoroutineWrapper *wrapper = opaque;
++ wrapper->entry(wrapper->entry_arg);
++ wrapper->finished = true;
++ aio_wait_kick();
++}
++
++void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
++{
++ assert(!qemu_in_coroutine());
++
++ AioContext *ctx = qemu_get_current_aio_context();
++ BlockOnCoroutineWrapper wrapper = {
++ .finished = false,
++ .entry = entry,
++ .entry_arg = entry_arg,
++ .ctx = ctx,
++ };
++ Coroutine *wrapper_co = qemu_coroutine_create(block_on_coroutine_wrapper, &wrapper);
++ aio_co_enter(ctx, wrapper_co);
++ AIO_WAIT_WHILE(ctx, !wrapper.finished);
++}
++
++// This is called from another thread, so we use aio_co_schedule()
++static void proxmox_backup_schedule_wake(void *data) {
++ CoCtxData *waker = (CoCtxData *)data;
++ aio_co_schedule(waker->ctx, waker->co);
++}
++
++int coroutine_fn
++proxmox_backup_co_connect(ProxmoxBackupHandle *pbs, Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ CoCtxData waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_connect_async(pbs, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup connect failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_add_config(
++ ProxmoxBackupHandle *pbs,
++ const char *name,
++ const uint8_t *data,
++ uint64_t size,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ CoCtxData waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_add_config_async(
++ pbs, name, data, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup add_config %s failed: %s", name, pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_register_image(
++ ProxmoxBackupHandle *pbs,
++ const char *device_name,
++ uint64_t size,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ CoCtxData waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_register_image_async(
++ pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_finish(
++ ProxmoxBackupHandle *pbs,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ CoCtxData waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_finish_async(
++ pbs, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup finish failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_close_image(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ CoCtxData waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_close_image_async(
++ pbs, dev_id, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup close image failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_write_data(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ const uint8_t *data,
++ uint64_t offset,
++ uint64_t size,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ CoCtxData waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_write_data_async(
++ pbs, dev_id, data, offset, size, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup write data failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
+diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
+new file mode 100644
+index 0000000000..1dda8b7d8f
+--- /dev/null
++++ b/proxmox-backup-client.h
+@@ -0,0 +1,59 @@
++#ifndef PROXMOX_BACKUP_CLIENT_H
++#define PROXMOX_BACKUP_CLIENT_H
++
++#include "qemu/osdep.h"
++#include "qemu/coroutine.h"
++#include "proxmox-backup-qemu.h"
++
++typedef struct CoCtxData {
++ Coroutine *co;
++ AioContext *ctx;
++ void *data;
++} CoCtxData;
++
++// FIXME: Remove once coroutines are supported for QMP
++void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg);
++
++int coroutine_fn
++proxmox_backup_co_connect(
++ ProxmoxBackupHandle *pbs,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_add_config(
++ ProxmoxBackupHandle *pbs,
++ const char *name,
++ const uint8_t *data,
++ uint64_t size,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_register_image(
++ ProxmoxBackupHandle *pbs,
++ const char *device_name,
++ uint64_t size,
++ Error **errp);
++
++
++int coroutine_fn
++proxmox_backup_co_finish(
++ ProxmoxBackupHandle *pbs,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_close_image(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_write_data(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ const uint8_t *data,
++ uint64_t offset,
++ uint64_t size,
++ Error **errp);
++
++
++#endif /* PROXMOX_BACKUP_CLIENT_H */
+diff --git a/pve-backup.c b/pve-backup.c
+new file mode 100644
+index 0000000000..88f5ee133f
+--- /dev/null
++++ b/pve-backup.c
+@@ -0,0 +1,959 @@
++#include "proxmox-backup-client.h"
++#include "vma.h"
++
++#include "qemu/osdep.h"
++#include "qemu/module.h"
++#include "sysemu/block-backend.h"
++#include "sysemu/blockdev.h"
++#include "block/blockjob.h"
++#include "qapi/qapi-commands-block.h"
++#include "qapi/qmp/qerror.h"
++
++/* PVE backup state and related function */
++
++/*
++ * Note: A resume from a qemu_coroutine_yield can happen in a different thread,
++ * so you may not use normal mutexes within coroutines:
++ *
++ * ---bad-example---
++ * qemu_rec_mutex_lock(lock)
++ * ...
++ * qemu_coroutine_yield() // wait for something
++ * // we are now inside a different thread
++ * qemu_rec_mutex_unlock(lock) // Crash - wrong thread!!
++ * ---end-bad-example--
++ *
++ * ==> Always use CoMutext inside coroutines.
++ * ==> Never acquire/release AioContext withing coroutines (because that use QemuRecMutex)
++ *
++ */
++
++static struct PVEBackupState {
++ struct {
++ // Everithing accessed from qmp_backup_query command is protected using lock
++ QemuMutex lock;
++ Error *error;
++ time_t start_time;
++ time_t end_time;
++ char *backup_file;
++ uuid_t uuid;
++ char uuid_str[37];
++ size_t total;
++ size_t transferred;
++ size_t zero_bytes;
++ } stat;
++ int64_t speed;
++ VmaWriter *vmaw;
++ ProxmoxBackupHandle *pbs;
++ GList *di_list;
++ QemuMutex backup_mutex;
++ CoMutex dump_callback_mutex;
++} backup_state;
++
++static void pvebackup_init(void)
++{
++ qemu_mutex_init(&backup_state.stat.lock);
++ qemu_mutex_init(&backup_state.backup_mutex);
++ qemu_co_mutex_init(&backup_state.dump_callback_mutex);
++}
++
++// initialize PVEBackupState at startup
++opts_init(pvebackup_init);
++
++typedef struct PVEBackupDevInfo {
++ BlockDriverState *bs;
++ size_t size;
++ uint8_t dev_id;
++ bool completed;
++ char targetfile[PATH_MAX];
++ BlockDriverState *target;
++} PVEBackupDevInfo;
++
++static void pvebackup_run_next_job(void);
++
++static BlockJob *
++lookup_active_block_job(PVEBackupDevInfo *di)
++{
++ if (!di->completed && di->bs) {
++ for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
++ if (job->job.driver->job_type != JOB_TYPE_BACKUP) {
++ continue;
++ }
++
++ BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
++ if (bjob && bjob->source_bs == di->bs) {
++ return job;
++ }
++ }
++ }
++ return NULL;
++}
++
++static void pvebackup_propagate_error(Error *err)
++{
++ qemu_mutex_lock(&backup_state.stat.lock);
++ error_propagate(&backup_state.stat.error, err);
++ qemu_mutex_unlock(&backup_state.stat.lock);
++}
++
++static bool pvebackup_error_or_canceled(void)
++{
++ qemu_mutex_lock(&backup_state.stat.lock);
++ bool error_or_canceled = !!backup_state.stat.error;
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
++ return error_or_canceled;
++}
++
++static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes)
++{
++ qemu_mutex_lock(&backup_state.stat.lock);
++ backup_state.stat.zero_bytes += zero_bytes;
++ backup_state.stat.transferred += transferred;
++ qemu_mutex_unlock(&backup_state.stat.lock);
++}
++
++// This may get called from multiple coroutines in multiple io-threads
++// Note1: this may get called after job_cancel()
++static int coroutine_fn
++pvebackup_co_dump_pbs_cb(
++ void *opaque,
++ uint64_t start,
++ uint64_t bytes,
++ const void *pbuf)
++{
++ assert(qemu_in_coroutine());
++
++ const uint64_t size = bytes;
++ const unsigned char *buf = pbuf;
++ PVEBackupDevInfo *di = opaque;
++
++ assert(backup_state.pbs);
++
++ Error *local_err = NULL;
++ int pbs_res = -1;
++
++ qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
++
++ // avoid deadlock if job is cancelled
++ if (pvebackup_error_or_canceled()) {
++ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++ return -1;
++ }
++
++ pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id, buf, start, size, &local_err);
++ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++
++ if (pbs_res < 0) {
++ pvebackup_propagate_error(local_err);
++ return pbs_res;
++ } else {
++ pvebackup_add_transfered_bytes(size, !buf ? size : 0);
++ }
++
++ return size;
++}
++
++// This may get called from multiple coroutines in multiple io-threads
++static int coroutine_fn
++pvebackup_co_dump_vma_cb(
++ void *opaque,
++ uint64_t start,
++ uint64_t bytes,
++ const void *pbuf)
++{
++ assert(qemu_in_coroutine());
++
++ const uint64_t size = bytes;
++ const unsigned char *buf = pbuf;
++ PVEBackupDevInfo *di = opaque;
++
++ int ret = -1;
++
++ assert(backup_state.vmaw);
++
++ uint64_t remaining = size;
++
++ uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
++ if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
++ Error *local_err = NULL;
++ error_setg(&local_err,
++ "got unaligned write inside backup dump "
++ "callback (sector %ld)", start);
++ pvebackup_propagate_error(local_err);
++ return -1; // not aligned to cluster size
++ }
++
++ while (remaining > 0) {
++ qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
++ // avoid deadlock if job is cancelled
++ if (pvebackup_error_or_canceled()) {
++ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++ return -1;
++ }
++
++ size_t zero_bytes = 0;
++ ret = vma_writer_write(backup_state.vmaw, di->dev_id, cluster_num, buf, &zero_bytes);
++ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++
++ ++cluster_num;
++ if (buf) {
++ buf += VMA_CLUSTER_SIZE;
++ }
++ if (ret < 0) {
++ Error *local_err = NULL;
++ vma_writer_error_propagate(backup_state.vmaw, &local_err);
++ pvebackup_propagate_error(local_err);
++ return ret;
++ } else {
++ if (remaining >= VMA_CLUSTER_SIZE) {
++ assert(ret == VMA_CLUSTER_SIZE);
++ pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes);
++ remaining -= VMA_CLUSTER_SIZE;
++ } else {
++ assert(ret == remaining);
++ pvebackup_add_transfered_bytes(remaining, zero_bytes);
++ remaining = 0;
++ }
++ }
++ }
++
++ return size;
++}
++
++// assumes the caller holds backup_mutex
++static void coroutine_fn pvebackup_co_cleanup(void *unused)
++{
++ assert(qemu_in_coroutine());
++
++ qemu_mutex_lock(&backup_state.stat.lock);
++ backup_state.stat.end_time = time(NULL);
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
++ if (backup_state.vmaw) {
++ Error *local_err = NULL;
++ vma_writer_close(backup_state.vmaw, &local_err);
++
++ if (local_err != NULL) {
++ pvebackup_propagate_error(local_err);
++ }
++
++ backup_state.vmaw = NULL;
++ }
++
++ if (backup_state.pbs) {
++ if (!pvebackup_error_or_canceled()) {
++ Error *local_err = NULL;
++ proxmox_backup_co_finish(backup_state.pbs, &local_err);
++ if (local_err != NULL) {
++ pvebackup_propagate_error(local_err);
++ }
++ }
++
++ proxmox_backup_disconnect(backup_state.pbs);
++ backup_state.pbs = NULL;
++ }
++
++ g_list_free(backup_state.di_list);
++ backup_state.di_list = NULL;
++}
++
++// assumes the caller holds backup_mutex
++static void coroutine_fn pvebackup_complete_stream(void *opaque)
++{
++ PVEBackupDevInfo *di = opaque;
++
++ bool error_or_canceled = pvebackup_error_or_canceled();
++
++ if (backup_state.vmaw) {
++ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
++ }
++
++ if (backup_state.pbs && !error_or_canceled) {
++ Error *local_err = NULL;
++ proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
++ if (local_err != NULL) {
++ pvebackup_propagate_error(local_err);
++ }
++ }
++}
++
++static void pvebackup_complete_cb(void *opaque, int ret)
++{
++ assert(!qemu_in_coroutine());
++
++ PVEBackupDevInfo *di = opaque;
++
++ qemu_mutex_lock(&backup_state.backup_mutex);
++
++ di->completed = true;
++
++ if (ret < 0) {
++ Error *local_err = NULL;
++ error_setg(&local_err, "job failed with err %d - %s", ret, strerror(-ret));
++ pvebackup_propagate_error(local_err);
++ }
++
++ di->bs = NULL;
++
++ assert(di->target == NULL);
++
++ block_on_coroutine_fn(pvebackup_complete_stream, di);
++
++ // remove self from job queue
++ backup_state.di_list = g_list_remove(backup_state.di_list, di);
++
++ g_free(di);
++
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++
++ pvebackup_run_next_job();
++}
++
++static void pvebackup_cancel(void)
++{
++ assert(!qemu_in_coroutine());
++
++ Error *cancel_err = NULL;
++ error_setg(&cancel_err, "backup canceled");
++ pvebackup_propagate_error(cancel_err);
++
++ qemu_mutex_lock(&backup_state.backup_mutex);
++
++ if (backup_state.vmaw) {
++ /* make sure vma writer does not block anymore */
++ vma_writer_set_error(backup_state.vmaw, "backup canceled");
++ }
++
++ if (backup_state.pbs) {
++ proxmox_backup_abort(backup_state.pbs, "backup canceled");
++ }
++
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++
++ for(;;) {
++
++ BlockJob *next_job = NULL;
++
++ qemu_mutex_lock(&backup_state.backup_mutex);
++
++ GList *l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ BlockJob *job = lookup_active_block_job(di);
++ if (job != NULL) {
++ next_job = job;
++ break;
++ }
++ }
++
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++
++ if (next_job) {
++ AioContext *aio_context = next_job->job.aio_context;
++ aio_context_acquire(aio_context);
++ job_cancel_sync(&next_job->job, true);
++ aio_context_release(aio_context);
++ } else {
++ break;
++ }
++ }
++}
++
++void qmp_backup_cancel(Error **errp)
++{
++ pvebackup_cancel();
++}
++
++// assumes the caller holds backup_mutex
++static int coroutine_fn pvebackup_co_add_config(
++ const char *file,
++ const char *name,
++ BackupFormat format,
++ const char *backup_dir,
++ VmaWriter *vmaw,
++ ProxmoxBackupHandle *pbs,
++ Error **errp)
++{
++ int res = 0;
++
++ char *cdata = NULL;
++ gsize clen = 0;
++ GError *err = NULL;
++ if (!g_file_get_contents(file, &cdata, &clen, &err)) {
++ error_setg(errp, "unable to read file '%s'", file);
++ return 1;
++ }
++
++ char *basename = g_path_get_basename(file);
++ if (name == NULL) name = basename;
++
++ if (format == BACKUP_FORMAT_VMA) {
++ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
++ error_setg(errp, "unable to add %s config data to vma archive", file);
++ goto err;
++ }
++ } else if (format == BACKUP_FORMAT_PBS) {
++ if (proxmox_backup_co_add_config(pbs, name, (unsigned char *)cdata, clen, errp) < 0)
++ goto err;
++ } else if (format == BACKUP_FORMAT_DIR) {
++ char config_path[PATH_MAX];
++ snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, name);
++ if (!g_file_set_contents(config_path, cdata, clen, &err)) {
++ error_setg(errp, "unable to write config file '%s'", config_path);
++ goto err;
++ }
++ }
++
++ out:
++ g_free(basename);
++ g_free(cdata);
++ return res;
++
++ err:
++ res = -1;
++ goto out;
++}
++
++bool job_should_pause(Job *job);
++
++static void pvebackup_run_next_job(void)
++{
++ assert(!qemu_in_coroutine());
++
++ qemu_mutex_lock(&backup_state.backup_mutex);
++
++ GList *l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ BlockJob *job = lookup_active_block_job(di);
++
++ if (job) {
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++
++ AioContext *aio_context = job->job.aio_context;
++ aio_context_acquire(aio_context);
++
++ if (job_should_pause(&job->job)) {
++ bool error_or_canceled = pvebackup_error_or_canceled();
++ if (error_or_canceled) {
++ job_cancel_sync(&job->job, true);
++ } else {
++ job_resume(&job->job);
++ }
++ }
++ aio_context_release(aio_context);
++ return;
++ }
++ }
++
++ block_on_coroutine_fn(pvebackup_co_cleanup, NULL); // no more jobs, run cleanup
++
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++}
++
++static bool create_backup_jobs(void) {
++
++ assert(!qemu_in_coroutine());
++
++ Error *local_err = NULL;
++
++ BackupPerf perf = { .max_workers = 16 };
++
++ /* create and start all jobs (paused state) */
++ GList *l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ assert(di->target != NULL);
++
++ AioContext *aio_context = bdrv_get_aio_context(di->bs);
++ aio_context_acquire(aio_context);
++
++ BlockJob *job = backup_job_create(
++ NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
++ BITMAP_SYNC_MODE_NEVER, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
++ JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
++
++ aio_context_release(aio_context);
++
++ if (!job || local_err != NULL) {
++ Error *create_job_err = NULL;
++ error_setg(&create_job_err, "backup_job_create failed: %s",
++ local_err ? error_get_pretty(local_err) : "null");
++
++ pvebackup_propagate_error(create_job_err);
++ break;
++ }
++ job_start(&job->job);
++
++ bdrv_unref(di->target);
++ di->target = NULL;
++ }
++
++ bool errors = pvebackup_error_or_canceled();
++
++ if (errors) {
++ l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ if (di->target) {
++ bdrv_unref(di->target);
++ di->target = NULL;
++ }
++ }
++ }
++
++ return errors;
++}
++
++typedef struct QmpBackupTask {
++ const char *backup_file;
++ bool has_password;
++ const char *password;
++ bool has_keyfile;
++ const char *keyfile;
++ bool has_key_password;
++ const char *key_password;
++ bool has_backup_id;
++ const char *backup_id;
++ bool has_backup_time;
++ const char *fingerprint;
++ bool has_fingerprint;
++ int64_t backup_time;
++ bool has_format;
++ BackupFormat format;
++ bool has_config_file;
++ const char *config_file;
++ bool has_firewall_file;
++ const char *firewall_file;
++ bool has_devlist;
++ const char *devlist;
++ bool has_speed;
++ int64_t speed;
++ Error **errp;
++ UuidInfo *result;
++} QmpBackupTask;
++
++// assumes the caller holds backup_mutex
++static void coroutine_fn pvebackup_co_prepare(void *opaque)
++{
++ assert(qemu_in_coroutine());
++
++ QmpBackupTask *task = opaque;
++
++ task->result = NULL; // just to be sure
++
++ BlockBackend *blk;
++ BlockDriverState *bs = NULL;
++ const char *backup_dir = NULL;
++ Error *local_err = NULL;
++ uuid_t uuid;
++ VmaWriter *vmaw = NULL;
++ ProxmoxBackupHandle *pbs = NULL;
++ gchar **devs = NULL;
++ GList *di_list = NULL;
++ GList *l;
++ UuidInfo *uuid_info;
++
++ const char *config_name = "qemu-server.conf";
++ const char *firewall_name = "qemu-server.fw";
++
++ if (backup_state.di_list) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ "previous backup not finished");
++ return;
++ }
++
++ /* Todo: try to auto-detect format based on file name */
++ BackupFormat format = task->has_format ? task->format : BACKUP_FORMAT_VMA;
++
++ if (task->has_devlist) {
++ devs = g_strsplit_set(task->devlist, ",;:", -1);
++
++ gchar **d = devs;
++ while (d && *d) {
++ blk = blk_by_name(*d);
++ if (blk) {
++ bs = blk_bs(blk);
++ if (!bdrv_is_inserted(bs)) {
++ error_setg(task->errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
++ goto err;
++ }
++ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
++ di->bs = bs;
++ di_list = g_list_append(di_list, di);
++ } else {
++ error_set(task->errp, ERROR_CLASS_DEVICE_NOT_FOUND,
++ "Device '%s' not found", *d);
++ goto err;
++ }
++ d++;
++ }
++
++ } else {
++ BdrvNextIterator it;
++
++ bs = NULL;
++ for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
++ if (!bdrv_is_inserted(bs) || bdrv_is_read_only(bs)) {
++ continue;
++ }
++
++ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
++ di->bs = bs;
++ di_list = g_list_append(di_list, di);
++ }
++ }
++
++ if (!di_list) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
++ goto err;
++ }
++
++ size_t total = 0;
++
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, task->errp)) {
++ goto err;
++ }
++
++ ssize_t size = bdrv_getlength(di->bs);
++ if (size < 0) {
++ error_setg_errno(task->errp, -di->size, "bdrv_getlength failed");
++ goto err;
++ }
++ di->size = size;
++ total += size;
++ }
++
++ uuid_generate(uuid);
++
++ if (format == BACKUP_FORMAT_PBS) {
++ if (!task->has_password) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
++ goto err;
++ }
++ if (!task->has_backup_id) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
++ goto err;
++ }
++ if (!task->has_backup_time) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
++ goto err;
++ }
++
++ int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
++ firewall_name = "fw.conf";
++
++ char *pbs_err = NULL;
++ pbs = proxmox_backup_new(
++ task->backup_file,
++ task->backup_id,
++ task->backup_time,
++ dump_cb_block_size,
++ task->has_password ? task->password : NULL,
++ task->has_keyfile ? task->keyfile : NULL,
++ task->has_key_password ? task->key_password : NULL,
++ task->has_fingerprint ? task->fingerprint : NULL,
++ &pbs_err);
++
++ if (!pbs) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ "proxmox_backup_new failed: %s", pbs_err);
++ proxmox_backup_free_error(pbs_err);
++ goto err;
++ }
++
++ if (proxmox_backup_co_connect(pbs, task->errp) < 0)
++ goto err;
++
++ /* register all devices */
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ const char *devname = bdrv_get_device_name(di->bs);
++
++ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, task->errp);
++ if (dev_id < 0)
++ goto err;
++
++ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
++ goto err;
++ }
++
++ di->dev_id = dev_id;
++ }
++ } else if (format == BACKUP_FORMAT_VMA) {
++ vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
++ if (!vmaw) {
++ if (local_err) {
++ error_propagate(task->errp, local_err);
++ }
++ goto err;
++ }
++
++ /* register all devices for vma writer */
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
++ goto err;
++ }
++
++ const char *devname = bdrv_get_device_name(di->bs);
++ di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
++ if (di->dev_id <= 0) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ "register_stream failed");
++ goto err;
++ }
++ }
++ } else if (format == BACKUP_FORMAT_DIR) {
++ if (mkdir(task->backup_file, 0640) != 0) {
++ error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
++ task->backup_file);
++ goto err;
++ }
++ backup_dir = task->backup_file;
++
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ const char *devname = bdrv_get_device_name(di->bs);
++ snprintf(di->targetfile, PATH_MAX, "%s/%s.raw", backup_dir, devname);
++
++ int flags = BDRV_O_RDWR;
++ bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
++ di->size, flags, false, &local_err);
++ if (local_err) {
++ error_propagate(task->errp, local_err);
++ goto err;
++ }
++
++ di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
++ if (!di->target) {
++ error_propagate(task->errp, local_err);
++ goto err;
++ }
++ }
++ } else {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
++ goto err;
++ }
++
++
++ /* add configuration file to archive */
++ if (task->has_config_file) {
++ if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
++ vmaw, pbs, task->errp) != 0) {
++ goto err;
++ }
++ }
++
++ /* add firewall file to archive */
++ if (task->has_firewall_file) {
++ if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
++ vmaw, pbs, task->errp) != 0) {
++ goto err;
++ }
++ }
++ /* initialize global backup_state now */
++
++ qemu_mutex_lock(&backup_state.stat.lock);
++
++ if (backup_state.stat.error) {
++ error_free(backup_state.stat.error);
++ backup_state.stat.error = NULL;
++ }
++
++ backup_state.stat.start_time = time(NULL);
++ backup_state.stat.end_time = 0;
++
++ if (backup_state.stat.backup_file) {
++ g_free(backup_state.stat.backup_file);
++ }
++ backup_state.stat.backup_file = g_strdup(task->backup_file);
++
++ uuid_copy(backup_state.stat.uuid, uuid);
++ uuid_unparse_lower(uuid, backup_state.stat.uuid_str);
++ char *uuid_str = g_strdup(backup_state.stat.uuid_str);
++
++ backup_state.stat.total = total;
++ backup_state.stat.transferred = 0;
++ backup_state.stat.zero_bytes = 0;
++
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
++ backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
++
++ backup_state.vmaw = vmaw;
++ backup_state.pbs = pbs;
++
++ backup_state.di_list = di_list;
++
++ uuid_info = g_malloc0(sizeof(*uuid_info));
++ uuid_info->UUID = uuid_str;
++
++ task->result = uuid_info;
++ return;
++
++err:
++
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ if (di->target) {
++ bdrv_unref(di->target);
++ }
++
++ if (di->targetfile[0]) {
++ unlink(di->targetfile);
++ }
++ g_free(di);
++ }
++ g_list_free(di_list);
++
++ if (devs) {
++ g_strfreev(devs);
++ }
++
++ if (vmaw) {
++ Error *err = NULL;
++ vma_writer_close(vmaw, &err);
++ unlink(task->backup_file);
++ }
++
++ if (pbs) {
++ proxmox_backup_disconnect(pbs);
++ }
++
++ if (backup_dir) {
++ rmdir(backup_dir);
++ }
++
++ task->result = NULL;
++ return;
++}
++
++UuidInfo *qmp_backup(
++ const char *backup_file,
++ bool has_password, const char *password,
++ bool has_keyfile, const char *keyfile,
++ bool has_key_password, const char *key_password,
++ bool has_fingerprint, const char *fingerprint,
++ bool has_backup_id, const char *backup_id,
++ bool has_backup_time, int64_t backup_time,
++ bool has_format, BackupFormat format,
++ bool has_config_file, const char *config_file,
++ bool has_firewall_file, const char *firewall_file,
++ bool has_devlist, const char *devlist,
++ bool has_speed, int64_t speed, Error **errp)
++{
++ QmpBackupTask task = {
++ .backup_file = backup_file,
++ .has_password = has_password,
++ .password = password,
++ .has_key_password = has_key_password,
++ .key_password = key_password,
++ .has_fingerprint = has_fingerprint,
++ .fingerprint = fingerprint,
++ .has_backup_id = has_backup_id,
++ .backup_id = backup_id,
++ .has_backup_time = has_backup_time,
++ .backup_time = backup_time,
++ .has_format = has_format,
++ .format = format,
++ .has_config_file = has_config_file,
++ .config_file = config_file,
++ .has_firewall_file = has_firewall_file,
++ .firewall_file = firewall_file,
++ .has_devlist = has_devlist,
++ .devlist = devlist,
++ .has_speed = has_speed,
++ .speed = speed,
++ .errp = errp,
++ };
++
++ qemu_mutex_lock(&backup_state.backup_mutex);
++
++ block_on_coroutine_fn(pvebackup_co_prepare, &task);
++
++ if (*errp == NULL) {
++ create_backup_jobs();
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++ pvebackup_run_next_job();
++ } else {
++ qemu_mutex_unlock(&backup_state.backup_mutex);
++ }
++
++ return task.result;
++}
++
++BackupStatus *qmp_query_backup(Error **errp)
++{
++ BackupStatus *info = g_malloc0(sizeof(*info));
++
++ qemu_mutex_lock(&backup_state.stat.lock);
++
++ if (!backup_state.stat.start_time) {
++ /* not started, return {} */
++ qemu_mutex_unlock(&backup_state.stat.lock);
++ return info;
++ }
++
++ info->has_status = true;
++ info->has_start_time = true;
++ info->start_time = backup_state.stat.start_time;
++
++ if (backup_state.stat.backup_file) {
++ info->has_backup_file = true;
++ info->backup_file = g_strdup(backup_state.stat.backup_file);
++ }
++
++ info->has_uuid = true;
++ info->uuid = g_strdup(backup_state.stat.uuid_str);
++
++ if (backup_state.stat.end_time) {
++ if (backup_state.stat.error) {
++ info->status = g_strdup("error");
++ info->has_errmsg = true;
++ info->errmsg = g_strdup(error_get_pretty(backup_state.stat.error));
++ } else {
++ info->status = g_strdup("done");
++ }
++ info->has_end_time = true;
++ info->end_time = backup_state.stat.end_time;
++ } else {
++ info->status = g_strdup("active");
++ }
++
++ info->has_total = true;
++ info->total = backup_state.stat.total;
++ info->has_zero_bytes = true;
++ info->zero_bytes = backup_state.stat.zero_bytes;
++ info->has_transferred = true;
++ info->transferred = backup_state.stat.transferred;
++
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
++ return info;
++}
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index ddac91e8f6..90ad07b7ee 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -740,6 +740,115 @@
+ { 'command': 'query-block', 'returns': ['BlockInfo'],
+ 'allow-preconfig': true }
+
++##
++# @BackupStatus:
++#
++# Detailed backup status.
++#
++# @status: string describing the current backup status.
++# This can be 'active', 'done', 'error'. If this field is not
++# returned, no backup process has been initiated
++#
++# @errmsg: error message (only returned if status is 'error')
++#
++# @total: total amount of bytes involved in the backup process
++#
++# @transferred: amount of bytes already backed up.
++#
++# @zero-bytes: amount of 'zero' bytes detected.
++#
++# @start-time: time (epoch) when backup job started.
++#
++# @end-time: time (epoch) when backup job finished.
++#
++# @backup-file: backup file name
++#
++# @uuid: uuid for this backup job
++#
++##
++{ 'struct': 'BackupStatus',
++ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
++ '*transferred': 'int', '*zero-bytes': 'int',
++ '*start-time': 'int', '*end-time': 'int',
++ '*backup-file': 'str', '*uuid': 'str' } }
++
++##
++# @BackupFormat:
++#
++# An enumeration of supported backup formats.
++#
++# @vma: Proxmox vma backup format
++##
++{ 'enum': 'BackupFormat',
++ 'data': [ 'vma', 'dir', 'pbs' ] }
++
++##
++# @backup:
++#
++# Starts a VM backup.
++#
++# @backup-file: the backup file name
++#
++# @format: format of the backup file
++#
++# @config-file: a configuration file to include into
++# the backup archive.
++#
++# @speed: the maximum speed, in bytes per second
++#
++# @devlist: list of block device names (separated by ',', ';'
++# or ':'). By default the backup includes all writable block devices.
++#
++# @password: backup server passsword (required for format 'pbs')
++#
++# @keyfile: keyfile used for encryption (optional for format 'pbs')
++#
++# @key-password: password for keyfile (optional for format 'pbs')
++#
++# @fingerprint: server cert fingerprint (optional for format 'pbs')
++#
++# @backup-id: backup ID (required for format 'pbs')
++#
++# @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
++#
++# Returns: the uuid of the backup job
++#
++##
++{ 'command': 'backup', 'data': { 'backup-file': 'str',
++ '*password': 'str',
++ '*keyfile': 'str',
++ '*key-password': 'str',
++ '*fingerprint': 'str',
++ '*backup-id': 'str',
++ '*backup-time': 'int',
++ '*format': 'BackupFormat',
++ '*config-file': 'str',
++ '*firewall-file': 'str',
++ '*devlist': 'str', '*speed': 'int' },
++ 'returns': 'UuidInfo' }
++
++##
++# @query-backup:
++#
++# Returns information about current/last backup task.
++#
++# Returns: @BackupStatus
++#
++##
++{ 'command': 'query-backup', 'returns': 'BackupStatus' }
++
++##
++# @backup-cancel:
++#
++# Cancel the current executing backup process.
++#
++# Returns: nothing on success
++#
++# Notes: This command succeeds even if there is no backup process running.
++#
++##
++{ 'command': 'backup-cancel' }
++
+ ##
+ # @BlockDeviceTimedStats:
+ #
+diff --git a/qapi/common.json b/qapi/common.json
+index 356db3f670..aae8a3b682 100644
+--- a/qapi/common.json
++++ b/qapi/common.json
+@@ -206,3 +206,16 @@
+ ##
+ { 'struct': 'HumanReadableText',
+ 'data': { 'human-readable-text': 'str' } }
++
++##
++# @UuidInfo:
++#
++# Guest UUID information (Universally Unique Identifier).
++#
++# @UUID: the UUID of the guest
++#
++# Since: 0.14.0
++#
++# Notes: If no UUID was specified for the guest, a null UUID is returned.
++##
++{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
+diff --git a/qapi/machine.json b/qapi/machine.json
+index d868e4d31d..a63d9a078d 100644
+--- a/qapi/machine.json
++++ b/qapi/machine.json
+@@ -4,6 +4,8 @@
+ # This work is licensed under the terms of the GNU GPL, version 2 or later.
+ # See the COPYING file in the top-level directory.
+
++{ 'include': 'common.json' }
++
+ ##
+ # = Machines
+ ##
+@@ -226,19 +228,6 @@
+ ##
+ { 'command': 'query-target', 'returns': 'TargetInfo' }
+
+-##
+-# @UuidInfo:
+-#
+-# Guest UUID information (Universally Unique Identifier).
+-#
+-# @UUID: the UUID of the guest
+-#
+-# Since: 0.14
+-#
+-# Notes: If no UUID was specified for the guest, a null UUID is returned.
+-##
+-{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
+-
+ ##
+ # @query-uuid:
+ #
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 29 Jun 2020 11:06:03 +0200
-Subject: [PATCH] PVE-Backup: Add dirty-bitmap tracking for incremental backups
-
-Uses QEMU's existing MIRROR_SYNC_MODE_BITMAP and a dirty-bitmap on top
-of all backed-up drives. This will only execute the data-write callback
-for any changed chunks, the PBS rust code will reuse chunks from the
-previous index for everything it doesn't receive if reuse_index is true.
-
-On error or cancellation, remove all dirty bitmaps to ensure
-consistency.
-
-Add PBS/incremental specific information to query backup info QMP and
-HMP commands.
-
-Only supported for PBS backups.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/monitor/block-hmp-cmds.c | 1 +
- monitor/hmp-cmds.c | 45 ++++++++++----
- proxmox-backup-client.c | 3 +-
- proxmox-backup-client.h | 1 +
- pve-backup.c | 103 ++++++++++++++++++++++++++++++---
- qapi/block-core.json | 12 +++-
- 6 files changed, 142 insertions(+), 23 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index 89ca64444d..45da74d7a0 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1042,6 +1042,7 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
- false, NULL, // PBS fingerprint
- false, NULL, // PBS backup-id
- false, 0, // PBS backup-time
-+ false, false, // PBS incremental
- true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
- false, NULL, false, NULL, !!devlist,
- devlist, qdict_haskey(qdict, "speed"), speed, &error);
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index ef654221c4..c5c74ac1dc 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -221,19 +221,42 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
- monitor_printf(mon, "End time: %s", ctime(&info->end_time));
- }
-
-- int per = (info->has_total && info->total &&
-- info->has_transferred && info->transferred) ?
-- (info->transferred * 100)/info->total : 0;
-- int zero_per = (info->has_total && info->total &&
-- info->has_zero_bytes && info->zero_bytes) ?
-- (info->zero_bytes * 100)/info->total : 0;
- monitor_printf(mon, "Backup file: %s\n", info->backup_file);
- monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
-- monitor_printf(mon, "Total size: %zd\n", info->total);
-- monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-- info->transferred, per);
-- monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-- info->zero_bytes, zero_per);
-+
-+ if (!(info->has_total && info->total)) {
-+ // this should not happen normally
-+ monitor_printf(mon, "Total size: %d\n", 0);
-+ } else {
-+ bool incremental = false;
-+ size_t total_or_dirty = info->total;
-+ if (info->has_transferred) {
-+ if (info->has_dirty && info->dirty) {
-+ if (info->dirty < info->total) {
-+ total_or_dirty = info->dirty;
-+ incremental = true;
-+ }
-+ }
-+ }
-+
-+ int per = (info->transferred * 100)/total_or_dirty;
-+
-+ monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
-+
-+ int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
-+ (info->zero_bytes * 100)/info->total : 0;
-+ monitor_printf(mon, "Total size: %zd\n", info->total);
-+ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-+ info->transferred, per);
-+ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-+ info->zero_bytes, zero_per);
-+
-+ if (info->has_reused) {
-+ int reused_per = (info->reused * 100)/total_or_dirty;
-+ monitor_printf(mon, "Reused bytes: %zd (%d%%)\n",
-+ info->reused, reused_per);
-+ }
-+ }
- }
-
- qapi_free_BackupStatus(info);
-diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
-index a8f6653a81..4ce7bc0b5e 100644
---- a/proxmox-backup-client.c
-+++ b/proxmox-backup-client.c
-@@ -89,6 +89,7 @@ proxmox_backup_co_register_image(
- ProxmoxBackupHandle *pbs,
- const char *device_name,
- uint64_t size,
-+ bool incremental,
- Error **errp)
- {
- Coroutine *co = qemu_coroutine_self();
-@@ -98,7 +99,7 @@ proxmox_backup_co_register_image(
- int pbs_res = -1;
-
- proxmox_backup_register_image_async(
-- pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+ pbs, device_name, size, incremental, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
- qemu_coroutine_yield();
- if (pbs_res < 0) {
- if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
-diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
-index 1dda8b7d8f..8cbf645b2c 100644
---- a/proxmox-backup-client.h
-+++ b/proxmox-backup-client.h
-@@ -32,6 +32,7 @@ proxmox_backup_co_register_image(
- ProxmoxBackupHandle *pbs,
- const char *device_name,
- uint64_t size,
-+ bool incremental,
- Error **errp);
-
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 88f5ee133f..1c49cd178d 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -28,6 +28,8 @@
- *
- */
-
-+const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
-+
- static struct PVEBackupState {
- struct {
- // Everithing accessed from qmp_backup_query command is protected using lock
-@@ -39,7 +41,9 @@ static struct PVEBackupState {
- uuid_t uuid;
- char uuid_str[37];
- size_t total;
-+ size_t dirty;
- size_t transferred;
-+ size_t reused;
- size_t zero_bytes;
- } stat;
- int64_t speed;
-@@ -66,6 +70,7 @@ typedef struct PVEBackupDevInfo {
- uint8_t dev_id;
- bool completed;
- char targetfile[PATH_MAX];
-+ BdrvDirtyBitmap *bitmap;
- BlockDriverState *target;
- } PVEBackupDevInfo;
-
-@@ -105,11 +110,12 @@ static bool pvebackup_error_or_canceled(void)
- return error_or_canceled;
- }
-
--static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes)
-+static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes, size_t reused)
- {
- qemu_mutex_lock(&backup_state.stat.lock);
- backup_state.stat.zero_bytes += zero_bytes;
- backup_state.stat.transferred += transferred;
-+ backup_state.stat.reused += reused;
- qemu_mutex_unlock(&backup_state.stat.lock);
- }
-
-@@ -148,7 +154,8 @@ pvebackup_co_dump_pbs_cb(
- pvebackup_propagate_error(local_err);
- return pbs_res;
- } else {
-- pvebackup_add_transfered_bytes(size, !buf ? size : 0);
-+ size_t reused = (pbs_res == 0) ? size : 0;
-+ pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
- }
-
- return size;
-@@ -208,11 +215,11 @@ pvebackup_co_dump_vma_cb(
- } else {
- if (remaining >= VMA_CLUSTER_SIZE) {
- assert(ret == VMA_CLUSTER_SIZE);
-- pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes);
-+ pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes, 0);
- remaining -= VMA_CLUSTER_SIZE;
- } else {
- assert(ret == remaining);
-- pvebackup_add_transfered_bytes(remaining, zero_bytes);
-+ pvebackup_add_transfered_bytes(remaining, zero_bytes, 0);
- remaining = 0;
- }
- }
-@@ -248,6 +255,18 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
- if (local_err != NULL) {
- pvebackup_propagate_error(local_err);
- }
-+ } else {
-+ // on error or cancel we cannot ensure synchronization of dirty
-+ // bitmaps with backup server, so remove all and do full backup next
-+ GList *l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ if (di->bitmap) {
-+ bdrv_release_dirty_bitmap(di->bitmap);
-+ }
-+ }
- }
-
- proxmox_backup_disconnect(backup_state.pbs);
-@@ -303,6 +322,12 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- // remove self from job queue
- backup_state.di_list = g_list_remove(backup_state.di_list, di);
-
-+ if (di->bitmap && ret < 0) {
-+ // on error or cancel we cannot ensure synchronization of dirty
-+ // bitmaps with backup server, so remove all and do full backup next
-+ bdrv_release_dirty_bitmap(di->bitmap);
-+ }
-+
- g_free(di);
-
- qemu_mutex_unlock(&backup_state.backup_mutex);
-@@ -472,12 +497,18 @@ static bool create_backup_jobs(void) {
-
- assert(di->target != NULL);
-
-+ MirrorSyncMode sync_mode = MIRROR_SYNC_MODE_FULL;
-+ BitmapSyncMode bitmap_mode = BITMAP_SYNC_MODE_NEVER;
-+ if (di->bitmap) {
-+ sync_mode = MIRROR_SYNC_MODE_BITMAP;
-+ bitmap_mode = BITMAP_SYNC_MODE_ON_SUCCESS;
-+ }
- AioContext *aio_context = bdrv_get_aio_context(di->bs);
- aio_context_acquire(aio_context);
-
- BlockJob *job = backup_job_create(
-- NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
-- BITMAP_SYNC_MODE_NEVER, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-+ NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
-+ bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
- JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
-
- aio_context_release(aio_context);
-@@ -528,6 +559,8 @@ typedef struct QmpBackupTask {
- const char *fingerprint;
- bool has_fingerprint;
- int64_t backup_time;
-+ bool has_use_dirty_bitmap;
-+ bool use_dirty_bitmap;
- bool has_format;
- BackupFormat format;
- bool has_config_file;
-@@ -619,6 +652,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
-
- size_t total = 0;
-+ size_t dirty = 0;
-
- l = di_list;
- while (l) {
-@@ -656,6 +690,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
- firewall_name = "fw.conf";
-
-+ bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
-+
- char *pbs_err = NULL;
- pbs = proxmox_backup_new(
- task->backup_file,
-@@ -675,7 +711,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- goto err;
- }
-
-- if (proxmox_backup_co_connect(pbs, task->errp) < 0)
-+ int connect_result = proxmox_backup_co_connect(pbs, task->errp);
-+ if (connect_result < 0)
- goto err;
-
- /* register all devices */
-@@ -686,9 +723,40 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-
- const char *devname = bdrv_get_device_name(di->bs);
-
-- int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, task->errp);
-- if (dev_id < 0)
-+ BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
-+ bool expect_only_dirty = false;
-+
-+ if (use_dirty_bitmap) {
-+ if (bitmap == NULL) {
-+ bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
-+ if (!bitmap) {
-+ goto err;
-+ }
-+ } else {
-+ expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
-+ }
-+
-+ if (expect_only_dirty) {
-+ dirty += bdrv_get_dirty_count(bitmap);
-+ } else {
-+ /* mark entire bitmap as dirty to make full backup */
-+ bdrv_set_dirty_bitmap(bitmap, 0, di->size);
-+ dirty += di->size;
-+ }
-+ di->bitmap = bitmap;
-+ } else {
-+ dirty += di->size;
-+
-+ /* after a full backup the old dirty bitmap is invalid anyway */
-+ if (bitmap != NULL) {
-+ bdrv_release_dirty_bitmap(bitmap);
-+ }
-+ }
-+
-+ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
-+ if (dev_id < 0) {
- goto err;
-+ }
-
- if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
- goto err;
-@@ -697,6 +765,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- di->dev_id = dev_id;
- }
- } else if (format == BACKUP_FORMAT_VMA) {
-+ dirty = total;
-+
- vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
- if (!vmaw) {
- if (local_err) {
-@@ -724,6 +794,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
- }
- } else if (format == BACKUP_FORMAT_DIR) {
-+ dirty = total;
-+
- if (mkdir(task->backup_file, 0640) != 0) {
- error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
- task->backup_file);
-@@ -796,8 +868,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- char *uuid_str = g_strdup(backup_state.stat.uuid_str);
-
- backup_state.stat.total = total;
-+ backup_state.stat.dirty = dirty;
- backup_state.stat.transferred = 0;
- backup_state.stat.zero_bytes = 0;
-+ backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
-
- qemu_mutex_unlock(&backup_state.stat.lock);
-
-@@ -821,6 +895,10 @@ err:
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-
-+ if (di->bitmap) {
-+ bdrv_release_dirty_bitmap(di->bitmap);
-+ }
-+
- if (di->target) {
- bdrv_unref(di->target);
- }
-@@ -862,6 +940,7 @@ UuidInfo *qmp_backup(
- bool has_fingerprint, const char *fingerprint,
- bool has_backup_id, const char *backup_id,
- bool has_backup_time, int64_t backup_time,
-+ bool has_use_dirty_bitmap, bool use_dirty_bitmap,
- bool has_format, BackupFormat format,
- bool has_config_file, const char *config_file,
- bool has_firewall_file, const char *firewall_file,
-@@ -880,6 +959,8 @@ UuidInfo *qmp_backup(
- .backup_id = backup_id,
- .has_backup_time = has_backup_time,
- .backup_time = backup_time,
-+ .has_use_dirty_bitmap = has_use_dirty_bitmap,
-+ .use_dirty_bitmap = use_dirty_bitmap,
- .has_format = has_format,
- .format = format,
- .has_config_file = has_config_file,
-@@ -948,10 +1029,14 @@ BackupStatus *qmp_query_backup(Error **errp)
-
- info->has_total = true;
- info->total = backup_state.stat.total;
-+ info->has_dirty = true;
-+ info->dirty = backup_state.stat.dirty;
- info->has_zero_bytes = true;
- info->zero_bytes = backup_state.stat.zero_bytes;
- info->has_transferred = true;
- info->transferred = backup_state.stat.transferred;
-+ info->has_reused = true;
-+ info->reused = backup_state.stat.reused;
-
- qemu_mutex_unlock(&backup_state.stat.lock);
-
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index ffcc25367b..b332d8a8da 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -757,8 +757,13 @@
- #
- # @total: total amount of bytes involved in the backup process
- #
-+# @dirty: with incremental mode (PBS) this is the amount of bytes involved
-+# in the backup process which are marked dirty.
-+#
- # @transferred: amount of bytes already backed up.
- #
-+# @reused: amount of bytes reused due to deduplication.
-+#
- # @zero-bytes: amount of 'zero' bytes detected.
- #
- # @start-time: time (epoch) when backup job started.
-@@ -771,8 +776,8 @@
- #
- ##
- { 'struct': 'BackupStatus',
-- 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
-- '*transferred': 'int', '*zero-bytes': 'int',
-+ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
-+ '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
- '*start-time': 'int', '*end-time': 'int',
- '*backup-file': 'str', '*uuid': 'str' } }
-
-@@ -815,6 +820,8 @@
- #
- # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
- #
-+# @use-dirty-bitmap: use dirty bitmap to detect incremental changes since last job (optional for format 'pbs')
-+#
- # Returns: the uuid of the backup job
- #
- ##
-@@ -825,6 +832,7 @@
- '*fingerprint': 'str',
- '*backup-id': 'str',
- '*backup-time': 'int',
-+ '*use-dirty-bitmap': 'bool',
- '*format': 'BackupFormat',
- '*config-file': 'str',
- '*firewall-file': 'str',
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Mon, 6 Apr 2020 12:17:01 +0200
+Subject: [PATCH] PVE-Backup: pbs-restore - new command to restore from proxmox
+ backup server
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ meson.build | 4 +
+ pbs-restore.c | 223 ++++++++++++++++++++++++++++++++++++++++++++++++++
+ 2 files changed, 227 insertions(+)
+ create mode 100644 pbs-restore.c
+
+diff --git a/meson.build b/meson.build
+index 0bc2fb5b10..f48d2e0457 100644
+--- a/meson.build
++++ b/meson.build
+@@ -3613,6 +3613,10 @@ if have_tools
+ vma = executable('vma', files('vma.c', 'vma-reader.c') + genh,
+ dependencies: [authz, block, crypto, io, qom], install: true)
+
++ pbs_restore = executable('pbs-restore', files('pbs-restore.c') + genh,
++ dependencies: [authz, block, crypto, io, qom,
++ libproxmox_backup_qemu], install: true)
++
+ subdir('storage-daemon')
+ subdir('contrib/rdmacm-mux')
+ subdir('contrib/elf2dmp')
+diff --git a/pbs-restore.c b/pbs-restore.c
+new file mode 100644
+index 0000000000..2f834cf42e
+--- /dev/null
++++ b/pbs-restore.c
+@@ -0,0 +1,223 @@
++/*
++ * Qemu image restore helper for Proxmox Backup
++ *
++ * Copyright (C) 2019 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++#include <getopt.h>
++#include <string.h>
++
++#include "qemu/module.h"
++#include "qemu/error-report.h"
++#include "qemu/main-loop.h"
++#include "qemu/cutils.h"
++#include "qapi/error.h"
++#include "qapi/qmp/qdict.h"
++#include "sysemu/block-backend.h"
++
++#include <proxmox-backup-qemu.h>
++
++static void help(void)
++{
++ const char *help_msg =
++ "usage: pbs-restore [--repository <repo>] snapshot archive-name target [command options]\n"
++ ;
++
++ printf("%s", help_msg);
++ exit(1);
++}
++
++typedef struct CallbackData {
++ BlockBackend *target;
++ uint64_t last_offset;
++ bool skip_zero;
++} CallbackData;
++
++static int write_callback(
++ void *callback_data_ptr,
++ uint64_t offset,
++ const unsigned char *data,
++ uint64_t data_len)
++{
++ int res = -1;
++
++ CallbackData *callback_data = (CallbackData *)callback_data_ptr;
++
++ uint64_t last_offset = callback_data->last_offset;
++ if (offset > last_offset) callback_data->last_offset = offset;
++
++ if (data == NULL) {
++ if (callback_data->skip_zero && offset > last_offset) {
++ return 0;
++ }
++ res = blk_pwrite_zeroes(callback_data->target, offset, data_len, 0);
++ } else {
++ res = blk_pwrite(callback_data->target, offset, data_len, data, 0);
++ }
++
++ if (res < 0) {
++ fprintf(stderr, "blk_pwrite failed at offset %ld length %ld (%d) - %s\n", offset, data_len, res, strerror(-res));
++ return res;
++ }
++
++ return 0;
++}
++
++int main(int argc, char **argv)
++{
++ Error *main_loop_err = NULL;
++ const char *format = "raw";
++ const char *repository = NULL;
++ const char *keyfile = NULL;
++ int verbose = false;
++ bool skip_zero = false;
++
++ error_init(argv[0]);
++
++ for (;;) {
++ static const struct option long_options[] = {
++ {"help", no_argument, 0, 'h'},
++ {"skip-zero", no_argument, 0, 'S'},
++ {"verbose", no_argument, 0, 'v'},
++ {"format", required_argument, 0, 'f'},
++ {"repository", required_argument, 0, 'r'},
++ {"keyfile", required_argument, 0, 'k'},
++ {0, 0, 0, 0}
++ };
++ int c = getopt_long(argc, argv, "hvf:r:k:", long_options, NULL);
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case ':':
++ fprintf(stderr, "missing argument for option '%s'\n", argv[optind - 1]);
++ return -1;
++ case '?':
++ fprintf(stderr, "unrecognized option '%s'\n", argv[optind - 1]);
++ return -1;
++ case 'f':
++ format = g_strdup(argv[optind - 1]);
++ break;
++ case 'r':
++ repository = g_strdup(argv[optind - 1]);
++ break;
++ case 'k':
++ keyfile = g_strdup(argv[optind - 1]);
++ break;
++ case 'v':
++ verbose = true;
++ break;
++ case 'S':
++ skip_zero = true;
++ break;
++ case 'h':
++ help();
++ return 0;
++ }
++ }
++
++ if (optind >= argc - 2) {
++ fprintf(stderr, "missing arguments\n");
++ help();
++ return -1;
++ }
++
++ if (repository == NULL) {
++ repository = getenv("PBS_REPOSITORY");
++ }
++
++ if (repository == NULL) {
++ fprintf(stderr, "no repository specified\n");
++ help();
++ return -1;
++ }
++
++ char *snapshot = argv[optind++];
++ char *archive_name = argv[optind++];
++ char *target = argv[optind++];
++
++ const char *password = getenv("PBS_PASSWORD");
++ const char *fingerprint = getenv("PBS_FINGERPRINT");
++ const char *key_password = getenv("PBS_ENCRYPTION_PASSWORD");
++
++ if (qemu_init_main_loop(&main_loop_err)) {
++ g_error("%s", error_get_pretty(main_loop_err));
++ }
++
++ bdrv_init();
++ module_call_init(MODULE_INIT_QOM);
++
++ if (verbose) {
++ fprintf(stderr, "connecting to repository '%s'\n", repository);
++ }
++ char *pbs_error = NULL;
++ ProxmoxRestoreHandle *conn = proxmox_restore_new(
++ repository, snapshot, password, keyfile, key_password, fingerprint, &pbs_error);
++ if (conn == NULL) {
++ fprintf(stderr, "restore failed: %s\n", pbs_error);
++ return -1;
++ }
++
++ int res = proxmox_restore_connect(conn, &pbs_error);
++ if (res < 0 || pbs_error) {
++ fprintf(stderr, "restore failed (connection error): %s\n", pbs_error);
++ return -1;
++ }
++
++ QDict *options = qdict_new();
++
++ if (format) {
++ qdict_put_str(options, "driver", format);
++ }
++
++
++ if (verbose) {
++ fprintf(stderr, "open block backend for target '%s'\n", target);
++ }
++ Error *local_err = NULL;
++ int flags = BDRV_O_RDWR;
++ BlockBackend *blk = blk_new_open(target, NULL, options, flags, &local_err);
++ if (!blk) {
++ fprintf(stderr, "%s\n", error_get_pretty(local_err));
++ return -1;
++ }
++
++ CallbackData *callback_data = calloc(sizeof(CallbackData), 1);
++
++ callback_data->target = blk;
++ callback_data->skip_zero = skip_zero;
++ callback_data->last_offset = 0;
++
++ // blk_set_enable_write_cache(blk, !writethrough);
++
++ if (verbose) {
++ fprintf(stderr, "starting to restore snapshot '%s'\n", snapshot);
++ fflush(stderr); // ensure we do not get printed after the progress log
++ }
++ res = proxmox_restore_image(
++ conn,
++ archive_name,
++ write_callback,
++ callback_data,
++ &pbs_error,
++ verbose);
++
++ proxmox_restore_disconnect(conn);
++ blk_unref(blk);
++
++ if (res < 0) {
++ fprintf(stderr, "restore failed: %s\n", pbs_error);
++ return -1;
++ }
++
++ return 0;
++}
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Mon, 29 Jun 2020 11:06:03 +0200
+Subject: [PATCH] PVE-Backup: Add dirty-bitmap tracking for incremental backups
+
+Uses QEMU's existing MIRROR_SYNC_MODE_BITMAP and a dirty-bitmap on top
+of all backed-up drives. This will only execute the data-write callback
+for any changed chunks, the PBS rust code will reuse chunks from the
+previous index for everything it doesn't receive if reuse_index is true.
+
+On error or cancellation, remove all dirty bitmaps to ensure
+consistency.
+
+Add PBS/incremental specific information to query backup info QMP and
+HMP commands.
+
+Only supported for PBS backups.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/monitor/block-hmp-cmds.c | 1 +
+ monitor/hmp-cmds.c | 45 ++++++++++----
+ proxmox-backup-client.c | 3 +-
+ proxmox-backup-client.h | 1 +
+ pve-backup.c | 103 ++++++++++++++++++++++++++++++---
+ qapi/block-core.json | 12 +++-
+ 6 files changed, 142 insertions(+), 23 deletions(-)
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index 89ca64444d..45da74d7a0 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1042,6 +1042,7 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
+ false, NULL, // PBS fingerprint
+ false, NULL, // PBS backup-id
+ false, 0, // PBS backup-time
++ false, false, // PBS incremental
+ true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
+ false, NULL, false, NULL, !!devlist,
+ devlist, qdict_haskey(qdict, "speed"), speed, &error);
+diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
+index 1168773da7..4c1671e289 100644
+--- a/monitor/hmp-cmds.c
++++ b/monitor/hmp-cmds.c
+@@ -223,19 +223,42 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
+ monitor_printf(mon, "End time: %s", ctime(&info->end_time));
+ }
+
+- int per = (info->has_total && info->total &&
+- info->has_transferred && info->transferred) ?
+- (info->transferred * 100)/info->total : 0;
+- int zero_per = (info->has_total && info->total &&
+- info->has_zero_bytes && info->zero_bytes) ?
+- (info->zero_bytes * 100)/info->total : 0;
+ monitor_printf(mon, "Backup file: %s\n", info->backup_file);
+ monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
+- monitor_printf(mon, "Total size: %zd\n", info->total);
+- monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
+- info->transferred, per);
+- monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
+- info->zero_bytes, zero_per);
++
++ if (!(info->has_total && info->total)) {
++ // this should not happen normally
++ monitor_printf(mon, "Total size: %d\n", 0);
++ } else {
++ bool incremental = false;
++ size_t total_or_dirty = info->total;
++ if (info->has_transferred) {
++ if (info->has_dirty && info->dirty) {
++ if (info->dirty < info->total) {
++ total_or_dirty = info->dirty;
++ incremental = true;
++ }
++ }
++ }
++
++ int per = (info->transferred * 100)/total_or_dirty;
++
++ monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
++
++ int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
++ (info->zero_bytes * 100)/info->total : 0;
++ monitor_printf(mon, "Total size: %zd\n", info->total);
++ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
++ info->transferred, per);
++ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
++ info->zero_bytes, zero_per);
++
++ if (info->has_reused) {
++ int reused_per = (info->reused * 100)/total_or_dirty;
++ monitor_printf(mon, "Reused bytes: %zd (%d%%)\n",
++ info->reused, reused_per);
++ }
++ }
+ }
+
+ qapi_free_BackupStatus(info);
+diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
+index a8f6653a81..4ce7bc0b5e 100644
+--- a/proxmox-backup-client.c
++++ b/proxmox-backup-client.c
+@@ -89,6 +89,7 @@ proxmox_backup_co_register_image(
+ ProxmoxBackupHandle *pbs,
+ const char *device_name,
+ uint64_t size,
++ bool incremental,
+ Error **errp)
+ {
+ Coroutine *co = qemu_coroutine_self();
+@@ -98,7 +99,7 @@ proxmox_backup_co_register_image(
+ int pbs_res = -1;
+
+ proxmox_backup_register_image_async(
+- pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ pbs, device_name, size, incremental, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
+ qemu_coroutine_yield();
+ if (pbs_res < 0) {
+ if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
+diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
+index 1dda8b7d8f..8cbf645b2c 100644
+--- a/proxmox-backup-client.h
++++ b/proxmox-backup-client.h
+@@ -32,6 +32,7 @@ proxmox_backup_co_register_image(
+ ProxmoxBackupHandle *pbs,
+ const char *device_name,
+ uint64_t size,
++ bool incremental,
+ Error **errp);
+
+
+diff --git a/pve-backup.c b/pve-backup.c
+index 88f5ee133f..1c49cd178d 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -28,6 +28,8 @@
+ *
+ */
+
++const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
++
+ static struct PVEBackupState {
+ struct {
+ // Everithing accessed from qmp_backup_query command is protected using lock
+@@ -39,7 +41,9 @@ static struct PVEBackupState {
+ uuid_t uuid;
+ char uuid_str[37];
+ size_t total;
++ size_t dirty;
+ size_t transferred;
++ size_t reused;
+ size_t zero_bytes;
+ } stat;
+ int64_t speed;
+@@ -66,6 +70,7 @@ typedef struct PVEBackupDevInfo {
+ uint8_t dev_id;
+ bool completed;
+ char targetfile[PATH_MAX];
++ BdrvDirtyBitmap *bitmap;
+ BlockDriverState *target;
+ } PVEBackupDevInfo;
+
+@@ -105,11 +110,12 @@ static bool pvebackup_error_or_canceled(void)
+ return error_or_canceled;
+ }
+
+-static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes)
++static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes, size_t reused)
+ {
+ qemu_mutex_lock(&backup_state.stat.lock);
+ backup_state.stat.zero_bytes += zero_bytes;
+ backup_state.stat.transferred += transferred;
++ backup_state.stat.reused += reused;
+ qemu_mutex_unlock(&backup_state.stat.lock);
+ }
+
+@@ -148,7 +154,8 @@ pvebackup_co_dump_pbs_cb(
+ pvebackup_propagate_error(local_err);
+ return pbs_res;
+ } else {
+- pvebackup_add_transfered_bytes(size, !buf ? size : 0);
++ size_t reused = (pbs_res == 0) ? size : 0;
++ pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
+ }
+
+ return size;
+@@ -208,11 +215,11 @@ pvebackup_co_dump_vma_cb(
+ } else {
+ if (remaining >= VMA_CLUSTER_SIZE) {
+ assert(ret == VMA_CLUSTER_SIZE);
+- pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes);
++ pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes, 0);
+ remaining -= VMA_CLUSTER_SIZE;
+ } else {
+ assert(ret == remaining);
+- pvebackup_add_transfered_bytes(remaining, zero_bytes);
++ pvebackup_add_transfered_bytes(remaining, zero_bytes, 0);
+ remaining = 0;
+ }
+ }
+@@ -248,6 +255,18 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
+ if (local_err != NULL) {
+ pvebackup_propagate_error(local_err);
+ }
++ } else {
++ // on error or cancel we cannot ensure synchronization of dirty
++ // bitmaps with backup server, so remove all and do full backup next
++ GList *l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ if (di->bitmap) {
++ bdrv_release_dirty_bitmap(di->bitmap);
++ }
++ }
+ }
+
+ proxmox_backup_disconnect(backup_state.pbs);
+@@ -303,6 +322,12 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+ // remove self from job queue
+ backup_state.di_list = g_list_remove(backup_state.di_list, di);
+
++ if (di->bitmap && ret < 0) {
++ // on error or cancel we cannot ensure synchronization of dirty
++ // bitmaps with backup server, so remove all and do full backup next
++ bdrv_release_dirty_bitmap(di->bitmap);
++ }
++
+ g_free(di);
+
+ qemu_mutex_unlock(&backup_state.backup_mutex);
+@@ -472,12 +497,18 @@ static bool create_backup_jobs(void) {
+
+ assert(di->target != NULL);
+
++ MirrorSyncMode sync_mode = MIRROR_SYNC_MODE_FULL;
++ BitmapSyncMode bitmap_mode = BITMAP_SYNC_MODE_NEVER;
++ if (di->bitmap) {
++ sync_mode = MIRROR_SYNC_MODE_BITMAP;
++ bitmap_mode = BITMAP_SYNC_MODE_ON_SUCCESS;
++ }
+ AioContext *aio_context = bdrv_get_aio_context(di->bs);
+ aio_context_acquire(aio_context);
+
+ BlockJob *job = backup_job_create(
+- NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
+- BITMAP_SYNC_MODE_NEVER, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
++ NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
++ bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
+ JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
+
+ aio_context_release(aio_context);
+@@ -528,6 +559,8 @@ typedef struct QmpBackupTask {
+ const char *fingerprint;
+ bool has_fingerprint;
+ int64_t backup_time;
++ bool has_use_dirty_bitmap;
++ bool use_dirty_bitmap;
+ bool has_format;
+ BackupFormat format;
+ bool has_config_file;
+@@ -619,6 +652,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+
+ size_t total = 0;
++ size_t dirty = 0;
+
+ l = di_list;
+ while (l) {
+@@ -656,6 +690,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
+ firewall_name = "fw.conf";
+
++ bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
++
+ char *pbs_err = NULL;
+ pbs = proxmox_backup_new(
+ task->backup_file,
+@@ -675,7 +711,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ goto err;
+ }
+
+- if (proxmox_backup_co_connect(pbs, task->errp) < 0)
++ int connect_result = proxmox_backup_co_connect(pbs, task->errp);
++ if (connect_result < 0)
+ goto err;
+
+ /* register all devices */
+@@ -686,9 +723,40 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+
+ const char *devname = bdrv_get_device_name(di->bs);
+
+- int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, task->errp);
+- if (dev_id < 0)
++ BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
++ bool expect_only_dirty = false;
++
++ if (use_dirty_bitmap) {
++ if (bitmap == NULL) {
++ bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
++ if (!bitmap) {
++ goto err;
++ }
++ } else {
++ expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
++ }
++
++ if (expect_only_dirty) {
++ dirty += bdrv_get_dirty_count(bitmap);
++ } else {
++ /* mark entire bitmap as dirty to make full backup */
++ bdrv_set_dirty_bitmap(bitmap, 0, di->size);
++ dirty += di->size;
++ }
++ di->bitmap = bitmap;
++ } else {
++ dirty += di->size;
++
++ /* after a full backup the old dirty bitmap is invalid anyway */
++ if (bitmap != NULL) {
++ bdrv_release_dirty_bitmap(bitmap);
++ }
++ }
++
++ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
++ if (dev_id < 0) {
+ goto err;
++ }
+
+ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
+ goto err;
+@@ -697,6 +765,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ di->dev_id = dev_id;
+ }
+ } else if (format == BACKUP_FORMAT_VMA) {
++ dirty = total;
++
+ vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
+ if (!vmaw) {
+ if (local_err) {
+@@ -724,6 +794,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+ }
+ } else if (format == BACKUP_FORMAT_DIR) {
++ dirty = total;
++
+ if (mkdir(task->backup_file, 0640) != 0) {
+ error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
+ task->backup_file);
+@@ -796,8 +868,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ char *uuid_str = g_strdup(backup_state.stat.uuid_str);
+
+ backup_state.stat.total = total;
++ backup_state.stat.dirty = dirty;
+ backup_state.stat.transferred = 0;
+ backup_state.stat.zero_bytes = 0;
++ backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
+
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+@@ -821,6 +895,10 @@ err:
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+ l = g_list_next(l);
+
++ if (di->bitmap) {
++ bdrv_release_dirty_bitmap(di->bitmap);
++ }
++
+ if (di->target) {
+ bdrv_unref(di->target);
+ }
+@@ -862,6 +940,7 @@ UuidInfo *qmp_backup(
+ bool has_fingerprint, const char *fingerprint,
+ bool has_backup_id, const char *backup_id,
+ bool has_backup_time, int64_t backup_time,
++ bool has_use_dirty_bitmap, bool use_dirty_bitmap,
+ bool has_format, BackupFormat format,
+ bool has_config_file, const char *config_file,
+ bool has_firewall_file, const char *firewall_file,
+@@ -880,6 +959,8 @@ UuidInfo *qmp_backup(
+ .backup_id = backup_id,
+ .has_backup_time = has_backup_time,
+ .backup_time = backup_time,
++ .has_use_dirty_bitmap = has_use_dirty_bitmap,
++ .use_dirty_bitmap = use_dirty_bitmap,
+ .has_format = has_format,
+ .format = format,
+ .has_config_file = has_config_file,
+@@ -948,10 +1029,14 @@ BackupStatus *qmp_query_backup(Error **errp)
+
+ info->has_total = true;
+ info->total = backup_state.stat.total;
++ info->has_dirty = true;
++ info->dirty = backup_state.stat.dirty;
+ info->has_zero_bytes = true;
+ info->zero_bytes = backup_state.stat.zero_bytes;
+ info->has_transferred = true;
+ info->transferred = backup_state.stat.transferred;
++ info->has_reused = true;
++ info->reused = backup_state.stat.reused;
+
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 90ad07b7ee..3ad9eb5d1a 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -753,8 +753,13 @@
+ #
+ # @total: total amount of bytes involved in the backup process
+ #
++# @dirty: with incremental mode (PBS) this is the amount of bytes involved
++# in the backup process which are marked dirty.
++#
+ # @transferred: amount of bytes already backed up.
+ #
++# @reused: amount of bytes reused due to deduplication.
++#
+ # @zero-bytes: amount of 'zero' bytes detected.
+ #
+ # @start-time: time (epoch) when backup job started.
+@@ -767,8 +772,8 @@
+ #
+ ##
+ { 'struct': 'BackupStatus',
+- 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
+- '*transferred': 'int', '*zero-bytes': 'int',
++ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
++ '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
+ '*start-time': 'int', '*end-time': 'int',
+ '*backup-file': 'str', '*uuid': 'str' } }
+
+@@ -811,6 +816,8 @@
+ #
+ # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
+ #
++# @use-dirty-bitmap: use dirty bitmap to detect incremental changes since last job (optional for format 'pbs')
++#
+ # Returns: the uuid of the backup job
+ #
+ ##
+@@ -821,6 +828,7 @@
+ '*fingerprint': 'str',
+ '*backup-id': 'str',
+ '*backup-time': 'int',
++ '*use-dirty-bitmap': 'bool',
+ '*format': 'BackupFormat',
+ '*config-file': 'str',
+ '*firewall-file': 'str',
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Thu, 9 Jul 2020 12:53:08 +0200
-Subject: [PATCH] PVE: various PBS fixes
-
-pbs: fix crypt and compress parameters
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-
-PVE: handle PBS write callback with big blocks correctly
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-
-PVE: add zero block handling to PBS dump callback
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/monitor/block-hmp-cmds.c | 4 ++-
- pve-backup.c | 57 +++++++++++++++++++++++++++-------
- qapi/block-core.json | 6 ++++
- 3 files changed, 54 insertions(+), 13 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index 45da74d7a0..ea7b665aa2 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1042,7 +1042,9 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
- false, NULL, // PBS fingerprint
- false, NULL, // PBS backup-id
- false, 0, // PBS backup-time
-- false, false, // PBS incremental
-+ false, false, // PBS use-dirty-bitmap
-+ false, false, // PBS compress
-+ false, false, // PBS encrypt
- true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
- false, NULL, false, NULL, !!devlist,
- devlist, qdict_haskey(qdict, "speed"), speed, &error);
-diff --git a/pve-backup.c b/pve-backup.c
-index 1c49cd178d..c15abefdda 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -8,6 +8,7 @@
- #include "block/blockjob.h"
- #include "qapi/qapi-commands-block.h"
- #include "qapi/qmp/qerror.h"
-+#include "qemu/cutils.h"
-
- /* PVE backup state and related function */
-
-@@ -67,6 +68,7 @@ opts_init(pvebackup_init);
- typedef struct PVEBackupDevInfo {
- BlockDriverState *bs;
- size_t size;
-+ uint64_t block_size;
- uint8_t dev_id;
- bool completed;
- char targetfile[PATH_MAX];
-@@ -135,10 +137,13 @@ pvebackup_co_dump_pbs_cb(
- PVEBackupDevInfo *di = opaque;
-
- assert(backup_state.pbs);
-+ assert(buf);
-
- Error *local_err = NULL;
- int pbs_res = -1;
-
-+ bool is_zero_block = size == di->block_size && buffer_is_zero(buf, size);
-+
- qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
-
- // avoid deadlock if job is cancelled
-@@ -147,17 +152,29 @@ pvebackup_co_dump_pbs_cb(
- return -1;
- }
-
-- pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id, buf, start, size, &local_err);
-- qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+ uint64_t transferred = 0;
-+ uint64_t reused = 0;
-+ while (transferred < size) {
-+ uint64_t left = size - transferred;
-+ uint64_t to_transfer = left < di->block_size ? left : di->block_size;
-
-- if (pbs_res < 0) {
-- pvebackup_propagate_error(local_err);
-- return pbs_res;
-- } else {
-- size_t reused = (pbs_res == 0) ? size : 0;
-- pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
-+ pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id,
-+ is_zero_block ? NULL : buf + transferred, start + transferred,
-+ to_transfer, &local_err);
-+ transferred += to_transfer;
-+
-+ if (pbs_res < 0) {
-+ pvebackup_propagate_error(local_err);
-+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+ return pbs_res;
-+ }
-+
-+ reused += pbs_res == 0 ? to_transfer : 0;
- }
-
-+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+ pvebackup_add_transfered_bytes(size, is_zero_block ? size : 0, reused);
-+
- return size;
- }
-
-@@ -178,6 +195,7 @@ pvebackup_co_dump_vma_cb(
- int ret = -1;
-
- assert(backup_state.vmaw);
-+ assert(buf);
-
- uint64_t remaining = size;
-
-@@ -204,9 +222,7 @@ pvebackup_co_dump_vma_cb(
- qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-
- ++cluster_num;
-- if (buf) {
-- buf += VMA_CLUSTER_SIZE;
-- }
-+ buf += VMA_CLUSTER_SIZE;
- if (ret < 0) {
- Error *local_err = NULL;
- vma_writer_error_propagate(backup_state.vmaw, &local_err);
-@@ -569,6 +585,10 @@ typedef struct QmpBackupTask {
- const char *firewall_file;
- bool has_devlist;
- const char *devlist;
-+ bool has_compress;
-+ bool compress;
-+ bool has_encrypt;
-+ bool encrypt;
- bool has_speed;
- int64_t speed;
- Error **errp;
-@@ -692,6 +712,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-
- bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
-
-+
- char *pbs_err = NULL;
- pbs = proxmox_backup_new(
- task->backup_file,
-@@ -701,8 +722,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- task->has_password ? task->password : NULL,
- task->has_keyfile ? task->keyfile : NULL,
- task->has_key_password ? task->key_password : NULL,
-+ task->has_compress ? task->compress : true,
-+ task->has_encrypt ? task->encrypt : task->has_keyfile,
- task->has_fingerprint ? task->fingerprint : NULL,
-- &pbs_err);
-+ &pbs_err);
-
- if (!pbs) {
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-@@ -721,6 +744,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-
-+ di->block_size = dump_cb_block_size;
-+
- const char *devname = bdrv_get_device_name(di->bs);
-
- BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
-@@ -941,6 +966,8 @@ UuidInfo *qmp_backup(
- bool has_backup_id, const char *backup_id,
- bool has_backup_time, int64_t backup_time,
- bool has_use_dirty_bitmap, bool use_dirty_bitmap,
-+ bool has_compress, bool compress,
-+ bool has_encrypt, bool encrypt,
- bool has_format, BackupFormat format,
- bool has_config_file, const char *config_file,
- bool has_firewall_file, const char *firewall_file,
-@@ -951,6 +978,8 @@ UuidInfo *qmp_backup(
- .backup_file = backup_file,
- .has_password = has_password,
- .password = password,
-+ .has_keyfile = has_keyfile,
-+ .keyfile = keyfile,
- .has_key_password = has_key_password,
- .key_password = key_password,
- .has_fingerprint = has_fingerprint,
-@@ -961,6 +990,10 @@ UuidInfo *qmp_backup(
- .backup_time = backup_time,
- .has_use_dirty_bitmap = has_use_dirty_bitmap,
- .use_dirty_bitmap = use_dirty_bitmap,
-+ .has_compress = has_compress,
-+ .compress = compress,
-+ .has_encrypt = has_encrypt,
-+ .encrypt = encrypt,
- .has_format = has_format,
- .format = format,
- .has_config_file = has_config_file,
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index b332d8a8da..abef91062a 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -822,6 +822,10 @@
- #
- # @use-dirty-bitmap: use dirty bitmap to detect incremental changes since last job (optional for format 'pbs')
- #
-+# @compress: use compression (optional for format 'pbs', defaults to true)
-+#
-+# @encrypt: use encryption ((optional for format 'pbs', defaults to true if there is a keyfile)
-+#
- # Returns: the uuid of the backup job
- #
- ##
-@@ -833,6 +837,8 @@
- '*backup-id': 'str',
- '*backup-time': 'int',
- '*use-dirty-bitmap': 'bool',
-+ '*compress': 'bool',
-+ '*encrypt': 'bool',
- '*format': 'BackupFormat',
- '*config-file': 'str',
- '*firewall-file': 'str',
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 8 Jul 2020 09:50:54 +0200
-Subject: [PATCH] PVE: Add PBS block driver to map backup archives into VMs
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-[error cleanups, file_open implementation]
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[adapt to changed function signatures]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/meson.build | 3 +
- block/pbs.c | 276 +++++++++++++++++++++++++++++++++++++++++++
- configure | 9 ++
- meson.build | 1 +
- qapi/block-core.json | 13 ++
- 5 files changed, 302 insertions(+)
- create mode 100644 block/pbs.c
-
-diff --git a/block/meson.build b/block/meson.build
-index 42aac96dbb..f94cc0cd25 100644
---- a/block/meson.build
-+++ b/block/meson.build
-@@ -53,6 +53,9 @@ block_ss.add(files(
- '../pve-backup.c',
- ), libproxmox_backup_qemu)
-
-+block_ss.add(when: 'CONFIG_PBS_BDRV', if_true: files('pbs.c'))
-+block_ss.add(when: 'CONFIG_PBS_BDRV', if_true: libproxmox_backup_qemu)
-+
-
- softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
-
-diff --git a/block/pbs.c b/block/pbs.c
-new file mode 100644
-index 0000000000..0b05ea9080
---- /dev/null
-+++ b/block/pbs.c
-@@ -0,0 +1,276 @@
-+/*
-+ * Proxmox Backup Server read-only block driver
-+ */
-+
-+#include "qemu/osdep.h"
-+#include "qapi/error.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qstring.h"
-+#include "qemu/module.h"
-+#include "qemu/option.h"
-+#include "qemu/cutils.h"
-+#include "block/block_int.h"
-+
-+#include <proxmox-backup-qemu.h>
-+
-+#define PBS_OPT_REPOSITORY "repository"
-+#define PBS_OPT_SNAPSHOT "snapshot"
-+#define PBS_OPT_ARCHIVE "archive"
-+#define PBS_OPT_KEYFILE "keyfile"
-+#define PBS_OPT_PASSWORD "password"
-+#define PBS_OPT_FINGERPRINT "fingerprint"
-+#define PBS_OPT_ENCRYPTION_PASSWORD "key_password"
-+
-+typedef struct {
-+ ProxmoxRestoreHandle *conn;
-+ char aid;
-+ int64_t length;
-+
-+ char *repository;
-+ char *snapshot;
-+ char *archive;
-+} BDRVPBSState;
-+
-+static QemuOptsList runtime_opts = {
-+ .name = "pbs",
-+ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
-+ .desc = {
-+ {
-+ .name = PBS_OPT_REPOSITORY,
-+ .type = QEMU_OPT_STRING,
-+ .help = "The server address and repository to connect to.",
-+ },
-+ {
-+ .name = PBS_OPT_SNAPSHOT,
-+ .type = QEMU_OPT_STRING,
-+ .help = "The snapshot to read.",
-+ },
-+ {
-+ .name = PBS_OPT_ARCHIVE,
-+ .type = QEMU_OPT_STRING,
-+ .help = "Which archive within the snapshot should be accessed.",
-+ },
-+ {
-+ .name = PBS_OPT_PASSWORD,
-+ .type = QEMU_OPT_STRING,
-+ .help = "Server password. Can be passed as env var 'PBS_PASSWORD'.",
-+ },
-+ {
-+ .name = PBS_OPT_FINGERPRINT,
-+ .type = QEMU_OPT_STRING,
-+ .help = "Server fingerprint. Can be passed as env var 'PBS_FINGERPRINT'.",
-+ },
-+ {
-+ .name = PBS_OPT_ENCRYPTION_PASSWORD,
-+ .type = QEMU_OPT_STRING,
-+ .help = "Optional: Key password. Can be passed as env var 'PBS_ENCRYPTION_PASSWORD'.",
-+ },
-+ {
-+ .name = PBS_OPT_KEYFILE,
-+ .type = QEMU_OPT_STRING,
-+ .help = "Optional: The path to the keyfile to use.",
-+ },
-+ { /* end of list */ }
-+ },
-+};
-+
-+
-+// filename format:
-+// pbs:repository=<repo>,snapshot=<snap>,password=<pw>,key_password=<kpw>,fingerprint=<fp>,archive=<archive>
-+static void pbs_parse_filename(const char *filename, QDict *options,
-+ Error **errp)
-+{
-+
-+ if (!strstart(filename, "pbs:", &filename)) {
-+ if (errp) error_setg(errp, "pbs_parse_filename failed - missing 'pbs:' prefix");
-+ }
-+
-+
-+ QemuOpts *opts = qemu_opts_parse_noisily(&runtime_opts, filename, false);
-+ if (!opts) {
-+ if (errp) error_setg(errp, "pbs_parse_filename failed");
-+ return;
-+ }
-+
-+ qemu_opts_to_qdict(opts, options);
-+
-+ qemu_opts_del(opts);
-+}
-+
-+static int pbs_open(BlockDriverState *bs, QDict *options, int flags,
-+ Error **errp)
-+{
-+ QemuOpts *opts;
-+ BDRVPBSState *s = bs->opaque;
-+ char *pbs_error = NULL;
-+
-+ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
-+ qemu_opts_absorb_qdict(opts, options, &error_abort);
-+
-+ s->repository = g_strdup(qemu_opt_get(opts, PBS_OPT_REPOSITORY));
-+ s->snapshot = g_strdup(qemu_opt_get(opts, PBS_OPT_SNAPSHOT));
-+ s->archive = g_strdup(qemu_opt_get(opts, PBS_OPT_ARCHIVE));
-+ const char *keyfile = qemu_opt_get(opts, PBS_OPT_KEYFILE);
-+ const char *password = qemu_opt_get(opts, PBS_OPT_PASSWORD);
-+ const char *fingerprint = qemu_opt_get(opts, PBS_OPT_FINGERPRINT);
-+ const char *key_password = qemu_opt_get(opts, PBS_OPT_ENCRYPTION_PASSWORD);
-+
-+ if (!password) {
-+ password = getenv("PBS_PASSWORD");
-+ }
-+ if (!fingerprint) {
-+ fingerprint = getenv("PBS_FINGERPRINT");
-+ }
-+ if (!key_password) {
-+ key_password = getenv("PBS_ENCRYPTION_PASSWORD");
-+ }
-+
-+ /* connect to PBS server in read mode */
-+ s->conn = proxmox_restore_new(s->repository, s->snapshot, password,
-+ keyfile, key_password, fingerprint, &pbs_error);
-+
-+ /* invalidates qemu_opt_get char pointers from above */
-+ qemu_opts_del(opts);
-+
-+ if (!s->conn) {
-+ if (pbs_error && errp) error_setg(errp, "PBS restore_new failed: %s", pbs_error);
-+ if (pbs_error) proxmox_backup_free_error(pbs_error);
-+ return -ENOMEM;
-+ }
-+
-+ int ret = proxmox_restore_connect(s->conn, &pbs_error);
-+ if (ret < 0) {
-+ if (pbs_error && errp) error_setg(errp, "PBS connect failed: %s", pbs_error);
-+ if (pbs_error) proxmox_backup_free_error(pbs_error);
-+ return -ECONNREFUSED;
-+ }
-+
-+ /* acquire handle and length */
-+ s->aid = proxmox_restore_open_image(s->conn, s->archive, &pbs_error);
-+ if (s->aid < 0) {
-+ if (pbs_error && errp) error_setg(errp, "PBS open_image failed: %s", pbs_error);
-+ if (pbs_error) proxmox_backup_free_error(pbs_error);
-+ return -ENODEV;
-+ }
-+ s->length = proxmox_restore_get_image_length(s->conn, s->aid, &pbs_error);
-+ if (s->length < 0) {
-+ if (pbs_error && errp) error_setg(errp, "PBS get_image_length failed: %s", pbs_error);
-+ if (pbs_error) proxmox_backup_free_error(pbs_error);
-+ return -EINVAL;
-+ }
-+
-+ return 0;
-+}
-+
-+static int pbs_file_open(BlockDriverState *bs, QDict *options, int flags,
-+ Error **errp)
-+{
-+ return pbs_open(bs, options, flags, errp);
-+}
-+
-+static void pbs_close(BlockDriverState *bs) {
-+ BDRVPBSState *s = bs->opaque;
-+ g_free(s->repository);
-+ g_free(s->snapshot);
-+ g_free(s->archive);
-+ proxmox_restore_disconnect(s->conn);
-+}
-+
-+static int64_t pbs_getlength(BlockDriverState *bs)
-+{
-+ BDRVPBSState *s = bs->opaque;
-+ return s->length;
-+}
-+
-+typedef struct ReadCallbackData {
-+ Coroutine *co;
-+ AioContext *ctx;
-+} ReadCallbackData;
-+
-+static void read_callback(void *callback_data)
-+{
-+ ReadCallbackData *rcb = callback_data;
-+ aio_co_schedule(rcb->ctx, rcb->co);
-+}
-+
-+static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes,
-+ QEMUIOVector *qiov, BdrvRequestFlags flags)
-+{
-+ BDRVPBSState *s = bs->opaque;
-+ int ret;
-+ char *pbs_error = NULL;
-+ uint8_t *buf = malloc(bytes);
-+
-+ if (offset < 0 || bytes < 0) {
-+ fprintf(stderr, "unexpected negative 'offset' or 'bytes' value!\n");
-+ return -EINVAL;
-+ }
-+
-+ ReadCallbackData rcb = {
-+ .co = qemu_coroutine_self(),
-+ .ctx = bdrv_get_aio_context(bs),
-+ };
-+
-+ proxmox_restore_read_image_at_async(s->conn, s->aid, buf, (uint64_t)offset, (uint64_t)bytes,
-+ read_callback, (void *) &rcb, &ret, &pbs_error);
-+
-+ qemu_coroutine_yield();
-+
-+ if (ret < 0) {
-+ fprintf(stderr, "error during PBS read: %s\n", pbs_error ? pbs_error : "unknown error");
-+ if (pbs_error) proxmox_backup_free_error(pbs_error);
-+ return -EIO;
-+ }
-+
-+ qemu_iovec_from_buf(qiov, 0, buf, bytes);
-+ free(buf);
-+
-+ return ret;
-+}
-+
-+static coroutine_fn int pbs_co_pwritev(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes,
-+ QEMUIOVector *qiov, BdrvRequestFlags flags)
-+{
-+ fprintf(stderr, "pbs-bdrv: cannot write to backup file, make sure "
-+ "any attached disk devices are set to read-only!\n");
-+ return -EPERM;
-+}
-+
-+static void pbs_refresh_filename(BlockDriverState *bs)
-+{
-+ BDRVPBSState *s = bs->opaque;
-+ snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s(%s)",
-+ s->repository, s->snapshot, s->archive);
-+}
-+
-+static const char *const pbs_strong_runtime_opts[] = {
-+ NULL
-+};
-+
-+static BlockDriver bdrv_pbs_co = {
-+ .format_name = "pbs",
-+ .protocol_name = "pbs",
-+ .instance_size = sizeof(BDRVPBSState),
-+
-+ .bdrv_parse_filename = pbs_parse_filename,
-+
-+ .bdrv_file_open = pbs_file_open,
-+ .bdrv_open = pbs_open,
-+ .bdrv_close = pbs_close,
-+ .bdrv_getlength = pbs_getlength,
-+
-+ .bdrv_co_preadv = pbs_co_preadv,
-+ .bdrv_co_pwritev = pbs_co_pwritev,
-+
-+ .bdrv_refresh_filename = pbs_refresh_filename,
-+ .strong_runtime_opts = pbs_strong_runtime_opts,
-+};
-+
-+static void bdrv_pbs_init(void)
-+{
-+ bdrv_register(&bdrv_pbs_co);
-+}
-+
-+block_init(bdrv_pbs_init);
-diff --git a/configure b/configure
-index 7c08c18358..08d164b4f5 100755
---- a/configure
-+++ b/configure
-@@ -325,6 +325,7 @@ trace_file="trace"
- opengl="$default_feature"
- coroutine=""
- tls_priority="NORMAL"
-+pbs_bdrv="yes"
- plugins="$default_feature"
- secret_keyring="$default_feature"
- meson=""
-@@ -991,6 +992,10 @@ for opt do
- ;;
- --disable-pvrdma) pvrdma="no"
- ;;
-+ --disable-pbs-bdrv) pbs_bdrv="no"
-+ ;;
-+ --enable-pbs-bdrv) pbs_bdrv="yes"
-+ ;;
- --disable-vhost-user) vhost_user="no"
- ;;
- --enable-vhost-user) vhost_user="yes"
-@@ -1265,6 +1270,7 @@ cat << EOF
- vhost-user vhost-user backend support
- vhost-vdpa vhost-vdpa kernel backend support
- opengl opengl support
-+ pbs-bdrv Proxmox backup server read-only block driver support
- gio libgio support
-
- NOTE: The object files are built at the place where configure is launched
-@@ -2848,6 +2854,9 @@ if test "$xen" = "enabled" ; then
- echo "XEN_CFLAGS=$xen_cflags" >> $config_host_mak
- echo "XEN_LIBS=$xen_libs" >> $config_host_mak
- fi
-+if test "$pbs_bdrv" = "yes" ; then
-+ echo "CONFIG_PBS_BDRV=y" >> $config_host_mak
-+fi
- if test "$vhost_scsi" = "yes" ; then
- echo "CONFIG_VHOST_SCSI=y" >> $config_host_mak
- fi
-diff --git a/meson.build b/meson.build
-index 9476ccbe07..d3783a83e4 100644
---- a/meson.build
-+++ b/meson.build
-@@ -3743,6 +3743,7 @@ summary_info += {'bzip2 support': libbzip2}
- summary_info += {'lzfse support': liblzfse}
- summary_info += {'zstd support': zstd}
- summary_info += {'NUMA host support': numa}
-+summary_info += {'PBS bdrv support': config_host.has_key('CONFIG_PBS_BDRV')}
- summary_info += {'capstone': capstone_opt == 'internal' ? capstone_opt : capstone}
- summary_info += {'libpmem support': libpmem}
- summary_info += {'libdaxctl support': libdaxctl}
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index abef91062a..68591cb343 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -3073,6 +3073,7 @@
- 'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
- 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
- { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
-+ 'pbs',
- 'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
-
- ##
-@@ -3145,6 +3146,17 @@
- { 'struct': 'BlockdevOptionsNull',
- 'data': { '*size': 'int', '*latency-ns': 'uint64', '*read-zeroes': 'bool' } }
-
-+##
-+# @BlockdevOptionsPbs:
-+#
-+# Driver specific block device options for the PBS backend.
-+#
-+##
-+{ 'struct': 'BlockdevOptionsPbs',
-+ 'data': { 'repository': 'str', 'snapshot': 'str', 'archive': 'str',
-+ '*keyfile': 'str', '*password': 'str', '*fingerprint': 'str',
-+ '*key_password': 'str' } }
-+
- ##
- # @BlockdevOptionsNVMe:
- #
-@@ -4405,6 +4417,7 @@
- 'nfs': 'BlockdevOptionsNfs',
- 'null-aio': 'BlockdevOptionsNull',
- 'null-co': 'BlockdevOptionsNull',
-+ 'pbs': 'BlockdevOptionsPbs',
- 'nvme': 'BlockdevOptionsNVMe',
- 'parallels': 'BlockdevOptionsGenericFormat',
- 'preallocate':'BlockdevOptionsPreallocate',
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Thu, 9 Jul 2020 12:53:08 +0200
+Subject: [PATCH] PVE: various PBS fixes
+
+pbs: fix crypt and compress parameters
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+
+PVE: handle PBS write callback with big blocks correctly
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+
+PVE: add zero block handling to PBS dump callback
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/monitor/block-hmp-cmds.c | 4 ++-
+ pve-backup.c | 57 +++++++++++++++++++++++++++-------
+ qapi/block-core.json | 6 ++++
+ 3 files changed, 54 insertions(+), 13 deletions(-)
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index 45da74d7a0..ea7b665aa2 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1042,7 +1042,9 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
+ false, NULL, // PBS fingerprint
+ false, NULL, // PBS backup-id
+ false, 0, // PBS backup-time
+- false, false, // PBS incremental
++ false, false, // PBS use-dirty-bitmap
++ false, false, // PBS compress
++ false, false, // PBS encrypt
+ true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
+ false, NULL, false, NULL, !!devlist,
+ devlist, qdict_haskey(qdict, "speed"), speed, &error);
+diff --git a/pve-backup.c b/pve-backup.c
+index 1c49cd178d..c15abefdda 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -8,6 +8,7 @@
+ #include "block/blockjob.h"
+ #include "qapi/qapi-commands-block.h"
+ #include "qapi/qmp/qerror.h"
++#include "qemu/cutils.h"
+
+ /* PVE backup state and related function */
+
+@@ -67,6 +68,7 @@ opts_init(pvebackup_init);
+ typedef struct PVEBackupDevInfo {
+ BlockDriverState *bs;
+ size_t size;
++ uint64_t block_size;
+ uint8_t dev_id;
+ bool completed;
+ char targetfile[PATH_MAX];
+@@ -135,10 +137,13 @@ pvebackup_co_dump_pbs_cb(
+ PVEBackupDevInfo *di = opaque;
+
+ assert(backup_state.pbs);
++ assert(buf);
+
+ Error *local_err = NULL;
+ int pbs_res = -1;
+
++ bool is_zero_block = size == di->block_size && buffer_is_zero(buf, size);
++
+ qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
+
+ // avoid deadlock if job is cancelled
+@@ -147,17 +152,29 @@ pvebackup_co_dump_pbs_cb(
+ return -1;
+ }
+
+- pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id, buf, start, size, &local_err);
+- qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++ uint64_t transferred = 0;
++ uint64_t reused = 0;
++ while (transferred < size) {
++ uint64_t left = size - transferred;
++ uint64_t to_transfer = left < di->block_size ? left : di->block_size;
+
+- if (pbs_res < 0) {
+- pvebackup_propagate_error(local_err);
+- return pbs_res;
+- } else {
+- size_t reused = (pbs_res == 0) ? size : 0;
+- pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
++ pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id,
++ is_zero_block ? NULL : buf + transferred, start + transferred,
++ to_transfer, &local_err);
++ transferred += to_transfer;
++
++ if (pbs_res < 0) {
++ pvebackup_propagate_error(local_err);
++ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++ return pbs_res;
++ }
++
++ reused += pbs_res == 0 ? to_transfer : 0;
+ }
+
++ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
++ pvebackup_add_transfered_bytes(size, is_zero_block ? size : 0, reused);
++
+ return size;
+ }
+
+@@ -178,6 +195,7 @@ pvebackup_co_dump_vma_cb(
+ int ret = -1;
+
+ assert(backup_state.vmaw);
++ assert(buf);
+
+ uint64_t remaining = size;
+
+@@ -204,9 +222,7 @@ pvebackup_co_dump_vma_cb(
+ qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
+
+ ++cluster_num;
+- if (buf) {
+- buf += VMA_CLUSTER_SIZE;
+- }
++ buf += VMA_CLUSTER_SIZE;
+ if (ret < 0) {
+ Error *local_err = NULL;
+ vma_writer_error_propagate(backup_state.vmaw, &local_err);
+@@ -569,6 +585,10 @@ typedef struct QmpBackupTask {
+ const char *firewall_file;
+ bool has_devlist;
+ const char *devlist;
++ bool has_compress;
++ bool compress;
++ bool has_encrypt;
++ bool encrypt;
+ bool has_speed;
+ int64_t speed;
+ Error **errp;
+@@ -692,6 +712,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+
+ bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
+
++
+ char *pbs_err = NULL;
+ pbs = proxmox_backup_new(
+ task->backup_file,
+@@ -701,8 +722,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ task->has_password ? task->password : NULL,
+ task->has_keyfile ? task->keyfile : NULL,
+ task->has_key_password ? task->key_password : NULL,
++ task->has_compress ? task->compress : true,
++ task->has_encrypt ? task->encrypt : task->has_keyfile,
+ task->has_fingerprint ? task->fingerprint : NULL,
+- &pbs_err);
++ &pbs_err);
+
+ if (!pbs) {
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
+@@ -721,6 +744,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+ l = g_list_next(l);
+
++ di->block_size = dump_cb_block_size;
++
+ const char *devname = bdrv_get_device_name(di->bs);
+
+ BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
+@@ -941,6 +966,8 @@ UuidInfo *qmp_backup(
+ bool has_backup_id, const char *backup_id,
+ bool has_backup_time, int64_t backup_time,
+ bool has_use_dirty_bitmap, bool use_dirty_bitmap,
++ bool has_compress, bool compress,
++ bool has_encrypt, bool encrypt,
+ bool has_format, BackupFormat format,
+ bool has_config_file, const char *config_file,
+ bool has_firewall_file, const char *firewall_file,
+@@ -951,6 +978,8 @@ UuidInfo *qmp_backup(
+ .backup_file = backup_file,
+ .has_password = has_password,
+ .password = password,
++ .has_keyfile = has_keyfile,
++ .keyfile = keyfile,
+ .has_key_password = has_key_password,
+ .key_password = key_password,
+ .has_fingerprint = has_fingerprint,
+@@ -961,6 +990,10 @@ UuidInfo *qmp_backup(
+ .backup_time = backup_time,
+ .has_use_dirty_bitmap = has_use_dirty_bitmap,
+ .use_dirty_bitmap = use_dirty_bitmap,
++ .has_compress = has_compress,
++ .compress = compress,
++ .has_encrypt = has_encrypt,
++ .encrypt = encrypt,
+ .has_format = has_format,
+ .format = format,
+ .has_config_file = has_config_file,
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 3ad9eb5d1a..4120052690 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -818,6 +818,10 @@
+ #
+ # @use-dirty-bitmap: use dirty bitmap to detect incremental changes since last job (optional for format 'pbs')
+ #
++# @compress: use compression (optional for format 'pbs', defaults to true)
++#
++# @encrypt: use encryption ((optional for format 'pbs', defaults to true if there is a keyfile)
++#
+ # Returns: the uuid of the backup job
+ #
+ ##
+@@ -829,6 +833,8 @@
+ '*backup-id': 'str',
+ '*backup-time': 'int',
+ '*use-dirty-bitmap': 'bool',
++ '*compress': 'bool',
++ '*encrypt': 'bool',
+ '*format': 'BackupFormat',
+ '*config-file': 'str',
+ '*firewall-file': 'str',
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 8 Jul 2020 09:50:54 +0200
+Subject: [PATCH] PVE: Add PBS block driver to map backup archives into VMs
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+[error cleanups, file_open implementation]
+Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[adapt to changed function signatures]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/meson.build | 3 +
+ block/pbs.c | 276 +++++++++++++++++++++++++++++++++++++++++++
+ configure | 9 ++
+ meson.build | 2 +-
+ qapi/block-core.json | 13 ++
+ 5 files changed, 302 insertions(+), 1 deletion(-)
+ create mode 100644 block/pbs.c
+
+diff --git a/block/meson.build b/block/meson.build
+index 2783b77e9c..a26a69434e 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -53,6 +53,9 @@ block_ss.add(files(
+ '../pve-backup.c',
+ ), libproxmox_backup_qemu)
+
++block_ss.add(when: 'CONFIG_PBS_BDRV', if_true: files('pbs.c'))
++block_ss.add(when: 'CONFIG_PBS_BDRV', if_true: libproxmox_backup_qemu)
++
+
+ softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+
+diff --git a/block/pbs.c b/block/pbs.c
+new file mode 100644
+index 0000000000..0b05ea9080
+--- /dev/null
++++ b/block/pbs.c
+@@ -0,0 +1,276 @@
++/*
++ * Proxmox Backup Server read-only block driver
++ */
++
++#include "qemu/osdep.h"
++#include "qapi/error.h"
++#include "qapi/qmp/qdict.h"
++#include "qapi/qmp/qstring.h"
++#include "qemu/module.h"
++#include "qemu/option.h"
++#include "qemu/cutils.h"
++#include "block/block_int.h"
++
++#include <proxmox-backup-qemu.h>
++
++#define PBS_OPT_REPOSITORY "repository"
++#define PBS_OPT_SNAPSHOT "snapshot"
++#define PBS_OPT_ARCHIVE "archive"
++#define PBS_OPT_KEYFILE "keyfile"
++#define PBS_OPT_PASSWORD "password"
++#define PBS_OPT_FINGERPRINT "fingerprint"
++#define PBS_OPT_ENCRYPTION_PASSWORD "key_password"
++
++typedef struct {
++ ProxmoxRestoreHandle *conn;
++ char aid;
++ int64_t length;
++
++ char *repository;
++ char *snapshot;
++ char *archive;
++} BDRVPBSState;
++
++static QemuOptsList runtime_opts = {
++ .name = "pbs",
++ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
++ .desc = {
++ {
++ .name = PBS_OPT_REPOSITORY,
++ .type = QEMU_OPT_STRING,
++ .help = "The server address and repository to connect to.",
++ },
++ {
++ .name = PBS_OPT_SNAPSHOT,
++ .type = QEMU_OPT_STRING,
++ .help = "The snapshot to read.",
++ },
++ {
++ .name = PBS_OPT_ARCHIVE,
++ .type = QEMU_OPT_STRING,
++ .help = "Which archive within the snapshot should be accessed.",
++ },
++ {
++ .name = PBS_OPT_PASSWORD,
++ .type = QEMU_OPT_STRING,
++ .help = "Server password. Can be passed as env var 'PBS_PASSWORD'.",
++ },
++ {
++ .name = PBS_OPT_FINGERPRINT,
++ .type = QEMU_OPT_STRING,
++ .help = "Server fingerprint. Can be passed as env var 'PBS_FINGERPRINT'.",
++ },
++ {
++ .name = PBS_OPT_ENCRYPTION_PASSWORD,
++ .type = QEMU_OPT_STRING,
++ .help = "Optional: Key password. Can be passed as env var 'PBS_ENCRYPTION_PASSWORD'.",
++ },
++ {
++ .name = PBS_OPT_KEYFILE,
++ .type = QEMU_OPT_STRING,
++ .help = "Optional: The path to the keyfile to use.",
++ },
++ { /* end of list */ }
++ },
++};
++
++
++// filename format:
++// pbs:repository=<repo>,snapshot=<snap>,password=<pw>,key_password=<kpw>,fingerprint=<fp>,archive=<archive>
++static void pbs_parse_filename(const char *filename, QDict *options,
++ Error **errp)
++{
++
++ if (!strstart(filename, "pbs:", &filename)) {
++ if (errp) error_setg(errp, "pbs_parse_filename failed - missing 'pbs:' prefix");
++ }
++
++
++ QemuOpts *opts = qemu_opts_parse_noisily(&runtime_opts, filename, false);
++ if (!opts) {
++ if (errp) error_setg(errp, "pbs_parse_filename failed");
++ return;
++ }
++
++ qemu_opts_to_qdict(opts, options);
++
++ qemu_opts_del(opts);
++}
++
++static int pbs_open(BlockDriverState *bs, QDict *options, int flags,
++ Error **errp)
++{
++ QemuOpts *opts;
++ BDRVPBSState *s = bs->opaque;
++ char *pbs_error = NULL;
++
++ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
++ qemu_opts_absorb_qdict(opts, options, &error_abort);
++
++ s->repository = g_strdup(qemu_opt_get(opts, PBS_OPT_REPOSITORY));
++ s->snapshot = g_strdup(qemu_opt_get(opts, PBS_OPT_SNAPSHOT));
++ s->archive = g_strdup(qemu_opt_get(opts, PBS_OPT_ARCHIVE));
++ const char *keyfile = qemu_opt_get(opts, PBS_OPT_KEYFILE);
++ const char *password = qemu_opt_get(opts, PBS_OPT_PASSWORD);
++ const char *fingerprint = qemu_opt_get(opts, PBS_OPT_FINGERPRINT);
++ const char *key_password = qemu_opt_get(opts, PBS_OPT_ENCRYPTION_PASSWORD);
++
++ if (!password) {
++ password = getenv("PBS_PASSWORD");
++ }
++ if (!fingerprint) {
++ fingerprint = getenv("PBS_FINGERPRINT");
++ }
++ if (!key_password) {
++ key_password = getenv("PBS_ENCRYPTION_PASSWORD");
++ }
++
++ /* connect to PBS server in read mode */
++ s->conn = proxmox_restore_new(s->repository, s->snapshot, password,
++ keyfile, key_password, fingerprint, &pbs_error);
++
++ /* invalidates qemu_opt_get char pointers from above */
++ qemu_opts_del(opts);
++
++ if (!s->conn) {
++ if (pbs_error && errp) error_setg(errp, "PBS restore_new failed: %s", pbs_error);
++ if (pbs_error) proxmox_backup_free_error(pbs_error);
++ return -ENOMEM;
++ }
++
++ int ret = proxmox_restore_connect(s->conn, &pbs_error);
++ if (ret < 0) {
++ if (pbs_error && errp) error_setg(errp, "PBS connect failed: %s", pbs_error);
++ if (pbs_error) proxmox_backup_free_error(pbs_error);
++ return -ECONNREFUSED;
++ }
++
++ /* acquire handle and length */
++ s->aid = proxmox_restore_open_image(s->conn, s->archive, &pbs_error);
++ if (s->aid < 0) {
++ if (pbs_error && errp) error_setg(errp, "PBS open_image failed: %s", pbs_error);
++ if (pbs_error) proxmox_backup_free_error(pbs_error);
++ return -ENODEV;
++ }
++ s->length = proxmox_restore_get_image_length(s->conn, s->aid, &pbs_error);
++ if (s->length < 0) {
++ if (pbs_error && errp) error_setg(errp, "PBS get_image_length failed: %s", pbs_error);
++ if (pbs_error) proxmox_backup_free_error(pbs_error);
++ return -EINVAL;
++ }
++
++ return 0;
++}
++
++static int pbs_file_open(BlockDriverState *bs, QDict *options, int flags,
++ Error **errp)
++{
++ return pbs_open(bs, options, flags, errp);
++}
++
++static void pbs_close(BlockDriverState *bs) {
++ BDRVPBSState *s = bs->opaque;
++ g_free(s->repository);
++ g_free(s->snapshot);
++ g_free(s->archive);
++ proxmox_restore_disconnect(s->conn);
++}
++
++static int64_t pbs_getlength(BlockDriverState *bs)
++{
++ BDRVPBSState *s = bs->opaque;
++ return s->length;
++}
++
++typedef struct ReadCallbackData {
++ Coroutine *co;
++ AioContext *ctx;
++} ReadCallbackData;
++
++static void read_callback(void *callback_data)
++{
++ ReadCallbackData *rcb = callback_data;
++ aio_co_schedule(rcb->ctx, rcb->co);
++}
++
++static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
++ int64_t offset, int64_t bytes,
++ QEMUIOVector *qiov, BdrvRequestFlags flags)
++{
++ BDRVPBSState *s = bs->opaque;
++ int ret;
++ char *pbs_error = NULL;
++ uint8_t *buf = malloc(bytes);
++
++ if (offset < 0 || bytes < 0) {
++ fprintf(stderr, "unexpected negative 'offset' or 'bytes' value!\n");
++ return -EINVAL;
++ }
++
++ ReadCallbackData rcb = {
++ .co = qemu_coroutine_self(),
++ .ctx = bdrv_get_aio_context(bs),
++ };
++
++ proxmox_restore_read_image_at_async(s->conn, s->aid, buf, (uint64_t)offset, (uint64_t)bytes,
++ read_callback, (void *) &rcb, &ret, &pbs_error);
++
++ qemu_coroutine_yield();
++
++ if (ret < 0) {
++ fprintf(stderr, "error during PBS read: %s\n", pbs_error ? pbs_error : "unknown error");
++ if (pbs_error) proxmox_backup_free_error(pbs_error);
++ return -EIO;
++ }
++
++ qemu_iovec_from_buf(qiov, 0, buf, bytes);
++ free(buf);
++
++ return ret;
++}
++
++static coroutine_fn int pbs_co_pwritev(BlockDriverState *bs,
++ int64_t offset, int64_t bytes,
++ QEMUIOVector *qiov, BdrvRequestFlags flags)
++{
++ fprintf(stderr, "pbs-bdrv: cannot write to backup file, make sure "
++ "any attached disk devices are set to read-only!\n");
++ return -EPERM;
++}
++
++static void pbs_refresh_filename(BlockDriverState *bs)
++{
++ BDRVPBSState *s = bs->opaque;
++ snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s(%s)",
++ s->repository, s->snapshot, s->archive);
++}
++
++static const char *const pbs_strong_runtime_opts[] = {
++ NULL
++};
++
++static BlockDriver bdrv_pbs_co = {
++ .format_name = "pbs",
++ .protocol_name = "pbs",
++ .instance_size = sizeof(BDRVPBSState),
++
++ .bdrv_parse_filename = pbs_parse_filename,
++
++ .bdrv_file_open = pbs_file_open,
++ .bdrv_open = pbs_open,
++ .bdrv_close = pbs_close,
++ .bdrv_getlength = pbs_getlength,
++
++ .bdrv_co_preadv = pbs_co_preadv,
++ .bdrv_co_pwritev = pbs_co_pwritev,
++
++ .bdrv_refresh_filename = pbs_refresh_filename,
++ .strong_runtime_opts = pbs_strong_runtime_opts,
++};
++
++static void bdrv_pbs_init(void)
++{
++ bdrv_register(&bdrv_pbs_co);
++}
++
++block_init(bdrv_pbs_init);
+diff --git a/configure b/configure
+index 72ab03f11a..7203c270ec 100755
+--- a/configure
++++ b/configure
+@@ -309,6 +309,7 @@ linux_user=""
+ bsd_user=""
+ pie=""
+ coroutine=""
++pbs_bdrv="yes"
+ plugins="$default_feature"
+ meson=""
+ meson_args=""
+@@ -902,6 +903,10 @@ for opt do
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;
++ --disable-pbs-bdrv) pbs_bdrv="no"
++ ;;
++ --enable-pbs-bdrv) pbs_bdrv="yes"
++ ;;
+ --with-git=*) git="$optarg"
+ ;;
+ --with-git-submodules=*)
+@@ -1087,6 +1092,7 @@ cat << EOF
+ debug-info debugging information
+ safe-stack SafeStack Stack Smash Protection. Depends on
+ clang/llvm >= 3.7 and requires coroutine backend ucontext.
++ pbs-bdrv Proxmox backup server read-only block driver support
+
+ NOTE: The object files are built at the place where configure is launched
+ EOF
+@@ -2463,6 +2469,9 @@ echo "TARGET_DIRS=$target_list" >> $config_host_mak
+ if test "$modules" = "yes"; then
+ echo "CONFIG_MODULES=y" >> $config_host_mak
+ fi
++if test "$pbs_bdrv" = "yes" ; then
++ echo "CONFIG_PBS_BDRV=y" >> $config_host_mak
++fi
+
+ # XXX: suppress that
+ if [ "$bsd" = "yes" ] ; then
+diff --git a/meson.build b/meson.build
+index f48d2e0457..be4785e2f6 100644
+--- a/meson.build
++++ b/meson.build
+@@ -3986,7 +3986,7 @@ summary_info += {'bzip2 support': libbzip2}
+ summary_info += {'lzfse support': liblzfse}
+ summary_info += {'zstd support': zstd}
+ summary_info += {'NUMA host support': numa}
+-summary_info += {'capstone': capstone}
++summary_info += {'PBS bdrv support': config_host.has_key('CONFIG_PBS_BDRV')}
+ summary_info += {'libpmem support': libpmem}
+ summary_info += {'libdaxctl support': libdaxctl}
+ summary_info += {'libudev': libudev}
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 4120052690..96bc696aaa 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -3099,6 +3099,7 @@
+ 'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
+ 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
+ { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
++ 'pbs',
+ 'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
+
+ ##
+@@ -3171,6 +3172,17 @@
+ { 'struct': 'BlockdevOptionsNull',
+ 'data': { '*size': 'int', '*latency-ns': 'uint64', '*read-zeroes': 'bool' } }
+
++##
++# @BlockdevOptionsPbs:
++#
++# Driver specific block device options for the PBS backend.
++#
++##
++{ 'struct': 'BlockdevOptionsPbs',
++ 'data': { 'repository': 'str', 'snapshot': 'str', 'archive': 'str',
++ '*keyfile': 'str', '*password': 'str', '*fingerprint': 'str',
++ '*key_password': 'str' } }
++
+ ##
+ # @BlockdevOptionsNVMe:
+ #
+@@ -4455,6 +4467,7 @@
+ 'nfs': 'BlockdevOptionsNfs',
+ 'null-aio': 'BlockdevOptionsNull',
+ 'null-co': 'BlockdevOptionsNull',
++ 'pbs': 'BlockdevOptionsPbs',
+ 'nvme': 'BlockdevOptionsNVMe',
+ 'parallels': 'BlockdevOptionsGenericFormat',
+ 'preallocate':'BlockdevOptionsPreallocate',
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 8 Jul 2020 11:57:53 +0200
-Subject: [PATCH] PVE: add query_proxmox_support QMP command
-
-Generic interface for future use, currently used for PBS dirty-bitmap
-backup support.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[PVE: query-proxmox-support: include library version]
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pve-backup.c | 9 +++++++++
- qapi/block-core.json | 29 +++++++++++++++++++++++++++++
- 2 files changed, 38 insertions(+)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index c15abefdda..4684789813 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -1075,3 +1075,12 @@ BackupStatus *qmp_query_backup(Error **errp)
-
- return info;
- }
-+
-+ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
-+{
-+ ProxmoxSupportStatus *ret = g_malloc0(sizeof(*ret));
-+ ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
-+ ret->pbs_dirty_bitmap = true;
-+ ret->pbs_dirty_bitmap_savevm = true;
-+ return ret;
-+}
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 68591cb343..7d25aa4fa1 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -867,6 +867,35 @@
- ##
- { 'command': 'backup-cancel' }
-
-+##
-+# @ProxmoxSupportStatus:
-+#
-+# Contains info about supported features added by Proxmox.
-+#
-+# @pbs-dirty-bitmap: True if dirty-bitmap-incremental backups to PBS are
-+# supported.
-+#
-+# @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
-+# safely be set for savevm-async.
-+#
-+# @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
-+#
-+##
-+{ 'struct': 'ProxmoxSupportStatus',
-+ 'data': { 'pbs-dirty-bitmap': 'bool',
-+ 'pbs-dirty-bitmap-savevm': 'bool',
-+ 'pbs-library-version': 'str' } }
-+
-+##
-+# @query-proxmox-support:
-+#
-+# Returns information about supported features added by Proxmox.
-+#
-+# Returns: @ProxmoxSupportStatus
-+#
-+##
-+{ 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
-+
- ##
- # @BlockDeviceTimedStats:
- #
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 19 Aug 2020 17:02:00 +0200
-Subject: [PATCH] PVE: add query-pbs-bitmap-info QMP call
-
-Returns advanced information about dirty bitmaps used (or not used) for
-the latest PBS backup.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- monitor/hmp-cmds.c | 28 ++++++-----
- pve-backup.c | 117 ++++++++++++++++++++++++++++++++-----------
- qapi/block-core.json | 56 +++++++++++++++++++++
- 3 files changed, 159 insertions(+), 42 deletions(-)
-
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index c5c74ac1dc..df273f41fb 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -198,6 +198,7 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
- void hmp_info_backup(Monitor *mon, const QDict *qdict)
- {
- BackupStatus *info;
-+ PBSBitmapInfoList *bitmap_info;
-
- info = qmp_query_backup(NULL);
-
-@@ -228,26 +229,29 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
- // this should not happen normally
- monitor_printf(mon, "Total size: %d\n", 0);
- } else {
-- bool incremental = false;
- size_t total_or_dirty = info->total;
-- if (info->has_transferred) {
-- if (info->has_dirty && info->dirty) {
-- if (info->dirty < info->total) {
-- total_or_dirty = info->dirty;
-- incremental = true;
-- }
-- }
-+ bitmap_info = qmp_query_pbs_bitmap_info(NULL);
-+
-+ while (bitmap_info) {
-+ monitor_printf(mon, "Drive %s:\n",
-+ bitmap_info->value->drive);
-+ monitor_printf(mon, " bitmap action: %s\n",
-+ PBSBitmapAction_str(bitmap_info->value->action));
-+ monitor_printf(mon, " size: %zd\n",
-+ bitmap_info->value->size);
-+ monitor_printf(mon, " dirty: %zd\n",
-+ bitmap_info->value->dirty);
-+ bitmap_info = bitmap_info->next;
- }
-
-- int per = (info->transferred * 100)/total_or_dirty;
--
-- monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
-+ qapi_free_PBSBitmapInfoList(bitmap_info);
-
- int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
- (info->zero_bytes * 100)/info->total : 0;
- monitor_printf(mon, "Total size: %zd\n", info->total);
-+ int trans_per = (info->transferred * 100)/total_or_dirty;
- monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-- info->transferred, per);
-+ info->transferred, trans_per);
- monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
- info->zero_bytes, zero_per);
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 4684789813..f90abaa50a 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -46,6 +46,7 @@ static struct PVEBackupState {
- size_t transferred;
- size_t reused;
- size_t zero_bytes;
-+ GList *bitmap_list;
- } stat;
- int64_t speed;
- VmaWriter *vmaw;
-@@ -672,7 +673,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
-
- size_t total = 0;
-- size_t dirty = 0;
-
- l = di_list;
- while (l) {
-@@ -693,18 +693,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-
- uuid_generate(uuid);
-
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ backup_state.stat.reused = 0;
-+
-+ /* clear previous backup's bitmap_list */
-+ if (backup_state.stat.bitmap_list) {
-+ GList *bl = backup_state.stat.bitmap_list;
-+ while (bl) {
-+ g_free(((PBSBitmapInfo *)bl->data)->drive);
-+ g_free(bl->data);
-+ bl = g_list_next(bl);
-+ }
-+ g_list_free(backup_state.stat.bitmap_list);
-+ backup_state.stat.bitmap_list = NULL;
-+ }
-+
- if (format == BACKUP_FORMAT_PBS) {
- if (!task->has_password) {
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
-- goto err;
-+ goto err_mutex;
- }
- if (!task->has_backup_id) {
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
-- goto err;
-+ goto err_mutex;
- }
- if (!task->has_backup_time) {
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
-- goto err;
-+ goto err_mutex;
- }
-
- int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
-@@ -731,12 +746,12 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
- "proxmox_backup_new failed: %s", pbs_err);
- proxmox_backup_free_error(pbs_err);
-- goto err;
-+ goto err_mutex;
- }
-
- int connect_result = proxmox_backup_co_connect(pbs, task->errp);
- if (connect_result < 0)
-- goto err;
-+ goto err_mutex;
-
- /* register all devices */
- l = di_list;
-@@ -747,6 +762,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- di->block_size = dump_cb_block_size;
-
- const char *devname = bdrv_get_device_name(di->bs);
-+ PBSBitmapAction action = PBS_BITMAP_ACTION_NOT_USED;
-+ size_t dirty = di->size;
-
- BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
- bool expect_only_dirty = false;
-@@ -755,49 +772,59 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- if (bitmap == NULL) {
- bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
- if (!bitmap) {
-- goto err;
-+ goto err_mutex;
- }
-+ action = PBS_BITMAP_ACTION_NEW;
- } else {
- expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
- }
-
- if (expect_only_dirty) {
-- dirty += bdrv_get_dirty_count(bitmap);
-+ /* track clean chunks as reused */
-+ dirty = MIN(bdrv_get_dirty_count(bitmap), di->size);
-+ backup_state.stat.reused += di->size - dirty;
-+ action = PBS_BITMAP_ACTION_USED;
- } else {
- /* mark entire bitmap as dirty to make full backup */
- bdrv_set_dirty_bitmap(bitmap, 0, di->size);
-- dirty += di->size;
-+ if (action != PBS_BITMAP_ACTION_NEW) {
-+ action = PBS_BITMAP_ACTION_INVALID;
-+ }
- }
- di->bitmap = bitmap;
- } else {
-- dirty += di->size;
--
- /* after a full backup the old dirty bitmap is invalid anyway */
- if (bitmap != NULL) {
- bdrv_release_dirty_bitmap(bitmap);
-+ action = PBS_BITMAP_ACTION_NOT_USED_REMOVED;
- }
- }
-
- int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
- if (dev_id < 0) {
-- goto err;
-+ goto err_mutex;
- }
-
- if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
-- goto err;
-+ goto err_mutex;
- }
-
- di->dev_id = dev_id;
-+
-+ PBSBitmapInfo *info = g_malloc(sizeof(*info));
-+ info->drive = g_strdup(devname);
-+ info->action = action;
-+ info->size = di->size;
-+ info->dirty = dirty;
-+ backup_state.stat.bitmap_list = g_list_append(backup_state.stat.bitmap_list, info);
- }
- } else if (format == BACKUP_FORMAT_VMA) {
-- dirty = total;
--
- vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
- if (!vmaw) {
- if (local_err) {
- error_propagate(task->errp, local_err);
- }
-- goto err;
-+ goto err_mutex;
- }
-
- /* register all devices for vma writer */
-@@ -807,7 +834,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- l = g_list_next(l);
-
- if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
-- goto err;
-+ goto err_mutex;
- }
-
- const char *devname = bdrv_get_device_name(di->bs);
-@@ -815,16 +842,14 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- if (di->dev_id <= 0) {
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
- "register_stream failed");
-- goto err;
-+ goto err_mutex;
- }
- }
- } else if (format == BACKUP_FORMAT_DIR) {
-- dirty = total;
--
- if (mkdir(task->backup_file, 0640) != 0) {
- error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
- task->backup_file);
-- goto err;
-+ goto err_mutex;
- }
- backup_dir = task->backup_file;
-
-@@ -841,18 +866,18 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- di->size, flags, false, &local_err);
- if (local_err) {
- error_propagate(task->errp, local_err);
-- goto err;
-+ goto err_mutex;
- }
-
- di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
- if (!di->target) {
- error_propagate(task->errp, local_err);
-- goto err;
-+ goto err_mutex;
- }
- }
- } else {
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
-- goto err;
-+ goto err_mutex;
- }
-
-
-@@ -860,7 +885,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- if (task->has_config_file) {
- if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
- vmaw, pbs, task->errp) != 0) {
-- goto err;
-+ goto err_mutex;
- }
- }
-
-@@ -868,12 +893,11 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- if (task->has_firewall_file) {
- if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
- vmaw, pbs, task->errp) != 0) {
-- goto err;
-+ goto err_mutex;
- }
- }
- /* initialize global backup_state now */
--
-- qemu_mutex_lock(&backup_state.stat.lock);
-+ /* note: 'reused' and 'bitmap_list' are initialized earlier */
-
- if (backup_state.stat.error) {
- error_free(backup_state.stat.error);
-@@ -893,10 +917,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- char *uuid_str = g_strdup(backup_state.stat.uuid_str);
-
- backup_state.stat.total = total;
-- backup_state.stat.dirty = dirty;
-+ backup_state.stat.dirty = total - backup_state.stat.reused;
- backup_state.stat.transferred = 0;
- backup_state.stat.zero_bytes = 0;
-- backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
-
- qemu_mutex_unlock(&backup_state.stat.lock);
-
-@@ -913,6 +936,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- task->result = uuid_info;
- return;
-
-+err_mutex:
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
- err:
-
- l = di_list;
-@@ -1076,11 +1102,42 @@ BackupStatus *qmp_query_backup(Error **errp)
- return info;
- }
-
-+PBSBitmapInfoList *qmp_query_pbs_bitmap_info(Error **errp)
-+{
-+ PBSBitmapInfoList *head = NULL, **p_next = &head;
-+
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+
-+ GList *l = backup_state.stat.bitmap_list;
-+ while (l) {
-+ PBSBitmapInfo *info = (PBSBitmapInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ /* clone bitmap info to avoid auto free after QMP marshalling */
-+ PBSBitmapInfo *info_ret = g_malloc0(sizeof(*info_ret));
-+ info_ret->drive = g_strdup(info->drive);
-+ info_ret->action = info->action;
-+ info_ret->size = info->size;
-+ info_ret->dirty = info->dirty;
-+
-+ PBSBitmapInfoList *info_list = g_malloc0(sizeof(*info_list));
-+ info_list->value = info_ret;
-+
-+ *p_next = info_list;
-+ p_next = &info_list->next;
-+ }
-+
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+ return head;
-+}
-+
- ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
- {
- ProxmoxSupportStatus *ret = g_malloc0(sizeof(*ret));
- ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
- ret->pbs_dirty_bitmap = true;
- ret->pbs_dirty_bitmap_savevm = true;
-+ ret->query_bitmap_info = true;
- return ret;
- }
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 7d25aa4fa1..69571d86eb 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -875,6 +875,8 @@
- # @pbs-dirty-bitmap: True if dirty-bitmap-incremental backups to PBS are
- # supported.
- #
-+# @query-bitmap-info: True if the 'query-pbs-bitmap-info' QMP call is supported.
-+#
- # @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
- # safely be set for savevm-async.
- #
-@@ -883,6 +885,7 @@
- ##
- { 'struct': 'ProxmoxSupportStatus',
- 'data': { 'pbs-dirty-bitmap': 'bool',
-+ 'query-bitmap-info': 'bool',
- 'pbs-dirty-bitmap-savevm': 'bool',
- 'pbs-library-version': 'str' } }
-
-@@ -896,6 +899,59 @@
- ##
- { 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
-
-+##
-+# @PBSBitmapAction:
-+#
-+# An action taken on a dirty-bitmap when a backup job was started.
-+#
-+# @not-used: Bitmap mode was not enabled.
-+#
-+# @not-used-removed: Bitmap mode was not enabled, but a bitmap from a
-+# previous backup still existed and was removed.
-+#
-+# @new: A new bitmap was attached to the drive for this backup.
-+#
-+# @used: An existing bitmap will be used to only backup changed data.
-+#
-+# @invalid: A bitmap existed, but had to be cleared since it's associated
-+# base snapshot did not match the base given for the current job or
-+# the crypt mode has changed.
-+#
-+##
-+{ 'enum': 'PBSBitmapAction',
-+ 'data': ['not-used', 'not-used-removed', 'new', 'used', 'invalid'] }
-+
-+##
-+# @PBSBitmapInfo:
-+#
-+# Contains information about dirty bitmaps used for each drive in a PBS backup.
-+#
-+# @drive: The underlying drive.
-+#
-+# @action: The action that was taken when the backup started.
-+#
-+# @size: The total size of the drive.
-+#
-+# @dirty: How much of the drive is considered dirty and will be backed up,
-+# or 'size' if everything will be.
-+#
-+##
-+{ 'struct': 'PBSBitmapInfo',
-+ 'data': { 'drive': 'str', 'action': 'PBSBitmapAction', 'size': 'int',
-+ 'dirty': 'int' } }
-+
-+##
-+# @query-pbs-bitmap-info:
-+#
-+# Returns information about dirty bitmaps used on the most recently started
-+# backup. Returns nothing when the last backup was not using PBS or if no
-+# backup occured in this session.
-+#
-+# Returns: @PBSBitmapInfo
-+#
-+##
-+{ 'command': 'query-pbs-bitmap-info', 'returns': ['PBSBitmapInfo'] }
-+
- ##
- # @BlockDeviceTimedStats:
- #
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 8 Jul 2020 11:57:53 +0200
+Subject: [PATCH] PVE: add query_proxmox_support QMP command
+
+Generic interface for future use, currently used for PBS dirty-bitmap
+backup support.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[PVE: query-proxmox-support: include library version]
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+---
+ pve-backup.c | 9 +++++++++
+ qapi/block-core.json | 29 +++++++++++++++++++++++++++++
+ 2 files changed, 38 insertions(+)
+
+diff --git a/pve-backup.c b/pve-backup.c
+index c15abefdda..4684789813 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -1075,3 +1075,12 @@ BackupStatus *qmp_query_backup(Error **errp)
+
+ return info;
+ }
++
++ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
++{
++ ProxmoxSupportStatus *ret = g_malloc0(sizeof(*ret));
++ ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
++ ret->pbs_dirty_bitmap = true;
++ ret->pbs_dirty_bitmap_savevm = true;
++ return ret;
++}
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 96bc696aaa..0b453c61d4 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -863,6 +863,35 @@
+ ##
+ { 'command': 'backup-cancel' }
+
++##
++# @ProxmoxSupportStatus:
++#
++# Contains info about supported features added by Proxmox.
++#
++# @pbs-dirty-bitmap: True if dirty-bitmap-incremental backups to PBS are
++# supported.
++#
++# @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
++# safely be set for savevm-async.
++#
++# @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
++#
++##
++{ 'struct': 'ProxmoxSupportStatus',
++ 'data': { 'pbs-dirty-bitmap': 'bool',
++ 'pbs-dirty-bitmap-savevm': 'bool',
++ 'pbs-library-version': 'str' } }
++
++##
++# @query-proxmox-support:
++#
++# Returns information about supported features added by Proxmox.
++#
++# Returns: @ProxmoxSupportStatus
++#
++##
++{ 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
++
+ ##
+ # @BlockDeviceTimedStats:
+ #
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 19 Aug 2020 17:02:00 +0200
+Subject: [PATCH] PVE: add query-pbs-bitmap-info QMP call
+
+Returns advanced information about dirty bitmaps used (or not used) for
+the latest PBS backup.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ monitor/hmp-cmds.c | 28 ++++++-----
+ pve-backup.c | 117 ++++++++++++++++++++++++++++++++-----------
+ qapi/block-core.json | 56 +++++++++++++++++++++
+ 3 files changed, 159 insertions(+), 42 deletions(-)
+
+diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
+index 4c1671e289..c1152f55a7 100644
+--- a/monitor/hmp-cmds.c
++++ b/monitor/hmp-cmds.c
+@@ -200,6 +200,7 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
+ void hmp_info_backup(Monitor *mon, const QDict *qdict)
+ {
+ BackupStatus *info;
++ PBSBitmapInfoList *bitmap_info;
+
+ info = qmp_query_backup(NULL);
+
+@@ -230,26 +231,29 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
+ // this should not happen normally
+ monitor_printf(mon, "Total size: %d\n", 0);
+ } else {
+- bool incremental = false;
+ size_t total_or_dirty = info->total;
+- if (info->has_transferred) {
+- if (info->has_dirty && info->dirty) {
+- if (info->dirty < info->total) {
+- total_or_dirty = info->dirty;
+- incremental = true;
+- }
+- }
++ bitmap_info = qmp_query_pbs_bitmap_info(NULL);
++
++ while (bitmap_info) {
++ monitor_printf(mon, "Drive %s:\n",
++ bitmap_info->value->drive);
++ monitor_printf(mon, " bitmap action: %s\n",
++ PBSBitmapAction_str(bitmap_info->value->action));
++ monitor_printf(mon, " size: %zd\n",
++ bitmap_info->value->size);
++ monitor_printf(mon, " dirty: %zd\n",
++ bitmap_info->value->dirty);
++ bitmap_info = bitmap_info->next;
+ }
+
+- int per = (info->transferred * 100)/total_or_dirty;
+-
+- monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
++ qapi_free_PBSBitmapInfoList(bitmap_info);
+
+ int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
+ (info->zero_bytes * 100)/info->total : 0;
+ monitor_printf(mon, "Total size: %zd\n", info->total);
++ int trans_per = (info->transferred * 100)/total_or_dirty;
+ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
+- info->transferred, per);
++ info->transferred, trans_per);
+ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
+ info->zero_bytes, zero_per);
+
+diff --git a/pve-backup.c b/pve-backup.c
+index 4684789813..f90abaa50a 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -46,6 +46,7 @@ static struct PVEBackupState {
+ size_t transferred;
+ size_t reused;
+ size_t zero_bytes;
++ GList *bitmap_list;
+ } stat;
+ int64_t speed;
+ VmaWriter *vmaw;
+@@ -672,7 +673,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+
+ size_t total = 0;
+- size_t dirty = 0;
+
+ l = di_list;
+ while (l) {
+@@ -693,18 +693,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+
+ uuid_generate(uuid);
+
++ qemu_mutex_lock(&backup_state.stat.lock);
++ backup_state.stat.reused = 0;
++
++ /* clear previous backup's bitmap_list */
++ if (backup_state.stat.bitmap_list) {
++ GList *bl = backup_state.stat.bitmap_list;
++ while (bl) {
++ g_free(((PBSBitmapInfo *)bl->data)->drive);
++ g_free(bl->data);
++ bl = g_list_next(bl);
++ }
++ g_list_free(backup_state.stat.bitmap_list);
++ backup_state.stat.bitmap_list = NULL;
++ }
++
+ if (format == BACKUP_FORMAT_PBS) {
+ if (!task->has_password) {
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
+- goto err;
++ goto err_mutex;
+ }
+ if (!task->has_backup_id) {
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
+- goto err;
++ goto err_mutex;
+ }
+ if (!task->has_backup_time) {
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
+- goto err;
++ goto err_mutex;
+ }
+
+ int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
+@@ -731,12 +746,12 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
+ "proxmox_backup_new failed: %s", pbs_err);
+ proxmox_backup_free_error(pbs_err);
+- goto err;
++ goto err_mutex;
+ }
+
+ int connect_result = proxmox_backup_co_connect(pbs, task->errp);
+ if (connect_result < 0)
+- goto err;
++ goto err_mutex;
+
+ /* register all devices */
+ l = di_list;
+@@ -747,6 +762,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ di->block_size = dump_cb_block_size;
+
+ const char *devname = bdrv_get_device_name(di->bs);
++ PBSBitmapAction action = PBS_BITMAP_ACTION_NOT_USED;
++ size_t dirty = di->size;
+
+ BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
+ bool expect_only_dirty = false;
+@@ -755,49 +772,59 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ if (bitmap == NULL) {
+ bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
+ if (!bitmap) {
+- goto err;
++ goto err_mutex;
+ }
++ action = PBS_BITMAP_ACTION_NEW;
+ } else {
+ expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
+ }
+
+ if (expect_only_dirty) {
+- dirty += bdrv_get_dirty_count(bitmap);
++ /* track clean chunks as reused */
++ dirty = MIN(bdrv_get_dirty_count(bitmap), di->size);
++ backup_state.stat.reused += di->size - dirty;
++ action = PBS_BITMAP_ACTION_USED;
+ } else {
+ /* mark entire bitmap as dirty to make full backup */
+ bdrv_set_dirty_bitmap(bitmap, 0, di->size);
+- dirty += di->size;
++ if (action != PBS_BITMAP_ACTION_NEW) {
++ action = PBS_BITMAP_ACTION_INVALID;
++ }
+ }
+ di->bitmap = bitmap;
+ } else {
+- dirty += di->size;
+-
+ /* after a full backup the old dirty bitmap is invalid anyway */
+ if (bitmap != NULL) {
+ bdrv_release_dirty_bitmap(bitmap);
++ action = PBS_BITMAP_ACTION_NOT_USED_REMOVED;
+ }
+ }
+
+ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
+ if (dev_id < 0) {
+- goto err;
++ goto err_mutex;
+ }
+
+ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
+- goto err;
++ goto err_mutex;
+ }
+
+ di->dev_id = dev_id;
++
++ PBSBitmapInfo *info = g_malloc(sizeof(*info));
++ info->drive = g_strdup(devname);
++ info->action = action;
++ info->size = di->size;
++ info->dirty = dirty;
++ backup_state.stat.bitmap_list = g_list_append(backup_state.stat.bitmap_list, info);
+ }
+ } else if (format == BACKUP_FORMAT_VMA) {
+- dirty = total;
+-
+ vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
+ if (!vmaw) {
+ if (local_err) {
+ error_propagate(task->errp, local_err);
+ }
+- goto err;
++ goto err_mutex;
+ }
+
+ /* register all devices for vma writer */
+@@ -807,7 +834,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ l = g_list_next(l);
+
+ if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
+- goto err;
++ goto err_mutex;
+ }
+
+ const char *devname = bdrv_get_device_name(di->bs);
+@@ -815,16 +842,14 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ if (di->dev_id <= 0) {
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
+ "register_stream failed");
+- goto err;
++ goto err_mutex;
+ }
+ }
+ } else if (format == BACKUP_FORMAT_DIR) {
+- dirty = total;
+-
+ if (mkdir(task->backup_file, 0640) != 0) {
+ error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
+ task->backup_file);
+- goto err;
++ goto err_mutex;
+ }
+ backup_dir = task->backup_file;
+
+@@ -841,18 +866,18 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ di->size, flags, false, &local_err);
+ if (local_err) {
+ error_propagate(task->errp, local_err);
+- goto err;
++ goto err_mutex;
+ }
+
+ di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
+ if (!di->target) {
+ error_propagate(task->errp, local_err);
+- goto err;
++ goto err_mutex;
+ }
+ }
+ } else {
+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
+- goto err;
++ goto err_mutex;
+ }
+
+
+@@ -860,7 +885,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ if (task->has_config_file) {
+ if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
+ vmaw, pbs, task->errp) != 0) {
+- goto err;
++ goto err_mutex;
+ }
+ }
+
+@@ -868,12 +893,11 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ if (task->has_firewall_file) {
+ if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
+ vmaw, pbs, task->errp) != 0) {
+- goto err;
++ goto err_mutex;
+ }
+ }
+ /* initialize global backup_state now */
+-
+- qemu_mutex_lock(&backup_state.stat.lock);
++ /* note: 'reused' and 'bitmap_list' are initialized earlier */
+
+ if (backup_state.stat.error) {
+ error_free(backup_state.stat.error);
+@@ -893,10 +917,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ char *uuid_str = g_strdup(backup_state.stat.uuid_str);
+
+ backup_state.stat.total = total;
+- backup_state.stat.dirty = dirty;
++ backup_state.stat.dirty = total - backup_state.stat.reused;
+ backup_state.stat.transferred = 0;
+ backup_state.stat.zero_bytes = 0;
+- backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
+
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+@@ -913,6 +936,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ task->result = uuid_info;
+ return;
+
++err_mutex:
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
+ err:
+
+ l = di_list;
+@@ -1076,11 +1102,42 @@ BackupStatus *qmp_query_backup(Error **errp)
+ return info;
+ }
+
++PBSBitmapInfoList *qmp_query_pbs_bitmap_info(Error **errp)
++{
++ PBSBitmapInfoList *head = NULL, **p_next = &head;
++
++ qemu_mutex_lock(&backup_state.stat.lock);
++
++ GList *l = backup_state.stat.bitmap_list;
++ while (l) {
++ PBSBitmapInfo *info = (PBSBitmapInfo *)l->data;
++ l = g_list_next(l);
++
++ /* clone bitmap info to avoid auto free after QMP marshalling */
++ PBSBitmapInfo *info_ret = g_malloc0(sizeof(*info_ret));
++ info_ret->drive = g_strdup(info->drive);
++ info_ret->action = info->action;
++ info_ret->size = info->size;
++ info_ret->dirty = info->dirty;
++
++ PBSBitmapInfoList *info_list = g_malloc0(sizeof(*info_list));
++ info_list->value = info_ret;
++
++ *p_next = info_list;
++ p_next = &info_list->next;
++ }
++
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
++ return head;
++}
++
+ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+ {
+ ProxmoxSupportStatus *ret = g_malloc0(sizeof(*ret));
+ ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
+ ret->pbs_dirty_bitmap = true;
+ ret->pbs_dirty_bitmap_savevm = true;
++ ret->query_bitmap_info = true;
+ return ret;
+ }
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 0b453c61d4..16e184dd28 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -871,6 +871,8 @@
+ # @pbs-dirty-bitmap: True if dirty-bitmap-incremental backups to PBS are
+ # supported.
+ #
++# @query-bitmap-info: True if the 'query-pbs-bitmap-info' QMP call is supported.
++#
+ # @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
+ # safely be set for savevm-async.
+ #
+@@ -879,6 +881,7 @@
+ ##
+ { 'struct': 'ProxmoxSupportStatus',
+ 'data': { 'pbs-dirty-bitmap': 'bool',
++ 'query-bitmap-info': 'bool',
+ 'pbs-dirty-bitmap-savevm': 'bool',
+ 'pbs-library-version': 'str' } }
+
+@@ -892,6 +895,59 @@
+ ##
+ { 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
+
++##
++# @PBSBitmapAction:
++#
++# An action taken on a dirty-bitmap when a backup job was started.
++#
++# @not-used: Bitmap mode was not enabled.
++#
++# @not-used-removed: Bitmap mode was not enabled, but a bitmap from a
++# previous backup still existed and was removed.
++#
++# @new: A new bitmap was attached to the drive for this backup.
++#
++# @used: An existing bitmap will be used to only backup changed data.
++#
++# @invalid: A bitmap existed, but had to be cleared since it's associated
++# base snapshot did not match the base given for the current job or
++# the crypt mode has changed.
++#
++##
++{ 'enum': 'PBSBitmapAction',
++ 'data': ['not-used', 'not-used-removed', 'new', 'used', 'invalid'] }
++
++##
++# @PBSBitmapInfo:
++#
++# Contains information about dirty bitmaps used for each drive in a PBS backup.
++#
++# @drive: The underlying drive.
++#
++# @action: The action that was taken when the backup started.
++#
++# @size: The total size of the drive.
++#
++# @dirty: How much of the drive is considered dirty and will be backed up,
++# or 'size' if everything will be.
++#
++##
++{ 'struct': 'PBSBitmapInfo',
++ 'data': { 'drive': 'str', 'action': 'PBSBitmapAction', 'size': 'int',
++ 'dirty': 'int' } }
++
++##
++# @query-pbs-bitmap-info:
++#
++# Returns information about dirty bitmaps used on the most recently started
++# backup. Returns nothing when the last backup was not using PBS or if no
++# backup occured in this session.
++#
++# Returns: @PBSBitmapInfo
++#
++##
++{ 'command': 'query-pbs-bitmap-info', 'returns': ['PBSBitmapInfo'] }
++
+ ##
+ # @BlockDeviceTimedStats:
+ #
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Tue, 12 Jan 2021 14:12:20 +0100
-Subject: [PATCH] PVE: redirect stderr to journal when daemonized
-
-QEMU uses the logging for error messages usually, so LOG_ERR is most
-fitting.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- meson.build | 2 ++
- os-posix.c | 7 +++++--
- 2 files changed, 7 insertions(+), 2 deletions(-)
-
-diff --git a/meson.build b/meson.build
-index d3783a83e4..715245ba20 100644
---- a/meson.build
-+++ b/meson.build
-@@ -1268,6 +1268,7 @@ keyutils = dependency('libkeyutils', required: false,
- has_gettid = cc.has_function('gettid')
-
- libuuid = cc.find_library('uuid', required: true)
-+libsystemd = cc.find_library('systemd', required: true)
- libproxmox_backup_qemu = cc.find_library('proxmox_backup_qemu', required: true)
-
- # libselinux
-@@ -2861,6 +2862,7 @@ if have_block
- # os-posix.c contains POSIX-specific functions used by qemu-storage-daemon,
- # os-win32.c does not
- blockdev_ss.add(when: 'CONFIG_POSIX', if_true: files('os-posix.c'))
-+ blockdev_ss.add(when: 'CONFIG_POSIX', if_true: libsystemd)
- softmmu_ss.add(when: 'CONFIG_WIN32', if_true: [files('os-win32.c')])
- endif
-
-diff --git a/os-posix.c b/os-posix.c
-index 24692c8593..d1bc5ac12d 100644
---- a/os-posix.c
-+++ b/os-posix.c
-@@ -28,6 +28,8 @@
- #include <pwd.h>
- #include <grp.h>
- #include <libgen.h>
-+#include <systemd/sd-journal.h>
-+#include <syslog.h>
-
- #include "qemu-common.h"
- /* Needed early for CONFIG_BSD etc. */
-@@ -291,9 +293,10 @@ void os_setup_post(void)
-
- dup2(fd, 0);
- dup2(fd, 1);
-- /* In case -D is given do not redirect stderr to /dev/null */
-+ /* In case -D is given do not redirect stderr to journal */
- if (!qemu_logfile) {
-- dup2(fd, 2);
-+ int journal_fd = sd_journal_stream_fd("QEMU", LOG_ERR, 0);
-+ dup2(journal_fd, 2);
- }
-
- close(fd);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 20 Aug 2020 14:31:59 +0200
-Subject: [PATCH] PVE: Add sequential job transaction support
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- include/qemu/job.h | 12 ++++++++++++
- job.c | 31 +++++++++++++++++++++++++++++++
- 2 files changed, 43 insertions(+)
-
-diff --git a/include/qemu/job.h b/include/qemu/job.h
-index c105b31076..5096679571 100644
---- a/include/qemu/job.h
-+++ b/include/qemu/job.h
-@@ -316,6 +316,18 @@ typedef enum JobCreateFlags {
- */
- JobTxn *job_txn_new(void);
-
-+/**
-+ * Create a new transaction and set it to sequential mode, i.e. run all jobs
-+ * one after the other instead of at the same time.
-+ */
-+JobTxn *job_txn_new_seq(void);
-+
-+/**
-+ * Helper method to start the first job in a sequential transaction to kick it
-+ * off. Other jobs will be run after this one completes.
-+ */
-+void job_txn_start_seq(JobTxn *txn);
-+
- /**
- * Release a reference that was previously acquired with job_txn_add_job or
- * job_txn_new. If it's the last reference to the object, it will be freed.
-diff --git a/job.c b/job.c
-index e5699ad200..34c9758349 100644
---- a/job.c
-+++ b/job.c
-@@ -72,6 +72,8 @@ struct JobTxn {
-
- /* Reference count */
- int refcnt;
-+
-+ bool sequential;
- };
-
- /* Right now, this mutex is only needed to synchronize accesses to job->busy
-@@ -102,6 +104,25 @@ JobTxn *job_txn_new(void)
- return txn;
- }
-
-+JobTxn *job_txn_new_seq(void)
-+{
-+ JobTxn *txn = job_txn_new();
-+ txn->sequential = true;
-+ return txn;
-+}
-+
-+void job_txn_start_seq(JobTxn *txn)
-+{
-+ assert(txn->sequential);
-+ assert(!txn->aborting);
-+
-+ Job *first = QLIST_FIRST(&txn->jobs);
-+ assert(first);
-+ assert(first->status == JOB_STATUS_CREATED);
-+
-+ job_start(first);
-+}
-+
- static void job_txn_ref(JobTxn *txn)
- {
- txn->refcnt++;
-@@ -897,6 +918,9 @@ static void job_completed_txn_success(Job *job)
- */
- QLIST_FOREACH(other_job, &txn->jobs, txn_list) {
- if (!job_is_completed(other_job)) {
-+ if (txn->sequential) {
-+ job_start(other_job);
-+ }
- return;
- }
- assert(other_job->ret == 0);
-@@ -1093,6 +1117,13 @@ int job_finish_sync(Job *job, void (*finish)(Job *, Error **errp), Error **errp)
- return -EBUSY;
- }
-
-+ /* in a sequential transaction jobs with status CREATED can appear at time
-+ * of cancelling, these have not begun work so job_enter won't do anything,
-+ * let's ensure they are marked as ABORTING if required */
-+ if (job->status == JOB_STATUS_CREATED && job->txn->sequential) {
-+ job_update_rc(job);
-+ }
-+
- AIO_WAIT_WHILE(job->aio_context,
- (job_enter(job), !job_is_completed(job)));
-
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Tue, 12 Jan 2021 14:12:20 +0100
+Subject: [PATCH] PVE: redirect stderr to journal when daemonized
+
+QEMU uses the logging for error messages usually, so LOG_ERR is most
+fitting.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ meson.build | 2 ++
+ os-posix.c | 7 +++++--
+ 2 files changed, 7 insertions(+), 2 deletions(-)
+
+diff --git a/meson.build b/meson.build
+index be4785e2f6..3fc7c8d435 100644
+--- a/meson.build
++++ b/meson.build
+@@ -1463,6 +1463,7 @@ keyutils = dependency('libkeyutils', required: false,
+ has_gettid = cc.has_function('gettid')
+
+ libuuid = cc.find_library('uuid', required: true)
++libsystemd = cc.find_library('systemd', required: true)
+ libproxmox_backup_qemu = cc.find_library('proxmox_backup_qemu', required: true)
+
+ # libselinux
+@@ -3105,6 +3106,7 @@ if have_block
+ # os-posix.c contains POSIX-specific functions used by qemu-storage-daemon,
+ # os-win32.c does not
+ blockdev_ss.add(when: 'CONFIG_POSIX', if_true: files('os-posix.c'))
++ blockdev_ss.add(when: 'CONFIG_POSIX', if_true: libsystemd)
+ softmmu_ss.add(when: 'CONFIG_WIN32', if_true: [files('os-win32.c')])
+ endif
+
+diff --git a/os-posix.c b/os-posix.c
+index 321fc4bd13..b1870d2690 100644
+--- a/os-posix.c
++++ b/os-posix.c
+@@ -28,6 +28,8 @@
+ #include <pwd.h>
+ #include <grp.h>
+ #include <libgen.h>
++#include <systemd/sd-journal.h>
++#include <syslog.h>
+
+ /* Needed early for CONFIG_BSD etc. */
+ #include "net/slirp.h"
+@@ -281,9 +283,10 @@ void os_setup_post(void)
+
+ dup2(fd, 0);
+ dup2(fd, 1);
+- /* In case -D is given do not redirect stderr to /dev/null */
++ /* In case -D is given do not redirect stderr to journal */
+ if (!qemu_log_enabled()) {
+- dup2(fd, 2);
++ int journal_fd = sd_journal_stream_fd("QEMU", LOG_ERR, 0);
++ dup2(journal_fd, 2);
+ }
+
+ close(fd);
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Thu, 20 Aug 2020 14:31:59 +0200
+Subject: [PATCH] PVE: Add sequential job transaction support
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ include/qemu/job.h | 12 ++++++++++++
+ job.c | 31 +++++++++++++++++++++++++++++++
+ 2 files changed, 43 insertions(+)
+
+diff --git a/include/qemu/job.h b/include/qemu/job.h
+index c105b31076..5096679571 100644
+--- a/include/qemu/job.h
++++ b/include/qemu/job.h
+@@ -316,6 +316,18 @@ typedef enum JobCreateFlags {
+ */
+ JobTxn *job_txn_new(void);
+
++/**
++ * Create a new transaction and set it to sequential mode, i.e. run all jobs
++ * one after the other instead of at the same time.
++ */
++JobTxn *job_txn_new_seq(void);
++
++/**
++ * Helper method to start the first job in a sequential transaction to kick it
++ * off. Other jobs will be run after this one completes.
++ */
++void job_txn_start_seq(JobTxn *txn);
++
+ /**
+ * Release a reference that was previously acquired with job_txn_add_job or
+ * job_txn_new. If it's the last reference to the object, it will be freed.
+diff --git a/job.c b/job.c
+index e5699ad200..34c9758349 100644
+--- a/job.c
++++ b/job.c
+@@ -72,6 +72,8 @@ struct JobTxn {
+
+ /* Reference count */
+ int refcnt;
++
++ bool sequential;
+ };
+
+ /* Right now, this mutex is only needed to synchronize accesses to job->busy
+@@ -102,6 +104,25 @@ JobTxn *job_txn_new(void)
+ return txn;
+ }
+
++JobTxn *job_txn_new_seq(void)
++{
++ JobTxn *txn = job_txn_new();
++ txn->sequential = true;
++ return txn;
++}
++
++void job_txn_start_seq(JobTxn *txn)
++{
++ assert(txn->sequential);
++ assert(!txn->aborting);
++
++ Job *first = QLIST_FIRST(&txn->jobs);
++ assert(first);
++ assert(first->status == JOB_STATUS_CREATED);
++
++ job_start(first);
++}
++
+ static void job_txn_ref(JobTxn *txn)
+ {
+ txn->refcnt++;
+@@ -897,6 +918,9 @@ static void job_completed_txn_success(Job *job)
+ */
+ QLIST_FOREACH(other_job, &txn->jobs, txn_list) {
+ if (!job_is_completed(other_job)) {
++ if (txn->sequential) {
++ job_start(other_job);
++ }
+ return;
+ }
+ assert(other_job->ret == 0);
+@@ -1093,6 +1117,13 @@ int job_finish_sync(Job *job, void (*finish)(Job *, Error **errp), Error **errp)
+ return -EBUSY;
+ }
+
++ /* in a sequential transaction jobs with status CREATED can appear at time
++ * of cancelling, these have not begun work so job_enter won't do anything,
++ * let's ensure they are marked as ABORTING if required */
++ if (job->status == JOB_STATUS_CREATED && job->txn->sequential) {
++ job_update_rc(job);
++ }
++
+ AIO_WAIT_WHILE(job->aio_context,
+ (job_enter(job), !job_is_completed(job)));
+
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 20 Aug 2020 14:25:00 +0200
-Subject: [PATCH] PVE-Backup: Use a transaction to synchronize job states
-
-By using a JobTxn, we can sync dirty bitmaps only when *all* jobs were
-successful - meaning we don't need to remove them when the backup fails,
-since QEMU's BITMAP_SYNC_MODE_ON_SUCCESS will now handle that for us.
-
-To keep the rate-limiting and IO impact from before, we use a sequential
-transaction, so drives will still be backed up one after the other.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[add new force parameter to job_cancel_sync calls]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- pve-backup.c | 167 +++++++++++++++------------------------------------
- 1 file changed, 49 insertions(+), 118 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index f90abaa50a..63c686463f 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -52,6 +52,7 @@ static struct PVEBackupState {
- VmaWriter *vmaw;
- ProxmoxBackupHandle *pbs;
- GList *di_list;
-+ JobTxn *txn;
- QemuMutex backup_mutex;
- CoMutex dump_callback_mutex;
- } backup_state;
-@@ -71,32 +72,12 @@ typedef struct PVEBackupDevInfo {
- size_t size;
- uint64_t block_size;
- uint8_t dev_id;
-- bool completed;
- char targetfile[PATH_MAX];
- BdrvDirtyBitmap *bitmap;
- BlockDriverState *target;
-+ BlockJob *job;
- } PVEBackupDevInfo;
-
--static void pvebackup_run_next_job(void);
--
--static BlockJob *
--lookup_active_block_job(PVEBackupDevInfo *di)
--{
-- if (!di->completed && di->bs) {
-- for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
-- if (job->job.driver->job_type != JOB_TYPE_BACKUP) {
-- continue;
-- }
--
-- BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
-- if (bjob && bjob->source_bs == di->bs) {
-- return job;
-- }
-- }
-- }
-- return NULL;
--}
--
- static void pvebackup_propagate_error(Error *err)
- {
- qemu_mutex_lock(&backup_state.stat.lock);
-@@ -272,18 +253,6 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
- if (local_err != NULL) {
- pvebackup_propagate_error(local_err);
- }
-- } else {
-- // on error or cancel we cannot ensure synchronization of dirty
-- // bitmaps with backup server, so remove all and do full backup next
-- GList *l = backup_state.di_list;
-- while (l) {
-- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-- l = g_list_next(l);
--
-- if (di->bitmap) {
-- bdrv_release_dirty_bitmap(di->bitmap);
-- }
-- }
- }
-
- proxmox_backup_disconnect(backup_state.pbs);
-@@ -322,8 +291,6 @@ static void pvebackup_complete_cb(void *opaque, int ret)
-
- qemu_mutex_lock(&backup_state.backup_mutex);
-
-- di->completed = true;
--
- if (ret < 0) {
- Error *local_err = NULL;
- error_setg(&local_err, "job failed with err %d - %s", ret, strerror(-ret));
-@@ -336,20 +303,17 @@ static void pvebackup_complete_cb(void *opaque, int ret)
-
- block_on_coroutine_fn(pvebackup_complete_stream, di);
-
-- // remove self from job queue
-+ // remove self from job list
- backup_state.di_list = g_list_remove(backup_state.di_list, di);
-
-- if (di->bitmap && ret < 0) {
-- // on error or cancel we cannot ensure synchronization of dirty
-- // bitmaps with backup server, so remove all and do full backup next
-- bdrv_release_dirty_bitmap(di->bitmap);
-- }
--
- g_free(di);
-
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ /* call cleanup if we're the last job */
-+ if (!g_list_first(backup_state.di_list)) {
-+ block_on_coroutine_fn(pvebackup_co_cleanup, NULL);
-+ }
-
-- pvebackup_run_next_job();
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
- }
-
- static void pvebackup_cancel(void)
-@@ -371,36 +335,28 @@ static void pvebackup_cancel(void)
- proxmox_backup_abort(backup_state.pbs, "backup canceled");
- }
-
-- qemu_mutex_unlock(&backup_state.backup_mutex);
--
-- for(;;) {
--
-- BlockJob *next_job = NULL;
--
-- qemu_mutex_lock(&backup_state.backup_mutex);
--
-- GList *l = backup_state.di_list;
-- while (l) {
-- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-- l = g_list_next(l);
-+ /* it's enough to cancel one job in the transaction, the rest will follow
-+ * automatically */
-+ GList *bdi = g_list_first(backup_state.di_list);
-+ BlockJob *cancel_job = bdi && bdi->data ?
-+ ((PVEBackupDevInfo *)bdi->data)->job :
-+ NULL;
-
-- BlockJob *job = lookup_active_block_job(di);
-- if (job != NULL) {
-- next_job = job;
-- break;
-- }
-- }
-+ /* ref the job before releasing the mutex, just to be safe */
-+ if (cancel_job) {
-+ job_ref(&cancel_job->job);
-+ }
-
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ /* job_cancel_sync may enter the job, so we need to release the
-+ * backup_mutex to avoid deadlock */
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-
-- if (next_job) {
-- AioContext *aio_context = next_job->job.aio_context;
-- aio_context_acquire(aio_context);
-- job_cancel_sync(&next_job->job, true);
-- aio_context_release(aio_context);
-- } else {
-- break;
-- }
-+ if (cancel_job) {
-+ AioContext *aio_context = cancel_job->job.aio_context;
-+ aio_context_acquire(aio_context);
-+ job_cancel_sync(&cancel_job->job, true);
-+ job_unref(&cancel_job->job);
-+ aio_context_release(aio_context);
- }
- }
-
-@@ -459,51 +415,19 @@ static int coroutine_fn pvebackup_co_add_config(
- goto out;
- }
-
--bool job_should_pause(Job *job);
--
--static void pvebackup_run_next_job(void)
--{
-- assert(!qemu_in_coroutine());
--
-- qemu_mutex_lock(&backup_state.backup_mutex);
--
-- GList *l = backup_state.di_list;
-- while (l) {
-- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-- l = g_list_next(l);
--
-- BlockJob *job = lookup_active_block_job(di);
--
-- if (job) {
-- qemu_mutex_unlock(&backup_state.backup_mutex);
--
-- AioContext *aio_context = job->job.aio_context;
-- aio_context_acquire(aio_context);
--
-- if (job_should_pause(&job->job)) {
-- bool error_or_canceled = pvebackup_error_or_canceled();
-- if (error_or_canceled) {
-- job_cancel_sync(&job->job, true);
-- } else {
-- job_resume(&job->job);
-- }
-- }
-- aio_context_release(aio_context);
-- return;
-- }
-- }
--
-- block_on_coroutine_fn(pvebackup_co_cleanup, NULL); // no more jobs, run cleanup
--
-- qemu_mutex_unlock(&backup_state.backup_mutex);
--}
--
- static bool create_backup_jobs(void) {
-
- assert(!qemu_in_coroutine());
-
- Error *local_err = NULL;
-
-+ /* create job transaction to synchronize bitmap commit and cancel all
-+ * jobs in case one errors */
-+ if (backup_state.txn) {
-+ job_txn_unref(backup_state.txn);
-+ }
-+ backup_state.txn = job_txn_new_seq();
-+
- BackupPerf perf = { .max_workers = 16 };
-
- /* create and start all jobs (paused state) */
-@@ -526,7 +450,7 @@ static bool create_backup_jobs(void) {
- BlockJob *job = backup_job_create(
- NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
- bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-- JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
-+ JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn, &local_err);
-
- aio_context_release(aio_context);
-
-@@ -538,7 +462,8 @@ static bool create_backup_jobs(void) {
- pvebackup_propagate_error(create_job_err);
- break;
- }
-- job_start(&job->job);
-+
-+ di->job = job;
-
- bdrv_unref(di->target);
- di->target = NULL;
-@@ -556,6 +481,10 @@ static bool create_backup_jobs(void) {
- bdrv_unref(di->target);
- di->target = NULL;
- }
-+
-+ if (di->job) {
-+ job_unref(&di->job->job);
-+ }
- }
- }
-
-@@ -946,10 +875,6 @@ err:
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-
-- if (di->bitmap) {
-- bdrv_release_dirty_bitmap(di->bitmap);
-- }
--
- if (di->target) {
- bdrv_unref(di->target);
- }
-@@ -1038,9 +963,15 @@ UuidInfo *qmp_backup(
- block_on_coroutine_fn(pvebackup_co_prepare, &task);
-
- if (*errp == NULL) {
-- create_backup_jobs();
-+ bool errors = create_backup_jobs();
- qemu_mutex_unlock(&backup_state.backup_mutex);
-- pvebackup_run_next_job();
-+
-+ if (!errors) {
-+ /* start the first job in the transaction
-+ * note: this might directly enter the job, so we need to do this
-+ * after unlocking the backup_mutex */
-+ job_txn_start_seq(backup_state.txn);
-+ }
- } else {
- qemu_mutex_unlock(&backup_state.backup_mutex);
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 28 Sep 2020 13:40:51 +0200
-Subject: [PATCH] PVE-Backup: Don't block on finishing and cleanup
- create_backup_jobs
-
-proxmox_backup_co_finish is already async, but previously we would wait
-for the coroutine using block_on_coroutine_fn(). Avoid this by
-scheduling pvebackup_co_complete_stream (and thus pvebackup_co_cleanup)
-as a real coroutine when calling from pvebackup_complete_cb. This is ok,
-since complete_stream uses the backup_mutex internally to synchronize,
-and other streams can happily continue writing in the meantime anyway.
-
-To accomodate, backup_mutex is converted to a CoMutex. This means
-converting every user to a coroutine. This is not just useful here, but
-will come in handy once this series[0] is merged, and QMP calls can be
-yield-able coroutines too. Then we can also finally get rid of
-block_on_coroutine_fn.
-
-Cases of aio_context_acquire/release from within what is now a coroutine
-are changed to aio_co_reschedule_self, which works since a running
-coroutine always holds the aio lock for the context it is running in.
-
-job_cancel_sync is called from a BH since it can't be run from a
-coroutine (uses AIO_WAIT_WHILE internally).
-
-Same thing for create_backup_jobs, which is converted to a BH too.
-
-To communicate the finishing state, a new property is introduced to
-query-backup: 'finishing'. A new state is explicitly not used, since
-that would break compatibility with older qemu-server versions.
-
-Also fix create_backup_jobs:
-
-No more weird bool returns, just the standard "errp" format used
-everywhere else too. With this, if backup_job_create fails, the error
-message is actually returned over QMP and can be shown to the user.
-
-To facilitate correct cleanup on such an error, we call
-create_backup_jobs as a bottom half directly from pvebackup_co_prepare.
-This additionally allows us to actually hold the backup_mutex during
-operation.
-
-Also add a job_cancel_sync before job_unref, since a job must be in
-STATUS_NULL to be deleted by unref, which could trigger an assert
-before.
-
-[0] https://lists.gnu.org/archive/html/qemu-devel/2020-09/msg03515.html
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[add new force parameter to job_cancel_sync calls]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- pve-backup.c | 217 ++++++++++++++++++++++++++++---------------
- qapi/block-core.json | 5 +-
- 2 files changed, 144 insertions(+), 78 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 63c686463f..6f05796fad 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -33,7 +33,9 @@ const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
-
- static struct PVEBackupState {
- struct {
-- // Everithing accessed from qmp_backup_query command is protected using lock
-+ // Everything accessed from qmp_backup_query command is protected using
-+ // this lock. Do NOT hold this lock for long times, as it is sometimes
-+ // acquired from coroutines, and thus any wait time may block the guest.
- QemuMutex lock;
- Error *error;
- time_t start_time;
-@@ -47,20 +49,22 @@ static struct PVEBackupState {
- size_t reused;
- size_t zero_bytes;
- GList *bitmap_list;
-+ bool finishing;
-+ bool starting;
- } stat;
- int64_t speed;
- VmaWriter *vmaw;
- ProxmoxBackupHandle *pbs;
- GList *di_list;
- JobTxn *txn;
-- QemuMutex backup_mutex;
-+ CoMutex backup_mutex;
- CoMutex dump_callback_mutex;
- } backup_state;
-
- static void pvebackup_init(void)
- {
- qemu_mutex_init(&backup_state.stat.lock);
-- qemu_mutex_init(&backup_state.backup_mutex);
-+ qemu_co_mutex_init(&backup_state.backup_mutex);
- qemu_co_mutex_init(&backup_state.dump_callback_mutex);
- }
-
-@@ -72,6 +76,7 @@ typedef struct PVEBackupDevInfo {
- size_t size;
- uint64_t block_size;
- uint8_t dev_id;
-+ int completed_ret; // INT_MAX if not completed
- char targetfile[PATH_MAX];
- BdrvDirtyBitmap *bitmap;
- BlockDriverState *target;
-@@ -227,12 +232,12 @@ pvebackup_co_dump_vma_cb(
- }
-
- // assumes the caller holds backup_mutex
--static void coroutine_fn pvebackup_co_cleanup(void *unused)
-+static void coroutine_fn pvebackup_co_cleanup(void)
- {
- assert(qemu_in_coroutine());
-
- qemu_mutex_lock(&backup_state.stat.lock);
-- backup_state.stat.end_time = time(NULL);
-+ backup_state.stat.finishing = true;
- qemu_mutex_unlock(&backup_state.stat.lock);
-
- if (backup_state.vmaw) {
-@@ -261,35 +266,29 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
-
- g_list_free(backup_state.di_list);
- backup_state.di_list = NULL;
-+
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ backup_state.stat.end_time = time(NULL);
-+ backup_state.stat.finishing = false;
-+ qemu_mutex_unlock(&backup_state.stat.lock);
- }
-
--// assumes the caller holds backup_mutex
--static void coroutine_fn pvebackup_complete_stream(void *opaque)
-+static void coroutine_fn pvebackup_co_complete_stream(void *opaque)
- {
- PVEBackupDevInfo *di = opaque;
-+ int ret = di->completed_ret;
-
-- bool error_or_canceled = pvebackup_error_or_canceled();
--
-- if (backup_state.vmaw) {
-- vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ bool starting = backup_state.stat.starting;
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+ if (starting) {
-+ /* in 'starting' state, no tasks have been run yet, meaning we can (and
-+ * must) skip all cleanup, as we don't know what has and hasn't been
-+ * initialized yet. */
-+ return;
- }
-
-- if (backup_state.pbs && !error_or_canceled) {
-- Error *local_err = NULL;
-- proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
-- if (local_err != NULL) {
-- pvebackup_propagate_error(local_err);
-- }
-- }
--}
--
--static void pvebackup_complete_cb(void *opaque, int ret)
--{
-- assert(!qemu_in_coroutine());
--
-- PVEBackupDevInfo *di = opaque;
--
-- qemu_mutex_lock(&backup_state.backup_mutex);
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-
- if (ret < 0) {
- Error *local_err = NULL;
-@@ -301,7 +300,19 @@ static void pvebackup_complete_cb(void *opaque, int ret)
-
- assert(di->target == NULL);
-
-- block_on_coroutine_fn(pvebackup_complete_stream, di);
-+ bool error_or_canceled = pvebackup_error_or_canceled();
-+
-+ if (backup_state.vmaw) {
-+ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-+ }
-+
-+ if (backup_state.pbs && !error_or_canceled) {
-+ Error *local_err = NULL;
-+ proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
-+ if (local_err != NULL) {
-+ pvebackup_propagate_error(local_err);
-+ }
-+ }
-
- // remove self from job list
- backup_state.di_list = g_list_remove(backup_state.di_list, di);
-@@ -310,21 +321,49 @@ static void pvebackup_complete_cb(void *opaque, int ret)
-
- /* call cleanup if we're the last job */
- if (!g_list_first(backup_state.di_list)) {
-- block_on_coroutine_fn(pvebackup_co_cleanup, NULL);
-+ pvebackup_co_cleanup();
- }
-
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
-
--static void pvebackup_cancel(void)
-+static void pvebackup_complete_cb(void *opaque, int ret)
- {
-- assert(!qemu_in_coroutine());
-+ PVEBackupDevInfo *di = opaque;
-+ di->completed_ret = ret;
-+
-+ /*
-+ * Schedule stream cleanup in async coroutine. close_image and finish might
-+ * take a while, so we can't block on them here. This way it also doesn't
-+ * matter if we're already running in a coroutine or not.
-+ * Note: di is a pointer to an entry in the global backup_state struct, so
-+ * it stays valid.
-+ */
-+ Coroutine *co = qemu_coroutine_create(pvebackup_co_complete_stream, di);
-+ aio_co_enter(qemu_get_aio_context(), co);
-+}
-+
-+/*
-+ * job_cancel(_sync) does not like to be called from coroutines, so defer to
-+ * main loop processing via a bottom half.
-+ */
-+static void job_cancel_bh(void *opaque) {
-+ CoCtxData *data = (CoCtxData*)opaque;
-+ Job *job = (Job*)data->data;
-+ AioContext *job_ctx = job->aio_context;
-+ aio_context_acquire(job_ctx);
-+ job_cancel_sync(job, true);
-+ aio_context_release(job_ctx);
-+ aio_co_enter(data->ctx, data->co);
-+}
-
-+static void coroutine_fn pvebackup_co_cancel(void *opaque)
-+{
- Error *cancel_err = NULL;
- error_setg(&cancel_err, "backup canceled");
- pvebackup_propagate_error(cancel_err);
-
-- qemu_mutex_lock(&backup_state.backup_mutex);
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-
- if (backup_state.vmaw) {
- /* make sure vma writer does not block anymore */
-@@ -342,27 +381,22 @@ static void pvebackup_cancel(void)
- ((PVEBackupDevInfo *)bdi->data)->job :
- NULL;
-
-- /* ref the job before releasing the mutex, just to be safe */
- if (cancel_job) {
-- job_ref(&cancel_job->job);
-+ CoCtxData data = {
-+ .ctx = qemu_get_current_aio_context(),
-+ .co = qemu_coroutine_self(),
-+ .data = &cancel_job->job,
-+ };
-+ aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
-+ qemu_coroutine_yield();
- }
-
-- /* job_cancel_sync may enter the job, so we need to release the
-- * backup_mutex to avoid deadlock */
-- qemu_mutex_unlock(&backup_state.backup_mutex);
--
-- if (cancel_job) {
-- AioContext *aio_context = cancel_job->job.aio_context;
-- aio_context_acquire(aio_context);
-- job_cancel_sync(&cancel_job->job, true);
-- job_unref(&cancel_job->job);
-- aio_context_release(aio_context);
-- }
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
-
- void qmp_backup_cancel(Error **errp)
- {
-- pvebackup_cancel();
-+ block_on_coroutine_fn(pvebackup_co_cancel, NULL);
- }
-
- // assumes the caller holds backup_mutex
-@@ -415,10 +449,18 @@ static int coroutine_fn pvebackup_co_add_config(
- goto out;
- }
-
--static bool create_backup_jobs(void) {
-+/*
-+ * backup_job_create can *not* be run from a coroutine (and requires an
-+ * acquired AioContext), so this can't either.
-+ * The caller is responsible that backup_mutex is held nonetheless.
-+ */
-+static void create_backup_jobs_bh(void *opaque) {
-
- assert(!qemu_in_coroutine());
-
-+ CoCtxData *data = (CoCtxData*)opaque;
-+ Error **errp = (Error**)data->data;
-+
- Error *local_err = NULL;
-
- /* create job transaction to synchronize bitmap commit and cancel all
-@@ -454,24 +496,19 @@ static bool create_backup_jobs(void) {
-
- aio_context_release(aio_context);
-
-- if (!job || local_err != NULL) {
-- Error *create_job_err = NULL;
-- error_setg(&create_job_err, "backup_job_create failed: %s",
-- local_err ? error_get_pretty(local_err) : "null");
-+ di->job = job;
-
-- pvebackup_propagate_error(create_job_err);
-+ if (!job || local_err) {
-+ error_setg(errp, "backup_job_create failed: %s",
-+ local_err ? error_get_pretty(local_err) : "null");
- break;
- }
-
-- di->job = job;
--
- bdrv_unref(di->target);
- di->target = NULL;
- }
-
-- bool errors = pvebackup_error_or_canceled();
--
-- if (errors) {
-+ if (*errp) {
- l = backup_state.di_list;
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-@@ -483,12 +520,17 @@ static bool create_backup_jobs(void) {
- }
-
- if (di->job) {
-+ AioContext *ctx = di->job->job.aio_context;
-+ aio_context_acquire(ctx);
-+ job_cancel_sync(&di->job->job, true);
- job_unref(&di->job->job);
-+ aio_context_release(ctx);
- }
- }
- }
-
-- return errors;
-+ /* return */
-+ aio_co_enter(data->ctx, data->co);
- }
-
- typedef struct QmpBackupTask {
-@@ -525,11 +567,12 @@ typedef struct QmpBackupTask {
- UuidInfo *result;
- } QmpBackupTask;
-
--// assumes the caller holds backup_mutex
- static void coroutine_fn pvebackup_co_prepare(void *opaque)
- {
- assert(qemu_in_coroutine());
-
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- QmpBackupTask *task = opaque;
-
- task->result = NULL; // just to be sure
-@@ -550,8 +593,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- const char *firewall_name = "qemu-server.fw";
-
- if (backup_state.di_list) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
- "previous backup not finished");
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return;
- }
-
-@@ -618,6 +662,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
- di->size = size;
- total += size;
-+
-+ di->completed_ret = INT_MAX;
- }
-
- uuid_generate(uuid);
-@@ -849,6 +895,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- backup_state.stat.dirty = total - backup_state.stat.reused;
- backup_state.stat.transferred = 0;
- backup_state.stat.zero_bytes = 0;
-+ backup_state.stat.finishing = false;
-+ backup_state.stat.starting = true;
-
- qemu_mutex_unlock(&backup_state.stat.lock);
-
-@@ -863,6 +911,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- uuid_info->UUID = uuid_str;
-
- task->result = uuid_info;
-+
-+ /* Run create_backup_jobs_bh outside of coroutine (in BH) but keep
-+ * backup_mutex locked. This is fine, a CoMutex can be held across yield
-+ * points, and we'll release it as soon as the BH reschedules us.
-+ */
-+ CoCtxData waker = {
-+ .co = qemu_coroutine_self(),
-+ .ctx = qemu_get_current_aio_context(),
-+ .data = &local_err,
-+ };
-+ aio_bh_schedule_oneshot(waker.ctx, create_backup_jobs_bh, &waker);
-+ qemu_coroutine_yield();
-+
-+ if (local_err) {
-+ error_propagate(task->errp, local_err);
-+ goto err;
-+ }
-+
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+
-+ qemu_mutex_lock(&backup_state.stat.lock);
-+ backup_state.stat.starting = false;
-+ qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+ /* start the first job in the transaction */
-+ job_txn_start_seq(backup_state.txn);
-+
- return;
-
- err_mutex:
-@@ -885,6 +960,7 @@ err:
- g_free(di);
- }
- g_list_free(di_list);
-+ backup_state.di_list = NULL;
-
- if (devs) {
- g_strfreev(devs);
-@@ -905,6 +981,8 @@ err:
- }
-
- task->result = NULL;
-+
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return;
- }
-
-@@ -958,24 +1036,8 @@ UuidInfo *qmp_backup(
- .errp = errp,
- };
-
-- qemu_mutex_lock(&backup_state.backup_mutex);
--
- block_on_coroutine_fn(pvebackup_co_prepare, &task);
-
-- if (*errp == NULL) {
-- bool errors = create_backup_jobs();
-- qemu_mutex_unlock(&backup_state.backup_mutex);
--
-- if (!errors) {
-- /* start the first job in the transaction
-- * note: this might directly enter the job, so we need to do this
-- * after unlocking the backup_mutex */
-- job_txn_start_seq(backup_state.txn);
-- }
-- } else {
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-- }
--
- return task.result;
- }
-
-@@ -1027,6 +1089,7 @@ BackupStatus *qmp_query_backup(Error **errp)
- info->transferred = backup_state.stat.transferred;
- info->has_reused = true;
- info->reused = backup_state.stat.reused;
-+ info->finishing = backup_state.stat.finishing;
-
- qemu_mutex_unlock(&backup_state.stat.lock);
-
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 69571d86eb..e6c3687bea 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -774,12 +774,15 @@
- #
- # @uuid: uuid for this backup job
- #
-+# @finishing: if status='active' and finishing=true, then the backup process is
-+# waiting for the target to finish.
-+#
- ##
- { 'struct': 'BackupStatus',
- 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
- '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
- '*start-time': 'int', '*end-time': 'int',
-- '*backup-file': 'str', '*uuid': 'str' } }
-+ '*backup-file': 'str', '*uuid': 'str', 'finishing': 'bool' } }
-
- ##
- # @BackupFormat:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Thu, 20 Aug 2020 14:25:00 +0200
+Subject: [PATCH] PVE-Backup: Use a transaction to synchronize job states
+
+By using a JobTxn, we can sync dirty bitmaps only when *all* jobs were
+successful - meaning we don't need to remove them when the backup fails,
+since QEMU's BITMAP_SYNC_MODE_ON_SUCCESS will now handle that for us.
+
+To keep the rate-limiting and IO impact from before, we use a sequential
+transaction, so drives will still be backed up one after the other.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[add new force parameter to job_cancel_sync calls]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ pve-backup.c | 167 +++++++++++++++------------------------------------
+ 1 file changed, 49 insertions(+), 118 deletions(-)
+
+diff --git a/pve-backup.c b/pve-backup.c
+index f90abaa50a..63c686463f 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -52,6 +52,7 @@ static struct PVEBackupState {
+ VmaWriter *vmaw;
+ ProxmoxBackupHandle *pbs;
+ GList *di_list;
++ JobTxn *txn;
+ QemuMutex backup_mutex;
+ CoMutex dump_callback_mutex;
+ } backup_state;
+@@ -71,32 +72,12 @@ typedef struct PVEBackupDevInfo {
+ size_t size;
+ uint64_t block_size;
+ uint8_t dev_id;
+- bool completed;
+ char targetfile[PATH_MAX];
+ BdrvDirtyBitmap *bitmap;
+ BlockDriverState *target;
++ BlockJob *job;
+ } PVEBackupDevInfo;
+
+-static void pvebackup_run_next_job(void);
+-
+-static BlockJob *
+-lookup_active_block_job(PVEBackupDevInfo *di)
+-{
+- if (!di->completed && di->bs) {
+- for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
+- if (job->job.driver->job_type != JOB_TYPE_BACKUP) {
+- continue;
+- }
+-
+- BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
+- if (bjob && bjob->source_bs == di->bs) {
+- return job;
+- }
+- }
+- }
+- return NULL;
+-}
+-
+ static void pvebackup_propagate_error(Error *err)
+ {
+ qemu_mutex_lock(&backup_state.stat.lock);
+@@ -272,18 +253,6 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
+ if (local_err != NULL) {
+ pvebackup_propagate_error(local_err);
+ }
+- } else {
+- // on error or cancel we cannot ensure synchronization of dirty
+- // bitmaps with backup server, so remove all and do full backup next
+- GList *l = backup_state.di_list;
+- while (l) {
+- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+- l = g_list_next(l);
+-
+- if (di->bitmap) {
+- bdrv_release_dirty_bitmap(di->bitmap);
+- }
+- }
+ }
+
+ proxmox_backup_disconnect(backup_state.pbs);
+@@ -322,8 +291,6 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+
+ qemu_mutex_lock(&backup_state.backup_mutex);
+
+- di->completed = true;
+-
+ if (ret < 0) {
+ Error *local_err = NULL;
+ error_setg(&local_err, "job failed with err %d - %s", ret, strerror(-ret));
+@@ -336,20 +303,17 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+
+ block_on_coroutine_fn(pvebackup_complete_stream, di);
+
+- // remove self from job queue
++ // remove self from job list
+ backup_state.di_list = g_list_remove(backup_state.di_list, di);
+
+- if (di->bitmap && ret < 0) {
+- // on error or cancel we cannot ensure synchronization of dirty
+- // bitmaps with backup server, so remove all and do full backup next
+- bdrv_release_dirty_bitmap(di->bitmap);
+- }
+-
+ g_free(di);
+
+- qemu_mutex_unlock(&backup_state.backup_mutex);
++ /* call cleanup if we're the last job */
++ if (!g_list_first(backup_state.di_list)) {
++ block_on_coroutine_fn(pvebackup_co_cleanup, NULL);
++ }
+
+- pvebackup_run_next_job();
++ qemu_mutex_unlock(&backup_state.backup_mutex);
+ }
+
+ static void pvebackup_cancel(void)
+@@ -371,36 +335,28 @@ static void pvebackup_cancel(void)
+ proxmox_backup_abort(backup_state.pbs, "backup canceled");
+ }
+
+- qemu_mutex_unlock(&backup_state.backup_mutex);
+-
+- for(;;) {
+-
+- BlockJob *next_job = NULL;
+-
+- qemu_mutex_lock(&backup_state.backup_mutex);
+-
+- GList *l = backup_state.di_list;
+- while (l) {
+- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+- l = g_list_next(l);
++ /* it's enough to cancel one job in the transaction, the rest will follow
++ * automatically */
++ GList *bdi = g_list_first(backup_state.di_list);
++ BlockJob *cancel_job = bdi && bdi->data ?
++ ((PVEBackupDevInfo *)bdi->data)->job :
++ NULL;
+
+- BlockJob *job = lookup_active_block_job(di);
+- if (job != NULL) {
+- next_job = job;
+- break;
+- }
+- }
++ /* ref the job before releasing the mutex, just to be safe */
++ if (cancel_job) {
++ job_ref(&cancel_job->job);
++ }
+
+- qemu_mutex_unlock(&backup_state.backup_mutex);
++ /* job_cancel_sync may enter the job, so we need to release the
++ * backup_mutex to avoid deadlock */
++ qemu_mutex_unlock(&backup_state.backup_mutex);
+
+- if (next_job) {
+- AioContext *aio_context = next_job->job.aio_context;
+- aio_context_acquire(aio_context);
+- job_cancel_sync(&next_job->job, true);
+- aio_context_release(aio_context);
+- } else {
+- break;
+- }
++ if (cancel_job) {
++ AioContext *aio_context = cancel_job->job.aio_context;
++ aio_context_acquire(aio_context);
++ job_cancel_sync(&cancel_job->job, true);
++ job_unref(&cancel_job->job);
++ aio_context_release(aio_context);
+ }
+ }
+
+@@ -459,51 +415,19 @@ static int coroutine_fn pvebackup_co_add_config(
+ goto out;
+ }
+
+-bool job_should_pause(Job *job);
+-
+-static void pvebackup_run_next_job(void)
+-{
+- assert(!qemu_in_coroutine());
+-
+- qemu_mutex_lock(&backup_state.backup_mutex);
+-
+- GList *l = backup_state.di_list;
+- while (l) {
+- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+- l = g_list_next(l);
+-
+- BlockJob *job = lookup_active_block_job(di);
+-
+- if (job) {
+- qemu_mutex_unlock(&backup_state.backup_mutex);
+-
+- AioContext *aio_context = job->job.aio_context;
+- aio_context_acquire(aio_context);
+-
+- if (job_should_pause(&job->job)) {
+- bool error_or_canceled = pvebackup_error_or_canceled();
+- if (error_or_canceled) {
+- job_cancel_sync(&job->job, true);
+- } else {
+- job_resume(&job->job);
+- }
+- }
+- aio_context_release(aio_context);
+- return;
+- }
+- }
+-
+- block_on_coroutine_fn(pvebackup_co_cleanup, NULL); // no more jobs, run cleanup
+-
+- qemu_mutex_unlock(&backup_state.backup_mutex);
+-}
+-
+ static bool create_backup_jobs(void) {
+
+ assert(!qemu_in_coroutine());
+
+ Error *local_err = NULL;
+
++ /* create job transaction to synchronize bitmap commit and cancel all
++ * jobs in case one errors */
++ if (backup_state.txn) {
++ job_txn_unref(backup_state.txn);
++ }
++ backup_state.txn = job_txn_new_seq();
++
+ BackupPerf perf = { .max_workers = 16 };
+
+ /* create and start all jobs (paused state) */
+@@ -526,7 +450,7 @@ static bool create_backup_jobs(void) {
+ BlockJob *job = backup_job_create(
+ NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
+ bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
+- JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
++ JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn, &local_err);
+
+ aio_context_release(aio_context);
+
+@@ -538,7 +462,8 @@ static bool create_backup_jobs(void) {
+ pvebackup_propagate_error(create_job_err);
+ break;
+ }
+- job_start(&job->job);
++
++ di->job = job;
+
+ bdrv_unref(di->target);
+ di->target = NULL;
+@@ -556,6 +481,10 @@ static bool create_backup_jobs(void) {
+ bdrv_unref(di->target);
+ di->target = NULL;
+ }
++
++ if (di->job) {
++ job_unref(&di->job->job);
++ }
+ }
+ }
+
+@@ -946,10 +875,6 @@ err:
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+ l = g_list_next(l);
+
+- if (di->bitmap) {
+- bdrv_release_dirty_bitmap(di->bitmap);
+- }
+-
+ if (di->target) {
+ bdrv_unref(di->target);
+ }
+@@ -1038,9 +963,15 @@ UuidInfo *qmp_backup(
+ block_on_coroutine_fn(pvebackup_co_prepare, &task);
+
+ if (*errp == NULL) {
+- create_backup_jobs();
++ bool errors = create_backup_jobs();
+ qemu_mutex_unlock(&backup_state.backup_mutex);
+- pvebackup_run_next_job();
++
++ if (!errors) {
++ /* start the first job in the transaction
++ * note: this might directly enter the job, so we need to do this
++ * after unlocking the backup_mutex */
++ job_txn_start_seq(backup_state.txn);
++ }
+ } else {
+ qemu_mutex_unlock(&backup_state.backup_mutex);
+ }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Mon, 28 Sep 2020 13:40:51 +0200
+Subject: [PATCH] PVE-Backup: Don't block on finishing and cleanup
+ create_backup_jobs
+
+proxmox_backup_co_finish is already async, but previously we would wait
+for the coroutine using block_on_coroutine_fn(). Avoid this by
+scheduling pvebackup_co_complete_stream (and thus pvebackup_co_cleanup)
+as a real coroutine when calling from pvebackup_complete_cb. This is ok,
+since complete_stream uses the backup_mutex internally to synchronize,
+and other streams can happily continue writing in the meantime anyway.
+
+To accomodate, backup_mutex is converted to a CoMutex. This means
+converting every user to a coroutine. This is not just useful here, but
+will come in handy once this series[0] is merged, and QMP calls can be
+yield-able coroutines too. Then we can also finally get rid of
+block_on_coroutine_fn.
+
+Cases of aio_context_acquire/release from within what is now a coroutine
+are changed to aio_co_reschedule_self, which works since a running
+coroutine always holds the aio lock for the context it is running in.
+
+job_cancel_sync is called from a BH since it can't be run from a
+coroutine (uses AIO_WAIT_WHILE internally).
+
+Same thing for create_backup_jobs, which is converted to a BH too.
+
+To communicate the finishing state, a new property is introduced to
+query-backup: 'finishing'. A new state is explicitly not used, since
+that would break compatibility with older qemu-server versions.
+
+Also fix create_backup_jobs:
+
+No more weird bool returns, just the standard "errp" format used
+everywhere else too. With this, if backup_job_create fails, the error
+message is actually returned over QMP and can be shown to the user.
+
+To facilitate correct cleanup on such an error, we call
+create_backup_jobs as a bottom half directly from pvebackup_co_prepare.
+This additionally allows us to actually hold the backup_mutex during
+operation.
+
+Also add a job_cancel_sync before job_unref, since a job must be in
+STATUS_NULL to be deleted by unref, which could trigger an assert
+before.
+
+[0] https://lists.gnu.org/archive/html/qemu-devel/2020-09/msg03515.html
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[add new force parameter to job_cancel_sync calls]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ pve-backup.c | 217 ++++++++++++++++++++++++++++---------------
+ qapi/block-core.json | 5 +-
+ 2 files changed, 144 insertions(+), 78 deletions(-)
+
+diff --git a/pve-backup.c b/pve-backup.c
+index 63c686463f..6f05796fad 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -33,7 +33,9 @@ const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
+
+ static struct PVEBackupState {
+ struct {
+- // Everithing accessed from qmp_backup_query command is protected using lock
++ // Everything accessed from qmp_backup_query command is protected using
++ // this lock. Do NOT hold this lock for long times, as it is sometimes
++ // acquired from coroutines, and thus any wait time may block the guest.
+ QemuMutex lock;
+ Error *error;
+ time_t start_time;
+@@ -47,20 +49,22 @@ static struct PVEBackupState {
+ size_t reused;
+ size_t zero_bytes;
+ GList *bitmap_list;
++ bool finishing;
++ bool starting;
+ } stat;
+ int64_t speed;
+ VmaWriter *vmaw;
+ ProxmoxBackupHandle *pbs;
+ GList *di_list;
+ JobTxn *txn;
+- QemuMutex backup_mutex;
++ CoMutex backup_mutex;
+ CoMutex dump_callback_mutex;
+ } backup_state;
+
+ static void pvebackup_init(void)
+ {
+ qemu_mutex_init(&backup_state.stat.lock);
+- qemu_mutex_init(&backup_state.backup_mutex);
++ qemu_co_mutex_init(&backup_state.backup_mutex);
+ qemu_co_mutex_init(&backup_state.dump_callback_mutex);
+ }
+
+@@ -72,6 +76,7 @@ typedef struct PVEBackupDevInfo {
+ size_t size;
+ uint64_t block_size;
+ uint8_t dev_id;
++ int completed_ret; // INT_MAX if not completed
+ char targetfile[PATH_MAX];
+ BdrvDirtyBitmap *bitmap;
+ BlockDriverState *target;
+@@ -227,12 +232,12 @@ pvebackup_co_dump_vma_cb(
+ }
+
+ // assumes the caller holds backup_mutex
+-static void coroutine_fn pvebackup_co_cleanup(void *unused)
++static void coroutine_fn pvebackup_co_cleanup(void)
+ {
+ assert(qemu_in_coroutine());
+
+ qemu_mutex_lock(&backup_state.stat.lock);
+- backup_state.stat.end_time = time(NULL);
++ backup_state.stat.finishing = true;
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+ if (backup_state.vmaw) {
+@@ -261,35 +266,29 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
+
+ g_list_free(backup_state.di_list);
+ backup_state.di_list = NULL;
++
++ qemu_mutex_lock(&backup_state.stat.lock);
++ backup_state.stat.end_time = time(NULL);
++ backup_state.stat.finishing = false;
++ qemu_mutex_unlock(&backup_state.stat.lock);
+ }
+
+-// assumes the caller holds backup_mutex
+-static void coroutine_fn pvebackup_complete_stream(void *opaque)
++static void coroutine_fn pvebackup_co_complete_stream(void *opaque)
+ {
+ PVEBackupDevInfo *di = opaque;
++ int ret = di->completed_ret;
+
+- bool error_or_canceled = pvebackup_error_or_canceled();
+-
+- if (backup_state.vmaw) {
+- vma_writer_close_stream(backup_state.vmaw, di->dev_id);
++ qemu_mutex_lock(&backup_state.stat.lock);
++ bool starting = backup_state.stat.starting;
++ qemu_mutex_unlock(&backup_state.stat.lock);
++ if (starting) {
++ /* in 'starting' state, no tasks have been run yet, meaning we can (and
++ * must) skip all cleanup, as we don't know what has and hasn't been
++ * initialized yet. */
++ return;
+ }
+
+- if (backup_state.pbs && !error_or_canceled) {
+- Error *local_err = NULL;
+- proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
+- if (local_err != NULL) {
+- pvebackup_propagate_error(local_err);
+- }
+- }
+-}
+-
+-static void pvebackup_complete_cb(void *opaque, int ret)
+-{
+- assert(!qemu_in_coroutine());
+-
+- PVEBackupDevInfo *di = opaque;
+-
+- qemu_mutex_lock(&backup_state.backup_mutex);
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
+
+ if (ret < 0) {
+ Error *local_err = NULL;
+@@ -301,7 +300,19 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+
+ assert(di->target == NULL);
+
+- block_on_coroutine_fn(pvebackup_complete_stream, di);
++ bool error_or_canceled = pvebackup_error_or_canceled();
++
++ if (backup_state.vmaw) {
++ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
++ }
++
++ if (backup_state.pbs && !error_or_canceled) {
++ Error *local_err = NULL;
++ proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
++ if (local_err != NULL) {
++ pvebackup_propagate_error(local_err);
++ }
++ }
+
+ // remove self from job list
+ backup_state.di_list = g_list_remove(backup_state.di_list, di);
+@@ -310,21 +321,49 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+
+ /* call cleanup if we're the last job */
+ if (!g_list_first(backup_state.di_list)) {
+- block_on_coroutine_fn(pvebackup_co_cleanup, NULL);
++ pvebackup_co_cleanup();
+ }
+
+- qemu_mutex_unlock(&backup_state.backup_mutex);
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ }
+
+-static void pvebackup_cancel(void)
++static void pvebackup_complete_cb(void *opaque, int ret)
+ {
+- assert(!qemu_in_coroutine());
++ PVEBackupDevInfo *di = opaque;
++ di->completed_ret = ret;
++
++ /*
++ * Schedule stream cleanup in async coroutine. close_image and finish might
++ * take a while, so we can't block on them here. This way it also doesn't
++ * matter if we're already running in a coroutine or not.
++ * Note: di is a pointer to an entry in the global backup_state struct, so
++ * it stays valid.
++ */
++ Coroutine *co = qemu_coroutine_create(pvebackup_co_complete_stream, di);
++ aio_co_enter(qemu_get_aio_context(), co);
++}
++
++/*
++ * job_cancel(_sync) does not like to be called from coroutines, so defer to
++ * main loop processing via a bottom half.
++ */
++static void job_cancel_bh(void *opaque) {
++ CoCtxData *data = (CoCtxData*)opaque;
++ Job *job = (Job*)data->data;
++ AioContext *job_ctx = job->aio_context;
++ aio_context_acquire(job_ctx);
++ job_cancel_sync(job, true);
++ aio_context_release(job_ctx);
++ aio_co_enter(data->ctx, data->co);
++}
+
++static void coroutine_fn pvebackup_co_cancel(void *opaque)
++{
+ Error *cancel_err = NULL;
+ error_setg(&cancel_err, "backup canceled");
+ pvebackup_propagate_error(cancel_err);
+
+- qemu_mutex_lock(&backup_state.backup_mutex);
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
+
+ if (backup_state.vmaw) {
+ /* make sure vma writer does not block anymore */
+@@ -342,27 +381,22 @@ static void pvebackup_cancel(void)
+ ((PVEBackupDevInfo *)bdi->data)->job :
+ NULL;
+
+- /* ref the job before releasing the mutex, just to be safe */
+ if (cancel_job) {
+- job_ref(&cancel_job->job);
++ CoCtxData data = {
++ .ctx = qemu_get_current_aio_context(),
++ .co = qemu_coroutine_self(),
++ .data = &cancel_job->job,
++ };
++ aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
++ qemu_coroutine_yield();
+ }
+
+- /* job_cancel_sync may enter the job, so we need to release the
+- * backup_mutex to avoid deadlock */
+- qemu_mutex_unlock(&backup_state.backup_mutex);
+-
+- if (cancel_job) {
+- AioContext *aio_context = cancel_job->job.aio_context;
+- aio_context_acquire(aio_context);
+- job_cancel_sync(&cancel_job->job, true);
+- job_unref(&cancel_job->job);
+- aio_context_release(aio_context);
+- }
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ }
+
+ void qmp_backup_cancel(Error **errp)
+ {
+- pvebackup_cancel();
++ block_on_coroutine_fn(pvebackup_co_cancel, NULL);
+ }
+
+ // assumes the caller holds backup_mutex
+@@ -415,10 +449,18 @@ static int coroutine_fn pvebackup_co_add_config(
+ goto out;
+ }
+
+-static bool create_backup_jobs(void) {
++/*
++ * backup_job_create can *not* be run from a coroutine (and requires an
++ * acquired AioContext), so this can't either.
++ * The caller is responsible that backup_mutex is held nonetheless.
++ */
++static void create_backup_jobs_bh(void *opaque) {
+
+ assert(!qemu_in_coroutine());
+
++ CoCtxData *data = (CoCtxData*)opaque;
++ Error **errp = (Error**)data->data;
++
+ Error *local_err = NULL;
+
+ /* create job transaction to synchronize bitmap commit and cancel all
+@@ -454,24 +496,19 @@ static bool create_backup_jobs(void) {
+
+ aio_context_release(aio_context);
+
+- if (!job || local_err != NULL) {
+- Error *create_job_err = NULL;
+- error_setg(&create_job_err, "backup_job_create failed: %s",
+- local_err ? error_get_pretty(local_err) : "null");
++ di->job = job;
+
+- pvebackup_propagate_error(create_job_err);
++ if (!job || local_err) {
++ error_setg(errp, "backup_job_create failed: %s",
++ local_err ? error_get_pretty(local_err) : "null");
+ break;
+ }
+
+- di->job = job;
+-
+ bdrv_unref(di->target);
+ di->target = NULL;
+ }
+
+- bool errors = pvebackup_error_or_canceled();
+-
+- if (errors) {
++ if (*errp) {
+ l = backup_state.di_list;
+ while (l) {
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+@@ -483,12 +520,17 @@ static bool create_backup_jobs(void) {
+ }
+
+ if (di->job) {
++ AioContext *ctx = di->job->job.aio_context;
++ aio_context_acquire(ctx);
++ job_cancel_sync(&di->job->job, true);
+ job_unref(&di->job->job);
++ aio_context_release(ctx);
+ }
+ }
+ }
+
+- return errors;
++ /* return */
++ aio_co_enter(data->ctx, data->co);
+ }
+
+ typedef struct QmpBackupTask {
+@@ -525,11 +567,12 @@ typedef struct QmpBackupTask {
+ UuidInfo *result;
+ } QmpBackupTask;
+
+-// assumes the caller holds backup_mutex
+ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ {
+ assert(qemu_in_coroutine());
+
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
+ QmpBackupTask *task = opaque;
+
+ task->result = NULL; // just to be sure
+@@ -550,8 +593,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ const char *firewall_name = "qemu-server.fw";
+
+ if (backup_state.di_list) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
+ "previous backup not finished");
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ return;
+ }
+
+@@ -618,6 +662,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+ di->size = size;
+ total += size;
++
++ di->completed_ret = INT_MAX;
+ }
+
+ uuid_generate(uuid);
+@@ -849,6 +895,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ backup_state.stat.dirty = total - backup_state.stat.reused;
+ backup_state.stat.transferred = 0;
+ backup_state.stat.zero_bytes = 0;
++ backup_state.stat.finishing = false;
++ backup_state.stat.starting = true;
+
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+@@ -863,6 +911,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ uuid_info->UUID = uuid_str;
+
+ task->result = uuid_info;
++
++ /* Run create_backup_jobs_bh outside of coroutine (in BH) but keep
++ * backup_mutex locked. This is fine, a CoMutex can be held across yield
++ * points, and we'll release it as soon as the BH reschedules us.
++ */
++ CoCtxData waker = {
++ .co = qemu_coroutine_self(),
++ .ctx = qemu_get_current_aio_context(),
++ .data = &local_err,
++ };
++ aio_bh_schedule_oneshot(waker.ctx, create_backup_jobs_bh, &waker);
++ qemu_coroutine_yield();
++
++ if (local_err) {
++ error_propagate(task->errp, local_err);
++ goto err;
++ }
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++
++ qemu_mutex_lock(&backup_state.stat.lock);
++ backup_state.stat.starting = false;
++ qemu_mutex_unlock(&backup_state.stat.lock);
++
++ /* start the first job in the transaction */
++ job_txn_start_seq(backup_state.txn);
++
+ return;
+
+ err_mutex:
+@@ -885,6 +960,7 @@ err:
+ g_free(di);
+ }
+ g_list_free(di_list);
++ backup_state.di_list = NULL;
+
+ if (devs) {
+ g_strfreev(devs);
+@@ -905,6 +981,8 @@ err:
+ }
+
+ task->result = NULL;
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ return;
+ }
+
+@@ -958,24 +1036,8 @@ UuidInfo *qmp_backup(
+ .errp = errp,
+ };
+
+- qemu_mutex_lock(&backup_state.backup_mutex);
+-
+ block_on_coroutine_fn(pvebackup_co_prepare, &task);
+
+- if (*errp == NULL) {
+- bool errors = create_backup_jobs();
+- qemu_mutex_unlock(&backup_state.backup_mutex);
+-
+- if (!errors) {
+- /* start the first job in the transaction
+- * note: this might directly enter the job, so we need to do this
+- * after unlocking the backup_mutex */
+- job_txn_start_seq(backup_state.txn);
+- }
+- } else {
+- qemu_mutex_unlock(&backup_state.backup_mutex);
+- }
+-
+ return task.result;
+ }
+
+@@ -1027,6 +1089,7 @@ BackupStatus *qmp_query_backup(Error **errp)
+ info->transferred = backup_state.stat.transferred;
+ info->has_reused = true;
+ info->reused = backup_state.stat.reused;
++ info->finishing = backup_state.stat.finishing;
+
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 16e184dd28..cb17d00fe0 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -770,12 +770,15 @@
+ #
+ # @uuid: uuid for this backup job
+ #
++# @finishing: if status='active' and finishing=true, then the backup process is
++# waiting for the target to finish.
++#
+ ##
+ { 'struct': 'BackupStatus',
+ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
+ '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
+ '*start-time': 'int', '*end-time': 'int',
+- '*backup-file': 'str', '*uuid': 'str' } }
++ '*backup-file': 'str', '*uuid': 'str', 'finishing': 'bool' } }
+
+ ##
+ # @BackupFormat:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 22 Oct 2020 17:34:18 +0200
-Subject: [PATCH] PVE: Migrate dirty bitmap state via savevm
-
-QEMU provides 'savevm' registrations as a mechanism for arbitrary state
-to be migrated along with a VM. Use this to send a serialized version of
-dirty bitmap state data from proxmox-backup-qemu, and restore it on the
-target node.
-
-Also add a flag to query-proxmox-support so qemu-server can determine if
-safe migration is possible and makes sense.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- include/migration/misc.h | 3 ++
- migration/meson.build | 2 +
- migration/migration.c | 1 +
- migration/pbs-state.c | 106 +++++++++++++++++++++++++++++++++++++++
- pve-backup.c | 1 +
- qapi/block-core.json | 6 +++
- 6 files changed, 119 insertions(+)
- create mode 100644 migration/pbs-state.c
-
-diff --git a/include/migration/misc.h b/include/migration/misc.h
-index 465906710d..4f0aeceb6f 100644
---- a/include/migration/misc.h
-+++ b/include/migration/misc.h
-@@ -75,4 +75,7 @@ bool migration_in_bg_snapshot(void);
- /* migration/block-dirty-bitmap.c */
- void dirty_bitmap_mig_init(void);
-
-+/* migration/pbs-state.c */
-+void pbs_state_mig_init(void);
-+
- #endif
-diff --git a/migration/meson.build b/migration/meson.build
-index 1e2aec8486..259e863a43 100644
---- a/migration/meson.build
-+++ b/migration/meson.build
-@@ -7,8 +7,10 @@ migration_files = files(
- 'qemu-file-channel.c',
- 'qemu-file.c',
- 'yank_functions.c',
-+ 'pbs-state.c',
- )
- softmmu_ss.add(migration_files)
-+softmmu_ss.add(libproxmox_backup_qemu)
-
- softmmu_ss.add(files(
- 'block-dirty-bitmap.c',
-diff --git a/migration/migration.c b/migration/migration.c
-index 695f0f2900..6e17f8e91c 100644
---- a/migration/migration.c
-+++ b/migration/migration.c
-@@ -214,6 +214,7 @@ void migration_object_init(void)
- blk_mig_init();
- ram_mig_init();
- dirty_bitmap_mig_init();
-+ pbs_state_mig_init();
- }
-
- void migration_cancel(const Error *error)
-diff --git a/migration/pbs-state.c b/migration/pbs-state.c
-new file mode 100644
-index 0000000000..29f2b3860d
---- /dev/null
-+++ b/migration/pbs-state.c
-@@ -0,0 +1,106 @@
-+/*
-+ * PBS (dirty-bitmap) state migration
-+ */
-+
-+#include "qemu/osdep.h"
-+#include "migration/misc.h"
-+#include "qemu-file.h"
-+#include "migration/vmstate.h"
-+#include "migration/register.h"
-+#include "proxmox-backup-qemu.h"
-+
-+typedef struct PBSState {
-+ bool active;
-+} PBSState;
-+
-+/* state is accessed via this static variable directly, 'opaque' is NULL */
-+static PBSState pbs_state;
-+
-+static void pbs_state_save_pending(QEMUFile *f, void *opaque,
-+ uint64_t max_size,
-+ uint64_t *res_precopy_only,
-+ uint64_t *res_compatible,
-+ uint64_t *res_postcopy_only)
-+{
-+ /* we send everything in save_setup, so nothing is ever pending */
-+}
-+
-+/* receive PBS state via f and deserialize, called on target */
-+static int pbs_state_load(QEMUFile *f, void *opaque, int version_id)
-+{
-+ /* safe cast, we cannot migrate to target with less bits than source */
-+ size_t buf_size = (size_t)qemu_get_be64(f);
-+
-+ uint8_t *buf = (uint8_t *)malloc(buf_size);
-+ size_t read = qemu_get_buffer(f, buf, buf_size);
-+
-+ if (read < buf_size) {
-+ fprintf(stderr, "error receiving PBS state: not enough data\n");
-+ return -EIO;
-+ }
-+
-+ proxmox_import_state(buf, buf_size);
-+
-+ free(buf);
-+ return 0;
-+}
-+
-+/* serialize PBS state and send to target via f, called on source */
-+static int pbs_state_save_setup(QEMUFile *f, void *opaque)
-+{
-+ size_t buf_size;
-+ uint8_t *buf = proxmox_export_state(&buf_size);
-+
-+ /* LV encoding */
-+ qemu_put_be64(f, buf_size);
-+ qemu_put_buffer(f, buf, buf_size);
-+
-+ proxmox_free_state_buf(buf);
-+ pbs_state.active = false;
-+ return 0;
-+}
-+
-+static bool pbs_state_is_active(void *opaque)
-+{
-+ /* we need to return active exactly once, else .save_setup is never called,
-+ * but if we'd just return true the migration doesn't make progress since
-+ * it'd be waiting for us */
-+ return pbs_state.active;
-+}
-+
-+static bool pbs_state_is_active_iterate(void *opaque)
-+{
-+ /* we don't iterate, everything is sent in save_setup */
-+ return pbs_state_is_active(opaque);
-+}
-+
-+static bool pbs_state_has_postcopy(void *opaque)
-+{
-+ /* PBS state can't change during a migration (since that's blocking any
-+ * potential backups), so we can copy everything before the VM is stopped */
-+ return false;
-+}
-+
-+static void pbs_state_save_cleanup(void *opaque)
-+{
-+ /* reset active after migration succeeds or fails */
-+ pbs_state.active = false;
-+}
-+
-+static SaveVMHandlers savevm_pbs_state_handlers = {
-+ .save_setup = pbs_state_save_setup,
-+ .has_postcopy = pbs_state_has_postcopy,
-+ .save_live_pending = pbs_state_save_pending,
-+ .is_active_iterate = pbs_state_is_active_iterate,
-+ .load_state = pbs_state_load,
-+ .is_active = pbs_state_is_active,
-+ .save_cleanup = pbs_state_save_cleanup,
-+};
-+
-+void pbs_state_mig_init(void)
-+{
-+ pbs_state.active = true;
-+ register_savevm_live("pbs-state", 0, 1,
-+ &savevm_pbs_state_handlers,
-+ NULL);
-+}
-diff --git a/pve-backup.c b/pve-backup.c
-index 6f05796fad..5fa3cc1352 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -1132,6 +1132,7 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
- ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
- ret->pbs_dirty_bitmap = true;
- ret->pbs_dirty_bitmap_savevm = true;
-+ ret->pbs_dirty_bitmap_migration = true;
- ret->query_bitmap_info = true;
- return ret;
- }
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index e6c3687bea..05382a1069 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -883,6 +883,11 @@
- # @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
- # safely be set for savevm-async.
- #
-+# @pbs-dirty-bitmap-migration: True if safe migration of dirty-bitmaps including
-+# PBS state is supported. Enabling 'dirty-bitmaps'
-+# migration cap if this is false/unset may lead
-+# to crashes on migration!
-+#
- # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
- #
- ##
-@@ -890,6 +895,7 @@
- 'data': { 'pbs-dirty-bitmap': 'bool',
- 'query-bitmap-info': 'bool',
- 'pbs-dirty-bitmap-savevm': 'bool',
-+ 'pbs-dirty-bitmap-migration': 'bool',
- 'pbs-library-version': 'str' } }
-
- ##
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Thu, 22 Oct 2020 17:34:18 +0200
+Subject: [PATCH] PVE: Migrate dirty bitmap state via savevm
+
+QEMU provides 'savevm' registrations as a mechanism for arbitrary state
+to be migrated along with a VM. Use this to send a serialized version of
+dirty bitmap state data from proxmox-backup-qemu, and restore it on the
+target node.
+
+Also add a flag to query-proxmox-support so qemu-server can determine if
+safe migration is possible and makes sense.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ include/migration/misc.h | 3 ++
+ migration/meson.build | 2 +
+ migration/migration.c | 1 +
+ migration/pbs-state.c | 106 +++++++++++++++++++++++++++++++++++++++
+ pve-backup.c | 1 +
+ qapi/block-core.json | 6 +++
+ 6 files changed, 119 insertions(+)
+ create mode 100644 migration/pbs-state.c
+
+diff --git a/include/migration/misc.h b/include/migration/misc.h
+index 465906710d..4f0aeceb6f 100644
+--- a/include/migration/misc.h
++++ b/include/migration/misc.h
+@@ -75,4 +75,7 @@ bool migration_in_bg_snapshot(void);
+ /* migration/block-dirty-bitmap.c */
+ void dirty_bitmap_mig_init(void);
+
++/* migration/pbs-state.c */
++void pbs_state_mig_init(void);
++
+ #endif
+diff --git a/migration/meson.build b/migration/meson.build
+index 0842d00cd2..d012f4d8d3 100644
+--- a/migration/meson.build
++++ b/migration/meson.build
+@@ -6,8 +6,10 @@ migration_files = files(
+ 'vmstate.c',
+ 'qemu-file.c',
+ 'yank_functions.c',
++ 'pbs-state.c',
+ )
+ softmmu_ss.add(migration_files)
++softmmu_ss.add(libproxmox_backup_qemu)
+
+ softmmu_ss.add(files(
+ 'block-dirty-bitmap.c',
+diff --git a/migration/migration.c b/migration/migration.c
+index bb8bbddfe4..8109e468eb 100644
+--- a/migration/migration.c
++++ b/migration/migration.c
+@@ -229,6 +229,7 @@ void migration_object_init(void)
+ blk_mig_init();
+ ram_mig_init();
+ dirty_bitmap_mig_init();
++ pbs_state_mig_init();
+ }
+
+ void migration_cancel(const Error *error)
+diff --git a/migration/pbs-state.c b/migration/pbs-state.c
+new file mode 100644
+index 0000000000..29f2b3860d
+--- /dev/null
++++ b/migration/pbs-state.c
+@@ -0,0 +1,106 @@
++/*
++ * PBS (dirty-bitmap) state migration
++ */
++
++#include "qemu/osdep.h"
++#include "migration/misc.h"
++#include "qemu-file.h"
++#include "migration/vmstate.h"
++#include "migration/register.h"
++#include "proxmox-backup-qemu.h"
++
++typedef struct PBSState {
++ bool active;
++} PBSState;
++
++/* state is accessed via this static variable directly, 'opaque' is NULL */
++static PBSState pbs_state;
++
++static void pbs_state_save_pending(QEMUFile *f, void *opaque,
++ uint64_t max_size,
++ uint64_t *res_precopy_only,
++ uint64_t *res_compatible,
++ uint64_t *res_postcopy_only)
++{
++ /* we send everything in save_setup, so nothing is ever pending */
++}
++
++/* receive PBS state via f and deserialize, called on target */
++static int pbs_state_load(QEMUFile *f, void *opaque, int version_id)
++{
++ /* safe cast, we cannot migrate to target with less bits than source */
++ size_t buf_size = (size_t)qemu_get_be64(f);
++
++ uint8_t *buf = (uint8_t *)malloc(buf_size);
++ size_t read = qemu_get_buffer(f, buf, buf_size);
++
++ if (read < buf_size) {
++ fprintf(stderr, "error receiving PBS state: not enough data\n");
++ return -EIO;
++ }
++
++ proxmox_import_state(buf, buf_size);
++
++ free(buf);
++ return 0;
++}
++
++/* serialize PBS state and send to target via f, called on source */
++static int pbs_state_save_setup(QEMUFile *f, void *opaque)
++{
++ size_t buf_size;
++ uint8_t *buf = proxmox_export_state(&buf_size);
++
++ /* LV encoding */
++ qemu_put_be64(f, buf_size);
++ qemu_put_buffer(f, buf, buf_size);
++
++ proxmox_free_state_buf(buf);
++ pbs_state.active = false;
++ return 0;
++}
++
++static bool pbs_state_is_active(void *opaque)
++{
++ /* we need to return active exactly once, else .save_setup is never called,
++ * but if we'd just return true the migration doesn't make progress since
++ * it'd be waiting for us */
++ return pbs_state.active;
++}
++
++static bool pbs_state_is_active_iterate(void *opaque)
++{
++ /* we don't iterate, everything is sent in save_setup */
++ return pbs_state_is_active(opaque);
++}
++
++static bool pbs_state_has_postcopy(void *opaque)
++{
++ /* PBS state can't change during a migration (since that's blocking any
++ * potential backups), so we can copy everything before the VM is stopped */
++ return false;
++}
++
++static void pbs_state_save_cleanup(void *opaque)
++{
++ /* reset active after migration succeeds or fails */
++ pbs_state.active = false;
++}
++
++static SaveVMHandlers savevm_pbs_state_handlers = {
++ .save_setup = pbs_state_save_setup,
++ .has_postcopy = pbs_state_has_postcopy,
++ .save_live_pending = pbs_state_save_pending,
++ .is_active_iterate = pbs_state_is_active_iterate,
++ .load_state = pbs_state_load,
++ .is_active = pbs_state_is_active,
++ .save_cleanup = pbs_state_save_cleanup,
++};
++
++void pbs_state_mig_init(void)
++{
++ pbs_state.active = true;
++ register_savevm_live("pbs-state", 0, 1,
++ &savevm_pbs_state_handlers,
++ NULL);
++}
+diff --git a/pve-backup.c b/pve-backup.c
+index 6f05796fad..5fa3cc1352 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -1132,6 +1132,7 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+ ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
+ ret->pbs_dirty_bitmap = true;
+ ret->pbs_dirty_bitmap_savevm = true;
++ ret->pbs_dirty_bitmap_migration = true;
+ ret->query_bitmap_info = true;
+ return ret;
+ }
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index cb17d00fe0..bd978ea562 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -879,6 +879,11 @@
+ # @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
+ # safely be set for savevm-async.
+ #
++# @pbs-dirty-bitmap-migration: True if safe migration of dirty-bitmaps including
++# PBS state is supported. Enabling 'dirty-bitmaps'
++# migration cap if this is false/unset may lead
++# to crashes on migration!
++#
+ # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
+ #
+ ##
+@@ -886,6 +891,7 @@
+ 'data': { 'pbs-dirty-bitmap': 'bool',
+ 'query-bitmap-info': 'bool',
+ 'pbs-dirty-bitmap-savevm': 'bool',
++ 'pbs-dirty-bitmap-migration': 'bool',
+ 'pbs-library-version': 'str' } }
+
+ ##
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Tue, 3 Nov 2020 14:57:32 +0100
-Subject: [PATCH] migration/block-dirty-bitmap: migrate other bitmaps even if
- one fails
-
-If the checks in bdrv_dirty_bitmap_check fail, that only means that this
-one specific bitmap cannot be migrated. That is not an error condition
-for any other bitmaps on the same block device.
-
-Fixes dirty-bitmap migration with sync=bitmap, as the bitmaps used for
-that are obviously marked as "busy", which would cause none at all to be
-transferred.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- migration/block-dirty-bitmap.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/migration/block-dirty-bitmap.c b/migration/block-dirty-bitmap.c
-index 9aba7d9c22..f4ecf9c9f9 100644
---- a/migration/block-dirty-bitmap.c
-+++ b/migration/block-dirty-bitmap.c
-@@ -538,7 +538,7 @@ static int add_bitmaps_to_list(DBMSaveState *s, BlockDriverState *bs,
-
- if (bdrv_dirty_bitmap_check(bitmap, BDRV_BITMAP_DEFAULT, &local_err)) {
- error_report_err(local_err);
-- return -1;
-+ continue;
- }
-
- if (bitmap_aliases) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Tue, 17 Nov 2020 10:51:05 +0100
-Subject: [PATCH] PVE: fall back to open-iscsi initiatorname
-
-When no explicit option is given, try reading the initiator name from
-/etc/iscsi/initiatorname.iscsi and only use the generic fallback, i.e.
-iqn.2008-11.org.linux-kvmXXX, as a third alternative.
-
-This avoids the need to add an explicit option for vma and to explicitly set it
-for each call to qemu that deals with iSCSI disks, while still allowing to set
-the option if a different name is needed.
-
-According to RFC 3720, an initiator name is at most 223 bytes long, so the
-4 KiB buffer is big enough, even if many whitespaces are used.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/iscsi.c | 30 ++++++++++++++++++++++++++++++
- 1 file changed, 30 insertions(+)
-
-diff --git a/block/iscsi.c b/block/iscsi.c
-index 51f2a5eeaa..f11591ac94 100644
---- a/block/iscsi.c
-+++ b/block/iscsi.c
-@@ -1386,12 +1386,42 @@ static char *get_initiator_name(QemuOpts *opts)
- const char *name;
- char *iscsi_name;
- UuidInfo *uuid_info;
-+ FILE *name_fh;
-
- name = qemu_opt_get(opts, "initiator-name");
- if (name) {
- return g_strdup(name);
- }
-
-+ name_fh = fopen("/etc/iscsi/initiatorname.iscsi", "r");
-+ if (name_fh) {
-+ const char *key = "InitiatorName";
-+ char buffer[4096];
-+ char *line;
-+
-+ while ((line = fgets(buffer, sizeof(buffer), name_fh))) {
-+ line = g_strstrip(line);
-+ if (!strncmp(line, key, strlen(key))) {
-+ line = strchr(line, '=');
-+ if (!line || strlen(line) == 1) {
-+ continue;
-+ }
-+ line++;
-+ g_strstrip(line);
-+ if (!strlen(line)) {
-+ continue;
-+ }
-+ name = line;
-+ break;
-+ }
-+ }
-+ fclose(name_fh);
-+
-+ if (name) {
-+ return g_strdup(name);
-+ }
-+ }
-+
- uuid_info = qmp_query_uuid(NULL);
- if (strcmp(uuid_info->UUID, UUID_NONE) == 0) {
- name = qemu_get_vm_name();
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Tue, 3 Nov 2020 14:57:32 +0100
+Subject: [PATCH] migration/block-dirty-bitmap: migrate other bitmaps even if
+ one fails
+
+If the checks in bdrv_dirty_bitmap_check fail, that only means that this
+one specific bitmap cannot be migrated. That is not an error condition
+for any other bitmaps on the same block device.
+
+Fixes dirty-bitmap migration with sync=bitmap, as the bitmaps used for
+that are obviously marked as "busy", which would cause none at all to be
+transferred.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ migration/block-dirty-bitmap.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/migration/block-dirty-bitmap.c b/migration/block-dirty-bitmap.c
+index 9aba7d9c22..f4ecf9c9f9 100644
+--- a/migration/block-dirty-bitmap.c
++++ b/migration/block-dirty-bitmap.c
+@@ -538,7 +538,7 @@ static int add_bitmaps_to_list(DBMSaveState *s, BlockDriverState *bs,
+
+ if (bdrv_dirty_bitmap_check(bitmap, BDRV_BITMAP_DEFAULT, &local_err)) {
+ error_report_err(local_err);
+- return -1;
++ continue;
+ }
+
+ if (bitmap_aliases) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Tue, 26 Jan 2021 15:45:30 +0100
-Subject: [PATCH] PVE: Use coroutine QMP for backup/cancel_backup
-
-Finally turn backup QMP calls into coroutines, now that it's possible.
-This has the benefit that calls are asynchronous to the main loop, i.e.
-long running operations like connecting to a PBS server will no longer
-hang the VM.
-
-Additionally, it allows us to get rid of block_on_coroutine_fn, which
-was always a hacky workaround.
-
-While we're already spring cleaning, also remove the QmpBackupTask
-struct, since we can now put the 'prepare' function directly into
-qmp_backup and thus no longer need those giant walls of text.
-
-(Note that for our patches to work with 5.2.0 this change is actually
-required, otherwise monitor_get_fd() fails as we're not in a QMP
-coroutine, but one we start ourselves - we could of course set the
-monitor for that coroutine ourselves, but let's just fix it the right
-way instead)
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/monitor/block-hmp-cmds.c | 4 +-
- hmp-commands.hx | 2 +
- proxmox-backup-client.c | 31 -----
- pve-backup.c | 232 ++++++++++-----------------------
- qapi/block-core.json | 4 +-
- 5 files changed, 77 insertions(+), 196 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index ea7b665aa2..ef45552e3b 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1016,7 +1016,7 @@ void hmp_info_snapshots(Monitor *mon, const QDict *qdict)
- g_free(global_snapshots);
- }
-
--void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
-+void coroutine_fn hmp_backup_cancel(Monitor *mon, const QDict *qdict)
- {
- Error *error = NULL;
-
-@@ -1025,7 +1025,7 @@ void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
- hmp_handle_error(mon, error);
- }
-
--void hmp_backup(Monitor *mon, const QDict *qdict)
-+void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
- {
- Error *error = NULL;
-
-diff --git a/hmp-commands.hx b/hmp-commands.hx
-index 994445f301..aaea875970 100644
---- a/hmp-commands.hx
-+++ b/hmp-commands.hx
-@@ -109,6 +109,7 @@ ERST
- "\n\t\t\t Use -d to dump data into a directory instead"
- "\n\t\t\t of using VMA format.",
- .cmd = hmp_backup,
-+ .coroutine = true,
- },
-
- SRST
-@@ -122,6 +123,7 @@ ERST
- .params = "",
- .help = "cancel the current VM backup",
- .cmd = hmp_backup_cancel,
-+ .coroutine = true,
- },
-
- SRST
-diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
-index 4ce7bc0b5e..0923037dec 100644
---- a/proxmox-backup-client.c
-+++ b/proxmox-backup-client.c
-@@ -5,37 +5,6 @@
-
- /* Proxmox Backup Server client bindings using coroutines */
-
--typedef struct BlockOnCoroutineWrapper {
-- AioContext *ctx;
-- CoroutineEntry *entry;
-- void *entry_arg;
-- bool finished;
--} BlockOnCoroutineWrapper;
--
--static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
--{
-- BlockOnCoroutineWrapper *wrapper = opaque;
-- wrapper->entry(wrapper->entry_arg);
-- wrapper->finished = true;
-- aio_wait_kick();
--}
--
--void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
--{
-- assert(!qemu_in_coroutine());
--
-- AioContext *ctx = qemu_get_current_aio_context();
-- BlockOnCoroutineWrapper wrapper = {
-- .finished = false,
-- .entry = entry,
-- .entry_arg = entry_arg,
-- .ctx = ctx,
-- };
-- Coroutine *wrapper_co = qemu_coroutine_create(block_on_coroutine_wrapper, &wrapper);
-- aio_co_enter(ctx, wrapper_co);
-- AIO_WAIT_WHILE(ctx, !wrapper.finished);
--}
--
- // This is called from another thread, so we use aio_co_schedule()
- static void proxmox_backup_schedule_wake(void *data) {
- CoCtxData *waker = (CoCtxData *)data;
-diff --git a/pve-backup.c b/pve-backup.c
-index 5fa3cc1352..323014744c 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -357,7 +357,7 @@ static void job_cancel_bh(void *opaque) {
- aio_co_enter(data->ctx, data->co);
- }
-
--static void coroutine_fn pvebackup_co_cancel(void *opaque)
-+void coroutine_fn qmp_backup_cancel(Error **errp)
- {
- Error *cancel_err = NULL;
- error_setg(&cancel_err, "backup canceled");
-@@ -394,11 +394,6 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
-
--void qmp_backup_cancel(Error **errp)
--{
-- block_on_coroutine_fn(pvebackup_co_cancel, NULL);
--}
--
- // assumes the caller holds backup_mutex
- static int coroutine_fn pvebackup_co_add_config(
- const char *file,
-@@ -533,50 +528,27 @@ static void create_backup_jobs_bh(void *opaque) {
- aio_co_enter(data->ctx, data->co);
- }
-
--typedef struct QmpBackupTask {
-- const char *backup_file;
-- bool has_password;
-- const char *password;
-- bool has_keyfile;
-- const char *keyfile;
-- bool has_key_password;
-- const char *key_password;
-- bool has_backup_id;
-- const char *backup_id;
-- bool has_backup_time;
-- const char *fingerprint;
-- bool has_fingerprint;
-- int64_t backup_time;
-- bool has_use_dirty_bitmap;
-- bool use_dirty_bitmap;
-- bool has_format;
-- BackupFormat format;
-- bool has_config_file;
-- const char *config_file;
-- bool has_firewall_file;
-- const char *firewall_file;
-- bool has_devlist;
-- const char *devlist;
-- bool has_compress;
-- bool compress;
-- bool has_encrypt;
-- bool encrypt;
-- bool has_speed;
-- int64_t speed;
-- Error **errp;
-- UuidInfo *result;
--} QmpBackupTask;
--
--static void coroutine_fn pvebackup_co_prepare(void *opaque)
-+UuidInfo coroutine_fn *qmp_backup(
-+ const char *backup_file,
-+ bool has_password, const char *password,
-+ bool has_keyfile, const char *keyfile,
-+ bool has_key_password, const char *key_password,
-+ bool has_fingerprint, const char *fingerprint,
-+ bool has_backup_id, const char *backup_id,
-+ bool has_backup_time, int64_t backup_time,
-+ bool has_use_dirty_bitmap, bool use_dirty_bitmap,
-+ bool has_compress, bool compress,
-+ bool has_encrypt, bool encrypt,
-+ bool has_format, BackupFormat format,
-+ bool has_config_file, const char *config_file,
-+ bool has_firewall_file, const char *firewall_file,
-+ bool has_devlist, const char *devlist,
-+ bool has_speed, int64_t speed, Error **errp)
- {
- assert(qemu_in_coroutine());
-
- qemu_co_mutex_lock(&backup_state.backup_mutex);
-
-- QmpBackupTask *task = opaque;
--
-- task->result = NULL; // just to be sure
--
- BlockBackend *blk;
- BlockDriverState *bs = NULL;
- const char *backup_dir = NULL;
-@@ -593,17 +565,17 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- const char *firewall_name = "qemu-server.fw";
-
- if (backup_state.di_list) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
- "previous backup not finished");
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-- return;
-+ return NULL;
- }
-
- /* Todo: try to auto-detect format based on file name */
-- BackupFormat format = task->has_format ? task->format : BACKUP_FORMAT_VMA;
-+ format = has_format ? format : BACKUP_FORMAT_VMA;
-
-- if (task->has_devlist) {
-- devs = g_strsplit_set(task->devlist, ",;:", -1);
-+ if (has_devlist) {
-+ devs = g_strsplit_set(devlist, ",;:", -1);
-
- gchar **d = devs;
- while (d && *d) {
-@@ -611,14 +583,14 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- if (blk) {
- bs = blk_bs(blk);
- if (!bdrv_is_inserted(bs)) {
-- error_setg(task->errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
-+ error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
- goto err;
- }
- PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
- di->bs = bs;
- di_list = g_list_append(di_list, di);
- } else {
-- error_set(task->errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-+ error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
- "Device '%s' not found", *d);
- goto err;
- }
-@@ -641,7 +613,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
-
- if (!di_list) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
- goto err;
- }
-
-@@ -651,13 +623,13 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-- if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, task->errp)) {
-+ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, errp)) {
- goto err;
- }
-
- ssize_t size = bdrv_getlength(di->bs);
- if (size < 0) {
-- error_setg_errno(task->errp, -di->size, "bdrv_getlength failed");
-+ error_setg_errno(errp, -di->size, "bdrv_getlength failed");
- goto err;
- }
- di->size = size;
-@@ -684,47 +656,44 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
-
- if (format == BACKUP_FORMAT_PBS) {
-- if (!task->has_password) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
-+ if (!has_password) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
- goto err_mutex;
- }
-- if (!task->has_backup_id) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
-+ if (!has_backup_id) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
- goto err_mutex;
- }
-- if (!task->has_backup_time) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
-+ if (!has_backup_time) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
- goto err_mutex;
- }
-
- int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
- firewall_name = "fw.conf";
-
-- bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
--
--
- char *pbs_err = NULL;
- pbs = proxmox_backup_new(
-- task->backup_file,
-- task->backup_id,
-- task->backup_time,
-+ backup_file,
-+ backup_id,
-+ backup_time,
- dump_cb_block_size,
-- task->has_password ? task->password : NULL,
-- task->has_keyfile ? task->keyfile : NULL,
-- task->has_key_password ? task->key_password : NULL,
-- task->has_compress ? task->compress : true,
-- task->has_encrypt ? task->encrypt : task->has_keyfile,
-- task->has_fingerprint ? task->fingerprint : NULL,
-+ has_password ? password : NULL,
-+ has_keyfile ? keyfile : NULL,
-+ has_key_password ? key_password : NULL,
-+ has_compress ? compress : true,
-+ has_encrypt ? encrypt : has_keyfile,
-+ has_fingerprint ? fingerprint : NULL,
- &pbs_err);
-
- if (!pbs) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
- "proxmox_backup_new failed: %s", pbs_err);
- proxmox_backup_free_error(pbs_err);
- goto err_mutex;
- }
-
-- int connect_result = proxmox_backup_co_connect(pbs, task->errp);
-+ int connect_result = proxmox_backup_co_connect(pbs, errp);
- if (connect_result < 0)
- goto err_mutex;
-
-@@ -743,9 +712,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
- bool expect_only_dirty = false;
-
-- if (use_dirty_bitmap) {
-+ if (has_use_dirty_bitmap && use_dirty_bitmap) {
- if (bitmap == NULL) {
-- bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
-+ bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, errp);
- if (!bitmap) {
- goto err_mutex;
- }
-@@ -775,12 +744,12 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- }
- }
-
-- int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
-+ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, errp);
- if (dev_id < 0) {
- goto err_mutex;
- }
-
-- if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
-+ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, errp))) {
- goto err_mutex;
- }
-
-@@ -794,10 +763,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- backup_state.stat.bitmap_list = g_list_append(backup_state.stat.bitmap_list, info);
- }
- } else if (format == BACKUP_FORMAT_VMA) {
-- vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
-+ vmaw = vma_writer_create(backup_file, uuid, &local_err);
- if (!vmaw) {
- if (local_err) {
-- error_propagate(task->errp, local_err);
-+ error_propagate(errp, local_err);
- }
- goto err_mutex;
- }
-@@ -808,25 +777,25 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-
-- if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
-+ if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, errp))) {
- goto err_mutex;
- }
-
- const char *devname = bdrv_get_device_name(di->bs);
- di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
- if (di->dev_id <= 0) {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
- "register_stream failed");
- goto err_mutex;
- }
- }
- } else if (format == BACKUP_FORMAT_DIR) {
-- if (mkdir(task->backup_file, 0640) != 0) {
-- error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
-- task->backup_file);
-+ if (mkdir(backup_file, 0640) != 0) {
-+ error_setg_errno(errp, errno, "can't create directory '%s'\n",
-+ backup_file);
- goto err_mutex;
- }
-- backup_dir = task->backup_file;
-+ backup_dir = backup_file;
-
- l = di_list;
- while (l) {
-@@ -840,34 +809,34 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
- di->size, flags, false, &local_err);
- if (local_err) {
-- error_propagate(task->errp, local_err);
-+ error_propagate(errp, local_err);
- goto err_mutex;
- }
-
- di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
- if (!di->target) {
-- error_propagate(task->errp, local_err);
-+ error_propagate(errp, local_err);
- goto err_mutex;
- }
- }
- } else {
-- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
- goto err_mutex;
- }
-
-
- /* add configuration file to archive */
-- if (task->has_config_file) {
-- if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
-- vmaw, pbs, task->errp) != 0) {
-+ if (has_config_file) {
-+ if (pvebackup_co_add_config(config_file, config_name, format, backup_dir,
-+ vmaw, pbs, errp) != 0) {
- goto err_mutex;
- }
- }
-
- /* add firewall file to archive */
-- if (task->has_firewall_file) {
-- if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
-- vmaw, pbs, task->errp) != 0) {
-+ if (has_firewall_file) {
-+ if (pvebackup_co_add_config(firewall_file, firewall_name, format, backup_dir,
-+ vmaw, pbs, errp) != 0) {
- goto err_mutex;
- }
- }
-@@ -885,7 +854,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- if (backup_state.stat.backup_file) {
- g_free(backup_state.stat.backup_file);
- }
-- backup_state.stat.backup_file = g_strdup(task->backup_file);
-+ backup_state.stat.backup_file = g_strdup(backup_file);
-
- uuid_copy(backup_state.stat.uuid, uuid);
- uuid_unparse_lower(uuid, backup_state.stat.uuid_str);
-@@ -900,7 +869,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-
- qemu_mutex_unlock(&backup_state.stat.lock);
-
-- backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
-+ backup_state.speed = (has_speed && speed > 0) ? speed : 0;
-
- backup_state.vmaw = vmaw;
- backup_state.pbs = pbs;
-@@ -910,8 +879,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- uuid_info = g_malloc0(sizeof(*uuid_info));
- uuid_info->UUID = uuid_str;
-
-- task->result = uuid_info;
--
- /* Run create_backup_jobs_bh outside of coroutine (in BH) but keep
- * backup_mutex locked. This is fine, a CoMutex can be held across yield
- * points, and we'll release it as soon as the BH reschedules us.
-@@ -925,7 +892,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- qemu_coroutine_yield();
-
- if (local_err) {
-- error_propagate(task->errp, local_err);
-+ error_propagate(errp, local_err);
- goto err;
- }
-
-@@ -938,7 +905,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- /* start the first job in the transaction */
- job_txn_start_seq(backup_state.txn);
-
-- return;
-+ return uuid_info;
-
- err_mutex:
- qemu_mutex_unlock(&backup_state.stat.lock);
-@@ -969,7 +936,7 @@ err:
- if (vmaw) {
- Error *err = NULL;
- vma_writer_close(vmaw, &err);
-- unlink(task->backup_file);
-+ unlink(backup_file);
- }
-
- if (pbs) {
-@@ -980,65 +947,8 @@ err:
- rmdir(backup_dir);
- }
-
-- task->result = NULL;
--
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-- return;
--}
--
--UuidInfo *qmp_backup(
-- const char *backup_file,
-- bool has_password, const char *password,
-- bool has_keyfile, const char *keyfile,
-- bool has_key_password, const char *key_password,
-- bool has_fingerprint, const char *fingerprint,
-- bool has_backup_id, const char *backup_id,
-- bool has_backup_time, int64_t backup_time,
-- bool has_use_dirty_bitmap, bool use_dirty_bitmap,
-- bool has_compress, bool compress,
-- bool has_encrypt, bool encrypt,
-- bool has_format, BackupFormat format,
-- bool has_config_file, const char *config_file,
-- bool has_firewall_file, const char *firewall_file,
-- bool has_devlist, const char *devlist,
-- bool has_speed, int64_t speed, Error **errp)
--{
-- QmpBackupTask task = {
-- .backup_file = backup_file,
-- .has_password = has_password,
-- .password = password,
-- .has_keyfile = has_keyfile,
-- .keyfile = keyfile,
-- .has_key_password = has_key_password,
-- .key_password = key_password,
-- .has_fingerprint = has_fingerprint,
-- .fingerprint = fingerprint,
-- .has_backup_id = has_backup_id,
-- .backup_id = backup_id,
-- .has_backup_time = has_backup_time,
-- .backup_time = backup_time,
-- .has_use_dirty_bitmap = has_use_dirty_bitmap,
-- .use_dirty_bitmap = use_dirty_bitmap,
-- .has_compress = has_compress,
-- .compress = compress,
-- .has_encrypt = has_encrypt,
-- .encrypt = encrypt,
-- .has_format = has_format,
-- .format = format,
-- .has_config_file = has_config_file,
-- .config_file = config_file,
-- .has_firewall_file = has_firewall_file,
-- .firewall_file = firewall_file,
-- .has_devlist = has_devlist,
-- .devlist = devlist,
-- .has_speed = has_speed,
-- .speed = speed,
-- .errp = errp,
-- };
--
-- block_on_coroutine_fn(pvebackup_co_prepare, &task);
--
-- return task.result;
-+ return NULL;
- }
-
- BackupStatus *qmp_query_backup(Error **errp)
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 05382a1069..f0399bf249 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -846,7 +846,7 @@
- '*config-file': 'str',
- '*firewall-file': 'str',
- '*devlist': 'str', '*speed': 'int' },
-- 'returns': 'UuidInfo' }
-+ 'returns': 'UuidInfo', 'coroutine': true }
-
- ##
- # @query-backup:
-@@ -868,7 +868,7 @@
- # Notes: This command succeeds even if there is no backup process running.
- #
- ##
--{ 'command': 'backup-cancel' }
-+{ 'command': 'backup-cancel', 'coroutine': true }
-
- ##
- # @ProxmoxSupportStatus:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Tue, 17 Nov 2020 10:51:05 +0100
+Subject: [PATCH] PVE: fall back to open-iscsi initiatorname
+
+When no explicit option is given, try reading the initiator name from
+/etc/iscsi/initiatorname.iscsi and only use the generic fallback, i.e.
+iqn.2008-11.org.linux-kvmXXX, as a third alternative.
+
+This avoids the need to add an explicit option for vma and to explicitly set it
+for each call to qemu that deals with iSCSI disks, while still allowing to set
+the option if a different name is needed.
+
+According to RFC 3720, an initiator name is at most 223 bytes long, so the
+4 KiB buffer is big enough, even if many whitespaces are used.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/iscsi.c | 30 ++++++++++++++++++++++++++++++
+ 1 file changed, 30 insertions(+)
+
+diff --git a/block/iscsi.c b/block/iscsi.c
+index d707d0b354..da6ed52323 100644
+--- a/block/iscsi.c
++++ b/block/iscsi.c
+@@ -1386,12 +1386,42 @@ static char *get_initiator_name(QemuOpts *opts)
+ const char *name;
+ char *iscsi_name;
+ UuidInfo *uuid_info;
++ FILE *name_fh;
+
+ name = qemu_opt_get(opts, "initiator-name");
+ if (name) {
+ return g_strdup(name);
+ }
+
++ name_fh = fopen("/etc/iscsi/initiatorname.iscsi", "r");
++ if (name_fh) {
++ const char *key = "InitiatorName";
++ char buffer[4096];
++ char *line;
++
++ while ((line = fgets(buffer, sizeof(buffer), name_fh))) {
++ line = g_strstrip(line);
++ if (!strncmp(line, key, strlen(key))) {
++ line = strchr(line, '=');
++ if (!line || strlen(line) == 1) {
++ continue;
++ }
++ line++;
++ g_strstrip(line);
++ if (!strlen(line)) {
++ continue;
++ }
++ name = line;
++ break;
++ }
++ }
++ fclose(name_fh);
++
++ if (name) {
++ return g_strdup(name);
++ }
++ }
++
+ uuid_info = qmp_query_uuid(NULL);
+ if (strcmp(uuid_info->UUID, UUID_NONE) == 0) {
+ name = qemu_get_vm_name();
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 10 Feb 2021 11:07:06 +0100
-Subject: [PATCH] PBS: add master key support
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-this requires a new enough libproxmox-backup-qemu0, and allows querying
-from the PVE side to avoid QMP calls with unsupported parameters.
-
-Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/monitor/block-hmp-cmds.c | 1 +
- pve-backup.c | 3 +++
- qapi/block-core.json | 7 +++++++
- 3 files changed, 11 insertions(+)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index ef45552e3b..4c799f00d9 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1039,6 +1039,7 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
- false, NULL, // PBS password
- false, NULL, // PBS keyfile
- false, NULL, // PBS key_password
-+ false, NULL, // PBS master_keyfile
- false, NULL, // PBS fingerprint
- false, NULL, // PBS backup-id
- false, 0, // PBS backup-time
-diff --git a/pve-backup.c b/pve-backup.c
-index 323014744c..9f6c04a512 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -533,6 +533,7 @@ UuidInfo coroutine_fn *qmp_backup(
- bool has_password, const char *password,
- bool has_keyfile, const char *keyfile,
- bool has_key_password, const char *key_password,
-+ bool has_master_keyfile, const char *master_keyfile,
- bool has_fingerprint, const char *fingerprint,
- bool has_backup_id, const char *backup_id,
- bool has_backup_time, int64_t backup_time,
-@@ -681,6 +682,7 @@ UuidInfo coroutine_fn *qmp_backup(
- has_password ? password : NULL,
- has_keyfile ? keyfile : NULL,
- has_key_password ? key_password : NULL,
-+ has_master_keyfile ? master_keyfile : NULL,
- has_compress ? compress : true,
- has_encrypt ? encrypt : has_keyfile,
- has_fingerprint ? fingerprint : NULL,
-@@ -1044,5 +1046,6 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
- ret->pbs_dirty_bitmap_savevm = true;
- ret->pbs_dirty_bitmap_migration = true;
- ret->query_bitmap_info = true;
-+ ret->pbs_masterkey = true;
- return ret;
- }
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index f0399bf249..13e03ca154 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -817,6 +817,8 @@
- #
- # @key-password: password for keyfile (optional for format 'pbs')
- #
-+# @master-keyfile: PEM-formatted master public keyfile (optional for format 'pbs')
-+#
- # @fingerprint: server cert fingerprint (optional for format 'pbs')
- #
- # @backup-id: backup ID (required for format 'pbs')
-@@ -836,6 +838,7 @@
- '*password': 'str',
- '*keyfile': 'str',
- '*key-password': 'str',
-+ '*master-keyfile': 'str',
- '*fingerprint': 'str',
- '*backup-id': 'str',
- '*backup-time': 'int',
-@@ -888,6 +891,9 @@
- # migration cap if this is false/unset may lead
- # to crashes on migration!
- #
-+# @pbs-masterkey: True if the QMP backup call supports the 'master_keyfile'
-+# parameter.
-+#
- # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
- #
- ##
-@@ -896,6 +902,7 @@
- 'query-bitmap-info': 'bool',
- 'pbs-dirty-bitmap-savevm': 'bool',
- 'pbs-dirty-bitmap-migration': 'bool',
-+ 'pbs-masterkey': 'bool',
- 'pbs-library-version': 'str' } }
-
- ##
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Tue, 26 Jan 2021 15:45:30 +0100
+Subject: [PATCH] PVE: Use coroutine QMP for backup/cancel_backup
+
+Finally turn backup QMP calls into coroutines, now that it's possible.
+This has the benefit that calls are asynchronous to the main loop, i.e.
+long running operations like connecting to a PBS server will no longer
+hang the VM.
+
+Additionally, it allows us to get rid of block_on_coroutine_fn, which
+was always a hacky workaround.
+
+While we're already spring cleaning, also remove the QmpBackupTask
+struct, since we can now put the 'prepare' function directly into
+qmp_backup and thus no longer need those giant walls of text.
+
+(Note that for our patches to work with 5.2.0 this change is actually
+required, otherwise monitor_get_fd() fails as we're not in a QMP
+coroutine, but one we start ourselves - we could of course set the
+monitor for that coroutine ourselves, but let's just fix it the right
+way instead)
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/monitor/block-hmp-cmds.c | 4 +-
+ hmp-commands.hx | 2 +
+ proxmox-backup-client.c | 31 -----
+ pve-backup.c | 232 ++++++++++-----------------------
+ qapi/block-core.json | 4 +-
+ 5 files changed, 77 insertions(+), 196 deletions(-)
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index ea7b665aa2..ef45552e3b 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1016,7 +1016,7 @@ void hmp_info_snapshots(Monitor *mon, const QDict *qdict)
+ g_free(global_snapshots);
+ }
+
+-void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
++void coroutine_fn hmp_backup_cancel(Monitor *mon, const QDict *qdict)
+ {
+ Error *error = NULL;
+
+@@ -1025,7 +1025,7 @@ void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
+ hmp_handle_error(mon, error);
+ }
+
+-void hmp_backup(Monitor *mon, const QDict *qdict)
++void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
+ {
+ Error *error = NULL;
+
+diff --git a/hmp-commands.hx b/hmp-commands.hx
+index 97f24942b3..7a2be816da 100644
+--- a/hmp-commands.hx
++++ b/hmp-commands.hx
+@@ -111,6 +111,7 @@ ERST
+ "\n\t\t\t Use -d to dump data into a directory instead"
+ "\n\t\t\t of using VMA format.",
+ .cmd = hmp_backup,
++ .coroutine = true,
+ },
+
+ SRST
+@@ -124,6 +125,7 @@ ERST
+ .params = "",
+ .help = "cancel the current VM backup",
+ .cmd = hmp_backup_cancel,
++ .coroutine = true,
+ },
+
+ SRST
+diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
+index 4ce7bc0b5e..0923037dec 100644
+--- a/proxmox-backup-client.c
++++ b/proxmox-backup-client.c
+@@ -5,37 +5,6 @@
+
+ /* Proxmox Backup Server client bindings using coroutines */
+
+-typedef struct BlockOnCoroutineWrapper {
+- AioContext *ctx;
+- CoroutineEntry *entry;
+- void *entry_arg;
+- bool finished;
+-} BlockOnCoroutineWrapper;
+-
+-static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
+-{
+- BlockOnCoroutineWrapper *wrapper = opaque;
+- wrapper->entry(wrapper->entry_arg);
+- wrapper->finished = true;
+- aio_wait_kick();
+-}
+-
+-void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
+-{
+- assert(!qemu_in_coroutine());
+-
+- AioContext *ctx = qemu_get_current_aio_context();
+- BlockOnCoroutineWrapper wrapper = {
+- .finished = false,
+- .entry = entry,
+- .entry_arg = entry_arg,
+- .ctx = ctx,
+- };
+- Coroutine *wrapper_co = qemu_coroutine_create(block_on_coroutine_wrapper, &wrapper);
+- aio_co_enter(ctx, wrapper_co);
+- AIO_WAIT_WHILE(ctx, !wrapper.finished);
+-}
+-
+ // This is called from another thread, so we use aio_co_schedule()
+ static void proxmox_backup_schedule_wake(void *data) {
+ CoCtxData *waker = (CoCtxData *)data;
+diff --git a/pve-backup.c b/pve-backup.c
+index 5fa3cc1352..323014744c 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -357,7 +357,7 @@ static void job_cancel_bh(void *opaque) {
+ aio_co_enter(data->ctx, data->co);
+ }
+
+-static void coroutine_fn pvebackup_co_cancel(void *opaque)
++void coroutine_fn qmp_backup_cancel(Error **errp)
+ {
+ Error *cancel_err = NULL;
+ error_setg(&cancel_err, "backup canceled");
+@@ -394,11 +394,6 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ }
+
+-void qmp_backup_cancel(Error **errp)
+-{
+- block_on_coroutine_fn(pvebackup_co_cancel, NULL);
+-}
+-
+ // assumes the caller holds backup_mutex
+ static int coroutine_fn pvebackup_co_add_config(
+ const char *file,
+@@ -533,50 +528,27 @@ static void create_backup_jobs_bh(void *opaque) {
+ aio_co_enter(data->ctx, data->co);
+ }
+
+-typedef struct QmpBackupTask {
+- const char *backup_file;
+- bool has_password;
+- const char *password;
+- bool has_keyfile;
+- const char *keyfile;
+- bool has_key_password;
+- const char *key_password;
+- bool has_backup_id;
+- const char *backup_id;
+- bool has_backup_time;
+- const char *fingerprint;
+- bool has_fingerprint;
+- int64_t backup_time;
+- bool has_use_dirty_bitmap;
+- bool use_dirty_bitmap;
+- bool has_format;
+- BackupFormat format;
+- bool has_config_file;
+- const char *config_file;
+- bool has_firewall_file;
+- const char *firewall_file;
+- bool has_devlist;
+- const char *devlist;
+- bool has_compress;
+- bool compress;
+- bool has_encrypt;
+- bool encrypt;
+- bool has_speed;
+- int64_t speed;
+- Error **errp;
+- UuidInfo *result;
+-} QmpBackupTask;
+-
+-static void coroutine_fn pvebackup_co_prepare(void *opaque)
++UuidInfo coroutine_fn *qmp_backup(
++ const char *backup_file,
++ bool has_password, const char *password,
++ bool has_keyfile, const char *keyfile,
++ bool has_key_password, const char *key_password,
++ bool has_fingerprint, const char *fingerprint,
++ bool has_backup_id, const char *backup_id,
++ bool has_backup_time, int64_t backup_time,
++ bool has_use_dirty_bitmap, bool use_dirty_bitmap,
++ bool has_compress, bool compress,
++ bool has_encrypt, bool encrypt,
++ bool has_format, BackupFormat format,
++ bool has_config_file, const char *config_file,
++ bool has_firewall_file, const char *firewall_file,
++ bool has_devlist, const char *devlist,
++ bool has_speed, int64_t speed, Error **errp)
+ {
+ assert(qemu_in_coroutine());
+
+ qemu_co_mutex_lock(&backup_state.backup_mutex);
+
+- QmpBackupTask *task = opaque;
+-
+- task->result = NULL; // just to be sure
+-
+ BlockBackend *blk;
+ BlockDriverState *bs = NULL;
+ const char *backup_dir = NULL;
+@@ -593,17 +565,17 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ const char *firewall_name = "qemu-server.fw";
+
+ if (backup_state.di_list) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+ "previous backup not finished");
+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+- return;
++ return NULL;
+ }
+
+ /* Todo: try to auto-detect format based on file name */
+- BackupFormat format = task->has_format ? task->format : BACKUP_FORMAT_VMA;
++ format = has_format ? format : BACKUP_FORMAT_VMA;
+
+- if (task->has_devlist) {
+- devs = g_strsplit_set(task->devlist, ",;:", -1);
++ if (has_devlist) {
++ devs = g_strsplit_set(devlist, ",;:", -1);
+
+ gchar **d = devs;
+ while (d && *d) {
+@@ -611,14 +583,14 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ if (blk) {
+ bs = blk_bs(blk);
+ if (!bdrv_is_inserted(bs)) {
+- error_setg(task->errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
++ error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
+ goto err;
+ }
+ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
+ di->bs = bs;
+ di_list = g_list_append(di_list, di);
+ } else {
+- error_set(task->errp, ERROR_CLASS_DEVICE_NOT_FOUND,
++ error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
+ "Device '%s' not found", *d);
+ goto err;
+ }
+@@ -641,7 +613,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+
+ if (!di_list) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
+ goto err;
+ }
+
+@@ -651,13 +623,13 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ while (l) {
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+ l = g_list_next(l);
+- if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, task->errp)) {
++ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, errp)) {
+ goto err;
+ }
+
+ ssize_t size = bdrv_getlength(di->bs);
+ if (size < 0) {
+- error_setg_errno(task->errp, -di->size, "bdrv_getlength failed");
++ error_setg_errno(errp, -di->size, "bdrv_getlength failed");
+ goto err;
+ }
+ di->size = size;
+@@ -684,47 +656,44 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+
+ if (format == BACKUP_FORMAT_PBS) {
+- if (!task->has_password) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
++ if (!has_password) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
+ goto err_mutex;
+ }
+- if (!task->has_backup_id) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
++ if (!has_backup_id) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
+ goto err_mutex;
+ }
+- if (!task->has_backup_time) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
++ if (!has_backup_time) {
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
+ goto err_mutex;
+ }
+
+ int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
+ firewall_name = "fw.conf";
+
+- bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
+-
+-
+ char *pbs_err = NULL;
+ pbs = proxmox_backup_new(
+- task->backup_file,
+- task->backup_id,
+- task->backup_time,
++ backup_file,
++ backup_id,
++ backup_time,
+ dump_cb_block_size,
+- task->has_password ? task->password : NULL,
+- task->has_keyfile ? task->keyfile : NULL,
+- task->has_key_password ? task->key_password : NULL,
+- task->has_compress ? task->compress : true,
+- task->has_encrypt ? task->encrypt : task->has_keyfile,
+- task->has_fingerprint ? task->fingerprint : NULL,
++ has_password ? password : NULL,
++ has_keyfile ? keyfile : NULL,
++ has_key_password ? key_password : NULL,
++ has_compress ? compress : true,
++ has_encrypt ? encrypt : has_keyfile,
++ has_fingerprint ? fingerprint : NULL,
+ &pbs_err);
+
+ if (!pbs) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+ "proxmox_backup_new failed: %s", pbs_err);
+ proxmox_backup_free_error(pbs_err);
+ goto err_mutex;
+ }
+
+- int connect_result = proxmox_backup_co_connect(pbs, task->errp);
++ int connect_result = proxmox_backup_co_connect(pbs, errp);
+ if (connect_result < 0)
+ goto err_mutex;
+
+@@ -743,9 +712,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
+ bool expect_only_dirty = false;
+
+- if (use_dirty_bitmap) {
++ if (has_use_dirty_bitmap && use_dirty_bitmap) {
+ if (bitmap == NULL) {
+- bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
++ bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, errp);
+ if (!bitmap) {
+ goto err_mutex;
+ }
+@@ -775,12 +744,12 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ }
+ }
+
+- int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
++ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, errp);
+ if (dev_id < 0) {
+ goto err_mutex;
+ }
+
+- if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
++ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, errp))) {
+ goto err_mutex;
+ }
+
+@@ -794,10 +763,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ backup_state.stat.bitmap_list = g_list_append(backup_state.stat.bitmap_list, info);
+ }
+ } else if (format == BACKUP_FORMAT_VMA) {
+- vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
++ vmaw = vma_writer_create(backup_file, uuid, &local_err);
+ if (!vmaw) {
+ if (local_err) {
+- error_propagate(task->errp, local_err);
++ error_propagate(errp, local_err);
+ }
+ goto err_mutex;
+ }
+@@ -808,25 +777,25 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+ l = g_list_next(l);
+
+- if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
++ if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, errp))) {
+ goto err_mutex;
+ }
+
+ const char *devname = bdrv_get_device_name(di->bs);
+ di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
+ if (di->dev_id <= 0) {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
+ "register_stream failed");
+ goto err_mutex;
+ }
+ }
+ } else if (format == BACKUP_FORMAT_DIR) {
+- if (mkdir(task->backup_file, 0640) != 0) {
+- error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
+- task->backup_file);
++ if (mkdir(backup_file, 0640) != 0) {
++ error_setg_errno(errp, errno, "can't create directory '%s'\n",
++ backup_file);
+ goto err_mutex;
+ }
+- backup_dir = task->backup_file;
++ backup_dir = backup_file;
+
+ l = di_list;
+ while (l) {
+@@ -840,34 +809,34 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
+ di->size, flags, false, &local_err);
+ if (local_err) {
+- error_propagate(task->errp, local_err);
++ error_propagate(errp, local_err);
+ goto err_mutex;
+ }
+
+ di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
+ if (!di->target) {
+- error_propagate(task->errp, local_err);
++ error_propagate(errp, local_err);
+ goto err_mutex;
+ }
+ }
+ } else {
+- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
++ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
+ goto err_mutex;
+ }
+
+
+ /* add configuration file to archive */
+- if (task->has_config_file) {
+- if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
+- vmaw, pbs, task->errp) != 0) {
++ if (has_config_file) {
++ if (pvebackup_co_add_config(config_file, config_name, format, backup_dir,
++ vmaw, pbs, errp) != 0) {
+ goto err_mutex;
+ }
+ }
+
+ /* add firewall file to archive */
+- if (task->has_firewall_file) {
+- if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
+- vmaw, pbs, task->errp) != 0) {
++ if (has_firewall_file) {
++ if (pvebackup_co_add_config(firewall_file, firewall_name, format, backup_dir,
++ vmaw, pbs, errp) != 0) {
+ goto err_mutex;
+ }
+ }
+@@ -885,7 +854,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ if (backup_state.stat.backup_file) {
+ g_free(backup_state.stat.backup_file);
+ }
+- backup_state.stat.backup_file = g_strdup(task->backup_file);
++ backup_state.stat.backup_file = g_strdup(backup_file);
+
+ uuid_copy(backup_state.stat.uuid, uuid);
+ uuid_unparse_lower(uuid, backup_state.stat.uuid_str);
+@@ -900,7 +869,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+
+ qemu_mutex_unlock(&backup_state.stat.lock);
+
+- backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
++ backup_state.speed = (has_speed && speed > 0) ? speed : 0;
+
+ backup_state.vmaw = vmaw;
+ backup_state.pbs = pbs;
+@@ -910,8 +879,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ uuid_info = g_malloc0(sizeof(*uuid_info));
+ uuid_info->UUID = uuid_str;
+
+- task->result = uuid_info;
+-
+ /* Run create_backup_jobs_bh outside of coroutine (in BH) but keep
+ * backup_mutex locked. This is fine, a CoMutex can be held across yield
+ * points, and we'll release it as soon as the BH reschedules us.
+@@ -925,7 +892,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ qemu_coroutine_yield();
+
+ if (local_err) {
+- error_propagate(task->errp, local_err);
++ error_propagate(errp, local_err);
+ goto err;
+ }
+
+@@ -938,7 +905,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+ /* start the first job in the transaction */
+ job_txn_start_seq(backup_state.txn);
+
+- return;
++ return uuid_info;
+
+ err_mutex:
+ qemu_mutex_unlock(&backup_state.stat.lock);
+@@ -969,7 +936,7 @@ err:
+ if (vmaw) {
+ Error *err = NULL;
+ vma_writer_close(vmaw, &err);
+- unlink(task->backup_file);
++ unlink(backup_file);
+ }
+
+ if (pbs) {
+@@ -980,65 +947,8 @@ err:
+ rmdir(backup_dir);
+ }
+
+- task->result = NULL;
+-
+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+- return;
+-}
+-
+-UuidInfo *qmp_backup(
+- const char *backup_file,
+- bool has_password, const char *password,
+- bool has_keyfile, const char *keyfile,
+- bool has_key_password, const char *key_password,
+- bool has_fingerprint, const char *fingerprint,
+- bool has_backup_id, const char *backup_id,
+- bool has_backup_time, int64_t backup_time,
+- bool has_use_dirty_bitmap, bool use_dirty_bitmap,
+- bool has_compress, bool compress,
+- bool has_encrypt, bool encrypt,
+- bool has_format, BackupFormat format,
+- bool has_config_file, const char *config_file,
+- bool has_firewall_file, const char *firewall_file,
+- bool has_devlist, const char *devlist,
+- bool has_speed, int64_t speed, Error **errp)
+-{
+- QmpBackupTask task = {
+- .backup_file = backup_file,
+- .has_password = has_password,
+- .password = password,
+- .has_keyfile = has_keyfile,
+- .keyfile = keyfile,
+- .has_key_password = has_key_password,
+- .key_password = key_password,
+- .has_fingerprint = has_fingerprint,
+- .fingerprint = fingerprint,
+- .has_backup_id = has_backup_id,
+- .backup_id = backup_id,
+- .has_backup_time = has_backup_time,
+- .backup_time = backup_time,
+- .has_use_dirty_bitmap = has_use_dirty_bitmap,
+- .use_dirty_bitmap = use_dirty_bitmap,
+- .has_compress = has_compress,
+- .compress = compress,
+- .has_encrypt = has_encrypt,
+- .encrypt = encrypt,
+- .has_format = has_format,
+- .format = format,
+- .has_config_file = has_config_file,
+- .config_file = config_file,
+- .has_firewall_file = has_firewall_file,
+- .firewall_file = firewall_file,
+- .has_devlist = has_devlist,
+- .devlist = devlist,
+- .has_speed = has_speed,
+- .speed = speed,
+- .errp = errp,
+- };
+-
+- block_on_coroutine_fn(pvebackup_co_prepare, &task);
+-
+- return task.result;
++ return NULL;
+ }
+
+ BackupStatus *qmp_query_backup(Error **errp)
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index bd978ea562..ca1966f54b 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -842,7 +842,7 @@
+ '*config-file': 'str',
+ '*firewall-file': 'str',
+ '*devlist': 'str', '*speed': 'int' },
+- 'returns': 'UuidInfo' }
++ 'returns': 'UuidInfo', 'coroutine': true }
+
+ ##
+ # @query-backup:
+@@ -864,7 +864,7 @@
+ # Notes: This command succeeds even if there is no backup process running.
+ #
+ ##
+-{ 'command': 'backup-cancel' }
++{ 'command': 'backup-cancel', 'coroutine': true }
+
+ ##
+ # @ProxmoxSupportStatus:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 10 Feb 2021 11:07:06 +0100
+Subject: [PATCH] PBS: add master key support
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+this requires a new enough libproxmox-backup-qemu0, and allows querying
+from the PVE side to avoid QMP calls with unsupported parameters.
+
+Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/monitor/block-hmp-cmds.c | 1 +
+ pve-backup.c | 3 +++
+ qapi/block-core.json | 7 +++++++
+ 3 files changed, 11 insertions(+)
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index ef45552e3b..4c799f00d9 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1039,6 +1039,7 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
+ false, NULL, // PBS password
+ false, NULL, // PBS keyfile
+ false, NULL, // PBS key_password
++ false, NULL, // PBS master_keyfile
+ false, NULL, // PBS fingerprint
+ false, NULL, // PBS backup-id
+ false, 0, // PBS backup-time
+diff --git a/pve-backup.c b/pve-backup.c
+index 323014744c..9f6c04a512 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -533,6 +533,7 @@ UuidInfo coroutine_fn *qmp_backup(
+ bool has_password, const char *password,
+ bool has_keyfile, const char *keyfile,
+ bool has_key_password, const char *key_password,
++ bool has_master_keyfile, const char *master_keyfile,
+ bool has_fingerprint, const char *fingerprint,
+ bool has_backup_id, const char *backup_id,
+ bool has_backup_time, int64_t backup_time,
+@@ -681,6 +682,7 @@ UuidInfo coroutine_fn *qmp_backup(
+ has_password ? password : NULL,
+ has_keyfile ? keyfile : NULL,
+ has_key_password ? key_password : NULL,
++ has_master_keyfile ? master_keyfile : NULL,
+ has_compress ? compress : true,
+ has_encrypt ? encrypt : has_keyfile,
+ has_fingerprint ? fingerprint : NULL,
+@@ -1044,5 +1046,6 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+ ret->pbs_dirty_bitmap_savevm = true;
+ ret->pbs_dirty_bitmap_migration = true;
+ ret->query_bitmap_info = true;
++ ret->pbs_masterkey = true;
+ return ret;
+ }
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index ca1966f54b..fc8a125451 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -813,6 +813,8 @@
+ #
+ # @key-password: password for keyfile (optional for format 'pbs')
+ #
++# @master-keyfile: PEM-formatted master public keyfile (optional for format 'pbs')
++#
+ # @fingerprint: server cert fingerprint (optional for format 'pbs')
+ #
+ # @backup-id: backup ID (required for format 'pbs')
+@@ -832,6 +834,7 @@
+ '*password': 'str',
+ '*keyfile': 'str',
+ '*key-password': 'str',
++ '*master-keyfile': 'str',
+ '*fingerprint': 'str',
+ '*backup-id': 'str',
+ '*backup-time': 'int',
+@@ -884,6 +887,9 @@
+ # migration cap if this is false/unset may lead
+ # to crashes on migration!
+ #
++# @pbs-masterkey: True if the QMP backup call supports the 'master_keyfile'
++# parameter.
++#
+ # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
+ #
+ ##
+@@ -892,6 +898,7 @@
+ 'query-bitmap-info': 'bool',
+ 'pbs-dirty-bitmap-savevm': 'bool',
+ 'pbs-dirty-bitmap-migration': 'bool',
++ 'pbs-masterkey': 'bool',
+ 'pbs-library-version': 'str' } }
+
+ ##
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 9 Dec 2020 11:46:57 +0100
-Subject: [PATCH] PVE: block/pbs: fast-path reads without allocation if
- possible
-
-...and switch over to g_malloc/g_free while at it to align with other
-QEMU code.
-
-Tracing shows the fast-path is taken almost all the time, though not
-100% so the slow one is still necessary.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/pbs.c | 17 ++++++++++++++---
- 1 file changed, 14 insertions(+), 3 deletions(-)
-
-diff --git a/block/pbs.c b/block/pbs.c
-index 0b05ea9080..c5eb4d5bad 100644
---- a/block/pbs.c
-+++ b/block/pbs.c
-@@ -200,7 +200,16 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
- BDRVPBSState *s = bs->opaque;
- int ret;
- char *pbs_error = NULL;
-- uint8_t *buf = malloc(bytes);
-+ uint8_t *buf;
-+ bool inline_buf = true;
-+
-+ /* for single-buffer IO vectors we can fast-path the write directly to it */
-+ if (qiov->niov == 1 && qiov->iov->iov_len >= bytes) {
-+ buf = qiov->iov->iov_base;
-+ } else {
-+ inline_buf = false;
-+ buf = g_malloc(bytes);
-+ }
-
- if (offset < 0 || bytes < 0) {
- fprintf(stderr, "unexpected negative 'offset' or 'bytes' value!\n");
-@@ -223,8 +232,10 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
- return -EIO;
- }
-
-- qemu_iovec_from_buf(qiov, 0, buf, bytes);
-- free(buf);
-+ if (!inline_buf) {
-+ qemu_iovec_from_buf(qiov, 0, buf, bytes);
-+ g_free(buf);
-+ }
-
- return ret;
- }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 9 Dec 2020 11:46:57 +0100
+Subject: [PATCH] PVE: block/pbs: fast-path reads without allocation if
+ possible
+
+...and switch over to g_malloc/g_free while at it to align with other
+QEMU code.
+
+Tracing shows the fast-path is taken almost all the time, though not
+100% so the slow one is still necessary.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/pbs.c | 17 ++++++++++++++---
+ 1 file changed, 14 insertions(+), 3 deletions(-)
+
+diff --git a/block/pbs.c b/block/pbs.c
+index 0b05ea9080..c5eb4d5bad 100644
+--- a/block/pbs.c
++++ b/block/pbs.c
+@@ -200,7 +200,16 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
+ BDRVPBSState *s = bs->opaque;
+ int ret;
+ char *pbs_error = NULL;
+- uint8_t *buf = malloc(bytes);
++ uint8_t *buf;
++ bool inline_buf = true;
++
++ /* for single-buffer IO vectors we can fast-path the write directly to it */
++ if (qiov->niov == 1 && qiov->iov->iov_len >= bytes) {
++ buf = qiov->iov->iov_base;
++ } else {
++ inline_buf = false;
++ buf = g_malloc(bytes);
++ }
+
+ if (offset < 0 || bytes < 0) {
+ fprintf(stderr, "unexpected negative 'offset' or 'bytes' value!\n");
+@@ -223,8 +232,10 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
+ return -EIO;
+ }
+
+- qemu_iovec_from_buf(qiov, 0, buf, bytes);
+- free(buf);
++ if (!inline_buf) {
++ qemu_iovec_from_buf(qiov, 0, buf, bytes);
++ g_free(buf);
++ }
+
+ return ret;
+ }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Tue, 2 Mar 2021 16:34:28 +0100
-Subject: [PATCH] PVE: block/stream: increase chunk size
-
-Ceph favors bigger chunks, so increase to 4M.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/stream.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/block/stream.c b/block/stream.c
-index 694709bd25..e09bd5c4ef 100644
---- a/block/stream.c
-+++ b/block/stream.c
-@@ -28,7 +28,7 @@ enum {
- * large enough to process multiple clusters in a single call, so
- * that populating contiguous regions of the image is efficient.
- */
-- STREAM_CHUNK = 512 * 1024, /* in bytes */
-+ STREAM_CHUNK = 4 * 1024 * 1024, /* in bytes */
- };
-
- typedef struct StreamBlockJob {
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Tue, 2 Mar 2021 16:34:28 +0100
+Subject: [PATCH] PVE: block/stream: increase chunk size
+
+Ceph favors bigger chunks, so increase to 4M.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/stream.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/block/stream.c b/block/stream.c
+index 694709bd25..e09bd5c4ef 100644
+--- a/block/stream.c
++++ b/block/stream.c
+@@ -28,7 +28,7 @@ enum {
+ * large enough to process multiple clusters in a single call, so
+ * that populating contiguous regions of the image is efficient.
+ */
+- STREAM_CHUNK = 512 * 1024, /* in bytes */
++ STREAM_CHUNK = 4 * 1024 * 1024, /* in bytes */
+ };
+
+ typedef struct StreamBlockJob {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Tue, 2 Mar 2021 16:11:54 +0100
-Subject: [PATCH] block/io: accept NULL qiov in bdrv_pad_request
-
-Some operations, e.g. block-stream, perform reads while discarding the
-results (only copy-on-read matters). In this case they will pass NULL as
-the target QEMUIOVector, which will however trip bdrv_pad_request, since
-it wants to extend its passed vector.
-
-Simply check for NULL and do nothing, there's no reason to pad the
-target if it will be discarded anyway.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/io.c | 4 ++++
- 1 file changed, 4 insertions(+)
-
-diff --git a/block/io.c b/block/io.c
-index 3280144a17..cd7ba08768 100644
---- a/block/io.c
-+++ b/block/io.c
-@@ -1794,6 +1794,10 @@ static int bdrv_pad_request(BlockDriverState *bs,
- {
- int ret;
-
-+ if (!qiov) {
-+ return 0;
-+ }
-+
- bdrv_check_qiov_request(*offset, *bytes, *qiov, *qiov_offset, &error_abort);
-
- if (!bdrv_init_padding(bs, *offset, *bytes, pad)) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 7 Dec 2020 15:21:03 +0100
-Subject: [PATCH] block: add alloc-track driver
-
-Add a new filter node 'alloc-track', which seperates reads and writes to
-different children, thus allowing to put a backing image behind any
-blockdev (regardless of driver support). Since we can't detect any
-pre-allocated blocks, we can only track new writes, hence the write
-target ('file') for this node must always be empty.
-
-Intended use case is for live restoring, i.e. add a backup image as a
-block device into a VM, then put an alloc-track on the restore target
-and set the backup as backing. With this, one can use a regular
-'block-stream' to restore the image, while the VM can already run in the
-background. Copy-on-read will help make progress as the VM reads as
-well.
-
-This only worked if the target supports backing images, so up until now
-only for qcow2, with alloc-track any driver for the target can be used.
-
-If 'auto-remove' is set, alloc-track will automatically detach itself
-once the backing image is removed. It will be replaced by 'file'.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[adapt to changed function signatures]
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/alloc-track.c | 350 ++++++++++++++++++++++++++++++++++++++++++++
- block/meson.build | 1 +
- 2 files changed, 351 insertions(+)
- create mode 100644 block/alloc-track.c
-
-diff --git a/block/alloc-track.c b/block/alloc-track.c
-new file mode 100644
-index 0000000000..6b50fbe537
---- /dev/null
-+++ b/block/alloc-track.c
-@@ -0,0 +1,350 @@
-+/*
-+ * Node to allow backing images to be applied to any node. Assumes a blank
-+ * image to begin with, only new writes are tracked as allocated, thus this
-+ * must never be put on a node that already contains data.
-+ *
-+ * Copyright (c) 2020 Proxmox Server Solutions GmbH
-+ * Copyright (c) 2020 Stefan Reiter <s.reiter@proxmox.com>
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ */
-+
-+#include "qemu/osdep.h"
-+#include "qapi/error.h"
-+#include "block/block_int.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qstring.h"
-+#include "qemu/cutils.h"
-+#include "qemu/option.h"
-+#include "qemu/module.h"
-+#include "sysemu/block-backend.h"
-+
-+#define TRACK_OPT_AUTO_REMOVE "auto-remove"
-+
-+typedef enum DropState {
-+ DropNone,
-+ DropRequested,
-+ DropInProgress,
-+} DropState;
-+
-+typedef struct {
-+ BdrvDirtyBitmap *bitmap;
-+ DropState drop_state;
-+ bool auto_remove;
-+} BDRVAllocTrackState;
-+
-+static QemuOptsList runtime_opts = {
-+ .name = "alloc-track",
-+ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
-+ .desc = {
-+ {
-+ .name = TRACK_OPT_AUTO_REMOVE,
-+ .type = QEMU_OPT_BOOL,
-+ .help = "automatically replace this node with 'file' when 'backing'"
-+ "is detached",
-+ },
-+ { /* end of list */ }
-+ },
-+};
-+
-+static void track_refresh_limits(BlockDriverState *bs, Error **errp)
-+{
-+ BlockDriverInfo bdi;
-+
-+ if (!bs->file) {
-+ return;
-+ }
-+
-+ /* always use alignment from underlying write device so RMW cycle for
-+ * bdrv_pwritev reads data from our backing via track_co_preadv (no partial
-+ * cluster allocation in 'file') */
-+ bdrv_get_info(bs->file->bs, &bdi);
-+ bs->bl.request_alignment = MAX(bs->file->bs->bl.request_alignment,
-+ MAX(bdi.cluster_size, BDRV_SECTOR_SIZE));
-+}
-+
-+static int track_open(BlockDriverState *bs, QDict *options, int flags,
-+ Error **errp)
-+{
-+ BDRVAllocTrackState *s = bs->opaque;
-+ QemuOpts *opts;
-+ Error *local_err = NULL;
-+ int ret = 0;
-+
-+ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
-+ qemu_opts_absorb_qdict(opts, options, &local_err);
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ ret = -EINVAL;
-+ goto fail;
-+ }
-+
-+ s->auto_remove = qemu_opt_get_bool(opts, TRACK_OPT_AUTO_REMOVE, false);
-+
-+ /* open the target (write) node, backing will be attached by block layer */
-+ bs->file = bdrv_open_child(NULL, options, "file", bs, &child_of_bds,
-+ BDRV_CHILD_DATA | BDRV_CHILD_METADATA, false,
-+ &local_err);
-+ if (local_err) {
-+ ret = -EINVAL;
-+ error_propagate(errp, local_err);
-+ goto fail;
-+ }
-+
-+ track_refresh_limits(bs, errp);
-+ uint64_t gran = bs->bl.request_alignment;
-+ s->bitmap = bdrv_create_dirty_bitmap(bs->file->bs, gran, NULL, &local_err);
-+ if (local_err) {
-+ ret = -EIO;
-+ error_propagate(errp, local_err);
-+ goto fail;
-+ }
-+
-+ s->drop_state = DropNone;
-+
-+fail:
-+ if (ret < 0) {
-+ bdrv_unref_child(bs, bs->file);
-+ if (s->bitmap) {
-+ bdrv_release_dirty_bitmap(s->bitmap);
-+ }
-+ }
-+ qemu_opts_del(opts);
-+ return ret;
-+}
-+
-+static void track_close(BlockDriverState *bs)
-+{
-+ BDRVAllocTrackState *s = bs->opaque;
-+ if (s->bitmap) {
-+ bdrv_release_dirty_bitmap(s->bitmap);
-+ }
-+}
-+
-+static int64_t track_getlength(BlockDriverState *bs)
-+{
-+ return bdrv_getlength(bs->file->bs);
-+}
-+
-+static int coroutine_fn track_co_preadv(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
-+{
-+ BDRVAllocTrackState *s = bs->opaque;
-+ QEMUIOVector local_qiov;
-+ int ret;
-+
-+ /* 'cur_offset' is relative to 'offset', 'local_offset' to image start */
-+ uint64_t cur_offset, local_offset;
-+ int64_t local_bytes;
-+ bool alloc;
-+
-+ if (offset < 0 || bytes < 0) {
-+ fprintf(stderr, "unexpected negative 'offset' or 'bytes' value!\n");
-+ return -EINVAL;
-+ }
-+
-+ /* a read request can span multiple granularity-sized chunks, and can thus
-+ * contain blocks with different allocation status - we could just iterate
-+ * granularity-wise, but for better performance use bdrv_dirty_bitmap_next_X
-+ * to find the next flip and consider everything up to that in one go */
-+ for (cur_offset = 0; cur_offset < bytes; cur_offset += local_bytes) {
-+ local_offset = offset + cur_offset;
-+ alloc = bdrv_dirty_bitmap_get(s->bitmap, local_offset);
-+ if (alloc) {
-+ local_bytes = bdrv_dirty_bitmap_next_zero(s->bitmap, local_offset,
-+ bytes - cur_offset);
-+ } else {
-+ local_bytes = bdrv_dirty_bitmap_next_dirty(s->bitmap, local_offset,
-+ bytes - cur_offset);
-+ }
-+
-+ /* _bitmap_next_X return is -1 if no end found within limit, otherwise
-+ * offset of next flip (to start of image) */
-+ local_bytes = local_bytes < 0 ?
-+ bytes - cur_offset :
-+ local_bytes - local_offset;
-+
-+ qemu_iovec_init_slice(&local_qiov, qiov, cur_offset, local_bytes);
-+
-+ if (alloc) {
-+ ret = bdrv_co_preadv(bs->file, local_offset, local_bytes,
-+ &local_qiov, flags);
-+ } else if (bs->backing) {
-+ ret = bdrv_co_preadv(bs->backing, local_offset, local_bytes,
-+ &local_qiov, flags);
-+ } else {
-+ ret = qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
-+ }
-+
-+ if (ret != 0) {
-+ break;
-+ }
-+ }
-+
-+ return ret;
-+}
-+
-+static int coroutine_fn track_co_pwritev(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
-+{
-+ return bdrv_co_pwritev(bs->file, offset, bytes, qiov, flags);
-+}
-+
-+static int coroutine_fn track_co_pwrite_zeroes(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes, BdrvRequestFlags flags)
-+{
-+ return bdrv_co_pwrite_zeroes(bs->file, offset, bytes, flags);
-+}
-+
-+static int coroutine_fn track_co_pdiscard(BlockDriverState *bs,
-+ int64_t offset, int64_t bytes)
-+{
-+ return bdrv_co_pdiscard(bs->file, offset, bytes);
-+}
-+
-+static coroutine_fn int track_co_flush(BlockDriverState *bs)
-+{
-+ return bdrv_co_flush(bs->file->bs);
-+}
-+
-+static int coroutine_fn track_co_block_status(BlockDriverState *bs,
-+ bool want_zero,
-+ int64_t offset,
-+ int64_t bytes,
-+ int64_t *pnum,
-+ int64_t *map,
-+ BlockDriverState **file)
-+{
-+ BDRVAllocTrackState *s = bs->opaque;
-+
-+ bool alloc = bdrv_dirty_bitmap_get(s->bitmap, offset);
-+ int64_t next_flipped;
-+ if (alloc) {
-+ next_flipped = bdrv_dirty_bitmap_next_zero(s->bitmap, offset, bytes);
-+ } else {
-+ next_flipped = bdrv_dirty_bitmap_next_dirty(s->bitmap, offset, bytes);
-+ }
-+
-+ /* in case not the entire region has the same state, we need to set pnum to
-+ * indicate for how many bytes our result is valid */
-+ *pnum = next_flipped == -1 ? bytes : next_flipped - offset;
-+ *map = offset;
-+
-+ if (alloc) {
-+ *file = bs->file->bs;
-+ return BDRV_BLOCK_RAW | BDRV_BLOCK_OFFSET_VALID;
-+ } else if (bs->backing) {
-+ *file = bs->backing->bs;
-+ }
-+ return 0;
-+}
-+
-+static void track_child_perm(BlockDriverState *bs, BdrvChild *c,
-+ BdrvChildRole role, BlockReopenQueue *reopen_queue,
-+ uint64_t perm, uint64_t shared,
-+ uint64_t *nperm, uint64_t *nshared)
-+{
-+ BDRVAllocTrackState *s = bs->opaque;
-+
-+ *nshared = BLK_PERM_ALL;
-+
-+ /* in case we're currently dropping ourselves, claim to not use any
-+ * permissions at all - which is fine, since from this point on we will
-+ * never issue a read or write anymore */
-+ if (s->drop_state == DropInProgress) {
-+ *nperm = 0;
-+ return;
-+ }
-+
-+ if (role & BDRV_CHILD_DATA) {
-+ *nperm = perm & DEFAULT_PERM_PASSTHROUGH;
-+ } else {
-+ /* 'backing' is also a child of our BDS, but we don't expect it to be
-+ * writeable, so we only forward 'consistent read' */
-+ *nperm = perm & BLK_PERM_CONSISTENT_READ;
-+ }
-+}
-+
-+static void track_drop(void *opaque)
-+{
-+ BlockDriverState *bs = (BlockDriverState*)opaque;
-+ BlockDriverState *file = bs->file->bs;
-+ BDRVAllocTrackState *s = bs->opaque;
-+
-+ assert(file);
-+
-+ /* we rely on the fact that we're not used anywhere else, so let's wait
-+ * until we're only used once - in the drive connected to the guest (and one
-+ * ref is held by bdrv_ref in track_change_backing_file) */
-+ if (bs->refcnt > 2) {
-+ aio_bh_schedule_oneshot(qemu_get_aio_context(), track_drop, opaque);
-+ return;
-+ }
-+ AioContext *aio_context = bdrv_get_aio_context(bs);
-+ aio_context_acquire(aio_context);
-+
-+ bdrv_drained_begin(bs);
-+
-+ /* now that we're drained, we can safely set 'DropInProgress' */
-+ s->drop_state = DropInProgress;
-+ bdrv_child_refresh_perms(bs, bs->file, &error_abort);
-+
-+ bdrv_replace_node(bs, file, &error_abort);
-+ bdrv_set_backing_hd(bs, NULL, &error_abort);
-+ bdrv_drained_end(bs);
-+ bdrv_unref(bs);
-+ aio_context_release(aio_context);
-+}
-+
-+static int track_change_backing_file(BlockDriverState *bs,
-+ const char *backing_file,
-+ const char *backing_fmt)
-+{
-+ BDRVAllocTrackState *s = bs->opaque;
-+ if (s->auto_remove && s->drop_state == DropNone &&
-+ backing_file == NULL && backing_fmt == NULL)
-+ {
-+ /* backing file has been disconnected, there's no longer any use for
-+ * this node, so let's remove ourselves from the block graph - we need
-+ * to schedule this for later however, since when this function is
-+ * called, the blockjob modifying us is probably not done yet and has a
-+ * blocker on 'bs' */
-+ s->drop_state = DropRequested;
-+ bdrv_ref(bs);
-+ aio_bh_schedule_oneshot(qemu_get_aio_context(), track_drop, (void*)bs);
-+ }
-+
-+ return 0;
-+}
-+
-+static BlockDriver bdrv_alloc_track = {
-+ .format_name = "alloc-track",
-+ .instance_size = sizeof(BDRVAllocTrackState),
-+
-+ .bdrv_file_open = track_open,
-+ .bdrv_close = track_close,
-+ .bdrv_getlength = track_getlength,
-+ .bdrv_child_perm = track_child_perm,
-+ .bdrv_refresh_limits = track_refresh_limits,
-+
-+ .bdrv_co_pwrite_zeroes = track_co_pwrite_zeroes,
-+ .bdrv_co_pwritev = track_co_pwritev,
-+ .bdrv_co_preadv = track_co_preadv,
-+ .bdrv_co_pdiscard = track_co_pdiscard,
-+
-+ .bdrv_co_flush = track_co_flush,
-+ .bdrv_co_flush_to_disk = track_co_flush,
-+
-+ .supports_backing = true,
-+
-+ .bdrv_co_block_status = track_co_block_status,
-+ .bdrv_change_backing_file = track_change_backing_file,
-+};
-+
-+static void bdrv_alloc_track_init(void)
-+{
-+ bdrv_register(&bdrv_alloc_track);
-+}
-+
-+block_init(bdrv_alloc_track_init);
-diff --git a/block/meson.build b/block/meson.build
-index f94cc0cd25..1716febb1d 100644
---- a/block/meson.build
-+++ b/block/meson.build
-@@ -2,6 +2,7 @@ block_ss.add(genh)
- block_ss.add(files(
- 'accounting.c',
- 'aio_task.c',
-+ 'alloc-track.c',
- 'amend.c',
- 'backup.c',
- 'backup-dump.c',
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Tue, 2 Mar 2021 16:11:54 +0100
+Subject: [PATCH] block/io: accept NULL qiov in bdrv_pad_request
+
+Some operations, e.g. block-stream, perform reads while discarding the
+results (only copy-on-read matters). In this case they will pass NULL as
+the target QEMUIOVector, which will however trip bdrv_pad_request, since
+it wants to extend its passed vector.
+
+Simply check for NULL and do nothing, there's no reason to pad the
+target if it will be discarded anyway.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/io.c | 4 ++++
+ 1 file changed, 4 insertions(+)
+
+diff --git a/block/io.c b/block/io.c
+index 0a8cbefe86..531b3b7a2d 100644
+--- a/block/io.c
++++ b/block/io.c
+@@ -1734,6 +1734,10 @@ static int bdrv_pad_request(BlockDriverState *bs,
+ {
+ int ret;
+
++ if (!qiov) {
++ return 0;
++ }
++
+ bdrv_check_qiov_request(*offset, *bytes, *qiov, *qiov_offset, &error_abort);
+
+ if (!bdrv_init_padding(bs, *offset, *bytes, pad)) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 26 May 2021 15:26:30 +0200
-Subject: [PATCH] PVE: whitelist 'invalid' QAPI names for backwards compat
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- qapi/pragma.json | 5 ++++-
- 1 file changed, 4 insertions(+), 1 deletion(-)
-
-diff --git a/qapi/pragma.json b/qapi/pragma.json
-index 41139d8645..a581acf7d4 100644
---- a/qapi/pragma.json
-+++ b/qapi/pragma.json
-@@ -15,6 +15,7 @@
- 'device_add',
- 'device_del',
- 'expire_password',
-+ 'get_link_status',
- 'migrate_cancel',
- 'netdev_add',
- 'netdev_del',
-@@ -63,6 +64,8 @@
- 'SysEmuTarget', # query-cpu-fast, query-target
- 'UuidInfo', # query-uuid
- 'VncClientInfo', # query-vnc, query-vnc-servers, ...
-- 'X86CPURegister32' # qom-get of x86 CPU properties
-+ 'X86CPURegister32', # qom-get of x86 CPU properties
- # feature-words, filtered-features
-+ 'BlockdevOptionsPbs', # for PBS backwards compat
-+ 'BalloonInfo'
- ] } }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Mon, 7 Dec 2020 15:21:03 +0100
+Subject: [PATCH] block: add alloc-track driver
+
+Add a new filter node 'alloc-track', which seperates reads and writes to
+different children, thus allowing to put a backing image behind any
+blockdev (regardless of driver support). Since we can't detect any
+pre-allocated blocks, we can only track new writes, hence the write
+target ('file') for this node must always be empty.
+
+Intended use case is for live restoring, i.e. add a backup image as a
+block device into a VM, then put an alloc-track on the restore target
+and set the backup as backing. With this, one can use a regular
+'block-stream' to restore the image, while the VM can already run in the
+background. Copy-on-read will help make progress as the VM reads as
+well.
+
+This only worked if the target supports backing images, so up until now
+only for qcow2, with alloc-track any driver for the target can be used.
+
+If 'auto-remove' is set, alloc-track will automatically detach itself
+once the backing image is removed. It will be replaced by 'file'.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[adapt to changed function signatures]
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/alloc-track.c | 350 ++++++++++++++++++++++++++++++++++++++++++++
+ block/meson.build | 1 +
+ 2 files changed, 351 insertions(+)
+ create mode 100644 block/alloc-track.c
+
+diff --git a/block/alloc-track.c b/block/alloc-track.c
+new file mode 100644
+index 0000000000..6b50fbe537
+--- /dev/null
++++ b/block/alloc-track.c
+@@ -0,0 +1,350 @@
++/*
++ * Node to allow backing images to be applied to any node. Assumes a blank
++ * image to begin with, only new writes are tracked as allocated, thus this
++ * must never be put on a node that already contains data.
++ *
++ * Copyright (c) 2020 Proxmox Server Solutions GmbH
++ * Copyright (c) 2020 Stefan Reiter <s.reiter@proxmox.com>
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ */
++
++#include "qemu/osdep.h"
++#include "qapi/error.h"
++#include "block/block_int.h"
++#include "qapi/qmp/qdict.h"
++#include "qapi/qmp/qstring.h"
++#include "qemu/cutils.h"
++#include "qemu/option.h"
++#include "qemu/module.h"
++#include "sysemu/block-backend.h"
++
++#define TRACK_OPT_AUTO_REMOVE "auto-remove"
++
++typedef enum DropState {
++ DropNone,
++ DropRequested,
++ DropInProgress,
++} DropState;
++
++typedef struct {
++ BdrvDirtyBitmap *bitmap;
++ DropState drop_state;
++ bool auto_remove;
++} BDRVAllocTrackState;
++
++static QemuOptsList runtime_opts = {
++ .name = "alloc-track",
++ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
++ .desc = {
++ {
++ .name = TRACK_OPT_AUTO_REMOVE,
++ .type = QEMU_OPT_BOOL,
++ .help = "automatically replace this node with 'file' when 'backing'"
++ "is detached",
++ },
++ { /* end of list */ }
++ },
++};
++
++static void track_refresh_limits(BlockDriverState *bs, Error **errp)
++{
++ BlockDriverInfo bdi;
++
++ if (!bs->file) {
++ return;
++ }
++
++ /* always use alignment from underlying write device so RMW cycle for
++ * bdrv_pwritev reads data from our backing via track_co_preadv (no partial
++ * cluster allocation in 'file') */
++ bdrv_get_info(bs->file->bs, &bdi);
++ bs->bl.request_alignment = MAX(bs->file->bs->bl.request_alignment,
++ MAX(bdi.cluster_size, BDRV_SECTOR_SIZE));
++}
++
++static int track_open(BlockDriverState *bs, QDict *options, int flags,
++ Error **errp)
++{
++ BDRVAllocTrackState *s = bs->opaque;
++ QemuOpts *opts;
++ Error *local_err = NULL;
++ int ret = 0;
++
++ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
++ qemu_opts_absorb_qdict(opts, options, &local_err);
++ if (local_err) {
++ error_propagate(errp, local_err);
++ ret = -EINVAL;
++ goto fail;
++ }
++
++ s->auto_remove = qemu_opt_get_bool(opts, TRACK_OPT_AUTO_REMOVE, false);
++
++ /* open the target (write) node, backing will be attached by block layer */
++ bs->file = bdrv_open_child(NULL, options, "file", bs, &child_of_bds,
++ BDRV_CHILD_DATA | BDRV_CHILD_METADATA, false,
++ &local_err);
++ if (local_err) {
++ ret = -EINVAL;
++ error_propagate(errp, local_err);
++ goto fail;
++ }
++
++ track_refresh_limits(bs, errp);
++ uint64_t gran = bs->bl.request_alignment;
++ s->bitmap = bdrv_create_dirty_bitmap(bs->file->bs, gran, NULL, &local_err);
++ if (local_err) {
++ ret = -EIO;
++ error_propagate(errp, local_err);
++ goto fail;
++ }
++
++ s->drop_state = DropNone;
++
++fail:
++ if (ret < 0) {
++ bdrv_unref_child(bs, bs->file);
++ if (s->bitmap) {
++ bdrv_release_dirty_bitmap(s->bitmap);
++ }
++ }
++ qemu_opts_del(opts);
++ return ret;
++}
++
++static void track_close(BlockDriverState *bs)
++{
++ BDRVAllocTrackState *s = bs->opaque;
++ if (s->bitmap) {
++ bdrv_release_dirty_bitmap(s->bitmap);
++ }
++}
++
++static int64_t track_getlength(BlockDriverState *bs)
++{
++ return bdrv_getlength(bs->file->bs);
++}
++
++static int coroutine_fn track_co_preadv(BlockDriverState *bs,
++ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++{
++ BDRVAllocTrackState *s = bs->opaque;
++ QEMUIOVector local_qiov;
++ int ret;
++
++ /* 'cur_offset' is relative to 'offset', 'local_offset' to image start */
++ uint64_t cur_offset, local_offset;
++ int64_t local_bytes;
++ bool alloc;
++
++ if (offset < 0 || bytes < 0) {
++ fprintf(stderr, "unexpected negative 'offset' or 'bytes' value!\n");
++ return -EINVAL;
++ }
++
++ /* a read request can span multiple granularity-sized chunks, and can thus
++ * contain blocks with different allocation status - we could just iterate
++ * granularity-wise, but for better performance use bdrv_dirty_bitmap_next_X
++ * to find the next flip and consider everything up to that in one go */
++ for (cur_offset = 0; cur_offset < bytes; cur_offset += local_bytes) {
++ local_offset = offset + cur_offset;
++ alloc = bdrv_dirty_bitmap_get(s->bitmap, local_offset);
++ if (alloc) {
++ local_bytes = bdrv_dirty_bitmap_next_zero(s->bitmap, local_offset,
++ bytes - cur_offset);
++ } else {
++ local_bytes = bdrv_dirty_bitmap_next_dirty(s->bitmap, local_offset,
++ bytes - cur_offset);
++ }
++
++ /* _bitmap_next_X return is -1 if no end found within limit, otherwise
++ * offset of next flip (to start of image) */
++ local_bytes = local_bytes < 0 ?
++ bytes - cur_offset :
++ local_bytes - local_offset;
++
++ qemu_iovec_init_slice(&local_qiov, qiov, cur_offset, local_bytes);
++
++ if (alloc) {
++ ret = bdrv_co_preadv(bs->file, local_offset, local_bytes,
++ &local_qiov, flags);
++ } else if (bs->backing) {
++ ret = bdrv_co_preadv(bs->backing, local_offset, local_bytes,
++ &local_qiov, flags);
++ } else {
++ ret = qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
++ }
++
++ if (ret != 0) {
++ break;
++ }
++ }
++
++ return ret;
++}
++
++static int coroutine_fn track_co_pwritev(BlockDriverState *bs,
++ int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++{
++ return bdrv_co_pwritev(bs->file, offset, bytes, qiov, flags);
++}
++
++static int coroutine_fn track_co_pwrite_zeroes(BlockDriverState *bs,
++ int64_t offset, int64_t bytes, BdrvRequestFlags flags)
++{
++ return bdrv_co_pwrite_zeroes(bs->file, offset, bytes, flags);
++}
++
++static int coroutine_fn track_co_pdiscard(BlockDriverState *bs,
++ int64_t offset, int64_t bytes)
++{
++ return bdrv_co_pdiscard(bs->file, offset, bytes);
++}
++
++static coroutine_fn int track_co_flush(BlockDriverState *bs)
++{
++ return bdrv_co_flush(bs->file->bs);
++}
++
++static int coroutine_fn track_co_block_status(BlockDriverState *bs,
++ bool want_zero,
++ int64_t offset,
++ int64_t bytes,
++ int64_t *pnum,
++ int64_t *map,
++ BlockDriverState **file)
++{
++ BDRVAllocTrackState *s = bs->opaque;
++
++ bool alloc = bdrv_dirty_bitmap_get(s->bitmap, offset);
++ int64_t next_flipped;
++ if (alloc) {
++ next_flipped = bdrv_dirty_bitmap_next_zero(s->bitmap, offset, bytes);
++ } else {
++ next_flipped = bdrv_dirty_bitmap_next_dirty(s->bitmap, offset, bytes);
++ }
++
++ /* in case not the entire region has the same state, we need to set pnum to
++ * indicate for how many bytes our result is valid */
++ *pnum = next_flipped == -1 ? bytes : next_flipped - offset;
++ *map = offset;
++
++ if (alloc) {
++ *file = bs->file->bs;
++ return BDRV_BLOCK_RAW | BDRV_BLOCK_OFFSET_VALID;
++ } else if (bs->backing) {
++ *file = bs->backing->bs;
++ }
++ return 0;
++}
++
++static void track_child_perm(BlockDriverState *bs, BdrvChild *c,
++ BdrvChildRole role, BlockReopenQueue *reopen_queue,
++ uint64_t perm, uint64_t shared,
++ uint64_t *nperm, uint64_t *nshared)
++{
++ BDRVAllocTrackState *s = bs->opaque;
++
++ *nshared = BLK_PERM_ALL;
++
++ /* in case we're currently dropping ourselves, claim to not use any
++ * permissions at all - which is fine, since from this point on we will
++ * never issue a read or write anymore */
++ if (s->drop_state == DropInProgress) {
++ *nperm = 0;
++ return;
++ }
++
++ if (role & BDRV_CHILD_DATA) {
++ *nperm = perm & DEFAULT_PERM_PASSTHROUGH;
++ } else {
++ /* 'backing' is also a child of our BDS, but we don't expect it to be
++ * writeable, so we only forward 'consistent read' */
++ *nperm = perm & BLK_PERM_CONSISTENT_READ;
++ }
++}
++
++static void track_drop(void *opaque)
++{
++ BlockDriverState *bs = (BlockDriverState*)opaque;
++ BlockDriverState *file = bs->file->bs;
++ BDRVAllocTrackState *s = bs->opaque;
++
++ assert(file);
++
++ /* we rely on the fact that we're not used anywhere else, so let's wait
++ * until we're only used once - in the drive connected to the guest (and one
++ * ref is held by bdrv_ref in track_change_backing_file) */
++ if (bs->refcnt > 2) {
++ aio_bh_schedule_oneshot(qemu_get_aio_context(), track_drop, opaque);
++ return;
++ }
++ AioContext *aio_context = bdrv_get_aio_context(bs);
++ aio_context_acquire(aio_context);
++
++ bdrv_drained_begin(bs);
++
++ /* now that we're drained, we can safely set 'DropInProgress' */
++ s->drop_state = DropInProgress;
++ bdrv_child_refresh_perms(bs, bs->file, &error_abort);
++
++ bdrv_replace_node(bs, file, &error_abort);
++ bdrv_set_backing_hd(bs, NULL, &error_abort);
++ bdrv_drained_end(bs);
++ bdrv_unref(bs);
++ aio_context_release(aio_context);
++}
++
++static int track_change_backing_file(BlockDriverState *bs,
++ const char *backing_file,
++ const char *backing_fmt)
++{
++ BDRVAllocTrackState *s = bs->opaque;
++ if (s->auto_remove && s->drop_state == DropNone &&
++ backing_file == NULL && backing_fmt == NULL)
++ {
++ /* backing file has been disconnected, there's no longer any use for
++ * this node, so let's remove ourselves from the block graph - we need
++ * to schedule this for later however, since when this function is
++ * called, the blockjob modifying us is probably not done yet and has a
++ * blocker on 'bs' */
++ s->drop_state = DropRequested;
++ bdrv_ref(bs);
++ aio_bh_schedule_oneshot(qemu_get_aio_context(), track_drop, (void*)bs);
++ }
++
++ return 0;
++}
++
++static BlockDriver bdrv_alloc_track = {
++ .format_name = "alloc-track",
++ .instance_size = sizeof(BDRVAllocTrackState),
++
++ .bdrv_file_open = track_open,
++ .bdrv_close = track_close,
++ .bdrv_getlength = track_getlength,
++ .bdrv_child_perm = track_child_perm,
++ .bdrv_refresh_limits = track_refresh_limits,
++
++ .bdrv_co_pwrite_zeroes = track_co_pwrite_zeroes,
++ .bdrv_co_pwritev = track_co_pwritev,
++ .bdrv_co_preadv = track_co_preadv,
++ .bdrv_co_pdiscard = track_co_pdiscard,
++
++ .bdrv_co_flush = track_co_flush,
++ .bdrv_co_flush_to_disk = track_co_flush,
++
++ .supports_backing = true,
++
++ .bdrv_co_block_status = track_co_block_status,
++ .bdrv_change_backing_file = track_change_backing_file,
++};
++
++static void bdrv_alloc_track_init(void)
++{
++ bdrv_register(&bdrv_alloc_track);
++}
++
++block_init(bdrv_alloc_track_init);
+diff --git a/block/meson.build b/block/meson.build
+index a26a69434e..74e5f49758 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -2,6 +2,7 @@ block_ss.add(genh)
+ block_ss.add(files(
+ 'accounting.c',
+ 'aio_task.c',
++ 'alloc-track.c',
+ 'amend.c',
+ 'backup.c',
+ 'backup-dump.c',
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 26 May 2021 17:36:55 +0200
-Subject: [PATCH] PVE: savevm-async: register yank before
- migration_incoming_state_destroy
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- migration/savevm-async.c | 5 +++++
- 1 file changed, 5 insertions(+)
-
-diff --git a/migration/savevm-async.c b/migration/savevm-async.c
-index 0bc5799cf8..10ebefef06 100644
---- a/migration/savevm-async.c
-+++ b/migration/savevm-async.c
-@@ -19,6 +19,7 @@
- #include "qemu/timer.h"
- #include "qemu/main-loop.h"
- #include "qemu/rcu.h"
-+#include "qemu/yank.h"
-
- /* #define DEBUG_SAVEVM_STATE */
-
-@@ -586,6 +587,10 @@ int load_snapshot_from_blockdev(const char *filename, Error **errp)
- dirty_bitmap_mig_before_vm_start();
-
- qemu_fclose(f);
-+
-+ /* state_destroy assumes a real migration which would have added a yank */
-+ yank_register_instance(MIGRATION_YANK_INSTANCE, &error_abort);
-+
- migration_incoming_state_destroy();
- if (ret < 0) {
- error_setg_errno(errp, -ret, "Error while loading VM state");
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 26 May 2021 15:26:30 +0200
+Subject: [PATCH] PVE: whitelist 'invalid' QAPI names for backwards compat
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ qapi/pragma.json | 5 ++++-
+ 1 file changed, 4 insertions(+), 1 deletion(-)
+
+diff --git a/qapi/pragma.json b/qapi/pragma.json
+index a2358e303a..9ff5c84ffd 100644
+--- a/qapi/pragma.json
++++ b/qapi/pragma.json
+@@ -15,6 +15,7 @@
+ 'device_add',
+ 'device_del',
+ 'expire_password',
++ 'get_link_status',
+ 'migrate_cancel',
+ 'netdev_add',
+ 'netdev_del',
+@@ -64,6 +65,8 @@
+ 'SysEmuTarget', # query-cpu-fast, query-target
+ 'UuidInfo', # query-uuid
+ 'VncClientInfo', # query-vnc, query-vnc-servers, ...
+- 'X86CPURegister32' # qom-get of x86 CPU properties
++ 'X86CPURegister32', # qom-get of x86 CPU properties
+ # feature-words, filtered-features
++ 'BlockdevOptionsPbs', # for PBS backwards compat
++ 'BalloonInfo'
+ ] } }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Wed, 26 May 2021 17:36:55 +0200
+Subject: [PATCH] PVE: savevm-async: register yank before
+ migration_incoming_state_destroy
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ migration/savevm-async.c | 5 +++++
+ 1 file changed, 5 insertions(+)
+
+diff --git a/migration/savevm-async.c b/migration/savevm-async.c
+index e65a5e3482..2ed2536816 100644
+--- a/migration/savevm-async.c
++++ b/migration/savevm-async.c
+@@ -20,6 +20,7 @@
+ #include "qemu/timer.h"
+ #include "qemu/main-loop.h"
+ #include "qemu/rcu.h"
++#include "qemu/yank.h"
+
+ /* #define DEBUG_SAVEVM_STATE */
+
+@@ -514,6 +515,10 @@ int load_snapshot_from_blockdev(const char *filename, Error **errp)
+ dirty_bitmap_mig_before_vm_start();
+
+ qemu_fclose(f);
++
++ /* state_destroy assumes a real migration which would have added a yank */
++ yank_register_instance(MIGRATION_YANK_INSTANCE, &error_abort);
++
+ migration_incoming_state_destroy();
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "Error while loading VM state");
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Mon, 7 Feb 2022 14:21:01 +0100
-Subject: [PATCH] qemu-img: dd: add -l option for loading a snapshot
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- docs/tools/qemu-img.rst | 6 +++---
- qemu-img-cmds.hx | 4 ++--
- qemu-img.c | 33 +++++++++++++++++++++++++++++++--
- 3 files changed, 36 insertions(+), 7 deletions(-)
-
-diff --git a/docs/tools/qemu-img.rst b/docs/tools/qemu-img.rst
-index 33979b7430..68e9c80788 100644
---- a/docs/tools/qemu-img.rst
-+++ b/docs/tools/qemu-img.rst
-@@ -492,10 +492,10 @@ Command description:
- it doesn't need to be specified separately in this case.
-
-
--.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
-+.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [-l SNAPSHOT_PARAM] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
-
-- dd copies from *INPUT* file to *OUTPUT* file converting it from
-- *FMT* format to *OUTPUT_FMT* format.
-+ dd copies from *INPUT* file or snapshot *SNAPSHOT_PARAM* to *OUTPUT* file
-+ converting it from *FMT* format to *OUTPUT_FMT* format.
-
- The data is by default read and written using blocks of 512 bytes but can be
- modified by specifying *BLOCK_SIZE*. If count=\ *BLOCKS* is specified
-diff --git a/qemu-img-cmds.hx b/qemu-img-cmds.hx
-index b5b0bb4467..36f97e1f19 100644
---- a/qemu-img-cmds.hx
-+++ b/qemu-img-cmds.hx
-@@ -58,9 +58,9 @@ SRST
- ERST
-
- DEF("dd", img_dd,
-- "dd [--image-opts] [-U] [-f fmt] [-O output_fmt] [-n] [bs=block_size] [count=blocks] [skip=blocks] [osize=output_size] if=input of=output")
-+ "dd [--image-opts] [-U] [-f fmt] [-O output_fmt] [-n] [-l snapshot_param] [bs=block_size] [count=blocks] [skip=blocks] [osize=output_size] if=input of=output")
- SRST
--.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] [osize=OUTPUT_SIZE] if=INPUT of=OUTPUT
-+.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [-l SNAPSHOT_PARAM] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] [osize=OUTPUT_SIZE] if=INPUT of=OUTPUT
- ERST
-
- DEF("info", img_info,
-diff --git a/qemu-img.c b/qemu-img.c
-index 35c2bdc95c..6e93bbd425 100644
---- a/qemu-img.c
-+++ b/qemu-img.c
-@@ -4938,6 +4938,7 @@ static int img_dd(int argc, char **argv)
- BlockDriver *drv = NULL, *proto_drv = NULL;
- BlockBackend *blk1 = NULL, *blk2 = NULL;
- QemuOpts *opts = NULL;
-+ QemuOpts *sn_opts = NULL;
- QemuOptsList *create_opts = NULL;
- Error *local_err = NULL;
- bool image_opts = false;
-@@ -4947,6 +4948,7 @@ static int img_dd(int argc, char **argv)
- int64_t size = 0, readsize = 0;
- int64_t block_count = 0, out_pos, in_pos;
- bool force_share = false, skip_create = false;
-+ const char *snapshot_name = NULL;
- struct DdInfo dd = {
- .flags = 0,
- .count = 0,
-@@ -4984,7 +4986,7 @@ static int img_dd(int argc, char **argv)
- { 0, 0, 0, 0 }
- };
-
-- while ((c = getopt_long(argc, argv, ":hf:O:Un", long_options, NULL))) {
-+ while ((c = getopt_long(argc, argv, ":hf:O:l:Un", long_options, NULL))) {
- if (c == EOF) {
- break;
- }
-@@ -5007,6 +5009,19 @@ static int img_dd(int argc, char **argv)
- case 'n':
- skip_create = true;
- break;
-+ case 'l':
-+ if (strstart(optarg, SNAPSHOT_OPT_BASE, NULL)) {
-+ sn_opts = qemu_opts_parse_noisily(&internal_snapshot_opts,
-+ optarg, false);
-+ if (!sn_opts) {
-+ error_report("Failed in parsing snapshot param '%s'",
-+ optarg);
-+ goto out;
-+ }
-+ } else {
-+ snapshot_name = optarg;
-+ }
-+ break;
- case 'U':
- force_share = true;
- break;
-@@ -5066,11 +5081,24 @@ static int img_dd(int argc, char **argv)
- if (dd.flags & C_IF) {
- blk1 = img_open(image_opts, in.filename, fmt, 0, false, false,
- force_share);
--
- if (!blk1) {
- ret = -1;
- goto out;
- }
-+ if (sn_opts) {
-+ bdrv_snapshot_load_tmp(blk_bs(blk1),
-+ qemu_opt_get(sn_opts, SNAPSHOT_OPT_ID),
-+ qemu_opt_get(sn_opts, SNAPSHOT_OPT_NAME),
-+ &local_err);
-+ } else if (snapshot_name != NULL) {
-+ bdrv_snapshot_load_tmp_by_id_or_name(blk_bs(blk1), snapshot_name,
-+ &local_err);
-+ }
-+ if (local_err) {
-+ error_reportf_err(local_err, "Failed to load snapshot: ");
-+ ret = -1;
-+ goto out;
-+ }
- }
-
- if (dd.flags & C_OSIZE) {
-@@ -5219,6 +5247,7 @@ static int img_dd(int argc, char **argv)
- out:
- g_free(arg);
- qemu_opts_del(opts);
-+ qemu_opts_del(sn_opts);
- qemu_opts_free(create_opts);
- blk_unref(blk1);
- blk_unref(blk2);
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Mon, 7 Feb 2022 14:21:01 +0100
+Subject: [PATCH] qemu-img: dd: add -l option for loading a snapshot
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ docs/tools/qemu-img.rst | 6 +++---
+ qemu-img-cmds.hx | 4 ++--
+ qemu-img.c | 33 +++++++++++++++++++++++++++++++--
+ 3 files changed, 36 insertions(+), 7 deletions(-)
+
+diff --git a/docs/tools/qemu-img.rst b/docs/tools/qemu-img.rst
+index 699229eef6..4189ced8bc 100644
+--- a/docs/tools/qemu-img.rst
++++ b/docs/tools/qemu-img.rst
+@@ -492,10 +492,10 @@ Command description:
+ it doesn't need to be specified separately in this case.
+
+
+-.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
++.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [-l SNAPSHOT_PARAM] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
+
+- dd copies from *INPUT* file to *OUTPUT* file converting it from
+- *FMT* format to *OUTPUT_FMT* format.
++ dd copies from *INPUT* file or snapshot *SNAPSHOT_PARAM* to *OUTPUT* file
++ converting it from *FMT* format to *OUTPUT_FMT* format.
+
+ The data is by default read and written using blocks of 512 bytes but can be
+ modified by specifying *BLOCK_SIZE*. If count=\ *BLOCKS* is specified
+diff --git a/qemu-img-cmds.hx b/qemu-img-cmds.hx
+index b5b0bb4467..36f97e1f19 100644
+--- a/qemu-img-cmds.hx
++++ b/qemu-img-cmds.hx
+@@ -58,9 +58,9 @@ SRST
+ ERST
+
+ DEF("dd", img_dd,
+- "dd [--image-opts] [-U] [-f fmt] [-O output_fmt] [-n] [bs=block_size] [count=blocks] [skip=blocks] [osize=output_size] if=input of=output")
++ "dd [--image-opts] [-U] [-f fmt] [-O output_fmt] [-n] [-l snapshot_param] [bs=block_size] [count=blocks] [skip=blocks] [osize=output_size] if=input of=output")
+ SRST
+-.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] [osize=OUTPUT_SIZE] if=INPUT of=OUTPUT
++.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [-l SNAPSHOT_PARAM] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] [osize=OUTPUT_SIZE] if=INPUT of=OUTPUT
+ ERST
+
+ DEF("info", img_info,
+diff --git a/qemu-img.c b/qemu-img.c
+index c6b4a5567d..041c203fc3 100644
+--- a/qemu-img.c
++++ b/qemu-img.c
+@@ -4943,6 +4943,7 @@ static int img_dd(int argc, char **argv)
+ BlockDriver *drv = NULL, *proto_drv = NULL;
+ BlockBackend *blk1 = NULL, *blk2 = NULL;
+ QemuOpts *opts = NULL;
++ QemuOpts *sn_opts = NULL;
+ QemuOptsList *create_opts = NULL;
+ Error *local_err = NULL;
+ bool image_opts = false;
+@@ -4952,6 +4953,7 @@ static int img_dd(int argc, char **argv)
+ int64_t size = 0, readsize = 0;
+ int64_t block_count = 0, out_pos, in_pos;
+ bool force_share = false, skip_create = false;
++ const char *snapshot_name = NULL;
+ struct DdInfo dd = {
+ .flags = 0,
+ .count = 0,
+@@ -4989,7 +4991,7 @@ static int img_dd(int argc, char **argv)
+ { 0, 0, 0, 0 }
+ };
+
+- while ((c = getopt_long(argc, argv, ":hf:O:Un", long_options, NULL))) {
++ while ((c = getopt_long(argc, argv, ":hf:O:l:Un", long_options, NULL))) {
+ if (c == EOF) {
+ break;
+ }
+@@ -5012,6 +5014,19 @@ static int img_dd(int argc, char **argv)
+ case 'n':
+ skip_create = true;
+ break;
++ case 'l':
++ if (strstart(optarg, SNAPSHOT_OPT_BASE, NULL)) {
++ sn_opts = qemu_opts_parse_noisily(&internal_snapshot_opts,
++ optarg, false);
++ if (!sn_opts) {
++ error_report("Failed in parsing snapshot param '%s'",
++ optarg);
++ goto out;
++ }
++ } else {
++ snapshot_name = optarg;
++ }
++ break;
+ case 'U':
+ force_share = true;
+ break;
+@@ -5071,11 +5086,24 @@ static int img_dd(int argc, char **argv)
+ if (dd.flags & C_IF) {
+ blk1 = img_open(image_opts, in.filename, fmt, 0, false, false,
+ force_share);
+-
+ if (!blk1) {
+ ret = -1;
+ goto out;
+ }
++ if (sn_opts) {
++ bdrv_snapshot_load_tmp(blk_bs(blk1),
++ qemu_opt_get(sn_opts, SNAPSHOT_OPT_ID),
++ qemu_opt_get(sn_opts, SNAPSHOT_OPT_NAME),
++ &local_err);
++ } else if (snapshot_name != NULL) {
++ bdrv_snapshot_load_tmp_by_id_or_name(blk_bs(blk1), snapshot_name,
++ &local_err);
++ }
++ if (local_err) {
++ error_reportf_err(local_err, "Failed to load snapshot: ");
++ ret = -1;
++ goto out;
++ }
+ }
+
+ if (dd.flags & C_OSIZE) {
+@@ -5230,6 +5258,7 @@ static int img_dd(int argc, char **argv)
+ out:
+ g_free(arg);
+ qemu_opts_del(opts);
++ qemu_opts_del(sn_opts);
+ qemu_opts_free(create_opts);
+ blk_unref(blk1);
+ blk_unref(blk2);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Thu, 21 Apr 2022 13:26:48 +0200
-Subject: [PATCH] vma: allow partial restore
-
-Introduce a new map line for skipping a certain drive, of the form
-skip=drive-scsi0
-
-Since in PVE, most archives are compressed and piped to vma for
-restore, it's not easily possible to skip reads.
-
-For the reader, a new skip flag for VmaRestoreState is added and the
-target is allowed to be NULL if skip is specified when registering. If
-the skip flag is set, no writes will be made as well as no check for
-duplicate clusters. Therefore, the flag is not set for verify.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Acked-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- vma-reader.c | 64 ++++++++++++---------
- vma.c | 157 +++++++++++++++++++++++++++++----------------------
- vma.h | 2 +-
- 3 files changed, 126 insertions(+), 97 deletions(-)
-
-diff --git a/vma-reader.c b/vma-reader.c
-index 4f4ee2b47b..844d95a5ba 100644
---- a/vma-reader.c
-+++ b/vma-reader.c
-@@ -29,6 +29,7 @@ typedef struct VmaRestoreState {
- bool write_zeroes;
- unsigned long *bitmap;
- int bitmap_size;
-+ bool skip;
- } VmaRestoreState;
-
- struct VmaReader {
-@@ -426,13 +427,14 @@ VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id)
- }
-
- static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
-- BlockBackend *target, bool write_zeroes)
-+ BlockBackend *target, bool write_zeroes, bool skip)
- {
- assert(vmar);
- assert(dev_id);
-
- vmar->rstate[dev_id].target = target;
- vmar->rstate[dev_id].write_zeroes = write_zeroes;
-+ vmar->rstate[dev_id].skip = skip;
-
- int64_t size = vmar->devinfo[dev_id].size;
-
-@@ -447,28 +449,30 @@ static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
- }
-
- int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id, BlockBackend *target,
-- bool write_zeroes, Error **errp)
-+ bool write_zeroes, bool skip, Error **errp)
- {
- assert(vmar);
-- assert(target != NULL);
-+ assert(target != NULL || skip);
- assert(dev_id);
-- assert(vmar->rstate[dev_id].target == NULL);
--
-- int64_t size = blk_getlength(target);
-- int64_t size_diff = size - vmar->devinfo[dev_id].size;
--
-- /* storage types can have different size restrictions, so it
-- * is not always possible to create an image with exact size.
-- * So we tolerate a size difference up to 4MB.
-- */
-- if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
-- error_setg(errp, "vma_reader_register_bs for stream %s failed - "
-- "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
-- size, vmar->devinfo[dev_id].size);
-- return -1;
-+ assert(vmar->rstate[dev_id].target == NULL && !vmar->rstate[dev_id].skip);
-+
-+ if (target != NULL) {
-+ int64_t size = blk_getlength(target);
-+ int64_t size_diff = size - vmar->devinfo[dev_id].size;
-+
-+ /* storage types can have different size restrictions, so it
-+ * is not always possible to create an image with exact size.
-+ * So we tolerate a size difference up to 4MB.
-+ */
-+ if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
-+ error_setg(errp, "vma_reader_register_bs for stream %s failed - "
-+ "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
-+ size, vmar->devinfo[dev_id].size);
-+ return -1;
-+ }
- }
-
-- allocate_rstate(vmar, dev_id, target, write_zeroes);
-+ allocate_rstate(vmar, dev_id, target, write_zeroes, skip);
-
- return 0;
- }
-@@ -561,19 +565,23 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
- VmaRestoreState *rstate = &vmar->rstate[dev_id];
- BlockBackend *target = NULL;
-
-+ bool skip = rstate->skip;
-+
- if (dev_id != vmar->vmstate_stream) {
- target = rstate->target;
-- if (!verify && !target) {
-+ if (!verify && !target && !skip) {
- error_setg(errp, "got wrong dev id %d", dev_id);
- return -1;
- }
-
-- if (vma_reader_get_bitmap(rstate, cluster_num)) {
-- error_setg(errp, "found duplicated cluster %zd for stream %s",
-- cluster_num, vmar->devinfo[dev_id].devname);
-- return -1;
-+ if (!skip) {
-+ if (vma_reader_get_bitmap(rstate, cluster_num)) {
-+ error_setg(errp, "found duplicated cluster %zd for stream %s",
-+ cluster_num, vmar->devinfo[dev_id].devname);
-+ return -1;
-+ }
-+ vma_reader_set_bitmap(rstate, cluster_num, 1);
- }
-- vma_reader_set_bitmap(rstate, cluster_num, 1);
-
- max_sector = vmar->devinfo[dev_id].size/BDRV_SECTOR_SIZE;
- } else {
-@@ -619,7 +627,7 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
- return -1;
- }
-
-- if (!verify) {
-+ if (!verify && !skip) {
- int nb_sectors = end_sector - sector_num;
- if (restore_write_data(vmar, dev_id, target, vmstate_fd,
- buf + start, sector_num, nb_sectors,
-@@ -655,7 +663,7 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
- return -1;
- }
-
-- if (!verify) {
-+ if (!verify && !skip) {
- int nb_sectors = end_sector - sector_num;
- if (restore_write_data(vmar, dev_id, target, vmstate_fd,
- buf + start, sector_num,
-@@ -680,7 +688,7 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
- vmar->partial_zero_cluster_data += zero_size;
- }
-
-- if (rstate->write_zeroes && !verify) {
-+ if (rstate->write_zeroes && !verify && !skip) {
- if (restore_write_data(vmar, dev_id, target, vmstate_fd,
- zero_vma_block, sector_num,
- nb_sectors, errp) < 0) {
-@@ -851,7 +859,7 @@ int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp)
-
- for (dev_id = 1; dev_id < 255; dev_id++) {
- if (vma_reader_get_device_info(vmar, dev_id)) {
-- allocate_rstate(vmar, dev_id, NULL, false);
-+ allocate_rstate(vmar, dev_id, NULL, false, false);
- }
- }
-
-diff --git a/vma.c b/vma.c
-index 91612d50a2..e1a750ff50 100644
---- a/vma.c
-+++ b/vma.c
-@@ -139,6 +139,7 @@ typedef struct RestoreMap {
- char *throttling_group;
- char *cache;
- bool write_zero;
-+ bool skip;
- } RestoreMap;
-
- static bool try_parse_option(char **line, const char *optname, char **out, const char *inbuf) {
-@@ -246,47 +247,61 @@ static int extract_content(int argc, char **argv)
- char *bps = NULL;
- char *group = NULL;
- char *cache = NULL;
-+ char *devname = NULL;
-+ bool skip = false;
-+ uint64_t bps_value = 0;
-+ const char *path = NULL;
-+ bool write_zero = true;
-+
- if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
- break;
- }
- int len = strlen(line);
- if (line[len - 1] == '\n') {
- line[len - 1] = '\0';
-- if (len == 1) {
-+ len = len - 1;
-+ if (len == 0) {
- break;
- }
- }
-
-- while (1) {
-- if (!try_parse_option(&line, "format", &format, inbuf) &&
-- !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
-- !try_parse_option(&line, "throttling.group", &group, inbuf) &&
-- !try_parse_option(&line, "cache", &cache, inbuf))
-- {
-- break;
-+ if (strncmp(line, "skip", 4) == 0) {
-+ if (len < 6 || line[4] != '=') {
-+ g_error("read map failed - option 'skip' has no value ('%s')",
-+ inbuf);
-+ } else {
-+ devname = line + 5;
-+ skip = true;
-+ }
-+ } else {
-+ while (1) {
-+ if (!try_parse_option(&line, "format", &format, inbuf) &&
-+ !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
-+ !try_parse_option(&line, "throttling.group", &group, inbuf) &&
-+ !try_parse_option(&line, "cache", &cache, inbuf))
-+ {
-+ break;
-+ }
- }
-- }
-
-- uint64_t bps_value = 0;
-- if (bps) {
-- bps_value = verify_u64(bps);
-- g_free(bps);
-- }
-+ if (bps) {
-+ bps_value = verify_u64(bps);
-+ g_free(bps);
-+ }
-
-- const char *path;
-- bool write_zero;
-- if (line[0] == '0' && line[1] == ':') {
-- path = line + 2;
-- write_zero = false;
-- } else if (line[0] == '1' && line[1] == ':') {
-- path = line + 2;
-- write_zero = true;
-- } else {
-- g_error("read map failed - parse error ('%s')", inbuf);
-+ if (line[0] == '0' && line[1] == ':') {
-+ path = line + 2;
-+ write_zero = false;
-+ } else if (line[0] == '1' && line[1] == ':') {
-+ path = line + 2;
-+ write_zero = true;
-+ } else {
-+ g_error("read map failed - parse error ('%s')", inbuf);
-+ }
-+
-+ path = extract_devname(path, &devname, -1);
- }
-
-- char *devname = NULL;
-- path = extract_devname(path, &devname, -1);
- if (!devname) {
- g_error("read map failed - no dev name specified ('%s')",
- inbuf);
-@@ -300,6 +315,7 @@ static int extract_content(int argc, char **argv)
- map->throttling_group = group;
- map->cache = cache;
- map->write_zero = write_zero;
-+ map->skip = skip;
-
- g_hash_table_insert(devmap, map->devname, map);
-
-@@ -329,6 +345,7 @@ static int extract_content(int argc, char **argv)
- const char *cache = NULL;
- int flags = BDRV_O_RDWR;
- bool write_zero = true;
-+ bool skip = false;
-
- BlockBackend *blk = NULL;
-
-@@ -344,6 +361,7 @@ static int extract_content(int argc, char **argv)
- throttling_group = map->throttling_group;
- cache = map->cache;
- write_zero = map->write_zero;
-+ skip = map->skip;
- } else {
- devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
- dirname, di->devname);
-@@ -362,57 +380,60 @@ static int extract_content(int argc, char **argv)
- write_zero = false;
- }
-
-- size_t devlen = strlen(devfn);
-- QDict *options = NULL;
-- bool writethrough;
-- if (format) {
-- /* explicit format from commandline */
-- options = qdict_new();
-- qdict_put_str(options, "driver", format);
-- } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
-- strncmp(devfn, "/dev/", 5) == 0)
-- {
-- /* This part is now deprecated for PVE as well (just as qemu
-- * deprecated not specifying an explicit raw format, too.
-- */
-- /* explicit raw format */
-- options = qdict_new();
-- qdict_put_str(options, "driver", "raw");
-- }
-- if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
-- g_error("invalid cache option: %s\n", cache);
-- }
-+ if (!skip) {
-+ size_t devlen = strlen(devfn);
-+ QDict *options = NULL;
-+ bool writethrough;
-+ if (format) {
-+ /* explicit format from commandline */
-+ options = qdict_new();
-+ qdict_put_str(options, "driver", format);
-+ } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
-+ strncmp(devfn, "/dev/", 5) == 0)
-+ {
-+ /* This part is now deprecated for PVE as well (just as qemu
-+ * deprecated not specifying an explicit raw format, too.
-+ */
-+ /* explicit raw format */
-+ options = qdict_new();
-+ qdict_put_str(options, "driver", "raw");
-+ }
-
-- if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
-- g_error("can't open file %s - %s", devfn,
-- error_get_pretty(errp));
-- }
-+ if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
-+ g_error("invalid cache option: %s\n", cache);
-+ }
-
-- if (cache) {
-- blk_set_enable_write_cache(blk, !writethrough);
-- }
-+ if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
-+ g_error("can't open file %s - %s", devfn,
-+ error_get_pretty(errp));
-+ }
-
-- if (throttling_group) {
-- blk_io_limits_enable(blk, throttling_group);
-- }
-+ if (cache) {
-+ blk_set_enable_write_cache(blk, !writethrough);
-+ }
-
-- if (throttling_bps) {
-- if (!throttling_group) {
-- blk_io_limits_enable(blk, devfn);
-+ if (throttling_group) {
-+ blk_io_limits_enable(blk, throttling_group);
- }
-
-- ThrottleConfig cfg;
-- throttle_config_init(&cfg);
-- cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
-- Error *err = NULL;
-- if (!throttle_is_valid(&cfg, &err)) {
-- error_report_err(err);
-- g_error("failed to apply throttling");
-+ if (throttling_bps) {
-+ if (!throttling_group) {
-+ blk_io_limits_enable(blk, devfn);
-+ }
-+
-+ ThrottleConfig cfg;
-+ throttle_config_init(&cfg);
-+ cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
-+ Error *err = NULL;
-+ if (!throttle_is_valid(&cfg, &err)) {
-+ error_report_err(err);
-+ g_error("failed to apply throttling");
-+ }
-+ blk_set_io_limits(blk, &cfg);
- }
-- blk_set_io_limits(blk, &cfg);
- }
-
-- if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
-+ if (vma_reader_register_bs(vmar, i, blk, write_zero, skip, &errp) < 0) {
- g_error("%s", error_get_pretty(errp));
- }
-
-diff --git a/vma.h b/vma.h
-index c895c97f6d..1b62859165 100644
---- a/vma.h
-+++ b/vma.h
-@@ -142,7 +142,7 @@ GList *vma_reader_get_config_data(VmaReader *vmar);
- VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id);
- int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id,
- BlockBackend *target, bool write_zeroes,
-- Error **errp);
-+ bool skip, Error **errp);
- int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
- Error **errp);
- int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Tue, 26 Apr 2022 16:06:28 +0200
-Subject: [PATCH] pbs: namespace support
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- block/monitor/block-hmp-cmds.c | 1 +
- block/pbs.c | 25 +++++++++++++++++++++----
- pbs-restore.c | 19 ++++++++++++++++---
- pve-backup.c | 6 +++++-
- qapi/block-core.json | 5 ++++-
- 5 files changed, 47 insertions(+), 9 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index 4c799f00d9..0502f42be6 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1041,6 +1041,7 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
- false, NULL, // PBS key_password
- false, NULL, // PBS master_keyfile
- false, NULL, // PBS fingerprint
-+ false, NULL, // PBS backup-ns
- false, NULL, // PBS backup-id
- false, 0, // PBS backup-time
- false, false, // PBS use-dirty-bitmap
-diff --git a/block/pbs.c b/block/pbs.c
-index c5eb4d5bad..7471e2ef9d 100644
---- a/block/pbs.c
-+++ b/block/pbs.c
-@@ -14,6 +14,7 @@
- #include <proxmox-backup-qemu.h>
-
- #define PBS_OPT_REPOSITORY "repository"
-+#define PBS_OPT_NAMESPACE "namespace"
- #define PBS_OPT_SNAPSHOT "snapshot"
- #define PBS_OPT_ARCHIVE "archive"
- #define PBS_OPT_KEYFILE "keyfile"
-@@ -27,6 +28,7 @@ typedef struct {
- int64_t length;
-
- char *repository;
-+ char *namespace;
- char *snapshot;
- char *archive;
- } BDRVPBSState;
-@@ -40,6 +42,11 @@ static QemuOptsList runtime_opts = {
- .type = QEMU_OPT_STRING,
- .help = "The server address and repository to connect to.",
- },
-+ {
-+ .name = PBS_OPT_NAMESPACE,
-+ .type = QEMU_OPT_STRING,
-+ .help = "Optional: The snapshot's namespace.",
-+ },
- {
- .name = PBS_OPT_SNAPSHOT,
- .type = QEMU_OPT_STRING,
-@@ -76,7 +83,7 @@ static QemuOptsList runtime_opts = {
-
-
- // filename format:
--// pbs:repository=<repo>,snapshot=<snap>,password=<pw>,key_password=<kpw>,fingerprint=<fp>,archive=<archive>
-+// pbs:repository=<repo>,namespace=<ns>,snapshot=<snap>,password=<pw>,key_password=<kpw>,fingerprint=<fp>,archive=<archive>
- static void pbs_parse_filename(const char *filename, QDict *options,
- Error **errp)
- {
-@@ -112,6 +119,7 @@ static int pbs_open(BlockDriverState *bs, QDict *options, int flags,
- s->archive = g_strdup(qemu_opt_get(opts, PBS_OPT_ARCHIVE));
- const char *keyfile = qemu_opt_get(opts, PBS_OPT_KEYFILE);
- const char *password = qemu_opt_get(opts, PBS_OPT_PASSWORD);
-+ const char *namespace = qemu_opt_get(opts, PBS_OPT_NAMESPACE);
- const char *fingerprint = qemu_opt_get(opts, PBS_OPT_FINGERPRINT);
- const char *key_password = qemu_opt_get(opts, PBS_OPT_ENCRYPTION_PASSWORD);
-
-@@ -124,9 +132,12 @@ static int pbs_open(BlockDriverState *bs, QDict *options, int flags,
- if (!key_password) {
- key_password = getenv("PBS_ENCRYPTION_PASSWORD");
- }
-+ if (namespace) {
-+ s->namespace = g_strdup(namespace);
-+ }
-
- /* connect to PBS server in read mode */
-- s->conn = proxmox_restore_new(s->repository, s->snapshot, password,
-+ s->conn = proxmox_restore_new_ns(s->repository, s->snapshot, s->namespace, password,
- keyfile, key_password, fingerprint, &pbs_error);
-
- /* invalidates qemu_opt_get char pointers from above */
-@@ -171,6 +182,7 @@ static int pbs_file_open(BlockDriverState *bs, QDict *options, int flags,
- static void pbs_close(BlockDriverState *bs) {
- BDRVPBSState *s = bs->opaque;
- g_free(s->repository);
-+ g_free(s->namespace);
- g_free(s->snapshot);
- g_free(s->archive);
- proxmox_restore_disconnect(s->conn);
-@@ -252,8 +264,13 @@ static coroutine_fn int pbs_co_pwritev(BlockDriverState *bs,
- static void pbs_refresh_filename(BlockDriverState *bs)
- {
- BDRVPBSState *s = bs->opaque;
-- snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s(%s)",
-- s->repository, s->snapshot, s->archive);
-+ if (s->namespace) {
-+ snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s:%s(%s)",
-+ s->repository, s->namespace, s->snapshot, s->archive);
-+ } else {
-+ snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s(%s)",
-+ s->repository, s->snapshot, s->archive);
-+ }
- }
-
- static const char *const pbs_strong_runtime_opts[] = {
-diff --git a/pbs-restore.c b/pbs-restore.c
-index 4d3f925a1b..62042bdd93 100644
---- a/pbs-restore.c
-+++ b/pbs-restore.c
-@@ -30,7 +30,7 @@
- static void help(void)
- {
- const char *help_msg =
-- "usage: pbs-restore [--repository <repo>] snapshot archive-name target [command options]\n"
-+ "usage: pbs-restore [--repository <repo>] [--ns namespace] snapshot archive-name target [command options]\n"
- ;
-
- printf("%s", help_msg);
-@@ -78,6 +78,7 @@ int main(int argc, char **argv)
- Error *main_loop_err = NULL;
- const char *format = "raw";
- const char *repository = NULL;
-+ const char *backup_ns = NULL;
- const char *keyfile = NULL;
- int verbose = false;
- bool skip_zero = false;
-@@ -91,6 +92,7 @@ int main(int argc, char **argv)
- {"verbose", no_argument, 0, 'v'},
- {"format", required_argument, 0, 'f'},
- {"repository", required_argument, 0, 'r'},
-+ {"ns", required_argument, 0, 'n'},
- {"keyfile", required_argument, 0, 'k'},
- {0, 0, 0, 0}
- };
-@@ -111,6 +113,9 @@ int main(int argc, char **argv)
- case 'r':
- repository = g_strdup(argv[optind - 1]);
- break;
-+ case 'n':
-+ backup_ns = g_strdup(argv[optind - 1]);
-+ break;
- case 'k':
- keyfile = g_strdup(argv[optind - 1]);
- break;
-@@ -161,8 +166,16 @@ int main(int argc, char **argv)
- fprintf(stderr, "connecting to repository '%s'\n", repository);
- }
- char *pbs_error = NULL;
-- ProxmoxRestoreHandle *conn = proxmox_restore_new(
-- repository, snapshot, password, keyfile, key_password, fingerprint, &pbs_error);
-+ ProxmoxRestoreHandle *conn = proxmox_restore_new_ns(
-+ repository,
-+ snapshot,
-+ backup_ns,
-+ password,
-+ keyfile,
-+ key_password,
-+ fingerprint,
-+ &pbs_error
-+ );
- if (conn == NULL) {
- fprintf(stderr, "restore failed: %s\n", pbs_error);
- return -1;
-diff --git a/pve-backup.c b/pve-backup.c
-index 9f6c04a512..f6a5f8c785 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -10,6 +10,8 @@
- #include "qapi/qmp/qerror.h"
- #include "qemu/cutils.h"
-
-+#include <proxmox-backup-qemu.h>
-+
- /* PVE backup state and related function */
-
- /*
-@@ -535,6 +537,7 @@ UuidInfo coroutine_fn *qmp_backup(
- bool has_key_password, const char *key_password,
- bool has_master_keyfile, const char *master_keyfile,
- bool has_fingerprint, const char *fingerprint,
-+ bool has_backup_ns, const char *backup_ns,
- bool has_backup_id, const char *backup_id,
- bool has_backup_time, int64_t backup_time,
- bool has_use_dirty_bitmap, bool use_dirty_bitmap,
-@@ -674,8 +677,9 @@ UuidInfo coroutine_fn *qmp_backup(
- firewall_name = "fw.conf";
-
- char *pbs_err = NULL;
-- pbs = proxmox_backup_new(
-+ pbs = proxmox_backup_new_ns(
- backup_file,
-+ has_backup_ns ? backup_ns : NULL,
- backup_id,
- backup_time,
- dump_cb_block_size,
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 13e03ca154..89875f309c 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -821,6 +821,8 @@
- #
- # @fingerprint: server cert fingerprint (optional for format 'pbs')
- #
-+# @backup-ns: backup namespace (required for format 'pbs')
-+#
- # @backup-id: backup ID (required for format 'pbs')
- #
- # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
-@@ -840,6 +842,7 @@
- '*key-password': 'str',
- '*master-keyfile': 'str',
- '*fingerprint': 'str',
-+ '*backup-ns': 'str',
- '*backup-id': 'str',
- '*backup-time': 'int',
- '*use-dirty-bitmap': 'bool',
-@@ -3256,7 +3259,7 @@
- { 'struct': 'BlockdevOptionsPbs',
- 'data': { 'repository': 'str', 'snapshot': 'str', 'archive': 'str',
- '*keyfile': 'str', '*password': 'str', '*fingerprint': 'str',
-- '*key_password': 'str' } }
-+ '*key_password': 'str', '*namespace': 'str' } }
-
- ##
- # @BlockdevOptionsNVMe:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Thu, 21 Apr 2022 13:26:48 +0200
+Subject: [PATCH] vma: allow partial restore
+
+Introduce a new map line for skipping a certain drive, of the form
+skip=drive-scsi0
+
+Since in PVE, most archives are compressed and piped to vma for
+restore, it's not easily possible to skip reads.
+
+For the reader, a new skip flag for VmaRestoreState is added and the
+target is allowed to be NULL if skip is specified when registering. If
+the skip flag is set, no writes will be made as well as no check for
+duplicate clusters. Therefore, the flag is not set for verify.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Acked-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ vma-reader.c | 64 ++++++++++++---------
+ vma.c | 157 +++++++++++++++++++++++++++++----------------------
+ vma.h | 2 +-
+ 3 files changed, 126 insertions(+), 97 deletions(-)
+
+diff --git a/vma-reader.c b/vma-reader.c
+index e65f1e8415..81a891c6b1 100644
+--- a/vma-reader.c
++++ b/vma-reader.c
+@@ -28,6 +28,7 @@ typedef struct VmaRestoreState {
+ bool write_zeroes;
+ unsigned long *bitmap;
+ int bitmap_size;
++ bool skip;
+ } VmaRestoreState;
+
+ struct VmaReader {
+@@ -425,13 +426,14 @@ VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id)
+ }
+
+ static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
+- BlockBackend *target, bool write_zeroes)
++ BlockBackend *target, bool write_zeroes, bool skip)
+ {
+ assert(vmar);
+ assert(dev_id);
+
+ vmar->rstate[dev_id].target = target;
+ vmar->rstate[dev_id].write_zeroes = write_zeroes;
++ vmar->rstate[dev_id].skip = skip;
+
+ int64_t size = vmar->devinfo[dev_id].size;
+
+@@ -446,28 +448,30 @@ static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
+ }
+
+ int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id, BlockBackend *target,
+- bool write_zeroes, Error **errp)
++ bool write_zeroes, bool skip, Error **errp)
+ {
+ assert(vmar);
+- assert(target != NULL);
++ assert(target != NULL || skip);
+ assert(dev_id);
+- assert(vmar->rstate[dev_id].target == NULL);
+-
+- int64_t size = blk_getlength(target);
+- int64_t size_diff = size - vmar->devinfo[dev_id].size;
+-
+- /* storage types can have different size restrictions, so it
+- * is not always possible to create an image with exact size.
+- * So we tolerate a size difference up to 4MB.
+- */
+- if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
+- error_setg(errp, "vma_reader_register_bs for stream %s failed - "
+- "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
+- size, vmar->devinfo[dev_id].size);
+- return -1;
++ assert(vmar->rstate[dev_id].target == NULL && !vmar->rstate[dev_id].skip);
++
++ if (target != NULL) {
++ int64_t size = blk_getlength(target);
++ int64_t size_diff = size - vmar->devinfo[dev_id].size;
++
++ /* storage types can have different size restrictions, so it
++ * is not always possible to create an image with exact size.
++ * So we tolerate a size difference up to 4MB.
++ */
++ if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
++ error_setg(errp, "vma_reader_register_bs for stream %s failed - "
++ "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
++ size, vmar->devinfo[dev_id].size);
++ return -1;
++ }
+ }
+
+- allocate_rstate(vmar, dev_id, target, write_zeroes);
++ allocate_rstate(vmar, dev_id, target, write_zeroes, skip);
+
+ return 0;
+ }
+@@ -560,19 +564,23 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
+ VmaRestoreState *rstate = &vmar->rstate[dev_id];
+ BlockBackend *target = NULL;
+
++ bool skip = rstate->skip;
++
+ if (dev_id != vmar->vmstate_stream) {
+ target = rstate->target;
+- if (!verify && !target) {
++ if (!verify && !target && !skip) {
+ error_setg(errp, "got wrong dev id %d", dev_id);
+ return -1;
+ }
+
+- if (vma_reader_get_bitmap(rstate, cluster_num)) {
+- error_setg(errp, "found duplicated cluster %zd for stream %s",
+- cluster_num, vmar->devinfo[dev_id].devname);
+- return -1;
++ if (!skip) {
++ if (vma_reader_get_bitmap(rstate, cluster_num)) {
++ error_setg(errp, "found duplicated cluster %zd for stream %s",
++ cluster_num, vmar->devinfo[dev_id].devname);
++ return -1;
++ }
++ vma_reader_set_bitmap(rstate, cluster_num, 1);
+ }
+- vma_reader_set_bitmap(rstate, cluster_num, 1);
+
+ max_sector = vmar->devinfo[dev_id].size/BDRV_SECTOR_SIZE;
+ } else {
+@@ -618,7 +626,7 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
+ return -1;
+ }
+
+- if (!verify) {
++ if (!verify && !skip) {
+ int nb_sectors = end_sector - sector_num;
+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
+ buf + start, sector_num, nb_sectors,
+@@ -654,7 +662,7 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
+ return -1;
+ }
+
+- if (!verify) {
++ if (!verify && !skip) {
+ int nb_sectors = end_sector - sector_num;
+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
+ buf + start, sector_num,
+@@ -679,7 +687,7 @@ static int restore_extent(VmaReader *vmar, unsigned char *buf,
+ vmar->partial_zero_cluster_data += zero_size;
+ }
+
+- if (rstate->write_zeroes && !verify) {
++ if (rstate->write_zeroes && !verify && !skip) {
+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
+ zero_vma_block, sector_num,
+ nb_sectors, errp) < 0) {
+@@ -850,7 +858,7 @@ int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp)
+
+ for (dev_id = 1; dev_id < 255; dev_id++) {
+ if (vma_reader_get_device_info(vmar, dev_id)) {
+- allocate_rstate(vmar, dev_id, NULL, false);
++ allocate_rstate(vmar, dev_id, NULL, false, false);
+ }
+ }
+
+diff --git a/vma.c b/vma.c
+index e8dffb43e0..e6e9ffc7fe 100644
+--- a/vma.c
++++ b/vma.c
+@@ -138,6 +138,7 @@ typedef struct RestoreMap {
+ char *throttling_group;
+ char *cache;
+ bool write_zero;
++ bool skip;
+ } RestoreMap;
+
+ static bool try_parse_option(char **line, const char *optname, char **out, const char *inbuf) {
+@@ -245,47 +246,61 @@ static int extract_content(int argc, char **argv)
+ char *bps = NULL;
+ char *group = NULL;
+ char *cache = NULL;
++ char *devname = NULL;
++ bool skip = false;
++ uint64_t bps_value = 0;
++ const char *path = NULL;
++ bool write_zero = true;
++
+ if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
+ break;
+ }
+ int len = strlen(line);
+ if (line[len - 1] == '\n') {
+ line[len - 1] = '\0';
+- if (len == 1) {
++ len = len - 1;
++ if (len == 0) {
+ break;
+ }
+ }
+
+- while (1) {
+- if (!try_parse_option(&line, "format", &format, inbuf) &&
+- !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
+- !try_parse_option(&line, "throttling.group", &group, inbuf) &&
+- !try_parse_option(&line, "cache", &cache, inbuf))
+- {
+- break;
++ if (strncmp(line, "skip", 4) == 0) {
++ if (len < 6 || line[4] != '=') {
++ g_error("read map failed - option 'skip' has no value ('%s')",
++ inbuf);
++ } else {
++ devname = line + 5;
++ skip = true;
++ }
++ } else {
++ while (1) {
++ if (!try_parse_option(&line, "format", &format, inbuf) &&
++ !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
++ !try_parse_option(&line, "throttling.group", &group, inbuf) &&
++ !try_parse_option(&line, "cache", &cache, inbuf))
++ {
++ break;
++ }
+ }
+- }
+
+- uint64_t bps_value = 0;
+- if (bps) {
+- bps_value = verify_u64(bps);
+- g_free(bps);
+- }
++ if (bps) {
++ bps_value = verify_u64(bps);
++ g_free(bps);
++ }
+
+- const char *path;
+- bool write_zero;
+- if (line[0] == '0' && line[1] == ':') {
+- path = line + 2;
+- write_zero = false;
+- } else if (line[0] == '1' && line[1] == ':') {
+- path = line + 2;
+- write_zero = true;
+- } else {
+- g_error("read map failed - parse error ('%s')", inbuf);
++ if (line[0] == '0' && line[1] == ':') {
++ path = line + 2;
++ write_zero = false;
++ } else if (line[0] == '1' && line[1] == ':') {
++ path = line + 2;
++ write_zero = true;
++ } else {
++ g_error("read map failed - parse error ('%s')", inbuf);
++ }
++
++ path = extract_devname(path, &devname, -1);
+ }
+
+- char *devname = NULL;
+- path = extract_devname(path, &devname, -1);
+ if (!devname) {
+ g_error("read map failed - no dev name specified ('%s')",
+ inbuf);
+@@ -299,6 +314,7 @@ static int extract_content(int argc, char **argv)
+ map->throttling_group = group;
+ map->cache = cache;
+ map->write_zero = write_zero;
++ map->skip = skip;
+
+ g_hash_table_insert(devmap, map->devname, map);
+
+@@ -328,6 +344,7 @@ static int extract_content(int argc, char **argv)
+ const char *cache = NULL;
+ int flags = BDRV_O_RDWR;
+ bool write_zero = true;
++ bool skip = false;
+
+ BlockBackend *blk = NULL;
+
+@@ -343,6 +360,7 @@ static int extract_content(int argc, char **argv)
+ throttling_group = map->throttling_group;
+ cache = map->cache;
+ write_zero = map->write_zero;
++ skip = map->skip;
+ } else {
+ devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
+ dirname, di->devname);
+@@ -361,57 +379,60 @@ static int extract_content(int argc, char **argv)
+ write_zero = false;
+ }
+
+- size_t devlen = strlen(devfn);
+- QDict *options = NULL;
+- bool writethrough;
+- if (format) {
+- /* explicit format from commandline */
+- options = qdict_new();
+- qdict_put_str(options, "driver", format);
+- } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
+- strncmp(devfn, "/dev/", 5) == 0)
+- {
+- /* This part is now deprecated for PVE as well (just as qemu
+- * deprecated not specifying an explicit raw format, too.
+- */
+- /* explicit raw format */
+- options = qdict_new();
+- qdict_put_str(options, "driver", "raw");
+- }
+- if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
+- g_error("invalid cache option: %s\n", cache);
+- }
++ if (!skip) {
++ size_t devlen = strlen(devfn);
++ QDict *options = NULL;
++ bool writethrough;
++ if (format) {
++ /* explicit format from commandline */
++ options = qdict_new();
++ qdict_put_str(options, "driver", format);
++ } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
++ strncmp(devfn, "/dev/", 5) == 0)
++ {
++ /* This part is now deprecated for PVE as well (just as qemu
++ * deprecated not specifying an explicit raw format, too.
++ */
++ /* explicit raw format */
++ options = qdict_new();
++ qdict_put_str(options, "driver", "raw");
++ }
+
+- if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
+- g_error("can't open file %s - %s", devfn,
+- error_get_pretty(errp));
+- }
++ if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
++ g_error("invalid cache option: %s\n", cache);
++ }
+
+- if (cache) {
+- blk_set_enable_write_cache(blk, !writethrough);
+- }
++ if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
++ g_error("can't open file %s - %s", devfn,
++ error_get_pretty(errp));
++ }
+
+- if (throttling_group) {
+- blk_io_limits_enable(blk, throttling_group);
+- }
++ if (cache) {
++ blk_set_enable_write_cache(blk, !writethrough);
++ }
+
+- if (throttling_bps) {
+- if (!throttling_group) {
+- blk_io_limits_enable(blk, devfn);
++ if (throttling_group) {
++ blk_io_limits_enable(blk, throttling_group);
+ }
+
+- ThrottleConfig cfg;
+- throttle_config_init(&cfg);
+- cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
+- Error *err = NULL;
+- if (!throttle_is_valid(&cfg, &err)) {
+- error_report_err(err);
+- g_error("failed to apply throttling");
++ if (throttling_bps) {
++ if (!throttling_group) {
++ blk_io_limits_enable(blk, devfn);
++ }
++
++ ThrottleConfig cfg;
++ throttle_config_init(&cfg);
++ cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
++ Error *err = NULL;
++ if (!throttle_is_valid(&cfg, &err)) {
++ error_report_err(err);
++ g_error("failed to apply throttling");
++ }
++ blk_set_io_limits(blk, &cfg);
+ }
+- blk_set_io_limits(blk, &cfg);
+ }
+
+- if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
++ if (vma_reader_register_bs(vmar, i, blk, write_zero, skip, &errp) < 0) {
+ g_error("%s", error_get_pretty(errp));
+ }
+
+diff --git a/vma.h b/vma.h
+index c895c97f6d..1b62859165 100644
+--- a/vma.h
++++ b/vma.h
+@@ -142,7 +142,7 @@ GList *vma_reader_get_config_data(VmaReader *vmar);
+ VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id);
+ int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id,
+ BlockBackend *target, bool write_zeroes,
+- Error **errp);
++ bool skip, Error **errp);
+ int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
+ Error **errp);
+ int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Thu, 23 Jun 2022 14:00:05 +0200
-Subject: [PATCH] Revert "block/rbd: workaround for ceph issue #53784"
-
-This reverts commit fc176116cdea816ceb8dd969080b2b95f58edbc0 in
-preparation to revert 0347a8fd4c3faaedf119be04c197804be40a384b.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/rbd.c | 42 ++----------------------------------------
- 1 file changed, 2 insertions(+), 40 deletions(-)
-
-diff --git a/block/rbd.c b/block/rbd.c
-index 0cec24c86d..0a9b97aa5e 100644
---- a/block/rbd.c
-+++ b/block/rbd.c
-@@ -1324,7 +1324,6 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
- int status, r;
- RBDDiffIterateReq req = { .offs = offset };
- uint64_t features, flags;
-- uint64_t head = 0;
-
- assert(offset + bytes <= s->image_size);
-
-@@ -1352,43 +1351,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
- return status;
- }
-
--#if LIBRBD_VERSION_CODE < LIBRBD_VERSION(1, 17, 0)
-- /*
-- * librbd had a bug until early 2022 that affected all versions of ceph that
-- * supported fast-diff. This bug results in reporting of incorrect offsets
-- * if the offset parameter to rbd_diff_iterate2 is not object aligned.
-- * Work around this bug by rounding down the offset to object boundaries.
-- * This is OK because we call rbd_diff_iterate2 with whole_object = true.
-- * However, this workaround only works for non cloned images with default
-- * striping.
-- *
-- * See: https://tracker.ceph.com/issues/53784
-- */
--
-- /* check if RBD image has non-default striping enabled */
-- if (features & RBD_FEATURE_STRIPINGV2) {
-- return status;
-- }
--
--#pragma GCC diagnostic push
--#pragma GCC diagnostic ignored "-Wdeprecated-declarations"
-- /*
-- * check if RBD image is a clone (= has a parent).
-- *
-- * rbd_get_parent_info is deprecated from Nautilus onwards, but the
-- * replacement rbd_get_parent is not present in Luminous and Mimic.
-- */
-- if (rbd_get_parent_info(s->image, NULL, 0, NULL, 0, NULL, 0) != -ENOENT) {
-- return status;
-- }
--#pragma GCC diagnostic pop
--
-- head = req.offs & (s->object_size - 1);
-- req.offs -= head;
-- bytes += head;
--#endif
--
-- r = rbd_diff_iterate2(s->image, NULL, req.offs, bytes, true, true,
-+ r = rbd_diff_iterate2(s->image, NULL, offset, bytes, true, true,
- qemu_rbd_diff_iterate_cb, &req);
- if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
- return status;
-@@ -1407,8 +1370,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
- status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
- }
-
-- assert(req.bytes > head);
-- *pnum = req.bytes - head;
-+ *pnum = req.bytes;
- return status;
- }
-
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Tue, 26 Apr 2022 16:06:28 +0200
+Subject: [PATCH] pbs: namespace support
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+---
+ block/monitor/block-hmp-cmds.c | 1 +
+ block/pbs.c | 25 +++++++++++++++++++++----
+ pbs-restore.c | 19 ++++++++++++++++---
+ pve-backup.c | 6 +++++-
+ qapi/block-core.json | 5 ++++-
+ 5 files changed, 47 insertions(+), 9 deletions(-)
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index 4c799f00d9..0502f42be6 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1041,6 +1041,7 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
+ false, NULL, // PBS key_password
+ false, NULL, // PBS master_keyfile
+ false, NULL, // PBS fingerprint
++ false, NULL, // PBS backup-ns
+ false, NULL, // PBS backup-id
+ false, 0, // PBS backup-time
+ false, false, // PBS use-dirty-bitmap
+diff --git a/block/pbs.c b/block/pbs.c
+index c5eb4d5bad..7471e2ef9d 100644
+--- a/block/pbs.c
++++ b/block/pbs.c
+@@ -14,6 +14,7 @@
+ #include <proxmox-backup-qemu.h>
+
+ #define PBS_OPT_REPOSITORY "repository"
++#define PBS_OPT_NAMESPACE "namespace"
+ #define PBS_OPT_SNAPSHOT "snapshot"
+ #define PBS_OPT_ARCHIVE "archive"
+ #define PBS_OPT_KEYFILE "keyfile"
+@@ -27,6 +28,7 @@ typedef struct {
+ int64_t length;
+
+ char *repository;
++ char *namespace;
+ char *snapshot;
+ char *archive;
+ } BDRVPBSState;
+@@ -40,6 +42,11 @@ static QemuOptsList runtime_opts = {
+ .type = QEMU_OPT_STRING,
+ .help = "The server address and repository to connect to.",
+ },
++ {
++ .name = PBS_OPT_NAMESPACE,
++ .type = QEMU_OPT_STRING,
++ .help = "Optional: The snapshot's namespace.",
++ },
+ {
+ .name = PBS_OPT_SNAPSHOT,
+ .type = QEMU_OPT_STRING,
+@@ -76,7 +83,7 @@ static QemuOptsList runtime_opts = {
+
+
+ // filename format:
+-// pbs:repository=<repo>,snapshot=<snap>,password=<pw>,key_password=<kpw>,fingerprint=<fp>,archive=<archive>
++// pbs:repository=<repo>,namespace=<ns>,snapshot=<snap>,password=<pw>,key_password=<kpw>,fingerprint=<fp>,archive=<archive>
+ static void pbs_parse_filename(const char *filename, QDict *options,
+ Error **errp)
+ {
+@@ -112,6 +119,7 @@ static int pbs_open(BlockDriverState *bs, QDict *options, int flags,
+ s->archive = g_strdup(qemu_opt_get(opts, PBS_OPT_ARCHIVE));
+ const char *keyfile = qemu_opt_get(opts, PBS_OPT_KEYFILE);
+ const char *password = qemu_opt_get(opts, PBS_OPT_PASSWORD);
++ const char *namespace = qemu_opt_get(opts, PBS_OPT_NAMESPACE);
+ const char *fingerprint = qemu_opt_get(opts, PBS_OPT_FINGERPRINT);
+ const char *key_password = qemu_opt_get(opts, PBS_OPT_ENCRYPTION_PASSWORD);
+
+@@ -124,9 +132,12 @@ static int pbs_open(BlockDriverState *bs, QDict *options, int flags,
+ if (!key_password) {
+ key_password = getenv("PBS_ENCRYPTION_PASSWORD");
+ }
++ if (namespace) {
++ s->namespace = g_strdup(namespace);
++ }
+
+ /* connect to PBS server in read mode */
+- s->conn = proxmox_restore_new(s->repository, s->snapshot, password,
++ s->conn = proxmox_restore_new_ns(s->repository, s->snapshot, s->namespace, password,
+ keyfile, key_password, fingerprint, &pbs_error);
+
+ /* invalidates qemu_opt_get char pointers from above */
+@@ -171,6 +182,7 @@ static int pbs_file_open(BlockDriverState *bs, QDict *options, int flags,
+ static void pbs_close(BlockDriverState *bs) {
+ BDRVPBSState *s = bs->opaque;
+ g_free(s->repository);
++ g_free(s->namespace);
+ g_free(s->snapshot);
+ g_free(s->archive);
+ proxmox_restore_disconnect(s->conn);
+@@ -252,8 +264,13 @@ static coroutine_fn int pbs_co_pwritev(BlockDriverState *bs,
+ static void pbs_refresh_filename(BlockDriverState *bs)
+ {
+ BDRVPBSState *s = bs->opaque;
+- snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s(%s)",
+- s->repository, s->snapshot, s->archive);
++ if (s->namespace) {
++ snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s:%s(%s)",
++ s->repository, s->namespace, s->snapshot, s->archive);
++ } else {
++ snprintf(bs->exact_filename, sizeof(bs->exact_filename), "%s/%s(%s)",
++ s->repository, s->snapshot, s->archive);
++ }
+ }
+
+ static const char *const pbs_strong_runtime_opts[] = {
+diff --git a/pbs-restore.c b/pbs-restore.c
+index 2f834cf42e..f03d9bab8d 100644
+--- a/pbs-restore.c
++++ b/pbs-restore.c
+@@ -29,7 +29,7 @@
+ static void help(void)
+ {
+ const char *help_msg =
+- "usage: pbs-restore [--repository <repo>] snapshot archive-name target [command options]\n"
++ "usage: pbs-restore [--repository <repo>] [--ns namespace] snapshot archive-name target [command options]\n"
+ ;
+
+ printf("%s", help_msg);
+@@ -77,6 +77,7 @@ int main(int argc, char **argv)
+ Error *main_loop_err = NULL;
+ const char *format = "raw";
+ const char *repository = NULL;
++ const char *backup_ns = NULL;
+ const char *keyfile = NULL;
+ int verbose = false;
+ bool skip_zero = false;
+@@ -90,6 +91,7 @@ int main(int argc, char **argv)
+ {"verbose", no_argument, 0, 'v'},
+ {"format", required_argument, 0, 'f'},
+ {"repository", required_argument, 0, 'r'},
++ {"ns", required_argument, 0, 'n'},
+ {"keyfile", required_argument, 0, 'k'},
+ {0, 0, 0, 0}
+ };
+@@ -110,6 +112,9 @@ int main(int argc, char **argv)
+ case 'r':
+ repository = g_strdup(argv[optind - 1]);
+ break;
++ case 'n':
++ backup_ns = g_strdup(argv[optind - 1]);
++ break;
+ case 'k':
+ keyfile = g_strdup(argv[optind - 1]);
+ break;
+@@ -160,8 +165,16 @@ int main(int argc, char **argv)
+ fprintf(stderr, "connecting to repository '%s'\n", repository);
+ }
+ char *pbs_error = NULL;
+- ProxmoxRestoreHandle *conn = proxmox_restore_new(
+- repository, snapshot, password, keyfile, key_password, fingerprint, &pbs_error);
++ ProxmoxRestoreHandle *conn = proxmox_restore_new_ns(
++ repository,
++ snapshot,
++ backup_ns,
++ password,
++ keyfile,
++ key_password,
++ fingerprint,
++ &pbs_error
++ );
+ if (conn == NULL) {
+ fprintf(stderr, "restore failed: %s\n", pbs_error);
+ return -1;
+diff --git a/pve-backup.c b/pve-backup.c
+index 9f6c04a512..f6a5f8c785 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -10,6 +10,8 @@
+ #include "qapi/qmp/qerror.h"
+ #include "qemu/cutils.h"
+
++#include <proxmox-backup-qemu.h>
++
+ /* PVE backup state and related function */
+
+ /*
+@@ -535,6 +537,7 @@ UuidInfo coroutine_fn *qmp_backup(
+ bool has_key_password, const char *key_password,
+ bool has_master_keyfile, const char *master_keyfile,
+ bool has_fingerprint, const char *fingerprint,
++ bool has_backup_ns, const char *backup_ns,
+ bool has_backup_id, const char *backup_id,
+ bool has_backup_time, int64_t backup_time,
+ bool has_use_dirty_bitmap, bool use_dirty_bitmap,
+@@ -674,8 +677,9 @@ UuidInfo coroutine_fn *qmp_backup(
+ firewall_name = "fw.conf";
+
+ char *pbs_err = NULL;
+- pbs = proxmox_backup_new(
++ pbs = proxmox_backup_new_ns(
+ backup_file,
++ has_backup_ns ? backup_ns : NULL,
+ backup_id,
+ backup_time,
+ dump_cb_block_size,
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index fc8a125451..cc2ead0b75 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -817,6 +817,8 @@
+ #
+ # @fingerprint: server cert fingerprint (optional for format 'pbs')
+ #
++# @backup-ns: backup namespace (required for format 'pbs')
++#
+ # @backup-id: backup ID (required for format 'pbs')
+ #
+ # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
+@@ -836,6 +838,7 @@
+ '*key-password': 'str',
+ '*master-keyfile': 'str',
+ '*fingerprint': 'str',
++ '*backup-ns': 'str',
+ '*backup-id': 'str',
+ '*backup-time': 'int',
+ '*use-dirty-bitmap': 'bool',
+@@ -3282,7 +3285,7 @@
+ { 'struct': 'BlockdevOptionsPbs',
+ 'data': { 'repository': 'str', 'snapshot': 'str', 'archive': 'str',
+ '*keyfile': 'str', '*password': 'str', '*fingerprint': 'str',
+- '*key_password': 'str' } }
++ '*key_password': 'str', '*namespace': 'str' } }
+
+ ##
+ # @BlockdevOptionsNVMe:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Thu, 23 Jun 2022 14:00:07 +0200
-Subject: [PATCH] Revert "block/rbd: fix handling of holes in
- .bdrv_co_block_status"
-
-This reverts commit 9e302f64bb407a9bb097b626da97228c2654cfee in
-preparation to revert 0347a8fd4c3faaedf119be04c197804be40a384b.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/rbd.c | 10 +++++-----
- 1 file changed, 5 insertions(+), 5 deletions(-)
-
-diff --git a/block/rbd.c b/block/rbd.c
-index 0a9b97aa5e..1c585d23e9 100644
---- a/block/rbd.c
-+++ b/block/rbd.c
-@@ -1283,11 +1283,11 @@ static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
- RBDDiffIterateReq *req = opaque;
-
- assert(req->offs + req->bytes <= offs);
--
-- /* treat a hole like an unallocated area and bail out */
-- if (!exists) {
-- return 0;
-- }
-+ /*
-+ * we do not diff against a snapshot so we should never receive a callback
-+ * for a hole.
-+ */
-+ assert(exists);
-
- if (!req->exists && offs > req->offs) {
- /*
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Thu, 23 Jun 2022 14:00:05 +0200
+Subject: [PATCH] Revert "block/rbd: workaround for ceph issue #53784"
+
+This reverts commit fc176116cdea816ceb8dd969080b2b95f58edbc0 in
+preparation to revert 0347a8fd4c3faaedf119be04c197804be40a384b.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/rbd.c | 42 ++----------------------------------------
+ 1 file changed, 2 insertions(+), 40 deletions(-)
+
+diff --git a/block/rbd.c b/block/rbd.c
+index 64a8d7d48b..9fc6dcb957 100644
+--- a/block/rbd.c
++++ b/block/rbd.c
+@@ -1348,7 +1348,6 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+ int status, r;
+ RBDDiffIterateReq req = { .offs = offset };
+ uint64_t features, flags;
+- uint64_t head = 0;
+
+ assert(offset + bytes <= s->image_size);
+
+@@ -1376,43 +1375,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+ return status;
+ }
+
+-#if LIBRBD_VERSION_CODE < LIBRBD_VERSION(1, 17, 0)
+- /*
+- * librbd had a bug until early 2022 that affected all versions of ceph that
+- * supported fast-diff. This bug results in reporting of incorrect offsets
+- * if the offset parameter to rbd_diff_iterate2 is not object aligned.
+- * Work around this bug by rounding down the offset to object boundaries.
+- * This is OK because we call rbd_diff_iterate2 with whole_object = true.
+- * However, this workaround only works for non cloned images with default
+- * striping.
+- *
+- * See: https://tracker.ceph.com/issues/53784
+- */
+-
+- /* check if RBD image has non-default striping enabled */
+- if (features & RBD_FEATURE_STRIPINGV2) {
+- return status;
+- }
+-
+-#pragma GCC diagnostic push
+-#pragma GCC diagnostic ignored "-Wdeprecated-declarations"
+- /*
+- * check if RBD image is a clone (= has a parent).
+- *
+- * rbd_get_parent_info is deprecated from Nautilus onwards, but the
+- * replacement rbd_get_parent is not present in Luminous and Mimic.
+- */
+- if (rbd_get_parent_info(s->image, NULL, 0, NULL, 0, NULL, 0) != -ENOENT) {
+- return status;
+- }
+-#pragma GCC diagnostic pop
+-
+- head = req.offs & (s->object_size - 1);
+- req.offs -= head;
+- bytes += head;
+-#endif
+-
+- r = rbd_diff_iterate2(s->image, NULL, req.offs, bytes, true, true,
++ r = rbd_diff_iterate2(s->image, NULL, offset, bytes, true, true,
+ qemu_rbd_diff_iterate_cb, &req);
+ if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
+ return status;
+@@ -1431,8 +1394,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+ status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
+ }
+
+- assert(req.bytes > head);
+- *pnum = req.bytes - head;
++ *pnum = req.bytes;
+ return status;
+ }
+
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Thu, 23 Jun 2022 14:00:07 +0200
+Subject: [PATCH] Revert "block/rbd: fix handling of holes in
+ .bdrv_co_block_status"
+
+This reverts commit 9e302f64bb407a9bb097b626da97228c2654cfee in
+preparation to revert 0347a8fd4c3faaedf119be04c197804be40a384b.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/rbd.c | 10 +++++-----
+ 1 file changed, 5 insertions(+), 5 deletions(-)
+
+diff --git a/block/rbd.c b/block/rbd.c
+index 9fc6dcb957..98f4ba2620 100644
+--- a/block/rbd.c
++++ b/block/rbd.c
+@@ -1307,11 +1307,11 @@ static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
+ RBDDiffIterateReq *req = opaque;
+
+ assert(req->offs + req->bytes <= offs);
+-
+- /* treat a hole like an unallocated area and bail out */
+- if (!exists) {
+- return 0;
+- }
++ /*
++ * we do not diff against a snapshot so we should never receive a callback
++ * for a hole.
++ */
++ assert(exists);
+
+ if (!req->exists && offs > req->offs) {
+ /*
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Tue, 17 May 2022 09:46:02 +0200
-Subject: [PATCH] Revert "block/rbd: implement bdrv_co_block_status"
-
-During backup, bdrv_co_block_status is called for each block copy
-chunk. When RBD is used, the current implementation with
-rbd_diff_iterate2() using whole_object=true takes about linearly more
-time, depending on the image size. Since there are linearly more
-chunks, the slowdown is quadratic, becoming unacceptable for large
-images (starting somewhere between 500-1000 GiB in my testing).
-
-This reverts commit 0347a8fd4c3faaedf119be04c197804be40a384b as a
-stop-gap measure, until it's clear how to make the implemenation
-more efficient.
-
-Upstream bug report:
-https://gitlab.com/qemu-project/qemu/-/issues/1026
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/rbd.c | 112 ----------------------------------------------------
- 1 file changed, 112 deletions(-)
-
-diff --git a/block/rbd.c b/block/rbd.c
-index 1c585d23e9..1704767041 100644
---- a/block/rbd.c
-+++ b/block/rbd.c
-@@ -97,12 +97,6 @@ typedef struct RBDTask {
- int64_t ret;
- } RBDTask;
-
--typedef struct RBDDiffIterateReq {
-- uint64_t offs;
-- uint64_t bytes;
-- bool exists;
--} RBDDiffIterateReq;
--
- static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
- BlockdevOptionsRbd *opts, bool cache,
- const char *keypairs, const char *secretid,
-@@ -1269,111 +1263,6 @@ static ImageInfoSpecific *qemu_rbd_get_specific_info(BlockDriverState *bs,
- return spec_info;
- }
-
--/*
-- * rbd_diff_iterate2 allows to interrupt the exection by returning a negative
-- * value in the callback routine. Choose a value that does not conflict with
-- * an existing exitcode and return it if we want to prematurely stop the
-- * execution because we detected a change in the allocation status.
-- */
--#define QEMU_RBD_EXIT_DIFF_ITERATE2 -9000
--
--static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
-- int exists, void *opaque)
--{
-- RBDDiffIterateReq *req = opaque;
--
-- assert(req->offs + req->bytes <= offs);
-- /*
-- * we do not diff against a snapshot so we should never receive a callback
-- * for a hole.
-- */
-- assert(exists);
--
-- if (!req->exists && offs > req->offs) {
-- /*
-- * we started in an unallocated area and hit the first allocated
-- * block. req->bytes must be set to the length of the unallocated area
-- * before the allocated area. stop further processing.
-- */
-- req->bytes = offs - req->offs;
-- return QEMU_RBD_EXIT_DIFF_ITERATE2;
-- }
--
-- if (req->exists && offs > req->offs + req->bytes) {
-- /*
-- * we started in an allocated area and jumped over an unallocated area,
-- * req->bytes contains the length of the allocated area before the
-- * unallocated area. stop further processing.
-- */
-- return QEMU_RBD_EXIT_DIFF_ITERATE2;
-- }
--
-- req->bytes += len;
-- req->exists = true;
--
-- return 0;
--}
--
--static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
-- bool want_zero, int64_t offset,
-- int64_t bytes, int64_t *pnum,
-- int64_t *map,
-- BlockDriverState **file)
--{
-- BDRVRBDState *s = bs->opaque;
-- int status, r;
-- RBDDiffIterateReq req = { .offs = offset };
-- uint64_t features, flags;
--
-- assert(offset + bytes <= s->image_size);
--
-- /* default to all sectors allocated */
-- status = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID;
-- *map = offset;
-- *file = bs;
-- *pnum = bytes;
--
-- /* check if RBD image supports fast-diff */
-- r = rbd_get_features(s->image, &features);
-- if (r < 0) {
-- return status;
-- }
-- if (!(features & RBD_FEATURE_FAST_DIFF)) {
-- return status;
-- }
--
-- /* check if RBD fast-diff result is valid */
-- r = rbd_get_flags(s->image, &flags);
-- if (r < 0) {
-- return status;
-- }
-- if (flags & RBD_FLAG_FAST_DIFF_INVALID) {
-- return status;
-- }
--
-- r = rbd_diff_iterate2(s->image, NULL, offset, bytes, true, true,
-- qemu_rbd_diff_iterate_cb, &req);
-- if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
-- return status;
-- }
-- assert(req.bytes <= bytes);
-- if (!req.exists) {
-- if (r == 0) {
-- /*
-- * rbd_diff_iterate2 does not invoke callbacks for unallocated
-- * areas. This here catches the case where no callback was
-- * invoked at all (req.bytes == 0).
-- */
-- assert(req.bytes == 0);
-- req.bytes = bytes;
-- }
-- status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
-- }
--
-- *pnum = req.bytes;
-- return status;
--}
--
- static int64_t qemu_rbd_getlength(BlockDriverState *bs)
- {
- BDRVRBDState *s = bs->opaque;
-@@ -1609,7 +1498,6 @@ static BlockDriver bdrv_rbd = {
- #ifdef LIBRBD_SUPPORTS_WRITE_ZEROES
- .bdrv_co_pwrite_zeroes = qemu_rbd_co_pwrite_zeroes,
- #endif
-- .bdrv_co_block_status = qemu_rbd_co_block_status,
-
- .bdrv_snapshot_create = qemu_rbd_snap_create,
- .bdrv_snapshot_delete = qemu_rbd_snap_remove,
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 25 May 2022 13:59:37 +0200
-Subject: [PATCH] PVE-Backup: create jobs: correctly cancel in error scenario
-
-The first call to job_cancel_sync() will cancel and free all jobs in
-the transaction, so ensure that it's called only once and get rid of
-the job_unref() that would operate on freed memory.
-
-It's also necessary to NULL backup_state.pbs in the error scenario,
-because a subsequent backup_cancel QMP call (as happens in PVE when
-the backup QMP command fails) would try to call proxmox_backup_abort()
-and run into a segfault.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- pve-backup.c | 10 ++++++++--
- 1 file changed, 8 insertions(+), 2 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index f6a5f8c785..5bed6f4014 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -506,6 +506,11 @@ static void create_backup_jobs_bh(void *opaque) {
- }
-
- if (*errp) {
-+ /*
-+ * It's enough to cancel one job in the transaction, the rest will
-+ * follow automatically.
-+ */
-+ bool canceled = false;
- l = backup_state.di_list;
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-@@ -516,12 +521,12 @@ static void create_backup_jobs_bh(void *opaque) {
- di->target = NULL;
- }
-
-- if (di->job) {
-+ if (!canceled && di->job) {
- AioContext *ctx = di->job->job.aio_context;
- aio_context_acquire(ctx);
- job_cancel_sync(&di->job->job, true);
-- job_unref(&di->job->job);
- aio_context_release(ctx);
-+ canceled = true;
- }
- }
- }
-@@ -947,6 +952,7 @@ err:
-
- if (pbs) {
- proxmox_backup_disconnect(pbs);
-+ backup_state.pbs = NULL;
- }
-
- if (backup_dir) {
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Tue, 17 May 2022 09:46:02 +0200
+Subject: [PATCH] Revert "block/rbd: implement bdrv_co_block_status"
+
+During backup, bdrv_co_block_status is called for each block copy
+chunk. When RBD is used, the current implementation with
+rbd_diff_iterate2() using whole_object=true takes about linearly more
+time, depending on the image size. Since there are linearly more
+chunks, the slowdown is quadratic, becoming unacceptable for large
+images (starting somewhere between 500-1000 GiB in my testing).
+
+This reverts commit 0347a8fd4c3faaedf119be04c197804be40a384b as a
+stop-gap measure, until it's clear how to make the implemenation
+more efficient.
+
+Upstream bug report:
+https://gitlab.com/qemu-project/qemu/-/issues/1026
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/rbd.c | 112 ----------------------------------------------------
+ 1 file changed, 112 deletions(-)
+
+diff --git a/block/rbd.c b/block/rbd.c
+index 98f4ba2620..efcbbe5949 100644
+--- a/block/rbd.c
++++ b/block/rbd.c
+@@ -97,12 +97,6 @@ typedef struct RBDTask {
+ int64_t ret;
+ } RBDTask;
+
+-typedef struct RBDDiffIterateReq {
+- uint64_t offs;
+- uint64_t bytes;
+- bool exists;
+-} RBDDiffIterateReq;
+-
+ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
+ BlockdevOptionsRbd *opts, bool cache,
+ const char *keypairs, const char *secretid,
+@@ -1293,111 +1287,6 @@ static ImageInfoSpecific *qemu_rbd_get_specific_info(BlockDriverState *bs,
+ return spec_info;
+ }
+
+-/*
+- * rbd_diff_iterate2 allows to interrupt the exection by returning a negative
+- * value in the callback routine. Choose a value that does not conflict with
+- * an existing exitcode and return it if we want to prematurely stop the
+- * execution because we detected a change in the allocation status.
+- */
+-#define QEMU_RBD_EXIT_DIFF_ITERATE2 -9000
+-
+-static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
+- int exists, void *opaque)
+-{
+- RBDDiffIterateReq *req = opaque;
+-
+- assert(req->offs + req->bytes <= offs);
+- /*
+- * we do not diff against a snapshot so we should never receive a callback
+- * for a hole.
+- */
+- assert(exists);
+-
+- if (!req->exists && offs > req->offs) {
+- /*
+- * we started in an unallocated area and hit the first allocated
+- * block. req->bytes must be set to the length of the unallocated area
+- * before the allocated area. stop further processing.
+- */
+- req->bytes = offs - req->offs;
+- return QEMU_RBD_EXIT_DIFF_ITERATE2;
+- }
+-
+- if (req->exists && offs > req->offs + req->bytes) {
+- /*
+- * we started in an allocated area and jumped over an unallocated area,
+- * req->bytes contains the length of the allocated area before the
+- * unallocated area. stop further processing.
+- */
+- return QEMU_RBD_EXIT_DIFF_ITERATE2;
+- }
+-
+- req->bytes += len;
+- req->exists = true;
+-
+- return 0;
+-}
+-
+-static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+- bool want_zero, int64_t offset,
+- int64_t bytes, int64_t *pnum,
+- int64_t *map,
+- BlockDriverState **file)
+-{
+- BDRVRBDState *s = bs->opaque;
+- int status, r;
+- RBDDiffIterateReq req = { .offs = offset };
+- uint64_t features, flags;
+-
+- assert(offset + bytes <= s->image_size);
+-
+- /* default to all sectors allocated */
+- status = BDRV_BLOCK_DATA | BDRV_BLOCK_OFFSET_VALID;
+- *map = offset;
+- *file = bs;
+- *pnum = bytes;
+-
+- /* check if RBD image supports fast-diff */
+- r = rbd_get_features(s->image, &features);
+- if (r < 0) {
+- return status;
+- }
+- if (!(features & RBD_FEATURE_FAST_DIFF)) {
+- return status;
+- }
+-
+- /* check if RBD fast-diff result is valid */
+- r = rbd_get_flags(s->image, &flags);
+- if (r < 0) {
+- return status;
+- }
+- if (flags & RBD_FLAG_FAST_DIFF_INVALID) {
+- return status;
+- }
+-
+- r = rbd_diff_iterate2(s->image, NULL, offset, bytes, true, true,
+- qemu_rbd_diff_iterate_cb, &req);
+- if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
+- return status;
+- }
+- assert(req.bytes <= bytes);
+- if (!req.exists) {
+- if (r == 0) {
+- /*
+- * rbd_diff_iterate2 does not invoke callbacks for unallocated
+- * areas. This here catches the case where no callback was
+- * invoked at all (req.bytes == 0).
+- */
+- assert(req.bytes == 0);
+- req.bytes = bytes;
+- }
+- status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
+- }
+-
+- *pnum = req.bytes;
+- return status;
+-}
+-
+ static int64_t qemu_rbd_getlength(BlockDriverState *bs)
+ {
+ BDRVRBDState *s = bs->opaque;
+@@ -1633,7 +1522,6 @@ static BlockDriver bdrv_rbd = {
+ #ifdef LIBRBD_SUPPORTS_WRITE_ZEROES
+ .bdrv_co_pwrite_zeroes = qemu_rbd_co_pwrite_zeroes,
+ #endif
+- .bdrv_co_block_status = qemu_rbd_co_block_status,
+
+ .bdrv_snapshot_create = qemu_rbd_snap_create,
+ .bdrv_snapshot_delete = qemu_rbd_snap_remove,
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 25 May 2022 13:59:37 +0200
+Subject: [PATCH] PVE-Backup: create jobs: correctly cancel in error scenario
+
+The first call to job_cancel_sync() will cancel and free all jobs in
+the transaction, so ensure that it's called only once and get rid of
+the job_unref() that would operate on freed memory.
+
+It's also necessary to NULL backup_state.pbs in the error scenario,
+because a subsequent backup_cancel QMP call (as happens in PVE when
+the backup QMP command fails) would try to call proxmox_backup_abort()
+and run into a segfault.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+---
+ pve-backup.c | 10 ++++++++--
+ 1 file changed, 8 insertions(+), 2 deletions(-)
+
+diff --git a/pve-backup.c b/pve-backup.c
+index f6a5f8c785..5bed6f4014 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -506,6 +506,11 @@ static void create_backup_jobs_bh(void *opaque) {
+ }
+
+ if (*errp) {
++ /*
++ * It's enough to cancel one job in the transaction, the rest will
++ * follow automatically.
++ */
++ bool canceled = false;
+ l = backup_state.di_list;
+ while (l) {
+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
+@@ -516,12 +521,12 @@ static void create_backup_jobs_bh(void *opaque) {
+ di->target = NULL;
+ }
+
+- if (di->job) {
++ if (!canceled && di->job) {
+ AioContext *ctx = di->job->job.aio_context;
+ aio_context_acquire(ctx);
+ job_cancel_sync(&di->job->job, true);
+- job_unref(&di->job->job);
+ aio_context_release(ctx);
++ canceled = true;
+ }
+ }
+ }
+@@ -947,6 +952,7 @@ err:
+
+ if (pbs) {
+ proxmox_backup_disconnect(pbs);
++ backup_state.pbs = NULL;
+ }
+
+ if (backup_dir) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 25 May 2022 13:59:38 +0200
-Subject: [PATCH] PVE-Backup: ensure jobs in di_list are referenced
-
-Ensures that qmp_backup_cancel doesn't pick a job that's already been
-freed. With unlucky timings it seems possible that:
-1. job_exit -> job_completed -> job_finalize_single starts
-2. pvebackup_co_complete_stream gets spawned in completion callback
-3. job finalize_single finishes -> job's refcount hits zero -> job is
- freed
-4. qmp_backup_cancel comes in and locks backup_state.backup_mutex
- before pvebackup_co_complete_stream can remove the job from the
- di_list
-5. qmp_backup_cancel will pick a job that's already been freed
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- pve-backup.c | 25 ++++++++++++++++++++-----
- 1 file changed, 20 insertions(+), 5 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 5bed6f4014..0c34428713 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -316,6 +316,14 @@ static void coroutine_fn pvebackup_co_complete_stream(void *opaque)
- }
- }
-
-+ if (di->job) {
-+ AioContext *ctx = di->job->job.aio_context;
-+ aio_context_acquire(ctx);
-+ job_unref(&di->job->job);
-+ di->job = NULL;
-+ aio_context_release(ctx);
-+ }
-+
- // remove self from job list
- backup_state.di_list = g_list_remove(backup_state.di_list, di);
-
-@@ -491,9 +499,12 @@ static void create_backup_jobs_bh(void *opaque) {
- bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
- JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn, &local_err);
-
-- aio_context_release(aio_context);
--
- di->job = job;
-+ if (job) {
-+ job_ref(&job->job);
-+ }
-+
-+ aio_context_release(aio_context);
-
- if (!job || local_err) {
- error_setg(errp, "backup_job_create failed: %s",
-@@ -521,12 +532,16 @@ static void create_backup_jobs_bh(void *opaque) {
- di->target = NULL;
- }
-
-- if (!canceled && di->job) {
-+ if (di->job) {
- AioContext *ctx = di->job->job.aio_context;
- aio_context_acquire(ctx);
-- job_cancel_sync(&di->job->job, true);
-+ if (!canceled) {
-+ job_cancel_sync(&di->job->job, true);
-+ canceled = true;
-+ }
-+ job_unref(&di->job->job);
-+ di->job = NULL;
- aio_context_release(ctx);
-- canceled = true;
- }
- }
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 25 May 2022 13:59:39 +0200
-Subject: [PATCH] PVE-Backup: avoid segfault issues upon backup-cancel
-
-When canceling a backup in PVE via a signal it's easy to run into a
-situation where the job is already failing when the backup_cancel QMP
-command comes in. With a bit of unlucky timing on top, it can happen
-that job_exit() runs between schedulung of job_cancel_bh() and
-execution of job_cancel_bh(). But job_cancel_sync() does not expect
-that the job is already finalized (in fact, the job might've been
-freed already, but even if it isn't, job_cancel_sync() would try to
-deref job->txn which would be NULL at that point).
-
-It is not possible to simply use the job_cancel() (which is advertised
-as being async but isn't in all cases) in qmp_backup_cancel() for the
-same reason job_cancel_sync() cannot be used. Namely, because it can
-invoke job_finish_sync() (which uses AIO_WAIT_WHILE and thus hangs if
-called from a coroutine). This happens when there's multiple jobs in
-the transaction and job->deferred_to_main_loop is true (is set before
-scheduling job_exit()) or if the job was not started yet.
-
-Fix the issue by selecting the job to cancel in job_cancel_bh() itself
-using the first job that's not completed yet. This is not necessarily
-the first job in the list, because pvebackup_co_complete_stream()
-might not yet have removed a completed job when job_cancel_bh() runs.
-
-An alternative would be to continue using only the first job and
-checking against JOB_STATUS_CONCLUDED or JOB_STATUS_NULL to decide if
-it's still necessary and possible to cancel, but the approach with
-using the first non-completed job seemed more robust.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- pve-backup.c | 61 +++++++++++++++++++++++++++++++++-------------------
- 1 file changed, 39 insertions(+), 22 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 0c34428713..2e22030eec 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -355,15 +355,42 @@ static void pvebackup_complete_cb(void *opaque, int ret)
-
- /*
- * job_cancel(_sync) does not like to be called from coroutines, so defer to
-- * main loop processing via a bottom half.
-+ * main loop processing via a bottom half. Assumes that caller holds
-+ * backup_mutex.
- */
- static void job_cancel_bh(void *opaque) {
- CoCtxData *data = (CoCtxData*)opaque;
-- Job *job = (Job*)data->data;
-- AioContext *job_ctx = job->aio_context;
-- aio_context_acquire(job_ctx);
-- job_cancel_sync(job, true);
-- aio_context_release(job_ctx);
-+
-+ /*
-+ * Be careful to pick a valid job to cancel:
-+ * 1. job_cancel_sync() does not expect the job to be finalized already.
-+ * 2. job_exit() might run between scheduling and running job_cancel_bh()
-+ * and pvebackup_co_complete_stream() might not have removed the job from
-+ * the list yet (in fact, cannot, because it waits for the backup_mutex).
-+ * Requiring !job_is_completed() ensures that no finalized job is picked.
-+ */
-+ GList *bdi = g_list_first(backup_state.di_list);
-+ while (bdi) {
-+ if (bdi->data) {
-+ BlockJob *bj = ((PVEBackupDevInfo *)bdi->data)->job;
-+ if (bj) {
-+ Job *job = &bj->job;
-+ if (!job_is_completed(job)) {
-+ AioContext *job_ctx = job->aio_context;
-+ aio_context_acquire(job_ctx);
-+ job_cancel_sync(job, true);
-+ aio_context_release(job_ctx);
-+ /*
-+ * It's enough to cancel one job in the transaction, the
-+ * rest will follow automatically.
-+ */
-+ break;
-+ }
-+ }
-+ }
-+ bdi = g_list_next(bdi);
-+ }
-+
- aio_co_enter(data->ctx, data->co);
- }
-
-@@ -384,22 +411,12 @@ void coroutine_fn qmp_backup_cancel(Error **errp)
- proxmox_backup_abort(backup_state.pbs, "backup canceled");
- }
-
-- /* it's enough to cancel one job in the transaction, the rest will follow
-- * automatically */
-- GList *bdi = g_list_first(backup_state.di_list);
-- BlockJob *cancel_job = bdi && bdi->data ?
-- ((PVEBackupDevInfo *)bdi->data)->job :
-- NULL;
--
-- if (cancel_job) {
-- CoCtxData data = {
-- .ctx = qemu_get_current_aio_context(),
-- .co = qemu_coroutine_self(),
-- .data = &cancel_job->job,
-- };
-- aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
-- qemu_coroutine_yield();
-- }
-+ CoCtxData data = {
-+ .ctx = qemu_get_current_aio_context(),
-+ .co = qemu_coroutine_self(),
-+ };
-+ aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
-+ qemu_coroutine_yield();
-
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 25 May 2022 13:59:38 +0200
+Subject: [PATCH] PVE-Backup: ensure jobs in di_list are referenced
+
+Ensures that qmp_backup_cancel doesn't pick a job that's already been
+freed. With unlucky timings it seems possible that:
+1. job_exit -> job_completed -> job_finalize_single starts
+2. pvebackup_co_complete_stream gets spawned in completion callback
+3. job finalize_single finishes -> job's refcount hits zero -> job is
+ freed
+4. qmp_backup_cancel comes in and locks backup_state.backup_mutex
+ before pvebackup_co_complete_stream can remove the job from the
+ di_list
+5. qmp_backup_cancel will pick a job that's already been freed
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+---
+ pve-backup.c | 25 ++++++++++++++++++++-----
+ 1 file changed, 20 insertions(+), 5 deletions(-)
+
+diff --git a/pve-backup.c b/pve-backup.c
+index 5bed6f4014..0c34428713 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -316,6 +316,14 @@ static void coroutine_fn pvebackup_co_complete_stream(void *opaque)
+ }
+ }
+
++ if (di->job) {
++ AioContext *ctx = di->job->job.aio_context;
++ aio_context_acquire(ctx);
++ job_unref(&di->job->job);
++ di->job = NULL;
++ aio_context_release(ctx);
++ }
++
+ // remove self from job list
+ backup_state.di_list = g_list_remove(backup_state.di_list, di);
+
+@@ -491,9 +499,12 @@ static void create_backup_jobs_bh(void *opaque) {
+ bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
+ JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn, &local_err);
+
+- aio_context_release(aio_context);
+-
+ di->job = job;
++ if (job) {
++ job_ref(&job->job);
++ }
++
++ aio_context_release(aio_context);
+
+ if (!job || local_err) {
+ error_setg(errp, "backup_job_create failed: %s",
+@@ -521,12 +532,16 @@ static void create_backup_jobs_bh(void *opaque) {
+ di->target = NULL;
+ }
+
+- if (!canceled && di->job) {
++ if (di->job) {
+ AioContext *ctx = di->job->job.aio_context;
+ aio_context_acquire(ctx);
+- job_cancel_sync(&di->job->job, true);
++ if (!canceled) {
++ job_cancel_sync(&di->job->job, true);
++ canceled = true;
++ }
++ job_unref(&di->job->job);
++ di->job = NULL;
+ aio_context_release(ctx);
+- canceled = true;
+ }
+ }
+ }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 25 May 2022 13:59:39 +0200
+Subject: [PATCH] PVE-Backup: avoid segfault issues upon backup-cancel
+
+When canceling a backup in PVE via a signal it's easy to run into a
+situation where the job is already failing when the backup_cancel QMP
+command comes in. With a bit of unlucky timing on top, it can happen
+that job_exit() runs between schedulung of job_cancel_bh() and
+execution of job_cancel_bh(). But job_cancel_sync() does not expect
+that the job is already finalized (in fact, the job might've been
+freed already, but even if it isn't, job_cancel_sync() would try to
+deref job->txn which would be NULL at that point).
+
+It is not possible to simply use the job_cancel() (which is advertised
+as being async but isn't in all cases) in qmp_backup_cancel() for the
+same reason job_cancel_sync() cannot be used. Namely, because it can
+invoke job_finish_sync() (which uses AIO_WAIT_WHILE and thus hangs if
+called from a coroutine). This happens when there's multiple jobs in
+the transaction and job->deferred_to_main_loop is true (is set before
+scheduling job_exit()) or if the job was not started yet.
+
+Fix the issue by selecting the job to cancel in job_cancel_bh() itself
+using the first job that's not completed yet. This is not necessarily
+the first job in the list, because pvebackup_co_complete_stream()
+might not yet have removed a completed job when job_cancel_bh() runs.
+
+An alternative would be to continue using only the first job and
+checking against JOB_STATUS_CONCLUDED or JOB_STATUS_NULL to decide if
+it's still necessary and possible to cancel, but the approach with
+using the first non-completed job seemed more robust.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+---
+ pve-backup.c | 61 +++++++++++++++++++++++++++++++++-------------------
+ 1 file changed, 39 insertions(+), 22 deletions(-)
+
+diff --git a/pve-backup.c b/pve-backup.c
+index 0c34428713..2e22030eec 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -355,15 +355,42 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+
+ /*
+ * job_cancel(_sync) does not like to be called from coroutines, so defer to
+- * main loop processing via a bottom half.
++ * main loop processing via a bottom half. Assumes that caller holds
++ * backup_mutex.
+ */
+ static void job_cancel_bh(void *opaque) {
+ CoCtxData *data = (CoCtxData*)opaque;
+- Job *job = (Job*)data->data;
+- AioContext *job_ctx = job->aio_context;
+- aio_context_acquire(job_ctx);
+- job_cancel_sync(job, true);
+- aio_context_release(job_ctx);
++
++ /*
++ * Be careful to pick a valid job to cancel:
++ * 1. job_cancel_sync() does not expect the job to be finalized already.
++ * 2. job_exit() might run between scheduling and running job_cancel_bh()
++ * and pvebackup_co_complete_stream() might not have removed the job from
++ * the list yet (in fact, cannot, because it waits for the backup_mutex).
++ * Requiring !job_is_completed() ensures that no finalized job is picked.
++ */
++ GList *bdi = g_list_first(backup_state.di_list);
++ while (bdi) {
++ if (bdi->data) {
++ BlockJob *bj = ((PVEBackupDevInfo *)bdi->data)->job;
++ if (bj) {
++ Job *job = &bj->job;
++ if (!job_is_completed(job)) {
++ AioContext *job_ctx = job->aio_context;
++ aio_context_acquire(job_ctx);
++ job_cancel_sync(job, true);
++ aio_context_release(job_ctx);
++ /*
++ * It's enough to cancel one job in the transaction, the
++ * rest will follow automatically.
++ */
++ break;
++ }
++ }
++ }
++ bdi = g_list_next(bdi);
++ }
++
+ aio_co_enter(data->ctx, data->co);
+ }
+
+@@ -384,22 +411,12 @@ void coroutine_fn qmp_backup_cancel(Error **errp)
+ proxmox_backup_abort(backup_state.pbs, "backup canceled");
+ }
+
+- /* it's enough to cancel one job in the transaction, the rest will follow
+- * automatically */
+- GList *bdi = g_list_first(backup_state.di_list);
+- BlockJob *cancel_job = bdi && bdi->data ?
+- ((PVEBackupDevInfo *)bdi->data)->job :
+- NULL;
+-
+- if (cancel_job) {
+- CoCtxData data = {
+- .ctx = qemu_get_current_aio_context(),
+- .co = qemu_coroutine_self(),
+- .data = &cancel_job->job,
+- };
+- aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
+- qemu_coroutine_yield();
+- }
++ CoCtxData data = {
++ .ctx = qemu_get_current_aio_context(),
++ .co = qemu_coroutine_self(),
++ };
++ aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
++ qemu_coroutine_yield();
+
+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 22 Jun 2022 10:45:11 +0200
-Subject: [PATCH] vma: create: support 64KiB-unaligned input images
-
-which fixes backing up templates with such disks in PVE, for example
-efitype=4m EFI disks on a file-based storage (size = 540672).
-
-If there is not enough left to read, blk_co_preadv will return -EIO,
-so limit the size in the last iteration.
-
-For writing, an unaligned end is already handled correctly.
-
-The call to memset is not strictly necessary, because writing also
-checks that it doesn't write data beyond the end of the image. But
-there are two reasons to do it:
-1. It's cleaner that way.
-2. It allows detecting when the final piece is all zeroes, which might
- not happen if the buffer still contains data from the previous
- iteration.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- vma.c | 12 ++++++++++--
- 1 file changed, 10 insertions(+), 2 deletions(-)
-
-diff --git a/vma.c b/vma.c
-index e1a750ff50..b1137e27a7 100644
---- a/vma.c
-+++ b/vma.c
-@@ -549,7 +549,7 @@ static void coroutine_fn backup_run(void *opaque)
- struct iovec iov;
- QEMUIOVector qiov;
-
-- int64_t start, end;
-+ int64_t start, end, readlen;
- int ret = 0;
-
- unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
-@@ -563,8 +563,16 @@ static void coroutine_fn backup_run(void *opaque)
- iov.iov_len = VMA_CLUSTER_SIZE;
- qemu_iovec_init_external(&qiov, &iov, 1);
-
-+ if (start + 1 == end) {
-+ memset(buf, 0, VMA_CLUSTER_SIZE);
-+ readlen = job->len - start * VMA_CLUSTER_SIZE;
-+ assert(readlen > 0 && readlen <= VMA_CLUSTER_SIZE);
-+ } else {
-+ readlen = VMA_CLUSTER_SIZE;
-+ }
-+
- ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
-- VMA_CLUSTER_SIZE, &qiov, 0);
-+ readlen, &qiov, 0);
- if (ret < 0) {
- vma_writer_set_error(job->vmaw, "read error", -1);
- goto out;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 22 Jun 2022 10:45:12 +0200
-Subject: [PATCH] vma: create: avoid triggering assertion in error case
-
-error_setg expects its argument to not be initialized yet.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- vma-writer.c | 2 ++
- 1 file changed, 2 insertions(+)
-
-diff --git a/vma-writer.c b/vma-writer.c
-index df4b20793d..ac7da237d0 100644
---- a/vma-writer.c
-+++ b/vma-writer.c
-@@ -311,6 +311,8 @@ VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
- }
-
- if (vmaw->fd < 0) {
-+ error_free(*errp);
-+ *errp = NULL;
- error_setg(errp, "can't open file %s - %s\n", filename,
- g_strerror(errno));
- goto err;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 22 Jun 2022 10:45:11 +0200
+Subject: [PATCH] vma: create: support 64KiB-unaligned input images
+
+which fixes backing up templates with such disks in PVE, for example
+efitype=4m EFI disks on a file-based storage (size = 540672).
+
+If there is not enough left to read, blk_co_preadv will return -EIO,
+so limit the size in the last iteration.
+
+For writing, an unaligned end is already handled correctly.
+
+The call to memset is not strictly necessary, because writing also
+checks that it doesn't write data beyond the end of the image. But
+there are two reasons to do it:
+1. It's cleaner that way.
+2. It allows detecting when the final piece is all zeroes, which might
+ not happen if the buffer still contains data from the previous
+ iteration.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ vma.c | 12 ++++++++++--
+ 1 file changed, 10 insertions(+), 2 deletions(-)
+
+diff --git a/vma.c b/vma.c
+index e6e9ffc7fe..304f02bc84 100644
+--- a/vma.c
++++ b/vma.c
+@@ -548,7 +548,7 @@ static void coroutine_fn backup_run(void *opaque)
+ struct iovec iov;
+ QEMUIOVector qiov;
+
+- int64_t start, end;
++ int64_t start, end, readlen;
+ int ret = 0;
+
+ unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
+@@ -562,8 +562,16 @@ static void coroutine_fn backup_run(void *opaque)
+ iov.iov_len = VMA_CLUSTER_SIZE;
+ qemu_iovec_init_external(&qiov, &iov, 1);
+
++ if (start + 1 == end) {
++ memset(buf, 0, VMA_CLUSTER_SIZE);
++ readlen = job->len - start * VMA_CLUSTER_SIZE;
++ assert(readlen > 0 && readlen <= VMA_CLUSTER_SIZE);
++ } else {
++ readlen = VMA_CLUSTER_SIZE;
++ }
++
+ ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
+- VMA_CLUSTER_SIZE, &qiov, 0);
++ readlen, &qiov, 0);
+ if (ret < 0) {
+ vma_writer_set_error(job->vmaw, "read error", -1);
+ goto out;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fabian Ebner <f.ebner@proxmox.com>
-Date: Wed, 22 Jun 2022 10:45:13 +0200
-Subject: [PATCH] block: alloc-track: avoid premature break
-
-While the bdrv_co_preadv() calls are expected to return 0 on success,
-qemu_iovec_memset() will return the number of bytes set (will be
-local_bytes, because the slice with that size was just initialized).
-
-Don't break out of the loop after the branch with qemu_iovec_memset(),
-because there might still be work to do. Additionally, ret is an int,
-which on 64-bit platforms is too small to hold the size_t returned by
-qemu_iovec_memset().
-
-The branch seems to be difficult to reach in practice, because the
-whole point of alloc-track is to be used with a backing device.
-
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
----
- block/alloc-track.c | 3 ++-
- 1 file changed, 2 insertions(+), 1 deletion(-)
-
-diff --git a/block/alloc-track.c b/block/alloc-track.c
-index 6b50fbe537..c1160af04b 100644
---- a/block/alloc-track.c
-+++ b/block/alloc-track.c
-@@ -174,7 +174,8 @@ static int coroutine_fn track_co_preadv(BlockDriverState *bs,
- ret = bdrv_co_preadv(bs->backing, local_offset, local_bytes,
- &local_qiov, flags);
- } else {
-- ret = qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
-+ qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
-+ ret = 0;
- }
-
- if (ret != 0) {
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 22 Jun 2022 10:45:12 +0200
+Subject: [PATCH] vma: create: avoid triggering assertion in error case
+
+error_setg expects its argument to not be initialized yet.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ vma-writer.c | 2 ++
+ 1 file changed, 2 insertions(+)
+
+diff --git a/vma-writer.c b/vma-writer.c
+index df4b20793d..ac7da237d0 100644
+--- a/vma-writer.c
++++ b/vma-writer.c
+@@ -311,6 +311,8 @@ VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
+ }
+
+ if (vmaw->fd < 0) {
++ error_free(*errp);
++ *errp = NULL;
+ error_setg(errp, "can't open file %s - %s\n", filename,
+ g_strerror(errno));
+ goto err;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Fiona Ebner <f.ebner@proxmox.com>
-Date: Mon, 3 Oct 2022 15:52:04 +0200
-Subject: [PATCH] PVE Backup: allow passing max-workers performance setting
-
-For query-proxmox-support, add an indication that it's possible to use
-the setting.
-
-For now, the other two BackupPerf settings are not exposed:
-
-* use-copy-range: would need to be implemented by the backup-dump
-block driver first, and in fact, the default for backup was changed,
-because it wasn't as fast for backup in QEMU, see commit
-6a30f663d4c0b3c45a544d541e0c4e214b2473a1.
-
-* max-chunk: enforced to be at least the backup cluster size, which is
-4 MiB for PBS and otherwise maximum of source and target cluster size.
-And block-copy has a maximum buffer size of 1 MiB, so setting a larger
-max-chunk doesn't even have an effect. To make the setting sensibly
-usable the check would need to be removed and optionally the
-block-copy max buffer size would need to be bumped. I tried doing just
-that, and tested different source/target combinations with different
-max-chunk settings, but there were no noticable improvements over the
-default "unlimited" (resulting in 1 MiB for block-copy).
-
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- block/monitor/block-hmp-cmds.c | 4 +++-
- pve-backup.c | 18 +++++++++++++-----
- qapi/block-core.json | 9 +++++++--
- 3 files changed, 23 insertions(+), 8 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index 0502f42be6..cc231ec3f2 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1049,7 +1049,9 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
- false, false, // PBS encrypt
- true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
- false, NULL, false, NULL, !!devlist,
-- devlist, qdict_haskey(qdict, "speed"), speed, &error);
-+ devlist, qdict_haskey(qdict, "speed"), speed,
-+ false, 0, // BackupPerf max-workers
-+ &error);
-
- hmp_handle_error(mon, error);
- }
-diff --git a/pve-backup.c b/pve-backup.c
-index 2e22030eec..e9aa7e0f49 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -55,6 +55,7 @@ static struct PVEBackupState {
- bool starting;
- } stat;
- int64_t speed;
-+ BackupPerf perf;
- VmaWriter *vmaw;
- ProxmoxBackupHandle *pbs;
- GList *di_list;
-@@ -492,8 +493,6 @@ static void create_backup_jobs_bh(void *opaque) {
- }
- backup_state.txn = job_txn_new_seq();
-
-- BackupPerf perf = { .max_workers = 16 };
--
- /* create and start all jobs (paused state) */
- GList *l = backup_state.di_list;
- while (l) {
-@@ -513,8 +512,9 @@ static void create_backup_jobs_bh(void *opaque) {
-
- BlockJob *job = backup_job_create(
- NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
-- bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-- JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn, &local_err);
-+ bitmap_mode, false, NULL, &backup_state.perf, BLOCKDEV_ON_ERROR_REPORT,
-+ BLOCKDEV_ON_ERROR_REPORT, JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn,
-+ &local_err);
-
- di->job = job;
- if (job) {
-@@ -584,7 +584,9 @@ UuidInfo coroutine_fn *qmp_backup(
- bool has_config_file, const char *config_file,
- bool has_firewall_file, const char *firewall_file,
- bool has_devlist, const char *devlist,
-- bool has_speed, int64_t speed, Error **errp)
-+ bool has_speed, int64_t speed,
-+ bool has_max_workers, int64_t max_workers,
-+ Error **errp)
- {
- assert(qemu_in_coroutine());
-
-@@ -914,6 +916,11 @@ UuidInfo coroutine_fn *qmp_backup(
-
- backup_state.speed = (has_speed && speed > 0) ? speed : 0;
-
-+ backup_state.perf = (BackupPerf){ .max_workers = 16 };
-+ if (has_max_workers) {
-+ backup_state.perf.max_workers = max_workers;
-+ }
-+
- backup_state.vmaw = vmaw;
- backup_state.pbs = pbs;
-
-@@ -1089,5 +1096,6 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
- ret->pbs_dirty_bitmap_migration = true;
- ret->query_bitmap_info = true;
- ret->pbs_masterkey = true;
-+ ret->backup_max_workers = true;
- return ret;
- }
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 89875f309c..43281aca79 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -833,6 +833,8 @@
- #
- # @encrypt: use encryption ((optional for format 'pbs', defaults to true if there is a keyfile)
- #
-+# @max-workers: see @BackupPerf for details. Default 16.
-+#
- # Returns: the uuid of the backup job
- #
- ##
-@@ -851,7 +853,9 @@
- '*format': 'BackupFormat',
- '*config-file': 'str',
- '*firewall-file': 'str',
-- '*devlist': 'str', '*speed': 'int' },
-+ '*devlist': 'str',
-+ '*speed': 'int',
-+ '*max-workers': 'int' },
- 'returns': 'UuidInfo', 'coroutine': true }
-
- ##
-@@ -906,7 +910,8 @@
- 'pbs-dirty-bitmap-savevm': 'bool',
- 'pbs-dirty-bitmap-migration': 'bool',
- 'pbs-masterkey': 'bool',
-- 'pbs-library-version': 'str' } }
-+ 'pbs-library-version': 'str',
-+ 'backup-max-workers': 'bool' } }
-
- ##
- # @query-proxmox-support:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fabian Ebner <f.ebner@proxmox.com>
+Date: Wed, 22 Jun 2022 10:45:13 +0200
+Subject: [PATCH] block: alloc-track: avoid premature break
+
+While the bdrv_co_preadv() calls are expected to return 0 on success,
+qemu_iovec_memset() will return the number of bytes set (will be
+local_bytes, because the slice with that size was just initialized).
+
+Don't break out of the loop after the branch with qemu_iovec_memset(),
+because there might still be work to do. Additionally, ret is an int,
+which on 64-bit platforms is too small to hold the size_t returned by
+qemu_iovec_memset().
+
+The branch seems to be difficult to reach in practice, because the
+whole point of alloc-track is to be used with a backing device.
+
+Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+---
+ block/alloc-track.c | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/block/alloc-track.c b/block/alloc-track.c
+index 6b50fbe537..c1160af04b 100644
+--- a/block/alloc-track.c
++++ b/block/alloc-track.c
+@@ -174,7 +174,8 @@ static int coroutine_fn track_co_preadv(BlockDriverState *bs,
+ ret = bdrv_co_preadv(bs->backing, local_offset, local_bytes,
+ &local_qiov, flags);
+ } else {
+- ret = qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
++ qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
++ ret = 0;
+ }
+
+ if (ret != 0) {
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Mon, 3 Oct 2022 15:52:04 +0200
+Subject: [PATCH] PVE Backup: allow passing max-workers performance setting
+
+For query-proxmox-support, add an indication that it's possible to use
+the setting.
+
+For now, the other two BackupPerf settings are not exposed:
+
+* use-copy-range: would need to be implemented by the backup-dump
+block driver first, and in fact, the default for backup was changed,
+because it wasn't as fast for backup in QEMU, see commit
+6a30f663d4c0b3c45a544d541e0c4e214b2473a1.
+
+* max-chunk: enforced to be at least the backup cluster size, which is
+4 MiB for PBS and otherwise maximum of source and target cluster size.
+And block-copy has a maximum buffer size of 1 MiB, so setting a larger
+max-chunk doesn't even have an effect. To make the setting sensibly
+usable the check would need to be removed and optionally the
+block-copy max buffer size would need to be bumped. I tried doing just
+that, and tested different source/target combinations with different
+max-chunk settings, but there were no noticable improvements over the
+default "unlimited" (resulting in 1 MiB for block-copy).
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ block/monitor/block-hmp-cmds.c | 4 +++-
+ pve-backup.c | 18 +++++++++++++-----
+ qapi/block-core.json | 9 +++++++--
+ 3 files changed, 23 insertions(+), 8 deletions(-)
+
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index 0502f42be6..cc231ec3f2 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1049,7 +1049,9 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
+ false, false, // PBS encrypt
+ true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
+ false, NULL, false, NULL, !!devlist,
+- devlist, qdict_haskey(qdict, "speed"), speed, &error);
++ devlist, qdict_haskey(qdict, "speed"), speed,
++ false, 0, // BackupPerf max-workers
++ &error);
+
+ hmp_handle_error(mon, error);
+ }
+diff --git a/pve-backup.c b/pve-backup.c
+index 2e22030eec..e9aa7e0f49 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -55,6 +55,7 @@ static struct PVEBackupState {
+ bool starting;
+ } stat;
+ int64_t speed;
++ BackupPerf perf;
+ VmaWriter *vmaw;
+ ProxmoxBackupHandle *pbs;
+ GList *di_list;
+@@ -492,8 +493,6 @@ static void create_backup_jobs_bh(void *opaque) {
+ }
+ backup_state.txn = job_txn_new_seq();
+
+- BackupPerf perf = { .max_workers = 16 };
+-
+ /* create and start all jobs (paused state) */
+ GList *l = backup_state.di_list;
+ while (l) {
+@@ -513,8 +512,9 @@ static void create_backup_jobs_bh(void *opaque) {
+
+ BlockJob *job = backup_job_create(
+ NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
+- bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
+- JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn, &local_err);
++ bitmap_mode, false, NULL, &backup_state.perf, BLOCKDEV_ON_ERROR_REPORT,
++ BLOCKDEV_ON_ERROR_REPORT, JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn,
++ &local_err);
+
+ di->job = job;
+ if (job) {
+@@ -584,7 +584,9 @@ UuidInfo coroutine_fn *qmp_backup(
+ bool has_config_file, const char *config_file,
+ bool has_firewall_file, const char *firewall_file,
+ bool has_devlist, const char *devlist,
+- bool has_speed, int64_t speed, Error **errp)
++ bool has_speed, int64_t speed,
++ bool has_max_workers, int64_t max_workers,
++ Error **errp)
+ {
+ assert(qemu_in_coroutine());
+
+@@ -914,6 +916,11 @@ UuidInfo coroutine_fn *qmp_backup(
+
+ backup_state.speed = (has_speed && speed > 0) ? speed : 0;
+
++ backup_state.perf = (BackupPerf){ .max_workers = 16 };
++ if (has_max_workers) {
++ backup_state.perf.max_workers = max_workers;
++ }
++
+ backup_state.vmaw = vmaw;
+ backup_state.pbs = pbs;
+
+@@ -1089,5 +1096,6 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+ ret->pbs_dirty_bitmap_migration = true;
+ ret->query_bitmap_info = true;
+ ret->pbs_masterkey = true;
++ ret->backup_max_workers = true;
+ return ret;
+ }
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index cc2ead0b75..e3f62faa81 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -829,6 +829,8 @@
+ #
+ # @encrypt: use encryption ((optional for format 'pbs', defaults to true if there is a keyfile)
+ #
++# @max-workers: see @BackupPerf for details. Default 16.
++#
+ # Returns: the uuid of the backup job
+ #
+ ##
+@@ -847,7 +849,9 @@
+ '*format': 'BackupFormat',
+ '*config-file': 'str',
+ '*firewall-file': 'str',
+- '*devlist': 'str', '*speed': 'int' },
++ '*devlist': 'str',
++ '*speed': 'int',
++ '*max-workers': 'int' },
+ 'returns': 'UuidInfo', 'coroutine': true }
+
+ ##
+@@ -902,7 +906,8 @@
+ 'pbs-dirty-bitmap-savevm': 'bool',
+ 'pbs-dirty-bitmap-migration': 'bool',
+ 'pbs-masterkey': 'bool',
+- 'pbs-library-version': 'str' } }
++ 'pbs-library-version': 'str',
++ 'backup-max-workers': 'bool' } }
+
+ ##
+ # @query-proxmox-support:
extra/0001-monitor-qmp-fix-race-with-clients-disconnecting-earl.patch
-extra/0002-block-gluster-correctly-set-max_pdiscard.patch
-extra/0003-block-vmdk-Fix-reopening-bs-file.patch
-extra/0004-linux-aio-fix-unbalanced-plugged-counter-in-laio_io_.patch
-extra/0005-pci-fix-overflow-in-snprintf-string-formatting.patch
-extra/0006-target-i386-kvm-Fix-disabling-MPX-on-cpu-host-with-M.patch
-extra/0007-coroutine-ucontext-use-QEMU_DEFINE_STATIC_CO_TLS.patch
-extra/0008-coroutine-use-QEMU_DEFINE_STATIC_CO_TLS.patch
-extra/0009-coroutine-Rename-qemu_coroutine_inc-dec_pool_size.patch
-extra/0010-coroutine-Revert-to-constant-batch-size.patch
-extra/0011-target-i386-do-not-consult-nonexistent-host-leaves.patch
-extra/0012-virtio-scsi-fix-ctrl-and-event-handler-functions-in-.patch
-extra/0013-virtio-scsi-don-t-waste-CPU-polling-the-event-virtqu.patch
-extra/0014-vhost-Track-descriptor-chain-in-private-at-SVQ.patch
-extra/0015-vhost-Fix-device-s-used-descriptor-dequeue.patch
-extra/0016-vdpa-Fix-bad-index-calculus-at-vhost_vdpa_get_vring_.patch
-extra/0017-vdpa-Fix-index-calculus-at-vhost_vdpa_svqs_start.patch
-extra/0018-hw-virtio-Replace-g_memdup-by-g_memdup2.patch
-extra/0019-vhost-Fix-element-in-vhost_svq_add-failure.patch
-extra/0020-io_uring-fix-short-read-slow-path.patch
-extra/0021-e1000-set-RX-descriptor-status-in-a-separate-operati.patch
bitmap-mirror/0001-drive-mirror-add-support-for-sync-bitmap-mode-never.patch
bitmap-mirror/0002-drive-mirror-add-support-for-conditional-and-always-.patch
bitmap-mirror/0003-mirror-add-check-for-bitmap-mode-without-bitmap.patch
pve/0013-PVE-virtio-balloon-improve-query-balloon.patch
pve/0014-PVE-qapi-modify-query-machines.patch
pve/0015-PVE-qapi-modify-spice-query.patch
-pve/0016-PVE-add-savevm-async-for-background-state-snapshots.patch
-pve/0017-PVE-add-optional-buffer-size-to-QEMUFile.patch
-pve/0018-PVE-block-add-the-zeroinit-block-driver-filter.patch
-pve/0019-PVE-Add-dummy-id-command-line-parameter.patch
-pve/0020-PVE-Config-Revert-target-i386-disable-LINT0-after-re.patch
-pve/0021-PVE-Up-Config-file-posix-make-locking-optiono-on-cre.patch
-pve/0022-PVE-monitor-disable-oob-capability.patch
-pve/0023-PVE-Compat-4.0-used-balloon-qemu-4-0-config-size-fal.patch
-pve/0024-PVE-Allow-version-code-in-machine-type.patch
-pve/0025-block-backup-move-bcs-bitmap-initialization-to-job-c.patch
-pve/0026-PVE-Backup-add-vma-backup-format-code.patch
-pve/0027-PVE-Backup-add-backup-dump-block-driver.patch
-pve/0028-PVE-Backup-proxmox-backup-patches-for-qemu.patch
-pve/0029-PVE-Backup-pbs-restore-new-command-to-restore-from-p.patch
-pve/0030-PVE-Backup-Add-dirty-bitmap-tracking-for-incremental.patch
-pve/0031-PVE-various-PBS-fixes.patch
-pve/0032-PVE-Add-PBS-block-driver-to-map-backup-archives-into.patch
-pve/0033-PVE-add-query_proxmox_support-QMP-command.patch
-pve/0034-PVE-add-query-pbs-bitmap-info-QMP-call.patch
-pve/0035-PVE-redirect-stderr-to-journal-when-daemonized.patch
-pve/0036-PVE-Add-sequential-job-transaction-support.patch
-pve/0037-PVE-Backup-Use-a-transaction-to-synchronize-job-stat.patch
-pve/0038-PVE-Backup-Don-t-block-on-finishing-and-cleanup-crea.patch
-pve/0039-PVE-Migrate-dirty-bitmap-state-via-savevm.patch
-pve/0040-migration-block-dirty-bitmap-migrate-other-bitmaps-e.patch
-pve/0041-PVE-fall-back-to-open-iscsi-initiatorname.patch
-pve/0042-PVE-Use-coroutine-QMP-for-backup-cancel_backup.patch
-pve/0043-PBS-add-master-key-support.patch
-pve/0044-PVE-block-pbs-fast-path-reads-without-allocation-if-.patch
-pve/0045-PVE-block-stream-increase-chunk-size.patch
-pve/0046-block-io-accept-NULL-qiov-in-bdrv_pad_request.patch
-pve/0047-block-add-alloc-track-driver.patch
-pve/0048-PVE-whitelist-invalid-QAPI-names-for-backwards-compa.patch
-pve/0049-PVE-savevm-async-register-yank-before-migration_inco.patch
-pve/0050-qemu-img-dd-add-l-option-for-loading-a-snapshot.patch
-pve/0051-vma-allow-partial-restore.patch
-pve/0052-pbs-namespace-support.patch
-pve/0053-Revert-block-rbd-workaround-for-ceph-issue-53784.patch
-pve/0054-Revert-block-rbd-fix-handling-of-holes-in-.bdrv_co_b.patch
-pve/0055-Revert-block-rbd-implement-bdrv_co_block_status.patch
-pve/0056-PVE-Backup-create-jobs-correctly-cancel-in-error-sce.patch
-pve/0057-PVE-Backup-ensure-jobs-in-di_list-are-referenced.patch
-pve/0058-PVE-Backup-avoid-segfault-issues-upon-backup-cancel.patch
-pve/0059-vma-create-support-64KiB-unaligned-input-images.patch
-pve/0060-vma-create-avoid-triggering-assertion-in-error-case.patch
-pve/0061-block-alloc-track-avoid-premature-break.patch
-pve/0062-PVE-Backup-allow-passing-max-workers-performance-set.patch
+pve/0016-PVE-add-IOChannel-implementation-for-savevm-async.patch
+pve/0017-PVE-add-savevm-async-for-background-state-snapshots.patch
+pve/0018-PVE-add-optional-buffer-size-to-QEMUFile.patch
+pve/0019-PVE-block-add-the-zeroinit-block-driver-filter.patch
+pve/0020-PVE-Add-dummy-id-command-line-parameter.patch
+pve/0021-PVE-Config-Revert-target-i386-disable-LINT0-after-re.patch
+pve/0022-PVE-Up-Config-file-posix-make-locking-optiono-on-cre.patch
+pve/0023-PVE-monitor-disable-oob-capability.patch
+pve/0024-PVE-Compat-4.0-used-balloon-qemu-4-0-config-size-fal.patch
+pve/0025-PVE-Allow-version-code-in-machine-type.patch
+pve/0026-block-backup-move-bcs-bitmap-initialization-to-job-c.patch
+pve/0027-PVE-Backup-add-vma-backup-format-code.patch
+pve/0028-PVE-Backup-add-backup-dump-block-driver.patch
+pve/0029-PVE-Backup-proxmox-backup-patches-for-qemu.patch
+pve/0030-PVE-Backup-pbs-restore-new-command-to-restore-from-p.patch
+pve/0031-PVE-Backup-Add-dirty-bitmap-tracking-for-incremental.patch
+pve/0032-PVE-various-PBS-fixes.patch
+pve/0033-PVE-Add-PBS-block-driver-to-map-backup-archives-into.patch
+pve/0034-PVE-add-query_proxmox_support-QMP-command.patch
+pve/0035-PVE-add-query-pbs-bitmap-info-QMP-call.patch
+pve/0036-PVE-redirect-stderr-to-journal-when-daemonized.patch
+pve/0037-PVE-Add-sequential-job-transaction-support.patch
+pve/0038-PVE-Backup-Use-a-transaction-to-synchronize-job-stat.patch
+pve/0039-PVE-Backup-Don-t-block-on-finishing-and-cleanup-crea.patch
+pve/0040-PVE-Migrate-dirty-bitmap-state-via-savevm.patch
+pve/0041-migration-block-dirty-bitmap-migrate-other-bitmaps-e.patch
+pve/0042-PVE-fall-back-to-open-iscsi-initiatorname.patch
+pve/0043-PVE-Use-coroutine-QMP-for-backup-cancel_backup.patch
+pve/0044-PBS-add-master-key-support.patch
+pve/0045-PVE-block-pbs-fast-path-reads-without-allocation-if-.patch
+pve/0046-PVE-block-stream-increase-chunk-size.patch
+pve/0047-block-io-accept-NULL-qiov-in-bdrv_pad_request.patch
+pve/0048-block-add-alloc-track-driver.patch
+pve/0049-PVE-whitelist-invalid-QAPI-names-for-backwards-compa.patch
+pve/0050-PVE-savevm-async-register-yank-before-migration_inco.patch
+pve/0051-qemu-img-dd-add-l-option-for-loading-a-snapshot.patch
+pve/0052-vma-allow-partial-restore.patch
+pve/0053-pbs-namespace-support.patch
+pve/0054-Revert-block-rbd-workaround-for-ceph-issue-53784.patch
+pve/0055-Revert-block-rbd-fix-handling-of-holes-in-.bdrv_co_b.patch
+pve/0056-Revert-block-rbd-implement-bdrv_co_block_status.patch
+pve/0057-PVE-Backup-create-jobs-correctly-cancel-in-error-sce.patch
+pve/0058-PVE-Backup-ensure-jobs-in-di_list-are-referenced.patch
+pve/0059-PVE-Backup-avoid-segfault-issues-upon-backup-cancel.patch
+pve/0060-vma-create-support-64KiB-unaligned-input-images.patch
+pve/0061-vma-create-avoid-triggering-assertion-in-error-case.patch
+pve/0062-block-alloc-track-avoid-premature-break.patch
+pve/0063-PVE-Backup-allow-passing-max-workers-performance-set.patch
-Subproject commit 823a3f11fb8f04c3c3cc0f95f968fef1bfc6534f
+Subproject commit 621da7789083b80d6f1ff1c0fb499334007b4f51