Add Vitastor support

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>

.gitignore

@@ -0,0 +1,7 @@
+/*.build
+/*.buildinfo
+/*.changes
+/*.deb
+/*.dsc
+/*.tar*
+/pve-qemu-kvm-*.*/

Makefile

@@ -1,38 +1,77 @@
-include /usr/share/dpkg/pkg-info.mk
-include /usr/share/dpkg/architecture.mk
+include /usr/share/dpkg/default.mk
 
 PACKAGE = pve-qemu-kvm
 
 SRCDIR := qemu
-BUILDDIR ?= ${PACKAGE}-${DEB_VERSION_UPSTREAM}
+BUILDDIR ?= $(PACKAGE)-$(DEB_VERSION_UPSTREAM)
+ORIG_SRC_TAR=$(PACKAGE)_$(DEB_VERSION_UPSTREAM).orig.tar.gz
 
 GITVERSION := $(shell git rev-parse HEAD)
 
-DEB = ${PACKAGE}_${DEB_VERSION_UPSTREAM_REVISION}_${DEB_BUILD_ARCH}.deb
-DEB_DBG = ${PACKAGE}-dbg_${DEB_VERSION_UPSTREAM_REVISION}_${DEB_BUILD_ARCH}.deb
+DSC=$(PACKAGE)_$(DEB_VERSION_UPSTREAM_REVISION).dsc
+DEB = $(PACKAGE)_$(DEB_VERSION_UPSTREAM_REVISION)_$(DEB_BUILD_ARCH).deb
+DEB_DBG = $(PACKAGE)-dbg_$(DEB_VERSION_UPSTREAM_REVISION)_$(DEB_BUILD_ARCH).deb
 DEBS = $(DEB) $(DEB_DBG)
 
 all: $(DEBS)
 
 .PHONY: submodule
 submodule:
-	test -f "${SRCDIR}/configure" || git submodule update --init --recursive
+	test -f "$(SRCDIR)/configure" || git submodule update --init --recursive
+
+PC_BIOS_FW_PURGE_LIST_IN = \
+	hppa-firmware.img \
+	openbios-ppc \
+	openbios-sparc32 \
+	openbios-sparc64 \
+	palcode-clipper \
+	s390-ccw.img \
+	s390-netboot.img \
+	u-boot.e500 \
+	.*\.dtb \
+	qemu_vga.ndrv \
+	slof.bin \
+	opensbi-riscv.*-generic-fw_dynamic.bin \
+
+BLOB_PURGE_SED_CMDS = $(foreach FILE,$(PC_BIOS_FW_PURGE_LIST_IN),-e "/$(FILE)/d")
+BLOB_PURGE_FILTER = $(foreach FILE,$(PC_BIOS_FW_PURGE_LIST_IN),-e "$(FILE)")
 
 $(BUILDDIR): keycodemapdb | submodule
-	rm -rf $(BUILDDIR)
-	cp -a $(SRCDIR) $(BUILDDIR)
-	cp -a debian $(BUILDDIR)/debian
-	rm -rf $(BUILDDIR)/ui/keycodemapdb
-	cp -a keycodemapdb $(BUILDDIR)/ui/
-	echo "git clone git://git.proxmox.com/git/pve-qemu.git\\ngit checkout $(GITVERSION)" > $(BUILDDIR)/debian/SOURCE
+	# check if qemu/ was used for a build
+	# if so, please run 'make distclean' in the submodule and try again
+	test ! -f $(SRCDIR)/build/config.status
+	rm -rf $@.tmp $@
+	cp -a $(SRCDIR) $@.tmp
+	cp -a debian $@.tmp/debian
+	rm -rf $@.tmp/ui/keycodemapdb
+	cp -a keycodemapdb $@.tmp/ui/
+	find $@.tmp/pc-bios -type f | grep $(BLOB_PURGE_FILTER) | xargs rm -f
+	sed -i $(BLOB_PURGE_SED_CMDS) $@.tmp/pc-bios/meson.build
+	echo "git clone git://git.proxmox.com/git/pve-qemu.git\\ngit checkout $(GITVERSION)" > $@.tmp/debian/SOURCE
+	mv $@.tmp $@
 
 .PHONY: deb kvm
 deb kvm: $(DEBS)
 $(DEB_DBG): $(DEB)
 $(DEB): $(BUILDDIR)
-	cd $(BUILDDIR); dpkg-buildpackage -b -us -uc -j
+	cd $(BUILDDIR); dpkg-buildpackage -b -us -uc -j32
 	lintian $(DEBS)
 
+sbuild: $(DSC)
+	sbuild $(DSC)
+
+$(ORIG_SRC_TAR): $(BUILDDIR)
+	tar czf $(ORIG_SRC_TAR) --exclude="$(BUILDDIR)/debian" $(BUILDDIR)
+
+.PHONY: dsc
+dsc:
+	rm -rf *.dsc $(BUILDDIR)
+	$(MAKE) $(DSC)
+	lintian $(DSC)
+
+$(DSC): $(ORIG_SRC_TAR) $(BUILDDIR)
+	cd $(BUILDDIR); dpkg-buildpackage -S -us -uc -d
+
 .PHONY: update
 update:
 	cd $(SRCDIR) && git submodule deinit ui/keycodemapdb || true
@@ -45,13 +84,14 @@ update:
 	git add keycodemapdb
 
 .PHONY: upload
+upload: UPLOAD_DIST ?= $(DEB_DISTRIBUTION)
 upload: $(DEBS)
-	tar cf - ${DEBS} | ssh repoman@repo.proxmox.com upload --product pve --dist buster
+	tar cf - $(DEBS) | ssh repoman@repo.proxmox.com upload --product pve --dist $(UPLOAD_DIST)
 
 .PHONY: distclean clean
 distclean: clean
 clean:
-	rm -rf $(BUILDDIR) $(PACKAGE)*.deb *.buildinfo *.changes
+	rm -rf $(PACKAGE)-[0-9]*/ $(PACKAGE)*.tar* *.deb *.dsc *.build *.buildinfo *.changes
 
 .PHONY: dinstall
 dinstall: $(DEBS)

debian/changelog

@@ -1,8 +1,381 @@
-pve-qemu-kvm (5.1.0-8+vitastor1) buster; urgency=medium
+pve-qemu-kvm (7.2.10-1+vitastor1) bullseye; urgency=medium
 
   * Add Vitastor support
 
- -- Vitaliy Filippov <vitalif@yourcmc.ru>  Sun, 05 Dec 2021 13:46:51 +0300
+ -- Vitaliy Filippov <vitalif@yourcmc.ru>  Sat, 22 Mar 2025 11:41:12 +0300
+
+pve-qemu-kvm (7.2.10-1) bullseye; urgency=medium
+
+  * update patches and submodule to QEMU stable 7.2.10
+
+  * pick up some extra fixes from upcoming 7.2.11 to, e.g., avoid releasing a
+    regression for i386 that happened in the 7.2.10 stable release.
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 10 Apr 2024 15:47:27 +0200
+
+pve-qemu-kvm (7.2.0-8) bullseye; urgency=medium
+
+  * backport fix for ACPI CPU hotplug issue with TCG
+
+  * cherry-pick TCG-related stable fixes for 7.2 for users that turned off KVM
+    HW acceleration
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 17 Mar 2023 15:47:08 +0100
+
+pve-qemu-kvm (7.2.0-7) bullseye; urgency=medium
+
+  * improve fix for potential deadlock with trim for IDE/SATA and draining
+
+  * backport stable fixes:
+    - hw/nvme: fix missing endian conversions for doorbell buffers
+    - hw/smbios: fix field corruption in type 4 table
+    - virtio-rng-pci: fix transitional migration compat for vectors
+    - hw/timer/hpet: Fix expiration time overflow
+    - vhost/vdpa: stop all svq on device deletion
+    - vhost: avoid a potential use of an uninitialized variable in the call to
+      vhost_svq_poll
+    - chardev/char-socket: set s->listener = NULL in char_socket_finalize to
+      fix a potential crash after live-migration
+    - intel-iommu: fail MAP notifier without caching mode
+    - intel-iommu: fail DEVIOTLB_UNMAP without dt mode
+
+  * fix a regression for when the LSI SCSI controller is used
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 13 Mar 2023 17:42:49 +0100
+
+pve-qemu-kvm (7.2.0-6) bullseye; urgency=medium
+
+  * fix 7.2 regression for Linux boot failures with megasas SCSI
+
+  * fix 7.0 regression for a potential deadlock with trim for IDE/SATA and
+    draining
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 08 Mar 2023 14:32:17 +0100
+
+pve-qemu-kvm (7.2.0-5) bullseye; urgency=medium
+
+  * fix #4476: savevm-async: avoid looping without progress
+
+  * savevm-async: decrease the boundary for free space for (memory) state left
+    on target from 30 MiB to 100 MiB, improving the heuristic for when to
+    enter the final "pause and sync" stage.
+
+  * QMP backup: use correct error number when getting blockdrive length fails
+
+  * backport fix for some DMA reentrancy issues, better protecting against
+    malicious guests
+
+  * backport fix for iSCSI double free issue leading to crashes
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 21 Feb 2023 13:49:43 +0100
+
+pve-qemu-kvm (7.2.0-4) bullseye; urgency=medium
+
+  * backport fix for a 7.2 regression when using VirtIO disk with
+    detect-zeroes=unmap
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 27 Jan 2023 09:37:49 +0100
+
+pve-qemu-kvm (7.2.0-3) bullseye; urgency=medium
+
+  * add fix for live-migration with virtio-rng devices, which regressed in
+    QEMU 7.2.0.
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 12 Jan 2023 13:13:14 +0100
+
+pve-qemu-kvm (7.2.0-2) bullseye; urgency=medium
+
+  * enable slirp again for now, as in qemu-server, user networking is
+    supported (via CLI/API) when no bridge is set on a virtual NIC
+
+  * cherry-pick stable fixes for 7.2. Two for virtio-mem and one for vIOMMU.
+    Both features are not yet exposed in PVE's qemu-server, but there's work
+    going on to change that.
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 10 Jan 2023 15:47:48 +0100
+
+pve-qemu-kvm (7.2.0-1) bullseye; urgency=medium
+
+  * update to QEMU stable release 7.2.0
+
+  * drop 'slirp' networking
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 16 Dec 2022 13:18:21 +0100
+
+pve-qemu-kvm (7.1.0-4) bullseye; urgency=medium
+
+  * cherry-pick "block/block-backend: blk_set_enable_write_cache is IO_CODE"
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 22 Nov 2022 09:21:06 +0100
+
+pve-qemu-kvm (7.1.0-3) bullseye; urgency=medium
+
+  * init: daemonize: defuse PID file resolve error to a warning at max, fixing
+    some usecases that regressed with 7.1, like tracking start up in our
+    file-restore VM.
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 28 Oct 2022 10:27:49 +0200
+
+pve-qemu-kvm (7.1.0-2) bullseye; urgency=medium
+
+  * fix an issue with error handling in async backup code
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 18 Oct 2022 15:33:44 +0200
+
+pve-qemu-kvm (7.1.0-1) bullseye; urgency=medium
+
+  * update to QEMU stable release 7.1.0
+
+  * add fix for io_uring_register_ring_fd from upstream
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 14 Oct 2022 14:54:09 +0200
+
+pve-qemu-kvm (7.0.0-4) bullseye; urgency=medium
+
+  * add revision to version output
+
+  * PVE Backup: allow passing max-workers performance setting
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 10 Oct 2022 11:55:37 +0200
+
+pve-qemu-kvm (7.0.0-3) bullseye; urgency=medium
+
+  * savevm-async: avoid segfault when aborting snapshot creation task
+
+  * savevm-async: set SAVE_STATE_DONE when closing state file was successful
+    allowing one to start a new snapshot task after aborting one.
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 30 Aug 2022 12:54:03 +0200
+
+pve-qemu-kvm (7.0.0-2) bullseye; urgency=medium
+
+  * backport "io_uring: fix short read slow path"
+
+  * backport "e1000: set RX descriptor status in a separate operation"
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 20 Jul 2022 09:17:07 +0200
+
+pve-qemu-kvm (7.0.0-1) bullseye; urgency=medium
+
+  * update to QEMU stable release 7.0.0
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 30 Jun 2022 11:07:37 +0200
+
+pve-qemu-kvm (6.2.0-11) bullseye; urgency=medium
+
+  * add 'namespace' to BlockdevOptionsPbs for live-restore support
+
+  * vma: create: support 64KiB-unaligned input images like to improve backing
+    up some VM templates
+
+  * block: alloc-track: avoid unlikely, but possible premature break
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 22 Jun 2022 15:54:54 +0200
+
+pve-qemu-kvm (6.2.0-10) bullseye; urgency=medium
+
+  * fix #4101: fix backup cancellation bug with iothreads
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 9 Jun 2022 16:35:51 +0200
+
+pve-qemu-kvm (6.2.0-9) bullseye; urgency=medium
+
+  * fix possible race conditions during cancellation of a PBS backup
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 08 Jun 2022 14:03:22 +0200
+
+pve-qemu-kvm (6.2.0-8) bullseye; urgency=medium
+
+  * revert "block/rbd: implement bdrv_co_block_status" to work around
+    performance regression when backing up large RBD disk
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 19 May 2022 09:24:45 +0200
+
+pve-qemu-kvm (6.2.0-7) bullseye; urgency=medium
+
+  * Proxmox Backup Server namespace support
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 12 May 2022 16:05:56 +0200
+
+pve-qemu-kvm (6.2.0-6) bullseye; urgency=medium
+
+  * block/gluster: correctly set max_pdiscard which is int64_t to avoid
+    triggering assertion
+
+  * ui/vnc.c: Fixed a deadlock bug
+
+  * display/qxl-render: fix race condition in qxl_cursor (CVE-2021-4207) and
+    integer overflow in cursor_alloc (CVE-2021-4206)
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 11 May 2022 10:42:53 +0200
+
+pve-qemu-kvm (6.2.0-5) bullseye; urgency=medium
+
+  * vma: allow partial restore by skipping some disk
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 25 Apr 2022 10:13:46 +0200
+
+pve-qemu-kvm (6.2.0-4) bullseye; urgency=medium
+
+  * d/control: add libgbm to build dependencies
+
+  * d/control: add suggest dependency-hint for libgl1
+
+  * various stable backports:
+    + virtio-net: fix map leaking on error during receive
+    + memory: Fix incorrect calls of log_global_start/stop
+    + acpi: fix OEM ID/OEM Table ID padding
+    + vhost-vsock: detach the virqueue element in case of error
+    + vhost-user: remove VirtQ notifier restore
+    + vhost-user: fix VirtQ notifier cleanup
+    + virtio: fix the condition for iommu_platform not supported
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 22 Apr 2022 11:52:30 +0200
+
+pve-qemu-kvm (6.2.0-3) bullseye; urgency=medium
+
+  *  cherry-pick fix for some manually added ACPI table SLIC entries via the
+     custom args flag.
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 15 Apr 2022 09:09:37 +0200
+
+pve-qemu-kvm (6.2.0-2) bullseye; urgency=medium
+
+  * compile in virgl support
+
+  * enable zstd support
+
+  * drop sdl dependency (it was disabled at compile time already)
+
+  * recommend 'numactl'
+
+  * fix an issue with multi-disk backups where chunks would be written
+    multiple times
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 03 Mar 2022 12:03:44 +0100
+
+pve-qemu-kvm (6.2.0-1) bullseye; urgency=medium
+
+  * update to QEMU stable release 6.2.0
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 17 Feb 2022 06:23:14 +0100
+
+pve-qemu-kvm (6.1.1-2) bullseye; urgency=medium
+
+  * vma: create: register all streams before entering coroutines to avoid that
+    an early stream starts to write already before all got registered.
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 14 Feb 2022 15:53:15 +0100
+
+pve-qemu-kvm (6.1.1-1) bullseye; urgency=medium
+
+  * update to 6.1.1 stable release
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 13 Jan 2022 10:57:43 +0100
+
+pve-qemu-kvm (6.1.0-3) bullseye; urgency=medium
+
+  * fix #3738: cherry-pick "block: introduce max_hw_iov for use in scsi-
+    generic
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 01 Dec 2021 15:35:43 +0100
+
+pve-qemu-kvm (6.1.0-2) bullseye; urgency=medium
+
+  * avoid a possible segmentation fault during block (disk) mirror
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 16 Nov 2021 09:38:10 +0100
+
+pve-qemu-kvm (6.1.0-1) bullseye; urgency=medium
+
+  * update to QEMU stable release 6.1.0
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 11 Oct 2021 15:15:19 +0200
+
+pve-qemu-kvm (6.0.0-4) bullseye; urgency=medium
+
+  * drop the ancient workaround that force disabled SMM due to observing VM
+    hangs on old kernel versions.
+
+  * monitor/qmp: fix race with clients disconnecting early resulting in other
+    clients receiving a message with the (now wrong) ID of the former
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 06 Sep 2021 07:30:00 +0200
+
+pve-qemu-kvm (6.0.0-3) bullseye; urgency=medium
+
+  * io_uring: resubmit when result is -EAGAIN
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 3 Aug 2021 15:01:31 +0200
+
+pve-qemu-kvm (6.0.0-2) bullseye; urgency=medium
+
+  * enable io-uring support in QEMU builds
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 23 Jun 2021 11:03:54 +0200
+
+pve-qemu-kvm (6.0.0-1) bullseye; urgency=medium
+
+  * update to QEMU stable release 6.0.0
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 28 May 2021 11:30:50 +0200
+
+pve-qemu-kvm (5.2.0-11) bullseye; urgency=medium
+
+  * re-build for Proxmox VE 7 / Debian Bullseye
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 13 May 2021 14:03:00 +0200
+
+pve-qemu-kvm (5.2.0-6) pve; urgency=medium
+
+  * improve the alloc-track and Proxmox Backup Server special block driver when
+    used with IO-threads for the new live-restore feature
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 15 Apr 2021 16:29:48 +0200
+
+pve-qemu-kvm (5.2.0-5) pve; urgency=medium
+
+  * cherry-pick fixes for a possible deadlock when resizing a disk while using
+    IO-threads
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 30 Mar 2021 18:18:18 +0200
+
+pve-qemu-kvm (5.2.0-4) pve; urgency=medium
+
+  * monitor/qmp: fix race on CHR_EVENT_CLOSED without OOB
+
+  * improve saving and loading dirty bitmaps in live-snapshots
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 23 Mar 2021 15:41:26 +0100
+
+pve-qemu-kvm (5.2.0-3) pve; urgency=medium
+
+  * backport "i386/acpi: restore device paths for pre-5.1 vms" patch
+
+  * ship list of 'i440fx' and 'q35' machine versions this QEMU version supports
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 05 Mar 2021 16:23:06 +0100
+
+pve-qemu-kvm (5.2.0-2) pve; urgency=medium
+
+  * Proxmox Backup restore: ensure all caches are flushed for userspace
+    attached storages (for example, Ceph when librbd and not KRBD is used)
+
+  * ship VirtIOFSd daemon
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 24 Feb 2021 18:25:14 +0100
+
+pve-qemu-kvm (5.2.0-1) pve; urgency=medium
+
+  * update to QEMU stable release 5.2.0
+
+  * fix #3084: fall back to open-iscsi initiatorname
+
+  * fix snapshot abort and improve performance in some edge cases
+
+  * add basis for Proxmox Backup Server master key support
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 12 Feb 2021 12:09:59 +0100
 
 pve-qemu-kvm (5.1.0-8) pve; urgency=medium
 

debian/control

@@ -10,26 +10,34 @@ Build-Depends: autotools-dev,
                libattr1-dev,
                libcap-ng-dev,
                libcurl4-gnutls-dev,
+               libepoxy-dev,
                libfdt-dev,
+               libgbm-dev,
                libglusterfs-dev (>= 5.2-2),
                libgnutls28-dev,
                libiscsi-dev (>= 1.12.0),
-               libjemalloc-dev,
                libjpeg-dev,
+               libjson-perl,
                libnuma-dev,
                libpci-dev,
                libpixman-1-dev,
-               libproxmox-backup-qemu0-dev (>= 1.0.2-1),
+               libproxmox-backup-qemu0-dev (>= 1.3.0),
                librbd-dev (>= 0.48),
                libsdl1.2-dev,
                libseccomp-dev,
+               libslirp-dev,
                libspice-protocol-dev (>= 0.12.14~),
                libspice-server-dev (>= 0.14.0~),
                libsystemd-dev,
-               libusb-1.0-0-dev (>= 1.0.17-1),
+               liburing-dev,
+               libusb-1.0-0-dev (>= 1.0.17),
                libusbredirparser-dev (>= 0.6-2),
+               libvirglrenderer-dev,
+               libzstd-dev,
+               meson,
                python3-minimal,
                python3-sphinx,
+               python3-sphinx-rtd-theme,
                quilt,
                texi2html,
                texinfo,
@@ -41,7 +49,6 @@ Package: pve-qemu-kvm
 Architecture: any
 Depends: ceph-common (>= 0.48),
          iproute2,
-         libaio1,
          libgfapi0 | glusterfs-common (>= 5.6),
          libgfchangelog0 | glusterfs-common (>= 5.6),
          libgfdb0 | glusterfs-common (>= 5.6),
@@ -50,18 +57,15 @@ Depends: ceph-common (>= 0.48),
          libglusterfs-dev | glusterfs-common (>= 5.6),
          libglusterfs0 | glusterfs-common (>= 5.6),
          libiscsi4 (>= 1.12.0) | libiscsi7,
-         libjemalloc2,
          libjpeg62-turbo,
-         libproxmox-backup-qemu0 (>= 1.0.2-1),
-         libsdl1.2debian,
          libspice-server1 (>= 0.14.0~),
          libusb-1.0-0 (>= 1.0.17-1),
          libusbredirparser1 (>= 0.6-2),
          libuuid1,
-         numactl,
-         python,
          ${misc:Depends},
          ${shlibs:Depends},
+Recommends: numactl,
+Suggests: libgl1,
 Conflicts: kvm,
            pve-kvm,
            pve-qemu-kvm-2.6.18,
@@ -69,9 +73,10 @@ Conflicts: kvm,
            qemu-kvm,
            qemu-system-arm,
            qemu-system-common,
+           qemu-system-data,
            qemu-system-x86,
            qemu-utils,
-Provides: qemu-system-arm, qemu-system-x86, qemu-utils
+Provides: qemu-system-arm, qemu-system-x86, qemu-utils,
 Replaces: pve-kvm,
           pve-qemu-kvm-2.6.18,
           qemu-system-arm,
@@ -85,6 +90,6 @@ Description: Full virtualization on x86 hardware
 Package: pve-qemu-kvm-dbg
 Architecture: any
 Section: debug
-Depends: pve-qemu-kvm (= ${binary:Version})
+Depends: pve-qemu-kvm (= ${binary:Version}),
 Description: pve qemu debugging symbols
  This package contains the debugging symbols for pve-qemu-kvm.

debian/copyright

@@ -25,7 +25,7 @@ License:
 
     In particular, the QEMU virtual CPU core library (libqemu.a) is
     released under the GNU Lesser General Public License version 2 or later.
-    On Debian systems, the complete text of the GNU Lesser General Public 
+    On Debian systems, the complete text of the GNU Lesser General Public
     License can be found in the file /usr/share/common-licenses/LGPL.
 
     Some hardware device emulation sources and other QEMU functionality are

debian/parse-machines.pl

@@ -0,0 +1,28 @@
+#!/usr/bin/perl
+
+use warnings;
+use strict;
+
+use JSON;
+
+my $machines = [];
+
+while (<STDIN>) {
+    if (/^\s*Supported machines are:/) {
+	next;
+    }
+
+    s/^\s+//;
+    my @machine = split(/\s+/);
+    next if $machine[0] !~ m/^pc-(i440fx|q35)-(.+)$/;
+    push @$machines, {
+        'id' => $machine[0],
+        'type' => $1,
+        'version' => $2,
+    };
+}
+
+die "no QEMU machine types detected from STDIN input" if scalar (@$machines) <= 0;
+
+print to_json($machines, { utf8 => 1, canonical => 1 })
+    or die "failed to encode detected machines as JSON - $!\n";

debian/patches/bitmap-mirror/0001-drive-mirror-add-support-for-sync-bitmap-mode-never.patch

@@ -26,19 +26,20 @@ Suggested-by: Ma Haocong <mahaocong@didichuxing.com>
 Signed-off-by: Ma Haocong <mahaocong@didichuxing.com>
 Signed-off-by: John Snow <jsnow@redhat.com>
 Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
- block/mirror.c              | 98 ++++++++++++++++++++++++++++++-------
- blockdev.c                  | 39 +++++++++++++--
- include/block/block_int.h   |  4 +-
- qapi/block-core.json        | 29 +++++++++--
- tests/test-block-iothread.c |  4 +-
+ block/mirror.c                         | 98 +++++++++++++++++++++-----
+ blockdev.c                             | 39 +++++++++-
+ include/block/block_int-global-state.h |  4 +-
+ qapi/block-core.json                   | 29 ++++++--
+ tests/unit/test-block-iothread.c       |  4 +-
  5 files changed, 145 insertions(+), 29 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index e8e8844afc..100e828639 100644
+index 251adc5ae0..8ead5f77a0 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -49,7 +49,7 @@ typedef struct MirrorBlockJob {
+@@ -51,7 +51,7 @@ typedef struct MirrorBlockJob {
      BlockDriverState *to_replace;
      /* Used to block operations on the drive-mirror-replace target */
      Error *replace_blocker;
@@ -47,7 +48,7 @@ index e8e8844afc..100e828639 100644
      BlockMirrorBackingMode backing_mode;
      /* Whether the target image requires explicit zero-initialization */
      bool zero_target;
-@@ -64,6 +64,8 @@ typedef struct MirrorBlockJob {
+@@ -65,6 +65,8 @@ typedef struct MirrorBlockJob {
      size_t buf_size;
      int64_t bdev_length;
      unsigned long *cow_bitmap;
@@ -56,17 +57,17 @@ index e8e8844afc..100e828639 100644
      BdrvDirtyBitmap *dirty_bitmap;
      BdrvDirtyBitmapIter *dbi;
      uint8_t *buf;
-@@ -676,7 +678,8 @@ static int mirror_exit_common(Job *job)
+@@ -699,7 +701,8 @@ static int mirror_exit_common(Job *job)
      bdrv_child_refresh_perms(mirror_top_bs, mirror_top_bs->backing,
                               &error_abort);
      if (!abort && s->backing_mode == MIRROR_SOURCE_BACKING_CHAIN) {
 -        BlockDriverState *backing = s->is_none_mode ? src : s->base;
 +        BlockDriverState *backing;
 +        backing = s->sync_mode == MIRROR_SYNC_MODE_NONE ? src : s->base;
-         if (backing_bs(target_bs) != backing) {
-             bdrv_set_backing_hd(target_bs, backing, &local_err);
-             if (local_err) {
-@@ -771,6 +774,16 @@ static void mirror_abort(Job *job)
+         BlockDriverState *unfiltered_target = bdrv_skip_filters(target_bs);
+ 
+         if (bdrv_cow_bs(unfiltered_target) != backing) {
+@@ -797,6 +800,16 @@ static void mirror_abort(Job *job)
      assert(ret == 0);
  }
  
@@ -83,7 +84,7 @@ index e8e8844afc..100e828639 100644
  static void coroutine_fn mirror_throttle(MirrorBlockJob *s)
  {
      int64_t now = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
-@@ -952,7 +965,8 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
+@@ -977,7 +990,8 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
      mirror_free_init(s);
  
      s->last_pause_ns = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
@@ -93,23 +94,23 @@ index e8e8844afc..100e828639 100644
          ret = mirror_dirty_init(s);
          if (ret < 0 || job_is_cancelled(&s->common.job)) {
              goto immediate_exit;
-@@ -1184,6 +1198,7 @@ static const BlockJobDriver mirror_job_driver = {
+@@ -1224,6 +1238,7 @@ static const BlockJobDriver mirror_job_driver = {
          .run                    = mirror_run,
          .prepare                = mirror_prepare,
          .abort                  = mirror_abort,
 +        .clean                  = mirror_clean,
          .pause                  = mirror_pause,
          .complete               = mirror_complete,
-     },
-@@ -1199,6 +1214,7 @@ static const BlockJobDriver commit_active_job_driver = {
+         .cancel                 = mirror_cancel,
+@@ -1240,6 +1255,7 @@ static const BlockJobDriver commit_active_job_driver = {
          .run                    = mirror_run,
          .prepare                = mirror_prepare,
          .abort                  = mirror_abort,
 +        .clean                  = mirror_clean,
          .pause                  = mirror_pause,
          .complete               = mirror_complete,
-     },
-@@ -1547,7 +1563,10 @@ static BlockJob *mirror_start_job(
+         .cancel                 = commit_active_cancel,
+@@ -1627,7 +1643,10 @@ static BlockJob *mirror_start_job(
                               BlockCompletionFunc *cb,
                               void *opaque,
                               const BlockJobDriver *driver,
@@ -121,8 +122,8 @@ index e8e8844afc..100e828639 100644
                               bool auto_complete, const char *filter_node_name,
                               bool is_mirror, MirrorCopyMode copy_mode,
                               Error **errp)
-@@ -1560,10 +1579,39 @@ static BlockJob *mirror_start_job(
-     Error *local_err = NULL;
+@@ -1639,10 +1658,39 @@ static BlockJob *mirror_start_job(
+     uint64_t target_perms, target_shared_perms;
      int ret;
  
 -    if (granularity == 0) {
@@ -163,7 +164,7 @@ index e8e8844afc..100e828639 100644
      assert(is_power_of_2(granularity));
  
      if (buf_size < 0) {
-@@ -1667,7 +1715,9 @@ static BlockJob *mirror_start_job(
+@@ -1774,7 +1822,9 @@ static BlockJob *mirror_start_job(
      s->replaces = g_strdup(replaces);
      s->on_source_error = on_source_error;
      s->on_target_error = on_target_error;
@@ -174,7 +175,7 @@ index e8e8844afc..100e828639 100644
      s->backing_mode = backing_mode;
      s->zero_target = zero_target;
      s->copy_mode = copy_mode;
-@@ -1687,6 +1737,18 @@ static BlockJob *mirror_start_job(
+@@ -1795,6 +1845,18 @@ static BlockJob *mirror_start_job(
          bdrv_disable_dirty_bitmap(s->dirty_bitmap);
      }
  
@@ -193,7 +194,7 @@ index e8e8844afc..100e828639 100644
      ret = block_job_add_bdrv(&s->common, "source", bs, 0,
                               BLK_PERM_WRITE_UNCHANGED | BLK_PERM_WRITE |
                               BLK_PERM_CONSISTENT_READ,
-@@ -1740,6 +1802,9 @@ fail:
+@@ -1872,6 +1934,9 @@ fail:
          if (s->dirty_bitmap) {
              bdrv_release_dirty_bitmap(s->dirty_bitmap);
          }
@@ -203,7 +204,7 @@ index e8e8844afc..100e828639 100644
          job_early_fail(&s->common.job);
      }
  
-@@ -1757,29 +1822,23 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
+@@ -1889,31 +1954,25 @@ 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,
@@ -220,6 +221,8 @@ index e8e8844afc..100e828639 100644
 -    bool is_none_mode;
      BlockDriverState *base;
  
+     GLOBAL_STATE_CODE();
+ 
 -    if ((mode == MIRROR_SYNC_MODE_INCREMENTAL) ||
 -        (mode == MIRROR_SYNC_MODE_BITMAP)) {
 -        error_setg(errp, "Sync mode '%s' not supported",
@@ -227,7 +230,7 @@ index e8e8844afc..100e828639 100644
 -        return;
 -    }
 -    is_none_mode = mode == MIRROR_SYNC_MODE_NONE;
-     base = mode == MIRROR_SYNC_MODE_TOP ? backing_bs(bs) : NULL;
+     base = mode == MIRROR_SYNC_MODE_TOP ? bdrv_backing_chain_next(bs) : NULL;
      mirror_start_job(job_id, bs, creation_flags, target, replaces,
                       speed, granularity, buf_size, backing_mode, zero_target,
                       on_source_error, on_target_error, unmap, NULL, NULL,
@@ -238,7 +241,7 @@ index e8e8844afc..100e828639 100644
  }
  
  BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
-@@ -1805,7 +1864,8 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
+@@ -1940,7 +1999,8 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
                       job_id, bs, creation_flags, base, NULL, speed, 0, 0,
                       MIRROR_LEAVE_BACKING_CHAIN, false,
                       on_error, on_error, true, cb, opaque,
@@ -246,13 +249,13 @@ index e8e8844afc..100e828639 100644
 +                     &commit_active_job_driver, MIRROR_SYNC_MODE_FULL,
 +                     NULL, 0, base, auto_complete,
                       filter_node_name, false, MIRROR_COPY_MODE_BACKGROUND,
-                      &local_err);
-     if (local_err) {
+                      errp);
+     if (!job) {
 diff --git a/blockdev.c b/blockdev.c
-index 681da7c8b6..02d58e7645 100644
+index ae27a41efa..a0c7e0c13b 100644
 --- a/blockdev.c
 +++ b/blockdev.c
-@@ -2876,6 +2876,10 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -2956,6 +2956,10 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
                                     BlockDriverState *target,
                                     bool has_replaces, const char *replaces,
                                     enum MirrorSyncMode sync,
@@ -263,15 +266,15 @@ index 681da7c8b6..02d58e7645 100644
                                     BlockMirrorBackingMode backing_mode,
                                     bool zero_target,
                                     bool has_speed, int64_t speed,
-@@ -2894,6 +2898,7 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
-                                    Error **errp)
+@@ -2975,6 +2979,7 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
  {
+     BlockDriverState *unfiltered_bs;
      int job_flags = JOB_DEFAULT;
 +    BdrvDirtyBitmap *bitmap = NULL;
  
      if (!has_speed) {
          speed = 0;
-@@ -2948,6 +2953,29 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -3029,6 +3034,29 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
          sync = MIRROR_SYNC_MODE_FULL;
      }
  
@@ -298,10 +301,10 @@ index 681da7c8b6..02d58e7645 100644
 +        }
 +    }
 +
-     if (has_replaces) {
-         BlockDriverState *to_replace_bs;
-         AioContext *replace_aio_context;
-@@ -2985,8 +3013,8 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+     if (!has_replaces) {
+         /* We want to mirror from @bs, but keep implicit filters on top */
+         unfiltered_bs = bdrv_skip_implicit_filters(bs);
+@@ -3075,8 +3103,8 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
       * and will allow to check whether the node still exist at mirror completion
       */
      mirror_start(job_id, bs, target,
@@ -312,7 +315,7 @@ index 681da7c8b6..02d58e7645 100644
                   on_source_error, on_target_error, unmap, filter_node_name,
                   copy_mode, errp);
  }
-@@ -3127,6 +3155,8 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
+@@ -3221,6 +3249,8 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
  
      blockdev_mirror_common(arg->has_job_id ? arg->job_id : NULL, bs, target_bs,
                             arg->has_replaces, arg->replaces, arg->sync,
@@ -321,7 +324,7 @@ index 681da7c8b6..02d58e7645 100644
                             backing_mode, zero_target,
                             arg->has_speed, arg->speed,
                             arg->has_granularity, arg->granularity,
-@@ -3148,6 +3178,8 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
+@@ -3242,6 +3272,8 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
                           const char *device, const char *target,
                           bool has_replaces, const char *replaces,
                           MirrorSyncMode sync,
@@ -330,7 +333,7 @@ index 681da7c8b6..02d58e7645 100644
                           bool has_speed, int64_t speed,
                           bool has_granularity, uint32_t granularity,
                           bool has_buf_size, int64_t buf_size,
-@@ -3197,7 +3229,8 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
+@@ -3291,7 +3323,8 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
      }
  
      blockdev_mirror_common(has_job_id ? job_id : NULL, bs, target_bs,
@@ -340,11 +343,11 @@ index 681da7c8b6..02d58e7645 100644
                             zero_target, has_speed, speed,
                             has_granularity, granularity,
                             has_buf_size, buf_size,
-diff --git a/include/block/block_int.h b/include/block/block_int.h
-index 8dda6f769d..279bd4ab61 100644
---- a/include/block/block_int.h
-+++ b/include/block/block_int.h
-@@ -1245,7 +1245,9 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
+diff --git a/include/block/block_int-global-state.h b/include/block/block_int-global-state.h
+index b49f4eb35b..9d744db618 100644
+--- a/include/block/block_int-global-state.h
++++ b/include/block/block_int-global-state.h
+@@ -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,
@@ -356,10 +359,10 @@ index 8dda6f769d..279bd4ab61 100644
                    BlockdevOnError on_source_error,
                    BlockdevOnError on_target_error,
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 69db270b1a..9db8e26517 100644
+index 95ac4fa634..7daaf545be 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -2064,10 +2064,19 @@
+@@ -2000,10 +2000,19 @@
  #        (all the disk, only the sectors allocated in the topmost image, or
  #        only new I/O).
  #
@@ -380,7 +383,7 @@ index 69db270b1a..9db8e26517 100644
  #
  # @buf-size: maximum amount of data in flight from source to
  #            target (since 1.4).
-@@ -2105,7 +2114,9 @@
+@@ -2043,7 +2052,9 @@
  { 'struct': 'DriveMirror',
    'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
              '*format': 'str', '*node-name': 'str', '*replaces': 'str',
@@ -391,7 +394,7 @@ index 69db270b1a..9db8e26517 100644
              '*speed': 'int', '*granularity': 'uint32',
              '*buf-size': 'int', '*on-source-error': 'BlockdevOnError',
              '*on-target-error': 'BlockdevOnError',
-@@ -2372,10 +2383,19 @@
+@@ -2322,10 +2333,19 @@
  #        (all the disk, only the sectors allocated in the topmost image, or
  #        only new I/O).
  #
@@ -412,7 +415,7 @@ index 69db270b1a..9db8e26517 100644
  #
  # @buf-size: maximum amount of data in flight from source to
  #            target
-@@ -2424,7 +2444,8 @@
+@@ -2375,7 +2395,8 @@
  { 'command': 'blockdev-mirror',
    'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
              '*replaces': 'str',
@@ -422,11 +425,11 @@ index 69db270b1a..9db8e26517 100644
              '*speed': 'int', '*granularity': 'uint32',
              '*buf-size': 'int', '*on-source-error': 'BlockdevOnError',
              '*on-target-error': 'BlockdevOnError',
-diff --git a/tests/test-block-iothread.c b/tests/test-block-iothread.c
-index 3f866a35c6..500ede71c8 100644
---- a/tests/test-block-iothread.c
-+++ b/tests/test-block-iothread.c
-@@ -623,8 +623,8 @@ static void test_propagate_mirror(void)
+diff --git a/tests/unit/test-block-iothread.c b/tests/unit/test-block-iothread.c
+index 8ca5adec5e..dae80e5a5f 100644
+--- a/tests/unit/test-block-iothread.c
++++ b/tests/unit/test-block-iothread.c
+@@ -755,8 +755,8 @@ static void test_propagate_mirror(void)
  
      /* Start a mirror job */
      mirror_start("job0", src, target, NULL, JOB_DEFAULT, 0, 0, 0,
@@ -436,4 +439,4 @@ index 3f866a35c6..500ede71c8 100644
 +                 false, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
                   false, "filter_node", MIRROR_COPY_MODE_BACKGROUND,
                   &error_abort);
-     job = job_get("job0");
+     WITH_JOB_LOCK_GUARD() {

debian/patches/bitmap-mirror/0002-drive-mirror-add-support-for-conditional-and-always-.patch

@@ -18,15 +18,16 @@ incremental backup modes; we can use this bitmap to later refresh a
 successfully created mirror.
 
 Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  block/mirror.c | 24 ++++++++++++++++++------
  1 file changed, 18 insertions(+), 6 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index 100e828639..7d3c3252f3 100644
+index 8ead5f77a0..35c1b8f25d 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -653,8 +653,6 @@ static int mirror_exit_common(Job *job)
+@@ -676,8 +676,6 @@ static int mirror_exit_common(Job *job)
          bdrv_unfreeze_backing_chain(mirror_top_bs, target_bs);
      }
  
@@ -35,9 +36,9 @@ index 100e828639..7d3c3252f3 100644
      /* Make sure that the source BDS doesn't go away during bdrv_replace_node,
       * before we can call bdrv_drained_end */
      bdrv_ref(src);
-@@ -752,6 +750,18 @@ static int mirror_exit_common(Job *job)
-     blk_set_perm(bjob->blk, 0, BLK_PERM_ALL, &error_abort);
-     blk_insert_bs(bjob->blk, mirror_top_bs, &error_abort);
+@@ -778,6 +776,18 @@ static int mirror_exit_common(Job *job)
+     block_job_remove_all_bdrv(bjob);
+     bdrv_replace_node(mirror_top_bs, mirror_top_bs->backing->bs, &error_abort);
  
 +    if (s->sync_bitmap) {
 +        if (s->bitmap_mode == BITMAP_SYNC_MODE_ALWAYS ||
@@ -54,7 +55,7 @@ index 100e828639..7d3c3252f3 100644
      bs_opaque->job = NULL;
  
      bdrv_drained_end(src);
-@@ -1589,10 +1599,6 @@ static BlockJob *mirror_start_job(
+@@ -1668,10 +1678,6 @@ static BlockJob *mirror_start_job(
                         " sync mode",
                         MirrorSyncMode_str(sync_mode));
              return NULL;
@@ -65,7 +66,7 @@ index 100e828639..7d3c3252f3 100644
          }
      } else if (bitmap) {
          error_setg(errp,
-@@ -1609,6 +1615,12 @@ static BlockJob *mirror_start_job(
+@@ -1688,6 +1694,12 @@ static BlockJob *mirror_start_job(
              return NULL;
          }
          granularity = bdrv_dirty_bitmap_granularity(bitmap);

debian/patches/bitmap-mirror/0003-mirror-add-check-for-bitmap-mode-without-bitmap.patch

@@ -10,15 +10,16 @@ as one without the other does not make much sense with the current set
 of modes.
 
 Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  blockdev.c | 3 +++
  1 file changed, 3 insertions(+)
 
 diff --git a/blockdev.c b/blockdev.c
-index 02d58e7645..0d480f02c7 100644
+index a0c7e0c13b..98b9dff154 100644
 --- a/blockdev.c
 +++ b/blockdev.c
-@@ -2974,6 +2974,9 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -3055,6 +3055,9 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
          if (bdrv_dirty_bitmap_check(bitmap, BDRV_BITMAP_ALLOW_RO, errp)) {
              return;
          }
@@ -27,4 +28,4 @@ index 02d58e7645..0d480f02c7 100644
 +        return;
      }
  
-     if (has_replaces) {
+     if (!has_replaces) {

debian/patches/bitmap-mirror/0004-mirror-switch-to-bdrv_dirty_bitmap_merge_internal.patch

@@ -10,15 +10,16 @@ since sync_bitmap is busy at the point of merging, and we checked access
 beforehand.
 
 Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
- block/mirror.c | 8 ++++----
- 1 file changed, 4 insertions(+), 4 deletions(-)
+ block/mirror.c | 11 ++++-------
+ 1 file changed, 4 insertions(+), 7 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index 7d3c3252f3..fb12ccb932 100644
+index 35c1b8f25d..4969c6833c 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -756,8 +756,8 @@ static int mirror_exit_common(Job *job)
+@@ -782,8 +782,8 @@ static int mirror_exit_common(Job *job)
               job->ret == 0 && ret == 0)) {
              /* Success; synchronize copy back to sync. */
              bdrv_clear_dirty_bitmap(s->sync_bitmap, NULL);
@@ -29,14 +30,17 @@ index 7d3c3252f3..fb12ccb932 100644
          }
      }
      bdrv_release_dirty_bitmap(s->dirty_bitmap);
-@@ -1754,8 +1754,8 @@ static BlockJob *mirror_start_job(
+@@ -1862,11 +1862,8 @@ static BlockJob *mirror_start_job(
      }
  
      if (s->sync_mode == MIRROR_SYNC_MODE_BITMAP) {
 -        bdrv_merge_dirty_bitmap(s->dirty_bitmap, s->sync_bitmap,
 -                                NULL, &local_err);
+-        if (local_err) {
+-            goto fail;
+-        }
 +        bdrv_dirty_bitmap_merge_internal(s->dirty_bitmap, s->sync_bitmap,
 +                                         NULL, true);
-         if (local_err) {
-             goto fail;
-         }
+     }
+ 
+     ret = block_job_add_bdrv(&s->common, "source", bs, 0,

debian/patches/bitmap-mirror/0005-iotests-add-test-for-bitmap-mirror.patch

@@ -20,11 +20,11 @@ intentionally keeping copyright and ownership of original test case to
 honor provenance.
 
 Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  tests/qemu-iotests/384     |  547 +++++++
  tests/qemu-iotests/384.out | 2846 ++++++++++++++++++++++++++++++++++++
- tests/qemu-iotests/group   |    1 +
- 3 files changed, 3394 insertions(+)
+ 2 files changed, 3393 insertions(+)
  create mode 100755 tests/qemu-iotests/384
  create mode 100644 tests/qemu-iotests/384.out
 
@@ -3433,15 +3433,3 @@ index 0000000000..9b7408b6d6
 +{"execute": "blockdev-mirror", "arguments": {"bitmap": "bitmap0", "device": "drive0", "filter-node-name": "mirror-top", "job-id": "api_job", "sync": "none", "target": "mirror_target"}}
 +{"error": {"class": "GenericError", "desc": "bitmap-mode must be specified if a bitmap is provided"}}
 +
-diff --git a/tests/qemu-iotests/group b/tests/qemu-iotests/group
-index 025ed5238d..bee527c012 100644
---- a/tests/qemu-iotests/group
-+++ b/tests/qemu-iotests/group
-@@ -270,6 +270,7 @@
- 253 rw quick
- 254 rw backing quick
- 255 rw quick
-+384 rw
- 256 rw auto quick
- 257 rw
- 258 rw quick

debian/patches/bitmap-mirror/0006-mirror-move-some-checks-to-qmp.patch

@@ -11,6 +11,7 @@ mode was never available for drive-mirror, it makes the interface more
 uniform w.r.t. backup block jobs.
 
 Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  block/mirror.c             | 28 +++------------
  blockdev.c                 | 29 +++++++++++++++
@@ -18,11 +19,11 @@ Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
  3 files changed, 70 insertions(+), 59 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index fb12ccb932..dfce442e97 100644
+index 4969c6833c..cf85ae1074 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -1589,31 +1589,13 @@ static BlockJob *mirror_start_job(
-     Error *local_err = NULL;
+@@ -1668,31 +1668,13 @@ static BlockJob *mirror_start_job(
+     uint64_t target_perms, target_shared_perms;
      int ret;
  
 -    if (sync_mode == MIRROR_SYNC_MODE_INCREMENTAL) {
@@ -59,10 +60,10 @@ index fb12ccb932..dfce442e97 100644
  
          if (bitmap_mode != BITMAP_SYNC_MODE_NEVER) {
 diff --git a/blockdev.c b/blockdev.c
-index 0d480f02c7..be87d65c02 100644
+index 98b9dff154..5b15a86bfa 100644
 --- a/blockdev.c
 +++ b/blockdev.c
-@@ -2953,7 +2953,36 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -3034,7 +3034,36 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
          sync = MIRROR_SYNC_MODE_FULL;
      }
  

debian/patches/extra/0001-block-block-copy-always-align-copied-region-to-clust.patch

@@ -1,34 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 6 Aug 2020 15:43:58 +0200
-Subject: [PATCH] block/block-copy: always align copied region to cluster size
-
-Since commit 42ac214406e0 (block/block-copy: refactor task creation)
-block_copy_task_create calculates the area to be copied via
-bdrv_dirty_bitmap_next_dirty_area, but that can return an unaligned byte
-count if the image's last cluster end is not aligned to the bitmap's
-granularity.
-
-Always ALIGN_UP the resulting bytes value to satisfy block_copy_do_copy,
-which requires the 'bytes' parameter to be aligned to cluster size.
-
-Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- block/block-copy.c | 3 +++
- 1 file changed, 3 insertions(+)
-
-diff --git a/block/block-copy.c b/block/block-copy.c
-index f7428a7c08..a30b9097ef 100644
---- a/block/block-copy.c
-+++ b/block/block-copy.c
-@@ -142,6 +142,9 @@ static BlockCopyTask *block_copy_task_create(BlockCopyState *s,
-         return NULL;
-     }
- 
-+    assert(QEMU_IS_ALIGNED(offset, s->cluster_size));
-+    bytes = QEMU_ALIGN_UP(bytes, s->cluster_size);
-+
-     /* region is dirty, so no existent tasks possible in it */
-     assert(!find_conflicting_task(s, offset, bytes));
- 

debian/patches/extra/0001-monitor-qmp-fix-race-with-clients-disconnecting-earl.patch

@@ -0,0 +1,206 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Mon, 23 Aug 2021 11:28:32 +0200
+Subject: [PATCH] monitor/qmp: fix race with clients disconnecting early
+
+The following sequence can produce a race condition that results in
+responses meant for different clients being sent to the wrong one:
+
+(QMP, no OOB)
+1) client A connects
+2) client A sends 'qmp_capabilities'
+3) 'qmp_dispatch' runs in coroutine, schedules out to
+   'do_qmp_dispatch_bh' and yields
+4) client A disconnects (i.e. aborts, crashes, etc...)
+5) client B connects
+6) 'do_qmp_dispatch_bh' runs 'qmp_capabilities' and wakes calling coroutine
+7) capabilities are now set and 'mon->commands' is set to '&qmp_commands'
+8) 'qmp_dispatch' returns to 'monitor_qmp_dispatch'
+9) success message is sent to client B *without it ever having sent
+   'qmp_capabilities' itself*
+9a) even if client B ignores it, it will now presumably send it's own
+   greeting, which will error because caps are already set
+
+The fix proposed here uses an atomic, sequential connection number
+stored in the MonitorQMP struct, which is incremented everytime a new
+client connects. Since it is not changed on CHR_EVENT_CLOSED, the
+behaviour of allowing a client to disconnect only one side of the
+connection is retained.
+
+The connection_nr needs to be exposed outside of the monitor subsystem,
+since qmp_dispatch lives in qapi code. It needs to be checked twice,
+once for actually running the command in the BH (fixes 7), and once for
+sending back a response (fixes 9).
+
+This satisfies my local reproducer - using multiple clients constantly
+looping to open a connection, send the greeting, then exiting no longer
+crashes other, normally behaving clients with unrelated responses.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ include/monitor/monitor.h  |  1 +
+ monitor/monitor-internal.h |  7 +++++++
+ monitor/monitor.c          | 15 +++++++++++++++
+ monitor/qmp.c              | 15 ++++++++++++++-
+ qapi/qmp-dispatch.c        | 21 +++++++++++++++++----
+ stubs/monitor-core.c       |  5 +++++
+ 6 files changed, 59 insertions(+), 5 deletions(-)
+
+diff --git a/include/monitor/monitor.h b/include/monitor/monitor.h
+index 737e750670..38804b8595 100644
+--- a/include/monitor/monitor.h
++++ b/include/monitor/monitor.h
+@@ -16,6 +16,7 @@ extern QemuOptsList qemu_mon_opts;
+ Monitor *monitor_cur(void);
+ Monitor *monitor_set_cur(Coroutine *co, Monitor *mon);
+ bool monitor_cur_is_qmp(void);
++int monitor_get_connection_nr(const Monitor *mon);
+ 
+ void monitor_init_globals(void);
+ void monitor_init_globals_core(void);
+diff --git a/monitor/monitor-internal.h b/monitor/monitor-internal.h
+index a2cdbbf646..b531bd50e7 100644
+--- a/monitor/monitor-internal.h
++++ b/monitor/monitor-internal.h
+@@ -152,6 +152,13 @@ typedef struct {
+     QemuMutex qmp_queue_lock;
+     /* Input queue that holds all the parsed QMP requests */
+     GQueue *qmp_requests;
++
++    /*
++     * A sequential number that gets incremented on every new CHR_EVENT_OPENED.
++     * Used to avoid leftover responses in BHs from being sent to the wrong
++     * client. Access with atomics.
++     */
++    int connection_nr;
+ } MonitorQMP;
+ 
+ /**
+diff --git a/monitor/monitor.c b/monitor/monitor.c
+index 86949024f6..c306cadcf4 100644
+--- a/monitor/monitor.c
++++ b/monitor/monitor.c
+@@ -135,6 +135,21 @@ bool monitor_cur_is_qmp(void)
+     return cur_mon && monitor_is_qmp(cur_mon);
+ }
+ 
++/**
++ * If @mon is a QMP monitor, return the connection_nr, otherwise -1.
++ */
++int monitor_get_connection_nr(const Monitor *mon)
++{
++    MonitorQMP *qmp_mon;
++
++    if (!monitor_is_qmp(mon)) {
++        return -1;
++    }
++
++    qmp_mon = container_of(mon, MonitorQMP, common);
++    return qatomic_read(&qmp_mon->connection_nr);
++}
++
+ /**
+  * Is @mon is using readline?
+  * Note: not all HMP monitors use readline, e.g., gdbserver has a
+diff --git a/monitor/qmp.c b/monitor/qmp.c
+index acd0a350c2..cc1407e4ac 100644
+--- a/monitor/qmp.c
++++ b/monitor/qmp.c
+@@ -141,6 +141,8 @@ static void monitor_qmp_dispatch(MonitorQMP *mon, QObject *req)
+     QDict *rsp;
+     QDict *error;
+ 
++    int conn_nr_before = qatomic_read(&mon->connection_nr);
++
+     rsp = qmp_dispatch(mon->commands, req, qmp_oob_enabled(mon),
+                        &mon->common);
+ 
+@@ -156,7 +158,17 @@ static void monitor_qmp_dispatch(MonitorQMP *mon, QObject *req)
+         }
+     }
+ 
+-    monitor_qmp_respond(mon, rsp);
++    /*
++     * qmp_dispatch might have yielded and waited for a BH, in which case there
++     * is a chance a new client connected in the meantime - if this happened,
++     * the command will not have been executed, but we also need to ensure that
++     * we don't send back a corresponding response on a line that no longer
++     * belongs to this request.
++     */
++    if (conn_nr_before == qatomic_read(&mon->connection_nr)) {
++        monitor_qmp_respond(mon, rsp);
++    }
++
+     qobject_unref(rsp);
+ }
+ 
+@@ -427,6 +439,7 @@ static void monitor_qmp_event(void *opaque, QEMUChrEvent event)
+ 
+     switch (event) {
+     case CHR_EVENT_OPENED:
++        qatomic_inc_fetch(&mon->connection_nr);
+         mon->commands = &qmp_cap_negotiation_commands;
+         monitor_qmp_caps_reset(mon);
+         data = qmp_greeting(mon);
+diff --git a/qapi/qmp-dispatch.c b/qapi/qmp-dispatch.c
+index 5d000fae87..404d428824 100644
+--- a/qapi/qmp-dispatch.c
++++ b/qapi/qmp-dispatch.c
+@@ -117,16 +117,28 @@ typedef struct QmpDispatchBH {
+     QObject **ret;
+     Error **errp;
+     Coroutine *co;
++    int conn_nr;
+ } QmpDispatchBH;
+ 
+ static void do_qmp_dispatch_bh(void *opaque)
+ {
+     QmpDispatchBH *data = opaque;
+ 
+-    assert(monitor_cur() == NULL);
+-    monitor_set_cur(qemu_coroutine_self(), data->cur_mon);
+-    data->cmd->fn(data->args, data->ret, data->errp);
+-    monitor_set_cur(qemu_coroutine_self(), NULL);
++    /*
++     * A QMP monitor tracks it's client with a connection number, if this
++     * changes during the scheduling delay of this BH, we must not execute the
++     * command. Otherwise a badly placed 'qmp_capabilities' might affect the
++     * connection state of a client it was never meant for.
++     */
++    if (data->conn_nr == monitor_get_connection_nr(data->cur_mon)) {
++        assert(monitor_cur() == NULL);
++        monitor_set_cur(qemu_coroutine_self(), data->cur_mon);
++        data->cmd->fn(data->args, data->ret, data->errp);
++        monitor_set_cur(qemu_coroutine_self(), NULL);
++    } else {
++        error_setg(data->errp, "active monitor connection changed");
++    }
++
+     aio_co_wake(data->co);
+ }
+ 
+@@ -253,6 +265,7 @@ QDict *qmp_dispatch(const QmpCommandList *cmds, QObject *request,
+             .ret        = &ret,
+             .errp       = &err,
+             .co         = qemu_coroutine_self(),
++            .conn_nr    = monitor_get_connection_nr(cur_mon),
+         };
+         aio_bh_schedule_oneshot(iohandler_get_aio_context(), do_qmp_dispatch_bh,
+                                 &data);
+diff --git a/stubs/monitor-core.c b/stubs/monitor-core.c
+index afa477aae6..d3ff124bf3 100644
+--- a/stubs/monitor-core.c
++++ b/stubs/monitor-core.c
+@@ -12,6 +12,11 @@ Monitor *monitor_set_cur(Coroutine *co, Monitor *mon)
+     return NULL;
+ }
+ 
++int monitor_get_connection_nr(const Monitor *mon)
++{
++    return -1;
++}
++
+ void monitor_init_qmp(Chardev *chr, bool pretty, Error **errp)
+ {
+ }

debian/patches/extra/0002-Revert-qemu-img-convert-Don-t-pre-zero-images.patch

@@ -1,33 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Mon, 14 Sep 2020 19:32:21 +0200
-Subject: [PATCH] Revert "qemu-img convert: Don't pre-zero images"
-
-This reverts commit edafc70c0c8510862f2f213a3acf7067113bcd08.
-
-As it correlates with causing issues on LVM allocation
-https://bugzilla.proxmox.com/show_bug.cgi?id=3002
----
- qemu-img.c | 9 +++++++++
- 1 file changed, 9 insertions(+)
-
-diff --git a/qemu-img.c b/qemu-img.c
-index 9635e9c001..c7884d248a 100644
---- a/qemu-img.c
-+++ b/qemu-img.c
-@@ -2079,6 +2079,15 @@ static int convert_do_copy(ImgConvertState *s)
-         s->has_zero_init = bdrv_has_zero_init(blk_bs(s->target));
-     }
- 
-+    if (!s->has_zero_init && !s->target_has_backing &&
-+        bdrv_can_write_zeroes_with_unmap(blk_bs(s->target)))
-+    {
-+        ret = blk_make_zero(s->target, BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK);
-+        if (ret == 0) {
-+            s->has_zero_init = true;
-+        }
-+    }
-+
-     /* Allocate buffer for copied data. For compressed images, only one cluster
-      * can be copied at a time. */
-     if (s->compressed) {

debian/patches/extra/0002-init-daemonize-defuse-PID-file-resolve-error.patch

@@ -0,0 +1,42 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Fri, 28 Oct 2022 10:09:46 +0200
+Subject: [PATCH] init: daemonize: defuse PID file resolve error
+
+When proxmox-file-restore invokes QEMU, the PID file is a (temporary)
+file that's already unlinked, so resolving the absolute path here
+failed.
+
+It should not be a critical error when the PID file unlink handler
+can't be registered, because the path can't be resolved for whatever
+reason. If the file is already gone from QEMU's perspective (i.e.
+errno is ENOENT), silently ignore the error. Otherwise, print a
+warning.
+
+Reported-by: Dominik Csapak <d.csapak@proxmox.com>
+Suggested-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ softmmu/vl.c | 9 +++++----
+ 1 file changed, 5 insertions(+), 4 deletions(-)
+
+diff --git a/softmmu/vl.c b/softmmu/vl.c
+index 38d76d6e51..7aa3eb5cf9 100644
+--- a/softmmu/vl.c
++++ b/softmmu/vl.c
+@@ -2468,10 +2468,11 @@ static void qemu_maybe_daemonize(const char *pid_file)
+ 
+         pid_file_realpath = g_malloc0(PATH_MAX);
+         if (!realpath(pid_file, pid_file_realpath)) {
+-            error_report("cannot resolve PID file path: %s: %s",
+-                         pid_file, strerror(errno));
+-            unlink(pid_file);
+-            exit(1);
++            if (errno != ENOENT) {
++                warn_report("not removing PID file on exit: cannot resolve PID "
++                            "file path: %s: %s", pid_file, strerror(errno));
++            }
++            return;
+         }
+ 
+         qemu_unlink_pidfile_notifier = (struct UnlinkPidfileNotifier) {

debian/patches/extra/0003-scsi-megasas-Internal-cdbs-have-16-byte-length.patch

@@ -0,0 +1,69 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Guenter Roeck <linux@roeck-us.net>
+Date: Tue, 28 Feb 2023 09:11:29 -0800
+Subject: [PATCH] scsi: megasas: Internal cdbs have 16-byte length
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+Host drivers do not necessarily set cdb_len in megasas io commands.
+With commits 6d1511cea0 ("scsi: Reject commands if the CDB length
+exceeds buf_len") and fe9d8927e2 ("scsi: Add buf_len parameter to
+scsi_req_new()"), this results in failures to boot Linux from affected
+SCSI drives because cdb_len is set to 0 by the host driver.
+Set the cdb length to its actual size to solve the problem.
+
+Signed-off-by: Guenter Roeck <linux@roeck-us.net>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+(picked-up from https://lists.nongnu.org/archive/html/qemu-devel/2023-02/msg08653.html)
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/scsi/megasas.c | 14 ++------------
+ 1 file changed, 2 insertions(+), 12 deletions(-)
+
+diff --git a/hw/scsi/megasas.c b/hw/scsi/megasas.c
+index 9cbbb16121..d624866bb6 100644
+--- a/hw/scsi/megasas.c
++++ b/hw/scsi/megasas.c
+@@ -1780,7 +1780,7 @@ static int megasas_handle_io(MegasasState *s, MegasasCmd *cmd, int frame_cmd)
+     uint8_t cdb[16];
+     int len;
+     struct SCSIDevice *sdev = NULL;
+-    int target_id, lun_id, cdb_len;
++    int target_id, lun_id;
+ 
+     lba_count = le32_to_cpu(cmd->frame->io.header.data_len);
+     lba_start_lo = le32_to_cpu(cmd->frame->io.lba_lo);
+@@ -1789,7 +1789,6 @@ static int megasas_handle_io(MegasasState *s, MegasasCmd *cmd, int frame_cmd)
+ 
+     target_id = cmd->frame->header.target_id;
+     lun_id = cmd->frame->header.lun_id;
+-    cdb_len = cmd->frame->header.cdb_len;
+ 
+     if (target_id < MFI_MAX_LD && lun_id == 0) {
+         sdev = scsi_device_find(&s->bus, 0, target_id, lun_id);
+@@ -1804,15 +1803,6 @@ static int megasas_handle_io(MegasasState *s, MegasasCmd *cmd, int frame_cmd)
+         return MFI_STAT_DEVICE_NOT_FOUND;
+     }
+ 
+-    if (cdb_len > 16) {
+-        trace_megasas_scsi_invalid_cdb_len(
+-            mfi_frame_desc(frame_cmd), 1, target_id, lun_id, cdb_len);
+-        megasas_write_sense(cmd, SENSE_CODE(INVALID_OPCODE));
+-        cmd->frame->header.scsi_status = CHECK_CONDITION;
+-        s->event_count++;
+-        return MFI_STAT_SCSI_DONE_WITH_ERROR;
+-    }
+-
+     cmd->iov_size = lba_count * sdev->blocksize;
+     if (megasas_map_sgl(s, cmd, &cmd->frame->io.sgl)) {
+         megasas_write_sense(cmd, SENSE_CODE(TARGET_FAILURE));
+@@ -1823,7 +1813,7 @@ static int megasas_handle_io(MegasasState *s, MegasasCmd *cmd, int frame_cmd)
+ 
+     megasas_encode_lba(cdb, lba_start, lba_count, is_write);
+     cmd->req = scsi_req_new(sdev, cmd->index,
+-                            lun_id, cdb, cdb_len, cmd);
++                            lun_id, cdb, sizeof(cdb), cmd);
+     if (!cmd->req) {
+         trace_megasas_scsi_req_alloc_failed(
+             mfi_frame_desc(frame_cmd), target_id, lun_id);

debian/patches/extra/0003-usb-fix-setup_len-init-CVE-2020-14364.patch

@@ -1,87 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Gerd Hoffmann <kraxel@redhat.com>
-Date: Tue, 25 Aug 2020 07:36:36 +0200
-Subject: [PATCH] usb: fix setup_len init (CVE-2020-14364)
-
-Store calculated setup_len in a local variable, verify it, and only
-write it to the struct (USBDevice->setup_len) in case it passed the
-sanity checks.
-
-This prevents other code (do_token_{in,out} functions specifically)
-from working with invalid USBDevice->setup_len values and overrunning
-the USBDevice->setup_buf[] buffer.
-
-Fixes: CVE-2020-14364
-Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
-Tested-by: Gonglei <arei.gonglei@huawei.com>
-Reviewed-by: Li Qiang <liq3ea@gmail.com>
-Message-id: 20200825053636.29648-1-kraxel@redhat.com
-(cherry picked from commit b946434f2659a182afc17e155be6791ebfb302eb)
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/usb/core.c | 16 ++++++++++------
- 1 file changed, 10 insertions(+), 6 deletions(-)
-
-diff --git a/hw/usb/core.c b/hw/usb/core.c
-index 5abd128b6b..5234dcc73f 100644
---- a/hw/usb/core.c
-+++ b/hw/usb/core.c
-@@ -129,6 +129,7 @@ void usb_wakeup(USBEndpoint *ep, unsigned int stream)
- static void do_token_setup(USBDevice *s, USBPacket *p)
- {
-     int request, value, index;
-+    unsigned int setup_len;
- 
-     if (p->iov.size != 8) {
-         p->status = USB_RET_STALL;
-@@ -138,14 +139,15 @@ static void do_token_setup(USBDevice *s, USBPacket *p)
-     usb_packet_copy(p, s->setup_buf, p->iov.size);
-     s->setup_index = 0;
-     p->actual_length = 0;
--    s->setup_len   = (s->setup_buf[7] << 8) | s->setup_buf[6];
--    if (s->setup_len > sizeof(s->data_buf)) {
-+    setup_len = (s->setup_buf[7] << 8) | s->setup_buf[6];
-+    if (setup_len > sizeof(s->data_buf)) {
-         fprintf(stderr,
-                 "usb_generic_handle_packet: ctrl buffer too small (%d > %zu)\n",
--                s->setup_len, sizeof(s->data_buf));
-+                setup_len, sizeof(s->data_buf));
-         p->status = USB_RET_STALL;
-         return;
-     }
-+    s->setup_len = setup_len;
- 
-     request = (s->setup_buf[0] << 8) | s->setup_buf[1];
-     value   = (s->setup_buf[3] << 8) | s->setup_buf[2];
-@@ -259,26 +261,28 @@ static void do_token_out(USBDevice *s, USBPacket *p)
- static void do_parameter(USBDevice *s, USBPacket *p)
- {
-     int i, request, value, index;
-+    unsigned int setup_len;
- 
-     for (i = 0; i < 8; i++) {
-         s->setup_buf[i] = p->parameter >> (i*8);
-     }
- 
-     s->setup_state = SETUP_STATE_PARAM;
--    s->setup_len   = (s->setup_buf[7] << 8) | s->setup_buf[6];
-     s->setup_index = 0;
- 
-     request = (s->setup_buf[0] << 8) | s->setup_buf[1];
-     value   = (s->setup_buf[3] << 8) | s->setup_buf[2];
-     index   = (s->setup_buf[5] << 8) | s->setup_buf[4];
- 
--    if (s->setup_len > sizeof(s->data_buf)) {
-+    setup_len = (s->setup_buf[7] << 8) | s->setup_buf[6];
-+    if (setup_len > sizeof(s->data_buf)) {
-         fprintf(stderr,
-                 "usb_generic_handle_packet: ctrl buffer too small (%d > %zu)\n",
--                s->setup_len, sizeof(s->data_buf));
-+                setup_len, sizeof(s->data_buf));
-         p->status = USB_RET_STALL;
-         return;
-     }
-+    s->setup_len = setup_len;
- 
-     if (p->pid == USB_TOKEN_OUT) {
-         usb_packet_copy(p, s->data_buf, s->setup_len);

debian/patches/extra/0004-ide-avoid-potential-deadlock-when-draining-during-tr.patch

@@ -0,0 +1,100 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Tue, 7 Mar 2023 15:03:02 +0100
+Subject: [PATCH] ide: avoid potential deadlock when draining during trim
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+The deadlock can happen as follows:
+1. ide_issue_trim is called, and increments the in_flight counter.
+2. ide_issue_trim_cb calls blk_aio_pdiscard.
+3. Somebody else starts draining (e.g. backup to insert the cbw node).
+4. ide_issue_trim_cb is called as the completion callback for
+   blk_aio_pdiscard.
+5. ide_issue_trim_cb issues yet another blk_aio_pdiscard request.
+6. The request is added to the wait queue via blk_wait_while_drained,
+   because draining has been started.
+7. Nobody ever decrements the in_flight counter and draining can't
+   finish. This would be done by ide_trim_bh_cb, which is called after
+   ide_issue_trim_cb has issued its last request, but
+   ide_issue_trim_cb is not called anymore, because it's the
+   completion callback of blk_aio_pdiscard, which waits on draining.
+
+Quoting Hanna Czenczek:
+> The point of 7e5cdb345f was that we need any in-flight count to
+> accompany a set s->bus->dma->aiocb. While blk_aio_pdiscard() is
+> happening, we don’t necessarily need another count. But we do need
+> it while there is no blk_aio_pdiscard().
+> ide_issue_trim_cb() returns in two cases (and, recursively through
+> its callers, leaves s->bus->dma->aiocb set):
+> 1. After calling blk_aio_pdiscard(), which will keep an in-flight
+>    count,
+> 2. After calling replay_bh_schedule_event() (i.e.
+>    qemu_bh_schedule()), which does not keep an in-flight count.
+
+Thus, even after moving the blk_inc_in_flight to above the
+replay_bh_schedule_event call, the invariant "ide_issue_trim_cb
+returns with an accompanying in-flight count" is still satisfied.
+
+However, the issue 7e5cdb345f fixed for canceling resurfaces, because
+ide_cancel_dma_sync assumes that it just needs to drain once. But now
+the in_flight count is not consistently > 0 during the trim operation.
+So, change it to drain until !s->bus->dma->aiocb, which means that the
+operation finished (s->bus->dma->aiocb is cleared by ide_set_inactive
+via the ide_dma_cb when the end of the transfer is reached).
+
+Discussion here:
+https://lists.nongnu.org/archive/html/qemu-devel/2023-03/msg02506.html
+
+Fixes: 7e5cdb345f ("ide: Increment BB in-flight counter for TRIM BH")
+Suggested-by: Hanna Czenczek <hreitz@redhat.com>
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/ide/core.c | 12 ++++++------
+ 1 file changed, 6 insertions(+), 6 deletions(-)
+
+diff --git a/hw/ide/core.c b/hw/ide/core.c
+index 3e97d665d9..a0f6801bce 100644
+--- a/hw/ide/core.c
++++ b/hw/ide/core.c
+@@ -443,7 +443,7 @@ static void ide_trim_bh_cb(void *opaque)
+     iocb->bh = NULL;
+     qemu_aio_unref(iocb);
+ 
+-    /* Paired with an increment in ide_issue_trim() */
++    /* Paired with an increment in ide_issue_trim_cb() */
+     blk_dec_in_flight(blk);
+ }
+ 
+@@ -503,6 +503,8 @@ static void ide_issue_trim_cb(void *opaque, int ret)
+ done:
+     iocb->aiocb = NULL;
+     if (iocb->bh) {
++        /* Paired with a decrement in ide_trim_bh_cb() */
++        blk_inc_in_flight(s->blk);
+         replay_bh_schedule_event(iocb->bh);
+     }
+ }
+@@ -515,9 +517,6 @@ BlockAIOCB *ide_issue_trim(
+     IDEDevice *dev = s->unit ? s->bus->slave : s->bus->master;
+     TrimAIOCB *iocb;
+ 
+-    /* Paired with a decrement in ide_trim_bh_cb() */
+-    blk_inc_in_flight(s->blk);
+-
+     iocb = blk_aio_get(&trim_aiocb_info, s->blk, cb, cb_opaque);
+     iocb->s = s;
+     iocb->bh = qemu_bh_new_guarded(ide_trim_bh_cb, iocb,
+@@ -741,8 +740,9 @@ void ide_cancel_dma_sync(IDEState *s)
+      */
+     if (s->bus->dma->aiocb) {
+         trace_ide_cancel_dma_sync_remaining();
+-        blk_drain(s->blk);
+-        assert(s->bus->dma->aiocb == NULL);
++        while (s->bus->dma->aiocb) {
++            blk_drain(s->blk);
++        }
+     }
+ }
+ 

debian/patches/extra/0005-target-arm-align-exposed-ID-registers-with-Linux.patch

@@ -0,0 +1,273 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Zhuojia Shen <chaosdefinition@hotmail.com>
+Date: Wed, 10 Apr 2024 08:43:24 +0300
+Subject: [PATCH] target/arm: align exposed ID registers with Linux
+
+In CPUID registers exposed to userspace, some registers were missing
+and some fields were not exposed.  This patch aligns exposed ID
+registers and their fields with what the upstream kernel currently
+exposes.
+
+Specifically, the following new ID registers/fields are exposed to
+userspace:
+
+ID_AA64PFR1_EL1.BT:       bits 3-0
+ID_AA64PFR1_EL1.MTE:      bits 11-8
+ID_AA64PFR1_EL1.SME:      bits 27-24
+
+ID_AA64ZFR0_EL1.SVEver:   bits 3-0
+ID_AA64ZFR0_EL1.AES:      bits 7-4
+ID_AA64ZFR0_EL1.BitPerm:  bits 19-16
+ID_AA64ZFR0_EL1.BF16:     bits 23-20
+ID_AA64ZFR0_EL1.SHA3:     bits 35-32
+ID_AA64ZFR0_EL1.SM4:      bits 43-40
+ID_AA64ZFR0_EL1.I8MM:     bits 47-44
+ID_AA64ZFR0_EL1.F32MM:    bits 55-52
+ID_AA64ZFR0_EL1.F64MM:    bits 59-56
+
+ID_AA64SMFR0_EL1.F32F32:  bit 32
+ID_AA64SMFR0_EL1.B16F32:  bit 34
+ID_AA64SMFR0_EL1.F16F32:  bit 35
+ID_AA64SMFR0_EL1.I8I32:   bits 39-36
+ID_AA64SMFR0_EL1.F64F64:  bit 48
+ID_AA64SMFR0_EL1.I16I64:  bits 55-52
+ID_AA64SMFR0_EL1.FA64:    bit 63
+
+ID_AA64MMFR0_EL1.ECV:     bits 63-60
+
+ID_AA64MMFR1_EL1.AFP:     bits 47-44
+
+ID_AA64MMFR2_EL1.AT:      bits 35-32
+
+ID_AA64ISAR0_EL1.RNDR:    bits 63-60
+
+ID_AA64ISAR1_EL1.FRINTTS: bits 35-32
+ID_AA64ISAR1_EL1.BF16:    bits 47-44
+ID_AA64ISAR1_EL1.DGH:     bits 51-48
+ID_AA64ISAR1_EL1.I8MM:    bits 55-52
+
+ID_AA64ISAR2_EL1.WFxT:    bits 3-0
+ID_AA64ISAR2_EL1.RPRES:   bits 7-4
+ID_AA64ISAR2_EL1.GPA3:    bits 11-8
+ID_AA64ISAR2_EL1.APA3:    bits 15-12
+
+The code is also refactored to use symbolic names for ID register fields
+for better readability and maintainability.
+
+The test case in tests/tcg/aarch64/sysregs.c is also updated to match
+the intended behavior.
+
+Signed-off-by: Zhuojia Shen <chaosdefinition@hotmail.com>
+Message-id: DS7PR12MB6309FB585E10772928F14271ACE79@DS7PR12MB6309.namprd12.prod.outlook.com
+Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+[PMM: use Sn_n_Cn_Cn_n syntax to work with older assemblers
+that don't recognize id_aa64isar2_el1 and id_aa64mmfr2_el1]
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+(cherry picked from commit bc6bd20ee3538347afb750c4bd06edca4a922897)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+(Mjt: pick this for v8.0.0-2361-g1f51573f79
+ "target/arm: Fix SME full tile indexing")
+---
+ target/arm/helper.c               | 96 +++++++++++++++++++++++++------
+ tests/tcg/aarch64/Makefile.target |  7 ++-
+ tests/tcg/aarch64/sysregs.c       | 24 ++++++--
+ 3 files changed, 103 insertions(+), 24 deletions(-)
+
+diff --git a/target/arm/helper.c b/target/arm/helper.c
+index 2e284e048c..acc0470e86 100644
+--- a/target/arm/helper.c
++++ b/target/arm/helper.c
+@@ -7852,31 +7852,89 @@ void register_cp_regs_for_features(ARMCPU *cpu)
+ #ifdef CONFIG_USER_ONLY
+         static const ARMCPRegUserSpaceInfo v8_user_idregs[] = {
+             { .name = "ID_AA64PFR0_EL1",
+-              .exported_bits = 0x000f000f00ff0000,
+-              .fixed_bits    = 0x0000000000000011 },
++              .exported_bits = R_ID_AA64PFR0_FP_MASK |
++                               R_ID_AA64PFR0_ADVSIMD_MASK |
++                               R_ID_AA64PFR0_SVE_MASK |
++                               R_ID_AA64PFR0_DIT_MASK,
++              .fixed_bits = (0x1u << R_ID_AA64PFR0_EL0_SHIFT) |
++                            (0x1u << R_ID_AA64PFR0_EL1_SHIFT) },
+             { .name = "ID_AA64PFR1_EL1",
+-              .exported_bits = 0x00000000000000f0 },
++              .exported_bits = R_ID_AA64PFR1_BT_MASK |
++                               R_ID_AA64PFR1_SSBS_MASK |
++                               R_ID_AA64PFR1_MTE_MASK |
++                               R_ID_AA64PFR1_SME_MASK },
+             { .name = "ID_AA64PFR*_EL1_RESERVED",
+-              .is_glob = true                     },
+-            { .name = "ID_AA64ZFR0_EL1"           },
++              .is_glob = true },
++            { .name = "ID_AA64ZFR0_EL1",
++              .exported_bits = R_ID_AA64ZFR0_SVEVER_MASK |
++                               R_ID_AA64ZFR0_AES_MASK |
++                               R_ID_AA64ZFR0_BITPERM_MASK |
++                               R_ID_AA64ZFR0_BFLOAT16_MASK |
++                               R_ID_AA64ZFR0_SHA3_MASK |
++                               R_ID_AA64ZFR0_SM4_MASK |
++                               R_ID_AA64ZFR0_I8MM_MASK |
++                               R_ID_AA64ZFR0_F32MM_MASK |
++                               R_ID_AA64ZFR0_F64MM_MASK },
++            { .name = "ID_AA64SMFR0_EL1",
++              .exported_bits = R_ID_AA64SMFR0_F32F32_MASK |
++                               R_ID_AA64SMFR0_B16F32_MASK |
++                               R_ID_AA64SMFR0_F16F32_MASK |
++                               R_ID_AA64SMFR0_I8I32_MASK |
++                               R_ID_AA64SMFR0_F64F64_MASK |
++                               R_ID_AA64SMFR0_I16I64_MASK |
++                               R_ID_AA64SMFR0_FA64_MASK },
+             { .name = "ID_AA64MMFR0_EL1",
+-              .fixed_bits    = 0x00000000ff000000 },
+-            { .name = "ID_AA64MMFR1_EL1"          },
++              .exported_bits = R_ID_AA64MMFR0_ECV_MASK,
++              .fixed_bits = (0xfu << R_ID_AA64MMFR0_TGRAN64_SHIFT) |
++                            (0xfu << R_ID_AA64MMFR0_TGRAN4_SHIFT) },
++            { .name = "ID_AA64MMFR1_EL1",
++              .exported_bits = R_ID_AA64MMFR1_AFP_MASK },
++            { .name = "ID_AA64MMFR2_EL1",
++              .exported_bits = R_ID_AA64MMFR2_AT_MASK },
+             { .name = "ID_AA64MMFR*_EL1_RESERVED",
+-              .is_glob = true                     },
++              .is_glob = true },
+             { .name = "ID_AA64DFR0_EL1",
+-              .fixed_bits    = 0x0000000000000006 },
+-            { .name = "ID_AA64DFR1_EL1"           },
++              .fixed_bits = (0x6u << R_ID_AA64DFR0_DEBUGVER_SHIFT) },
++            { .name = "ID_AA64DFR1_EL1" },
+             { .name = "ID_AA64DFR*_EL1_RESERVED",
+-              .is_glob = true                     },
++              .is_glob = true },
+             { .name = "ID_AA64AFR*",
+-              .is_glob = true                     },
++              .is_glob = true },
+             { .name = "ID_AA64ISAR0_EL1",
+-              .exported_bits = 0x00fffffff0fffff0 },
++              .exported_bits = R_ID_AA64ISAR0_AES_MASK |
++                               R_ID_AA64ISAR0_SHA1_MASK |
++                               R_ID_AA64ISAR0_SHA2_MASK |
++                               R_ID_AA64ISAR0_CRC32_MASK |
++                               R_ID_AA64ISAR0_ATOMIC_MASK |
++                               R_ID_AA64ISAR0_RDM_MASK |
++                               R_ID_AA64ISAR0_SHA3_MASK |
++                               R_ID_AA64ISAR0_SM3_MASK |
++                               R_ID_AA64ISAR0_SM4_MASK |
++                               R_ID_AA64ISAR0_DP_MASK |
++                               R_ID_AA64ISAR0_FHM_MASK |
++                               R_ID_AA64ISAR0_TS_MASK |
++                               R_ID_AA64ISAR0_RNDR_MASK },
+             { .name = "ID_AA64ISAR1_EL1",
+-              .exported_bits = 0x000000f0ffffffff },
++              .exported_bits = R_ID_AA64ISAR1_DPB_MASK |
++                               R_ID_AA64ISAR1_APA_MASK |
++                               R_ID_AA64ISAR1_API_MASK |
++                               R_ID_AA64ISAR1_JSCVT_MASK |
++                               R_ID_AA64ISAR1_FCMA_MASK |
++                               R_ID_AA64ISAR1_LRCPC_MASK |
++                               R_ID_AA64ISAR1_GPA_MASK |
++                               R_ID_AA64ISAR1_GPI_MASK |
++                               R_ID_AA64ISAR1_FRINTTS_MASK |
++                               R_ID_AA64ISAR1_SB_MASK |
++                               R_ID_AA64ISAR1_BF16_MASK |
++                               R_ID_AA64ISAR1_DGH_MASK |
++                               R_ID_AA64ISAR1_I8MM_MASK },
++            { .name = "ID_AA64ISAR2_EL1",
++              .exported_bits = R_ID_AA64ISAR2_WFXT_MASK |
++                               R_ID_AA64ISAR2_RPRES_MASK |
++                               R_ID_AA64ISAR2_GPA3_MASK |
++                               R_ID_AA64ISAR2_APA3_MASK },
+             { .name = "ID_AA64ISAR*_EL1_RESERVED",
+-              .is_glob = true                     },
++              .is_glob = true },
+         };
+         modify_arm_cp_regs(v8_idregs, v8_user_idregs);
+ #endif
+@@ -8194,8 +8252,12 @@ void register_cp_regs_for_features(ARMCPU *cpu)
+ #ifdef CONFIG_USER_ONLY
+         static const ARMCPRegUserSpaceInfo id_v8_user_midr_cp_reginfo[] = {
+             { .name = "MIDR_EL1",
+-              .exported_bits = 0x00000000ffffffff },
+-            { .name = "REVIDR_EL1"                },
++              .exported_bits = R_MIDR_EL1_REVISION_MASK |
++                               R_MIDR_EL1_PARTNUM_MASK |
++                               R_MIDR_EL1_ARCHITECTURE_MASK |
++                               R_MIDR_EL1_VARIANT_MASK |
++                               R_MIDR_EL1_IMPLEMENTER_MASK },
++            { .name = "REVIDR_EL1" },
+         };
+         modify_arm_cp_regs(id_v8_midr_cp_reginfo, id_v8_user_midr_cp_reginfo);
+ #endif
+diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
+index a72578fccb..fc6d5d824d 100644
+--- a/tests/tcg/aarch64/Makefile.target
++++ b/tests/tcg/aarch64/Makefile.target
+@@ -23,7 +23,8 @@ config-cc.mak: Makefile
+ 	    $(call cc-option,-march=armv8.1-a+sve2,         CROSS_CC_HAS_SVE2); \
+ 	    $(call cc-option,-march=armv8.3-a,              CROSS_CC_HAS_ARMV8_3); \
+ 	    $(call cc-option,-mbranch-protection=standard,  CROSS_CC_HAS_ARMV8_BTI); \
+-	    $(call cc-option,-march=armv8.5-a+memtag,       CROSS_CC_HAS_ARMV8_MTE)) 3> config-cc.mak
++	    $(call cc-option,-march=armv8.5-a+memtag,       CROSS_CC_HAS_ARMV8_MTE); \
++	    $(call cc-option,-march=armv9-a+sme,            CROSS_CC_HAS_ARMV9_SME)) 3> config-cc.mak
+ -include config-cc.mak
+ 
+ # Pauth Tests
+@@ -53,7 +54,11 @@ endif
+ ifneq ($(CROSS_CC_HAS_SVE),)
+ # System Registers Tests
+ AARCH64_TESTS += sysregs
++ifneq ($(CROSS_CC_HAS_ARMV9_SME),)
++sysregs: CFLAGS+=-march=armv9-a+sme -DHAS_ARMV9_SME
++else
+ sysregs: CFLAGS+=-march=armv8.1-a+sve
++endif
+ 
+ # SVE ioctl test
+ AARCH64_TESTS += sve-ioctls
+diff --git a/tests/tcg/aarch64/sysregs.c b/tests/tcg/aarch64/sysregs.c
+index 40cf8d2877..46b931f781 100644
+--- a/tests/tcg/aarch64/sysregs.c
++++ b/tests/tcg/aarch64/sysregs.c
+@@ -22,6 +22,13 @@
+ #define HWCAP_CPUID (1 << 11)
+ #endif
+ 
++/*
++ * Older assemblers don't recognize newer system register names,
++ * but we can still access them by the Sn_n_Cn_Cn_n syntax.
++ */
++#define SYS_ID_AA64ISAR2_EL1 S3_0_C0_C6_2
++#define SYS_ID_AA64MMFR2_EL1 S3_0_C0_C7_2
++
+ int failed_bit_count;
+ 
+ /* Read and print system register `id' value */
+@@ -112,18 +119,23 @@ int main(void)
+      * minimum valid fields - for the purposes of this check allowed
+      * to have non-zero values.
+      */
+-    get_cpu_reg_check_mask(id_aa64isar0_el1, _m(00ff,ffff,f0ff,fff0));
+-    get_cpu_reg_check_mask(id_aa64isar1_el1, _m(0000,00f0,ffff,ffff));
++    get_cpu_reg_check_mask(id_aa64isar0_el1, _m(f0ff,ffff,f0ff,fff0));
++    get_cpu_reg_check_mask(id_aa64isar1_el1, _m(00ff,f0ff,ffff,ffff));
++    get_cpu_reg_check_mask(SYS_ID_AA64ISAR2_EL1, _m(0000,0000,0000,ffff));
+     /* TGran4 & TGran64 as pegged to -1 */
+-    get_cpu_reg_check_mask(id_aa64mmfr0_el1, _m(0000,0000,ff00,0000));
+-    get_cpu_reg_check_zero(id_aa64mmfr1_el1);
++    get_cpu_reg_check_mask(id_aa64mmfr0_el1, _m(f000,0000,ff00,0000));
++    get_cpu_reg_check_mask(id_aa64mmfr1_el1, _m(0000,f000,0000,0000));
++    get_cpu_reg_check_mask(SYS_ID_AA64MMFR2_EL1, _m(0000,000f,0000,0000));
+     /* EL1/EL0 reported as AA64 only */
+     get_cpu_reg_check_mask(id_aa64pfr0_el1,  _m(000f,000f,00ff,0011));
+-    get_cpu_reg_check_mask(id_aa64pfr1_el1,  _m(0000,0000,0000,00f0));
++    get_cpu_reg_check_mask(id_aa64pfr1_el1,  _m(0000,0000,0f00,0fff));
+     /* all hidden, DebugVer fixed to 0x6 (ARMv8 debug architecture) */
+     get_cpu_reg_check_mask(id_aa64dfr0_el1,  _m(0000,0000,0000,0006));
+     get_cpu_reg_check_zero(id_aa64dfr1_el1);
+-    get_cpu_reg_check_zero(id_aa64zfr0_el1);
++    get_cpu_reg_check_mask(id_aa64zfr0_el1,  _m(0ff0,ff0f,00ff,00ff));
++#ifdef HAS_ARMV9_SME
++    get_cpu_reg_check_mask(id_aa64smfr0_el1, _m(80f1,00fd,0000,0000));
++#endif
+ 
+     get_cpu_reg_check_zero(id_aa64afr0_el1);
+     get_cpu_reg_check_zero(id_aa64afr1_el1);

debian/patches/extra/0006-tests-tcg-aarch64-sysregs.c-Use-S-syntax-for-id_aa64.patch

@@ -0,0 +1,91 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Peter Maydell <peter.maydell@linaro.org>
+Date: Wed, 10 Apr 2024 08:43:25 +0300
+Subject: [PATCH] tests/tcg/aarch64/sysregs.c: Use S syntax for id_aa64zfr0_el1
+ and id_aa64smfr0_el1
+
+Some assemblers will complain about attempts to access
+id_aa64zfr0_el1 and id_aa64smfr0_el1 by name if the test
+binary isn't built for the right processor type:
+
+ /tmp/ccASXpLo.s:782: Error: selected processor does not support system register name 'id_aa64zfr0_el1'
+ /tmp/ccASXpLo.s:829: Error: selected processor does not support system register name 'id_aa64smfr0_el1'
+
+However, these registers are in the ID space and are guaranteed to
+read-as-zero on older CPUs, so the access is both safe and sensible.
+Switch to using the S syntax, as we already do for ID_AA64ISAR2_EL1
+and ID_AA64MMFR2_EL1.  This allows us to drop the HAS_ARMV9_SME check
+and the makefile machinery to adjust the CFLAGS for this test, so we
+don't rely on having a sufficiently new compiler to be able to check
+these registers.
+
+This means we're actually testing the SME ID register: no released
+GCC yet recognizes -march=armv9-a+sme, so that was always skipped.
+It also avoids a future problem if we try to switch the "do we have
+SME support in the toolchain" check from "in the compiler" to "in the
+assembler" (at which point we would otherwise run into the above
+errors).
+
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+(cherry picked from commit 3dc2afeab2964b54848715b913b6c605f36be3e1)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+(Mjt: pick this for v8.0.0-2361-g1f51573f79
+ "target/arm: Fix SME full tile indexing")
+---
+ tests/tcg/aarch64/Makefile.target |  7 +------
+ tests/tcg/aarch64/sysregs.c       | 11 +++++++----
+ 2 files changed, 8 insertions(+), 10 deletions(-)
+
+diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
+index fc6d5d824d..118d069073 100644
+--- a/tests/tcg/aarch64/Makefile.target
++++ b/tests/tcg/aarch64/Makefile.target
+@@ -51,15 +51,10 @@ AARCH64_TESTS += mte-1 mte-2 mte-3 mte-4 mte-5 mte-6 mte-7
+ mte-%: CFLAGS += -march=armv8.5-a+memtag
+ endif
+ 
+-ifneq ($(CROSS_CC_HAS_SVE),)
+ # System Registers Tests
+ AARCH64_TESTS += sysregs
+-ifneq ($(CROSS_CC_HAS_ARMV9_SME),)
+-sysregs: CFLAGS+=-march=armv9-a+sme -DHAS_ARMV9_SME
+-else
+-sysregs: CFLAGS+=-march=armv8.1-a+sve
+-endif
+ 
++ifneq ($(CROSS_CC_HAS_SVE),)
+ # SVE ioctl test
+ AARCH64_TESTS += sve-ioctls
+ sve-ioctls: CFLAGS+=-march=armv8.1-a+sve
+diff --git a/tests/tcg/aarch64/sysregs.c b/tests/tcg/aarch64/sysregs.c
+index 46b931f781..d8eb06abcf 100644
+--- a/tests/tcg/aarch64/sysregs.c
++++ b/tests/tcg/aarch64/sysregs.c
+@@ -25,9 +25,14 @@
+ /*
+  * Older assemblers don't recognize newer system register names,
+  * but we can still access them by the Sn_n_Cn_Cn_n syntax.
++ * This also means we don't need to specifically request that the
++ * assembler enables whatever architectural features the ID registers
++ * syntax might be gated behind.
+  */
+ #define SYS_ID_AA64ISAR2_EL1 S3_0_C0_C6_2
+ #define SYS_ID_AA64MMFR2_EL1 S3_0_C0_C7_2
++#define SYS_ID_AA64ZFR0_EL1 S3_0_C0_C4_4
++#define SYS_ID_AA64SMFR0_EL1 S3_0_C0_C4_5
+ 
+ int failed_bit_count;
+ 
+@@ -132,10 +137,8 @@ int main(void)
+     /* all hidden, DebugVer fixed to 0x6 (ARMv8 debug architecture) */
+     get_cpu_reg_check_mask(id_aa64dfr0_el1,  _m(0000,0000,0000,0006));
+     get_cpu_reg_check_zero(id_aa64dfr1_el1);
+-    get_cpu_reg_check_mask(id_aa64zfr0_el1,  _m(0ff0,ff0f,00ff,00ff));
+-#ifdef HAS_ARMV9_SME
+-    get_cpu_reg_check_mask(id_aa64smfr0_el1, _m(80f1,00fd,0000,0000));
+-#endif
++    get_cpu_reg_check_mask(SYS_ID_AA64ZFR0_EL1,  _m(0ff0,ff0f,00ff,00ff));
++    get_cpu_reg_check_mask(SYS_ID_AA64SMFR0_EL1, _m(80f1,00fd,0000,0000));
+ 
+     get_cpu_reg_check_zero(id_aa64afr0_el1);
+     get_cpu_reg_check_zero(id_aa64afr1_el1);

debian/patches/extra/0007-target-arm-Fix-SME-full-tile-indexing.patch

@@ -0,0 +1,199 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Richard Henderson <richard.henderson@linaro.org>
+Date: Wed, 10 Apr 2024 08:43:26 +0300
+Subject: [PATCH] target/arm: Fix SME full tile indexing
+
+For the outer product set of insns, which take an entire matrix
+tile as output, the argument is not a combined tile+column.
+Therefore using get_tile_rowcol was incorrect, as we extracted
+the tile number from itself.
+
+The test case relies only on assembler support for SME, since
+no release of GCC recognizes -march=armv9-a+sme yet.
+
+Cc: qemu-stable@nongnu.org
+Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1620
+Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+Message-id: 20230622151201.1578522-5-richard.henderson@linaro.org
+Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
+[PMM: dropped now-unneeded changes to sysregs CFLAGS]
+Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
+(cherry picked from commit 1f51573f7925b80e79a29f87c7d9d6ead60960c0)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ target/arm/translate-sme.c        | 24 ++++++---
+ tests/tcg/aarch64/Makefile.target |  7 ++-
+ tests/tcg/aarch64/sme-outprod1.c  | 83 +++++++++++++++++++++++++++++++
+ 3 files changed, 107 insertions(+), 7 deletions(-)
+ create mode 100644 tests/tcg/aarch64/sme-outprod1.c
+
+diff --git a/target/arm/translate-sme.c b/target/arm/translate-sme.c
+index 7b87a9df63..65f8495bdd 100644
+--- a/target/arm/translate-sme.c
++++ b/target/arm/translate-sme.c
+@@ -103,6 +103,21 @@ static TCGv_ptr get_tile_rowcol(DisasContext *s, int esz, int rs,
+     return addr;
+ }
+ 
++/*
++ * Resolve tile.size[0] to a host pointer.
++ * Used by e.g. outer product insns where we require the entire tile.
++ */
++static TCGv_ptr get_tile(DisasContext *s, int esz, int tile)
++{
++    TCGv_ptr addr = tcg_temp_new_ptr();
++    int offset;
++
++    offset = tile * sizeof(ARMVectorReg) + offsetof(CPUARMState, zarray);
++
++    tcg_gen_addi_ptr(addr, cpu_env, offset);
++    return addr;
++}
++
+ static bool trans_ZERO(DisasContext *s, arg_ZERO *a)
+ {
+     if (!dc_isar_feature(aa64_sme, s)) {
+@@ -279,8 +294,7 @@ static bool do_adda(DisasContext *s, arg_adda *a, MemOp esz,
+         return true;
+     }
+ 
+-    /* Sum XZR+zad to find ZAd. */
+-    za = get_tile_rowcol(s, esz, 31, a->zad, false);
++    za = get_tile(s, esz, a->zad);
+     zn = vec_full_reg_ptr(s, a->zn);
+     pn = pred_full_reg_ptr(s, a->pn);
+     pm = pred_full_reg_ptr(s, a->pm);
+@@ -310,8 +324,7 @@ static bool do_outprod(DisasContext *s, arg_op *a, MemOp esz,
+         return true;
+     }
+ 
+-    /* Sum XZR+zad to find ZAd. */
+-    za = get_tile_rowcol(s, esz, 31, a->zad, false);
++    za = get_tile(s, esz, a->zad);
+     zn = vec_full_reg_ptr(s, a->zn);
+     zm = vec_full_reg_ptr(s, a->zm);
+     pn = pred_full_reg_ptr(s, a->pn);
+@@ -337,8 +350,7 @@ static bool do_outprod_fpst(DisasContext *s, arg_op *a, MemOp esz,
+         return true;
+     }
+ 
+-    /* Sum XZR+zad to find ZAd. */
+-    za = get_tile_rowcol(s, esz, 31, a->zad, false);
++    za = get_tile(s, esz, a->zad);
+     zn = vec_full_reg_ptr(s, a->zn);
+     zm = vec_full_reg_ptr(s, a->zm);
+     pn = pred_full_reg_ptr(s, a->pn);
+diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
+index 118d069073..5e4ea7c998 100644
+--- a/tests/tcg/aarch64/Makefile.target
++++ b/tests/tcg/aarch64/Makefile.target
+@@ -24,7 +24,7 @@ config-cc.mak: Makefile
+ 	    $(call cc-option,-march=armv8.3-a,              CROSS_CC_HAS_ARMV8_3); \
+ 	    $(call cc-option,-mbranch-protection=standard,  CROSS_CC_HAS_ARMV8_BTI); \
+ 	    $(call cc-option,-march=armv8.5-a+memtag,       CROSS_CC_HAS_ARMV8_MTE); \
+-	    $(call cc-option,-march=armv9-a+sme,            CROSS_CC_HAS_ARMV9_SME)) 3> config-cc.mak
++	    $(call cc-option,-Wa$(COMMA)-march=armv9-a+sme, CROSS_AS_HAS_ARMV9_SME)) 3> config-cc.mak
+ -include config-cc.mak
+ 
+ # Pauth Tests
+@@ -51,6 +51,11 @@ AARCH64_TESTS += mte-1 mte-2 mte-3 mte-4 mte-5 mte-6 mte-7
+ mte-%: CFLAGS += -march=armv8.5-a+memtag
+ endif
+ 
++# SME Tests
++ifneq ($(CROSS_AS_HAS_ARMV9_SME),)
++AARCH64_TESTS += sme-outprod1
++endif
++
+ # System Registers Tests
+ AARCH64_TESTS += sysregs
+ 
+diff --git a/tests/tcg/aarch64/sme-outprod1.c b/tests/tcg/aarch64/sme-outprod1.c
+new file mode 100644
+index 0000000000..6e5972d75e
+--- /dev/null
++++ b/tests/tcg/aarch64/sme-outprod1.c
+@@ -0,0 +1,83 @@
++/*
++ * SME outer product, 1 x 1.
++ * SPDX-License-Identifier: GPL-2.0-or-later
++ */
++
++#include <stdio.h>
++
++extern void foo(float *dst);
++
++asm(
++"	.arch_extension sme\n"
++"	.type foo, @function\n"
++"foo:\n"
++"	stp x29, x30, [sp, -80]!\n"
++"	mov x29, sp\n"
++"	stp d8, d9, [sp, 16]\n"
++"	stp d10, d11, [sp, 32]\n"
++"	stp d12, d13, [sp, 48]\n"
++"	stp d14, d15, [sp, 64]\n"
++"	smstart\n"
++"	ptrue p0.s, vl4\n"
++"	fmov z0.s, #1.0\n"
++/*
++ * An outer product of a vector of 1.0 by itself should be a matrix of 1.0.
++ * Note that we are using tile 1 here (za1.s) rather than tile 0.
++ */
++"	zero {za}\n"
++"	fmopa za1.s, p0/m, p0/m, z0.s, z0.s\n"
++/*
++ * Read the first 4x4 sub-matrix of elements from tile 1:
++ * Note that za1h should be interchangable here.
++ */
++"	mov w12, #0\n"
++"	mova z0.s, p0/m, za1v.s[w12, #0]\n"
++"	mova z1.s, p0/m, za1v.s[w12, #1]\n"
++"	mova z2.s, p0/m, za1v.s[w12, #2]\n"
++"	mova z3.s, p0/m, za1v.s[w12, #3]\n"
++/*
++ * And store them to the input pointer (dst in the C code):
++ */
++"	st1w {z0.s}, p0, [x0]\n"
++"	add x0, x0, #16\n"
++"	st1w {z1.s}, p0, [x0]\n"
++"	add x0, x0, #16\n"
++"	st1w {z2.s}, p0, [x0]\n"
++"	add x0, x0, #16\n"
++"	st1w {z3.s}, p0, [x0]\n"
++"	smstop\n"
++"	ldp d8, d9, [sp, 16]\n"
++"	ldp d10, d11, [sp, 32]\n"
++"	ldp d12, d13, [sp, 48]\n"
++"	ldp d14, d15, [sp, 64]\n"
++"	ldp x29, x30, [sp], 80\n"
++"	ret\n"
++"	.size foo, . - foo"
++);
++
++int main()
++{
++    float dst[16];
++    int i, j;
++
++    foo(dst);
++
++    for (i = 0; i < 16; i++) {
++        if (dst[i] != 1.0f) {
++            break;
++        }
++    }
++
++    if (i == 16) {
++        return 0; /* success */
++    }
++
++    /* failure */
++    for (i = 0; i < 4; ++i) {
++        for (j = 0; j < 4; ++j) {
++            printf("%f ", (double)dst[i * 4 + j]);
++        }
++        printf("\n");
++    }
++    return 1;
++}

debian/patches/extra/0008-system-qdev-monitor-move-drain_call_rcu-call-under-i.patch

@@ -0,0 +1,61 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dmitrii Gavrilov <ds-gavr@yandex-team.ru>
+Date: Wed, 10 Apr 2024 08:43:28 +0300
+Subject: [PATCH] system/qdev-monitor: move drain_call_rcu call under if (!dev)
+ in qmp_device_add()
+
+Original goal of addition of drain_call_rcu to qmp_device_add was to cover
+the failure case of qdev_device_add. It seems call of drain_call_rcu was
+misplaced in 7bed89958bfbf40df what led to waiting for pending RCU callbacks
+under happy path too. What led to overall performance degradation of
+qmp_device_add.
+
+In this patch call of drain_call_rcu moved under handling of failure of
+qdev_device_add.
+
+Signed-off-by: Dmitrii Gavrilov <ds-gavr@yandex-team.ru>
+Message-ID: <20231103105602.90475-1-ds-gavr@yandex-team.ru>
+Fixes: 7bed89958bf ("device_core: use drain_call_rcu in in qmp_device_add", 2020-10-12)
+Cc: qemu-stable@nongnu.org
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit 012b170173bcaa14b9bc26209e0813311ac78489)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ softmmu/qdev-monitor.c | 23 +++++++++++------------
+ 1 file changed, 11 insertions(+), 12 deletions(-)
+
+diff --git a/softmmu/qdev-monitor.c b/softmmu/qdev-monitor.c
+index 4b0ef65780..f4348443b0 100644
+--- a/softmmu/qdev-monitor.c
++++ b/softmmu/qdev-monitor.c
+@@ -853,19 +853,18 @@ void qmp_device_add(QDict *qdict, QObject **ret_data, Error **errp)
+         return;
+     }
+     dev = qdev_device_add(opts, errp);
+-
+-    /*
+-     * Drain all pending RCU callbacks. This is done because
+-     * some bus related operations can delay a device removal
+-     * (in this case this can happen if device is added and then
+-     * removed due to a configuration error)
+-     * to a RCU callback, but user might expect that this interface
+-     * will finish its job completely once qmp command returns result
+-     * to the user
+-     */
+-    drain_call_rcu();
+-
+     if (!dev) {
++        /*
++         * Drain all pending RCU callbacks. This is done because
++         * some bus related operations can delay a device removal
++         * (in this case this can happen if device is added and then
++         * removed due to a configuration error)
++         * to a RCU callback, but user might expect that this interface
++         * will finish its job completely once qmp command returns result
++         * to the user
++         */
++        drain_call_rcu();
++
+         qemu_opts_del(opts);
+         return;
+     }

debian/patches/extra/0009-hw-scsi-lsi53c895a-stop-script-on-phase-mismatch.patch

@@ -0,0 +1,85 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Sven Schnelle <svens@stackframe.org>
+Date: Wed, 10 Apr 2024 08:43:29 +0300
+Subject: [PATCH] hw/scsi/lsi53c895a: stop script on phase mismatch
+
+Netbsd isn't happy with qemu lsi53c895a emulation:
+
+cd0(esiop0:0:2:0): command with tag id 0 reset
+esiop0: autoconfiguration error: phase mismatch without command
+esiop0: autoconfiguration error: unhandled scsi interrupt, sist=0x80 sstat1=0x0 DSA=0x23a64b1 DSP=0x50
+
+This is because lsi_bad_phase() triggers a phase mismatch, which
+stops SCRIPT processing. However, after returning to
+lsi_command_complete(), SCRIPT is restarted with lsi_resume_script().
+Fix this by adding a return value to lsi_bad_phase(), and only resume
+script processing when lsi_bad_phase() didn't trigger a host interrupt.
+
+Signed-off-by: Sven Schnelle <svens@stackframe.org>
+Tested-by: Helge Deller <deller@gmx.de>
+Message-ID: <20240302214453.2071388-1-svens@stackframe.org>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit a9198b3132d81a6bfc9fdbf6f3d3a514c2864674)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ hw/scsi/lsi53c895a.c | 16 ++++++++++++----
+ 1 file changed, 12 insertions(+), 4 deletions(-)
+
+diff --git a/hw/scsi/lsi53c895a.c b/hw/scsi/lsi53c895a.c
+index ca619ed564..905f5ef237 100644
+--- a/hw/scsi/lsi53c895a.c
++++ b/hw/scsi/lsi53c895a.c
+@@ -570,8 +570,9 @@ static inline void lsi_set_phase(LSIState *s, int phase)
+     s->sstat1 = (s->sstat1 & ~PHASE_MASK) | phase;
+ }
+ 
+-static void lsi_bad_phase(LSIState *s, int out, int new_phase)
++static int lsi_bad_phase(LSIState *s, int out, int new_phase)
+ {
++    int ret = 0;
+     /* Trigger a phase mismatch.  */
+     if (s->ccntl0 & LSI_CCNTL0_ENPMJ) {
+         if ((s->ccntl0 & LSI_CCNTL0_PMJCTL)) {
+@@ -584,8 +585,10 @@ static void lsi_bad_phase(LSIState *s, int out, int new_phase)
+         trace_lsi_bad_phase_interrupt();
+         lsi_script_scsi_interrupt(s, LSI_SIST0_MA, 0);
+         lsi_stop_script(s);
++        ret = 1;
+     }
+     lsi_set_phase(s, new_phase);
++    return ret;
+ }
+ 
+ 
+@@ -789,7 +792,7 @@ static int lsi_queue_req(LSIState *s, SCSIRequest *req, uint32_t len)
+ static void lsi_command_complete(SCSIRequest *req, size_t resid)
+ {
+     LSIState *s = LSI53C895A(req->bus->qbus.parent);
+-    int out;
++    int out, stop = 0;
+ 
+     out = (s->sstat1 & PHASE_MASK) == PHASE_DO;
+     trace_lsi_command_complete(req->status);
+@@ -797,7 +800,10 @@ static void lsi_command_complete(SCSIRequest *req, size_t resid)
+     s->command_complete = 2;
+     if (s->waiting && s->dbc != 0) {
+         /* Raise phase mismatch for short transfers.  */
+-        lsi_bad_phase(s, out, PHASE_ST);
++        stop = lsi_bad_phase(s, out, PHASE_ST);
++        if (stop) {
++            s->waiting = 0;
++        }
+     } else {
+         lsi_set_phase(s, PHASE_ST);
+     }
+@@ -807,7 +813,9 @@ static void lsi_command_complete(SCSIRequest *req, size_t resid)
+         lsi_request_free(s, s->current);
+         scsi_req_unref(req);
+     }
+-    lsi_resume_script(s);
++    if (!stop) {
++        lsi_resume_script(s);
++    }
+ }
+ 
+  /* Callback to indicate that the SCSI layer has completed a transfer.  */

debian/patches/extra/0010-hw-scsi-lsi53c895a-add-missing-decrement-of-reentran.patch

@@ -0,0 +1,38 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Sven Schnelle <svens@stackframe.org>
+Date: Wed, 10 Apr 2024 08:43:30 +0300
+Subject: [PATCH] hw/scsi/lsi53c895a: add missing decrement of reentrancy
+ counter
+
+When the maximum count of SCRIPTS instructions is reached, the code
+stops execution and returns, but fails to decrement the reentrancy
+counter. This effectively renders the SCSI controller unusable
+because on next entry the reentrancy counter is still above the limit.
+
+This bug was seen on HP-UX 10.20 which seems to trigger SCRIPTS
+loops.
+
+Fixes: b987718bbb ("hw/scsi/lsi53c895a: Fix reentrancy issues in the LSI controller (CVE-2023-0330)")
+Signed-off-by: Sven Schnelle <svens@stackframe.org>
+Message-ID: <20240128202214.2644768-1-svens@stackframe.org>
+Reviewed-by: Thomas Huth <thuth@redhat.com>
+Tested-by: Helge Deller <deller@gmx.de>
+Signed-off-by: Thomas Huth <thuth@redhat.com>
+(cherry picked from commit 8b09b7fe47082c69295a0fc0cc01b041b6385025)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ hw/scsi/lsi53c895a.c | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/hw/scsi/lsi53c895a.c b/hw/scsi/lsi53c895a.c
+index 905f5ef237..c7a3964b5f 100644
+--- a/hw/scsi/lsi53c895a.c
++++ b/hw/scsi/lsi53c895a.c
+@@ -1167,6 +1167,7 @@ again:
+         lsi_script_scsi_interrupt(s, LSI_SIST0_UDC, 0);
+         lsi_disconnect(s);
+         trace_lsi_execute_script_stop();
++        reentrancy_level--;
+         return;
+     }
+     insn = read_dword(s, s->dsp);

debian/patches/extra/0011-hw-scsi-lsi53c895a-add-timer-to-scripts-processing.patch

@@ -0,0 +1,173 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Sven Schnelle <svens@stackframe.org>
+Date: Wed, 10 Apr 2024 08:43:31 +0300
+Subject: [PATCH] hw/scsi/lsi53c895a: add timer to scripts processing
+
+HP-UX 10.20 seems to make the lsi53c895a spinning on a memory location
+under certain circumstances. As the SCSI controller and CPU are not
+running at the same time this loop will never finish. After some
+time, the check loop interrupts with a unexpected device disconnect.
+This works, but is slow because the kernel resets the scsi controller.
+Instead of signaling UDC, start a timer and exit the loop. Until the
+timer fires, the CPU can process instructions which might changes the
+memory location.
+
+The limit of instructions is also reduced because scripts running on
+the SCSI processor are usually very short. This keeps the time until
+the loop is exit short.
+
+Suggested-by: Peter Maydell <peter.maydell@linaro.org>
+Signed-off-by: Sven Schnelle <svens@stackframe.org>
+Message-ID: <20240229204407.1699260-1-svens@stackframe.org>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit 9876359990dd4c8a48de65cf5e1c3d13e96a7f4e)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ hw/scsi/lsi53c895a.c | 43 +++++++++++++++++++++++++++++++++----------
+ hw/scsi/trace-events |  2 ++
+ 2 files changed, 35 insertions(+), 10 deletions(-)
+
+diff --git a/hw/scsi/lsi53c895a.c b/hw/scsi/lsi53c895a.c
+index c7a3964b5f..48c85d479c 100644
+--- a/hw/scsi/lsi53c895a.c
++++ b/hw/scsi/lsi53c895a.c
+@@ -188,7 +188,7 @@ static const char *names[] = {
+ #define LSI_TAG_VALID     (1 << 16)
+ 
+ /* Maximum instructions to process. */
+-#define LSI_MAX_INSN    10000
++#define LSI_MAX_INSN    100
+ 
+ typedef struct lsi_request {
+     SCSIRequest *req;
+@@ -205,6 +205,7 @@ enum {
+     LSI_WAIT_RESELECT, /* Wait Reselect instruction has been issued */
+     LSI_DMA_SCRIPTS, /* processing DMA from lsi_execute_script */
+     LSI_DMA_IN_PROGRESS, /* DMA operation is in progress */
++    LSI_WAIT_SCRIPTS, /* SCRIPTS stopped because of instruction count limit */
+ };
+ 
+ enum {
+@@ -224,6 +225,7 @@ struct LSIState {
+     MemoryRegion ram_io;
+     MemoryRegion io_io;
+     AddressSpace pci_io_as;
++    QEMUTimer *scripts_timer;
+ 
+     int carry; /* ??? Should this be an a visible register somewhere?  */
+     int status;
+@@ -415,6 +417,7 @@ static void lsi_soft_reset(LSIState *s)
+     s->sbr = 0;
+     assert(QTAILQ_EMPTY(&s->queue));
+     assert(!s->current);
++    timer_del(s->scripts_timer);
+ }
+ 
+ static int lsi_dma_40bit(LSIState *s)
+@@ -1135,6 +1138,12 @@ static void lsi_wait_reselect(LSIState *s)
+     }
+ }
+ 
++static void lsi_scripts_timer_start(LSIState *s)
++{
++    trace_lsi_scripts_timer_start();
++    timer_mod(s->scripts_timer, qemu_clock_get_us(QEMU_CLOCK_VIRTUAL) + 500);
++}
++
+ static void lsi_execute_script(LSIState *s)
+ {
+     PCIDevice *pci_dev = PCI_DEVICE(s);
+@@ -1144,6 +1153,11 @@ static void lsi_execute_script(LSIState *s)
+     int insn_processed = 0;
+     static int reentrancy_level;
+ 
++    if (s->waiting == LSI_WAIT_SCRIPTS) {
++        timer_del(s->scripts_timer);
++        s->waiting = LSI_NOWAIT;
++    }
++
+     reentrancy_level++;
+ 
+     s->istat1 |= LSI_ISTAT1_SRUN;
+@@ -1151,8 +1165,8 @@ again:
+     /*
+      * Some windows drivers make the device spin waiting for a memory location
+      * to change. If we have executed more than LSI_MAX_INSN instructions then
+-     * assume this is the case and force an unexpected device disconnect. This
+-     * is apparently sufficient to beat the drivers into submission.
++     * assume this is the case and start a timer. Until the timer fires, the
++     * host CPU has a chance to run and change the memory location.
+      *
+      * Another issue (CVE-2023-0330) can occur if the script is programmed to
+      * trigger itself again and again. Avoid this problem by stopping after
+@@ -1160,13 +1174,8 @@ again:
+      * which should be enough for all valid use cases).
+      */
+     if (++insn_processed > LSI_MAX_INSN || reentrancy_level > 8) {
+-        if (!(s->sien0 & LSI_SIST0_UDC)) {
+-            qemu_log_mask(LOG_GUEST_ERROR,
+-                          "lsi_scsi: inf. loop with UDC masked");
+-        }
+-        lsi_script_scsi_interrupt(s, LSI_SIST0_UDC, 0);
+-        lsi_disconnect(s);
+-        trace_lsi_execute_script_stop();
++        s->waiting = LSI_WAIT_SCRIPTS;
++        lsi_scripts_timer_start(s);
+         reentrancy_level--;
+         return;
+     }
+@@ -2205,6 +2214,9 @@ static int lsi_post_load(void *opaque, int version_id)
+         return -EINVAL;
+     }
+ 
++    if (s->waiting == LSI_WAIT_SCRIPTS) {
++        lsi_scripts_timer_start(s);
++    }
+     return 0;
+ }
+ 
+@@ -2302,6 +2314,15 @@ static const struct SCSIBusInfo lsi_scsi_info = {
+     .cancel = lsi_request_cancelled
+ };
+ 
++static void scripts_timer_cb(void *opaque)
++{
++    LSIState *s = opaque;
++
++    trace_lsi_scripts_timer_triggered();
++    s->waiting = LSI_NOWAIT;
++    lsi_execute_script(s);
++}
++
+ static void lsi_scsi_realize(PCIDevice *dev, Error **errp)
+ {
+     LSIState *s = LSI53C895A(dev);
+@@ -2321,6 +2342,7 @@ static void lsi_scsi_realize(PCIDevice *dev, Error **errp)
+                           "lsi-ram", 0x2000);
+     memory_region_init_io(&s->io_io, OBJECT(s), &lsi_io_ops, s,
+                           "lsi-io", 256);
++    s->scripts_timer = timer_new_us(QEMU_CLOCK_VIRTUAL, scripts_timer_cb, s);
+ 
+     /*
+      * Since we use the address-space API to interact with ram_io, disable the
+@@ -2345,6 +2367,7 @@ static void lsi_scsi_exit(PCIDevice *dev)
+     LSIState *s = LSI53C895A(dev);
+ 
+     address_space_destroy(&s->pci_io_as);
++    timer_del(s->scripts_timer);
+ }
+ 
+ static void lsi_class_init(ObjectClass *klass, void *data)
+diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events
+index ab238293f0..131af99d91 100644
+--- a/hw/scsi/trace-events
++++ b/hw/scsi/trace-events
+@@ -299,6 +299,8 @@ lsi_execute_script_stop(void) "SCRIPTS execution stopped"
+ lsi_awoken(void) "Woken by SIGP"
+ lsi_reg_read(const char *name, int offset, uint8_t ret) "Read reg %s 0x%x = 0x%02x"
+ lsi_reg_write(const char *name, int offset, uint8_t val) "Write reg %s 0x%x = 0x%02x"
++lsi_scripts_timer_triggered(void) "SCRIPTS timer triggered"
++lsi_scripts_timer_start(void) "SCRIPTS timer started"
+ 
+ # virtio-scsi.c
+ virtio_scsi_cmd_req(int lun, uint32_t tag, uint8_t cmd) "virtio_scsi_cmd_req lun=%u tag=0x%x cmd=0x%x"

debian/patches/extra/0012-e1000e-fix-link-state-on-resume.patch

@@ -0,0 +1,161 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Laurent Vivier <lvivier@redhat.com>
+Date: Wed, 10 Apr 2024 08:43:33 +0300
+Subject: [PATCH] e1000e: fix link state on resume
+
+On resume e1000e_vm_state_change() always calls e1000e_autoneg_resume()
+that sets link_down to false, and thus activates the link even
+if we have disabled it.
+
+The problem can be reproduced starting qemu in paused state (-S) and
+then set the link to down. When we resume the machine the link appears
+to be up.
+
+Reproducer:
+
+   # qemu-system-x86_64 ... -device e1000e,netdev=netdev0,id=net0 -S
+
+   {"execute": "qmp_capabilities" }
+   {"execute": "set_link", "arguments": {"name": "net0", "up": false}}
+   {"execute": "cont" }
+
+To fix the problem, merge the content of e1000e_vm_state_change()
+into e1000e_core_post_load() as e1000 does.
+
+Buglink: https://issues.redhat.com/browse/RHEL-21867
+Fixes: 6f3fbe4ed06a ("net: Introduce e1000e device emulation")
+Suggested-by: Akihiko Odaki <akihiko.odaki@daynix.com>
+Signed-off-by: Laurent Vivier <lvivier@redhat.com>
+Signed-off-by: Jason Wang <jasowang@redhat.com>
+(cherry picked from commit 4cadf10234989861398e19f3bb441d3861f3bb7c)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ hw/net/e1000e_core.c | 60 ++++++--------------------------------------
+ hw/net/e1000e_core.h |  2 --
+ 2 files changed, 7 insertions(+), 55 deletions(-)
+
+diff --git a/hw/net/e1000e_core.c b/hw/net/e1000e_core.c
+index c71d82ce1d..742f5ec800 100644
+--- a/hw/net/e1000e_core.c
++++ b/hw/net/e1000e_core.c
+@@ -108,14 +108,6 @@ e1000e_intmgr_timer_resume(E1000IntrDelayTimer *timer)
+     }
+ }
+ 
+-static void
+-e1000e_intmgr_timer_pause(E1000IntrDelayTimer *timer)
+-{
+-    if (timer->running) {
+-        timer_del(timer->timer);
+-    }
+-}
+-
+ static inline void
+ e1000e_intrmgr_stop_timer(E1000IntrDelayTimer *timer)
+ {
+@@ -397,24 +389,6 @@ e1000e_intrmgr_resume(E1000ECore *core)
+     }
+ }
+ 
+-static void
+-e1000e_intrmgr_pause(E1000ECore *core)
+-{
+-    int i;
+-
+-    e1000e_intmgr_timer_pause(&core->radv);
+-    e1000e_intmgr_timer_pause(&core->rdtr);
+-    e1000e_intmgr_timer_pause(&core->raid);
+-    e1000e_intmgr_timer_pause(&core->tidv);
+-    e1000e_intmgr_timer_pause(&core->tadv);
+-
+-    e1000e_intmgr_timer_pause(&core->itr);
+-
+-    for (i = 0; i < E1000E_MSIX_VEC_NUM; i++) {
+-        e1000e_intmgr_timer_pause(&core->eitr[i]);
+-    }
+-}
+-
+ static void
+ e1000e_intrmgr_reset(E1000ECore *core)
+ {
+@@ -3336,12 +3310,6 @@ e1000e_core_read(E1000ECore *core, hwaddr addr, unsigned size)
+     return 0;
+ }
+ 
+-static inline void
+-e1000e_autoneg_pause(E1000ECore *core)
+-{
+-    timer_del(core->autoneg_timer);
+-}
+-
+ static void
+ e1000e_autoneg_resume(E1000ECore *core)
+ {
+@@ -3353,22 +3321,6 @@ e1000e_autoneg_resume(E1000ECore *core)
+     }
+ }
+ 
+-static void
+-e1000e_vm_state_change(void *opaque, bool running, RunState state)
+-{
+-    E1000ECore *core = opaque;
+-
+-    if (running) {
+-        trace_e1000e_vm_state_running();
+-        e1000e_intrmgr_resume(core);
+-        e1000e_autoneg_resume(core);
+-    } else {
+-        trace_e1000e_vm_state_stopped();
+-        e1000e_autoneg_pause(core);
+-        e1000e_intrmgr_pause(core);
+-    }
+-}
+-
+ void
+ e1000e_core_pci_realize(E1000ECore     *core,
+                         const uint16_t *eeprom_templ,
+@@ -3381,9 +3333,6 @@ e1000e_core_pci_realize(E1000ECore     *core,
+                                        e1000e_autoneg_timer, core);
+     e1000e_intrmgr_pci_realize(core);
+ 
+-    core->vmstate =
+-        qemu_add_vm_change_state_handler(e1000e_vm_state_change, core);
+-
+     for (i = 0; i < E1000E_NUM_QUEUES; i++) {
+         net_tx_pkt_init(&core->tx[i].tx_pkt, core->owner,
+                         E1000E_MAX_TX_FRAGS, core->has_vnet);
+@@ -3408,8 +3357,6 @@ e1000e_core_pci_uninit(E1000ECore *core)
+ 
+     e1000e_intrmgr_pci_unint(core);
+ 
+-    qemu_del_vm_change_state_handler(core->vmstate);
+-
+     for (i = 0; i < E1000E_NUM_QUEUES; i++) {
+         net_tx_pkt_reset(core->tx[i].tx_pkt);
+         net_tx_pkt_uninit(core->tx[i].tx_pkt);
+@@ -3561,5 +3508,12 @@ e1000e_core_post_load(E1000ECore *core)
+      */
+     nc->link_down = (core->mac[STATUS] & E1000_STATUS_LU) == 0;
+ 
++    /*
++     * we need to restart intrmgr timers, as an older version of
++     * QEMU can have stopped them before migration
++     */
++    e1000e_intrmgr_resume(core);
++    e1000e_autoneg_resume(core);
++
+     return 0;
+ }
+diff --git a/hw/net/e1000e_core.h b/hw/net/e1000e_core.h
+index 4ddb4d2c39..f2a8ff4a33 100644
+--- a/hw/net/e1000e_core.h
++++ b/hw/net/e1000e_core.h
+@@ -100,8 +100,6 @@ struct E1000Core {
+     E1000IntrDelayTimer eitr[E1000E_MSIX_VEC_NUM];
+     bool eitr_intr_pending[E1000E_MSIX_VEC_NUM];
+ 
+-    VMChangeStateEntry *vmstate;
+-
+     uint32_t itr_guest_value;
+     uint32_t eitr_guest_value[E1000E_MSIX_VEC_NUM];
+ 

debian/patches/extra/0013-target-i386-introduce-function-to-query-MMU-indices.patch

@@ -0,0 +1,61 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Paolo Bonzini <pbonzini@redhat.com>
+Date: Wed, 10 Apr 2024 08:43:49 +0300
+Subject: [PATCH] target/i386: introduce function to query MMU indices
+
+Remove knowledge of specific MMU indexes (other than MMU_NESTED_IDX and
+MMU_PHYS_IDX) from mmu_translate().  This will make it possible to split
+32-bit and 64-bit MMU indexes.
+
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit 5f97afe2543f09160a8d123ab6e2e8c6d98fa9ce)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+(Mjt: context fixup in target/i386/cpu.h due to other changes in that area)
+---
+ target/i386/cpu.h                    | 10 ++++++++++
+ target/i386/tcg/sysemu/excp_helper.c |  4 ++--
+ 2 files changed, 12 insertions(+), 2 deletions(-)
+
+diff --git a/target/i386/cpu.h b/target/i386/cpu.h
+index 7be047ce33..f175e18768 100644
+--- a/target/i386/cpu.h
++++ b/target/i386/cpu.h
+@@ -2195,6 +2195,16 @@ static inline int cpu_mmu_index(CPUX86State *env, bool ifetch)
+         ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
+ }
+ 
++static inline bool is_mmu_index_smap(int mmu_index)
++{
++    return mmu_index == MMU_KSMAP_IDX;
++}
++
++static inline bool is_mmu_index_user(int mmu_index)
++{
++    return mmu_index == MMU_USER_IDX;
++}
++
+ static inline bool is_mmu_index_32(int mmu_index)
+ {
+     assert(mmu_index < MMU_PHYS_IDX);
+diff --git a/target/i386/tcg/sysemu/excp_helper.c b/target/i386/tcg/sysemu/excp_helper.c
+index 5999cdedf5..553a60d976 100644
+--- a/target/i386/tcg/sysemu/excp_helper.c
++++ b/target/i386/tcg/sysemu/excp_helper.c
+@@ -135,7 +135,7 @@ static bool mmu_translate(CPUX86State *env, const TranslateParams *in,
+ {
+     const target_ulong addr = in->addr;
+     const int pg_mode = in->pg_mode;
+-    const bool is_user = (in->mmu_idx == MMU_USER_IDX);
++    const bool is_user = is_mmu_index_user(in->mmu_idx);
+     const MMUAccessType access_type = in->access_type;
+     uint64_t ptep, pte, rsvd_mask;
+     PTETranslate pte_trans = {
+@@ -355,7 +355,7 @@ do_check_protect_pse36:
+     }
+ 
+     int prot = 0;
+-    if (in->mmu_idx != MMU_KSMAP_IDX || !(ptep & PG_USER_MASK)) {
++    if (!is_mmu_index_smap(in->mmu_idx) || !(ptep & PG_USER_MASK)) {
+         prot |= PAGE_READ;
+         if ((ptep & PG_RW_MASK) || !(is_user || (pg_mode & PG_MODE_WP))) {
+             prot |= PAGE_WRITE;

debian/patches/extra/0014-target-i386-use-separate-MMU-indexes-for-32-bit-acce.patch

@@ -0,0 +1,130 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Paolo Bonzini <pbonzini@redhat.com>
+Date: Wed, 10 Apr 2024 08:43:50 +0300
+Subject: [PATCH] target/i386: use separate MMU indexes for 32-bit accesses
+
+Accesses from a 32-bit environment (32-bit code segment for instruction
+accesses, EFER.LMA==0 for processor accesses) have to mask away the
+upper 32 bits of the address.  While a bit wasteful, the easiest way
+to do so is to use separate MMU indexes.  These days, QEMU anyway is
+compiled with a fixed value for NB_MMU_MODES.  Split MMU_USER_IDX,
+MMU_KSMAP_IDX and MMU_KNOSMAP_IDX in two.
+
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit 90f641531c782c873a05895f411c05fbbbef3c49)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+(Mjt: move changes for x86_cpu_mmu_index() to cpu_mmu_index() due to missing
+ v8.2.0-1030-gace0c5fe5950 "target/i386: Populate CPUClass.mmu_index"
+ Increase NB_MMU_MODES from 5 to 8 in target/i386/cpu-param.h due to missing
+ v7.2.0-2640-gffd824f3f32d "include/exec: Set default NB_MMU_MODES to 16"
+ v7.2.0-2647-g6787318a5d86 "target/i386: Remove NB_MMU_MODES define"
+ which relaxed upper limit of MMU index for i386, since this commit starts
+ using MMU_NESTED_IDX=7.
+ Thanks Zhao Liu and Paolo Bonzini for the analisys and suggestions.
+)
+---
+ target/i386/cpu-param.h              |  2 +-
+ target/i386/cpu.h                    | 44 ++++++++++++++++++++--------
+ target/i386/tcg/sysemu/excp_helper.c |  3 +-
+ 3 files changed, 34 insertions(+), 15 deletions(-)
+
+diff --git a/target/i386/cpu-param.h b/target/i386/cpu-param.h
+index f579b16bd2..e21e472e1e 100644
+--- a/target/i386/cpu-param.h
++++ b/target/i386/cpu-param.h
+@@ -23,7 +23,7 @@
+ # define TARGET_VIRT_ADDR_SPACE_BITS  32
+ #endif
+ #define TARGET_PAGE_BITS 12
+-#define NB_MMU_MODES 5
++#define NB_MMU_MODES 8
+ 
+ #ifndef CONFIG_USER_ONLY
+ # define TARGET_TB_PCREL 1
+diff --git a/target/i386/cpu.h b/target/i386/cpu.h
+index f175e18768..73eee08f3f 100644
+--- a/target/i386/cpu.h
++++ b/target/i386/cpu.h
+@@ -2182,27 +2182,42 @@ uint64_t cpu_get_tsc(CPUX86State *env);
+ #define cpu_list x86_cpu_list
+ 
+ /* MMU modes definitions */
+-#define MMU_KSMAP_IDX   0
+-#define MMU_USER_IDX    1
+-#define MMU_KNOSMAP_IDX 2
+-#define MMU_NESTED_IDX  3
+-#define MMU_PHYS_IDX    4
++#define MMU_KSMAP64_IDX    0
++#define MMU_KSMAP32_IDX    1
++#define MMU_USER64_IDX     2
++#define MMU_USER32_IDX     3
++#define MMU_KNOSMAP64_IDX  4
++#define MMU_KNOSMAP32_IDX  5
++#define MMU_PHYS_IDX       6
++#define MMU_NESTED_IDX     7
++
++#ifdef CONFIG_USER_ONLY
++#ifdef TARGET_X86_64
++#define MMU_USER_IDX MMU_USER64_IDX
++#else
++#define MMU_USER_IDX MMU_USER32_IDX
++#endif
++#endif
+ 
+ static inline int cpu_mmu_index(CPUX86State *env, bool ifetch)
+ {
+-    return (env->hflags & HF_CPL_MASK) == 3 ? MMU_USER_IDX :
+-        (!(env->hflags & HF_SMAP_MASK) || (env->eflags & AC_MASK))
+-        ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
++    int mmu_index_32 = (env->hflags & HF_CS64_MASK) ? 1 : 0;
++    int mmu_index_base =
++        (env->hflags & HF_CPL_MASK) == 3 ? MMU_USER64_IDX :
++        !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP64_IDX :
++        (env->eflags & AC_MASK) ? MMU_KNOSMAP64_IDX : MMU_KSMAP64_IDX;
++
++    return mmu_index_base + mmu_index_32;
+ }
+ 
+ static inline bool is_mmu_index_smap(int mmu_index)
+ {
+-    return mmu_index == MMU_KSMAP_IDX;
++    return (mmu_index & ~1) == MMU_KSMAP64_IDX;
+ }
+ 
+ static inline bool is_mmu_index_user(int mmu_index)
+ {
+-    return mmu_index == MMU_USER_IDX;
++    return (mmu_index & ~1) == MMU_USER64_IDX;
+ }
+ 
+ static inline bool is_mmu_index_32(int mmu_index)
+@@ -2213,9 +2228,12 @@ static inline bool is_mmu_index_32(int mmu_index)
+ 
+ static inline int cpu_mmu_index_kernel(CPUX86State *env)
+ {
+-    return !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP_IDX :
+-        ((env->hflags & HF_CPL_MASK) < 3 && (env->eflags & AC_MASK))
+-        ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
++    int mmu_index_32 = (env->hflags & HF_LMA_MASK) ? 1 : 0;
++    int mmu_index_base =
++        !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP64_IDX :
++        ((env->hflags & HF_CPL_MASK) < 3 && (env->eflags & AC_MASK)) ? MMU_KNOSMAP64_IDX : MMU_KSMAP64_IDX;
++
++    return mmu_index_base + mmu_index_32;
+ }
+ 
+ #define CC_DST  (env->cc_dst)
+diff --git a/target/i386/tcg/sysemu/excp_helper.c b/target/i386/tcg/sysemu/excp_helper.c
+index 553a60d976..5f13252d68 100644
+--- a/target/i386/tcg/sysemu/excp_helper.c
++++ b/target/i386/tcg/sysemu/excp_helper.c
+@@ -541,7 +541,8 @@ static bool get_physical_address(CPUX86State *env, vaddr addr,
+         if (likely(use_stage2)) {
+             in.cr3 = env->nested_cr3;
+             in.pg_mode = env->nested_pg_mode;
+-            in.mmu_idx = MMU_USER_IDX;
++            in.mmu_idx =
++                env->nested_pg_mode & PG_MODE_LMA ? MMU_USER64_IDX : MMU_USER32_IDX;
+             in.ptw_idx = MMU_PHYS_IDX;
+ 
+             if (!mmu_translate(env, &in, out, err)) {

debian/patches/extra/0015-target-i386-fix-direction-of-32-bit-MMU-test.patch

@@ -0,0 +1,46 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Paolo Bonzini <pbonzini@redhat.com>
+Date: Wed, 10 Apr 2024 08:43:51 +0300
+Subject: [PATCH] target/i386: fix direction of "32-bit MMU" test
+
+The low bit of MMU indices for x86 TCG indicates whether the processor is
+in 32-bit mode and therefore linear addresses have to be masked to 32 bits.
+However, the index was computed incorrectly, leading to possible conflicts
+in the TLB for any address above 4G.
+
+Analyzed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
+Fixes: b1661801c18 ("target/i386: Fix physical address truncation", 2024-02-28)
+Fixes: 1c15f97b4f1 ("target/i386: Fix physical address truncation" in stable-7.2)
+Cc: qemu-stable@nongnu.org
+Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2206
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit 2cc68629a6fc198f4a972698bdd6477f883aedfb)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+(Mjt: move changes for x86_cpu_mmu_index() to cpu_mmu_index() due to missing
+ v8.2.0-1030-gace0c5fe59 "target/i386: Populate CPUClass.mmu_index")
+---
+ target/i386/cpu.h | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/target/i386/cpu.h b/target/i386/cpu.h
+index 73eee08f3f..326649ca99 100644
+--- a/target/i386/cpu.h
++++ b/target/i386/cpu.h
+@@ -2201,7 +2201,7 @@ uint64_t cpu_get_tsc(CPUX86State *env);
+ 
+ static inline int cpu_mmu_index(CPUX86State *env, bool ifetch)
+ {
+-    int mmu_index_32 = (env->hflags & HF_CS64_MASK) ? 1 : 0;
++    int mmu_index_32 = (env->hflags & HF_CS64_MASK) ? 0 : 1;
+     int mmu_index_base =
+         (env->hflags & HF_CPL_MASK) == 3 ? MMU_USER64_IDX :
+         !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP64_IDX :
+@@ -2228,7 +2228,7 @@ static inline bool is_mmu_index_32(int mmu_index)
+ 
+ static inline int cpu_mmu_index_kernel(CPUX86State *env)
+ {
+-    int mmu_index_32 = (env->hflags & HF_LMA_MASK) ? 1 : 0;
++    int mmu_index_32 = (env->hflags & HF_LMA_MASK) ? 0 : 1;
+     int mmu_index_base =
+         !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP64_IDX :
+         ((env->hflags & HF_CPL_MASK) < 3 && (env->eflags & AC_MASK)) ? MMU_KNOSMAP64_IDX : MMU_KSMAP64_IDX;

debian/patches/extra/0016-target-i386-Revert-monitor_puts-in-do_inject_x86_mce.patch

@@ -0,0 +1,35 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Tao Su <tao1.su@linux.intel.com>
+Date: Wed, 10 Apr 2024 08:43:52 +0300
+Subject: [PATCH] target/i386: Revert monitor_puts() in do_inject_x86_mce()
+
+monitor_puts() doesn't check the monitor pointer, but do_inject_x86_mce()
+may have a parameter with NULL monitor pointer. Revert monitor_puts() in
+do_inject_x86_mce() to fix, then the fact that we send the same message to
+monitor and log is again more obvious.
+
+Fixes: bf0c50d4aa85 (monitor: expose monitor_puts to rest of code)
+Reviwed-by: Xiaoyao Li <xiaoyao.li@intel.com>
+Reviewed-by: Markus Armbruster <armbru@redhat.com>
+Signed-off-by: Tao Su <tao1.su@linux.intel.com>
+Message-ID: <20240320083640.523287-1-tao1.su@linux.intel.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+(cherry picked from commit 7fd226b04746f0be0b636de5097f1b42338951a0)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ target/i386/helper.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/target/i386/helper.c b/target/i386/helper.c
+index 0ac2da066d..290d9d309c 100644
+--- a/target/i386/helper.c
++++ b/target/i386/helper.c
+@@ -427,7 +427,7 @@ static void do_inject_x86_mce(CPUState *cs, run_on_cpu_data data)
+         if (need_reset) {
+             emit_guest_memory_failure(MEMORY_FAILURE_ACTION_RESET, ar,
+                                       recursive);
+-            monitor_puts(params->mon, msg);
++            monitor_printf(params->mon, "%s", msg);
+             qemu_log_mask(CPU_LOG_RESET, "%s\n", msg);
+             qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET);
+             return;

debian/patches/extra/0017-tcg-optimize-Fix-sign_mask-for-logical-right-shift.patch

@@ -0,0 +1,86 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Richard Henderson <richard.henderson@linaro.org>
+Date: Wed, 10 Apr 2024 08:43:57 +0300
+Subject: [PATCH] tcg/optimize: Fix sign_mask for logical right-shift
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+The 'sign' computation is attempting to locate the sign bit that has
+been repeated, so that we can test if that bit is known zero.  That
+computation can be zero if there are no known sign repetitions.
+
+Cc: qemu-stable@nongnu.org
+Fixes: 93a967fbb57 ("tcg/optimize: Propagate sign info for shifting")
+Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2248
+Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
+Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
+(cherry picked from commit 2911e9b95f3bb03783ae5ca3e2494dc3b44a9161)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+(Mjt: trivial context fixup in tests/tcg/aarch64/Makefile.target)
+---
+ tcg/optimize.c                    |  2 +-
+ tests/tcg/aarch64/Makefile.target |  1 +
+ tests/tcg/aarch64/test-2248.c     | 28 ++++++++++++++++++++++++++++
+ 3 files changed, 30 insertions(+), 1 deletion(-)
+ create mode 100644 tests/tcg/aarch64/test-2248.c
+
+diff --git a/tcg/optimize.c b/tcg/optimize.c
+index ae081ab29c..b6f6436c74 100644
+--- a/tcg/optimize.c
++++ b/tcg/optimize.c
+@@ -1907,7 +1907,7 @@ static bool fold_shift(OptContext *ctx, TCGOp *op)
+          * will not reduced the number of input sign repetitions.
+          */
+         sign = (s_mask & -s_mask) >> 1;
+-        if (!(z_mask & sign)) {
++        if (sign && !(z_mask & sign)) {
+             ctx->s_mask = s_mask;
+         }
+         break;
+diff --git a/tests/tcg/aarch64/Makefile.target b/tests/tcg/aarch64/Makefile.target
+index 5e4ea7c998..474f61bc30 100644
+--- a/tests/tcg/aarch64/Makefile.target
++++ b/tests/tcg/aarch64/Makefile.target
+@@ -10,6 +10,7 @@ VPATH 		+= $(AARCH64_SRC)
+ 
+ # Base architecture tests
+ AARCH64_TESTS=fcvt pcalign-a64
++AARCH64_TESTS += test-2248
+ 
+ fcvt: LDFLAGS+=-lm
+ 
+diff --git a/tests/tcg/aarch64/test-2248.c b/tests/tcg/aarch64/test-2248.c
+new file mode 100644
+index 0000000000..aac2e17836
+--- /dev/null
++++ b/tests/tcg/aarch64/test-2248.c
+@@ -0,0 +1,28 @@
++/* SPDX-License-Identifier: GPL-2.0-or-later */
++/* See https://gitlab.com/qemu-project/qemu/-/issues/2248 */
++
++#include <assert.h>
++
++__attribute__((noinline))
++long test(long x, long y, long sh)
++{
++    long r;
++    asm("cmp   %1, %2\n\t"
++        "cset  x12, lt\n\t"
++        "and   w11, w12, #0xff\n\t"
++        "cmp   w11, #0\n\t"
++        "csetm x14, ne\n\t"
++        "lsr   x13, x14, %3\n\t"
++        "sxtb  %0, w13"
++        : "=r"(r)
++        : "r"(x), "r"(y), "r"(sh)
++        : "x11", "x12", "x13", "x14");
++    return r;
++}
++
++int main()
++{
++    long r = test(0, 1, 2);
++    assert(r == -1);
++    return 0;
++}

debian/patches/extra/0018-hw-virtio-Fix-packed-virtqueue-flush-used_idx.patch

@@ -0,0 +1,66 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wafer <wafer@jaguarmicro.com>
+Date: Wed, 10 Apr 2024 08:44:02 +0300
+Subject: [PATCH] hw/virtio: Fix packed virtqueue flush used_idx
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+In the event of writing many chains of descriptors, the device must
+write just the id of the last buffer in the descriptor chain, skip
+forward the number of descriptors in the chain, and then repeat the
+operations for the rest of chains.
+
+Current QEMU code writes all the buffer ids consecutively, and then
+skips all the buffers altogether. This is a bug, and can be reproduced
+with a VirtIONet device with _F_MRG_RXBUB and without
+_F_INDIRECT_DESC:
+
+If a virtio-net device has the VIRTIO_NET_F_MRG_RXBUF feature
+but not the VIRTIO_RING_F_INDIRECT_DESC feature,
+'VirtIONetQueue->rx_vq' will use the merge feature
+to store data in multiple 'elems'.
+The 'num_buffers' in the virtio header indicates how many elements are merged.
+If the value of 'num_buffers' is greater than 1,
+all the merged elements will be filled into the descriptor ring.
+The 'idx' of the elements should be the value of 'vq->used_idx' plus 'ndescs'.
+
+Fixes: 86044b24e8 ("virtio: basic packed virtqueue support")
+Acked-by: Eugenio Pérez <eperezma@redhat.com>
+Signed-off-by: Wafer <wafer@jaguarmicro.com>
+Message-Id: <20240407015451.5228-2-wafer@jaguarmicro.com>
+Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
+Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
+(cherry picked from commit 2d9a31b3c27311eca1682cb2c076d7a300441960)
+Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
+---
+ hw/virtio/virtio.c | 12 ++++++++++--
+ 1 file changed, 10 insertions(+), 2 deletions(-)
+
+diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
+index b7da7f074d..e4f8ed1e63 100644
+--- a/hw/virtio/virtio.c
++++ b/hw/virtio/virtio.c
+@@ -1367,12 +1367,20 @@ static void virtqueue_packed_flush(VirtQueue *vq, unsigned int count)
+         return;
+     }
+ 
++    /*
++     * For indirect element's 'ndescs' is 1.
++     * For all other elemment's 'ndescs' is the
++     * number of descriptors chained by NEXT (as set in virtqueue_packed_pop).
++     * So When the 'elem' be filled into the descriptor ring,
++     * The 'idx' of this 'elem' shall be
++     * the value of 'vq->used_idx' plus the 'ndescs'.
++     */
++    ndescs += vq->used_elems[0].ndescs;
+     for (i = 1; i < count; i++) {
+-        virtqueue_packed_fill_desc(vq, &vq->used_elems[i], i, false);
++        virtqueue_packed_fill_desc(vq, &vq->used_elems[i], ndescs, false);
+         ndescs += vq->used_elems[i].ndescs;
+     }
+     virtqueue_packed_fill_desc(vq, &vq->used_elems[0], 0, true);
+-    ndescs += vq->used_elems[0].ndescs;
+ 
+     vq->inuse -= ndescs;
+     vq->used_idx += ndescs;

debian/patches/pve-qemu-5.1-vitastor.patch

@@ -1,778 +0,0 @@
-Index: pve-qemu-kvm-5.1.0/qapi/block-core.json
-===================================================================
---- pve-qemu-kvm-5.1.0.orig/qapi/block-core.json
-+++ pve-qemu-kvm-5.1.0/qapi/block-core.json
-@@ -3041,7 +3041,7 @@
-             'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
-             'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
-             { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' },
--            'sheepdog', 'pbs',
-+            'sheepdog', 'pbs', 'vitastor',
-             'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
- 
- ##
-@@ -3889,6 +3889,28 @@
-             '*tag': 'str' } }
- 
- ##
-+# @BlockdevOptionsVitastor:
-+#
-+# Driver specific block device options for vitastor
-+#
-+# @image:       Image name
-+# @inode:       Inode number
-+# @pool:        Pool ID
-+# @size:        Desired image size in bytes
-+# @config-path: Path to Vitastor configuration
-+# @etcd-host:   etcd connection address(es)
-+# @etcd-prefix: etcd key/value prefix
-+##
-+{ 'struct': 'BlockdevOptionsVitastor',
-+  'data': { '*inode': 'uint64',
-+            '*pool': 'uint64',
-+            '*size': 'uint64',
-+            '*image': 'str',
-+            '*config-path': 'str',
-+            '*etcd-host': 'str',
-+            '*etcd-prefix': 'str' } }
-+
-+##
- # @ReplicationMode:
- #
- # An enumeration of replication modes.
-@@ -4234,6 +4256,7 @@
-       'replication': { 'type': 'BlockdevOptionsReplication',
-                        'if': 'defined(CONFIG_REPLICATION)' },
-       'sheepdog':   'BlockdevOptionsSheepdog',
-+      'vitastor':   'BlockdevOptionsVitastor',
-       'ssh':        'BlockdevOptionsSsh',
-       'throttle':   'BlockdevOptionsThrottle',
-       'vdi':        'BlockdevOptionsGenericFormat',
-@@ -4623,6 +4646,17 @@
-             '*cluster-size' :   'size' } }
- 
- ##
-+# @BlockdevCreateOptionsVitastor:
-+#
-+# Driver specific image creation options for Vitastor.
-+#
-+# @size: Size of the virtual disk in bytes
-+##
-+{ 'struct': 'BlockdevCreateOptionsVitastor',
-+  'data': { 'location':         'BlockdevOptionsVitastor',
-+            'size':             'size' } }
-+
-+##
- # @BlockdevVmdkSubformat:
- #
- # Subformat options for VMDK images
-@@ -4884,6 +4918,7 @@
-       'qed':            'BlockdevCreateOptionsQed',
-       'rbd':            'BlockdevCreateOptionsRbd',
-       'sheepdog':       'BlockdevCreateOptionsSheepdog',
-+      'vitastor':       'BlockdevCreateOptionsVitastor',
-       'ssh':            'BlockdevCreateOptionsSsh',
-       'vdi':            'BlockdevCreateOptionsVdi',
-       'vhdx':           'BlockdevCreateOptionsVhdx',
-Index: pve-qemu-kvm-5.1.0/configure
-===================================================================
---- pve-qemu-kvm-5.1.0.orig/configure
-+++ pve-qemu-kvm-5.1.0/configure
-@@ -446,6 +446,7 @@ trace_backends="log"
- trace_file="trace"
- spice=""
- rbd=""
-+vitastor=""
- smartcard=""
- libusb=""
- usb_redir=""
-@@ -1383,6 +1384,10 @@ for opt do
-   ;;
-   --enable-rbd) rbd="yes"
-   ;;
-+  --disable-vitastor) vitastor="no"
-+  ;;
-+  --enable-vitastor) vitastor="yes"
-+  ;;
-   --disable-xfsctl) xfs="no"
-   ;;
-   --enable-xfsctl) xfs="yes"
-@@ -1901,6 +1906,7 @@ disabled with --disable-FEATURE, default
-   vhost-vdpa      vhost-vdpa kernel backend support
-   spice           spice
-   rbd             rados block device (rbd)
-+  vitastor        vitastor block device
-   libiscsi        iscsi support
-   libnfs          nfs support
-   smartcard       smartcard support (libcacard)
-@@ -4234,6 +4240,27 @@ EOF
- fi
- 
- ##########################################
-+# vitastor probe
-+if test "$vitastor" != "no" ; then
-+  cat > $TMPC <<EOF
-+#include <vitastor_c.h>
-+int main(void) {
-+  vitastor_c_create_qemu(0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
-+  return 0;
-+}
-+EOF
-+  vitastor_libs="-lvitastor_client"
-+  if compile_prog "" "$vitastor_libs" ; then
-+    vitastor=yes
-+  else
-+    if test "$vitastor" = "yes" ; then
-+      feature_not_found "vitastor block device" "Install vitastor-client-dev"
-+    fi
-+    vitastor=no
-+  fi
-+fi
-+
-+##########################################
- # libssh probe
- if test "$libssh" != "no" ; then
-   if $pkg_config --exists libssh; then
-@@ -6969,6 +6996,7 @@ echo "Trace output file $trace_file-<pid
- fi
- echo "spice support     $spice $(echo_version $spice $spice_protocol_version/$spice_server_version)"
- echo "rbd support       $rbd"
-+echo "vitastor support  $vitastor"
- echo "xfsctl support    $xfs"
- echo "smartcard support $smartcard"
- echo "libusb            $libusb"
-@@ -7644,6 +7672,10 @@ if test "$rbd" = "yes" ; then
-   echo "RBD_CFLAGS=$rbd_cflags" >> $config_host_mak
-   echo "RBD_LIBS=$rbd_libs" >> $config_host_mak
- fi
-+if test "$vitastor" = "yes" ; then
-+  echo "CONFIG_VITASTOR=y" >> $config_host_mak
-+  echo "VITASTOR_LIBS=$vitastor_libs" >> $config_host_mak
-+fi
- 
- echo "CONFIG_COROUTINE_BACKEND=$coroutine" >> $config_host_mak
- if test "$coroutine_pool" = "yes" ; then
-Index: pve-qemu-kvm-5.1.0/block/vitastor.c
-===================================================================
---- /dev/null
-+++ pve-qemu-kvm-5.1.0/block/vitastor.c
-@@ -0,0 +1,598 @@
-+// Copyright (c) Vitaliy Filippov, 2019+
-+// License: VNPL-1.1 or GNU GPL-2.0+ (see README.md for details)
-+
-+// QEMU block driver
-+
-+#ifdef VITASTOR_SOURCE_TREE
-+#define BUILD_DSO
-+#define _GNU_SOURCE
-+#endif
-+#include "qemu/osdep.h"
-+#include "qemu/main-loop.h"
-+#include "block/block_int.h"
-+#include "qapi/error.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qerror.h"
-+#include "qemu/uri.h"
-+#include "qemu/error-report.h"
-+#include "qemu/module.h"
-+#include "qemu/option.h"
-+
-+#if QEMU_VERSION_MAJOR >= 3
-+#include "qemu/units.h"
-+#include "block/qdict.h"
-+#include "qemu/cutils.h"
-+#elif QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 10
-+#include "qemu/cutils.h"
-+#include "qapi/qmp/qstring.h"
-+#include "qapi/qmp/qjson.h"
-+#else
-+#include "qapi/qmp/qint.h"
-+#define qdict_put_int(options, name, num_val) qdict_put_obj(options, name, QOBJECT(qint_from_int(num_val)))
-+#define qdict_put_str(options, name, value) qdict_put_obj(options, name, QOBJECT(qstring_from_str(value)))
-+#define qobject_unref QDECREF
-+#endif
-+
-+#include "vitastor_c.h"
-+
-+#ifdef VITASTOR_SOURCE_TREE
-+void qemu_module_dummy(void)
-+{
-+}
-+
-+void DSO_STAMP_FUN(void)
-+{
-+}
-+#endif
-+
-+typedef struct VitastorClient
-+{
-+    void *proxy;
-+    void *watch;
-+    char *config_path;
-+    char *etcd_host;
-+    char *etcd_prefix;
-+    char *image;
-+    uint64_t inode;
-+    uint64_t pool;
-+    uint64_t size;
-+    long readonly;
-+    int use_rdma;
-+    char *rdma_device;
-+    int rdma_port_num;
-+    int rdma_gid_index;
-+    int rdma_mtu;
-+    QemuMutex mutex;
-+} VitastorClient;
-+
-+typedef struct VitastorRPC
-+{
-+    BlockDriverState *bs;
-+    Coroutine *co;
-+    QEMUIOVector *iov;
-+    long ret;
-+    int complete;
-+} VitastorRPC;
-+
-+static void vitastor_co_init_task(BlockDriverState *bs, VitastorRPC *task);
-+static void vitastor_co_generic_bh_cb(void *opaque, long retval);
-+static void vitastor_co_read_cb(void *opaque, long retval, uint64_t version);
-+static void vitastor_close(BlockDriverState *bs);
-+
-+static char *qemu_vitastor_next_tok(char *src, char delim, char **p)
-+{
-+    char *end;
-+    *p = NULL;
-+    for (end = src; *end; ++end)
-+    {
-+        if (*end == delim)
-+            break;
-+        if (*end == '\\' && end[1] != '\0')
-+            end++;
-+    }
-+    if (*end == delim)
-+    {
-+        *p = end + 1;
-+        *end = '\0';
-+    }
-+    return src;
-+}
-+
-+static void qemu_vitastor_unescape(char *src)
-+{
-+    char *p;
-+    for (p = src; *src; ++src, ++p)
-+    {
-+        if (*src == '\\' && src[1] != '\0')
-+            src++;
-+        *p = *src;
-+    }
-+    *p = '\0';
-+}
-+
-+// vitastor[:key=value]*
-+// vitastor[:etcd_host=127.0.0.1]:inode=1:pool=1[:rdma_gid_index=3]
-+// vitastor:config_path=/etc/vitastor/vitastor.conf:image=testimg
-+static void vitastor_parse_filename(const char *filename, QDict *options, Error **errp)
-+{
-+    const char *start;
-+    char *p, *buf;
-+
-+    if (!strstart(filename, "vitastor:", &start))
-+    {
-+        error_setg(errp, "File name must start with 'vitastor:'");
-+        return;
-+    }
-+
-+    buf = g_strdup(start);
-+    p = buf;
-+
-+    // The following are all key/value pairs
-+    while (p)
-+    {
-+        int i;
-+        char *name, *value;
-+        name = qemu_vitastor_next_tok(p, '=', &p);
-+        if (!p)
-+        {
-+            error_setg(errp, "conf option %s has no value", name);
-+            break;
-+        }
-+        for (i = 0; i < strlen(name); i++)
-+            if (name[i] == '_')
-+                name[i] = '-';
-+        qemu_vitastor_unescape(name);
-+        value = qemu_vitastor_next_tok(p, ':', &p);
-+        qemu_vitastor_unescape(value);
-+        if (!strcmp(name, "inode") ||
-+            !strcmp(name, "pool") ||
-+            !strcmp(name, "size") ||
-+            !strcmp(name, "use-rdma") ||
-+            !strcmp(name, "rdma-port_num") ||
-+            !strcmp(name, "rdma-gid-index") ||
-+            !strcmp(name, "rdma-mtu"))
-+        {
-+            unsigned long long num_val;
-+            if (parse_uint_full(value, &num_val, 0))
-+            {
-+                error_setg(errp, "Illegal %s: %s", name, value);
-+                goto out;
-+            }
-+            qdict_put_int(options, name, num_val);
-+        }
-+        else
-+        {
-+            qdict_put_str(options, name, value);
-+        }
-+    }
-+    if (!qdict_get_try_str(options, "image"))
-+    {
-+        if (!qdict_get_try_int(options, "inode", 0))
-+        {
-+            error_setg(errp, "one of image (name) and inode (number) must be specified");
-+            goto out;
-+        }
-+        if (!(qdict_get_try_int(options, "inode", 0) >> (64-POOL_ID_BITS)) &&
-+            !qdict_get_try_int(options, "pool", 0))
-+        {
-+            error_setg(errp, "pool number must be specified or included in the inode number");
-+            goto out;
-+        }
-+        if (!qdict_get_try_int(options, "size", 0))
-+        {
-+            error_setg(errp, "size must be specified when inode number is used instead of image name");
-+            goto out;
-+        }
-+    }
-+
-+out:
-+    g_free(buf);
-+    return;
-+}
-+
-+static void coroutine_fn vitastor_co_get_metadata(VitastorRPC *task)
-+{
-+    BlockDriverState *bs = task->bs;
-+    VitastorClient *client = bs->opaque;
-+    task->co = qemu_coroutine_self();
-+
-+    qemu_mutex_lock(&client->mutex);
-+    vitastor_c_watch_inode(client->proxy, client->image, vitastor_co_generic_bh_cb, task);
-+    qemu_mutex_unlock(&client->mutex);
-+
-+    while (!task->complete)
-+    {
-+        qemu_coroutine_yield();
-+    }
-+}
-+
-+static int vitastor_file_open(BlockDriverState *bs, QDict *options, int flags, Error **errp)
-+{
-+    VitastorClient *client = bs->opaque;
-+    int64_t ret = 0;
-+    qemu_mutex_init(&client->mutex);
-+    client->config_path = g_strdup(qdict_get_try_str(options, "config-path"));
-+    // FIXME: Rename to etcd_address
-+    client->etcd_host = g_strdup(qdict_get_try_str(options, "etcd-host"));
-+    client->etcd_prefix = g_strdup(qdict_get_try_str(options, "etcd-prefix"));
-+    client->use_rdma = qdict_get_try_int(options, "use-rdma", -1);
-+    client->rdma_device = g_strdup(qdict_get_try_str(options, "rdma-device"));
-+    client->rdma_port_num = qdict_get_try_int(options, "rdma-port-num", 0);
-+    client->rdma_gid_index = qdict_get_try_int(options, "rdma-gid-index", 0);
-+    client->rdma_mtu = qdict_get_try_int(options, "rdma-mtu", 0);
-+    client->proxy = vitastor_c_create_qemu(
-+        (QEMUSetFDHandler*)aio_set_fd_handler, bdrv_get_aio_context(bs), client->config_path, client->etcd_host, client->etcd_prefix,
-+        client->use_rdma, client->rdma_device, client->rdma_port_num, client->rdma_gid_index, client->rdma_mtu, 0
-+    );
-+    client->image = g_strdup(qdict_get_try_str(options, "image"));
-+    client->readonly = (flags & BDRV_O_RDWR) ? 1 : 0;
-+    if (client->image)
-+    {
-+        // Get image metadata (size and readonly flag)
-+        VitastorRPC task;
-+        task.complete = 0;
-+        task.bs = bs;
-+        if (qemu_in_coroutine())
-+        {
-+            vitastor_co_get_metadata(&task);
-+        }
-+        else
-+        {
-+            qemu_coroutine_enter(qemu_coroutine_create((void(*)(void*))vitastor_co_get_metadata, &task));
-+        }
-+        BDRV_POLL_WHILE(bs, !task.complete);
-+        client->watch = (void*)task.ret;
-+        client->readonly = client->readonly || vitastor_c_inode_get_readonly(client->watch);
-+        client->size = vitastor_c_inode_get_size(client->watch);
-+        if (!vitastor_c_inode_get_num(client->watch))
-+        {
-+            error_setg(errp, "image does not exist");
-+            vitastor_close(bs);
-+            return -1;
-+        }
-+        if (!client->size)
-+        {
-+            client->size = qdict_get_try_int(options, "size", 0);
-+        }
-+    }
-+    else
-+    {
-+        client->watch = NULL;
-+        client->inode = qdict_get_try_int(options, "inode", 0);
-+        client->pool = qdict_get_try_int(options, "pool", 0);
-+        if (client->pool)
-+        {
-+            client->inode = (client->inode & ((1l << (64-POOL_ID_BITS)) - 1)) | (client->pool << (64-POOL_ID_BITS));
-+        }
-+        client->size = qdict_get_try_int(options, "size", 0);
-+    }
-+    if (!client->size)
-+    {
-+        error_setg(errp, "image size not specified");
-+        vitastor_close(bs);
-+        return -1;
-+    }
-+    bs->total_sectors = client->size / BDRV_SECTOR_SIZE;
-+    //client->aio_context = bdrv_get_aio_context(bs);
-+    qdict_del(options, "use-rdma");
-+    qdict_del(options, "rdma-mtu");
-+    qdict_del(options, "rdma-gid-index");
-+    qdict_del(options, "rdma-port-num");
-+    qdict_del(options, "rdma-device");
-+    qdict_del(options, "config-path");
-+    qdict_del(options, "etcd-host");
-+    qdict_del(options, "etcd-prefix");
-+    qdict_del(options, "image");
-+    qdict_del(options, "inode");
-+    qdict_del(options, "pool");
-+    qdict_del(options, "size");
-+    return ret;
-+}
-+
-+static void vitastor_close(BlockDriverState *bs)
-+{
-+    VitastorClient *client = bs->opaque;
-+    vitastor_c_destroy(client->proxy);
-+    qemu_mutex_destroy(&client->mutex);
-+    if (client->config_path)
-+        g_free(client->config_path);
-+    if (client->etcd_host)
-+        g_free(client->etcd_host);
-+    if (client->etcd_prefix)
-+        g_free(client->etcd_prefix);
-+    if (client->image)
-+        g_free(client->image);
-+}
-+
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 2
-+static int vitastor_probe_blocksizes(BlockDriverState *bs, BlockSizes *bsz)
-+{
-+    bsz->phys = 4096;
-+    bsz->log = 512;
-+    return 0;
-+}
-+#endif
-+
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 12
-+static int coroutine_fn vitastor_co_create_opts(
-+#if QEMU_VERSION_MAJOR >= 4
-+    BlockDriver *drv,
-+#endif
-+    const char *url, QemuOpts *opts, Error **errp)
-+{
-+    QDict *options;
-+    int ret;
-+
-+    options = qdict_new();
-+    vitastor_parse_filename(url, options, errp);
-+    if (*errp)
-+    {
-+        ret = -1;
-+        goto out;
-+    }
-+
-+    // inodes don't require creation in Vitastor. FIXME: They will when there will be some metadata
-+
-+    ret = 0;
-+out:
-+    qobject_unref(options);
-+    return ret;
-+}
-+#endif
-+
-+#if QEMU_VERSION_MAJOR >= 3
-+static int coroutine_fn vitastor_co_truncate(BlockDriverState *bs, int64_t offset,
-+#if QEMU_VERSION_MAJOR >= 4
-+    bool exact,
-+#endif
-+    PreallocMode prealloc,
-+#if QEMU_VERSION_MAJOR >= 5 && QEMU_VERSION_MINOR >= 1 || QEMU_VERSION_MAJOR > 5 || defined RHEL_BDRV_CO_TRUNCATE_FLAGS
-+    BdrvRequestFlags flags,
-+#endif
-+    Error **errp)
-+{
-+    VitastorClient *client = bs->opaque;
-+
-+    if (prealloc != PREALLOC_MODE_OFF)
-+    {
-+        error_setg(errp, "Unsupported preallocation mode '%s'", PreallocMode_str(prealloc));
-+        return -ENOTSUP;
-+    }
-+
-+    // TODO: Resize inode to <offset> bytes
-+    client->size = offset / BDRV_SECTOR_SIZE;
-+
-+    return 0;
-+}
-+#endif
-+
-+static int vitastor_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
-+{
-+    bdi->cluster_size = 4096;
-+    return 0;
-+}
-+
-+static int64_t vitastor_getlength(BlockDriverState *bs)
-+{
-+    VitastorClient *client = bs->opaque;
-+    return client->size;
-+}
-+
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 0
-+static void vitastor_refresh_limits(BlockDriverState *bs, Error **errp)
-+#else
-+static int vitastor_refresh_limits(BlockDriverState *bs)
-+#endif
-+{
-+    bs->bl.request_alignment = 4096;
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 3
-+    bs->bl.min_mem_alignment = 4096;
-+#endif
-+    bs->bl.opt_mem_alignment = 4096;
-+#if QEMU_VERSION_MAJOR < 2 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR == 0
-+    return 0;
-+#endif
-+}
-+
-+//static int64_t vitastor_get_allocated_file_size(BlockDriverState *bs)
-+//{
-+//    return 0;
-+//}
-+
-+static void vitastor_co_init_task(BlockDriverState *bs, VitastorRPC *task)
-+{
-+    *task = (VitastorRPC) {
-+        .co     = qemu_coroutine_self(),
-+        .bs     = bs,
-+    };
-+}
-+
-+static void vitastor_co_generic_bh_cb(void *opaque, long retval)
-+{
-+    VitastorRPC *task = opaque;
-+    task->ret = retval;
-+    task->complete = 1;
-+    if (qemu_coroutine_self() != task->co)
-+    {
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 8
-+        aio_co_wake(task->co);
-+#else
-+        qemu_coroutine_enter(task->co, NULL);
-+        qemu_aio_release(task);
-+#endif
-+    }
-+}
-+
-+static void vitastor_co_read_cb(void *opaque, long retval, uint64_t version)
-+{
-+    vitastor_co_generic_bh_cb(opaque, retval);
-+}
-+
-+static int coroutine_fn vitastor_co_preadv(BlockDriverState *bs, uint64_t offset, uint64_t bytes, QEMUIOVector *iov, int flags)
-+{
-+    VitastorClient *client = bs->opaque;
-+    VitastorRPC task;
-+    vitastor_co_init_task(bs, &task);
-+    task.iov = iov;
-+
-+    uint64_t inode = client->watch ? vitastor_c_inode_get_num(client->watch) : client->inode;
-+    qemu_mutex_lock(&client->mutex);
-+    vitastor_c_read(client->proxy, inode, offset, bytes, iov->iov, iov->niov, vitastor_co_read_cb, &task);
-+    qemu_mutex_unlock(&client->mutex);
-+
-+    while (!task.complete)
-+    {
-+        qemu_coroutine_yield();
-+    }
-+
-+    return task.ret;
-+}
-+
-+static int coroutine_fn vitastor_co_pwritev(BlockDriverState *bs, uint64_t offset, uint64_t bytes, QEMUIOVector *iov, int flags)
-+{
-+    VitastorClient *client = bs->opaque;
-+    VitastorRPC task;
-+    vitastor_co_init_task(bs, &task);
-+    task.iov = iov;
-+
-+    uint64_t inode = client->watch ? vitastor_c_inode_get_num(client->watch) : client->inode;
-+    qemu_mutex_lock(&client->mutex);
-+    vitastor_c_write(client->proxy, inode, offset, bytes, 0, iov->iov, iov->niov, vitastor_co_generic_bh_cb, &task);
-+    qemu_mutex_unlock(&client->mutex);
-+
-+    while (!task.complete)
-+    {
-+        qemu_coroutine_yield();
-+    }
-+
-+    return task.ret;
-+}
-+
-+#if !( QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 7 )
-+static int coroutine_fn vitastor_co_readv(BlockDriverState *bs, int64_t sector_num, int nb_sectors, QEMUIOVector *iov)
-+{
-+    return vitastor_co_preadv(bs, sector_num*BDRV_SECTOR_SIZE, nb_sectors*BDRV_SECTOR_SIZE, iov, 0);
-+}
-+
-+static int coroutine_fn vitastor_co_writev(BlockDriverState *bs, int64_t sector_num, int nb_sectors, QEMUIOVector *iov)
-+{
-+    return vitastor_co_pwritev(bs, sector_num*BDRV_SECTOR_SIZE, nb_sectors*BDRV_SECTOR_SIZE, iov, 0);
-+}
-+#endif
-+
-+static int coroutine_fn vitastor_co_flush(BlockDriverState *bs)
-+{
-+    VitastorClient *client = bs->opaque;
-+    VitastorRPC task;
-+    vitastor_co_init_task(bs, &task);
-+
-+    qemu_mutex_lock(&client->mutex);
-+    vitastor_c_sync(client->proxy, vitastor_co_generic_bh_cb, &task);
-+    qemu_mutex_unlock(&client->mutex);
-+
-+    while (!task.complete)
-+    {
-+        qemu_coroutine_yield();
-+    }
-+
-+    return task.ret;
-+}
-+
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 0
-+static QemuOptsList vitastor_create_opts = {
-+    .name = "vitastor-create-opts",
-+    .head = QTAILQ_HEAD_INITIALIZER(vitastor_create_opts.head),
-+    .desc = {
-+        {
-+            .name = BLOCK_OPT_SIZE,
-+            .type = QEMU_OPT_SIZE,
-+            .help = "Virtual disk size"
-+        },
-+        { /* end of list */ }
-+    }
-+};
-+#else
-+static QEMUOptionParameter vitastor_create_opts[] = {
-+    {
-+        .name = BLOCK_OPT_SIZE,
-+        .type = OPT_SIZE,
-+        .help = "Virtual disk size"
-+    },
-+    { NULL }
-+};
-+#endif
-+
-+#if QEMU_VERSION_MAJOR >= 4
-+static const char *vitastor_strong_runtime_opts[] = {
-+    "inode",
-+    "pool",
-+    "config-path",
-+    "etcd-host",
-+    "etcd-prefix",
-+
-+    NULL
-+};
-+#endif
-+
-+static BlockDriver bdrv_vitastor = {
-+    .format_name                    = "vitastor",
-+    .protocol_name                  = "vitastor",
-+
-+    .instance_size                  = sizeof(VitastorClient),
-+    .bdrv_parse_filename            = vitastor_parse_filename,
-+
-+    .bdrv_has_zero_init             = bdrv_has_zero_init_1,
-+    .bdrv_get_info                  = vitastor_get_info,
-+    .bdrv_getlength                 = vitastor_getlength,
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 2
-+    .bdrv_probe_blocksizes          = vitastor_probe_blocksizes,
-+#endif
-+    .bdrv_refresh_limits            = vitastor_refresh_limits,
-+
-+    // FIXME: Implement it along with per-inode statistics
-+    //.bdrv_get_allocated_file_size   = vitastor_get_allocated_file_size,
-+
-+    .bdrv_file_open                 = vitastor_file_open,
-+    .bdrv_close                     = vitastor_close,
-+
-+    // Option list for the create operation
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR > 0
-+    .create_opts                    = &vitastor_create_opts,
-+#else
-+    .create_options                 = vitastor_create_opts,
-+#endif
-+
-+    // For qmp_blockdev_create(), used by the qemu monitor / QAPI
-+    // Requires patching QAPI IDL, thus unimplemented
-+    //.bdrv_co_create                 = vitastor_co_create,
-+
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 12
-+    // For bdrv_create(), used by qemu-img
-+    .bdrv_co_create_opts            = vitastor_co_create_opts,
-+#endif
-+
-+#if QEMU_VERSION_MAJOR >= 3
-+    .bdrv_co_truncate               = vitastor_co_truncate,
-+#endif
-+
-+#if QEMU_VERSION_MAJOR >= 3 || QEMU_VERSION_MAJOR == 2 && QEMU_VERSION_MINOR >= 7
-+    .bdrv_co_preadv                 = vitastor_co_preadv,
-+    .bdrv_co_pwritev                = vitastor_co_pwritev,
-+#else
-+    .bdrv_co_readv                  = vitastor_co_readv,
-+    .bdrv_co_writev                 = vitastor_co_writev,
-+#endif
-+
-+    .bdrv_co_flush_to_disk          = vitastor_co_flush,
-+
-+#if QEMU_VERSION_MAJOR >= 4
-+    .strong_runtime_opts            = vitastor_strong_runtime_opts,
-+#endif
-+};
-+
-+static void vitastor_block_init(void)
-+{
-+    bdrv_register(&bdrv_vitastor);
-+}
-+
-+block_init(vitastor_block_init);
-Index: pve-qemu-kvm-5.1.0/block/Makefile.objs
-===================================================================
---- pve-qemu-kvm-5.1.0.orig/block/Makefile.objs
-+++ pve-qemu-kvm-5.1.0/block/Makefile.objs
-@@ -32,6 +32,7 @@ block-obj-$(if $(CONFIG_LIBISCSI),y,n) +
- block-obj-$(CONFIG_LIBNFS) += nfs.o
- block-obj-$(CONFIG_CURL) += curl.o
- block-obj-$(CONFIG_RBD) += rbd.o
-+block-obj-$(CONFIG_VITASTOR) += vitastor.o
- block-obj-$(CONFIG_GLUSTERFS) += gluster.o
- block-obj-$(CONFIG_LIBSSH) += ssh.o
- block-obj-y += backup-dump.o
-@@ -61,6 +62,8 @@ curl.o-cflags      := $(CURL_CFLAGS)
- curl.o-libs        := $(CURL_LIBS)
- rbd.o-cflags       := $(RBD_CFLAGS)
- rbd.o-libs         := $(RBD_LIBS)
-+vitastor.o-cflags  := $(VITASTOR_CFLAGS)
-+vitastor.o-libs    := $(VITASTOR_LIBS)
- gluster.o-cflags   := $(GLUSTERFS_CFLAGS)
- gluster.o-libs     := $(GLUSTERFS_LIBS)
- ssh.o-cflags       := $(LIBSSH_CFLAGS)

debian/patches/pve/0001-PVE-Config-block-file-change-locking-default-to-off.patch

@@ -14,10 +14,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 2 insertions(+), 2 deletions(-)
 
 diff --git a/block/file-posix.c b/block/file-posix.c
-index 9a00d4190a..bb72e1e5ca 100644
+index b9647c5ffc..9a16d86344 100644
 --- a/block/file-posix.c
 +++ b/block/file-posix.c
-@@ -508,7 +508,7 @@ static QemuOptsList raw_runtime_opts = {
+@@ -552,7 +552,7 @@ static QemuOptsList raw_runtime_opts = {
          {
              .name = "locking",
              .type = QEMU_OPT_STRING,
@@ -26,7 +26,7 @@ index 9a00d4190a..bb72e1e5ca 100644
          },
          {
              .name = "pr-manager",
-@@ -606,7 +606,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
+@@ -652,7 +652,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
          s->use_lock = false;
          break;
      case ON_OFF_AUTO_AUTO:

debian/patches/pve/0002-PVE-Config-Adjust-network-script-path-to-etc-kvm.patch

@@ -5,22 +5,21 @@ Subject: [PATCH] PVE: [Config] Adjust network script path to /etc/kvm/
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
- include/net/net.h | 5 +++--
- 1 file changed, 3 insertions(+), 2 deletions(-)
+ include/net/net.h | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
 
 diff --git a/include/net/net.h b/include/net/net.h
-index e7ef42d62b..4fd1144e58 100644
+index 5a7c0e9ebf..59dde996f9 100644
 --- a/include/net/net.h
 +++ b/include/net/net.h
-@@ -209,8 +209,9 @@ void netdev_add(QemuOpts *opts, Error **errp);
+@@ -238,8 +238,8 @@ void netdev_add(QemuOpts *opts, Error **errp);
  int net_hub_id_for_client(NetClientState *nc, int *id);
  NetClientState *net_hub_port_find(int hub_id);
  
--#define DEFAULT_NETWORK_SCRIPT "/etc/qemu-ifup"
--#define DEFAULT_NETWORK_DOWN_SCRIPT "/etc/qemu-ifdown"
-+#define DEFAULT_NETWORK_SCRIPT "/etc/kvm/kvm-ifup"
-+#define DEFAULT_NETWORK_DOWN_SCRIPT "/etc/kvm/kvm-ifdown"
-+
+-#define DEFAULT_NETWORK_SCRIPT CONFIG_SYSCONFDIR "/qemu-ifup"
+-#define DEFAULT_NETWORK_DOWN_SCRIPT CONFIG_SYSCONFDIR "/qemu-ifdown"
++#define DEFAULT_NETWORK_SCRIPT CONFIG_SYSCONFDIR "/kvm/kvm-ifup"
++#define DEFAULT_NETWORK_DOWN_SCRIPT CONFIG_SYSCONFDIR "/kvm/kvm-ifdown"
  #define DEFAULT_BRIDGE_HELPER CONFIG_QEMU_HELPERDIR "/qemu-bridge-helper"
  #define DEFAULT_BRIDGE_INTERFACE "br0"
  

debian/patches/pve/0003-PVE-Config-set-the-CPU-model-to-kvm64-32-instead-of-.patch

@@ -10,10 +10,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 2 insertions(+), 2 deletions(-)
 
 diff --git a/target/i386/cpu.h b/target/i386/cpu.h
-index e1a5c174dc..8973d3160f 100644
+index 326649ca99..24d21486bc 100644
 --- a/target/i386/cpu.h
 +++ b/target/i386/cpu.h
-@@ -1975,9 +1975,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
+@@ -2174,9 +2174,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
  #define CPU_RESOLVING_TYPE TYPE_X86_CPU
  
  #ifdef TARGET_X86_64
@@ -24,4 +24,4 @@ index e1a5c174dc..8973d3160f 100644
 +#define TARGET_DEFAULT_CPU_TYPE X86_CPU_TYPE_NAME("kvm32")
  #endif
  
- #define cpu_signal_handler cpu_x86_signal_handler
+ #define cpu_list x86_cpu_list

debian/patches/pve/0004-PVE-Config-ui-spice-default-to-pve-certificates.patch

@@ -9,10 +9,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 9 insertions(+), 6 deletions(-)
 
 diff --git a/ui/spice-core.c b/ui/spice-core.c
-index ecc2ec2c55..ca04965ead 100644
+index c3ac20ad43..37774f1c0a 100644
 --- a/ui/spice-core.c
 +++ b/ui/spice-core.c
-@@ -668,32 +668,35 @@ void qemu_spice_init(void)
+@@ -689,32 +689,35 @@ static void qemu_spice_init(void)
  
      if (tls_port) {
          x509_dir = qemu_opt_get(opts, "x509-dir");

debian/patches/pve/0005-PVE-Config-glusterfs-no-default-logfile-if-daemonize.patch

@@ -9,7 +9,7 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 11 insertions(+), 4 deletions(-)
 
 diff --git a/block/gluster.c b/block/gluster.c
-index 4f1448e2bc..715f06394c 100644
+index 7c90f7ba4b..2e03102f00 100644
 --- a/block/gluster.c
 +++ b/block/gluster.c
 @@ -42,7 +42,7 @@

debian/patches/pve/0005-PVE-Config-smm_available-false.patch

@@ -1,24 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Alexandre Derumier <aderumier@odiso.com>
-Date: Mon, 6 Apr 2020 12:16:34 +0200
-Subject: [PATCH] PVE: [Config] smm_available = false
-
-Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/i386/x86.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/hw/i386/x86.c b/hw/i386/x86.c
-index 67bee1bcb8..d954bf77b9 100644
---- a/hw/i386/x86.c
-+++ b/hw/i386/x86.c
-@@ -856,7 +856,7 @@ bool x86_machine_is_smm_enabled(X86MachineState *x86ms)
-     if (tcg_enabled() || qtest_enabled()) {
-         smm_available = true;
-     } else if (kvm_enabled()) {
--        smm_available = kvm_has_smm();
-+        smm_available = false;
-     }
- 
-     if (smm_available) {

debian/patches/pve/0006-PVE-Config-rbd-block-rbd-disable-rbd_cache_writethro.patch

@@ -18,10 +18,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 2 insertions(+)
 
 diff --git a/block/rbd.c b/block/rbd.c
-index 688074c64b..8ae39abb46 100644
+index f826410f40..64a8d7d48b 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
-@@ -651,6 +651,8 @@ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
+@@ -820,6 +820,8 @@ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
          rados_conf_set(*cluster, "rbd_cache", "false");
      }
  

debian/patches/pve/0007-PVE-Up-qmp-add-get_link_status.patch

@@ -4,17 +4,19 @@ Date: Mon, 6 Apr 2020 12:16:37 +0200
 Subject: [PATCH] PVE: [Up] qmp: add get_link_status
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: add get_link_status to command name exceptions]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  net/net.c        | 27 +++++++++++++++++++++++++++
  qapi/net.json    | 15 +++++++++++++++
- qapi/pragma.json |  1 +
- 3 files changed, 43 insertions(+)
+ qapi/pragma.json |  2 ++
+ 3 files changed, 44 insertions(+)
 
 diff --git a/net/net.c b/net/net.c
-index bbaedb3c7a..9de23ec834 100644
+index c3391168f6..f7d984f6f5 100644
 --- a/net/net.c
 +++ b/net/net.c
-@@ -1276,6 +1276,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
+@@ -1387,6 +1387,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
      }
  }
  
@@ -49,10 +51,10 @@ index bbaedb3c7a..9de23ec834 100644
  {
      NetClientState *nc;
 diff --git a/qapi/net.json b/qapi/net.json
-index ddb113e5e5..eb3b785984 100644
+index 522ac582ed..327d7c5a37 100644
 --- a/qapi/net.json
 +++ b/qapi/net.json
-@@ -35,6 +35,21 @@
+@@ -36,6 +36,21 @@
  ##
  { 'command': 'set_link', 'data': {'name': 'str', 'up': 'bool'} }
  
@@ -75,14 +77,22 @@ index ddb113e5e5..eb3b785984 100644
  # @netdev_add:
  #
 diff --git a/qapi/pragma.json b/qapi/pragma.json
-index cffae27666..5a3e3de95f 100644
+index 7f810b0e97..29233db825 100644
 --- a/qapi/pragma.json
 +++ b/qapi/pragma.json
-@@ -5,6 +5,7 @@
- { 'pragma': {
-     # Commands allowed to return a non-dictionary:
-     'returns-whitelist': [
-+	'get_link_status',
+@@ -15,6 +15,7 @@
+         'device_add',
+         'device_del',
+         'expire_password',
++        'get_link_status',
+         'migrate_cancel',
+         'netdev_add',
+         'netdev_del',
+@@ -26,6 +27,7 @@
+         'system_wakeup' ],
+     # Commands allowed to return a non-dictionary
+     'command-returns-exceptions': [
++        'get_link_status',
          'human-monitor-command',
          'qom-get',
-         'query-migrate-cache-size',
+         'query-tpm-models',

debian/patches/pve/0008-PVE-Up-glusterfs-allow-partial-reads.patch

@@ -16,7 +16,7 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 9 insertions(+), 1 deletion(-)
 
 diff --git a/block/gluster.c b/block/gluster.c
-index 715f06394c..6d02170d9b 100644
+index 2e03102f00..7886c5fe8c 100644
 --- a/block/gluster.c
 +++ b/block/gluster.c
 @@ -57,6 +57,7 @@ typedef struct GlusterAIOCB {
@@ -27,7 +27,7 @@ index 715f06394c..6d02170d9b 100644
  } GlusterAIOCB;
  
  typedef struct BDRVGlusterState {
-@@ -759,8 +760,10 @@ static void gluster_finish_aiocb(struct glfs_fd *fd, ssize_t ret,
+@@ -752,8 +753,10 @@ static void gluster_finish_aiocb(struct glfs_fd *fd, ssize_t ret,
          acb->ret = 0; /* Success */
      } else if (ret < 0) {
          acb->ret = -errno; /* Read/Write failed */
@@ -39,15 +39,15 @@ index 715f06394c..6d02170d9b 100644
      }
  
      aio_co_schedule(acb->aio_context, acb->coroutine);
-@@ -1028,6 +1031,7 @@ static coroutine_fn int qemu_gluster_co_pwrite_zeroes(BlockDriverState *bs,
+@@ -1022,6 +1025,7 @@ static coroutine_fn int qemu_gluster_co_pwrite_zeroes(BlockDriverState *bs,
      acb.ret = 0;
      acb.coroutine = qemu_coroutine_self();
      acb.aio_context = bdrv_get_aio_context(bs);
 +    acb.is_write = true;
  
-     ret = glfs_zerofill_async(s->fd, offset, size, gluster_finish_aiocb, &acb);
+     ret = glfs_zerofill_async(s->fd, offset, bytes, gluster_finish_aiocb, &acb);
      if (ret < 0) {
-@@ -1209,9 +1213,11 @@ static coroutine_fn int qemu_gluster_co_rw(BlockDriverState *bs,
+@@ -1203,9 +1207,11 @@ static coroutine_fn int qemu_gluster_co_rw(BlockDriverState *bs,
      acb.aio_context = bdrv_get_aio_context(bs);
  
      if (write) {
@@ -59,7 +59,7 @@ index 715f06394c..6d02170d9b 100644
          ret = glfs_preadv_async(s->fd, qiov->iov, qiov->niov, offset, 0,
                                  gluster_finish_aiocb, &acb);
      }
-@@ -1275,6 +1281,7 @@ static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs)
+@@ -1268,6 +1274,7 @@ static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs)
      acb.ret = 0;
      acb.coroutine = qemu_coroutine_self();
      acb.aio_context = bdrv_get_aio_context(bs);
@@ -67,11 +67,11 @@ index 715f06394c..6d02170d9b 100644
  
      ret = glfs_fsync_async(s->fd, gluster_finish_aiocb, &acb);
      if (ret < 0) {
-@@ -1321,6 +1328,7 @@ static coroutine_fn int qemu_gluster_co_pdiscard(BlockDriverState *bs,
+@@ -1316,6 +1323,7 @@ static coroutine_fn int qemu_gluster_co_pdiscard(BlockDriverState *bs,
      acb.ret = 0;
      acb.coroutine = qemu_coroutine_self();
      acb.aio_context = bdrv_get_aio_context(bs);
 +    acb.is_write = true;
  
-     ret = glfs_discard_async(s->fd, offset, size, gluster_finish_aiocb, &acb);
+     ret = glfs_discard_async(s->fd, offset, bytes, gluster_finish_aiocb, &acb);
      if (ret < 0) {

debian/patches/pve/0009-PVE-Up-qemu-img-return-success-on-info-without-snaps.patch

@@ -9,10 +9,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 2 insertions(+), 1 deletion(-)
 
 diff --git a/qemu-img.c b/qemu-img.c
-index 5308773811..45aa024acc 100644
+index 2c32d9da4e..b9636714f6 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -2955,7 +2955,8 @@ static int img_info(int argc, char **argv)
+@@ -3013,7 +3013,8 @@ static int img_info(int argc, char **argv)
      list = collect_image_info_list(image_opts, filename, fmt, chain,
                                     force_share);
      if (!list) {

debian/patches/pve/0010-PVE-Up-qemu-img-dd-add-osize-and-read-from-to-stdin-.patch

@@ -31,13 +31,14 @@ override the output file's size.
 
 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       | 191 +++++++++++++++++++++++++++++------------------
- 2 files changed, 121 insertions(+), 74 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 b89c019b76..91d18a4819 100644
+index 1b1dab5b17..d1616c045a 100644
 --- a/qemu-img-cmds.hx
 +++ b/qemu-img-cmds.hx
 @@ -58,9 +58,9 @@ SRST
@@ -53,10 +54,10 @@ index b89c019b76..91d18a4819 100644
  
  DEF("info", img_info,
 diff --git a/qemu-img.c b/qemu-img.c
-index 45aa024acc..af54d0896e 100644
+index b9636714f6..0f6a4e4e57 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -4819,10 +4819,12 @@ static int img_bitmap(int argc, char **argv)
+@@ -4840,10 +4840,12 @@ static int img_bitmap(int argc, char **argv)
  #define C_IF      04
  #define C_OF      010
  #define C_SKIP    020
@@ -69,7 +70,7 @@ index 45aa024acc..af54d0896e 100644
  };
  
  struct DdIo {
-@@ -4898,6 +4900,19 @@ static int img_dd_skip(const char *arg,
+@@ -4919,6 +4921,19 @@ static int img_dd_skip(const char *arg,
      return 0;
  }
  
@@ -89,7 +90,7 @@ index 45aa024acc..af54d0896e 100644
  static int img_dd(int argc, char **argv)
  {
      int ret = 0;
-@@ -4938,6 +4953,7 @@ static int img_dd(int argc, char **argv)
+@@ -4959,6 +4974,7 @@ 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 },
@@ -97,7 +98,7 @@ index 45aa024acc..af54d0896e 100644
          { NULL, NULL, 0 }
      };
      const struct option long_options[] = {
-@@ -5016,8 +5032,13 @@ static int img_dd(int argc, char **argv)
+@@ -5034,91 +5050,112 @@ static int img_dd(int argc, char **argv)
          arg = NULL;
      }
  
@@ -105,53 +106,30 @@ index 45aa024acc..af54d0896e 100644
 -        error_report("Must specify both input and output files");
 +    if (!(dd.flags & C_IF) && (!fmt || strcmp(fmt, "raw") != 0)) {
 +        error_report("Input format must be raw when readin from stdin");
-+        ret = -1;
-+        goto out;
-+    }
+         ret = -1;
+         goto out;
+     }
+-
+-    blk1 = img_open(image_opts, in.filename, fmt, 0, false, false,
+-                    force_share);
+-
+-    if (!blk1) {
 +    if (!(dd.flags & C_OF) && strcmp(out_fmt, "raw") != 0) {
 +        error_report("Output format must be raw when writing to stdout");
          ret = -1;
          goto out;
      }
-@@ -5029,85 +5050,101 @@ static int img_dd(int argc, char **argv)
-         goto out;
-     }
- 
--    blk1 = img_open(image_opts, in.filename, fmt, 0, false, false,
--                    force_share);
-+    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 (!blk1) {
-+            ret = -1;
-+            goto out;
-+        }
-     }
  
 -    drv = bdrv_find_format(out_fmt);
 -    if (!drv) {
 -        error_report("Unknown file format");
-+    if (dd.flags & C_OSIZE) {
-+        size = dd.osize;
-+    } else if (dd.flags & C_IF) {
-+        size = blk_getlength(blk1);
-+        if (size < 0) {
-+            error_report("Failed to get size for '%s'", in.filename);
-+            ret = -1;
-+            goto out;
-+        }
-+    } else if (dd.flags & C_COUNT) {
-+        size = dd.count * in.bsz;
-+    } else {
-+        error_report("Output size must be known when reading from stdin");
-         ret = -1;
-         goto out;
-     }
+-        ret = -1;
+-        goto out;
+-    }
 -    proto_drv = bdrv_find_protocol(out.filename, true, &local_err);
++    if (dd.flags & C_IF) {
++        blk1 = img_open(image_opts, in.filename, fmt, 0, false, false,
++                        force_share);
  
 -    if (!proto_drv) {
 -        error_report_err(local_err);
@@ -169,14 +147,50 @@ index 45aa024acc..af54d0896e 100644
 -                     proto_drv->format_name);
 -        ret = -1;
 -        goto out;
-+    if (!(dd.flags & C_OSIZE) && dd.flags & C_COUNT && dd.count <= INT64_MAX / in.bsz &&
-+        dd.count * in.bsz < size) {
-+        size = dd.count * in.bsz;
++        if (!blk1) {
++            ret = -1;
++            goto out;
++        }
      }
 -    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);
+ 
+-    size = blk_getlength(blk1);
+-    if (size < 0) {
+-        error_report("Failed to get size for '%s'", in.filename);
++    if (dd.flags & C_OSIZE) {
++        size = dd.osize;
++    } else if (dd.flags & C_IF) {
++        size = blk_getlength(blk1);
++        if (size < 0) {
++            error_report("Failed to get size for '%s'", in.filename);
++            ret = -1;
++            goto out;
++        }
++    } else if (dd.flags & C_COUNT) {
++        size = dd.count * in.bsz;
++    } else {
++        error_report("Output size must be known when reading from stdin");
+         ret = -1;
+         goto out;
+     }
+ 
+-    if (dd.flags & C_COUNT && dd.count <= INT64_MAX / in.bsz &&
++    if (!(dd.flags & C_OSIZE) && dd.flags & C_COUNT && dd.count <= INT64_MAX / in.bsz &&
+         dd.count * in.bsz < size) {
+         size = dd.count * in.bsz;
+     }
+ 
+-    /* Overflow means the specified offset is beyond input image's size */
+-    if (dd.flags & C_SKIP && (in.offset > INT64_MAX / in.bsz ||
+-                              size < in.bsz * in.offset)) {
+-        qemu_opt_set_number(opts, BLOCK_OPT_SIZE, 0, &error_abort);
+-    } else {
+-        qemu_opt_set_number(opts, BLOCK_OPT_SIZE,
+-                            size - in.bsz * in.offset, &error_abort);
+-    }
 +    if (dd.flags & C_OF) {
 +        drv = bdrv_find_format(out_fmt);
 +        if (!drv) {
@@ -186,9 +200,11 @@ index 45aa024acc..af54d0896e 100644
 +        }
 +        proto_drv = bdrv_find_protocol(out.filename, true, &local_err);
  
--    size = blk_getlength(blk1);
--    if (size < 0) {
--        error_report("Failed to get size for '%s'", in.filename);
+-    ret = bdrv_create(drv, out.filename, opts, &local_err);
+-    if (ret < 0) {
+-        error_reportf_err(local_err,
+-                          "%s: error while creating output image: ",
+-                          out.filename);
 -        ret = -1;
 -        goto out;
 -    }
@@ -212,20 +228,18 @@ index 45aa024acc..af54d0896e 100644
 +        create_opts = qemu_opts_append(create_opts, drv->create_opts);
 +        create_opts = qemu_opts_append(create_opts, proto_drv->create_opts);
  
--    if (dd.flags & C_COUNT && dd.count <= INT64_MAX / in.bsz &&
--        dd.count * in.bsz < size) {
--        size = dd.count * in.bsz;
--    }
+-    /* TODO, we can't honour --image-opts for the target,
+-     * since it needs to be given in a format compatible
+-     * with the bdrv_create() call above which does not
+-     * support image-opts style.
+-     */
+-    blk2 = img_open_file(out.filename, NULL, out_fmt, BDRV_O_RDWR,
+-                         false, false, false);
 +        opts = qemu_opts_create(create_opts, NULL, 0, &error_abort);
  
--    /* Overflow means the specified offset is beyond input image's size */
--    if (dd.flags & C_SKIP && (in.offset > INT64_MAX / in.bsz ||
--                              size < in.bsz * in.offset)) {
--        qemu_opt_set_number(opts, BLOCK_OPT_SIZE, 0, &error_abort);
--    } else {
--        qemu_opt_set_number(opts, BLOCK_OPT_SIZE,
--                            size - in.bsz * in.offset, &error_abort);
--    }
+-    if (!blk2) {
+-        ret = -1;
+-        goto out;
 +        /* Overflow means the specified offset is beyond input image's size */
 +        if (dd.flags & C_OSIZE) {
 +            qemu_opt_set_number(opts, BLOCK_OPT_SIZE, size, &error_abort);
@@ -236,15 +250,7 @@ index 45aa024acc..af54d0896e 100644
 +            qemu_opt_set_number(opts, BLOCK_OPT_SIZE,
 +                                size - in.bsz * in.offset, &error_abort);
 +        }
- 
--    ret = bdrv_create(drv, out.filename, opts, &local_err);
--    if (ret < 0) {
--        error_reportf_err(local_err,
--                          "%s: error while creating output image: ",
--                          out.filename);
--        ret = -1;
--        goto out;
--    }
++
 +        ret = bdrv_create(drv, out.filename, opts, &local_err);
 +        if (ret < 0) {
 +            error_reportf_err(local_err,
@@ -253,14 +259,7 @@ index 45aa024acc..af54d0896e 100644
 +            ret = -1;
 +            goto out;
 +        }
- 
--    /* TODO, we can't honour --image-opts for the target,
--     * since it needs to be given in a format compatible
--     * with the bdrv_create() call above which does not
--     * support image-opts style.
--     */
--    blk2 = img_open_file(out.filename, NULL, out_fmt, BDRV_O_RDWR,
--                         false, false, false);
++
 +        /* TODO, we can't honour --image-opts for the target,
 +         * since it needs to be given in a format compatible
 +         * with the bdrv_create() call above which does not
@@ -268,10 +267,7 @@ index 45aa024acc..af54d0896e 100644
 +         */
 +        blk2 = img_open_file(out.filename, NULL, out_fmt, BDRV_O_RDWR,
 +                             false, false, false);
- 
--    if (!blk2) {
--        ret = -1;
--        goto out;
++
 +        if (!blk2) {
 +            ret = -1;
 +            goto out;
@@ -279,41 +275,54 @@ index 45aa024acc..af54d0896e 100644
      }
  
      if (dd.flags & C_SKIP && (in.offset > INT64_MAX / in.bsz ||
-@@ -5125,11 +5162,17 @@ static int img_dd(int argc, char **argv)
+@@ -5135,20 +5172,43 @@ 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;
+     for (out_pos = 0; in_pos < size; ) {
++        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",
-@@ -5139,9 +5182,13 @@ 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;

debian/patches/pve/0011-PVE-Up-qemu-img-dd-add-isize-parameter.patch

@@ -10,15 +10,16 @@ an expected end of input.
 
 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 af54d0896e..0f1d464392 100644
+index 0f6a4e4e57..a3cd66e56c 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -4820,11 +4820,13 @@ static int img_bitmap(int argc, char **argv)
+@@ -4841,11 +4841,13 @@ static int img_bitmap(int argc, char **argv)
  #define C_OF      010
  #define C_SKIP    020
  #define C_OSIZE   040
@@ -32,7 +33,7 @@ index af54d0896e..0f1d464392 100644
  };
  
  struct DdIo {
-@@ -4913,6 +4915,19 @@ static int img_dd_osize(const char *arg,
+@@ -4934,6 +4936,19 @@ static int img_dd_osize(const char *arg,
      return 0;
  }
  
@@ -52,13 +53,13 @@ index af54d0896e..0f1d464392 100644
  static int img_dd(int argc, char **argv)
  {
      int ret = 0;
-@@ -4927,12 +4942,14 @@ static int img_dd(int argc, char **argv)
+@@ -4948,12 +4963,14 @@ static int img_dd(int argc, char **argv)
      int c, i;
      const char *out_fmt = "raw";
      const char *fmt = NULL;
 -    int64_t size = 0;
 +    int64_t size = 0, readsize = 0;
-     int64_t block_count = 0, out_pos, in_pos;
+     int64_t out_pos, in_pos;
      bool force_share = false;
      struct DdInfo dd = {
          .flags = 0,
@@ -68,7 +69,7 @@ index af54d0896e..0f1d464392 100644
      };
      struct DdIo in = {
          .bsz = 512, /* Block size is by default 512 bytes */
-@@ -4954,6 +4971,7 @@ static int img_dd(int argc, char **argv)
+@@ -4975,6 +4992,7 @@ 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 },
@@ -76,20 +77,22 @@ index af54d0896e..0f1d464392 100644
          { NULL, NULL, 0 }
      };
      const struct option long_options[] = {
-@@ -5160,14 +5178,18 @@ static int img_dd(int argc, char **argv)
+@@ -5171,9 +5189,10 @@ 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++) {
+-    for (out_pos = 0; in_pos < size; ) {
 +    readsize = (dd.isize > 0) ? dd.isize : size;
-+    for (out_pos = 0; in_pos < readsize; block_count++) {
++    for (out_pos = 0; in_pos < readsize; ) {
          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) {
+@@ -5182,6 +5201,9 @@ 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;

debian/patches/pve/0012-PVE-Up-qemu-img-dd-add-n-skip_create.patch

@@ -0,0 +1,121 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Alexandre Derumier <aderumier@odiso.com>
+Date: Mon, 6 Apr 2020 12:16:42 +0200
+Subject: [PATCH] PVE: [Up] qemu-img dd: add -n skip_create
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: fix getopt-string + add documentation]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ docs/tools/qemu-img.rst | 11 ++++++++++-
+ qemu-img-cmds.hx        |  4 ++--
+ qemu-img.c              | 23 ++++++++++++++---------
+ 3 files changed, 26 insertions(+), 12 deletions(-)
+
+diff --git a/docs/tools/qemu-img.rst b/docs/tools/qemu-img.rst
+index 15aeddc6d8..5e713e231d 100644
+--- a/docs/tools/qemu-img.rst
++++ b/docs/tools/qemu-img.rst
+@@ -208,6 +208,10 @@ Parameters to convert subcommand:
+ 
+ Parameters to dd subcommand:
+ 
++.. option:: -n
++
++  Skip the creation of the target volume
++
+ .. program:: qemu-img-dd
+ 
+ .. option:: bs=BLOCK_SIZE
+@@ -488,7 +492,7 @@ Command description:
+   it doesn't need to be specified separately in this case.
+ 
+ 
+-.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [bs=BLOCK_SIZE] [count=BLOCKS] [skip=BLOCKS] if=INPUT of=OUTPUT
++.. option:: dd [--image-opts] [-U] [-f FMT] [-O OUTPUT_FMT] [-n] [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.
+@@ -499,6 +503,11 @@ Command description:
+ 
+   The size syntax is similar to :manpage:`dd(1)`'s size syntax.
+ 
++  If the ``-n`` option is specified, the target volume creation will be
++  skipped. This is useful for formats such as ``rbd`` if the target
++  volume has already been created with site specific options that cannot
++  be supplied through ``qemu-img``.
++
+ .. option:: info [--object OBJECTDEF] [--image-opts] [-f FMT] [--output=OFMT] [--backing-chain] [-U] FILENAME
+ 
+   Give information about the disk image *FILENAME*. Use it in
+diff --git a/qemu-img-cmds.hx b/qemu-img-cmds.hx
+index d1616c045a..b5b0bb4467 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] [bs=block_size] [count=blocks] [skip=blocks] [osize=output_size] if=input of=output")
++    "dd [--image-opts] [-U] [-f fmt] [-O output_fmt] [-n] [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] [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] [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 a3cd66e56c..4f5ef5b887 100644
+--- a/qemu-img.c
++++ b/qemu-img.c
+@@ -4965,7 +4965,7 @@ static int img_dd(int argc, char **argv)
+     const char *fmt = NULL;
+     int64_t size = 0, readsize = 0;
+     int64_t out_pos, in_pos;
+-    bool force_share = false;
++    bool force_share = false, skip_create = false;
+     struct DdInfo dd = {
+         .flags = 0,
+         .count = 0,
+@@ -5003,7 +5003,7 @@ static int img_dd(int argc, char **argv)
+         { 0, 0, 0, 0 }
+     };
+ 
+-    while ((c = getopt_long(argc, argv, ":hf:O:U", long_options, NULL))) {
++    while ((c = getopt_long(argc, argv, ":hf:O:Un", long_options, NULL))) {
+         if (c == EOF) {
+             break;
+         }
+@@ -5023,6 +5023,9 @@ static int img_dd(int argc, char **argv)
+         case 'h':
+             help();
+             break;
++        case 'n':
++            skip_create = true;
++            break;
+         case 'U':
+             force_share = true;
+             break;
+@@ -5153,13 +5156,15 @@ static int img_dd(int argc, char **argv)
+                                 size - in.bsz * in.offset, &error_abort);
+         }
+ 
+-        ret = bdrv_create(drv, out.filename, opts, &local_err);
+-        if (ret < 0) {
+-            error_reportf_err(local_err,
+-                              "%s: error while creating output image: ",
+-                              out.filename);
+-            ret = -1;
+-            goto out;
++        if (!skip_create) {
++            ret = bdrv_create(drv, out.filename, opts, &local_err);
++            if (ret < 0) {
++                error_reportf_err(local_err,
++                                  "%s: error while creating output image: ",
++                                  out.filename);
++                ret = -1;
++                goto out;
++            }
+         }
+ 
+         /* TODO, we can't honour --image-opts for the target,

debian/patches/pve/0013-PVE-Up-qemu-img-dd-add-n-skip_create.patch

@@ -1,65 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Alexandre Derumier <aderumier@odiso.com>
-Date: Mon, 6 Apr 2020 12:16:42 +0200
-Subject: [PATCH] PVE: [Up] qemu-img dd: add -n skip_create
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- qemu-img.c | 23 ++++++++++++++---------
- 1 file changed, 14 insertions(+), 9 deletions(-)
-
-diff --git a/qemu-img.c b/qemu-img.c
-index 0f1d464392..9635e9c001 100644
---- a/qemu-img.c
-+++ b/qemu-img.c
-@@ -4944,7 +4944,7 @@ 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;
--    bool force_share = false;
-+    bool force_share = false, skip_create = false;
-     struct DdInfo dd = {
-         .flags = 0,
-         .count = 0,
-@@ -4982,7 +4982,7 @@ static int img_dd(int argc, char **argv)
-         { 0, 0, 0, 0 }
-     };
- 
--    while ((c = getopt_long(argc, argv, ":hf:O:U", long_options, NULL))) {
-+    while ((c = getopt_long(argc, argv, ":hf:O:U:n", long_options, NULL))) {
-         if (c == EOF) {
-             break;
-         }
-@@ -5002,6 +5002,9 @@ static int img_dd(int argc, char **argv)
-         case 'h':
-             help();
-             break;
-+        case 'n':
-+            skip_create = true;
-+            break;
-         case 'U':
-             force_share = true;
-             break;
-@@ -5142,13 +5145,15 @@ static int img_dd(int argc, char **argv)
-                                 size - in.bsz * in.offset, &error_abort);
-         }
- 
--        ret = bdrv_create(drv, out.filename, opts, &local_err);
--        if (ret < 0) {
--            error_reportf_err(local_err,
--                              "%s: error while creating output image: ",
--                              out.filename);
--            ret = -1;
--            goto out;
-+        if (!skip_create) {
-+            ret = bdrv_create(drv, out.filename, opts, &local_err);
-+            if (ret < 0) {
-+                error_reportf_err(local_err,
-+                                  "%s: error while creating output image: ",
-+                                  out.filename);
-+                ret = -1;
-+                goto out;
-+            }
-         }
- 
-         /* TODO, we can't honour --image-opts for the target,

debian/patches/pve/0013-PVE-virtio-balloon-improve-query-balloon.patch

@@ -7,17 +7,20 @@ Actually provide memory information via the query-balloon
 command.
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: add BalloonInfo to member name exceptions list]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  hw/virtio/virtio-balloon.c | 33 +++++++++++++++++++++++++++++++--
  monitor/hmp-cmds.c         | 30 +++++++++++++++++++++++++++++-
- qapi/misc.json             | 22 +++++++++++++++++++++-
- 3 files changed, 81 insertions(+), 4 deletions(-)
+ qapi/machine.json          | 22 +++++++++++++++++++++-
+ qapi/pragma.json           |  1 +
+ 4 files changed, 82 insertions(+), 4 deletions(-)
 
 diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
-index 22cb5df717..6513adb0a6 100644
+index e4c4c2d3c8..49874569b1 100644
 --- a/hw/virtio/virtio-balloon.c
 +++ b/hw/virtio/virtio-balloon.c
-@@ -805,8 +805,37 @@ 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;
@@ -58,10 +61,10 @@ index 22cb5df717..6513adb0a6 100644
  
  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 ae4b6a4246..6e26ea2cd0 100644
+index 01b789a79e..480b798963 100644
 --- a/monitor/hmp-cmds.c
 +++ b/monitor/hmp-cmds.c
-@@ -660,7 +660,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
+@@ -696,7 +696,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
          return;
      }
  
@@ -98,13 +101,13 @@ index ae4b6a4246..6e26ea2cd0 100644
  
      qapi_free_BalloonInfo(info);
  }
-diff --git a/qapi/misc.json b/qapi/misc.json
-index 9d32820dc1..44b1fb6fa7 100644
---- a/qapi/misc.json
-+++ b/qapi/misc.json
-@@ -226,10 +226,30 @@
- #
- # @actual: the number of bytes the balloon currently contains
+diff --git a/qapi/machine.json b/qapi/machine.json
+index b9228a5e46..10e77a9af3 100644
+--- a/qapi/machine.json
++++ b/qapi/machine.json
+@@ -1054,9 +1054,29 @@
+ # @actual: the logical size of the VM in bytes
+ #          Formula used: logical_vm_size = vm_ram_size - balloon_size
  #
 +# @last_update: time when stats got updated from guest
 +#
@@ -122,8 +125,7 @@ index 9d32820dc1..44b1fb6fa7 100644
 +#
 +# @max_mem: amount of memory (in bytes) assigned to the guest
 +#
- # Since: 0.14.0
- #
+ # Since: 0.14
  ##
 -{ 'struct': 'BalloonInfo', 'data': {'actual': 'int' } }
 +{ 'struct': 'BalloonInfo',
@@ -134,3 +136,15 @@ index 9d32820dc1..44b1fb6fa7 100644
  
  ##
  # @query-balloon:
+diff --git a/qapi/pragma.json b/qapi/pragma.json
+index 29233db825..f2097b9020 100644
+--- a/qapi/pragma.json
++++ b/qapi/pragma.json
+@@ -37,6 +37,7 @@
+     'member-name-exceptions': [     # visible in:
+         'ACPISlotType',             # query-acpi-ospm-status
+         'AcpiTableOptions',         # -acpitable
++        'BalloonInfo',              # query-balloon
+         'BlkdebugEvent',            # blockdev-add, -blockdev
+         'BlkdebugSetStateOptions',  # blockdev-add, -blockdev
+         'BlockDeviceInfo',          # query-block

debian/patches/pve/0014-PVE-qapi-modify-query-machines.patch

@@ -13,10 +13,10 @@ Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
  2 files changed, 9 insertions(+), 1 deletion(-)
 
 diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
-index 963088b798..32f630549e 100644
+index 4f4ab30f8c..76fff60a6b 100644
 --- a/hw/core/machine-qmp-cmds.c
 +++ b/hw/core/machine-qmp-cmds.c
-@@ -234,6 +234,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
+@@ -99,6 +99,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
          info->hotpluggable_cpus = mc->has_hotpluggable_cpus;
          info->numa_mem_supported = mc->numa_mem_supported;
          info->deprecated = !!mc->deprecation_reason;
@@ -30,24 +30,24 @@ index 963088b798..32f630549e 100644
              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 481b1f07ec..268044a34b 100644
+index 10e77a9af3..9156103c8f 100644
 --- a/qapi/machine.json
 +++ b/qapi/machine.json
-@@ -342,6 +342,8 @@
+@@ -138,6 +138,8 @@
  #
  # @is-default: whether the machine is default
  #
 +# @is-current: whether this machine is currently used
 +#
  # @cpu-max: maximum number of CPUs supported by the machine type
- #           (since 1.5.0)
+ #           (since 1.5)
  #
-@@ -361,7 +363,7 @@
+@@ -159,7 +161,7 @@
  ##
  { 'struct': 'MachineInfo',
    'data': { 'name': 'str', '*alias': 'str',
 -            '*is-default': 'bool', 'cpu-max': 'int',
 +            '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
              'hotpluggable-cpus': 'bool',  'numa-mem-supported': 'bool',
-             'deprecated': 'bool', '*default-cpu-type': 'str' } }
- 
+             'deprecated': 'bool', '*default-cpu-type': 'str',
+             '*default-ram-id': 'str' } }

debian/patches/pve/0015-PVE-qapi-modify-spice-query.patch

@@ -12,29 +12,29 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  2 files changed, 8 insertions(+)
 
 diff --git a/qapi/ui.json b/qapi/ui.json
-index 9d6721037f..af87b18db9 100644
+index 0abba3e930..bf8f441227 100644
 --- a/qapi/ui.json
 +++ b/qapi/ui.json
-@@ -214,11 +214,14 @@
+@@ -310,11 +310,14 @@
  #
  # @channels: a list of @SpiceChannel for each active spice channel
  #
 +# @ticket: The last ticket set with set_password
 +#
- # Since: 0.14.0
+ # Since: 0.14
  ##
  { 'struct': 'SpiceInfo',
    'data': {'enabled': 'bool', 'migrated': 'bool', '*host': 'str', '*port': 'int',
             '*tls-port': 'int', '*auth': 'str', '*compiled-version': 'str',
 +           '*ticket': 'str',
             'mouse-mode': 'SpiceQueryMouseMode', '*channels': ['SpiceChannel']},
-   'if': 'defined(CONFIG_SPICE)' }
+   'if': 'CONFIG_SPICE' }
  
 diff --git a/ui/spice-core.c b/ui/spice-core.c
-index ca04965ead..243466c13d 100644
+index 37774f1c0a..367f77f2b4 100644
 --- a/ui/spice-core.c
 +++ b/ui/spice-core.c
-@@ -539,6 +539,11 @@ SpiceInfo *qmp_query_spice(Error **errp)
+@@ -534,6 +534,11 @@ static SpiceInfo *qmp_query_spice_real(Error **errp)
      micro = SPICE_SERVER_VERSION & 0xff;
      info->compiled_version = g_strdup_printf("%d.%d.%d", major, minor, micro);
  

debian/patches/pve/0016-PVE-add-IOChannel-implementation-for-savevm-async.patch

@@ -0,0 +1,281 @@
+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 | 182 +++++++++++++++++++++++++++++++
+ migration/channel-savevm-async.h |  51 +++++++++
+ migration/meson.build            |   1 +
+ 3 files changed, 234 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..06d5484778
+--- /dev/null
++++ b/migration/channel-savevm-async.c
+@@ -0,0 +1,182 @@
++/*
++ * 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_errno(errp, -ret, "blk_preadv failed");
++        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) {
++        error_setg_errno(errp, -ret, "blk(_co)_pwritev failed");
++        return -1;
++    }
++
++    *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',

debian/patches/pve/0017-PVE-add-savevm-async-for-background-state-snapshots.patch

@@ -1,9 +1,7 @@
 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: internal snapshot async
-
-Truncate at 1024 boundary (Fabian Ebner will send a patch for stable)
+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
@@ -11,44 +9,45 @@ 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
+     improve condition for entering final stage]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- Makefile.objs                |   1 +
  hmp-commands-info.hx         |  13 +
- hmp-commands.hx              |  32 +++
- include/block/aio.h          |  10 +
- include/migration/snapshot.h |   1 +
+ hmp-commands.hx              |  33 +++
+ include/migration/snapshot.h |   2 +
  include/monitor/hmp.h        |   5 +
+ migration/meson.build        |   1 +
+ migration/savevm-async.c     | 538 +++++++++++++++++++++++++++++++++++
  monitor/hmp-cmds.c           |  57 ++++
  qapi/migration.json          |  34 +++
  qapi/misc.json               |  32 +++
  qemu-options.hx              |  12 +
- savevm-async.c               | 542 +++++++++++++++++++++++++++++++++++
  softmmu/vl.c                 |  10 +
- util/async.c                 |  30 ++
- 13 files changed, 779 insertions(+)
- create mode 100644 savevm-async.c
+ 11 files changed, 737 insertions(+)
+ create mode 100644 migration/savevm-async.c
 
-diff --git a/Makefile.objs b/Makefile.objs
-index d22b3b45d7..a1307c12a8 100644
---- a/Makefile.objs
-+++ b/Makefile.objs
-@@ -46,6 +46,7 @@ common-obj-y += bootdevice.o iothread.o
- common-obj-y += dump/
- common-obj-y += job-qmp.o
- common-obj-y += monitor/
-+common-obj-y += savevm-async.o
- common-obj-y += net/
- common-obj-y += qdev-monitor.o
- common-obj-$(CONFIG_WIN32) += os-win32.o
 diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
-index 30209e3903..ae8ff21789 100644
+index 754b1e8408..489c524e9e 100644
 --- a/hmp-commands-info.hx
 +++ b/hmp-commands-info.hx
-@@ -580,6 +580,19 @@ SRST
-     Show current migration xbzrle cache size.
+@@ -540,6 +540,19 @@ SRST
+     Show current migration parameters.
  ERST
  
 +    {
@@ -68,13 +67,13 @@ index 30209e3903..ae8ff21789 100644
          .name       = "balloon",
          .args_type  = "",
 diff --git a/hmp-commands.hx b/hmp-commands.hx
-index 60f395c276..2b58ac4a1c 100644
+index 673e39a697..039be0033d 100644
 --- a/hmp-commands.hx
 +++ b/hmp-commands.hx
-@@ -1829,3 +1829,35 @@ ERST
-         .flags      = "p",
-     },
- 
+@@ -1815,3 +1815,36 @@ SRST
+   Dump the FDT in dtb format to *filename*.
+ ERST
+ #endif
 +
 +    {
 +        .name       = "savevm-start",
@@ -105,52 +104,25 @@ index 60f395c276..2b58ac4a1c 100644
 +        .args_type  = "",
 +        .params     = "",
 +        .help       = "Resume VM after snaphot.",
-+        .cmd = hmp_savevm_end,
++        .cmd        = hmp_savevm_end,
++        .coroutine  = true,
 +    },
-diff --git a/include/block/aio.h b/include/block/aio.h
-index b2f703fa3f..c37617b404 100644
---- a/include/block/aio.h
-+++ b/include/block/aio.h
-@@ -17,6 +17,7 @@
- #ifdef CONFIG_LINUX_IO_URING
- #include <liburing.h>
- #endif
-+#include "qemu/coroutine.h"
- #include "qemu/queue.h"
- #include "qemu/event_notifier.h"
- #include "qemu/thread.h"
-@@ -654,6 +655,15 @@ static inline bool aio_node_check(AioContext *ctx, bool is_external)
-  */
- void aio_co_schedule(AioContext *ctx, struct Coroutine *co);
- 
-+/**
-+ * aio_co_reschedule_self:
-+ * @new_ctx: the new context
-+ *
-+ * Move the currently running coroutine to new_ctx. If the coroutine is already
-+ * running in new_ctx, do nothing.
-+ */
-+void coroutine_fn aio_co_reschedule_self(AioContext *new_ctx);
-+
- /**
-  * aio_co_wake:
-  * @co: the coroutine
 diff --git a/include/migration/snapshot.h b/include/migration/snapshot.h
-index c85b6ec75b..4411b7121d 100644
+index e72083b117..c846d37806 100644
 --- a/include/migration/snapshot.h
 +++ b/include/migration/snapshot.h
-@@ -17,5 +17,6 @@
+@@ -61,4 +61,6 @@ bool delete_snapshot(const char *name,
+                     bool has_devices, strList *devices,
+                     Error **errp);
  
- int save_snapshot(const char *name, Error **errp);
- int load_snapshot(const char *name, 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 c986cfd28b..243952d32f 100644
+index dfbc0c9a2f..440f86aba8 100644
 --- a/include/monitor/hmp.h
 +++ b/include/monitor/hmp.h
-@@ -25,6 +25,7 @@ void hmp_info_status(Monitor *mon, const QDict *qdict);
+@@ -27,6 +27,7 @@ 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);
@@ -158,7 +130,7 @@ index c986cfd28b..243952d32f 100644
  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);
-@@ -83,6 +84,10 @@ void hmp_netdev_add(Monitor *mon, const QDict *qdict);
+@@ -81,6 +82,10 @@ 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);
@@ -167,194 +139,28 @@ index c986cfd28b..243952d32f 100644
 +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 coroutine_fn hmp_screendump(Monitor *mon, const QDict *qdict);
  void hmp_chardev_add(Monitor *mon, const QDict *qdict);
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 6e26ea2cd0..280bb447a6 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -1904,6 +1904,63 @@ 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 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 ea53b23dca..c556257544 100644
---- a/qapi/migration.json
-+++ b/qapi/migration.json
-@@ -225,6 +225,40 @@
-            '*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 44b1fb6fa7..9895899f8b 100644
---- a/qapi/misc.json
-+++ b/qapi/misc.json
-@@ -1168,6 +1168,38 @@
- ##
- { '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' }
-+
- ##
- # @AcpiTableOptions:
- #
-diff --git a/qemu-options.hx b/qemu-options.hx
-index 708583b4ce..d32995cc50 100644
---- a/qemu-options.hx
-+++ b/qemu-options.hx
-@@ -3866,6 +3866,18 @@ 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/savevm-async.c b/savevm-async.c
+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..f918e18dce
+index 0000000000..dc30558713
 --- /dev/null
-+++ b/savevm-async.c
-@@ -0,0 +1,542 @@
++++ b/migration/savevm-async.c
+@@ -0,0 +1,538 @@
 +#include "qemu/osdep.h"
++#include "migration/channel-savevm-async.h"
 +#include "migration/migration.h"
 +#include "migration/savevm.h"
 +#include "migration/snapshot.h"
@@ -372,14 +178,12 @@ index 0000000000..f918e18dce
 +#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)
@@ -408,8 +212,15 @@ index 0000000000..f918e18dce
 +    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));
@@ -459,20 +270,25 @@ index 0000000000..f918e18dce
 +
 +    if (snap_state.file) {
 +        ret = qemu_fclose(snap_state.file);
++        snap_state.file = NULL;
 +    }
 +
 +    if (snap_state.target) {
-+        /* 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);
++        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;
@@ -498,66 +314,14 @@ index 0000000000..f918e18dce
 +    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
@@ -579,10 +343,13 @@ index 0000000000..f918e18dce
 +        save_snapshot_error("vm_stop_force_state error %d", ret);
 +    }
 +
-+    (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);
++    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_complete_precopy error %d", ret);
++        }
 +    }
 +
 +    DPRINTF("state saving complete\n");
@@ -591,7 +358,7 @@ index 0000000000..f918e18dce
 +
 +    /* clear migration state */
 +    migrate_set_state(&ms->state, MIGRATION_STATUS_SETUP,
-+                      ret ? MIGRATION_STATUS_FAILED : MIGRATION_STATUS_COMPLETED);
++        ret || aborted ? MIGRATION_STATUS_FAILED : MIGRATION_STATUS_COMPLETED);
 +    ms->to_dst_file = NULL;
 +
 +    qemu_savevm_state_cleanup();
@@ -601,6 +368,12 @@ index 0000000000..f918e18dce
 +        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);
@@ -634,12 +407,23 @@ index 0000000000..f918e18dce
 +    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;
++        /*
++         * A guest reaching this cutoff is dirtying lots of RAM. It should be
++         * large enough so that the guest can't dirty this much between the
++         * check and the guest actually being stopped, but it should be small
++         * enough to avoid long downtimes for non-hibernation snapshots.
++         */
++        maxlen = blk_getlength(snap_state.target) - 100*1024*1024;
 +
-+        if (pending_size > 400000 && snap_state.bs_pos + pending_size < maxlen) {
++        /* Note that there is no progress for pend_postcopy when iterating */
++        if (pending_size - pend_postcopy > 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);
@@ -722,6 +506,7 @@ index 0000000000..f918e18dce
 +    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 = (QemuCoSleep){ .to_wake = NULL };
 +
 +    if (snap_state.error) {
 +        error_free(snap_state.error);
@@ -748,7 +533,9 @@ index 0000000000..f918e18dce
 +        goto restart;
 +    }
 +
-+    snap_state.file = qemu_fopen_ops(&snap_state, &block_file_ops);
++    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);
@@ -790,11 +577,14 @@ index 0000000000..f918e18dce
 +
 +    if (snap_state.saved_vm_running) {
 +        vm_start();
++        snap_state.saved_vm_running = false;
 +    }
 +}
 +
-+void qmp_savevm_end(Error **errp)
++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");
@@ -803,14 +593,42 @@ index 0000000000..f918e18dce
 +
 +    if (snap_state.state == SAVE_STATE_ACTIVE) {
 +        snap_state.state = SAVE_STATE_CANCELLED;
-+        return;
++        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
@@ -829,27 +647,6 @@ index 0000000000..f918e18dce
 +                                                     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;
@@ -857,6 +654,7 @@ index 0000000000..f918e18dce
 +    Error *blocker = NULL;
 +
 +    QEMUFile *f;
++    size_t bs_pos = 0;
 +    int ret = -EINVAL;
 +
 +    be = blk_new_open(filename, NULL, NULL, 0, &local_err);
@@ -870,7 +668,7 @@ index 0000000000..f918e18dce
 +    blk_op_block_all(be, blocker);
 +
 +    /* restore the VM state */
-+    f = qemu_fopen_ops(be, &loadstate_file_ops);
++    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;
@@ -879,6 +677,9 @@ index 0000000000..f918e18dce
 +    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) {
@@ -896,32 +697,201 @@ index 0000000000..f918e18dce
 +    }
 +    return ret;
 +}
+diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
+index 480b798963..cfebfd1db5 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 88ecf86ac8..4435866379 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 7f798ce47e..9e3de34143 100644
+--- a/qemu-options.hx
++++ b/qemu-options.hx
+@@ -4423,6 +4423,18 @@ 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 4eb9d1f7fd..670b7e427c 100644
+index 7aa3eb5cf9..c94fe3d778 100644
 --- a/softmmu/vl.c
 +++ b/softmmu/vl.c
-@@ -2844,6 +2844,7 @@ void qemu_init(int argc, char **argv, char **envp)
-     int optind;
-     const char *optarg;
-     const char *loadvm = NULL;
-+    const char *loadstate = NULL;
-     MachineClass *machine_class;
-     const char *cpu_option;
-     const char *vga_model = NULL;
-@@ -3408,6 +3409,9 @@ 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;
-@@ -4464,6 +4468,12 @@ void qemu_init(int argc, char **argv, char **envp)
-             autostart = 0;
-             exit(1);
-         }
+@@ -164,6 +164,7 @@ 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;
+@@ -2615,6 +2616,12 @@ 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) {
@@ -931,44 +901,13 @@ index 4eb9d1f7fd..670b7e427c 100644
      }
      if (replay_mode != REPLAY_MODE_NONE) {
          replay_vmstate_init();
-diff --git a/util/async.c b/util/async.c
-index 1319eee3bc..b68e73f488 100644
---- a/util/async.c
-+++ b/util/async.c
-@@ -559,6 +559,36 @@ void aio_co_schedule(AioContext *ctx, Coroutine *co)
-     aio_context_unref(ctx);
- }
- 
-+typedef struct AioCoRescheduleSelf {
-+    Coroutine *co;
-+    AioContext *new_ctx;
-+} AioCoRescheduleSelf;
-+
-+static void aio_co_reschedule_self_bh(void *opaque)
-+{
-+    AioCoRescheduleSelf *data = opaque;
-+    aio_co_schedule(data->new_ctx, data->co);
-+}
-+
-+void coroutine_fn aio_co_reschedule_self(AioContext *new_ctx)
-+{
-+    AioContext *old_ctx = qemu_get_current_aio_context();
-+
-+    if (old_ctx != new_ctx) {
-+        AioCoRescheduleSelf data = {
-+            .co = qemu_coroutine_self(),
-+            .new_ctx = new_ctx,
-+        };
-+        /*
-+         * We can't directly schedule the coroutine in the target context
-+         * because this would be racy: The other thread could try to enter the
-+         * coroutine before it has yielded in this one.
-+         */
-+        aio_bh_schedule_oneshot(old_ctx, aio_co_reschedule_self_bh, &data);
-+        qemu_coroutine_yield();
-+    }
-+}
-+
- void aio_co_wake(struct Coroutine *co)
- {
-     AioContext *ctx;
+@@ -3159,6 +3166,9 @@ void qemu_init(int argc, char **argv)
+             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;

debian/patches/pve/0018-PVE-add-optional-buffer-size-to-QEMUFile.patch

@@ -0,0 +1,216 @@
+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 2d5f74ffc2..9fd97e6fe1 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];
+@@ -127,7 +128,9 @@ 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;
+ 
+@@ -136,6 +139,8 @@ 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;
+ }
+@@ -146,17 +151,27 @@ 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)
+@@ -414,7 +429,7 @@ 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()) {
+@@ -464,6 +479,8 @@ 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.
+      */
+@@ -518,7 +535,7 @@ 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);
+         }
+     }
+@@ -544,7 +561,7 @@ 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;
+         }
+@@ -591,8 +608,8 @@ 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;
+@@ -642,7 +659,7 @@ 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;
+         }
+@@ -676,7 +693,7 @@ 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;
+ 
+@@ -701,7 +718,7 @@ 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);
+@@ -853,7 +870,7 @@ 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 dc30558713..a38e7351c1 100644
+--- a/migration/savevm-async.c
++++ b/migration/savevm-async.c
+@@ -374,7 +374,7 @@ 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);
+@@ -507,7 +507,8 @@ 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;

debian/patches/pve/0018-add-optional-buffer-size-to-QEMUFile.patch

@@ -1,183 +0,0 @@
-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] 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>
----
- migration/qemu-file.c | 36 ++++++++++++++++++++++++------------
- migration/qemu-file.h |  1 +
- savevm-async.c        |  4 ++--
- 3 files changed, 27 insertions(+), 14 deletions(-)
-
-diff --git a/migration/qemu-file.c b/migration/qemu-file.c
-index be21518c57..a4d2e2c8ff 100644
---- a/migration/qemu-file.c
-+++ b/migration/qemu-file.c
-@@ -30,7 +30,7 @@
- #include "trace.h"
- #include "qapi/error.h"
- 
--#define IO_BUF_SIZE 32768
-+#define DEFAULT_IO_BUF_SIZE 32768
- #define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 64)
- 
- struct QEMUFile {
-@@ -45,7 +45,8 @@ 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];
-@@ -101,7 +102,7 @@ bool qemu_file_mode_is_not_valid(const char *mode)
-     return false;
- }
- 
--QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops)
-+QEMUFile *qemu_fopen_ops_sized(void *opaque, const QEMUFileOps *ops, size_t buffer_size)
- {
-     QEMUFile *f;
- 
-@@ -109,9 +110,17 @@ QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops)
- 
-     f->opaque = opaque;
-     f->ops = ops;
-+    f->buf_allocated_size = buffer_size;
-+    f->buf = malloc(buffer_size);
-+
-     return f;
- }
- 
-+QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops)
-+{
-+    return qemu_fopen_ops_sized(opaque, ops, DEFAULT_IO_BUF_SIZE);
-+}
-+
- 
- void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks)
- {
-@@ -346,7 +355,7 @@ 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;
-@@ -386,6 +395,9 @@ 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.
-      */
-@@ -435,7 +447,7 @@ 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);
-         }
-     }
-@@ -461,7 +473,7 @@ 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;
-         }
-@@ -508,8 +520,8 @@ 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;
-@@ -559,7 +571,7 @@ 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;
-         }
-@@ -593,7 +605,7 @@ 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;
- 
-@@ -618,7 +630,7 @@ 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);
-@@ -770,7 +782,7 @@ 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 a9b6d6ccb7..8752d27c74 100644
---- a/migration/qemu-file.h
-+++ b/migration/qemu-file.h
-@@ -120,6 +120,7 @@ typedef struct QEMUFileHooks {
- } QEMUFileHooks;
- 
- QEMUFile *qemu_fopen_ops(void *opaque, const QEMUFileOps *ops);
-+QEMUFile *qemu_fopen_ops_sized(void *opaque, const QEMUFileOps *ops, 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/savevm-async.c b/savevm-async.c
-index f918e18dce..156b7a030e 100644
---- a/savevm-async.c
-+++ b/savevm-async.c
-@@ -392,7 +392,7 @@ 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, 4 * 1024 * 1024);
- 
-     if (!snap_state.file) {
-         error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
-@@ -514,7 +514,7 @@ 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, 4 * 1024 * 1024);
-     if (!f) {
-         error_setg(errp, "Could not open VM state file");
-         goto the_end;

debian/patches/pve/0019-PVE-block-add-the-zeroinit-block-driver-filter.patch

@@ -4,27 +4,29 @@ 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/Makefile.objs |   1 +
- block/zeroinit.c    | 198 ++++++++++++++++++++++++++++++++++++++++++++
+ block/meson.build |   1 +
+ block/zeroinit.c  | 198 ++++++++++++++++++++++++++++++++++++++++++++++
  2 files changed, 199 insertions(+)
  create mode 100644 block/zeroinit.c
 
-diff --git a/block/Makefile.objs b/block/Makefile.objs
-index 19c6f371c9..d1a9227b8f 100644
---- a/block/Makefile.objs
-+++ b/block/Makefile.objs
-@@ -11,6 +11,7 @@ block-obj-$(CONFIG_QED) += qed.o qed-l2-cache.o qed-table.o qed-cluster.o
- block-obj-$(CONFIG_QED) += qed-check.o
- block-obj-y += vhdx.o vhdx-endian.o vhdx-log.o
- block-obj-y += quorum.o
-+block-obj-y += zeroinit.o
- block-obj-y += blkdebug.o blkverify.o blkreplay.o
- block-obj-$(CONFIG_PARALLELS) += parallels.o
- block-obj-y += blklogwrites.o
+diff --git a/block/meson.build b/block/meson.build
+index b7c68b83a3..020a89ae07 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..4fbb80eab0
+index 0000000000..b60e1b84dc
 --- /dev/null
 +++ b/block/zeroinit.c
 @@ -0,0 +1,198 @@
@@ -107,7 +109,9 @@ index 0000000000..4fbb80eab0
 +
 +    /* 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);
++                               bs, &child_of_bds,
++                               BDRV_CHILD_FILTERED | BDRV_CHILD_PRIMARY,
++                               false, &local_err);
 +    if (local_err) {
 +        ret = -EINVAL;
 +        error_propagate(errp, local_err);
@@ -138,22 +142,22 @@ index 0000000000..4fbb80eab0
 +}
 +
 +static int coroutine_fn zeroinit_co_preadv(BlockDriverState *bs,
-+    uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, int flags)
++    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,
-+                                                 int count, BdrvRequestFlags flags)
++                                                 int64_t bytes, BdrvRequestFlags flags)
 +{
 +    BDRVZeroinitState *s = bs->opaque;
 +    if (offset >= s->extents)
 +        return 0;
-+    return bdrv_pwrite_zeroes(bs->file, offset, count, flags);
++    return bdrv_pwrite_zeroes(bs->file, offset, bytes, flags);
 +}
 +
 +static int coroutine_fn zeroinit_co_pwritev(BlockDriverState *bs,
-+    uint64_t offset, uint64_t bytes, QEMUIOVector *qiov, int flags)
++    int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
 +{
 +    BDRVZeroinitState *s = bs->opaque;
 +    int64_t extents = offset + bytes;
@@ -174,9 +178,9 @@ index 0000000000..4fbb80eab0
 +}
 +
 +static int coroutine_fn zeroinit_co_pdiscard(BlockDriverState *bs,
-+                                             int64_t offset, int count)
++                                             int64_t offset, int64_t bytes)
 +{
-+    return bdrv_co_pdiscard(bs->file, offset, count);
++    return bdrv_co_pdiscard(bs->file, offset, bytes);
 +}
 +
 +static int zeroinit_co_truncate(BlockDriverState *bs, int64_t offset,
@@ -212,8 +216,6 @@ index 0000000000..4fbb80eab0
 +
 +    .bdrv_has_zero_init               = zeroinit_has_zero_init,
 +
-+    .bdrv_co_block_status             = bdrv_co_block_status_from_file,
-+
 +    .bdrv_co_pdiscard                 = zeroinit_co_pdiscard,
 +
 +    .bdrv_co_truncate                 = zeroinit_co_truncate,

debian/patches/pve/0020-PVE-Add-dummy-id-command-line-parameter.patch

@@ -14,12 +14,12 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  2 files changed, 11 insertions(+)
 
 diff --git a/qemu-options.hx b/qemu-options.hx
-index d32995cc50..abfde19ce0 100644
+index 9e3de34143..1ff8905127 100644
 --- a/qemu-options.hx
 +++ b/qemu-options.hx
-@@ -914,6 +914,9 @@ DEFHEADING()
+@@ -1159,6 +1159,9 @@ legacy PC, they are not recommended for modern configurations.
  
- DEFHEADING(Block device options:)
+ ERST
  
 +DEF("id", HAS_ARG, QEMU_OPTION_id,
 +    "-id n           set the VMID", QEMU_ARCH_ALL)
@@ -28,20 +28,20 @@ index d32995cc50..abfde19ce0 100644
      "-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 670b7e427c..366e30e594 100644
+index c94fe3d778..a6f7a422ec 100644
 --- a/softmmu/vl.c
 +++ b/softmmu/vl.c
-@@ -2832,6 +2832,7 @@ static void create_default_memdev(MachineState *ms, const char *path)
- void qemu_init(int argc, char **argv, char **envp)
- {
-     int i;
+@@ -2651,6 +2651,7 @@ void qemu_init(int argc, char **argv)
+     MachineClass *machine_class;
+     bool userconfig = true;
+     FILE *vmstate_dump_file = NULL;
 +    long vm_id;
-     int snapshot, linux_boot;
-     const char *initrd_filename;
-     const char *kernel_filename, *kernel_cmdline;
-@@ -3530,6 +3531,13 @@ void qemu_init(int argc, char **argv, char **envp)
-                     exit(1);
-                 }
+ 
+     qemu_add_opts(&qemu_drive_opts);
+     qemu_add_drive_opts(&qemu_legacy_drive_opts);
+@@ -3271,6 +3272,13 @@ void qemu_init(int argc, char **argv)
+                 machine_parse_property_opt(qemu_find_opts("smp-opts"),
+                                            "smp", optarg);
                  break;
 +            case QEMU_OPTION_id:
 +                vm_id = strtol(optarg, (char **)&optarg, 10);
@@ -51,5 +51,5 @@ index 670b7e427c..366e30e594 100644
 +                }
 +                break;
              case QEMU_OPTION_vnc:
-                 vnc_parse(optarg, &error_fatal);
+                 vnc_parse(optarg);
                  break;

debian/patches/pve/0021-PVE-Config-Revert-target-i386-disable-LINT0-after-re.patch

@@ -11,7 +11,7 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 9 insertions(+)
 
 diff --git a/hw/intc/apic_common.c b/hw/intc/apic_common.c
-index 81addd6390..c2026b07c5 100644
+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)

debian/patches/pve/0022-PVE-Up-Config-file-posix-make-locking-optiono-on-cre.patch

@@ -8,15 +8,15 @@ 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   | 61 +++++++++++++++++++++++++++++---------------
+ block/file-posix.c   | 59 ++++++++++++++++++++++++++++++--------------
  qapi/block-core.json |  3 ++-
- 2 files changed, 43 insertions(+), 21 deletions(-)
+ 2 files changed, 42 insertions(+), 20 deletions(-)
 
 diff --git a/block/file-posix.c b/block/file-posix.c
-index bb72e1e5ca..914bd1f367 100644
+index 9a16d86344..bd68df57ad 100644
 --- a/block/file-posix.c
 +++ b/block/file-posix.c
-@@ -2390,6 +2390,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+@@ -2487,6 +2487,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
      int fd;
      uint64_t perm, shared;
      int result = 0;
@@ -24,7 +24,7 @@ index bb72e1e5ca..914bd1f367 100644
  
      /* Validate options and set default values */
      assert(options->driver == BLOCKDEV_DRIVER_FILE);
-@@ -2431,19 +2432,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+@@ -2527,19 +2528,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
      perm = BLK_PERM_WRITE | BLK_PERM_RESIZE;
      shared = BLK_PERM_ALL & ~BLK_PERM_RESIZE;
  
@@ -59,7 +59,7 @@ index bb72e1e5ca..914bd1f367 100644
      }
  
      /* Clear the file by truncating it to 0 */
-@@ -2497,13 +2501,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+@@ -2593,13 +2597,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
      }
  
  out_unlock:
@@ -82,7 +82,7 @@ index bb72e1e5ca..914bd1f367 100644
      }
  
  out_close:
-@@ -2528,6 +2534,7 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+@@ -2624,6 +2630,7 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
      PreallocMode prealloc;
      char *buf = NULL;
      Error *local_err = NULL;
@@ -90,7 +90,7 @@ index bb72e1e5ca..914bd1f367 100644
  
      /* Skip file: protocol prefix */
      strstart(filename, "file:", &filename);
-@@ -2550,6 +2557,18 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+@@ -2646,6 +2653,18 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
          return -EINVAL;
      }
  
@@ -109,7 +109,7 @@ index bb72e1e5ca..914bd1f367 100644
      options = (BlockdevCreateOptions) {
          .driver     = BLOCKDEV_DRIVER_FILE,
          .u.file     = {
-@@ -2561,6 +2580,8 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+@@ -2657,6 +2676,8 @@ 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,
@@ -118,20 +118,11 @@ index bb72e1e5ca..914bd1f367 100644
          },
      };
      return raw_co_create(&options, errp);
-@@ -3107,7 +3128,7 @@ static int raw_check_perm(BlockDriverState *bs, uint64_t perm, uint64_t shared,
-     }
- 
-     /* Copy locks to the new fd */
--    if (s->perm_change_fd) {
-+    if (s->use_lock && s->perm_change_fd) {
-         ret = raw_apply_lock_bytes(NULL, s->perm_change_fd, perm, ~shared,
-                                    false, errp);
-         if (ret < 0) {
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 197bdc1c36..ea5fae22ae 100644
+index 7daaf545be..9e902b96bb 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -4178,7 +4178,8 @@
+@@ -4624,7 +4624,8 @@
              'size':                 'size',
              '*preallocation':       'PreallocMode',
              '*nocow':               'bool',

debian/patches/pve/0023-PVE-monitor-disable-oob-capability.patch

@@ -18,10 +18,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 1 insertion(+), 2 deletions(-)
 
 diff --git a/monitor/qmp.c b/monitor/qmp.c
-index d433ceae5b..a16cf3532d 100644
+index cc1407e4ac..c34fa2e0e3 100644
 --- a/monitor/qmp.c
 +++ b/monitor/qmp.c
-@@ -409,8 +409,7 @@ void monitor_init_qmp(Chardev *chr, bool pretty, Error **errp)
+@@ -502,8 +502,7 @@ 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 */

debian/patches/pve/0024-PVE-Compat-4.0-used-balloon-qemu-4-0-config-size-fal.patch

@@ -26,10 +26,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 2 insertions(+), 1 deletion(-)
 
 diff --git a/hw/core/machine.c b/hw/core/machine.c
-index 8d1a90c6cf..413902777e 100644
+index 19f42450f5..39ef2b6fe6 100644
 --- a/hw/core/machine.c
 +++ b/hw/core/machine.c
-@@ -66,7 +66,8 @@ GlobalProperty hw_compat_4_0[] = {
+@@ -135,7 +135,8 @@ GlobalProperty hw_compat_4_0[] = {
      { "virtio-vga",     "edid", "false" },
      { "virtio-gpu-device", "edid", "false" },
      { "virtio-device", "use-started", "false" },

debian/patches/pve/0025-PVE-Allow-version-code-in-machine-type.patch

@@ -10,18 +10,19 @@ 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          |  3 ++-
- softmmu/vl.c               | 15 ++++++++++++++-
- 4 files changed, 24 insertions(+), 2 deletions(-)
+ 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 32f630549e..71e19db4e1 100644
+index 76fff60a6b..ec9201fb9a 100644
 --- a/hw/core/machine-qmp-cmds.c
 +++ b/hw/core/machine-qmp-cmds.c
-@@ -238,6 +238,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
+@@ -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;
@@ -35,67 +36,93 @@ index 32f630549e..71e19db4e1 100644
  
          if (mc->default_cpu_type) {
 diff --git a/include/hw/boards.h b/include/hw/boards.h
-index 426ce5f625..3bce25a25f 100644
+index ca2f0d3592..acc3b62b6e 100644
 --- a/include/hw/boards.h
 +++ b/include/hw/boards.h
-@@ -170,6 +170,8 @@ struct MachineClass {
+@@ -232,6 +232,8 @@ struct MachineClass {
      const char *desc;
      const char *deprecation_reason;
  
 +    const char *pve_version;
 +
      void (*init)(MachineState *state);
-     void (*reset)(MachineState *state);
+     void (*reset)(MachineState *state, ShutdownCause reason);
      void (*wakeup)(MachineState *state);
 diff --git a/qapi/machine.json b/qapi/machine.json
-index 268044a34b..7a811a5860 100644
+index 9156103c8f..f4fb1b2c9c 100644
 --- a/qapi/machine.json
 +++ b/qapi/machine.json
-@@ -365,7 +365,8 @@
-   'data': { 'name': 'str', '*alias': 'str',
+@@ -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' } }
-+            'deprecated': 'bool', '*default-cpu-type': 'str',
-+            '*pve-version': 'str' } }
+             '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 366e30e594..16aa2186b0 100644
+index a6f7a422ec..8b0b35b6b4 100644
 --- a/softmmu/vl.c
 +++ b/softmmu/vl.c
-@@ -2322,6 +2322,8 @@ static MachineClass *machine_parse(const char *name, GSList *machines)
+@@ -1582,6 +1582,7 @@ static const QEMUOption *lookup_opt(int argc, char **argv,
+ static MachineClass *select_machine(QDict *qdict, Error **errp)
  {
-     MachineClass *mc;
-     GSList *el;
-+    size_t pvever_index = 0;
-+    gchar *name_clean;
- 
-     if (is_help_option(name)) {
-         printf("Supported machines are:\n");
-@@ -2338,12 +2340,23 @@ static MachineClass *machine_parse(const char *name, GSList *machines)
-         exit(0);
+     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;
+@@ -1599,6 +1600,11 @@ static MachineClass *select_machine(QDict *qdict, Error **errp)
+         }
      }
  
--    mc = find_machine(name, machines);
-+    // PVE version is specified with '+' as seperator, e.g. pc-i440fx+pvever
-+    pvever_index = strcspn(name, "+");
-+
-+    name_clean = g_strndup(name, pvever_index);
-+    mc = find_machine(name_clean, machines);
-+    g_free(name_clean);
-+
-     if (!mc) {
-         error_report("unsupported machine type");
-         error_printf("Use -machine help to list supported machines\n");
-         exit(1);
-     }
-+
-+    if (pvever_index < strlen(name)) {
-+        mc->pve_version = &name[pvever_index+1];
++    if (machine_class) {
++        machine_class->pve_version = g_strdup(pvever);
++        qdict_del(qdict, "pvever");
 +    }
 +
-     return mc;
- }
+     g_slist_free(machines);
+     if (local_err) {
+         error_append_hint(&local_err, "Use -machine help to list supported machines\n");
+@@ -3213,12 +3219,31 @@ void qemu_init(int argc, char **argv)
+             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:

debian/patches/pve/0026-block-backup-move-bcs-bitmap-initialization-to-job-c.patch

@@ -0,0 +1,59 @@
+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 6a9ad97a53..9b0151c5be 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:

debian/patches/pve/0027-PVE-Backup-add-vma-backup-format-code.patch

@@ -3,58 +3,64 @@ 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>
 ---
- Makefile      |   3 +-
- Makefile.objs |   1 +
- vma-reader.c  | 857 ++++++++++++++++++++++++++++++++++++++++++++++++++
- vma-writer.c  | 790 ++++++++++++++++++++++++++++++++++++++++++++++
- vma.c         | 839 ++++++++++++++++++++++++++++++++++++++++++++++++
- vma.h         | 150 +++++++++
- 6 files changed, 2639 insertions(+), 1 deletion(-)
+ 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/Makefile b/Makefile
-index 13dd708c4a..7b8c17ce2d 100644
---- a/Makefile
-+++ b/Makefile
-@@ -479,7 +479,7 @@ dummy := $(call unnest-vars,, \
+diff --git a/block/meson.build b/block/meson.build
+index 020a89ae07..4feae20e37 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -46,6 +46,8 @@ block_ss.add(files(
+   'zeroinit.c',
+ ), zstd, zlib, gnutls)
  
- include $(SRC_PATH)/tests/Makefile.include
++block_ss.add(files('../vma-writer.c'), libuuid)
++
+ softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+ softmmu_ss.add(files('block-ram-registrar.c'))
  
--all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
-+all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
+diff --git a/meson.build b/meson.build
+index 787f91855e..a496f6fe10 100644
+--- a/meson.build
++++ b/meson.build
+@@ -1525,6 +1525,8 @@ keyutils = dependency('libkeyutils', required: false,
  
- qemu-version.h: FORCE
- 	$(call quiet-command, \
-@@ -602,6 +602,7 @@ qemu-img$(EXESUF): qemu-img.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io
- qemu-nbd$(EXESUF): qemu-nbd.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
- qemu-io$(EXESUF): qemu-io.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
- qemu-storage-daemon$(EXESUF): qemu-storage-daemon.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(chardev-obj-y) $(io-obj-y) $(qom-obj-y) $(storage-daemon-obj-y) $(COMMON_LDADDS)
-+vma$(EXESUF): vma.o vma-reader.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+ has_gettid = cc.has_function('gettid')
  
- qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o $(COMMON_LDADDS)
- 
-diff --git a/Makefile.objs b/Makefile.objs
-index a1307c12a8..ade7b17a69 100644
---- a/Makefile.objs
-+++ b/Makefile.objs
-@@ -17,6 +17,7 @@ block-obj-y = block/ nbd/ scsi/
- block-obj-y += block.o blockjob.o job.o
- block-obj-y += qemu-io-cmds.o
- block-obj-$(CONFIG_REPLICATION) += replication.o
-+block-obj-y += vma-writer.o
- 
- block-obj-m = block/
++libuuid = cc.find_library('uuid', required: true)
++
+ # libselinux
+ selinux = dependency('libselinux',
+                      required: get_option('selinux'),
+@@ -3598,6 +3600,9 @@ 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..2b1d1cdab3
+index 0000000000..e65f1e8415
 --- /dev/null
 +++ b/vma-reader.c
-@@ -0,0 +1,857 @@
+@@ -0,0 +1,859 @@
 +/*
 + * VMA: Virtual Machine Archive
 + *
@@ -72,7 +78,6 @@ index 0000000000..2b1d1cdab3
 +#include <glib.h>
 +#include <uuid/uuid.h>
 +
-+#include "qemu-common.h"
 +#include "qemu/timer.h"
 +#include "qemu/ratelimit.h"
 +#include "vma.h"
@@ -249,6 +254,9 @@ index 0000000000..2b1d1cdab3
 +        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) {
@@ -580,7 +588,7 @@ index 0000000000..2b1d1cdab3
 +            }
 +        }
 +    } else {
-+        int res = blk_pwrite(target, sector_num * BDRV_SECTOR_SIZE, buf, nb_sectors * BDRV_SECTOR_SIZE, 0);
++        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);
@@ -914,10 +922,10 @@ index 0000000000..2b1d1cdab3
 +
 diff --git a/vma-writer.c b/vma-writer.c
 new file mode 100644
-index 0000000000..f5d2c5d23c
+index 0000000000..df4b20793d
 --- /dev/null
 +++ b/vma-writer.c
-@@ -0,0 +1,790 @@
+@@ -0,0 +1,791 @@
 +/*
 + * VMA: Virtual Machine Archive
 + *
@@ -941,6 +949,7 @@ index 0000000000..f5d2c5d23c
 +#include "qemu/main-loop.h"
 +#include "qemu/coroutine.h"
 +#include "qemu/cutils.h"
++#include "qemu/memalign.h"
 +
 +#define DEBUG_VMA 0
 +
@@ -1124,9 +1133,9 @@ index 0000000000..f5d2c5d23c
 +    assert(qemu_in_coroutine());
 +    AioContext *ctx = qemu_get_current_aio_context();
 +    aio_set_fd_handler(ctx, fd, false, NULL, (IOHandler *)qemu_coroutine_enter,
-+                       NULL, qemu_coroutine_self());
++                       NULL, NULL, qemu_coroutine_self());
 +    qemu_coroutine_yield();
-+    aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL);
++    aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL, NULL);
 +}
 +
 +static ssize_t coroutine_fn
@@ -1213,20 +1222,20 @@ index 0000000000..f5d2c5d23c
 +
 +        if ((stat(filename, &st) == 0) && S_ISFIFO(st.st_mode)) {
 +            oflags = O_NONBLOCK|O_WRONLY;
-+            vmaw->fd = qemu_open(filename, oflags, 0644);
++            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, 0644);
++            vmaw->fd = qemu_open(filename, oflags, errp);
 +        } else if (strstart(filename, "/dev/fdname/", &tmp_id_str)) {
-+            vmaw->fd = monitor_get_fd(cur_mon, tmp_id_str, errp);
++            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_CREAT|O_EXCL;
-+            vmaw->fd = qemu_open(filename, oflags, 0644);
++            oflags = O_NONBLOCK|O_DIRECT|O_WRONLY|O_EXCL;
++            vmaw->fd = qemu_create(filename, oflags, 0644, errp);
 +        }
 +
 +        if (vmaw->fd < 0) {
@@ -1710,10 +1719,10 @@ index 0000000000..f5d2c5d23c
 +}
 diff --git a/vma.c b/vma.c
 new file mode 100644
-index 0000000000..2eea2fc281
+index 0000000000..e8dffb43e0
 --- /dev/null
 +++ b/vma.c
-@@ -0,0 +1,839 @@
+@@ -0,0 +1,849 @@
 +/*
 + * VMA: Virtual Machine Archive
 + *
@@ -1731,11 +1740,11 @@ index 0000000000..2eea2fc281
 +#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"
 +
@@ -2025,8 +2034,6 @@ index 0000000000..2eea2fc281
 +    int vmstate_fd = -1;
 +    guint8 vmstate_stream = 0;
 +
-+    BlockBackend *blk = NULL;
-+
 +    for (i = 1; i < 255; i++) {
 +        VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
 +        if (di && (strcmp(di->devname, "vmstate") == 0)) {
@@ -2047,6 +2054,8 @@ index 0000000000..2eea2fc281
 +            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);
@@ -2159,8 +2168,6 @@ index 0000000000..2eea2fc281
 +
 +    vma_reader_destroy(vmar);
 +
-+    blk_unref(blk);
-+
 +    bdrv_close_all();
 +
 +    return ret;
@@ -2289,6 +2296,7 @@ index 0000000000..2eea2fc281
 +    int i, c;
 +    int verbose = 0;
 +    const char *archivename;
++    GList *backup_coroutines = NULL;
 +    GList *config_files = NULL;
 +
 +    for (;;) {
@@ -2377,7 +2385,9 @@ index 0000000000..2eea2fc281
 +        job->dev_id = dev_id;
 +
 +        Coroutine *co = qemu_coroutine_create(backup_run, job);
-+        qemu_coroutine_enter(co);
++        // 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;
@@ -2385,6 +2395,13 @@ index 0000000000..2eea2fc281
 +    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);
@@ -2449,6 +2466,8 @@ index 0000000000..2eea2fc281
 +        g_error("creating vma archive failed");
 +    }
 +
++    g_list_free(backup_coroutines);
++    g_list_free(config_files);
 +    vma_writer_destroy(vmaw);
 +    return 0;
 +}

debian/patches/pve/0028-PVE-Backup-add-backup-dump-block-driver.patch

@@ -7,33 +7,23 @@ Subject: [PATCH] PVE-Backup: add backup-dump block driver
 - 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/Makefile.objs       |   1 +
- block/backup-dump.c       | 168 ++++++++++++++++++++++++++++++++++++++
- block/backup.c            |  23 ++----
- include/block/block_int.h |  30 +++++++
- job.c                     |   3 +-
- 5 files changed, 206 insertions(+), 19 deletions(-)
+ 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/Makefile.objs b/block/Makefile.objs
-index d1a9227b8f..9ea0477d0b 100644
---- a/block/Makefile.objs
-+++ b/block/Makefile.objs
-@@ -33,6 +33,7 @@ block-obj-$(CONFIG_CURL) += curl.o
- block-obj-$(CONFIG_RBD) += rbd.o
- block-obj-$(CONFIG_GLUSTERFS) += gluster.o
- block-obj-$(CONFIG_LIBSSH) += ssh.o
-+block-obj-y += backup-dump.o
- block-obj-y += accounting.o dirty-bitmap.o
- block-obj-y += write-threshold.o
- block-obj-y += backup.o
 diff --git a/block/backup-dump.c b/block/backup-dump.c
 new file mode 100644
-index 0000000000..93d7f46950
+index 0000000000..04718a94e2
 --- /dev/null
 +++ b/block/backup-dump.c
-@@ -0,0 +1,168 @@
+@@ -0,0 +1,167 @@
 +/*
 + * BlockDriver to send backup data stream to a callback function
 + *
@@ -45,7 +35,6 @@ index 0000000000..93d7f46950
 + */
 +
 +#include "qemu/osdep.h"
-+#include "qemu-common.h"
 +#include "qom/object_interfaces.h"
 +#include "block/block_int.h"
 +
@@ -203,17 +192,18 @@ index 0000000000..93d7f46950
 +    return bs;
 +}
 diff --git a/block/backup.c b/block/backup.c
-index 4f13bb20a5..cd42236b79 100644
+index 9b0151c5be..6e8f6e67b3 100644
 --- a/block/backup.c
 +++ b/block/backup.c
-@@ -32,24 +32,6 @@
+@@ -29,28 +29,6 @@
  
- #define BACKUP_CLUSTER_SIZE_DEFAULT (1 << 16)
+ #include "block/copy-before-write.h"
  
 -typedef struct BackupBlockJob {
 -    BlockJob common;
--    BlockDriverState *backup_top;
+-    BlockDriverState *cbw;
 -    BlockDriverState *source_bs;
+-    BlockDriverState *target_bs;
 -
 -    BdrvDirtyBitmap *sync_bitmap;
 -
@@ -222,38 +212,64 @@ index 4f13bb20a5..cd42236b79 100644
 -    BlockdevOnError on_source_error;
 -    BlockdevOnError on_target_error;
 -    uint64_t len;
--    uint64_t bytes_read;
 -    int64_t cluster_size;
+-    BackupPerf perf;
 -
 -    BlockCopyState *bcs;
+-
+-    bool wait;
+-    BlockCopyCallState *bg_bcs_call;
 -} BackupBlockJob;
 -
  static const BlockJobDriver backup_job_driver;
  
- static void backup_progress_bytes_callback(int64_t bytes, void *opaque)
-@@ -422,6 +404,11 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
-         goto error;
+ 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 source is in backing chain of target assume that target is going to be
-      * used for "image fleecing", i.e. it should represent a kind of snapshot of
-diff --git a/include/block/block_int.h b/include/block/block_int.h
-index 38dec0275b..1efb1f527c 100644
---- a/include/block/block_int.h
-+++ b/include/block/block_int.h
-@@ -62,6 +62,36 @@
+ 
+     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 4feae20e37..0d7023fc82 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 31ae91e56e..37b64bcd93 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_backuo_dump_create(
++BlockDriverState *bdrv_backup_dump_create(
 +    int dump_cb_block_size,
 +    uint64_t byte_size,
 +    BackupDumpFunc *dump_cb,
@@ -265,8 +281,9 @@ index 38dec0275b..1efb1f527c 100644
 +typedef struct BlockCopyState BlockCopyState;
 +typedef struct BackupBlockJob {
 +    BlockJob common;
-+    BlockDriverState *backup_top;
++    BlockDriverState *cbw;
 +    BlockDriverState *source_bs;
++    BlockDriverState *target_bs;
 +
 +    BdrvDirtyBitmap *sync_bitmap;
 +
@@ -275,26 +292,29 @@ index 38dec0275b..1efb1f527c 100644
 +    BlockdevOnError on_source_error;
 +    BlockdevOnError on_target_error;
 +    uint64_t len;
-+    uint64_t bytes_read;
 +    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 53be57a3a0..b8139c80a4 100644
+index 72d57f0934..93e22d180b 100644
 --- a/job.c
 +++ b/job.c
-@@ -269,7 +269,8 @@ static bool job_started(Job *job)
-     return job->co;
+@@ -330,7 +330,8 @@ static bool job_started_locked(Job *job)
  }
  
--static bool job_should_pause(Job *job)
-+bool job_should_pause(Job *job);
-+bool job_should_pause(Job *job)
+ /* Called with job_mutex held. */
+-static bool job_should_pause_locked(Job *job)
++bool job_should_pause_locked(Job *job);
++bool job_should_pause_locked(Job *job)
  {
      return job->pause_count > 0;
  }

debian/patches/pve/0030-PVE-Backup-avoid-coroutines-to-fix-AIO-freeze-cleanu.patch

@@ -1,914 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 6 Apr 2020 12:17:00 +0200
-Subject: [PATCH] PVE-Backup: avoid coroutines to fix AIO freeze, cleanups
-
-We observed various AIO pool loop freezes, so we decided to avoid
-coroutines and restrict ourselfes using similar code as upstream
-(see blockdev.c: do_backup_common).
-
-* avoid coroutine for job related code (causes hangs with iothreads)
-    - We then acquire/release all mutexes outside coroutines now, so we can now
-      correctly use a normal mutex.
-
-* split pvebackup_co_dump_cb into:
-    - pvebackup_co_dump_pbs_cb and
-    - pvebackup_co_dump_pbs_cb
-
-* new helper functions
-    - pvebackup_propagate_error
-    - pvebackup_error_or_canceled
-    - pvebackup_add_transfered_bytes
-
-* avoid cancel flag (not needed)
-
-* simplify backup_cancel logic
-
-There is progress on upstream to support running qmp commands inside
-coroutines, see:
-https://lists.gnu.org/archive/html/qemu-devel/2020-02/msg04852.html
-
-We should consider using that when it is available in upstream qemu.
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- pve-backup.c | 638 ++++++++++++++++++++++++++-------------------------
- 1 file changed, 320 insertions(+), 318 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 55441eb9d1..d40f3f2fd6 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -11,11 +11,27 @@
- 
- /* 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 command, protected using rwlock
--        CoRwlock rwlock;
-+        // Everithing accessed from qmp_backup_query command is protected using lock
-+        QemuMutex lock;
-         Error *error;
-         time_t start_time;
-         time_t end_time;
-@@ -25,19 +41,20 @@ static struct PVEBackupState {
-         size_t total;
-         size_t transferred;
-         size_t zero_bytes;
--        bool cancel;
-     } stat;
-     int64_t speed;
-     VmaWriter *vmaw;
-     ProxmoxBackupHandle *pbs;
-     GList *di_list;
--    CoMutex backup_mutex;
-+    QemuMutex backup_mutex;
-+    CoMutex dump_callback_mutex;
- } backup_state;
- 
- static void pvebackup_init(void)
- {
--    qemu_co_rwlock_init(&backup_state.stat.rwlock);
--    qemu_co_mutex_init(&backup_state.backup_mutex);
-+    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
-@@ -52,10 +69,54 @@ typedef struct PVEBackupDevInfo {
-     BlockDriverState *target;
- } PVEBackupDevInfo;
- 
--static void pvebackup_co_run_next_job(void);
-+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_cb(
-+pvebackup_co_dump_pbs_cb(
-     void *opaque,
-     uint64_t start,
-     uint64_t bytes,
-@@ -67,137 +128,127 @@ pvebackup_co_dump_cb(
-     const unsigned char *buf = pbuf;
-     PVEBackupDevInfo *di = opaque;
- 
--    qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--    bool cancel = backup_state.stat.cancel;
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+    assert(backup_state.pbs);
-+
-+    Error *local_err = NULL;
-+    int pbs_res = -1;
-+
-+    qemu_co_mutex_lock(&backup_state.dump_callback_mutex);
- 
--    if (cancel) {
--        return size; // return success
-+    // avoid deadlock if job is cancelled
-+    if (pvebackup_error_or_canceled()) {
-+        qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
-+        return -1;
-     }
- 
--    qemu_co_mutex_lock(&backup_state.backup_mutex);
-+    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);
- 
--    int ret = -1;
-+    if (pbs_res < 0) {
-+        pvebackup_propagate_error(local_err);
-+        return pbs_res;
-+    } else {
-+        pvebackup_add_transfered_bytes(size, !buf ? size : 0);
-+    }
- 
--    if (backup_state.vmaw) {
--        size_t zero_bytes = 0;
--        uint64_t remaining = size;
--
--        uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
--        if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
--            qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--            if (!backup_state.stat.error) {
--                qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
--                error_setg(&backup_state.stat.error,
--                           "got unaligned write inside backup dump "
--                           "callback (sector %ld)", start);
--            }
--            qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--            qemu_co_mutex_unlock(&backup_state.backup_mutex);
--            return -1; // not aligned to cluster size
--        }
-+    return size;
-+}
- 
--        while (remaining > 0) {
--            ret = vma_writer_write(backup_state.vmaw, di->dev_id, cluster_num,
--                                   buf, &zero_bytes);
--            ++cluster_num;
--            if (buf) {
--                buf += VMA_CLUSTER_SIZE;
--            }
--            if (ret < 0) {
--                qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--                if (!backup_state.stat.error) {
--                    qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
--                    vma_writer_error_propagate(backup_state.vmaw, &backup_state.stat.error);
--                }
--                qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+// 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());
- 
--                qemu_co_mutex_unlock(&backup_state.backup_mutex);
--                return ret;
--            } else {
--                qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--                backup_state.stat.zero_bytes += zero_bytes;
--                if (remaining >= VMA_CLUSTER_SIZE) {
--                    backup_state.stat.transferred += VMA_CLUSTER_SIZE;
--                    remaining -= VMA_CLUSTER_SIZE;
--                } else {
--                    backup_state.stat.transferred += remaining;
--                    remaining = 0;
--                }
--                qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--            }
--        }
--    } else if (backup_state.pbs) {
--        Error *local_err = NULL;
--        int pbs_res = -1;
-+    const uint64_t size = bytes;
-+    const unsigned char *buf = pbuf;
-+    PVEBackupDevInfo *di = opaque;
- 
--        pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id, buf, start, size, &local_err);
-+    int ret = -1;
- 
--        qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+    assert(backup_state.vmaw);
- 
--        if (pbs_res < 0) {
--            error_propagate(&backup_state.stat.error, local_err);
--            qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--            qemu_co_mutex_unlock(&backup_state.backup_mutex);
--            return pbs_res;
--        } else {
--            if (!buf) {
--                backup_state.stat.zero_bytes += size;
--            }
--            backup_state.stat.transferred += size;
-+    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;
-         }
- 
--        qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+        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);
- 
--    } else {
--        qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--        if (!buf) {
--            backup_state.stat.zero_bytes += size;
-+        ++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;
-+            }
-         }
--        backup_state.stat.transferred += size;
--        qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-     }
- 
--    qemu_co_mutex_unlock(&backup_state.backup_mutex);
--
-     return size;
- }
- 
--static void coroutine_fn pvebackup_co_cleanup(void)
-+// assumes the caller holds backup_mutex
-+static void coroutine_fn pvebackup_co_cleanup(void *unused)
- {
-     assert(qemu_in_coroutine());
- 
--    qemu_co_mutex_lock(&backup_state.backup_mutex);
--
--    qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+    qemu_mutex_lock(&backup_state.stat.lock);
-     backup_state.stat.end_time = time(NULL);
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+    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) {
--            qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--            error_propagate(&backup_state.stat.error, local_err);
--            qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--        }
-+            pvebackup_propagate_error(local_err);
-+         }
- 
-         backup_state.vmaw = NULL;
-     }
- 
-     if (backup_state.pbs) {
--        qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--        bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
--        if (!error_or_canceled) {
-+        if (!pvebackup_error_or_canceled()) {
-             Error *local_err = NULL;
-             proxmox_backup_co_finish(backup_state.pbs, &local_err);
-             if (local_err != NULL) {
--                qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
--                error_propagate(&backup_state.stat.error, local_err);
--             }
-+                pvebackup_propagate_error(local_err);
-+            }
-         }
--        qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- 
-         proxmox_backup_disconnect(backup_state.pbs);
-         backup_state.pbs = NULL;
-@@ -205,43 +256,14 @@ static void coroutine_fn pvebackup_co_cleanup(void)
- 
-     g_list_free(backup_state.di_list);
-     backup_state.di_list = NULL;
--    qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
- 
--typedef struct PVEBackupCompeteCallbackData {
--    PVEBackupDevInfo *di;
--    int result;
--} PVEBackupCompeteCallbackData;
--
--static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
-+// assumes the caller holds backup_mutex
-+static void coroutine_fn pvebackup_complete_stream(void *opaque)
- {
--    assert(qemu_in_coroutine());
--
--    PVEBackupCompeteCallbackData *cb_data = opaque;
--
--    qemu_co_mutex_lock(&backup_state.backup_mutex);
--
--    PVEBackupDevInfo *di = cb_data->di;
--    int ret = cb_data->result;
--
--    di->completed = true;
--
--    qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--    bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
--
--    if (ret < 0 && !backup_state.stat.error) {
--        qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
--        error_setg(&backup_state.stat.error, "job failed with err %d - %s",
--                   ret, strerror(-ret));
--    }
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--
--    di->bs = NULL;
-+    PVEBackupDevInfo *di = opaque;
- 
--    if (di->target) {
--        bdrv_unref(di->target);
--        di->target = NULL;
--    }
-+    bool error_or_canceled = pvebackup_error_or_canceled();
- 
-     if (backup_state.vmaw) {
-         vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-@@ -251,110 +273,101 @@ static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
-         Error *local_err = NULL;
-         proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
-         if (local_err != NULL) {
--            qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--            error_propagate(&backup_state.stat.error, local_err);
--            qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+            pvebackup_propagate_error(local_err);
-         }
-     }
-+}
- 
--    // remove self from job queue
--    backup_state.di_list = g_list_remove(backup_state.di_list, di);
--    g_free(di);
-+static void pvebackup_complete_cb(void *opaque, int ret)
-+{
-+    assert(!qemu_in_coroutine());
- 
--    int pending_jobs = g_list_length(backup_state.di_list);
-+    PVEBackupDevInfo *di = opaque;
- 
--    qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+    qemu_mutex_lock(&backup_state.backup_mutex);
- 
--    if (pending_jobs > 0) {
--        pvebackup_co_run_next_job();
--    } else {
--        pvebackup_co_cleanup();
-+    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);
-     }
--}
- 
--static void pvebackup_complete_cb(void *opaque, int ret)
--{
--    // This can be called from the main loop, or from a coroutine
--    PVEBackupCompeteCallbackData cb_data = {
--        .di = opaque,
--        .result = ret,
--    };
-+    di->bs = NULL;
- 
--    if (qemu_in_coroutine()) {
--        pvebackup_co_complete_cb(&cb_data);
--    } else {
--        block_on_coroutine_fn(pvebackup_co_complete_cb, &cb_data);
--    }
--}
-+    assert(di->target == NULL);
- 
--static void coroutine_fn pvebackup_co_cancel(void *opaque)
--{
--    assert(qemu_in_coroutine());
-+    block_on_coroutine_fn(pvebackup_complete_stream, di);
- 
--    qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--    backup_state.stat.cancel = true;
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+    // remove self from job queue
-+    backup_state.di_list = g_list_remove(backup_state.di_list, di);
- 
--    qemu_co_mutex_lock(&backup_state.backup_mutex);
-+    g_free(di);
- 
--    // Avoid race between block jobs and backup-cancel command:
--    if (!(backup_state.vmaw || backup_state.pbs)) {
--        qemu_co_mutex_unlock(&backup_state.backup_mutex);
--        return;
--    }
-+    qemu_mutex_unlock(&backup_state.backup_mutex);
- 
--    qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--    if (!backup_state.stat.error) {
--        qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
--        error_setg(&backup_state.stat.error, "backup cancelled");
--    }
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+    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 cancelled");
-+        vma_writer_set_error(backup_state.vmaw, "backup canceled");
-     }
- 
-     if (backup_state.pbs) {
--        proxmox_backup_abort(backup_state.pbs, "backup cancelled");
-+        proxmox_backup_abort(backup_state.pbs, "backup canceled");
-     }
- 
--    bool running_jobs = 0;
--    GList *l = backup_state.di_list;
--    while (l) {
--        PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
--        l = g_list_next(l);
--        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;
--                }
-+    qemu_mutex_unlock(&backup_state.backup_mutex);
- 
--                BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
--                if (bjob && bjob->source_bs == di->bs) {
--                    AioContext *aio_context = job->job.aio_context;
--                    aio_context_acquire(aio_context);
-+    for(;;) {
- 
--                    if (!di->completed) {
--                        running_jobs += 1;
--                        job_cancel(&job->job, false);
--                    }
--                    aio_context_release(aio_context);
--                }
-+        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_co_mutex_unlock(&backup_state.backup_mutex);
-+        qemu_mutex_unlock(&backup_state.backup_mutex);
- 
--    if (running_jobs == 0) pvebackup_co_cleanup(); // else job will call completion handler
-+        if (next_job) {
-+            AioContext *aio_context = next_job->job.aio_context;
-+            aio_context_acquire(aio_context);
-+            job_cancel_sync(&next_job->job);
-+            aio_context_release(aio_context);
-+        } else {
-+            break;
-+        }
-+    }
- }
- 
- void qmp_backup_cancel(Error **errp)
- {
--    block_on_coroutine_fn(pvebackup_co_cancel, NULL);
-+    pvebackup_cancel();
- }
- 
-+// assumes the caller holds backup_mutex
- static int coroutine_fn pvebackup_co_add_config(
-     const char *file,
-     const char *name,
-@@ -406,46 +419,97 @@ static int coroutine_fn pvebackup_co_add_config(
- 
- bool job_should_pause(Job *job);
- 
--static void coroutine_fn pvebackup_co_run_next_job(void)
-+static void pvebackup_run_next_job(void)
- {
--    assert(qemu_in_coroutine());
-+    assert(!qemu_in_coroutine());
- 
--    qemu_co_mutex_lock(&backup_state.backup_mutex);
-+    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);
--        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) {
--                    AioContext *aio_context = job->job.aio_context;
--                    qemu_co_mutex_unlock(&backup_state.backup_mutex);
--                    aio_context_acquire(aio_context);
--
--                    if (job_should_pause(&job->job)) {
--                        qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--                        bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
--                        qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--
--                        if (error_or_canceled) {
--                            job_cancel(&job->job, false);
--                        } else {
--                            job_resume(&job->job);
--                        }
--                    }
--                    aio_context_release(aio_context);
--                    return;
-+        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);
-+                } 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 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, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-+            JOB_DEFAULT, pvebackup_complete_cb, di, 1, 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;
-+            }
-         }
-     }
--    qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+
-+    return errors;
- }
- 
- typedef struct QmpBackupTask {
-@@ -476,7 +540,8 @@ typedef struct QmpBackupTask {
-     UuidInfo *result;
- } QmpBackupTask;
- 
--static void coroutine_fn pvebackup_co_start(void *opaque)
-+// assumes the caller holds backup_mutex
-+static void coroutine_fn pvebackup_co_prepare(void *opaque)
- {
-     assert(qemu_in_coroutine());
- 
-@@ -495,16 +560,12 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
-     GList *di_list = NULL;
-     GList *l;
-     UuidInfo *uuid_info;
--    BlockJob *job;
- 
-     const char *config_name = "qemu-server.conf";
-     const char *firewall_name = "qemu-server.fw";
- 
--    qemu_co_mutex_lock(&backup_state.backup_mutex);
--
-     if (backup_state.di_list) {
--        qemu_co_mutex_unlock(&backup_state.backup_mutex);
--        error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+         error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-                   "previous backup not finished");
-         return;
-     }
-@@ -627,7 +688,7 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
-             if (dev_id < 0)
-                 goto err;
- 
--            if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_cb, di, task->errp))) {
-+            if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
-                 goto err;
-             }
- 
-@@ -648,7 +709,7 @@ static void coroutine_fn pvebackup_co_start(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_cb, di, task->errp))) {
-+            if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_vma_cb, di, task->errp))) {
-                 goto err;
-             }
- 
-@@ -713,9 +774,7 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
-     }
-     /* initialize global backup_state now */
- 
--    qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--
--    backup_state.stat.cancel = false;
-+    qemu_mutex_lock(&backup_state.stat.lock);
- 
-     if (backup_state.stat.error) {
-         error_free(backup_state.stat.error);
-@@ -738,7 +797,7 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
-     backup_state.stat.transferred = 0;
-     backup_state.stat.zero_bytes = 0;
- 
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+    qemu_mutex_unlock(&backup_state.stat.lock);
- 
-     backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
- 
-@@ -747,48 +806,6 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- 
-     backup_state.di_list = di_list;
- 
--    /* start all jobs (paused state) */
--    l = di_list;
--    while (l) {
--        PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
--        l = g_list_next(l);
--
--        // make sure target runs in same aoi_context as source
--        AioContext *aio_context = bdrv_get_aio_context(di->bs);
--        aio_context_acquire(aio_context);
--        GSList *ignore = NULL;
--        bdrv_set_aio_context_ignore(di->target, aio_context, &ignore);
--        g_slist_free(ignore);
--        aio_context_release(aio_context);
--
--        job = backup_job_create(NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
--                                BITMAP_SYNC_MODE_NEVER, false, NULL, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
--                                JOB_DEFAULT, pvebackup_complete_cb, di, 1, NULL, &local_err);
--        if (!job || local_err != NULL) {
--            qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
--            error_setg(&backup_state.stat.error, "backup_job_create failed");
--            qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--            break;
--        }
--        job_start(&job->job);
--        if (di->target) {
--            bdrv_unref(di->target);
--            di->target = NULL;
--        }
--    }
--
--    qemu_co_mutex_unlock(&backup_state.backup_mutex);
--
--    qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
--    bool no_errors = !backup_state.stat.error;
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--
--    if (no_errors) {
--        pvebackup_co_run_next_job(); // run one job
--    } else {
--        pvebackup_co_cancel(NULL);
--    }
--
-     uuid_info = g_malloc0(sizeof(*uuid_info));
-     uuid_info->UUID = uuid_str;
- 
-@@ -831,8 +848,6 @@ err:
-         rmdir(backup_dir);
-     }
- 
--    qemu_co_mutex_unlock(&backup_state.backup_mutex);
--
-     task->result = NULL;
-     return;
- }
-@@ -876,32 +891,31 @@ UuidInfo *qmp_backup(
-         .errp = errp,
-     };
- 
--    block_on_coroutine_fn(pvebackup_co_start, &task);
-+    qemu_mutex_lock(&backup_state.backup_mutex);
- 
--    return task.result;
--}
-+    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);
-+    }
- 
--typedef struct QmpQueryBackupTask {
--    Error **errp;
--    BackupStatus *result;
--} QmpQueryBackupTask;
-+    return task.result;
-+}
- 
--static void coroutine_fn pvebackup_co_query(void *opaque)
-+BackupStatus *qmp_query_backup(Error **errp)
- {
--    assert(qemu_in_coroutine());
--
--    QmpQueryBackupTask *task = opaque;
--
-     BackupStatus *info = g_malloc0(sizeof(*info));
- 
--    qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+    qemu_mutex_lock(&backup_state.stat.lock);
- 
-     if (!backup_state.stat.start_time) {
-         /* not started, return {} */
--        task->result = info;
--        qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--        return;
-+        qemu_mutex_unlock(&backup_state.stat.lock);
-+        return info;
-     }
- 
-     info->has_status = true;
-@@ -937,19 +951,7 @@ static void coroutine_fn pvebackup_co_query(void *opaque)
-     info->has_transferred = true;
-     info->transferred = backup_state.stat.transferred;
- 
--    task->result = info;
-+    qemu_mutex_unlock(&backup_state.stat.lock);
- 
--    qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
--}
--
--BackupStatus *qmp_query_backup(Error **errp)
--{
--    QmpQueryBackupTask task = {
--        .errp = errp,
--        .result = NULL,
--    };
--
--    block_on_coroutine_fn(pvebackup_co_query, &task);
--
--    return task.result;
-+    return info;
- }

debian/patches/pve/0030-PVE-Backup-pbs-restore-new-command-to-restore-from-p.patch

@@ -6,39 +6,32 @@ Subject: [PATCH] PVE-Backup: pbs-restore - new command to restore from proxmox
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
- Makefile      |   4 +-
- pbs-restore.c | 217 ++++++++++++++++++++++++++++++++++++++++++++++++++
- 2 files changed, 220 insertions(+), 1 deletion(-)
+ meson.build   |   4 +
+ pbs-restore.c | 223 ++++++++++++++++++++++++++++++++++++++++++++++++++
+ 2 files changed, 227 insertions(+)
  create mode 100644 pbs-restore.c
 
-diff --git a/Makefile b/Makefile
-index aec216968d..b73da29f24 100644
---- a/Makefile
-+++ b/Makefile
-@@ -479,7 +479,7 @@ dummy := $(call unnest-vars,, \
- 
- include $(SRC_PATH)/tests/Makefile.include
- 
--all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
-+all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) pbs-restore$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
- 
- qemu-version.h: FORCE
- 	$(call quiet-command, \
-@@ -604,6 +604,8 @@ qemu-io$(EXESUF): qemu-io.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-o
- qemu-storage-daemon$(EXESUF): qemu-storage-daemon.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(chardev-obj-y) $(io-obj-y) $(qom-obj-y) $(storage-daemon-obj-y) $(COMMON_LDADDS)
- qemu-storage-daemon$(EXESUF): LIBS += -lproxmox_backup_qemu
- vma$(EXESUF): vma.o vma-reader.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
-+pbs-restore$(EXESUF): pbs-restore.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
-+pbs-restore$(EXESUF): LIBS += -lproxmox_backup_qemu
- 
- qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o $(COMMON_LDADDS)
+diff --git a/meson.build b/meson.build
+index 406112d96f..9c46881eb7 100644
+--- a/meson.build
++++ b/meson.build
+@@ -3604,6 +3604,10 @@ 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..4bf37ef1fa
+index 0000000000..2f834cf42e
 --- /dev/null
 +++ b/pbs-restore.c
-@@ -0,0 +1,217 @@
+@@ -0,0 +1,223 @@
 +/*
 + * Qemu image restore helper for Proxmox Backup
 + *
@@ -57,7 +50,6 @@ index 0000000000..4bf37ef1fa
 +#include <getopt.h>
 +#include <string.h>
 +
-+#include "qemu-common.h"
 +#include "qemu/module.h"
 +#include "qemu/error-report.h"
 +#include "qemu/main-loop.h"
@@ -103,7 +95,7 @@ index 0000000000..4bf37ef1fa
 +        }
 +        res = blk_pwrite_zeroes(callback_data->target, offset, data_len, 0);
 +    } else {
-+        res = blk_pwrite(callback_data->target, offset, data, data_len, 0);
++        res = blk_pwrite(callback_data->target, offset, data_len, data, 0);
 +    }
 +
 +    if (res < 0) {
@@ -202,13 +194,19 @@ index 0000000000..4bf37ef1fa
 +        fprintf(stderr, "connecting to repository '%s'\n", repository);
 +    }
 +    char *pbs_error = NULL;
-+    ProxmoxRestoreHandle *conn = proxmox_restore_connect(
++    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) {
@@ -239,7 +237,7 @@ index 0000000000..4bf37ef1fa
 +        fprintf(stderr, "starting to restore snapshot '%s'\n", snapshot);
 +        fflush(stderr); // ensure we do not get printed after the progress log
 +    }
-+    int res = proxmox_restore_image(
++    res = proxmox_restore_image(
 +        conn,
 +        archive_name,
 +        write_callback,
@@ -248,6 +246,7 @@ index 0000000000..4bf37ef1fa
 +        verbose);
 +
 +    proxmox_restore_disconnect(conn);
++    blk_unref(blk);
 +
 +    if (res < 0) {
 +        fprintf(stderr, "restore failed: %s\n", pbs_error);

debian/patches/pve/0031-PVE-Backup-Add-dirty-bitmap-tracking-for-incremental.patch

@@ -20,19 +20,19 @@ 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                   | 95 ++++++++++++++++++++++++++++++----
- qapi/block-core.json           | 12 ++++-
- 6 files changed, 134 insertions(+), 23 deletions(-)
+ 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 d485c3ac79..fdc85a5c0e 100644
+index 60fa93c85e..8b23bdedac 100644
 --- a/block/monitor/block-hmp-cmds.c
 +++ b/block/monitor/block-hmp-cmds.c
-@@ -1038,6 +1038,7 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
+@@ -1044,6 +1044,7 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
          false, NULL, // PBS fingerprint
          false, NULL, // PBS backup-id
          false, 0, // PBS backup-time
@@ -41,10 +41,10 @@ index d485c3ac79..fdc85a5c0e 100644
          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 0e2d166552..3ff014d32a 100644
+index a40b25e906..670f783515 100644
 --- a/monitor/hmp-cmds.c
 +++ b/monitor/hmp-cmds.c
-@@ -218,19 +218,42 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
+@@ -225,19 +225,42 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
              monitor_printf(mon, "End time: %s", ctime(&info->end_time));
          }
  
@@ -132,7 +132,7 @@ index 1dda8b7d8f..8cbf645b2c 100644
  
  
 diff --git a/pve-backup.c b/pve-backup.c
-index d40f3f2fd6..d50f03a050 100644
+index 6af212b9b4..3f97cf6532 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
 @@ -28,6 +28,8 @@
@@ -162,7 +162,7 @@ index d40f3f2fd6..d50f03a050 100644
      BlockDriverState *target;
  } PVEBackupDevInfo;
  
-@@ -105,11 +110,12 @@ static bool pvebackup_error_or_canceled(void)
+@@ -107,11 +112,12 @@ static bool pvebackup_error_or_canceled(void)
      return error_or_canceled;
  }
  
@@ -176,7 +176,7 @@ index d40f3f2fd6..d50f03a050 100644
      qemu_mutex_unlock(&backup_state.stat.lock);
  }
  
-@@ -148,7 +154,8 @@ pvebackup_co_dump_pbs_cb(
+@@ -150,7 +156,8 @@ pvebackup_co_dump_pbs_cb(
          pvebackup_propagate_error(local_err);
          return pbs_res;
      } else {
@@ -186,7 +186,7 @@ index d40f3f2fd6..d50f03a050 100644
      }
  
      return size;
-@@ -208,11 +215,11 @@ pvebackup_co_dump_vma_cb(
+@@ -210,11 +217,11 @@ pvebackup_co_dump_vma_cb(
          } else {
              if (remaining >= VMA_CLUSTER_SIZE) {
                  assert(ret == VMA_CLUSTER_SIZE);
@@ -200,7 +200,7 @@ index d40f3f2fd6..d50f03a050 100644
                  remaining = 0;
              }
          }
-@@ -248,6 +255,18 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
+@@ -250,6 +257,18 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
              if (local_err != NULL) {
                  pvebackup_propagate_error(local_err);
              }
@@ -219,7 +219,7 @@ index d40f3f2fd6..d50f03a050 100644
          }
  
          proxmox_backup_disconnect(backup_state.pbs);
-@@ -303,6 +322,12 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+@@ -305,6 +324,12 @@ 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);
  
@@ -232,7 +232,7 @@ index d40f3f2fd6..d50f03a050 100644
      g_free(di);
  
      qemu_mutex_unlock(&backup_state.backup_mutex);
-@@ -470,12 +495,18 @@ static bool create_backup_jobs(void) {
+@@ -469,12 +494,18 @@ static bool create_backup_jobs(void) {
  
          assert(di->target != NULL);
  
@@ -247,22 +247,22 @@ index d40f3f2fd6..d50f03a050 100644
  
          BlockJob *job = backup_job_create(
 -            NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
--            BITMAP_SYNC_MODE_NEVER, false, NULL, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
+-            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, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-             JOB_DEFAULT, pvebackup_complete_cb, di, 1, NULL, &local_err);
++            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);
-@@ -526,6 +557,8 @@ typedef struct QmpBackupTask {
+@@ -525,6 +556,8 @@ typedef struct QmpBackupTask {
      const char *fingerprint;
      bool has_fingerprint;
      int64_t backup_time;
-+    bool has_incremental;
-+    bool incremental;
++    bool has_use_dirty_bitmap;
++    bool use_dirty_bitmap;
      bool has_format;
      BackupFormat format;
      bool has_config_file;
-@@ -617,6 +650,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -616,6 +649,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
      }
  
      size_t total = 0;
@@ -270,16 +270,16 @@ index d40f3f2fd6..d50f03a050 100644
  
      l = di_list;
      while (l) {
-@@ -654,6 +688,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -653,6 +687,8 @@ 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 incremental = task->has_incremental && task->incremental;
++        bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
 +
          char *pbs_err = NULL;
          pbs = proxmox_backup_new(
              task->backup_file,
-@@ -673,7 +709,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -672,7 +708,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
              goto err;
          }
  
@@ -289,42 +289,50 @@ index d40f3f2fd6..d50f03a050 100644
              goto err;
  
          /* register all devices */
-@@ -684,9 +721,32 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -683,9 +720,40 @@ 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;
 +
-+            bool use_incremental = false;
-+            if (incremental) {
++            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;
 +                    }
-+                    /* mark entire bitmap as dirty to make full backup first */
++                } 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;
-+                } else {
-+                    use_incremental = true;
-+                    dirty += bdrv_get_dirty_count(bitmap);
 +                }
 +                di->bitmap = bitmap;
-+            } else if (bitmap != NULL) {
++            } else {
 +                dirty += di->size;
-+                bdrv_release_dirty_bitmap(bitmap);
++
++                /* 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, use_incremental, task->errp);
++            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;
-@@ -695,6 +755,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -694,6 +762,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
              di->dev_id = dev_id;
          }
      } else if (format == BACKUP_FORMAT_VMA) {
@@ -333,7 +341,7 @@ index d40f3f2fd6..d50f03a050 100644
          vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
          if (!vmaw) {
              if (local_err) {
-@@ -722,6 +784,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -721,6 +791,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
              }
          }
      } else if (format == BACKUP_FORMAT_DIR) {
@@ -342,18 +350,18 @@ index d40f3f2fd6..d50f03a050 100644
          if (mkdir(task->backup_file, 0640) != 0) {
              error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
                               task->backup_file);
-@@ -794,8 +858,10 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -793,8 +865,10 @@ 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 = dirty >= total ? 0 : total - dirty;
++    backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
  
      qemu_mutex_unlock(&backup_state.stat.lock);
  
-@@ -819,6 +885,10 @@ err:
+@@ -818,6 +892,10 @@ err:
          PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
          l = g_list_next(l);
  
@@ -364,24 +372,24 @@ index d40f3f2fd6..d50f03a050 100644
          if (di->target) {
              bdrv_unref(di->target);
          }
-@@ -860,6 +930,7 @@ UuidInfo *qmp_backup(
+@@ -859,6 +937,7 @@ 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_incremental, bool incremental,
++    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,
-@@ -878,6 +949,8 @@ UuidInfo *qmp_backup(
+@@ -877,6 +956,8 @@ UuidInfo *qmp_backup(
          .backup_id = backup_id,
          .has_backup_time = has_backup_time,
          .backup_time = backup_time,
-+        .has_incremental = has_incremental,
-+        .incremental = incremental,
++        .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,
-@@ -946,10 +1019,14 @@ BackupStatus *qmp_query_backup(Error **errp)
+@@ -945,10 +1026,14 @@ BackupStatus *qmp_query_backup(Error **errp)
  
      info->has_total = true;
      info->total = backup_state.stat.total;
@@ -397,14 +405,14 @@ index d40f3f2fd6..d50f03a050 100644
      qemu_mutex_unlock(&backup_state.stat.lock);
  
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 9db8e26517..6cad1e0e38 100644
+index c3b6b93472..992e6c1e3f 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -767,8 +767,13 @@
+@@ -753,8 +753,13 @@
  #
  # @total: total amount of bytes involved in the backup process
  #
-+# @dirty: with incremental mode, this is the amount of bytes involved
++# @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.
@@ -414,7 +422,7 @@ index 9db8e26517..6cad1e0e38 100644
  # @zero-bytes: amount of 'zero' bytes detected.
  #
  # @start-time: time (epoch) when backup job started.
-@@ -781,8 +786,8 @@
+@@ -767,8 +772,8 @@
  #
  ##
  { 'struct': 'BackupStatus',
@@ -425,20 +433,20 @@ index 9db8e26517..6cad1e0e38 100644
             '*start-time': 'int', '*end-time': 'int',
             '*backup-file': 'str', '*uuid': 'str' } }
  
-@@ -825,6 +830,8 @@
+@@ -811,6 +816,8 @@
  #
  # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
  #
-+# @incremental: sync incremental changes since last job (optional 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
  #
  ##
-@@ -835,6 +842,7 @@
+@@ -821,6 +828,7 @@
                                      '*fingerprint': 'str',
                                      '*backup-id': 'str',
                                      '*backup-time': 'int',
-+                                    '*incremental': 'bool',
++                                    '*use-dirty-bitmap': 'bool',
                                      '*format': 'BackupFormat',
                                      '*config-file': 'str',
                                      '*firewall-file': 'str',

debian/patches/pve/0032-PVE-various-PBS-fixes.patch

@@ -0,0 +1,219 @@
+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 8b23bdedac..f59b02592e 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1044,7 +1044,9 @@ 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 3f97cf6532..a275a1d4e1 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];
+@@ -137,10 +139,13 @@ 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
+@@ -149,17 +154,29 @@ 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;
+ }
+ 
+@@ -180,6 +197,7 @@ pvebackup_co_dump_vma_cb(
+     int ret = -1;
+ 
+     assert(backup_state.vmaw);
++    assert(buf);
+ 
+     uint64_t remaining = size;
+ 
+@@ -206,9 +224,7 @@ 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);
+@@ -566,6 +582,10 @@ 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;
+@@ -689,6 +709,7 @@ 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,
+@@ -698,8 +719,10 @@ 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,
+@@ -718,6 +741,8 @@ 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);
+@@ -938,6 +963,8 @@ 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,
+@@ -948,6 +975,8 @@ 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,
+@@ -958,6 +987,10 @@ 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 992e6c1e3f..5ac6276dc1 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',

debian/patches/pve/0033-PVE-Add-PBS-block-driver-to-map-backup-archives-into.patch

@@ -6,37 +6,40 @@ 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>
+[FE: adapt to changed function signatures
+     make pbs_co_preadv return values consistent with QEMU]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- block/Makefile.objs  |   2 +
- block/pbs.c          | 271 +++++++++++++++++++++++++++++++++++++++++++
- configure            |  10 ++
- qapi/block-core.json |  14 ++-
- 4 files changed, 296 insertions(+), 1 deletion(-)
+ block/meson.build    |   3 +
+ block/pbs.c          | 276 +++++++++++++++++++++++++++++++++++++++++++
+ configure            |   9 ++
+ meson.build          |   2 +-
+ qapi/block-core.json |  13 ++
+ qapi/pragma.json     |   1 +
+ 6 files changed, 303 insertions(+), 1 deletion(-)
  create mode 100644 block/pbs.c
 
-diff --git a/block/Makefile.objs b/block/Makefile.objs
-index 9ea0477d0b..28fb3b7f7c 100644
---- a/block/Makefile.objs
-+++ b/block/Makefile.objs
-@@ -5,6 +5,7 @@ block-obj-$(CONFIG_CLOOP) += cloop.o
- block-obj-$(CONFIG_BOCHS) += bochs.o
- block-obj-$(CONFIG_VVFAT) += vvfat.o
- block-obj-$(CONFIG_DMG) += dmg.o
-+block-obj-$(CONFIG_PBS_BDRV) += pbs.o
+diff --git a/block/meson.build b/block/meson.build
+index e995ae72b9..7ef2fa72d5 100644
+--- a/block/meson.build
++++ b/block/meson.build
+@@ -53,6 +53,9 @@ block_ss.add(files(
+   '../pve-backup.c',
+ ), libproxmox_backup_qemu)
  
- block-obj-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o qcow2-cache.o qcow2-bitmap.o qcow2-threads.o
- block-obj-$(CONFIG_QED) += qed.o qed-l2-cache.o qed-table.o qed-cluster.o
-@@ -75,3 +76,4 @@ io_uring.o-cflags  := $(LINUX_IO_URING_CFLAGS)
- io_uring.o-libs    := $(LINUX_IO_URING_LIBS)
- parallels.o-cflags := $(LIBXML2_CFLAGS)
- parallels.o-libs   := $(LIBXML2_LIBS)
-+pbs.o-libs         := -lproxmox_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'))
+ softmmu_ss.add(files('block-ram-registrar.c'))
 diff --git a/block/pbs.c b/block/pbs.c
 new file mode 100644
-index 0000000000..1481a2bfd1
+index 0000000000..9d1f1f39d4
 --- /dev/null
 +++ b/block/pbs.c
-@@ -0,0 +1,271 @@
+@@ -0,0 +1,276 @@
 +/*
 + * Proxmox Backup Server read-only block driver
 + */
@@ -233,20 +236,25 @@ index 0000000000..1481a2bfd1
 +}
 +
 +static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
-+                                      uint64_t offset, uint64_t bytes,
-+                                      QEMUIOVector *qiov, int flags)
++                                      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 -EIO;
++    }
++
 +    ReadCallbackData rcb = {
 +        .co = qemu_coroutine_self(),
-+        .ctx = qemu_get_current_aio_context(),
++        .ctx = bdrv_get_aio_context(bs),
 +    };
 +
-+    proxmox_restore_read_image_at_async(s->conn, s->aid, buf, offset, bytes,
++    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();
@@ -260,12 +268,12 @@ index 0000000000..1481a2bfd1
 +    qemu_iovec_from_buf(qiov, 0, buf, bytes);
 +    free(buf);
 +
-+    return ret;
++    return 0;
 +}
 +
 +static coroutine_fn int pbs_co_pwritev(BlockDriverState *bs,
-+                                       uint64_t offset, uint64_t bytes,
-+                                       QEMUIOVector *qiov, int flags)
++                                       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");
@@ -309,68 +317,72 @@ index 0000000000..1481a2bfd1
 +
 +block_init(bdrv_pbs_init);
 diff --git a/configure b/configure
-index 2acc4d1465..3fc80d0c2c 100755
+index 5f1828f1ec..c9b70b5e9a 100755
 --- a/configure
 +++ b/configure
-@@ -510,6 +510,7 @@ vvfat="yes"
- qed="yes"
- parallels="yes"
- sheepdog="yes"
+@@ -285,6 +285,7 @@ linux_user=""
+ bsd_user=""
+ pie=""
+ coroutine=""
 +pbs_bdrv="yes"
- libxml2=""
- debug_mutex="no"
- libpmem=""
-@@ -1576,6 +1577,10 @@ for opt do
-   ;;
-   --enable-sheepdog) sheepdog="yes"
+ plugins="$default_feature"
+ meson=""
+ ninja=""
+@@ -864,6 +865,10 @@ for opt do
+   --enable-uuid|--disable-uuid)
+       echo "$0: $opt is obsolete, UUID support is always built" >&2
    ;;
 +  --disable-pbs-bdrv) pbs_bdrv="no"
 +  ;;
 +  --enable-pbs-bdrv) pbs_bdrv="yes"
 +  ;;
-   --disable-vhost-user) vhost_user="no"
+   --with-git=*) git="$optarg"
    ;;
-   --enable-vhost-user) vhost_user="yes"
-@@ -1936,6 +1941,7 @@ disabled with --disable-FEATURE, default is enabled if available:
-   qed             qed image format support
-   parallels       parallels image format support
-   sheepdog        sheepdog block driver support
+   --with-git-submodules=*)
+@@ -1049,6 +1054,7 @@ 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
-   crypto-afalg    Linux AF_ALG crypto backend driver
-   capstone        capstone disassembler support
-   debug-mutex     mutex debugging support
-@@ -7009,6 +7015,7 @@ echo "vvfat support     $vvfat"
- echo "qed support       $qed"
- echo "parallels support $parallels"
- echo "sheepdog support  $sheepdog"
-+echo "pbs-bdrv support  $pbs_bdrv"
- echo "capstone          $capstone"
- echo "libpmem support   $libpmem"
- echo "libdaxctl support $libdaxctl"
-@@ -7885,6 +7892,9 @@ fi
- if test "$sheepdog" = "yes" ; then
-   echo "CONFIG_SHEEPDOG=y" >> $config_host_mak
+ 
+ NOTE: The object files are built at the place where configure is launched
+ EOF
+@@ -2372,6 +2378,9 @@ echo "TARGET_DIRS=$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" >> $config_host_mak
 +fi
- if test "$pty_h" = "yes" ; then
-   echo "HAVE_PTY_H=y" >> $config_host_mak
- fi
+ 
+ # XXX: suppress that
+ if [ "$bsd" = "yes" ] ; then
+diff --git a/meson.build b/meson.build
+index 9c46881eb7..93ebda47af 100644
+--- a/meson.build
++++ b/meson.build
+@@ -3980,7 +3980,7 @@ 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 a177fea6cd..f782c2cf96 100644
+index 5ac6276dc1..45b63dfe26 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -2951,7 +2951,7 @@
-             'luks', 'nbd', 'nfs', 'null-aio', 'null-co', 'nvme', 'parallels',
-             'qcow', 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
-             { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' },
--            'sheepdog',
-+            'sheepdog', 'pbs',
-             'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat' ] }
- 
- ##
-@@ -3015,6 +3015,17 @@
+@@ -3103,6 +3103,7 @@
+             'parallels', 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum',
+             'raw', 'rbd',
+             { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
++            'pbs',
+             'ssh', 'throttle', 'vdi', 'vhdx',
+             { 'name': 'virtio-blk-vfio-pci', 'if': 'CONFIG_BLKIO' },
+             { 'name': 'virtio-blk-vhost-user', 'if': 'CONFIG_BLKIO' },
+@@ -3179,6 +3180,17 @@
  { 'struct': 'BlockdevOptionsNull',
    'data': { '*size': 'int', '*latency-ns': 'uint64', '*read-zeroes': 'bool' } }
  
@@ -388,11 +400,23 @@ index a177fea6cd..f782c2cf96 100644
  ##
  # @BlockdevOptionsNVMe:
  #
-@@ -4121,6 +4132,7 @@
+@@ -4531,6 +4543,7 @@
        'nfs':        'BlockdevOptionsNfs',
        'null-aio':   'BlockdevOptionsNull',
        'null-co':    'BlockdevOptionsNull',
 +      'pbs':        'BlockdevOptionsPbs',
        'nvme':       'BlockdevOptionsNVMe',
-       'parallels':  'BlockdevOptionsGenericFormat',
-       'qcow2':      'BlockdevOptionsQcow2',
+       'nvme-io_uring': { 'type': 'BlockdevOptionsNvmeIoUring',
+                          'if': 'CONFIG_BLKIO' },
+diff --git a/qapi/pragma.json b/qapi/pragma.json
+index f2097b9020..5ab1890519 100644
+--- a/qapi/pragma.json
++++ b/qapi/pragma.json
+@@ -47,6 +47,7 @@
+         'BlockInfo',                # query-block
+         'BlockdevAioOptions',       # blockdev-add, -blockdev
+         'BlockdevDriver',           # blockdev-add, query-blockstats, ...
++        'BlockdevOptionsPbs',       # for PBS backwards compat
+         'BlockdevVmdkAdapterType',  # blockdev-create (to match VMDK spec)
+         'BlockdevVmdkSubformat',    # blockdev-create (to match VMDK spec)
+         'ColoCompareProperties',    # object_add, -object

debian/patches/pve/0034-PVE-add-query_proxmox_support-QMP-command.patch

@@ -11,15 +11,15 @@ 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         |  8 ++++++++
- qapi/block-core.json | 25 +++++++++++++++++++++++++
- 2 files changed, 33 insertions(+)
+ 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 bfb648d6b5..6bf138cfc6 100644
+index a275a1d4e1..b10373cd8a 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
-@@ -1051,3 +1051,11 @@ BackupStatus *qmp_query_backup(Error **errp)
+@@ -1072,3 +1072,12 @@ BackupStatus *qmp_query_backup(Error **errp)
  
      return info;
  }
@@ -29,13 +29,14 @@ index bfb648d6b5..6bf138cfc6 100644
 +    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 f782c2cf96..1ed5987c88 100644
+index 45b63dfe26..8b0e0d92de 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -877,6 +877,31 @@
+@@ -863,6 +863,35 @@
  ##
  { 'command': 'backup-cancel' }
  
@@ -47,11 +48,15 @@ index f782c2cf96..1ed5987c88 100644
 +# @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' } }
 +
 +##

debian/patches/pve/0035-PVE-add-query-pbs-bitmap-info-QMP-call.patch

@@ -7,6 +7,7 @@ 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 ++++++++++++++++++++++++++++++++-----------
@@ -14,10 +15,10 @@ Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
  3 files changed, 159 insertions(+), 42 deletions(-)
 
 diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 3ff014d32a..c3227a1498 100644
+index 670f783515..d819e5fc36 100644
 --- a/monitor/hmp-cmds.c
 +++ b/monitor/hmp-cmds.c
-@@ -195,6 +195,7 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
+@@ -202,6 +202,7 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
  void hmp_info_backup(Monitor *mon, const QDict *qdict)
  {
      BackupStatus *info;
@@ -25,7 +26,7 @@ index 3ff014d32a..c3227a1498 100644
  
      info = qmp_query_backup(NULL);
  
-@@ -225,26 +226,29 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
+@@ -232,26 +233,29 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
              // this should not happen normally
              monitor_printf(mon, "Total size: %d\n", 0);
          } else {
@@ -68,7 +69,7 @@ index 3ff014d32a..c3227a1498 100644
                             info->zero_bytes, zero_per);
  
 diff --git a/pve-backup.c b/pve-backup.c
-index bd802c6205..2db4a62580 100644
+index b10373cd8a..8ae50e06c3 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
 @@ -46,6 +46,7 @@ static struct PVEBackupState {
@@ -79,7 +80,7 @@ index bd802c6205..2db4a62580 100644
      } stat;
      int64_t speed;
      VmaWriter *vmaw;
-@@ -670,7 +671,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -669,7 +670,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
      }
  
      size_t total = 0;
@@ -87,7 +88,7 @@ index bd802c6205..2db4a62580 100644
  
      l = di_list;
      while (l) {
-@@ -691,18 +691,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -690,18 +690,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
  
      uuid_generate(uuid);
  
@@ -124,7 +125,7 @@ index bd802c6205..2db4a62580 100644
          }
  
          int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
-@@ -729,12 +744,12 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -728,12 +743,12 @@ 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);
@@ -139,7 +140,7 @@ index bd802c6205..2db4a62580 100644
  
          /* register all devices */
          l = di_list;
-@@ -745,6 +760,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -744,6 +759,8 @@ 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);
@@ -148,7 +149,7 @@ index bd802c6205..2db4a62580 100644
  
              BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
              bool expect_only_dirty = false;
-@@ -753,49 +770,59 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -752,49 +769,59 @@ 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) {
@@ -218,7 +219,7 @@ index bd802c6205..2db4a62580 100644
          }
  
          /* register all devices for vma writer */
-@@ -805,7 +832,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -804,7 +831,7 @@ 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))) {
@@ -227,7 +228,7 @@ index bd802c6205..2db4a62580 100644
              }
  
              const char *devname = bdrv_get_device_name(di->bs);
-@@ -813,16 +840,14 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -812,16 +839,14 @@ 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");
@@ -246,7 +247,7 @@ index bd802c6205..2db4a62580 100644
          }
          backup_dir = task->backup_file;
  
-@@ -839,18 +864,18 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -838,18 +863,18 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
                              di->size, flags, false, &local_err);
              if (local_err) {
                  error_propagate(task->errp, local_err);
@@ -268,7 +269,7 @@ index bd802c6205..2db4a62580 100644
      }
  
  
-@@ -858,7 +883,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -857,7 +882,7 @@ 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) {
@@ -277,7 +278,7 @@ index bd802c6205..2db4a62580 100644
          }
      }
  
-@@ -866,12 +891,11 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -865,12 +890,11 @@ 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) {
@@ -292,7 +293,7 @@ index bd802c6205..2db4a62580 100644
  
      if (backup_state.stat.error) {
          error_free(backup_state.stat.error);
-@@ -891,10 +915,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -890,10 +914,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
      char *uuid_str = g_strdup(backup_state.stat.uuid_str);
  
      backup_state.stat.total = total;
@@ -304,7 +305,7 @@ index bd802c6205..2db4a62580 100644
  
      qemu_mutex_unlock(&backup_state.stat.lock);
  
-@@ -911,6 +934,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -910,6 +933,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
      task->result = uuid_info;
      return;
  
@@ -314,7 +315,7 @@ index bd802c6205..2db4a62580 100644
  err:
  
      l = di_list;
-@@ -1074,10 +1100,41 @@ BackupStatus *qmp_query_backup(Error **errp)
+@@ -1073,11 +1099,42 @@ BackupStatus *qmp_query_backup(Error **errp)
      return info;
  }
  
@@ -353,29 +354,32 @@ index bd802c6205..2db4a62580 100644
      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 1ed5987c88..03fc0af99b 100644
+index 8b0e0d92de..7fde927621 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -885,11 +885,14 @@
+@@ -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-library-version: Running version of libproxmox-backup-qemu0 library.
+ # @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' } }
  
- ##
-@@ -902,6 +905,59 @@
+@@ -892,6 +895,59 @@
  ##
  { 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
  

debian/patches/pve/0036-PVE-redirect-stderr-to-journal-when-daemonized.patch

@@ -0,0 +1,61 @@
+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 93ebda47af..d47ce5fe64 100644
+--- a/meson.build
++++ b/meson.build
+@@ -1526,6 +1526,7 @@ 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
+@@ -3094,6 +3095,7 @@ 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 4858650c3e..c5cb12226a 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"
+@@ -287,9 +289,10 @@ 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);

debian/patches/pve/0037-PVE-Add-sequential-job-transaction-support.patch

@@ -4,16 +4,17 @@ 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              | 24 ++++++++++++++++++++++++
- 2 files changed, 36 insertions(+)
+ job.c              | 34 ++++++++++++++++++++++++++++++++++
+ 2 files changed, 46 insertions(+)
 
 diff --git a/include/qemu/job.h b/include/qemu/job.h
-index 32aabb1c60..f7a6a0926a 100644
+index e502787dd8..963cf2bef5 100644
 --- a/include/qemu/job.h
 +++ b/include/qemu/job.h
-@@ -280,6 +280,18 @@ typedef enum JobCreateFlags {
+@@ -381,6 +381,18 @@ void job_unlock(void);
   */
  JobTxn *job_txn_new(void);
  
@@ -33,10 +34,10 @@ index 32aabb1c60..f7a6a0926a 100644
   * 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 b8139c80a4..97ee97a192 100644
+index 93e22d180b..2b31f1e14f 100644
 --- a/job.c
 +++ b/job.c
-@@ -72,6 +72,8 @@ struct JobTxn {
+@@ -93,6 +93,8 @@ struct JobTxn {
  
      /* Reference count */
      int refcnt;
@@ -44,8 +45,8 @@ index b8139c80a4..97ee97a192 100644
 +    bool sequential;
  };
  
- /* Right now, this mutex is only needed to synchronize accesses to job->busy
-@@ -102,6 +104,25 @@ JobTxn *job_txn_new(void)
+ void job_lock(void)
+@@ -118,6 +120,25 @@ JobTxn *job_txn_new(void)
      return txn;
  }
  
@@ -68,16 +69,33 @@ index b8139c80a4..97ee97a192 100644
 +    job_start(first);
 +}
 +
- static void job_txn_ref(JobTxn *txn)
+ /* Called with job_mutex held. */
+ static void job_txn_ref_locked(JobTxn *txn)
  {
-     txn->refcnt++;
-@@ -841,6 +862,9 @@ static void job_completed_txn_success(Job *job)
+@@ -1057,6 +1078,12 @@ static void job_completed_txn_success_locked(Job *job)
       */
      QLIST_FOREACH(other_job, &txn->jobs, txn_list) {
-         if (!job_is_completed(other_job)) {
+         if (!job_is_completed_locked(other_job)) {
 +            if (txn->sequential) {
++                job_unlock();
++                /* Needs to be called without holding the job lock */
 +                job_start(other_job);
++                job_lock();
 +            }
              return;
          }
          assert(other_job->ret == 0);
+@@ -1268,6 +1295,13 @@ int job_finish_sync_locked(Job *job,
+         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_locked(job);
++    }
++
+     job_unlock();
+     AIO_WAIT_WHILE_UNLOCKED(job->aio_context,
+                             (job_enter(job), !job_is_completed(job)));

debian/patches/pve/0038-PVE-Backup-Use-a-transaction-to-synchronize-job-stat.patch

@@ -11,12 +11,16 @@ 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>
+[FE: add new force parameter to job_cancel_sync calls
+     adapt for new job lock mechanism replacing AioContext locks]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- pve-backup.c | 167 +++++++++++++++------------------------------------
- 1 file changed, 49 insertions(+), 118 deletions(-)
+ pve-backup.c | 163 ++++++++++++++++-----------------------------------
+ 1 file changed, 50 insertions(+), 113 deletions(-)
 
 diff --git a/pve-backup.c b/pve-backup.c
-index 2db4a62580..b52f4a9364 100644
+index 8ae50e06c3..eedac335ec 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
 @@ -52,6 +52,7 @@ static struct PVEBackupState {
@@ -27,7 +31,7 @@ index 2db4a62580..b52f4a9364 100644
      QemuMutex backup_mutex;
      CoMutex dump_callback_mutex;
  } backup_state;
-@@ -71,32 +72,12 @@ typedef struct PVEBackupDevInfo {
+@@ -71,34 +72,12 @@ typedef struct PVEBackupDevInfo {
      size_t size;
      uint64_t block_size;
      uint8_t dev_id;
@@ -44,14 +48,16 @@ index 2db4a62580..b52f4a9364 100644
 -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;
--            }
+-        WITH_JOB_LOCK_GUARD() {
+-            for (BlockJob *job = block_job_next_locked(NULL); job; job = block_job_next_locked(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;
+-                BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
+-                if (bjob && bjob->source_bs == di->bs) {
+-                    return job;
+-                }
 -            }
 -        }
 -    }
@@ -61,7 +67,7 @@ index 2db4a62580..b52f4a9364 100644
  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)
+@@ -274,18 +253,6 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
              if (local_err != NULL) {
                  pvebackup_propagate_error(local_err);
              }
@@ -80,7 +86,7 @@ index 2db4a62580..b52f4a9364 100644
          }
  
          proxmox_backup_disconnect(backup_state.pbs);
-@@ -322,8 +291,6 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+@@ -324,8 +291,6 @@ static void pvebackup_complete_cb(void *opaque, int ret)
  
      qemu_mutex_lock(&backup_state.backup_mutex);
  
@@ -89,7 +95,7 @@ index 2db4a62580..b52f4a9364 100644
      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)
+@@ -338,20 +303,17 @@ static void pvebackup_complete_cb(void *opaque, int ret)
  
      block_on_coroutine_fn(pvebackup_complete_stream, di);
  
@@ -116,7 +122,7 @@ index 2db4a62580..b52f4a9364 100644
  }
  
  static void pvebackup_cancel(void)
-@@ -371,36 +335,28 @@ static void pvebackup_cancel(void)
+@@ -373,32 +335,28 @@ static void pvebackup_cancel(void)
          proxmox_backup_abort(backup_state.pbs, "backup canceled");
      }
  
@@ -125,13 +131,6 @@ index 2db4a62580..b52f4a9364 100644
 -    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);
@@ -139,15 +138,23 @@ index 2db4a62580..b52f4a9364 100644
 +        ((PVEBackupDevInfo *)bdi->data)->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;
 -            }
--        }
 +    /* ref the job before releasing the mutex, just to be safe */
 +    if (cancel_job) {
-+        job_ref(&cancel_job->job);
++        WITH_JOB_LOCK_GUARD() {
++            job_ref_locked(&cancel_job->job);
+         }
 +    }
  
 -        qemu_mutex_unlock(&backup_state.backup_mutex);
@@ -156,27 +163,21 @@ index 2db4a62580..b52f4a9364 100644
 +    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);
--            aio_context_release(aio_context);
+-            job_cancel_sync(&next_job->job, true);
 -        } else {
 -            break;
--        }
 +    if (cancel_job) {
-+        AioContext *aio_context = cancel_job->job.aio_context;
-+        aio_context_acquire(aio_context);
-+        job_cancel_sync(&cancel_job->job);
-+        job_unref(&cancel_job->job);
-+        aio_context_release(aio_context);
++        WITH_JOB_LOCK_GUARD() {
++            job_cancel_sync_locked(&cancel_job->job, true);
++            job_unref_locked(&cancel_job->job);
+         }
      }
  }
- 
-@@ -459,51 +415,19 @@ static int coroutine_fn pvebackup_co_add_config(
+@@ -458,49 +416,19 @@ static int coroutine_fn pvebackup_co_add_config(
      goto out;
  }
  
--bool job_should_pause(Job *job);
+-bool job_should_pause_locked(Job *job);
 -
 -static void pvebackup_run_next_job(void)
 -{
@@ -194,18 +195,16 @@ index 2db4a62580..b52f4a9364 100644
 -        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);
--                } else {
--                    job_resume(&job->job);
+-            WITH_JOB_LOCK_GUARD() {
+-                if (job_should_pause_locked(&job->job)) {
+-                    bool error_or_canceled = pvebackup_error_or_canceled();
+-                    if (error_or_canceled) {
+-                        job_cancel_sync_locked(&job->job, true);
+-                    } else {
+-                        job_resume_locked(&job->job);
+-                    }
 -                }
 -            }
--            aio_context_release(aio_context);
 -            return;
 -        }
 -    }
@@ -228,19 +227,19 @@ index 2db4a62580..b52f4a9364 100644
 +    }
 +    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) {
-@@ -524,7 +448,7 @@ static bool create_backup_jobs(void) {
+@@ -523,7 +451,7 @@ 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, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
--            JOB_DEFAULT, pvebackup_complete_cb, di, 1, NULL, &local_err);
+             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);
  
-@@ -536,7 +460,8 @@ static bool create_backup_jobs(void) {
+@@ -535,7 +463,8 @@ static bool create_backup_jobs(void) {
              pvebackup_propagate_error(create_job_err);
              break;
          }
@@ -250,18 +249,20 @@ index 2db4a62580..b52f4a9364 100644
  
          bdrv_unref(di->target);
          di->target = NULL;
-@@ -554,6 +479,10 @@ static bool create_backup_jobs(void) {
+@@ -553,6 +482,12 @@ static bool create_backup_jobs(void) {
                  bdrv_unref(di->target);
                  di->target = NULL;
              }
 +
 +            if (di->job) {
-+                job_unref(&di->job->job);
++                WITH_JOB_LOCK_GUARD() {
++                    job_unref_locked(&di->job->job);
++                }
 +            }
          }
      }
  
-@@ -944,10 +873,6 @@ err:
+@@ -943,10 +878,6 @@ err:
          PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
          l = g_list_next(l);
  
@@ -272,7 +273,7 @@ index 2db4a62580..b52f4a9364 100644
          if (di->target) {
              bdrv_unref(di->target);
          }
-@@ -1036,9 +961,15 @@ UuidInfo *qmp_backup(
+@@ -1035,9 +966,15 @@ UuidInfo *qmp_backup(
      block_on_coroutine_fn(pvebackup_co_prepare, &task);
  
      if (*errp == NULL) {

debian/patches/pve/0038-PVE-backup-rename-incremental-to-use-dirty-bitmap.patch

@@ -1,126 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Mon, 6 Jul 2020 20:05:16 +0200
-Subject: [PATCH] PVE backup: rename incremental to use-dirty-bitmap
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pve-backup.c         | 22 +++++++++++-----------
- qapi/block-core.json |  6 +++---
- 2 files changed, 14 insertions(+), 14 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index d50f03a050..7bf54b4c5d 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -557,8 +557,8 @@ typedef struct QmpBackupTask {
-     const char *fingerprint;
-     bool has_fingerprint;
-     int64_t backup_time;
--    bool has_incremental;
--    bool incremental;
-+    bool has_use_dirty_bitmap;
-+    bool use_dirty_bitmap;
-     bool has_format;
-     BackupFormat format;
-     bool has_config_file;
-@@ -688,7 +688,7 @@ 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 incremental = task->has_incremental && task->incremental;
-+        bool use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
- 
-         char *pbs_err = NULL;
-         pbs = proxmox_backup_new(
-@@ -722,9 +722,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-             const char *devname = bdrv_get_device_name(di->bs);
- 
-             BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
-+            bool expect_only_dirty = false;
- 
--            bool use_incremental = false;
--            if (incremental) {
-+            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) {
-@@ -734,7 +734,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-                     bdrv_set_dirty_bitmap(bitmap, 0, di->size);
-                     dirty += di->size;
-                 } else {
--                    use_incremental = true;
-+                    expect_only_dirty = true;
-                     dirty += bdrv_get_dirty_count(bitmap);
-                 }
-                 di->bitmap = bitmap;
-@@ -743,7 +743,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-                 bdrv_release_dirty_bitmap(bitmap);
-             }
- 
--            int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, use_incremental, task->errp);
-+            int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
-             if (dev_id < 0) {
-                 goto err;
-             }
-@@ -861,7 +861,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-     backup_state.stat.dirty = dirty;
-     backup_state.stat.transferred = 0;
-     backup_state.stat.zero_bytes = 0;
--    backup_state.stat.reused = dirty >= total ? 0 : total - dirty;
-+    backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
- 
-     qemu_mutex_unlock(&backup_state.stat.lock);
- 
-@@ -930,7 +930,7 @@ 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_incremental, bool incremental,
-+    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,
-@@ -949,8 +949,8 @@ UuidInfo *qmp_backup(
-         .backup_id = backup_id,
-         .has_backup_time = has_backup_time,
-         .backup_time = backup_time,
--        .has_incremental = has_incremental,
--        .incremental = incremental,
-+        .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,
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 6cad1e0e38..e00e577c6c 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -767,7 +767,7 @@
- #
- # @total: total amount of bytes involved in the backup process
- #
--# @dirty: with incremental mode, this is the amount of bytes involved
-+# @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.
-@@ -830,7 +830,7 @@
- #
- # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
- #
--# @incremental: sync incremental changes since last job (optional 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
- #
-@@ -842,7 +842,7 @@
-                                     '*fingerprint': 'str',
-                                     '*backup-id': 'str',
-                                     '*backup-time': 'int',
--                                    '*incremental': 'bool',
-+                                    '*use-dirty-bitmap': 'bool',
-                                     '*format': 'BackupFormat',
-                                     '*config-file': 'str',
-                                     '*firewall-file': 'str',

debian/patches/pve/0039-PVE-Backup-Don-t-block-on-finishing-and-cleanup-crea.patch

@@ -1,7 +1,8 @@
 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: Use more coroutines and don't block on finishing
+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
@@ -29,16 +30,34 @@ 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>
+[FE: add new force parameter to job_cancel_sync calls]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- pve-backup.c         | 148 ++++++++++++++++++++++++++-----------------
+ pve-backup.c         | 212 +++++++++++++++++++++++++++----------------
  qapi/block-core.json |   5 +-
- 2 files changed, 95 insertions(+), 58 deletions(-)
+ 2 files changed, 138 insertions(+), 79 deletions(-)
 
 diff --git a/pve-backup.c b/pve-backup.c
-index b52f4a9364..4402c0cb17 100644
+index eedac335ec..7bd9d06346 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
 @@ -33,7 +33,9 @@ const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
@@ -52,11 +71,12 @@ index b52f4a9364..4402c0cb17 100644
          QemuMutex lock;
          Error *error;
          time_t start_time;
-@@ -47,20 +49,21 @@ static struct PVEBackupState {
+@@ -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;
@@ -76,7 +96,7 @@ index b52f4a9364..4402c0cb17 100644
      qemu_co_mutex_init(&backup_state.dump_callback_mutex);
  }
  
-@@ -72,6 +75,7 @@ typedef struct PVEBackupDevInfo {
+@@ -72,6 +76,7 @@ typedef struct PVEBackupDevInfo {
      size_t size;
      uint64_t block_size;
      uint8_t dev_id;
@@ -84,7 +104,7 @@ index b52f4a9364..4402c0cb17 100644
      char targetfile[PATH_MAX];
      BdrvDirtyBitmap *bitmap;
      BlockDriverState *target;
-@@ -227,12 +231,12 @@ pvebackup_co_dump_vma_cb(
+@@ -227,12 +232,12 @@ pvebackup_co_dump_vma_cb(
  }
  
  // assumes the caller holds backup_mutex
@@ -99,7 +119,7 @@ index b52f4a9364..4402c0cb17 100644
      qemu_mutex_unlock(&backup_state.stat.lock);
  
      if (backup_state.vmaw) {
-@@ -261,12 +265,29 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
+@@ -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;
@@ -116,25 +136,28 @@ index b52f4a9364..4402c0cb17 100644
  {
      PVEBackupDevInfo *di = opaque;
 +    int ret = di->completed_ret;
-+
-+    qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
-+    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);
  
-     bool error_or_canceled = pvebackup_error_or_canceled();
- 
-@@ -281,27 +302,6 @@ static void coroutine_fn pvebackup_complete_stream(void *opaque)
-             pvebackup_propagate_error(local_err);
-         }
+-    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)
@@ -144,22 +167,32 @@ index b52f4a9364..4402c0cb17 100644
 -    PVEBackupDevInfo *di = opaque;
 -
 -    qemu_mutex_lock(&backup_state.backup_mutex);
--
--    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);
--
++    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 +310,49 @@ static void pvebackup_complete_cb(void *opaque, int ret)
+@@ -310,21 +321,46 @@ 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)) {
@@ -188,7 +221,7 @@ index b52f4a9364..4402c0cb17 100644
 +    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.
@@ -196,13 +229,10 @@ index b52f4a9364..4402c0cb17 100644
 +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);
-+    aio_context_release(job_ctx);
++    job_cancel_sync(job, true);
 +    aio_co_enter(data->ctx, data->co);
 +}
-+
+ 
 +static void coroutine_fn pvebackup_co_cancel(void *opaque)
 +{
      Error *cancel_err = NULL;
@@ -214,13 +244,15 @@ index b52f4a9364..4402c0cb17 100644
  
      if (backup_state.vmaw) {
          /* make sure vma writer does not block anymore */
-@@ -342,27 +370,22 @@ static void pvebackup_cancel(void)
+@@ -342,28 +378,22 @@ 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);
+-        WITH_JOB_LOCK_GUARD() {
+-            job_ref_locked(&cancel_job->job);
+-        }
 +        CoCtxData data = {
 +            .ctx = qemu_get_current_aio_context(),
 +            .co = qemu_coroutine_self(),
@@ -235,11 +267,10 @@ index b52f4a9364..4402c0cb17 100644
 -    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);
--        job_unref(&cancel_job->job);
--        aio_context_release(aio_context);
+-        WITH_JOB_LOCK_GUARD() {
+-            job_cancel_sync_locked(&cancel_job->job, true);
+-            job_unref_locked(&cancel_job->job);
+-        }
 -    }
 +    qemu_co_mutex_unlock(&backup_state.backup_mutex);
  }
@@ -251,22 +282,74 @@ index b52f4a9364..4402c0cb17 100644
  }
  
  // assumes the caller holds backup_mutex
-@@ -415,6 +438,14 @@ static int coroutine_fn pvebackup_co_add_config(
+@@ -416,10 +446,18 @@ 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.
-+ * This does imply that this function cannot run with backup_mutex acquired.
-+ * That is ok because it is only ever called between setting up the backup_state
-+ * struct and starting the jobs, and from within a QMP call. This means that no
-+ * other QMP call can interrupt, and no background job is running yet.
++ * The caller is responsible that backup_mutex is held nonetheless.
 + */
- static bool create_backup_jobs(void) {
++static void create_backup_jobs_bh(void *opaque) {
  
      assert(!qemu_in_coroutine());
-@@ -523,11 +554,12 @@ typedef struct QmpBackupTask {
+ 
++    CoCtxData *data = (CoCtxData*)opaque;
++    Error **errp = (Error**)data->data;
++
+     Error *local_err = NULL;
+ 
+     /* create job transaction to synchronize bitmap commit and cancel all
+@@ -455,24 +493,19 @@ 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;
+@@ -485,13 +518,15 @@ static bool create_backup_jobs(void) {
+ 
+             if (di->job) {
+                 WITH_JOB_LOCK_GUARD() {
++                    job_cancel_sync_locked(&di->job->job, true);
+                     job_unref_locked(&di->job->job);
+                 }
+             }
+         }
+     }
+ 
+-    return errors;
++    /* return */
++    aio_co_enter(data->ctx, data->co);
+ }
+ 
+ typedef struct QmpBackupTask {
+@@ -528,11 +563,12 @@ typedef struct QmpBackupTask {
      UuidInfo *result;
  } QmpBackupTask;
  
@@ -280,7 +363,7 @@ index b52f4a9364..4402c0cb17 100644
      QmpBackupTask *task = opaque;
  
      task->result = NULL; // just to be sure
-@@ -548,8 +580,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -553,8 +589,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
      const char *firewall_name = "qemu-server.fw";
  
      if (backup_state.di_list) {
@@ -291,7 +374,7 @@ index b52f4a9364..4402c0cb17 100644
          return;
      }
  
-@@ -616,6 +649,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -621,6 +658,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
          }
          di->size = size;
          total += size;
@@ -300,24 +383,58 @@ index b52f4a9364..4402c0cb17 100644
      }
  
      uuid_generate(uuid);
-@@ -847,6 +882,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -852,6 +891,8 @@ 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);
  
-@@ -861,6 +897,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
+@@ -866,6 +907,33 @@ 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:
-@@ -903,6 +941,8 @@ err:
+@@ -888,6 +956,7 @@ err:
+         g_free(di);
+     }
+     g_list_free(di_list);
++    backup_state.di_list = NULL;
+ 
+     if (devs) {
+         g_strfreev(devs);
+@@ -908,6 +977,8 @@ err:
      }
  
      task->result = NULL;
@@ -326,7 +443,7 @@ index b52f4a9364..4402c0cb17 100644
      return;
  }
  
-@@ -956,22 +996,15 @@ UuidInfo *qmp_backup(
+@@ -961,24 +1032,8 @@ UuidInfo *qmp_backup(
          .errp = errp,
      };
  
@@ -334,23 +451,24 @@ index b52f4a9364..4402c0cb17 100644
 -
      block_on_coroutine_fn(pvebackup_co_prepare, &task);
  
-     if (*errp == NULL) {
-         bool errors = create_backup_jobs();
+-    if (*errp == NULL) {
+-        bool errors = create_backup_jobs();
 -        qemu_mutex_unlock(&backup_state.backup_mutex);
- 
-         if (!errors) {
+-
+-        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 */
-+            // start the first job in the transaction
-             job_txn_start_seq(backup_state.txn);
-         }
+-            job_txn_start_seq(backup_state.txn);
+-        }
 -    } else {
 -        qemu_mutex_unlock(&backup_state.backup_mutex);
-     }
- 
+-    }
+-
      return task.result;
-@@ -1025,6 +1058,7 @@ BackupStatus *qmp_query_backup(Error **errp)
+ }
+ 
+@@ -1030,6 +1085,7 @@ BackupStatus *qmp_query_backup(Error **errp)
      info->transferred = backup_state.stat.transferred;
      info->has_reused = true;
      info->reused = backup_state.stat.reused;
@@ -359,10 +477,10 @@ index b52f4a9364..4402c0cb17 100644
      qemu_mutex_unlock(&backup_state.stat.lock);
  
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 03fc0af99b..29650896e2 100644
+index 7fde927621..bf559c6d52 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -784,12 +784,15 @@
+@@ -770,12 +770,15 @@
  #
  # @uuid: uuid for this backup job
  #

debian/patches/pve/0039-PVE-fixup-pbs-restore-API.patch

@@ -1,44 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 6 Jul 2020 14:40:12 +0200
-Subject: [PATCH] PVE: fixup pbs-restore API
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pbs-restore.c | 10 ++++++++--
- 1 file changed, 8 insertions(+), 2 deletions(-)
-
-diff --git a/pbs-restore.c b/pbs-restore.c
-index 4bf37ef1fa..ff3fc38e8a 100644
---- a/pbs-restore.c
-+++ b/pbs-restore.c
-@@ -161,13 +161,19 @@ int main(int argc, char **argv)
-         fprintf(stderr, "connecting to repository '%s'\n", repository);
-     }
-     char *pbs_error = NULL;
--    ProxmoxRestoreHandle *conn = proxmox_restore_connect(
-+    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) {
-@@ -198,7 +204,7 @@ int main(int argc, char **argv)
-         fprintf(stderr, "starting to restore snapshot '%s'\n", snapshot);
-         fflush(stderr); // ensure we do not get printed after the progress log
-     }
--    int res = proxmox_restore_image(
-+    res = proxmox_restore_image(
-         conn,
-         archive_name,
-         write_callback,

debian/patches/pve/0040-PVE-Migrate-dirty-bitmap-state-via-savevm.patch

@@ -12,21 +12,22 @@ 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/Makefile.objs  |   1 +
+ migration/meson.build    |   2 +
+ migration/migration.c    |   1 +
  migration/pbs-state.c    | 106 +++++++++++++++++++++++++++++++++++++++
  pve-backup.c             |   1 +
  qapi/block-core.json     |   6 +++
- softmmu/vl.c             |   1 +
- 6 files changed, 118 insertions(+)
+ 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 34e7d75713..f83816dd3c 100644
+index 465906710d..4f0aeceb6f 100644
 --- a/include/migration/misc.h
 +++ b/include/migration/misc.h
-@@ -75,4 +75,7 @@ bool migration_in_incoming_postcopy(void);
+@@ -75,4 +75,7 @@ bool migration_in_bg_snapshot(void);
  /* migration/block-dirty-bitmap.c */
  void dirty_bitmap_mig_init(void);
  
@@ -34,18 +35,33 @@ index 34e7d75713..f83816dd3c 100644
 +void pbs_state_mig_init(void);
 +
  #endif
-diff --git a/migration/Makefile.objs b/migration/Makefile.objs
-index 0fc619e380..20b3792599 100644
---- a/migration/Makefile.objs
-+++ b/migration/Makefile.objs
-@@ -9,6 +9,7 @@ common-obj-y += qjson.o
- common-obj-y += block-dirty-bitmap.o
- common-obj-y += multifd.o
- common-obj-y += multifd-zlib.o
-+common-obj-y += pbs-state.o
- common-obj-$(CONFIG_ZSTD) += multifd-zstd.o
+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)
  
- common-obj-$(CONFIG_RDMA) += rdma.o
+ softmmu_ss.add(files(
+   'block-dirty-bitmap.c',
+diff --git a/migration/migration.c b/migration/migration.c
+index 9b496cce1d..421b4ee225 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
@@ -159,23 +175,24 @@ index 0000000000..29f2b3860d
 +                         NULL);
 +}
 diff --git a/pve-backup.c b/pve-backup.c
-index c7cde0fb0e..f65f1dda26 100644
+index 7bd9d06346..5662f48b72 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
-@@ -1130,5 +1130,6 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+@@ -1128,6 +1128,7 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
      ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
      ret->pbs_dirty_bitmap = true;
-     ret->query_bitmap_info = 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 29650896e2..0da4b35028 100644
+index bf559c6d52..24f30260c8 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -890,12 +890,18 @@
- #
- # @query-bitmap-info: True if the 'query-pbs-bitmap-info' QMP call is supported.
+@@ -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'
@@ -185,22 +202,11 @@ index 29650896e2..0da4b35028 100644
  # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
  #
  ##
- { 'struct': 'ProxmoxSupportStatus',
+@@ -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' } }
  
  ##
-diff --git a/softmmu/vl.c b/softmmu/vl.c
-index 16aa2186b0..88b13871fd 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -4288,6 +4288,7 @@ void qemu_init(int argc, char **argv, char **envp)
-     blk_mig_init();
-     ram_mig_init();
-     dirty_bitmap_mig_init();
-+    pbs_state_mig_init();
- 
-     qemu_opts_foreach(qemu_find_opts("mon"),
-                       mon_init_func, NULL, &error_fatal);

debian/patches/pve/0040-PVE-always-set-dirty-counter-for-non-incremental-bac.patch

@@ -1,30 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 6 Jul 2020 14:40:13 +0200
-Subject: [PATCH] PVE: always set dirty counter for non-incremental backups
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pve-backup.c | 8 ++++++--
- 1 file changed, 6 insertions(+), 2 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 7bf54b4c5d..1f2a0bbe8c 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -738,9 +738,13 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-                     dirty += bdrv_get_dirty_count(bitmap);
-                 }
-                 di->bitmap = bitmap;
--            } else if (bitmap != NULL) {
-+            } else {
-                 dirty += di->size;
--                bdrv_release_dirty_bitmap(bitmap);
-+
-+                /* 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);

debian/patches/pve/0041-PVE-use-proxmox_backup_check_incremental.patch

@@ -1,36 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 6 Jul 2020 14:40:14 +0200
-Subject: [PATCH] PVE: use proxmox_backup_check_incremental
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- pve-backup.c | 12 ++++++++----
- 1 file changed, 8 insertions(+), 4 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 1f2a0bbe8c..1cd9d31d7c 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -730,12 +730,16 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-                     if (!bitmap) {
-                         goto err;
-                     }
--                    /* mark entire bitmap as dirty to make full backup first */
--                    bdrv_set_dirty_bitmap(bitmap, 0, di->size);
--                    dirty += di->size;
-                 } else {
--                    expect_only_dirty = true;
-+                    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 {

debian/patches/pve/0041-migration-block-dirty-bitmap-migrate-other-bitmaps-e.patch

@@ -13,15 +13,16 @@ 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 5bf0d9fbc6..1070c19181 100644
+index 9aba7d9c22..f4ecf9c9f9 100644
 --- a/migration/block-dirty-bitmap.c
 +++ b/migration/block-dirty-bitmap.c
-@@ -323,7 +323,7 @@ static int add_bitmaps_to_list(DBMSaveState *s, BlockDriverState *bs,
+@@ -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);
@@ -29,4 +30,4 @@ index 5bf0d9fbc6..1070c19181 100644
 +            continue;
          }
  
-         bdrv_ref(bs);
+         if (bitmap_aliases) {

debian/patches/pve/0042-PVE-fall-back-to-open-iscsi-initiatorname.patch

@@ -0,0 +1,69 @@
+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 1bba42a71b..89cd032c3a 100644
+--- a/block/iscsi.c
++++ b/block/iscsi.c
+@@ -1388,12 +1388,42 @@ 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();

debian/patches/pve/0042-PVE-fixup-pbs-backup-add-compress-and-encrypt-option.patch

@@ -1,103 +0,0 @@
-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: fixup pbs backup, add compress and encrypt options
-
----
- block/monitor/block-hmp-cmds.c |  4 +++-
- pve-backup.c                   | 13 ++++++++++++-
- qapi/block-core.json           |  6 ++++++
- 3 files changed, 21 insertions(+), 2 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index fdc85a5c0e..43aa87487b 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1038,7 +1038,9 @@ 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 1cd9d31d7c..bfb648d6b5 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -567,6 +567,10 @@ 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;
-@@ -690,6 +694,7 @@ 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,
-@@ -699,8 +704,10 @@ 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,
-@@ -939,6 +946,8 @@ 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,
-@@ -967,6 +976,8 @@ UuidInfo *qmp_backup(
-         .firewall_file = firewall_file,
-         .has_devlist = has_devlist,
-         .devlist = devlist,
-+        .has_compress = has_compress,
-+        .has_encrypt = has_encrypt,
-         .has_speed = has_speed,
-         .speed = speed,
-         .errp = errp,
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index e00e577c6c..a177fea6cd 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -832,6 +832,10 @@
- #
- # @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
- #
- ##
-@@ -843,6 +847,8 @@
-                                     '*backup-id': 'str',
-                                     '*backup-time': 'int',
-                                     '*use-dirty-bitmap': 'bool',
-+                                    '*compress': 'bool',
-+                                    '*encrypt': 'bool',
-                                     '*format': 'BackupFormat',
-                                     '*config-file': 'str',
-                                     '*firewall-file': 'str',

debian/patches/pve/0043-PVE-Use-coroutine-QMP-for-backup-cancel_backup.patch

@@ -0,0 +1,598 @@
+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 f59b02592e..2e53cb65df 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1018,7 +1018,7 @@ 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;
+ 
+@@ -1027,7 +1027,7 @@ 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 fcf9461295..5fdb198ca4 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 5662f48b72..e4fe1b601d 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -354,7 +354,7 @@ 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");
+@@ -391,11 +391,6 @@ 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,
+@@ -529,50 +524,27 @@ 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;
+@@ -589,17 +561,17 @@ 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) {
+@@ -607,14 +579,14 @@ 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;
+             }
+@@ -637,7 +609,7 @@ 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;
+     }
+ 
+@@ -647,13 +619,13 @@ 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, -size, "bdrv_getlength failed");
++            error_setg_errno(errp, -size, "bdrv_getlength failed");
+             goto err;
+         }
+         di->size = size;
+@@ -680,47 +652,44 @@ 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;
+ 
+@@ -739,9 +708,9 @@ 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;
+                     }
+@@ -771,12 +740,12 @@ 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;
+             }
+ 
+@@ -790,10 +759,10 @@ 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;
+         }
+@@ -804,25 +773,25 @@ 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) {
+@@ -836,34 +805,34 @@ 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;
+         }
+     }
+@@ -881,7 +850,7 @@ 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);
+@@ -896,7 +865,7 @@ 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;
+@@ -906,8 +875,6 @@ 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.
+@@ -921,7 +888,7 @@ 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;
+     }
+ 
+@@ -934,7 +901,7 @@ 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);
+@@ -965,7 +932,7 @@ err:
+     if (vmaw) {
+         Error *err = NULL;
+         vma_writer_close(vmaw, &err);
+-        unlink(task->backup_file);
++        unlink(backup_file);
+     }
+ 
+     if (pbs) {
+@@ -976,65 +943,8 @@ 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 24f30260c8..4e8c35a3a2 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:

debian/patches/pve/0044-PBS-add-master-key-support.patch

@@ -0,0 +1,98 @@
+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 2e53cb65df..f98f4cf7e6 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 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 e4fe1b601d..41e8effa01 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -529,6 +529,7 @@ 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,
+@@ -677,6 +678,7 @@ 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,
+@@ -1040,5 +1042,6 @@ 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 4e8c35a3a2..d8c7331090 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' } }
+ 
+ ##

debian/patches/pve/0045-PVE-block-pbs-fast-path-reads-without-allocation-if-.patch

@@ -0,0 +1,53 @@
+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 9d1f1f39d4..ce9a870885 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 0;
+ }

debian/patches/pve/0045-pbs-fix-missing-crypt-and-compress-parameters.patch

@@ -1,43 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Fri, 10 Jul 2020 13:22:35 +0200
-Subject: [PATCH] pbs: fix missing crypt and compress parameters
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- pve-backup.c | 8 ++++++--
- 1 file changed, 6 insertions(+), 2 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index 6bf138cfc6..cd3a132d8b 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -958,6 +958,8 @@ 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,
-@@ -968,6 +970,10 @@ 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,
-@@ -976,8 +982,6 @@ UuidInfo *qmp_backup(
-         .firewall_file = firewall_file,
-         .has_devlist = has_devlist,
-         .devlist = devlist,
--        .has_compress = has_compress,
--        .has_encrypt = has_encrypt,
-         .has_speed = has_speed,
-         .speed = speed,
-         .errp = errp,

debian/patches/pve/0046-PVE-block-stream-increase-chunk-size.patch

@@ -0,0 +1,25 @@
+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 {

debian/patches/pve/0046-PVE-handle-PBS-write-callback-with-big-blocks-correc.patch

@@ -1,76 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 19 Aug 2020 12:33:17 +0200
-Subject: [PATCH] PVE: handle PBS write callback with big blocks correctly
-
-Under certain conditions QEMU will push more than the given blocksize
-into the callback at once. Handle it like VMA does, by iterating the
-data until all is written.
-
-The block size is stored per backup device to be used in the callback.
-This avoids relying on PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE, in case it is
-made configurable in the future.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pve-backup.c | 30 ++++++++++++++++++++++--------
- 1 file changed, 22 insertions(+), 8 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index cd3a132d8b..f14273645a 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -67,6 +67,7 @@ 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];
-@@ -147,17 +148,28 @@ 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,
-+            buf ? buf + transferred : NULL, 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, !buf ? size : 0, reused);
-+
-     return size;
- }
- 
-@@ -726,6 +738,8 @@ 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);

debian/patches/pve/0047-PVE-add-zero-block-handling-to-PBS-dump-callback.patch

@@ -1,85 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 13 Aug 2020 13:50:27 +0200
-Subject: [PATCH] PVE: add zero block handling to PBS dump callback
-
-Both the PBS and VMA dump callbacks assume that a NULL pointer can be
-passed as *pbuf, but that never happens, as backup-dump.c calls this
-function with contents of an iovec.
-
-So first, remove that assumption and add an 'assert' to verify.
-
-Secondly, while the vma-writer already does the buffer_is_zero check
-internally, for PBS we relied on that non-existant behaviour for zero
-chunks, so do the buffer_is_zero check manually and pass NULL to the
-rust lib in case it is true.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pve-backup.c | 14 +++++++++-----
- 1 file changed, 9 insertions(+), 5 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index f14273645a..bd802c6205 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 */
- 
-@@ -136,10 +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
-@@ -155,7 +159,8 @@ pvebackup_co_dump_pbs_cb(
-         uint64_t to_transfer = left < di->block_size ? left : di->block_size;
- 
-         pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id,
--            buf ? buf + transferred : NULL, start + transferred, to_transfer, &local_err);
-+            is_zero_block ? NULL : buf + transferred, start + transferred,
-+            to_transfer, &local_err);
-         transferred += to_transfer;
- 
-         if (pbs_res < 0) {
-@@ -168,7 +173,7 @@ pvebackup_co_dump_pbs_cb(
-     }
- 
-     qemu_co_mutex_unlock(&backup_state.dump_callback_mutex);
--    pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
-+    pvebackup_add_transfered_bytes(size, is_zero_block ? size : 0, reused);
- 
-     return size;
- }
-@@ -190,6 +195,7 @@ pvebackup_co_dump_vma_cb(
-     int ret = -1;
- 
-     assert(backup_state.vmaw);
-+    assert(buf);
- 
-     uint64_t remaining = size;
- 
-@@ -216,9 +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);

debian/patches/pve/0047-block-io-accept-NULL-qiov-in-bdrv_pad_request.patch

@@ -0,0 +1,33 @@
+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 4589a58917..20167d61b7 100644
+--- a/block/io.c
++++ b/block/io.c
+@@ -1730,6 +1730,10 @@ 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)) {

debian/patches/pve/0048-block-add-alloc-track-driver.patch

@@ -0,0 +1,403 @@
+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>
+[FE: adapt to changed function signatures
+     make error return value consistent with QEMU]
+Signed-off-by: Fiona 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..43d40d11af
+--- /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 -EIO;
++    }
++
++    /* 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 7ef2fa72d5..15352f579f 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',

debian/patches/pve/0049-PVE-redirect-stderr-to-journal-when-daemonized.patch

@@ -1,52 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Tue, 30 Jun 2020 13:10:10 +0200
-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>
----
- Makefile.objs | 1 +
- os-posix.c    | 7 +++++--
- 2 files changed, 6 insertions(+), 2 deletions(-)
-
-diff --git a/Makefile.objs b/Makefile.objs
-index 240eb503f2..c7ba4e11e7 100644
---- a/Makefile.objs
-+++ b/Makefile.objs
-@@ -54,6 +54,7 @@ common-obj-y += net/
- common-obj-y += qdev-monitor.o
- common-obj-$(CONFIG_WIN32) += os-win32.o
- common-obj-$(CONFIG_POSIX) += os-posix.o
-+os-posix.o-libs := -lsystemd
- 
- common-obj-$(CONFIG_LINUX) += fsdev/
- 
-diff --git a/os-posix.c b/os-posix.c
-index 3572db3f44..b45dde63ac 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. */
-@@ -312,9 +314,10 @@ 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);

debian/patches/pve/0049-PVE-savevm-async-register-yank-before-migration_inco.patch

@@ -0,0 +1,35 @@
+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 a38e7351c1..0b1b60c6ae 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 */
+ 
+@@ -521,6 +522,10 @@ 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");

debian/patches/pve/0050-qemu-img-dd-add-l-option-for-loading-a-snapshot.patch

@@ -0,0 +1,130 @@
+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 5e713e231d..9390d5e5cf 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 4f5ef5b887..4894016ad2 100644
+--- a/qemu-img.c
++++ b/qemu-img.c
+@@ -4957,6 +4957,7 @@ 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;
+@@ -4966,6 +4967,7 @@ static int img_dd(int argc, char **argv)
+     int64_t size = 0, readsize = 0;
+     int64_t out_pos, in_pos;
+     bool force_share = false, skip_create = false;
++    const char *snapshot_name = NULL;
+     struct DdInfo dd = {
+         .flags = 0,
+         .count = 0,
+@@ -5003,7 +5005,7 @@ 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;
+         }
+@@ -5026,6 +5028,19 @@ 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;
+@@ -5085,11 +5100,24 @@ 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) {
+@@ -5244,6 +5272,7 @@ 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);

debian/patches/pve/0051-vma-allow-partial-restore.patch

@@ -0,0 +1,407 @@
+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);

debian/patches/pve/0052-pbs-namespace-support.patch

@@ -0,0 +1,233 @@
+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 f98f4cf7e6..55ef4f5965 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1043,6 +1043,7 @@ 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 ce9a870885..9192f3e41b 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 41e8effa01..1c25ae98bd 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 */
+ 
+ /*
+@@ -531,6 +533,7 @@ 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,
+@@ -670,8 +673,9 @@ 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 d8c7331090..889726fc26 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',
+@@ -3290,7 +3293,7 @@
+ { 'struct': 'BlockdevOptionsPbs',
+   'data': { 'repository': 'str', 'snapshot': 'str', 'archive': 'str',
+             '*keyfile': 'str', '*password': 'str', '*fingerprint': 'str',
+-            '*key_password': 'str' } }
++            '*key_password': 'str', '*namespace': 'str' } }
+ 
+ ##
+ # @BlockdevOptionsNVMe: